ptq q3 2013
TRANSCRIPT
special features
instrumentation & control
rotating equipment
refininggas processingpetrochemicals
petroleum technology quarterly
ptqQ3 2013
cover and spine copy 16.indd 1 10/06/2013 14:05
1219
_e
Whatever the impurity, whatever the composition, Air Liquide Global E&C Solutionshas the right treatment.
Does your raw natural gas contain hydrogen sulfi de,carbon dioxide, mercaptans or more?
The composition of natural gas varies tremendously: almost every source contains a diff erent blend of impurities. The options for treatment are almostas diverse. That’s why off ering a solution specifi cally designed for your gas fi eld
is crucial. We as your partner of choice provide solutions for all types of natural gas, including associated and unconventional gas, from a single source. Customised and effi cient.
Come and visit us at: MIOGE 2013, 25 - 28 June in Moscow, RussiaPavilion 1 | Booth A 524
www.engineering-solutions.airliquide.com
air liquide.indd 1 06/06/2013 10:13
©2013. The entire content of this publication is protected by copyright full details of which are available from the publishers. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means – electronic, mechanical, photocopying, recording or otherwise – without the prior permission of the copyright owner.The opinions and views expressed by the authors in this publication are not necessarily those of the editor or publisher and while every care has been taken in the preparation of all material included in Petroleum Technology Quarterly the publisher cannot be held responsible for any statements, opinions or views or for any inaccuracies.
3 The shake-out from shale ChrisCunningham 5 Technology in Action
9 ptq&a 11 Towards a zero gasoline production refinery: part 1 BlasisStamaterisFoster Wheeler
21 Refinery power failures: causes, costs and solutions PatrickJChristensen,WilliamHGrafandThomasWYeung Hydrocarbon Publishing Company
29 Combating green oil formation in a CCR reformer OsmanKubilayKaran,MehmetAsimAyandKorayKahramanTüpras Kirikkale refinery
35 Role of fired heater safety systems NikkiBishopandDavidSheppard Emerson Process Management
43 Manufacturing execution systems for the refining industry MartyMoranAspenTech
47 Advances in engineering and design technologies SimonBennett Aveva
53 A novel approach to cleaning furnace coils RupaliSahu,ShyamKishoreChoudhary,UgrasenYadavandMKEPrasad Technip KT India
63 Extending gasification runtime with antifoulant BertholdOtziskKurita Europe
67 Improving reliability and efficiency in centrifugal pumps GeoffLewisDuPont
71 Oil sands-derived feed processing MaxOvchinnikov,JosianeGinestra,DorianRauschning,BillGillespie andKevinCarlson Criterion Catalysts & Technologies
85 Predicting catalyst lifetime SRezaSeifMohaddecyandSepehrSadighiRIPI
95 Additives provide flexibility for FCC units and delayed cokers AlanKramerandRaulArriagaAlbemarle Corporation 103 Preventing ingress of HCN into amine systems RalphWeiland,NathanHatcherandClaytonJonesOptimized Gas Treating, Inc
109 Optimised hydrogen production by steam reforming: part 2 KedarVPatwardhan,SankeRajyalakshmiandPBalaramakrishnaLarsen & Toubro
115 Corrosion control with high-acid crudes IndiaNagi-Hanspal,MaheshSubramaniyamandParagShahDorf Ketal Chemicals
123 Chemical analysis in amine system operations ScottWaite,ArthurCummingsandGlenSmith MPR Services
133 Industry News
OMV’sSchwechatrefineryinVienna,wherebutadieneproductionistobeexpandedinresponsetomarketgrowth. Photo: OMV/Flickr
Q3 (Jul, Aug, Sept) 2013www.eptq.com
ptqYLRETRAUQYGOLONHCET MUELORTEP
ed com copy.indd 1 11/06/2013 12:16
AMER: +1 281 293 8200 ASIA: +65 6735 5488 EMEA: +44 1932 242424
[email protected] | www.kbcat.com | blog.kbcat.com
KBC Advanced Technologies plc
SOLUTIONS ACROSS THE VALUE CHAIN
GLOBAL SOFTWARE AND CONSULTING SOLUTIONS FOR
UPSTREAM | GAS PROCESSING | REFINING | PETROCHEMICAL
Linking cutting edge technology with consulting expertise for over 30 years
KBC Advanced Technologies plcKBC Advanced Technologies plcKBC Advanced Technologies plc
GLOBAL SOFTWARE AND CONSULTING SOLUTIONS FORGLOBAL SOFTWARE AND CONSULTING SOLUTIONS FORGLOBAL SOFTWARE AND CONSULTING SOLUTIONS FORGLOBAL SOFTWARE AND CONSULTING SOLUTIONS FORGLOBAL SOFTWARE AND CONSULTING SOLUTIONS FORGLOBAL SOFTWARE AND CONSULTING SOLUTIONS FORGLOBAL SOFTWARE AND CONSULTING SOLUTIONS FOR
UPSTREAM UPSTREAM || GAS PROCESSING GAS PROCESSING GAS PROCESSING GAS PROCESSING GAS PROCESSING GAS PROCESSING | | REFINING REFINING REFINING REFINING | PETROCHEMICAL PETROCHEMICAL PETROCHEMICAL
| Strategic Investment Support
| Asset Optimisation
| Sustainable Workforce Development
C
M
Y
CM
MY
CY
CMY
K
KBC Adv DS - PTQ.pdf 1 6/7/2013 12:54:59 PM
kbc.indd 1 10/06/2013 11:18
The European Union has arguably been the global leader in biodiesel production and use, with overall
biodiesel production increasing from 1.9 million tonnes in 2004 to nearly 10.3 million tonnes in 2007. Biodiesel production in the US has also increased dramatically in the past few years from 2 million gallons in 2000 to approximately 450 million gallons in 2007. According to the National Biodiesel Board, 171 companies own biodiesel manufacturing plants and are actively marketing biodiesel.1. The global biodiesel market is estimated to reach 37 billion gallons by 2016, with an average annual growth rate of 42%. Europe will continue to be the major biodiesel market for the next decade, followed closely by the US market.
Although high energy prices, increasing global demand, drought and other factors are the primary drivers for higher food prices, food competitive feedstocks have long been and will continue to be a major concern for the development of biofu-els. To compete, the industry has responded by developing methods to increase process efficiency, utilise or upgrade by-products and operate with lower quality lipids as feedstocks.
Feedstocks
Biodiesel refers to a diesel-equivalent fuel consisting of short-chain alkyl (methyl or ethyl) esters, made by the transesterification of triglycerides, commonly known as vegetable oils or animal fats. The most common form uses methanol, the cheapest alcohol available, to produce methyl esters. The molecules in biodiesel are pri-marily fatty acid methyl esters (FAME), usually created by trans-esterification between fats and metha-nol. Currently, biodiesel is produced from various vegetable and plant oils. First-generation food-based feedstocks are straight vegetable oils such as soybean oil and animal fats such as tallow, lard, yellow grease, chicken fat and the by-products of the production of Omega-3 fatty acids from fish oil. Soybean oil and rapeseeds oil are the common source for biodiesel produc-tion in the US and Europe in quanti-ties that can produce enough biodie-sel to be used in a commercial market with currently applicable
PTQ Q3 2013 3
Editor Chris Cunningham [email protected]
Production EditorRachel [email protected]
Graphics EditorRob Fris [email protected]
Editorial tel +44 844 5888 773fax +44 844 5888 667
Business Development DirectorPaul [email protected] Advertising SalesBob [email protected]
Advertising Sales Officetel +44 844 5888 771 fax +44 844 5888 662
PublisherNic [email protected]
CirculationJacki [email protected]
Crambeth Allen Publishing LtdHopesay, Craven Arms SY7 8HD, UKtel +44 844 5888 776fax +44 844 5888 667
ptq (petroleum technology quarterly) (ISSN No: 1632-363X, USPS No: 014-781) is published quarterly plus annual Catalysis edition by Crambeth Allen Publishing Ltd and is distributed in the US by SP/Asendia, 17B South Middlesex Avenue, Monroe NJ 08831. Periodicals postage paid at New Brunswick, NJ. Postmaster: send address changes to ptq (petroleum technology quarterly), 17B South Middlesex Avenue, Monroe NJ 08831.Back numbers available from the Publisher at $30 per copy inc postage.
Vol 18 No 4
Q3 (Jul, Aug, Sept) 2013
The shake-out from shale
The reshaping of global crude supply in response to the shale oil boom is beginning to take shape. According to the International Energy Agency (IEA) in its latest biannual report (Medium-Term Oil Market Report 2013)
on energy supply covering five years ahead, upgraded estimates of the US’s reserves of shale oil mean that the nation will be the source of a third of all new oil supplies during the period. This reinforces a general trend for the US to morph from being the world’s biggest refiner of imported crude to becoming a potential net exporter. Combined advances in shale oil and gas production will mean that by 2035 the nation will be all but self sufficient in energy supplies, says the IEA.
US production is expected to grow by close to four million b/d above its 2012 level by 2018, which is 60-70% of all growth in oil output outside the OPEC bloc. Meanwhile, expected growth in Canadian oil sands production will reinforce the North American surge in supply.
Talk of North America as a major exporter of crude remains pointless while the law remains firmly on the side of maintaining strategic reserves of oil over the earnings potential of exports. However, the IEA points out that the chief reason for the defensive American view — global reliance on imports from the Middle East — is set to end pretty soon. The agency says that it expects production capacity in the Middle East to continue to grow over the next five years, but this will be at a rate slower than in recent years. The IEA has downgraded its estimate of OPEC production in 2018 by 750 000 b/d, citing insecurity in Africa following political turmoil in the north of the continent as the chief reason for a slowing in investment and capacity growth.
On the basis that one bout of fortune is balanced by an attack of misfor-tune elsewhere, Middle East producers might envisage a more productive outlook in response to the continuing grim story for refining in Europe. According to France’s Total, the region’s flatlining economic performance, a general desire simply to consume less oil, and the pressures of meeting car-bon emissions targets will lead to further closures of refineries in the coming few years. In France so far this year, Petroplus’s Petit-Couronne refinery is in the process of closure, while Total’s Dunkirk plant has stopped production altogether. With Middle East exports increasingly geared towards Euro standard products, Europe could be an important stabilising influence on OPEC’s performance.
Another export market that all major producers will be keeping an eye on is China. The PRC’s shale energy supply majors on gas; although the nation is looking to expand shale oil supplies, the outlook is for its refining industry to remain largely reliant on imports. The local refining sector has been under-performing recently, in part because of a slow-down in the general economy, but also because of restrictive regulations concerning oil imports. Only the state-owned majors are permitted to import crude, while imports for privately owned refiners are limited to heavy fuel oil, which does not meet their requirements. In short, bureaucracy impedes progress in China’s refining sector. Some liberalisation in that particular market should set a few exporters’ pulses racing.
CHRIS CUNNINGHAM
ptqYLRETRAUQYGOLONHCET MUELORTEP
ed com copy.indd 2 10/06/2013 14:59
“ There’s only one way to stay ahead of the competition: always be evolving.”
Meet Max Ovchinnikov:Industry Game-Changer.
When you’re leading the world in catalyst development, your work is never done – and that’s just fi ne by Max Ovchinnikov, CRITERION’s Senior Research Chemist and co-developer of such advanced catalyst platforms as SENTRY and CENTERA®. He, along with fellow scientists from Shell’s three R&D centers, has made a career developing high throughput techniques that lead to the continual improvement of catalysts. This means as Max is evolving the company’s process and products, and launching a new catalyst every three or four years, CRITERION is always in front when it comes to innovation.
Leading minds. Advanced technologies.
www.CRITERIONCatalysts.com
criterion.indd 1 07/06/2013 16:11
A refiner took advantage of the relatively high differ-ential in price between hydrogen and transportation fuel to convert an existing distillate hydrotreater to a mild hydrocracking operation, with resulting increased overall volume gain as well as increased flexibility to optimise the refinery’s product mix.
The conversion involved Valero’s Memphis refinery, which was originally commissioned with a distillate hydrotreater (DHT) as part of the refinery’s ULSD proj-ect. The original design fed a blend of light cycle oil (LCO), kerosene and crude unit gas oil (CUGO) at a rate of 33 000 b/d. The unit includes a single four-bed reactor and operates at a nominal pressure of 1200 psig.
Diesel product from the bottom of the stripper goes directly into the ULSD blending pool. The naphtha product from the stripper overhead goes to a naphtha hydrotreating (NHT) fractionator, and the NHT fractionator overhead goes to gasoline blending, while the fractionator’s bottoms go to a UOP CCR Platforming Unit.
In 2010, Valero asked UOP to carry out a revamp process study to evaluate the modifications required to convert the unit into a mild hydrocracker (MHC). Valero wanted the MHC to process 21 000 b/d of LCO to produce ULSD blend stock and additional naphtha. Since the refinery had spare hydrotreating capacity else-where in the refinery, it was not necessary to continue to co-process kerosene and CUGO at the MHC. The results of the process study indicated that the conversion could be accomplished simply by the addition of a high-rate depressuring system and the replacement of the hydro-treating catalyst in the bottom bed and 40% of Bed 3 of the reactor with UOP’s Unicracking catalyst.
In addition to these two modifications, the refinery decided to install a steam generator at the reactor outlet for additional effluent heat removal and to replace several relief valves in order to allow the unit pressure to be raised slightly. The revamp was completed early in 2012 and the unit was recommissioned as a MHC soon afterwards.
Before the revamp, makeup hydrogen availability was often a limiting factor for the DHT. During much of that time, the refinery was octane-long as a result of running the reformer at high severity in order to produce adequate hydrogen for the DHT. In 2012, along with the conversion of the unit to a MHC a new steam-methane hydrogen plant was added, enabling the refinery to reduce the naphtha reformer’s severity and optimise the octane balance. The refinery’s economics were therefore enhanced by replacing expensive hydrogen from the naphtha reformer with cheaper hydrogen from the new hydrogen plant.
After two-and-a-half months of operation, the lower bed temperatures were increased to approximately 370°C and the conversion increased to approximately 20%. During periods of high diesel-to-gasoline price differentials or high natural gas prices, the unit can be operated as a DHT by quenching the bottom beds. During periods of lower diesel-gasoline price differen-tials or low natural gas prices, the lower bed temperatures can be increased to effect higher volume swell.
Applying a bulk metal catalyst in the pretreat section of a once-through heavy feed hydrocracker process-ing vacuum gas oil (VGO) enabled a refiner to achieve gains including reduced nitrogen slip, increased distillate yields, unconverted bottoms reduced by 50% and savings from reduced furnace firing. The application involved an ExxonMobil refin-ery replacing alumina-supported catalyst with Nebula, a bulk metal catalyst developed jointly by ExxonMobil Research and Engineering Company and Albemarle Corporation. The principle was that the bulk metal catalyst contains a higher proportion of more active Type II sites.
Before loading, the hydrocracker’s performance and economics were evaluated. This modelling study indi-cated that replacing 25% of supported catalyst with Nebula would deliver the best balance of benefit and cost.
Nebula was fed into the lower of two pretreat reactor catalyst beds, and the unit has operated well since startup, processing feed that is similar in qual-ity and feed rate to the material processed in previous cycles.
Capturing value from deploying high-activity bulk metal catalysts depends on the economic priorities under which the unit is operated, but can include increased challenge feed processing, improved product quality, increased unit throughput, increased cycle length and reduced operating expense.
Technology in Action
Bulk metal catalyst boosts hydrocracking
MHC benefits from low gas prices
A new steam-methane hydrogen plant enabled the refinery to reduce the naphtha reformer’s severity and optimise the octane balance
www.eptq.com PTQ Q3 2013 5
case studies copy 7.indd 1 10/06/2013 12:12
Alon was able to increase feed throughput, even at the height of summer. The unit processed a higher feed rate while on the Rive catalyst at a given ambient temperature. On average while on the Rive catalyst, the FCC unit was processing over 700 b/d of addi-tional feed.
A permitted 125 tons of catalyst were produced for the trial. Using this combination of Rive zeolite and Grace matrix technology, formulations were developed for ACE testing with Alon’s feedstock. The ACE unit is an industry-accepted tool for evaluating FCC catalysts in the laboratory and then predicting commercial performance. The results of these tests were used to build an economic model for the refinery that predicted a $2.00/FCC bbl value increase over the incumbent catalyst, substantial enough to justify a commercial trial.
As the Rive catalyst changed out in the unit, coke selectivity improved and feed rate to the FCC unit increased steadily despite an increase in the ambient temperatures. Slurry yields decreased with an increase in slurry density due to improved bottoms upgrading. Based on the improvement in coke selectivity and improved bottoms upgrading, the refinery progres-sively increased the feed rate and lowered the riser and regenerator temperatures by a total of 20°C over several weeks while maintaining a carbon on regener-ated catalyst level less than 0.3%.
Technip has appointed Emerson Process Management as the main supplier of control valves for the Algiers Refinery rehabilitation and adaptation project. The Algiers refinery is located 10 km south of Algiers and is being revamped to increase crude oil processing capacity and ensure it can produce gasoline at specifi-cations similar to those in Europe.
Emerson will supply approximately 600 of its Fisher control valves and regulators, including rotary, globe and butterfly-type devices. The valves will be deployed throughout the refinery, including within the residual FCC unit, naphtha processing units and new LPG storage unit. According to Emerson, the moderni-sation of its instrumentation and control valves will help the refinery to optimise process units and reduce maintenance and plant shutdowns. The Fisher valves will be supplied with a range of trims and silencers for severe service applications. The company’s ValveLink software will work with the existing control system to provide online diagnostics to help identify potential problems.
The Algiers refinery was built in 1964 and produces motor fuels and LPG for the local market, and naphtha and fuel oil for export. The refinery is owned by Sonatrach. Technip is the EPC contractor for the Rehabilitation and Adaptation project. The project is part of a broader refinery upgrade programme for the production of clean fuels in Algeria.
Trials with the second generation of a new FCC cata-lyst achieved over $2.50/FCC bbl of additional value in mild resid operation along with gains in perfor-mance, including improved coke selectivity, improved bottoms upgrading and increased olefinicity of the cracked products. No capital investment in plant and equipment was needed to achieve these gains and operational changes are reported to be within the normal range of practice.
The trials involved the FCC unit at the Alon Big Spring plant in Texas and featured Rive Technology’s Molecular Highway zeolite and Grace matrix technology. The feed to the FCC unit is a mix of VGOs and mildly hydrotreated propane deasphalting (PDA) oil.
Prior to the trial, the unit typically operated with a riser temperature of around 540°C, to obtain maximum LCO conversion. Reactor product vapours were quenched with LCO to minimise the dry gas rate. The regenerator bed temperature was controlled to maintain a carbon on regenerated catalyst level of below 0.3 wt%.
The unit is a UOP stacked design, revamped to include an external vertical riser with updated injec-tion nozzles. The riser terminates into a pair of primary cyclones that discharge the spent catalyst via dip legs in the stripper. The spent catalyst is subjected to steam for stripping absorbed product vapours before it flows down the spent catalyst standpipe into the regenerator. The riser product vapours exiting the primary cyclone gas tubes are quenched by LCO sprays. The quenched product vapours and stripping steam leave the reactor via a pair of secondary cyclones, which further separate entrained catalyst.
Grace manufactured 328 tons of MH-1 catalyst for Alon. Rive MH-1 had an average zeolite surface area of 227 m2/g, a matrix surface area of 100 m2/g and a Grace Davison Attrition Index (DI) of 6. The existing catalyst had a similar fresh zeolite surface area and a slightly lower matrix surface area. Catalyst quantity was sufficient for a 109-day trial (at 3 tons/day) and was projected to yield an 80% change-out.
The FCC unit at Alon Big Spring is constrained by air blower capacity, particularly in summer. The constraint is severe enough that during summer the FCC feed rate cycles daily in sync with the ambient temperature, as air density affects the air blower rate. Consequently, the refinery builds up an inventory of unprocessed FCC feedstock over the summer.
Control valves selected for North African refiner
On average while on the Rive catalyst, the FCC unit was processing over 700 b/d of additional feed
Zeolite matrix improves FCC performance
6 PTQ Q3 2013 www.eptq.com
case studies copy 7.indd 2 11/06/2013 12:16
Nasty StuffHeavy crudes are here to stay. As long as oil prices remain high,Canadian, Venezuelan, Deep WaterGulf of Mexico, Mexican andother low API gravity crude oilswill play an ever more importantrole in supplying world refineries.And prices promise to remain highbecause gainsayers notwithstanding,Hubbert was right.
A big question is how to best handlethese nasty crudes? Do you revamp existing units or invest innew capacity? With refineries nowrunning flat out, the balance mightseem to favor grass roots expansion,but given the substantial cost multiplier over revamps, this could
be questioned. Whichever the case,however, an inescapable fact is that the process design of the projectwill prove crucial. Between thecharge pump, the desalter and theunits' distillation columns there aremany places where miscalcula-tions in the process design couldwreck the entire project.
Can you really be sure of attainingdesired crude rates? Desalting viscous crude is extremely difficult.Minimizing coking or asphalteneprecipitation in the heaters demandsextreme care. Can you reasonablyexpect high diesel and HVGOrecoveries, acceptable levels ofnickel, vanadium, and microcarbon
residue (MCR)? Refiners who cutdeep should not be surprised whenthe HVGO product MCR is over 2wt % and the vanadium content isin excess of 10 ppmw. Any one ofsuch difficulties can result in lowerrevenue, unstable operation oreven unit shutdown. It is critical tounderstand that the inherent properties of these low API gravitycrudes dictate that exact processdesign is of paramount importance.
The point of this litany of possibleproblems is to remind you not toskimp in the early phases of engineering. From the start of theLP work through the completion offront-end process engineering,actual product yield and qualitiesdepend on the process design.
The message is clear. Nasty crudeswill continue to make up anincreasing proportion of refineries'crude slates. But time is precious.The sooner we face this fact, unwelcome as it may be, the moreexpeditiously we can adapt.
For a more in depth review ofheavy crude challenges, ask us forTechnical Papers 173, 185 and197.
3400 BissonnetSuite 130Houston, Texas 77005USA
Ph: [1] (713) 665-7046Fx: [1] (713) [email protected]
This exchanger has seen better crude slates
pcs nasty.indd 1 1/6/12 11:58:35
Q What are the main features I should expect of a safety instrumented system for refinery fired heaters?
A Luis Duran, Senior Specialist, ABB, [email protected], the industry has asked for design and certification to industry or application standards such as API 560, 556 or NFPA 87. However, while those standards are applicable, they should be the starting point rather than the end of the journey. In the process industries such as refining, it is important to look for compliance to IEC 61511/ISA 84, too.
Applicable NFPA and API standards do an excellent job of defining the protective functions that are required in the logic system to mitigate risk. However, performance-based standards start with an assessment of the actual risk in a particular installation and, by adopting the standard, users will gain a solid design, validation methodologies, and a maintenance and test-ing regime in addition to the protective functions.
Q Is electrostatic desalting a useful step to take for crude?
A Berthold Otzisk, Consulting Engineer, Kurita Europe, [email protected] desalting significantly reduces the corrosion potential in the CDU overhead system. Crude oil contains water, solids and salts. The main target is the removal of these contaminants to the lowest level before crude oil is fed to the CDU. The electrostatic coalescing in the desalter vessel will provide a good dehydration of the crude oil with significant removal of salts such as NaCl, MgCl2, CaCl2. Single-stage desalter systems will give >90% desalting efficiency with <10 ppm chlorides in the desalted crude oil. Two-stage desalter systems will give >98% desalting efficiency with <2 ppm chlo-rides in the desalted crude oil. Mainly MgCl2 and some CaCl2 will hydrolyse above 150°C to form corrosive HCl gas, which can enter the overhead system.
Q Under what conditions can a steam turbine act as the prime mover for a reciprocating compressor?
A Miralem Okanovic, Product Manager Compressor Systems, Burckhardt, [email protected] factors need to be considered when selecting a driver for a reciprocating compressor: economically based factors (capital and operating costs); applica-tion/process-specific requirements (demand for steam); availability of electricity on site; reliability of electricity supply (ie, continued operation when electrical service is lost); and local prices for electricity.
Steam turbines can be desirable when steam is required as a source of hydrogen in steam methane reforming (SMR), which is an important process for many chemical and petroleum refining applications. Steam is also required to control the pressures and temperatures of many chemical processes. However, for these applications, electrical motors are replacing steam turbines as the prime mover for reciprocating compressors. One reason is that they have lower capi-tal costs. By comparison, a steam turbine system has a lot of auxiliaries such a local boiler, piping, instrumen-tation, condensate return and water treatment.
Q What is the most effective way to prevent plugging in resid lines: steam or electric tracing?
A Koen Verleyen, Product & Marketing Manager/Industrial Solutions, Pentair Thermal Management, [email protected] is the type of liquid in the residue line, what is the melting temperature, and where is the line located? When the lines are installed inside the battery and there is free process steam available at medium pres-sure, steam tracing could be considered, unless temperatures are low (steam at 1 bar = 100°C) or steam tracing pressure is high (pressure directives in design).
Installation and maintenance costs are typically much higher on a steam tracing system as return trac-ing lines are needed, there are design implications at higher pressures and corrosion issues, while electric tracing systems are more energy efficient, better to monitor and control, and provide long-term reliability.
Q What catalyst/additive options are there for maximising butylene from our FCC operations?
A Ken Bruno, Global Applications Technology Manager, FCC, Albemarle, [email protected] key to maximising butylene (and propylene) from the FCC unit is a combination of optimum operating conditions and catalyst selection. The ideal catalyst to maximise butylene (and propylene) is designed with high accessibility (superior diffusion character), low hydrogen transfer, a zeolite-to-matrix ratio targeting the overall product slate, and specialised zeolite technol-ogy. Albemarle’s AFX and Action catalyst families are formulated to maximise LPG olefins. After selecting the optimal catalyst, incremental trimming of butylene (and propylene) is possible with additives. Albemarle offers a range of additives to maximise LPG olefins, including DuraZoom and other speciality products.
www.eptq.com PTQ Q3 2013 9
ptq&a
Additional Q&A can be found at www.eptq.com/QandA
Q&A copy 14.indd 1 10/06/2013 19:20
3400 BissonnetSuite 130Houston, Texas 77005USA
Ph: [1] (713) 665-7046Fx: [1] (713) [email protected]
PROCESSCONSULTINGSERVICES,INC.
upsets from water slugs andother unpredictable situationsthat have damaged internals,resulting in diluent losses andhigh vacuum unit overhead con-densable oil. Diluent is neithercheap nor plentiful, and highvacuum column operating pres-sure will reduce overall liquidvolume yields. And if the designof the delayed coker fractionatoris based on today’s experiencewith conventional heavy feed-stocks you will be lucky to runsix months.What all this means is thatspecial process and equipmentdesigns are needed to satisfythe special demands of pro-cessing oil sands crudes. Suchprocesses are not generated bycomputer based designers whohave little or no experience andnever leave the office. They aredeveloped only by engineerswith know-how who have realexperience wearing Nomex® suitsand measuring true unit per-formance in Northern Alberta.Shouldn’t this be kept in mindby those considering long termsupply agreements?
Oil Sands Crude– Profits andProblems?Canadian bitumen productioncurrently runs about 1 MMbpd,with some being sold as Synbitand Dilbit. Over the next 10-12years output is expected toincrease to 3.5 MMbpd and morerefiners will begin investing toprocess it and come to dependon the Synbit and Dilbit for asignificant part of their supply.Few today, however, have everprocessed these feeds at highblend ratios, and are unawarethat conventional process andequipment designs are not upto the job. Canadian oil sands
feedstocks are extremely hardto desalt, difficult to vaporize,thermally unstable, corrosive, andproduce high di-olefin productfrom the coker. If you intend tolock into a long-term supply,therefore, it is imperative that youconsider reliability and run lengthfrom a particular design.Too low tube velocity in thevacuum heater tubes will lead toprecipitation of asphaltenes. Toofast a flow rate will erode thetube bends. If coil layout, burnerconfiguration and steam rate arenot correct, run length will bemeasured in months, not years.Diluent recovery unit designsmust take into account possible
For a discussion of factorsinvolved in designing refinery unitsto process difficult oil sands feed-stocks, ask for Technical Papers#234 and 238.
10 PTQ 01:10 01 PC PTQ 0107 ADF 10/19/07 4:42 PM Page 1
pcs 2.indd 1 1/6/12 21:05:28
Towards a zero gasoline production refinery: part 1
In some regions, the traditional export markets for gasoline- focused refineries are diminish-
ing. In this environment, investments in new refining projects (or in significant refinery upgrades) need to be optimised to fulfil market requirements. An opportunity exists to integrate the value chain from crude oil process-ing in refineries to the production of petrochemicals in line with market requirements.
The objective of this article is to present alternative refinery configu-rations that are able to process relatively heavy crudes, producing middle distillates, petrochemicals and aromatics, without producing any gasoline at all. The article will demonstrate how, by relying on well-proven refining process tech-nologies, configurations could be adapted in an existing FCC-based refinery or can be applied to a grassroots project. It will show how to integrate products from the steam cracker, aromatics complex and FCC unit while rationalising investments.
Part 1 of the article presents ways of upgrading streams produced in refineries that have traditionally been oriented towards the produc-tion of transportation fuels to produce petrochemicals, which may offer more attractive margins, allow diversification of product slates and reduce the impact of refined product market volatility. Streams from the FCC unit offer building blocks towards propylene and aromatics production. Part 2, to be published in a forthcoming issue of PTQ, will show the impact of these processes on the overall
Integrating products from the steam cracker, aromatics complex and FCC unit to produce petrochemicals without gasoline may offer more attractive margins
BLASIS STAMATERISFoster Wheeler
product slate of the refinery for different refinery configurations.
FCC unit: not just a gasoline-making machine In addition to gasoline, the refinery FCC process also produces one-third of the global propylene supply. Propylene yields from the FCC unit when operating in petro-chemical mode can be at least 10 wt% of the feed.
In order to maximise propylene production in the FCC unit, the following technical aspects need to be considered:1 • Feed quality: hydrogen content of the feedstock strongly correlates to the propylene yield• Increased riser outlet temperature (ROT) and high catalyst-to-oil ratio: increased severity yields higher conversion and a higher propylene yield • Catalyst system: use of ZSM-5 zeolite, which converts the C7+ olefins into light olefins with high catalyst Micro Activity Testing (MAT) activity• Hydrocarbon partial pressure in the reactor: shifts the reaction equi-librium to favour low molecular weight olefins. This is achieved though a low operating pressure and the addition of steam, and
should be balanced with increased plant costs due to larger vessel requirements.
Propylene is produced in the FCC unit by cracking of olefinic naphtha to lighter olefins. The cracking reac-tions that take place in the initial reaction step on the lower section of the riser with the feed and hot regenerated catalyst are endother-mic. The catalyst supplies the necessary heat to reaction tempera-ture. The riser is no more than a straight pipe, the diameter of which is set to provide the feed with a certain residence time.
Other converting reactions occur in a later step in the middle or upper section of the riser. In the sequential reactions, olefins that are initially produced from cracking reactions are consumed by subse-quent secondary reactions yielding iso-paraffins and/or aromatics. The reactions producing light olefins are controlled by an equilibrium mecha-nism and thermodynamics limit the propylene production from the FCC unit. Table 1 summarises the FCC unit’s operating conditions as compared with steam cracking.
As Table 1 shows, cracking in the FCC/DCC units occurs at moderate temperatures compared with steam cracking, which makes
www.eptq.com PTQ Q3 2013 11
Parameters Units FCC unit Deep catalytic cracking Steam crackingResidence time Seconds 1-3 10-16 0.1-0.2Catalyst/oil wt/wt 5-10 9-15 -Steam wt% of the feed 1-10 10-30 30-80Reaction temp °C 510-550 550-590 760-870Pressure KPa 15-30 10-20 15
FCC unit operating conditions as compared with steam cracking2
Table 1
fw.indd 1 07/06/2013 17:32
12 PTQ Q3 2013 www.eptq.com
synergy that can be exploited for aromatics production.3
Processing of light catalyticnaphtha (LCN)The direct cracking of C5-C6-type molecules contained in the LCN to form light olefins requires initial dehydrogenation to form olefins that can then be cracked through olefin reaction pathways and require high severity (temperatures of about 650°C) and a high catalyst-to-oil ratio. So, because of the selectivity of ZSM-5 zeolites to crack larger molecules, the cracking of the lighter molecules is regarded more as a thermal cracking process. Coke make when cracking LCN is low. Therefore, in order to achieve a significant increase in propylene production, large amounts of LCN need to be processed.
One option is to process the LCN into the same riser as the main feedstock. In this case, the LCN would be injected slightly below the main feedstock for it to detect a high catalyst-to-oil ratio. This option, although relatively low in cost, is not recommended, as the relatively large amounts of naphtha required to significantly increase the propylene yield (say by about 2%) would cool down the catalyst
the process very efficient from an energy consumption perspective. Coke that is deposited in the cata-lyst is burnt to regenerate the catalyst and provides the heat required by the cracking reaction. Also, since the feedstocks employed, such as gas oils and residues, are relatively cheap compared to steam cracking (tradi-tionally naphtha fed in Europe), the process can be economically attractive. In addition, the DCC unit requires very clean (desul-phurised, low metals content, low Conradson carbon) feedstocks with a high hydrogen content.
The increase in the ROT brings increased production of olefinic liquefied petroleum gas (LPG), dry gas and coke. Ethylene produced in the FCC unit (typical yield <2 wt% of the feed) could be recovered instead of using it in the fuel gas system.
The addition of ZSM-5 zeolite, with its characteristic pore size, which provides shape selectivity by limiting access to the interior of the catalyst to mostly linear non-branched paraffin and olefin molecules, gears the resulting equi-librium distribution of the C3 and lighter olefins towards propylene, the olefin product with higher yields.
Ethylene is also produced, but its yield is largely dependent on reac-tion conditions, as mentioned above. The reaction chemistry and the use of ZSM-5 catalyst favour conversion of the olefinic molecules in the C7-C10 range to olefinic LPG. This depletes the catalytic naphtha of olefins, which, along with the fractionation of the light catalytic naphtha (LCN) typically composed of C5-C6 molecules, contributes to the high aromatics content of the heavy catalytic naphtha — another
Products FCC, wt% Steamcracking,wt%Ethylene 1.5-4.5 25-30.8Propylene 12-17 17-14C
4s 14.4-17 About12
Naphthas 65.4-51.2 11.3-4.7
Yield comparison of naphtha cracking in FCC units versus steam cracking4,5
Table 2
Selective hydrogenation1
Total hydrogenation
MTBE decomposition
Butadiene extraction
Butene-1fractionation
Butene-2fractionation
Mixed C4s from refinery and petrochemicals
Raffinate-1
Mixed C4s
Raffinate-2
C4s processing into FCC.Direct recycle vs. oligomerisation.Simplicity vs. selectivity
Butene-1Co-monomer for polyethylene production
Steam cracker feed/LPG
Butadiene
Metathesis feed
Isobutylene
1 Butadiene saturation
Figure 1 C4sprocessingoptions
fw.indd 2 07/06/2013 17:32
To learn more about ISOMIX®-e and other CLG hydroprocessing innovations, visit www.chevronlummus.com
ISOMIX®-eIntroducing ISOMIX®-e.
Hydroprocessing with an edge.
enhanced reactor internals.
exclusively from CLG.
84
02
0-1
04
82
ISO
MIX
-e A
d v1
Title: ISOMIX®-e Advertiser: Chevron Lummus Global Client: Lori De Amaral, 1 510 242 3177, [email protected] Designer: John Lind, 1 510 242 2131, [email protected] Ad size: Full page (bleed), Trim: 210mm width x 297mm depth Publication: PTQ Q# 2011 Representative:
Live
TrimBleed
Live
TrimB
leed
Live
TrimBleed
Live
Trim
Ble
ed
clg.indd 1 26/02/2013 17:02
and the lower temperature would not allow complete vaporisation of the main feedstock, leading to unnecessary coke formation and deposition in the feed zone.
Another option is to use a sepa-rate riser, where the temperature and catalyst-to-oil ratio can be opti-mised and where the processing of the LCN does not interfere with the cracking of the main feedstock. In this case, the increase in propylene yield is still relatively small (about 2 wt% on top of the one achieved with the main feedstock), with low conversion per pass, a high gas yield and a great portion of the olefi ns produced being further converted due to secondary reac-tions into aromatics. Another option to dispose of the LCN is to send it along with the straight-run light naphtha to a steam cracker.
Table 2 compares the yields that can be obtained when cracking naphtha on a steam cracker versus in a separate riser on a FCC unit.
The trends are similar in both cases. As the severity increases, the
14 PTQ Q3 2013 www.eptq.com
ethylene yield increases; the propylene yield in the steam cracker decreases with increased severity, whereas it increases in the FCC unit. C4 yields remain about the same, but the naphtha yield decreases with increased severity, hence the once-through conversion of naphtha on the steam cracker is higher than on the FCC unit.
From a yields perspective, dispos-ing of the LCN through the steam cracker versus a separate riser in the FCC could yield better economic returns, but this needs to be further investigated for each specifi c case.
Processing C4s
There are several commercially proven ways to upgrade C4s produced from the FCC unit and steam crackers.6 Figure 1 illustrates some of the options.
Although the butadiene content of C4s from the steam cracker is signifi cantly higher than that of the FCC unit, the potential to recover butadiene is lost when the light
recovery section of the steam cracker and FCC unit are integrated, because of the dilution effect that the C4s from the FCC unit have on the combined C4s stream.
The main options for the produc-tion of on-purpose propylene are:• Direct recycle of C4s to a separate FCC riser • Via metathesis • Oligomerisation combined with recycle of the oligomerate to be processed in a separate riser of the FCC unit.
The direct recycle of the C4s cut involves a secondary high-severity riser parallel to the main riser designed to upgrade the C4s into ethylene, propylene and catalytic naphtha. Some fuel gas is also produced. The products from both risers merge at the reactor outlet and travel as a common stream to the main fractionator.
An advantage of this process is that selective hydrotreatment for the diene conversion of the C4s stream as well as the removal of
fw.indd 3 10/06/2013 12:33
www.eptq.com PTQ Q3 2013 15
The oligomerisation process can be applied to the C4s from the steam cracker and FCC units.
The reaction steps are: • Convert C4s olefins into C8-C12 oligomers in a polynaphtha unit• Recycle the oligomers into a separate riser in the FCC unit to selectively crack them into propyl-ene. The separate riser allows adjustment of the operating condi-tions to maximise propylene yield through a high reaction tempera-ture and high catalyst-to-oil ratio.
As an alternative, since the polynaphtha unit could be run to produce 100% gasoline-type mate-rial or 30% gasoline/70% distillates, this option offers the flexibility to increase the production of kero-sene-type material, which, after hydrotreating to saturate the olefins, would have a relatively low freezing point (-60°C) and high smoke point (>33 mm).
With this option, the increase in propylene yield would be 3-5 percentage points over the propyl-ene yield provided by cracking the heavy feed into the main riser, depending on the amount of oligomers recycled.
heavier naphtha-type products to a steam cracker.
In order to maximise propylene yields, the C4s need to be selec-tively hydrogenated to convert the butadiene into butenes and to isomerise the isobutylene and 1-butene into 2-butenes. The isobu-tylenes isomerisation reaction is equilibrium limited with a conver-sion of isobutylenes of about 62% and selectivity to n-butenes of about 90%.
To maximise the propylene yield, the remaining isobutylene should be removed from the metathesis feed either through fractionation or methyl tertiary butyl ether (MTBE) decomposition (if pure isobutylene is to be produced) because the isobutylenes react with the available n-butenes competing with ethylene in the production of propylene.
In the recycle of oligomerate, the idea behind the process is that oligomer cracking, with a catalyst that has ZSM-5 zeolite additive, is more selective, resulting in higher propylene yields when recycled to extinction than the recycle and cracking of C4s in a separate FCC riser.8
impurities is not required. Also, as with the cracking of the LCN, the aromatics content of the catalytic naphtha increases significantly compared with the catalytic naph-tha produced from the cracking of vacuum gas oil (VGO) and residue feedstocks, which helps increase the production of benzene, toluene and xylenes (BTX).7 From the investment perspective, it is a rela-tively low-cost option, but the cracking towards propylene is not selective.
In the metathesis case, which is essentially an equilibrium dispro-portionation reaction between two olefins, the n-butenes react with ethylene to produce mainly propyl-ene and other byproducts. The vapour phase reactions take place in a single fixed-bed reactor. The per-pass conversion is greater than 60%, with overall selectivity to propylene exceeding 90% when the feed is rich in 2-butene. This is a simple process to upgrade the value of n-butenes to high-value propylene. The paraffins pass through the system as inerts and, once recovered as non-reactive light materials, can be sent along with
Continuous catalytic reformerCCRHydrotreaterHTHeavy catalytic naphthaHCN
HT
CCR
FCC HT
Aromatics
Trans alkylation
Mid/heavy naphtha PyGas
HCN
Raffinate
Heavy reformate C8+
C7+
Light reformate C5–
Medium reformate C6/C7
Mixed aromatics
C8+
C9+
Raffinate
Reformate stabiliser
Reformate splitter
Benzene/toluene splitter
Toluene column
Xylene column
Xylene loop
C6–
C7–
C8+
Mixed xylene
Heavy aromatics
Benzene mixed xylene
Light ends
C9–C10–C11
Purify
Light ends
Naphtha/gasoline
Benzene
Toluene
Paraxylene
C8
Aromatics extraction
Figure 2 Configuration of the aromatics plant
fw.indd 4 07/06/2013 17:32
16 PTQ Q3 2013 www.eptq.com
Figure 2 shows the configuration of the aromatics plant.
The salient features of the configu-ration of the aromatics complex are: • Hydrotreat the straight-run naph-tha to remove sulphur and nitrogen compounds and send it to a CCR, where paraffins and naphthenes are converted to aromatics• Stabilise the reformate product and send it along with the pygas to a reformate splitter column. The C7 fraction from the overhead is sent to the extractive distillation unit for extraction of benzene and toluene. The C8+ fraction from the bottom of the reformate splitter is clay-treated and then sent directly to the xylene recovery section of the complex• Hydrotreat the heavy catalytic naphtha to remove impurities and send to a separate aromatics extrac-tion unit. The extract is combined with extract from the reformate and sent to the BTX fractionation section to recover high-purity benzene and toluene products• Send the raffinate from the heavy catalytic naphtha aromatics extrac-tion unit to the straight-run naphtha hydrotreater to remove traces of extraction solvent and convert the naphthenes into aromatics in the CCR, whereas the raffinate from the reformate aromatics extraction unit is used as feedstock for the steam cracker plant • The rest is a conventional aromat-ics plant, which includes a C8 aromatics isomerisation unit, parax-ylene purification unit, a heavy aromatics column and transalkyla-tion unit, where toluene is blended with C9 and C10 aromatics (A9-) from the overhead of the heavy aromatics column and processed for the production of additional xylenes and benzene, which are recovered in the BTX fractionation section.
The incorporation of a transalkyl-ation unit into the aromatics complex allows the C9 aromatics of the heavier straight-run/catalytic naphtha with an end point of above 170°C to be converted into addi-tional xylenes.
Overall, the steam cracker effi-ciently converts light paraffins and rejects aromatics and unconverted naphthenes in pygas, whereas an aromatics unit efficiently converts
The incremental capital invest-ment required by the polynaphtha unit is moderate and its economic justification competes with the direct recycle of the C4s to a separate riser. However, the process is selective towards the production of propyl-ene mainly because of the operating conditions and catalyst additive (ZSM-5) used to crack the oligomers.
Configuration of the aromatics plantThe configuration of the aromatics complex, which is used to convert naphtha and pyrolysis gasoline (pygas) into BTX, includes the following process technologies:• Continuous catalytic reformer (CCR) for the production of aromatics from naphtha at high severity• Extractive distillation for the recovery of benzene and toluene• Paraxylene purification for the recovery of paraxylene by continu-ous adsorptive separation from a mixed xylenes stream• C8 aromatics isomerisation for the isomerisation of xylenes and the conversion of ethylbenzene• Transalkylation for the conver-sion of toluene and heavy aromatics to xylenes and benzene.
In order to exploit potential synergies with the refinery and steam cracker, the heavy FCC naphtha, which is highly aromatic and naphthenic, as well as the pygas, are attractive feedstocks for aromatics production. Coupling aromatics production with a high olefin yield FCC maximises the value added from the FCC unit; with a aromatics content of >55 wt%, FCC gasoline becomes a desirable aromatics feedstock.
On the other hand, pygas compo-sition varies widely with the type of feedstock being cracked in an ethylene plant. Light steam cracker feedstocks tend to produce a pygas that is rich in benzene but contains almost no C8 aromatics. Substantial amounts of C8 aromatics are found only in pygas from ethylene plants cracking naphtha feedstocks. All pygas contains significant amounts of sulphur, nitrogen and dienes that must be removed by two-stage hydrotreating before being processed in an aromatics complex.
naphthenes and efficiently recovers aromatics, but rejects light paraffins in fuel gas, light ends and raffinate streams, allowing exploitation of the composition of the streams to maximise the production of valua-ble products while minimising the investment cost.
Integration of refining and petrochemicalsThe drivers for integration between refining and petrochemical facilities have been extensively discussed elsewhere.9 While olefins require further integration into polyolefins or other olefin derivatives, since they are not readily transportable, the aromatics are directly marketable. The key is to have refining units, steam crackers and aromatics complexes on the same site.
From the revenue perspective, having integrated refining and petrochemicals complexes mini-mises the impact of the volatility of crude oil price, the primary driver of petrochemical costs and prices.10
From the pricing perspective, it is observed that: • Price differentials between naphtha and gasoline are relatively narrow, compared with naphtha and aromatics and naphtha and ethylene or propylene• The spread between these price differentials has been increasing over time.
Many new refinery projects, including some of those under development in China and the Middle East, include integrated refinery and petrochemicals complexes, addressing growth needs in the Asia Pacific, Middle East and European markets.
The other aspect that needs to be considered for refinery/petrochem-ical integration to be successful is the yields versus capital required, since it affects the revenues and the profitability of the facility and, in the case of the steam cracker, the yields are heavily influenced by feedstock selection.
Figure 3 shows the yields for steam cracker feeds versus FCC and CCR for petrochemicals. It shows that naphtha cracking gives a far wider range of products than gas-based steam cracking. From the
fw.indd 5 07/06/2013 17:32
How Can KBR Answer Your Refining Challenges?
Owners of refineries continue to confront
many challenges – rising feedstock prices,
shrinking margins, varying global demands
and a changing regulatory landscape that
includes ever-more stringent specifications
on sulfur and carbon footprints. As refinery
owners debottleneck and enhance existing
facilities, they call on KBR to deliver.
To learn how KBR can address your refining
challenges, go to:
refining.KBR.com/PTQ
refining.kbr.com / PTQ
K13009 © 2013 KBR, Inc. All Rights Reserved.
kbr.indd 1 08/03/2013 10:45
18 PTQ Q3 2013 www.eptq.com
cracker can be processed in the aromatics complex to recover BTX components.
So, liquid-based steam crackers can be economically competitive because of the credits obtained with the byproduct slate that the naphtha-based cracker produces, which maximise the revenues.
Another aspect of integration that leads to lower investments but needs careful consideration during design is the use of combined light ends recovery facilities between the steam cracker and the FCC unit.
In terms of investment, there are different levels of integration. For instance, the minimum would be to send off-gas from the FCC unit to
capital investment perspective, naphtha cracking is also far more complex. However, it makes sense when integrated with refineries that produce aromatics, because of the synergies that can be exploited with other facilities within the refinery, such as:• Hydrogen produced in the steam cracker in the refinery, the CCR and in steam reforming can be used to supply the requirements of the hydrotreating units• The raffinate stream produced in the aromatics complex is a perfect steam cracker feedstock: light mate-rial, low in aromatics, high in paraffins• The pygas produced in the steam
Steam cracker
Petrochemical FCC
Unsaturated gas plant
Refinery
Ethane
Crude
Wet gas
Light ends
FCC gasoline
Fuel oil
Diesel
Kerosene
Gasoline
Ethylene
C3+
VGO C5+
MIxed C4s to C4 complex
Propylene
Hydrogen fuel shared utilities
Figure 4 Integration opportunities to maximise olefins production
0 10 20 30 40 50 60 70 80 90
BTX
Butenes
Propylene
Ethylene
AGO (atmospheric gas oil)
CCR (continuous catalytic reformer)
n-C4 (normal butane)
FCC (fluid catalytic cracker)
Light naphthai-C4 (ISO butane)
n-C3 (propane)n-C2 (ethane)
Figure 3 Yields for steam cracker feeds vs FCC and CCR for petrochemicals
Heavy crude Oil
Atmospheric Distillation
Vacuum Distillation
Coker & Visbreaker Feed
Fluidized Catalytic Cracker
Bitumen
Non-intrusive fl ow measurement
up to 400°C
Field-Proven at Refi neries
Trouble free operation at extreme pipe temperatures
No clogging, no abrasion, no pressure losses
Installation and maintenance without process interruption
Independent of fl uid or pressure
Hazardous area approved
Non-intrusive fl ow
refi nery.fl exim.com
fw.indd 6 07/06/2013 17:33
www.eptq.com PTQ Q3 2013 19
9 Allen A, Refinery/petrochemical integration: past, present and look into the future, Offshore World, 29 Dec 2007-Jan 2008, 29-34.10 Scott A, et al, Using microeconomics to guide investments in petrochemicals, McKinsey on Chemicals, No 4, Spring 2012, 47.11 Dharia D, et al, Catalytic cracking for integration of refinery and steam cracker, Advances in Fluid Catalytic Cracking, CRC Press, 2010, 119-126.
Blasis Stamateris is Downstream Business Consultant in the Business Solutions Group of Foster Wheeler, Reading, UK. He has over 25 years’ experience in the oil refining and upgrading business, and holds a degree in chemical engineering.
FCC naphtha in a secondary riser of the FCC unit for maximum propylene production, Fuel Processing Technology, 2008 (89), 864-873.5 Ethylene, Chemsystems PERP program, PERP 08/09-5, Sept 2009.6 Kantorowicz S, C
4 processing options to
upgrade steam cracker and FCC streams, 2nd Asian Petrochemicals Technology Conference, Korea, 7-8 May 2002.7 Niccum P K, et al, KBR catalytic olefins technologies provide refinery/petrochemical balance, 25th JPI Petroleum Refining Conference, Recent Progress in Petroleum Process Technology, 26-27 Oct 2010, Tokyo. 8 Dupraz C, (R)FCC product flexibility with FlexEne, ARTC 2012, Bangkok.
the steam cracker to recover ethylene or use a common propyl-ene-propane splitter for full integration with only the front end of the FCC unit (consisting of reac-tor, regenerator and main fractionation) with overhead sent to the wet gas compressor, whereas the olefins unit water quench tower goes to the cracker gas compressor sharing the product recovery systems.11
Figure 4 shows examples of streams that are produced in refiner-ies that can be used on steam crackers such as LPG and/or the use of common facilities for the recovery of ethylene and propylene.
In summary, refinery and petro-chemical integration: • Allows the upgrade of low-value streams to high-value products• Minimises the cost of petrochemi-cal feedstocks, since they are readily available from the refinery• Provides stability over the value creation chain by diversifying the product slate, which dampens cyclic profitability impact• Reduces hydrogen production in steam reformers by recovering the hydrogen produced by the steam cracker and catalytic reformer• Optimises capital, operating costs and resources through shared infra-structure for utilities supply, off-sites (tankage allows transfer of refinery intermediate products to petrochemicals, common flare, wastewater treating facilities) and infrastructure (buildings, labo-ratory), leading to lower investments.
The second part of this article will demonstrate that the full inte-grated scheme leads to significant savings in investment and operat-ing costs, but has a lot of design challenges to guarantee the opera-bility of all process units.
References1 Knight J, Mehlberg R, Maximise propylene from your FCC unit, Hydrocarbon Processing, Sep 2011, 91-95.2 Kayode Coker A, Modelling of Chemical Kinetics and Reactor Design, Gulf Professional Publishing, 237.3 Bedell M, Ruziska P A, Steffen T R, On purpose propylene from olefinic streams, Tulane Engineering Forum, Sept 2003, 3-4.4 Wang G, C Xu, Jinsen G, Study of cracking
www.eptq.com PTQ Q2 2013 35
MBA from the University of Houston. He is a registered professional engineer in the state of Texas, and is an inventor or co-inventor on eight US patents and has published nine technical papers.
Robert Žajdlík is a Slovnaft Bratislava Refinery Technologist with the MOL Group. During the past two years, he has been a technologist for the refinery’s LC Fining unit and FCC pretreater unit. He is a graduate of the Slovak Technical University, Bratislava, Faculty of Chemical Technology, and holds a PhD in chemical engineering and process control. Email: [email protected]
she holds a master’s in chemistry from the University of Novi Sad, Serbia, and a PhD in chemistry from the University of Vienna. Email: [email protected]
Bruce Wright is a Baker Hughes Senior Technical Support Engineer in the Industrial Technology Department in Sugar Land, Texas, specialising in the hydrocarbon process industries. He has more than 30 years’ industry experience and is currently involved in technical support and troubleshooting of refinery fouling problems. He holds a BS in chemical engineering from Rensselaer Polytechnic Institute, Troy, New York, and an
control values to drive the process close to the fouling threshold where rapid exponential fouling starts, thus maximising severity/conver-sion with respect to an acceptable and controllable rate of fouling. The approach can be used to determine the best economical operating window, even with changes in feedstock and operations, because the demonstrated methods can rapidly detect the impact of fouling.
LC Fining is a mark of Chevron Lummus Global.This article is based on a paper presented at the International Bottom of the Barrel Technology Conference & Exhibition, Rome, May 2012.
References1 Putek S, Gragnani A, First resid hydrocracker to produce stable, low-sulphur diesel fuel from ural vacuum residue, ERTC 10th Annual Meeting, Vienna, 2005.2 Sherwood D E Jr, Barriers to high conversion operations in an ebullated-bed unit relationship between sedimentation and operability, NCUT Workshop, Edmonton, 2000.3 McNamara D J, Sherwood D E Jr, Bhan Opinder K, Getting more out of your resid upgrading unit, 6th BBTC, Barcelona, 2008.4 Bartholdy J, Andersen S I, Changes in asphaltene stability during hydrotreating, Energy and Fuels, 2000.
Marco Respini is a Baker Hughes Technology Development Specialist in the Industrial Technology Group in Europe, specialising in refinery process fouling control. He has 14 years’ refining experience and is currently involved in developing new technologies for monitoring fouling and severity control in resid conversion processes. He holds a degree in industrial chemistry from Milan University and is a registered professional chemist in Italy. An inventor of two US patents, he has published five technical papers and four conference papers on visbreakers and heavy fuel oil stability problems. Email: [email protected]
Silvia Ekres is a Baker Hughes Account Manager responsible for Downstream Chemicals, Industrial Portfolio product line sales, and specialises in refinery and petrochemical applications, including mitigating fouling-related issues in LC Fining units. The author of several publications,
Contact us at +1.801.984.1000 or visit http://deltavalve.cwfc.com
Our innovative coke drum unheading valves have been in service for over 12 years, and are installed in more refineries worldwide than any other unheading valve. Introduced in 2001, our coke drum unheading valves have completed over one million combined cycles and have a safety and reliability record unmatched by any traditional coke drum unheading system worldwide.
Our recently upgraded coke drum bottom unheading valve is now 15% smaller, Our recently upgraded coke drum bottom unheading valve is now 15% smaller, 12% lighter, has minimal steam consumption, no cooling water, and can be equiped with our new optional electric actuator featuring state-of-the-art planetary roller screw technology.
DeltaValve’s coke drum unheading valves;most proven, most preferred, worldwide.
Most proven. Most preferred.The global standard in coke drum unheading valves.
baker hughes.indd 10 08/03/2013 12:12fw.indd 7 07/06/2013 17:33
relentless
UOP has been serving catalyst customers for
decades, and constant innovation ensures we’ll
be there for decades to come.
Pilot plants. A true commitment to fast, thorough research and
development. A vast team of experienced, dedicated engineers
and scientists. The reasons to choose UOP catalysts are many,
but perhaps the biggest of all is that UOP never quits innovating. When you use UOP
catalysts, you’ll know you’re using products specifically designed to help optimize
your operation. UOP invented the refining and petrochemical technology used in most
operations today, and that knowledge and expertise comes through in every catalyst we
develop. You’ll always get the high stability and global support your business needs, and
you’ll be working with a company that will keep you on the leading edge year after year.
For more information about UOP catalysts, visit www.uop.com/catalysts.© 2013 UOP LLC. All rights reserved
SPM-UOP-35-Catalysts-Ad-A4_UOP Ad 297mm x 210mm A4 size UOP6098 6/6/13 10:55 AM Page 1
uop.indd 1 07/06/2013 09:58
Refinery power failures: causes, costs and solutions
About 90 seconds into the second half of the 2013 Super Bowl, the American football
championship game, half of the stadium lost power — a 34-minute delay caused by a malfunction in a faulty electrical relay that was supposed to monitor the electrical load. This event not only caused much frustration between the teams, fans and workers trying to fix the problem but also cost the broadcast network advertising revenue. The point is that we all take electricity for granted until the lights go out. For most of us, we just endure until the power comes back on. For refineries, however, there are more serious conse-quences that occur when power is lost. This article is the first of a two-part series examining tech-nology-driven strategies to mitigate refinery power failures and mini-mise impacts on company earnings. Part 1 discusses the causes and financial costs when a refinery suffers a power disruption. Part 2 focuses on mitigation technologies available to refiners for minimising power disruptions and unplanned shutdowns as well as overall strate-gies to reduce financial impacts.
Refinery shutdowns happen on a daily basisHydrocarbon Publishing Company collected data from reports published by the US Department of Energy’s Energy Assurance Daily. The information analysed focused on power failures and disruptions at US refineries.
From 2009 to 2012, there were over 1700 refinery shutdowns, which equates to an average of 1.2
Power outages typically lead to damaging costs for refiners. Strategies are needed to minimise them
PATRICK J CHRISTENSEN, WILLIAM H GRAF and THOMAS W YEUNG Hydrocarbon Publishing Company
shutdowns per day. As Figure 1 shows, 46% of the shutdowns were due to mechanical breakdowns, 19% were caused by electrical disruptions and power failures, 23% were the result of mainte-nance, and the last 12% were because of other causes, mostly fires that occurred at the refinery. While some incidences last only a couple of hours, many last multiple days and even weeks. About 92% of maintenance-related shutdowns
were unplanned, many due to leaks in piping and different units.
Major causes of power disruptionsAs Figure 3 shows, 17.6% of refin-ery power disruptions were the result of electrical equipment fail-ures or refinery processing units having electrical problems. This includes transformers malfunction-ing or the FCC unit suffering an electrical failure. Some 16.4% of causes were due to weather events. This includes hurricanes, lightning strikes and wind causing power lines to fall or units to be knocked out of service. While accidents due to weather cannot be avoided, proper maintenance of processing units and electrical equipment can help avoid breakdowns.
Unfortunately, over 60% of the causes of electrical disruptions are not specified. Most of the unspeci-fied power failures were listed as “a power failure occurred at the refinery and caused flaring” or similar statements. There are multiple reasons why the reports were so vague: the refiner did not know the reason at the time of the report, there are confidentiality policies that prevented the refinery from reporting the causes, or the refinery for whatever reason did not want to be very specific in its report. The fact that so many are unspecified makes it difficult to minimise power disruptions. Therefore, any prevention and protection strategies deployed by a refinery should include detailed bookkeeping of failed equipment so that statistical analyses or predictive analytics can be performed in identifying exact
www.eptq.com PTQ Q3 2013 21
Electrical: 19%
Mechanical: 46%
Other: 12%
Maintenance: 23%
Figure 1 Overall shutdowns 2009-2012, %
Weather: 16.4%
Unspecified: 60.4%
Fire: 1.9%
Other: 1.9%
Breakdown: 17.6%
Surges: 1.9%
Figure 3 Causes of power disruptions 2009-2012, %
hpc.indd 1 11/06/2013 12:17
22 PTQ Q3 2013 www.eptq.com
are the most commonly cited causes of hurricane damage, for Gulf Coast refineries in the US, where hurricanes hit most often and hardest, the greatest hurricane damage is the result of flooding. Hurricane Isaac, which touched down in Louisiana 28-29 August 2012, caused extensive flooding, forcing nine Louisiana oil refineries with a collective capacity of 2.2 million b/d to shutter about 42% of that capacity by 29 August due to power outages. Backup generators are typically only used for refiner-ies’ most important units. Flood control measures commonly employed at refineries and petro-chemical plants on the Gulf Coast include dikes and levees to guard against rising waters, but each storm is different. Normal shut-downs usually take three days, but hurricanes are often unpredictable, making it difficult to schedule a proper shutdown procedure. Restarting after a hurricane is determined by electrical outages on and off-site, so it could take days or weeks. As a result of Superstorm Sandy in late October 2012, Phillips 66’s 238 000 b/d Bayway refinery in the state of New Jersey was idle for over three weeks due to electri-cal damage sustained from salt water flooding.
Electrical equipment failuresElectrical breakdowns are nothing new to any industry. Many units require simple maintenance checks that can help avoid sudden break-downs. Fires can easily occur in oil or gas-insulated units such as trans-formers. In January 2009, ExxonMobil’s Beaumont, Texas, coker was shut for unplanned elec-trical repairs. In July 2010, Citgo’s Lake Charles, Los Angeles, refinery had a fire break out due to faulty electrical wiring, which led to an emissions release. In February 2012, PBF Energy’s Delaware City, Delaware, plant had an electrical issue with the CDU. Two compres-sors in the catalytic cracker were inoperable due to an electrical issue, which led to flaring. These electrical problems can be easily seen and fixed during routine maintenance before the failures occur.
causes to reduce future incidents, as discussed in part 2 of this article.
In Figure 4, 56.1% of the failing units were electrical equipment such as circuit breakers, switch-gears, transformers and substations. Some 24.6% of power disruptions were caused by refinery processing units such as the FCC unit having electrical issues. About 7% of prob-lems were the result of rotary equipment such as motors and compressors having electrical prob-lems. About 12% did not specify the unit or equipment that was having problems.
From 2009 through 2012, there were about 320 power disruptions at refineries in the US. In this section, some of their experiences are reviewed. All of these disrup-tion events were gathered from the US Department of Energy’s Energy Assurance Daily publication. We chose events that provided a clear picture of the causes and impacts from power disruptions.
Weather problemsThe most common weather event that causes power failures is a thunderstorm. High winds, light-ning strikes, heavy rain and flooding can all cause problems. Power lines can be knocked down, lightning strikes can disrupt units, and rain can interfere with the steam supply. These are a few of the examples that show how weather can cause a refinery shutdown.
In February 2010, Valero’s Ardmore, Oklahoma, refinery expe-rienced thunderstorms that caused temporary power outages at several units, reducing runs. Valero reported that it was restoring full production at the plant the next day. On 1 December 2011, a wind storm caused a power outage and forced several units offline at Chevron’s El Segundo, California, refinery, halting production until 7 December. Among the affected units was a CDU that caught on fire in the aftermath of the electrical failure. Flaring occurred as a result of the incident and during the subsequent restart process. Fire damage to the CDU, six days of
inactivity and flaring all adversely affected the refinery’s bottom line.
In March 2009, Motiva’s Port Arthur, Texas, facility experienced lightning that led to a power outage and also stated a fire. The lightning caused a crude unit, two hydrotreaters and a delayed coker to shut down. Unlike Motiva’s experience, in July 2009, Pasadena Refining’s Pasadena, Texas, complex experienced a lightning strike that disabled all power to the
Red Bluff Tank farm, which resulted in loss of feed to the refin-ery’s crude unit. The feed was restored with a backup generator, and the refinery was able to run using backup power and keep operations online despite the loss of power to the tank farm. In April 2012, Valero’s Norco, Los Angeles, facility experienced a power surge caused by lightning, which caused the hydrocracking unit to trip. Emissions of sulphur dioxide and hydrogen sulphide were released during the flaring caused by the shutdown.
Although high wind and debris
Unspecified equipment:
12.3%
Electrical equipment: 56.1%
Refinery processing units: 24.6%
Rotary equipment: 7.0%
Figure 4 Breakdowns causing power disruptions 2009-2012
When a power failure occurs, a refinery unit or units or an entire facility must be shut down and production is lost
www.eptq.com PTQ Q2 2013 41
the ebullated-bed reactor or fed to a coker unit. For this study, it was considered that this stream is recy-cled back to the hydroconversion reactor, meaning that there is complete overall residue conver-sion even when the once-through conversion reported in Table 2 is 78%.
The fractionated products from the VTB conversion step, along with the corresponding VDU distil-lates, require further hydrotreating to reduce sulphur, nitrogen and aromatic contents for producing suitable blending components for SCO. The properties of the hydro-treated products and final SCO are provided in Table 4. The SCO has no residue and very low sulphur and nitrogen contents. The coking-derived SCO is slightly more aromatic than that derived from the hydroconversion-based scheme. Table 5 summarises the details of the HDT units. It is observed that the hydrotreating of coker products requires higher
Coker HydroconverterProperty Naphtha LGO HGO Naphtha LGO HGO Vacuum residueYield, wt% 26.8 34.1 39.1 14.0 37.9 27.8 20.4SG 60/60°F 0.7363 0.8715 0.9736 0.7365 0.8710 0.9715 1.1240 API gravity 60.7 30.9 13.8 60.6 31.0 14.2 -5.6Sulphur, wt% 1.79 3.67 4.43 0.25 0.71 1.37 6.90Nitrogen, wppm 316 1694 3973 393 1582 3063 8905Aromatics, wt% 27.1 58.0 66.0 20.5 42.5 53.0 99.8Nickel, wppm - - 8.2 - - 3.0 262.9Vanadium, wppm - - 27.4 - - 6.2 458.3CCR, wt% - - 1.5 - - 0.2 14.3
a) Yield values are based on total liquid product.
Properties of coker and hydroconverter liquid productsa
Table 3
Coker-based scheme Hydroconverter-based schemeProperty Naphtha LGO HGO SCO Naphtha LGO HGO SCOSG 60/60°F 0.7375 0.8597 0.9131 0.8680 0.7375 0.8573 0.9115 0.8665API gravity 60.4 33.1 23.5 31.5 60.4 33.6 23.7 31.8Sulphur, wt% 0.018 0.112 0.16 0.13 0.014 0.044 0.15 0.10Nitrogen, wppm 12 204 740 482 21 264 455 336Aromatics, wt% 19.9 41.8 54.2 45.9 15.8 36.5 50.9 41.7
Properties of hydrotreated products and SCO
Table 4
test run’s results, it could be concluded that the revamp targets for the CDU-1 main fractionator (C-150) were achieved. No hydrau-lic constraint was experienced in achieving the design intake of 13 000 t/d and the required prod-uct quality was achieved.
ConclusionsThe performance of Shell ConSep trays in the HGO pumparound section of the CDU-1 main fractiona-tor met the target of capacity enhancement without any drawback compared to the pre-revamp condi-tions. During the test run, the trays were operating at 10-15% lower than the design capacity even at the design intake of 13 000 t/d due to heavier crude feed and lower feed temperature. However, the built-in capacity margin enabled stable oper-ation for the trays at much above the capacity limit of the first genera-tion of high-capacity trays.
The options to debottleneck columns already equipped with the first generation of high-capacity trays are limited. ConSep trays provide an attractive solution for
www.eptq.com PTQ Q1 2013 77
such cases. In this revamp project, use of only three of these trays in the most capacity-constrained section of the column made it possi-ble to retrofit the existing column and made the capex option more attractive over the other debottle-necking options.
* Shell ConSep, Shell CS and Shell HiFi are Shell trademarks. ** Mellapak Plus 252Y is a Sulzer Chemtech trademark.
References1 Refinery expansion means NZ more self reliant, media release by NZRC, 16 July 2010.2 Wilkinson P M, De Villiers W E, Mosca G, Tonon L, Achieve challenging targets in propylene yield using ultra system fractionation trays, ERTC 2006.
3 De Villiers W E, Bravo J L, Wilkinson P M, Summers D R, Further advances in light hydrocarbon fractionation, PTQ Q3 2004.
KaushikMajumder is Distillation Team Lead of Shell Projects & Technology in Bangalore, India. He holds a bachelor’s degree from Jadavpur University, India, and a master’s and doctorate from Indian Institute of Technology, Delhi. Email: [email protected] Mosca is the Global Refinery Technology Manager of Sulzer Chemtech. He holds BS and MS degrees in chemical engineering from the University “La Sapienza” Rome, Italy. Email: [email protected] is a Process Engineer at Refining NZ. He was the Senior Process Engineer and Commissioning Process Engineer during the Point Forward Project. Email: [email protected]
Parameters Design TestrunFroth backup/CS height, % 68 60Tray pressure drop, mbar 12.3 9.2Tube flood , % 73 60Flow parameter 0.17 0.19Overall column load factor, m/s 0.12 0.10Flooding (CS tray), % 133 112
KeyperformanceindicatorsforConSeptrays
Table3
sulzer.indd 5 11/12/12 18:05:58
canmet.indd 4 08/03/2013 13:04hpc.indd 2 11/06/2013 12:18
Substations, whether owned by the utility company or the refinery, tend to break down if not properly maintained. In October 2009, Valero’s McKee, Texas, plant had a substation malfunction, which forced units offline. Substation malfunctions have occurred more frequently than many people think. In March 2010, Chevron’s El Segundo, California, refinery shut down due to fire at a substation. And, in August 2011, Tesoro’s Kapolei, Hawaii, complex had a power outage due to a failure at one of the local electric utility’s substations.
In May 2010, BP’s Texas City, Texas, refinery had a power blip caused by switchgear failure. Switchgear failures also occur quite frequently. The relay that failed at the Super Bowl was a part of the switchgear setup that supplied power to the stadium. Switchgears operate as a protective device against overcurrent and arc flashes. They also need to be properly maintained to avoid failures. A few
www.eptq.com PTQ Q3 2013 23www.eptq.com PTQ Q2 2013 41
the ebullated-bed reactor or fed to a coker unit. For this study, it was considered that this stream is recy-cled back to the hydroconversion reactor, meaning that there is complete overall residue conver-sion even when the once-through conversion reported in Table 2 is 78%.
The fractionated products from the VTB conversion step, along with the corresponding VDU distil-lates, require further hydrotreating to reduce sulphur, nitrogen and aromatic contents for producing suitable blending components for SCO. The properties of the hydro-treated products and final SCO are provided in Table 4. The SCO has no residue and very low sulphur and nitrogen contents. The coking-derived SCO is slightly more aromatic than that derived from the hydroconversion-based scheme. Table 5 summarises the details of the HDT units. It is observed that the hydrotreating of coker products requires higher
Coker HydroconverterProperty Naphtha LGO HGO Naphtha LGO HGO Vacuum residueYield, wt% 26.8 34.1 39.1 14.0 37.9 27.8 20.4SG 60/60°F 0.7363 0.8715 0.9736 0.7365 0.8710 0.9715 1.1240 API gravity 60.7 30.9 13.8 60.6 31.0 14.2 -5.6Sulphur, wt% 1.79 3.67 4.43 0.25 0.71 1.37 6.90Nitrogen, wppm 316 1694 3973 393 1582 3063 8905Aromatics, wt% 27.1 58.0 66.0 20.5 42.5 53.0 99.8Nickel, wppm - - 8.2 - - 3.0 262.9Vanadium, wppm - - 27.4 - - 6.2 458.3CCR, wt% - - 1.5 - - 0.2 14.3
a) Yield values are based on total liquid product.
Properties of coker and hydroconverter liquid productsa
Table 3
Coker-based scheme Hydroconverter-based schemeProperty Naphtha LGO HGO SCO Naphtha LGO HGO SCOSG 60/60°F 0.7375 0.8597 0.9131 0.8680 0.7375 0.8573 0.9115 0.8665API gravity 60.4 33.1 23.5 31.5 60.4 33.6 23.7 31.8Sulphur, wt% 0.018 0.112 0.16 0.13 0.014 0.044 0.15 0.10Nitrogen, wppm 12 204 740 482 21 264 455 336Aromatics, wt% 19.9 41.8 54.2 45.9 15.8 36.5 50.9 41.7
Properties of hydrotreated products and SCO
Table 4
test run’s results, it could be concluded that the revamp targets for the CDU-1 main fractionator (C-150) were achieved. No hydrau-lic constraint was experienced in achieving the design intake of 13 000 t/d and the required prod-uct quality was achieved.
ConclusionsThe performance of Shell ConSep trays in the HGO pumparound section of the CDU-1 main fractiona-tor met the target of capacity enhancement without any drawback compared to the pre-revamp condi-tions. During the test run, the trays were operating at 10-15% lower than the design capacity even at the design intake of 13 000 t/d due to heavier crude feed and lower feed temperature. However, the built-in capacity margin enabled stable oper-ation for the trays at much above the capacity limit of the first genera-tion of high-capacity trays.
The options to debottleneck columns already equipped with the first generation of high-capacity trays are limited. ConSep trays provide an attractive solution for
www.eptq.com PTQ Q1 2013 77
such cases. In this revamp project, use of only three of these trays in the most capacity-constrained section of the column made it possi-ble to retrofit the existing column and made the capex option more attractive over the other debottle-necking options.
* Shell ConSep, Shell CS and Shell HiFi are Shell trademarks. ** Mellapak Plus 252Y is a Sulzer Chemtech trademark.
References1 Refinery expansion means NZ more self reliant, media release by NZRC, 16 July 2010.2 Wilkinson P M, De Villiers W E, Mosca G, Tonon L, Achieve challenging targets in propylene yield using ultra system fractionation trays, ERTC 2006.
3 De Villiers W E, Bravo J L, Wilkinson P M, Summers D R, Further advances in light hydrocarbon fractionation, PTQ Q3 2004.
KaushikMajumder is Distillation Team Lead of Shell Projects & Technology in Bangalore, India. He holds a bachelor’s degree from Jadavpur University, India, and a master’s and doctorate from Indian Institute of Technology, Delhi. Email: [email protected] Mosca is the Global Refinery Technology Manager of Sulzer Chemtech. He holds BS and MS degrees in chemical engineering from the University “La Sapienza” Rome, Italy. Email: [email protected] is a Process Engineer at Refining NZ. He was the Senior Process Engineer and Commissioning Process Engineer during the Point Forward Project. Email: [email protected]
Parameters Design TestrunFroth backup/CS height, % 68 60Tray pressure drop, mbar 12.3 9.2Tube flood , % 73 60Flow parameter 0.17 0.19Overall column load factor, m/s 0.12 0.10Flooding (CS tray), % 133 112
KeyperformanceindicatorsforConSeptrays
Table3
sulzer.indd 5 11/12/12 18:05:58
canmet.indd 4 08/03/2013 13:04
types of switchgear are oil or gas insulated. A stray spark can cause these to catch on fire and fail.
Then there are the obscure reasons that cannot be predicted and happen so infrequently that it is hard to protect against. In November 2012, Valero’s Corpus Christi, Texas, had a power loss that led to flaring due to a rodent contacting the primary power transformer. Transformers that can be accessed should be protected to allow only those working on them to get to them. In February 2010, Western Refining’s Yorktown, Virginia, had an unidentified unit shutdown due a temporary power blip caused by a goose flying into a nearby power line. Unless power lines are put underground, it is difficult to protect them from animals and debris.
Power surges and fluctuationsPower surges also occur frequently and can be prevented with circuit breakers or switchgear. In August 2009, ExxonMobil’s Baytown,
Texas, complex had a power surge that triggered a small fire in a pipe rack at the refinery’s chemical plant. Power surges can cause fires and cause breakdown in different units. In April 2011, Sunoco’s Philadelphia, Pennsylvania, refin-ery had a power surge that knocked compressors offline. In May 2011, Phillips 66’s Wilmington, California, facility had a power fluctuation occur, temporarily shut-ting down several refinery units. Normal operations were restored the next day.
Multiple unit shutdownsSince most refinery units are inte-grated and sometimes share the same power supply, power failures could lead to the shutdown of these integrated units and the production loss of many refined products, thereby magnifying potential damages.
In June 2009, Tesoro’s Kenai, Alaska, facility experienced a power outage. The hydrocracker and isomerisation unit were shut to
hpc.indd 3 07/06/2013 17:53
24 PTQ Q3 2013 www.eptq.com
specific. Nowadays, the loss is even higher, as the FCC unit is consid-ered an important source for polymer-grade propylene and light cycle oil (LCO) for the production of middle distillates.2 Moreover, steam and power from the regener-ator turbo-expander contribute part of the plant utilities.3
Case studyPhillips 66’s Bayway, New Jersey, refinery was down from 28 October to 20 November 2012 due to Superstorm Sandy. According to the company, the expenses related to the storm were $56 million before tax.4 This loss did not include missed production for over three weeks. Based on the refinery’s nameplate gasoline production capacity of 145 000 b/d and distil-late production capacity of 115 000 b/d5, an utilisation rate of 85% and average spot market prices of gaso-line of $2.812/gal and $3.084/gal for distillates in New York Harbor during the shutdown period, esti-mated revenue loss was over $650 million. Net profit loss ranged from $5.3 million for cash margins of $1/bbl to $26.5 million for cash margins of $5/bbl. US refineries outside the East Coast should expect a bigger financial impact, since East Coast refiners are known to have much lower refining margins than their peers in other parts of the country.
From a business point of view, missed shipments to retail outlets at a time of strong demand can lead to a price surge at the pump, resulting in public outcry and possible governmental investiga-tions.6 Undoubtedly, both environmental and retail price issues will generate negative media coverage and severely damage a company’s public relations.
A rapid shutdown also increases the danger of mechanical damage. Subsequent repairs to damaged mechanical parts as well as costs of downtime can make power failures even more costly than the earlier estimates. Furthermore, unit shut-downs and restarts are known to reduce energy efficiency and increase the carbon footprint of the operation.
tower and providing feedstock to the alkylation unit.
For a rough estimate, a US Gulf Coast refinery with an average- sized FCC unit of 80K b/d will lose $68K a day for a downed unit if cash margins (which is defined as gross margins minus operating costs, before interest, taxes, and
depreciation and amortisation) of $1 a barrel and a utilisation rate of 85% are assumed. Depending on cash margins, profit penalties on the refinery could be considerable, as projected in Figure 5. A three-day shutdown of a FCC unit with a cash margin of $1/bbl at the time of the incident could cost a refinery over $200 000. On the other hand, when the cash margin was $5/bbl prior to a shutdown for a week, the monetary penalty could be as high as $2.4 million. Refineries on the US West Coast and in the Mid-West can see cash margins as high as $20/bbl and sometimes higher. A seven-day outage during margins this high can lead to a loss of almost $12 million. Of course, exact loss for each refinery is plant
make repairs. In April 2011, BP’s Texas City, Texas, plant experi-enced an external power failure that knocked BP’s plant and other refineries in the area offline. Several units, including an alkylation unit, coker and FCC unit were out of commission for more than a week. This amount of downtime leads to a substantial loss in profits.
Production loss and profit penaltiesWhen a power failure occurs, a refinery unit or units, or an entire facility must be shut down and production is lost. A refiner could post a loss instead of profit in a quarter, and adverse financial impacts are further magnified when refining margins are poor.
In a gasoline-centric refinery, gasoline is a mixture of FCC gasoline, alkylate, reformate, hydrocrackate gasoline and also possibly some straight-run gasoline from the still tower. However, both FCC and alkylation units contribute to about one half of the gasoline volumes. According to a study initiated by the US Department of Energy, a loss of one barrel of FCC input will result in a loss of one barrel of gasoline based on statisti-cal analysis.1 The one-for-one relationship is said to be appropri-ate because of the physical tie-ins between different operating units and limitations in storage and distribution systems that transfer intermediate feedstocks between units. The FCC unit is the primary production unit in a refinery, draw-ing feedstock from the distillation
About 19% of refinery emergency shutdowns in the US between 2009 and 2012 were caused by power disruptions
0 1 2 3 4 5 6 7
Production-time lost, days
12000
14000
10000
8000
6000
4000
2000Net
pro
fit/
loss
, $000’s
0
Cash margins = 25 $/b
Cash margins = 10 $/b
Cash margins = 20 $/bCash margins = 15 $/b
Cash margins = 5 $/bCash margins = 1 $/b
Figure 5 Net profit loss due to shutdown
hpc.indd 4 07/06/2013 17:53
thermal oxidizers
UOP Callidus building thermal oxidizers to exceed
expectations and protect the environment.
At UOP Callidus our broad base of thermal oxidizer designs and wealth
of experience allow us to provide customer solutions to meet your
specifications. You’ll find UOP Callidus thermal oxidizers in service in a
variety of gas and liquid waste applications including halogenated,
fume, tail gas, downfired salt and nitrogen bearing wastes. Each
system is built to meet the highest quality standards and to meet or exceed the latest environmental
regulations including NOX standards.
7130 South Lewis, Suite 335, Tulsa, OK 74136, call 1-918-496-7599 or visit our website www.callidus.com©2013 Honeywell International Inc. All rights reserved.
callidus.indd 1 07/06/2013 16:14
26 PTQ Q3 2013 www.eptq.com
Potential legal liabilitiesAside from missed production, outages prompt unnecessary flaring of hydrocarbons to avoid unsafe conditions. Over the past 10 years, the US EPA has entered into settlements with 28 different refin-eries that are aimed at restricting emissions by the oil industry. The EPA has acquired consent decrees from 105 US refineries in 32 states and territories since December 2000. All of the settlements have involved at least one of four primary pollution types: NOx, SOx, benzene and volatile organic compounds (VOCs). Furthermore, all of the violations involved one or more four key refinery components: the FCC unit, SRU, flares and heat-ers/boilers. Excessive flaring can lead to environmental concerns and incur fines imposed by environ-mental agencies.
The potential liability due to a prolonged flaring can cost a refiner a huge sum. For illustration purposes, a fire caused by a corroded pipe that led to subse-quent excessive emission at its
245 000 b/d Richmond, California, refinery in August 2012 had forced Chevron to pay $10 million to indi-viduals, area hospitals, city agencies and the Hazardous Materials Program as of January 2013. There are approximately 24 000 civil claims that have been levied against the firm as a result of the flaring at Richmond.
Emergency shutdowns because of power failure can also pose safety issues. Sometimes a refinery shut-down will force the need to evacuate workers for safety reasons. In April 2010, BP’s Texas City, Texas, plant experienced an electrical outage that shut off all power and steam to the refinery and forced the evacuation of non-essential workers from the plant. Safety-related incidents can result in the loss of life along with the potential for millions of dollars in fines and civil lawsuits, not to mention any negative publicity from the event. Previously, BP paid over $87 million in fines issued by OSHA along with undisclosed amounts in civil suits related to an
explosion and subsequent fire at its Texas City, Texas, refinery on 23 March 2005 that resulted in 15 worker deaths and over 170 injuries.7
ConclusionAs previously noted, about 19% of refinery emergency shutdowns in the US between 2009 and 2012 were caused by power disruptions because of severe weather, poor power quality and electrical equip-ment malfunctions. Costs to refiners amount to millions of dollars every year due to lost production, repairs to damaged equipment, sending valuable material to flare, possible fines for excessive emissions, and so on. Therefore, equipment vendors, maintenance and reliabil-ity servicing companies and refinery technology developers have been seeking technical approaches to prevent power fail-ures, protect equipment, speed up restarts and salvage damaged components. A few of the tech-niques available include approaches to examine transformer failure due
A new multi-client report:
WHEN THE POWER GOES DOWN, SO DO THE PROFITS!
Refinery Power OutageMitigations: LatestTechnologies and Strategiesto Minimize FinancialImpacts
A report prospectus is available atwww.hydrocarbonpublishing.com/ReportP/report13/power.pdf
The study analyzes causes of refinery powerfailures, evaluates associated costs, andrecommends solutions. Primary focuses are:
• Prevention & Protection
• Recovery & Restart
• Maintenance & Safety
• Combined Heat & Power
• Microgrid/Smart Grid
• Refinery Experiences
• Near- and Long-Term Approaches
hpc.indd 5 07/06/2013 17:54
www.eptq.com PTQ Q3 2013 27
Patrick J Christensen is Project Manager, Hydrocarbon Publishing Company, with seven years’ refi ning experience. He holds a BS degree in chemical engineering from Drexel University. William H Graf is Technology Analyst, Hydrocarbon Publishing Company, and holds a BS degree in physics from Hampden-Sydney College. Thomas W Yeung is Principal and Managing Consultant, Hydrocarbon Publishing Company.He is a licensed professional engineer in New York State and holds a BS degree in chemical engineering from University of Wisconsin-Madison, a MS degree in chemical engineering from University of Connecticut-Storrs, and a MBA from New York University. Email [email protected]
3 Carbonetto B, Pecchi P, Going ‘green’ with FCC expander technology, Hydrocarbon Processing, Jan 2011, 79.4 www.phil l ips66.com/EN/newsroom/n e w s _ r e l e a s e s / 2 0 1 3 N e w s Re l e a s e s /Pages/01-30-2013.aspx5 www.phi l l ips66.com/EN/about/our-businesses/refi ning-marketing/refi ning/Pages/index.aspx6 Feinstein to FTC: Investigate Spikes in California Gas Prices, www.feinstein.s e n a t e . gov / p u b l i c / i n d ex . c f m / p re s s -releases?ID=d5f32f54-4bec-490e-a350-f68797cef1d77 BP Texas City Violations and Settlement Agreements. OSHA website. www.osha.gov/dep/bp/bp.html
to switching transients, perform transformer end-of-life evaluations, undertake early fault detection and diagnosis in an FCC unit, reduce arc fl ash energy, predict and diag-nose medium-voltage switchgear and rotating machines, and other latest techniques.
Power grid reliability can be threatened by anything from severe weather to cyber-terrorism. CHP/cogeneration is said to be a good way to preserve power reliability through natural disasters. In the near future, microgrids will play an important role in refi nery power reliability and security. Since there are many options available and refi ners are often limited by tight budgets, the cost-effective strategy is fi rst to combine electrical outage data and process reliability models and identify the most vulnerable equipment and units. The results assist in sound deci-sion-making on what options and investments to choose in minimising power failures. Finally, disciplines such as actuarial science and enter-prise asset management should be included in overall refi nery opera-tions management. The objective is to ensure that the plant can achieve utmost reliability and highest energy effi ciency to maximise profi t-ability while fulfi lling safety and environmental requirements.
This article is the fi rst of a two-part series from a white paper called Refi nery Power Failures: Causes, Costs, and Solutions from Hydrocarbon Publishing Company. The paper is an excerpt from a multi-client strategic report called Refi nery Power Outage Mitigations: Latest Technologies and Strategies to Minimize Financial Impacts, to be published in June 2013.
Special thanks to Baldwin A Yeung, PE of SAIC, who provided technical assistance for this paper.
1 US Energy Information Administration, Refi nery Outages: Description and Potential Impact on Petroleum Product Prices, March 2007.2 Fluid Catalytic Cracking, 4Q2012 issue of Worldwide Refi nery Processing Review, Hydrocarbon Publishing Company (http://hydrocarbonpublishing.com/store10/product.php?productid=B21204&srchkey=)
RecipCOMIt sees what you can’t
HOERBIGER’s new monitoring, protection and diagnostic system:
Find out more:www.hoerbiger.com/[email protected]
When reciprocating compressors are critical to your process, inefficient
operation or breakdowns put the profitability and safety of the entire
plant at risk. A monitoring and protection system is the best investment
to ensure safe and reliable operation and cut maintenance costs.
On every revolution, RecipCOM’s continuous monitoring of critical
equipment parameters gives early warning of potentially catastrophic
failures, in real time and with SIL certified reliability.
hpc.indd 6 07/06/2013 17:54
limiting factors lovelimitless possibilitiesOvercome limiting factors affecting re� nery capacity and operating � exibility with BASF innovative FCC products, services and solutions. Our products deliver value to enhance sustainability and performance.
At BASF, we create chemistry for a sustainable future.
� Catalysts � Co-Catalysts� Additives � Services and Solutions
www.catalysts.basf.com/re� ning
basf.indd 1 6/12/12 13:23:11
Combating green oil formation in a CCR reformer
CCR reformer units need chloride injection during regeneration to promote plati-
num dispersion and to restore catalyst acidity. This leads to hydrogen chloride formation, and a highly viscous green oil may also be formed at some locations in the unit. This is one of the main reasons why the operation of a CCR reformer’s rich gas compres-sors can affect unit reliability. This article looks at a real-world exam-ple of green oil formation, and shows how the main cause was identified and then overcome.
When looking for a high rate of unit utilisation in a refinery, the availability and reliability of all equipment is of utmost importance to prevent loss of production and negative effects on profitability. So a refiner should either perform proper maintenance or find a way to deal with any kind of problem that threatens unit reliability.
Catalyst deactivation It has been known for many years that acid gases are present in the petroleum industries in liquid or gas streams. These gases include hydrogen halides such as HCl, HF, HBr, HI and mixtures thereof. From an acid gas point of view, one of the key processes of the petroleum industry are reforming reactions such as those in CCR reformer units. In the catalytic reforming process, sweet heavy naphtha is processed in contact with a platini-um-based catalyst to produce a high-octane product. Hydrogen is a byproduct of the catalytic reform-ing process and some of this product is recycled to the reaction
The addition of a chloride adsorber guard bed solved a refiner’s issues with contamination affecting a CCR reformer’s rich gas compressor
OSMAN KUBILAY KARAN, MEHMET ASIM AY and KORAY KAHRAMAN Tüpras Kirikkale refineryARNAUD SELMEN Axens Technology &Technical Services
section to maintain catalyst stabil-ity. This reforming catalyst is promoted with chloride in the pres-ence of water, resulting in the production of hydrogen chloride. Thus, the gas that is not recycled but sent to downstream catalytic processes and known as net gas contains hydrogen chloride. As a result, this chloride-containing gas can deactivate downstream cataly-sis because it can poison catalysts and cause undesired reactions.
Even the presence of a small amount of HCl in the net hydrogen gas can seriously interfere with the operation of downstream processes that use hydrogen. It can also cause corrosion problems in equipment such as pipes, valves and compres-sors. In addition, the formation of polymerised long-chain hydrocar-bons, generally called green oils, is a common problem in CCR reformer units.
Green oils are actually the end products of undesirable polymerisa-tion reactions taking place over the catalyst surface area, in which the reaction of HCl with hydrocarbons leads to chlorinated hydrocarbons. The presence of HCl will promote olefin polymerisation reactions with green oil downstream of the reac-tion section. These reactions are mainly chemical combinations of relatively small molecules with huge chain-like or network-structured molecules. Polymerised molecules formed in this fashion have complex multi-chain chemistries and high boiling points, and are typically waxy in nature. These molecules are green or red in colour and contain mainly C6-C18 hydrocarbons, with a potential tail above C40, and are
believed to be oligomers of light olefinic hydrocarbons, with some aromatic nuclei included in the structures. HCl in gas or liquid hydrocarbon streams must be removed, since it may cause unde-sired catalytic reactions and poison the catalyst systems of the down-stream units. Moreover, HCl is considered a hazardous material, so the release of this substance to the environment must be avoided.
Chlorine sourceFor the time being, the exact mech-anism of green oil formation is unknown, but it is believed that it is formed by the catalytic reaction of HCl with hydrocarbons, which leads to chlorinated hydrocarbons. Classically, a chlorination agent is injected during catalyst regenera-tion in the oxychlorination part of the regenerator to restore the opti-mal metallic phase dispersion of the platinum-based catalyst and to restore a normal chlorine content of 0.9-1.1 wt% on the catalyst. This leads typically to a recycle gas chlo-rine content of 1 ppmwt with a water content of less than 30 ppm. The HCl content of the recycle gas is kept under control, but hydrogen gas from the reduction section is also a chlorine source and both streams enter the net gas booster compressor section. The main chlo-rine contributor in the net gas comes from the reduction gas, which leads to exacerbated issues of green oil formation.
Tüpras Kirikkale refinery’s CCR reformer unit is licensed by Axens and was put into operation in 2008. The basic flow scheme is shown in Figure 1. Shortly after unit startup,
www.eptq.com PTQ Q3 2013 29
tupras.indd 1 07/06/2013 17:57
30 PTQ Q3 2013 www.eptq.com
• When draining the suction line of the third stage, a small amount of green and yellow oil was found. This oil was very sticky and gum-like.
For analysis purposes, the sticky oil samples were sent to an accred-ited laboratory. Inductively coupled analysis (ICP) was carried out
the CCR reformer’s H2-rich gas compressor was encountering frequent emergency shutdowns. These shutdowns were initiated by serious vibrations threatening the operation and reliability of the compressor. When it was disman-tled for investigation, some deposits were found (see Figures 2a-e).
The findings of this investigation included:• Dark green oil was adhering to the bottom of the first suction snubber• Liquid oil was found in the third-stage cylinder, and the valve plate was covered with coked hydrocar-bon. Part of the valve ring was damaged
Feed
H2-rich gas
Unstabilised reformate
Chilling system
Reactors and heaters
Regenerator
Recycle compressor
Booster compressor
Recontacting drumSeparator
Figure 1 Flow scheme for Tüpras Kirikkale refinery’s CCR reformer unit
Figure 2a First-stage No.1 suction snubber Figure 2b First-stage No.2 suction snubber
tupras.indd 2 07/06/2013 17:58
because there was not enough sample available for X-ray fluorescence (XRF) analysis. The ICP results are shown in Table 1. The dominant metal in the foul-ing material is iron at 8 wt% content. This suggests that the residue of the particulate analysis is mainly iron oxides. The iron can be present in the sample as metal particles or as Fe2O3. During analysis, metallic iron is oxidised so a distinction cannot be made.
Laboratory analysisThe laboratory reported that “the sample from the CCR reformer compressor contains 73 wt % C and 10 wt % H. No sulphur or nitrogen is detected. Next to that, several metals were detected, with iron being the most dominant at 8 wt%. The molar H/C ratio is 1.64, indi-cating very likely an olefin structure together with an aromatic nucleus. There are no signs of oxygen present in the sample. Approximately 8 wt% of the elemental contents cannot be explained, but based on the process this is possibly organically bonded chloride. However, this needs to be confirmed with a different analysis, because the current results are not conclusive. Further structural detailed information can be obtained by means of GC-MS.”
The same kind of liquid was found in a Korean refinery complex designed by Axens, and the
www.eptq.com PTQ Q3 2013 31
problem was solved by the installa-tion of a chloride adsorber at the outlet line of the reduction chamber.
Discussions with Axens about this specific issue were initiated and the H2-rich gas compressor was dismantled to discover the extent of the deposit. Finally, an adsorber drum was recommended, and an action plan has been defined to implement a chloride guard bed on the hydrogen gas from the reduc-tion chamber.
It is clear that green oil formation is highly dependent on the chlorine content of the H2-rich gas handled by the rich gas compressor. In addi-tion, hydrogen gas used for catalyst reduction is the main contributor of chloride. Besides good chloride management, the strategy to tackle this issue was to remove chloride from this process stream by means of a specific adsorbent before the suction section of the compressor.
Axens carried out the design of
the hydrogen chloride guard bed to be imple-mented at the outlet line of the reduction chamber. A non-regenerative promoted alumina was selected for the removal of HCl in the gas phase. A simple flow sheet for the scheme is shown in Figure 3.
Alumina-based adsorbent is widely used in the petro-leum refining industry as a trapping material for vari-ous impurities. Some special formulations loaded in a fixed-bed vessel enable
the removal of small quantities of chloride in liquid or gas streams. The adsorbent used for this process is generally disposed of at the end of its useful life; in other words, it is not regenerable. As the alumina- based adsorbents pick up HCl, the sodium or calcium promotor, as well as aluminium, reacts with HCl to form chloride salts. Green oil may also be formed in adsorption beds, and when these green oils are formed in fixed-bed adsorbents they cause fouling and result in the premature failure of the sorbent, which can be understood by a delta P increase through the bed.
Dräger tube chloride analyses are a simple way to check on a weekly basis the adsorbent’s performance and to check if there is any Cl breakthrough to determine adsor-bent replacement time, reported to be typically more than one year.
This new arrangement to mitigate the problem was put on stream in January 2012. Weekly Dräger tube
Figure 2c Third-stage discharge valve cover Figure 2d Third-stage discharge valve
Figure 2e Rod packing carbon seal rings
tupras.indd 3 07/06/2013 17:58
Unlike a phony cowboy who is all hat with no cattle, a boiler from RENTECH will pass muster.
Each boiler is designed and built to meet its demanding specifications and operate in its unique conditions in a variety of industries, including refining, petro-chemical and power generation. Our quality control system assures you that RENTECH boilers are safe, reliable and efficient.
For a real, genuine, original boiler, you can depend on RENTECH. Honestly.
www.RENTECHBOILERS.COM
rentech.indd 1 07/06/2013 16:13
www.eptq.com PTQ Q3 2013 33
and DHP units. He holds a degree in chemical engineering from Middle East Technical University, Turkey, and is a certified Energy Supervisor for industrial plants. Email: [email protected] Asim Ay is CCR/NHT/ISOM Units Process Superintendent with Tüpras Kirikkale refinery. He holds a degree in chemical engineering from Middle East Technical University, Turkey. Email: [email protected] Kahraman is CCR/NHT/ISOM Units Process Chief Engineer with Tüpras Kirikkale refinery. His six years of refinery experience includes the process side of hydrocracker and hydrogen production plants, sulphur recovery, NHT, ISOM, CCR and DHP units. He holds a degree in chemical engineering from Middle East Technical University, Turkey. Email: [email protected] Selmen is Axens’ Technology Manager for Naphtha Hydrotreatment and Reforming Technologies. He has worked mainly with bottom-of the-barrel technologies, specialising in heavy crude oil upgrading. He has also been involved in the process design of aromatics complexes and NHT, as well as reforming units startup and troubleshooting. He holds an engineering degree from the ENSGTI engineering school and a DEA in refinery process modelling from IFP School. Email: [email protected]
year without a major interruption, indicating that the vibration prob-lem was correctly identified. This case shows that a good licensor and refinery relationship is essential for solving problems that require both technological and operational experience.Osman Kubilay Karan is the Hydroprocessing Units Process Superintendent with Tüpras Kirikkale refinery. His 25 years of refinery experience includes the operational and process sides of crude, vacuum units, hydrocracker, hydrogen production plants, CCR
measurements are taken at the outlet stream of this adsorber and the results are 0 ppm HCl, whereas inlet concentration averages 30 ppm HCl.
ConclusionCurrently, the H2-rich gas compres-sors are running smoothly without any problem. The compressors were recently opened by the mechanical maintenance group and no green oil formation was found, although they ran for almost one
Reduction chamber
Reaction section
Dedicated chlorine trap
Separator drum
H2 to reduction
H2 from reduction
To booster compressor
Recycle gas
Catalyst streamProcess stream
Figure 3 PFD after adsorber drum
www.eptq.com PTQ Q2 2013 21
refi ner and particularly the ability to anticipate market needs in differ-ent regions as constraints evolve.
PolyFuel and PolyNaphtha are trademarks of Axens.
Marielle Gagnière is Technology Manager for hydroprocessing and olefi ns technologies downstream FCC, especially oligomerisation and etherifi cation technologies, in Axens’ Marketing, Technology and Technical Assistance Department. She is an engineering graduate from the Ecole Nationale Supérieure de Chimie de Paris, and holds a post-graduate engineering degree from the IFP School.Annick Pucci is Deputy Product Line Manager in the fi eld of light ends hydrotreatment and a specialist in refi ning olefi ns processing, particularly for FCC effl uent upgrading. She holds a bachelor’s degree in chemical engineering from Ecole Nationale Supérieure des Industries Chimiques de Nancy, France.Emilie Rousseau is a Strategic Marketing Engineer in Axens’ Marketing Department. She holds a chemical engineering degree from the Ecole Nationale Supérieure des Ingénieurs en Arts Chimiques et Technologiques de Toulouse, a master’s in chemical engineering from Imperial College in London and a master’s in energy economics and corporate management from IFP School.
It is in Europe where the difference between refi nery yield structure and market demand is critical, especially since conven-tional refi ning tools do not have the fl exibility to reduce excess gasoline production and to increase the amount of middle distillates. Moreover, with European refi neries facing increasing diffi culty in fi nd-ing export markets for their excess gasoline and given the tensions in middle distillate supply, PolyFuel should fulfi l a primordial role in adjusting the gasoline-distillate production to better fi t market demand.
In other regions, new tendencies such as shale oil and shale gas are revolutionising the US market, providing additional light products and consequently infl uencing market balance and prices. Today in the US, as a result of the impact of shale gas on the cost of LPG, PolyFuel is already profi table for a mixed feed of LPG and C5/C6 cut.
The fl exibility of the new process offers many advantages to the
The new technology is profi table taking into account today’s US market prices even if the middle distillate price is lower than the gasoline price. Indeed, as a result of shale gas production, LPG prices are low. Adding LPG (C3 and/or C4) cut in a PolyFuel unit lowers feedstock costs and contributes to increased profi tability, while maximising middle distillates production in the refi nery.
To reach 15% IRR for PolyFuel with prices based in 2012 in the US Gulf Coast, the middle distillate price can be $96/t lower than gaso-line. If the middle distillate price were equal to the gasoline price ($1129/t) and the LPG price kept at $636/t, the IRR would reach 28%.
ConclusionWith the world market for middle distillates growing and a reduced demand for gasoline in certain regions, the new process for olefi nic gasoline oligomerisation allows the refi nery scheme to be adapted to a maximum distillate mode.
Taking the Guesswork Out of CO2 RemovalProTreat® Simulation Software
Get RealDon’t Guess!
SyngaS, Lng and Carbon Capture
ProTreat accurately predicts column performance.
100% Mass-Transfer-Rate based — no ideal stages.
Grassroots? Revamp? Designing for Packing? Using MDEA with piperazine? HotPot?
Build a virtual plant with the industry’s most advanced gas treating simulator.
www.ogtrt.com +1 281 970 2700 [email protected]
Learn how state-of-the-art simulation science can benefit you.
OGT-0117 PTQ Half Horizontal Ad.indd 1 11/13/12 10:40 PM
axens.indd 7 08/03/2013 11:49tupras.indd 4 07/06/2013 17:58
blow...Our heater doesn’t
...the flames downward, against the natural buoyancy of the flue gases. Our steam methane reformer design fires upward,
reducing fan power consumption by 40 - 50%, to save operating costs.
Choose the Foster Wheeler Terrace Wall™ Steam Reformer for your hydrogen plant.
Learn more at www.fwc.com/terracewall
foster wheeler.indd 1 08/03/2013 10:49
Role of fired heater safety systems
Safety and risk mitigation have always been, and always will be, an important topic for any
operating company. Safety risks can be lurking anywhere through-out a facility and the consequences of an event range from minor inci-dents to catastrophic failures that lead to a loss of life. Protective systems are put in place to reduce the likelihood of occurrence of safety incidents and to take action in the event that unsafe conditions arise. While the safety and well being of personnel is of utmost importance, the financial impact of safety systems cannot be ignored. With so many options available, selection of a safety system can prove challenging but can also be rewarding.
Minimum safety goals must be met, but a safety system can go beyond meeting safety require-ments to improve overall operations and profitability. This article will discuss common risks associated with fired heaters, the role of the safety system, applicable standards and practices, safety instrumented systems (SIS) and the benefits of an automated burner management system (BMS), includ-ing an example of cost savings.
Fired heater operation and riskProcess fired heaters present signif-icant safety risks. Common in refineries and chemical facilities, they are used for heating, vaporisa-tion and thermal cracking of various process fluids. Heat energy, provided by the combustion of fuel, is transferred to a charge or feed in a controlled manner. The primary functions of a process fired heater
A fully automated burner management system operating as a SIS for burner control can meet minimum safety targets, improve system availability and lower costs
NIKKI BISHOP and DAVID SHEPPARD Emerson Process Management
are to maintain the desired outlet temperature at the desired charge rate. Besides maintaining tempera-ture and charge rate, control and safety systems are designed to maintain efficient combustion of fuel and safe operation throughout the full range of conditions the heater experiences. Figure 1 shows a typical process fired heater.
Burners in the heater transfer energy into the process through the combustion of fuel. As with any combustion process, care must be
taken to ensure safe operation. Fuel can accumulate when burners are off but should be on and also from substoichiometric conditions. Fuel must not be allowed to accumulate in the firebox as subsequent intro-duction of an ignition source could be catastrophic. In addition to combustion risks, fired heaters present risks associated with the process. Unlike boilers, where the process stream is water, the process stream for most fired heaters is highly flammable hydrocarbons.
www.eptq.com PTQ Q3 2013 35
FC
FC
FC
FC
Coil outlet temperature
Firebox temperature
Stack temperature
Total charge flow
Oxygen
CO
Draught pressurePC
TI TI
TI
TI
TI
Pass 1
Pass 2
Pass 3
Pass 4
TI
TC
TC
TI
TI
TI
AC
AC
FI
FC
Fuel gas
PI
PI
Fuel gas
Pilot gasCombustion
air flow
Figure 1 Typical process fired heater
emerson.indd 1 07/06/2013 18:01
36 PTQ Q3 2013 www.eptq.com
can be quite a daunting task to assess risk factors, assemble the interlock and permissive conditions and then determine which safety system to implement that meets both safety and financial targets. Fortunately, organisations such as NFPA, ISA, IEC and API publish documents that offer guidance for protective systems, which inform a user how to avoid a situation where the fuel supply should be off but is not, where the flame should be on but is off, where the process equipment is overheated, and where the protective system itself is prevented from working as it should. These standards also describe possible actions that the protective system can perform when it detects any of these situa-tions. Each document has been developed based on experience and offers valuable information. The fact that accidents and disasters are as infrequent as they are is due to the long experience that has been incorporated into the various stand-ards and recommended practices.
The National Fire Protection Association (NFPA) publishes NFPA 86 Standard for Ovens and Furnaces, which covers protective systems for process fired heaters. It applies to heated enclosures regard-less of heat source. NFPA 86 is a prescriptive, conservative standard written like instructions with few options. While NFPA develops the standards, it does not enforce compliance to the standard. Insurers or local authorities may, in certain cases, enforce compliance.
IEC 61508 Functional Safety of Electrical/Electronic/Programmable Electronic Safety-related Systems, developed by the International Electrotechnical Commission (IEC), provides the framework and core requirements for safety-related system design of hardware and software, independent of industry sector. IEC also released the document IEC 61511 Functional Safety – Safety Instrumented Systems for the Process Industry Sector, which defines the functional safety requirements established by IEC 61508 for the process industry sector specifically. The International Society of Automation (ISA)
Overheating or overfiring can cause process tubes to exceed metallurgi-cal limits and rupture. In cases where tube leaks occur, resulting explosions can destroy process equipment and pose a threat to human life. The release of the process stream into the surround-ings can pose an environmental threat. Even minor events can result in extended downtime for repair, impacting production.
The purpose of the fired heater safety system is to prevent disas-trous combustion of accumulated fuel and to prevent overheating and the subsequent catastrophic release of the process stream. It sounds simple enough to inhibit the admission of fuel when unsafe conditions exist, but the determina-tion of a safe state requires careful monitoring of many conditions. In particular, conditions such as fuel gas pressure and flow, furnace
draft pressure, flame detection, process stream flow, combustion air flow, tube skin temperature, stack temperature, per cent oxygen and combustibles all pose an opera-tional threat if limits are exceeded. The safety system must continu-ously monitor for unsafe conditions and take action when necessary, making it critical to understand all of the possible equipment failure modes and the potential impact to both the operating unit and personnel.
Safety system design and selectionEvery fired heater must have some type of safety system in place. It may be as simple as a written procedure for manual intervention or it may be a fully automated emergency shutdown system. The design and selection of a safety system starts with the evaluation of risk factors and risk tolerance. It
Analysis
Implementation
Hazard and risk assessment
Allocation of safety functions to protection
layers
Safety requirements Specifications for
the SIS
Design and engineering of
safety instrumenated
system
Design and development of
other means of risk reduction
Installation, commissioning and
validation
Operation
Operation and maintenance
Modification
Decommissioning
Safety lifecycle structure
and planning
Verification
Management of functional safety, and functional
safety assessment and auditing
Figure 2 Safety lifecycle as per ISA 84.00.01 2004
Source: IEC 61511-1 ed. 1.0 Copyright 2003 Geneva, Switzerland. www.iec.ch
emerson.indd 2 07/06/2013 18:02
MAN Diesel & Turbo has unique compression solutions in its portfolio for the complete range of CO2, N2, propylene and vapour related applications, with single machines or complete train solutions. Others talk about CO2 compression. We have the credentials! Over 200 RG Gas compressor units are in operation or on order. We are world`s number one in CO2 high pressure applications, thanks to sophisticated test-ing facilities and proven track records in the fi eld. With an RG Gas compressor from MAN Diesel & Turbo you will gain an exceptionally fl exible and optimized solution to maintain your business economic.Find out more at www.mandieselturbo.com
Engines and Marine Systems Power Plants Turbomachinery PrimeServ
Actions speak louder than wordsHigh pressure CO2 gas compression
man turbo.indd 1 10/06/2013 16:49
released a document very similar to IEC 61511, ANSI/ISA 84.00.01-2004, and the two documents were merged into one standard, IEC 61511 - Mod. This standard is a performance-based, rather than prescriptive, standard that applies to SIS regardless of application, with no specifi c functions defi ned. S84-2004, as the merged standard is more commonly known, focuses on the safety lifecycle. Steps include identifying risks, assessing the risks and then reducing the risk by means of a SIS. Figure 2 shows the safety lifecycle as defi ned by S84-2004. The standard clearly defi nes the steps for designing the SIS and requires that users have a good understanding of their process hazards and risks. ISA also published a technical report, TR84.00.05 Guidance on the Identifi cation of Safety Instrumented Functions (SIF) in Burner Management Systems (BMS), which offers specifi c guidance on SIS used as BMS. This technical report offers recommendations for assessing SIF within a BMS and provides some example safety assessments.
The American Petroleum Institute (API) issued the second version of recommended practice, RP 556 Instrumentation, Control, and Protective Systems for Gas Fired Heaters, in April 2011. RP 556 applies only to gas-fi red heaters and excludes boilers. In contrast to RP 560 Fired Heaters for General Refi nery Services, which applies to
38 PTQ Q3 2013 www.eptq.com
design and construction of heaters with little focus on instrumentation, the scope of RP 556 includes process measurement, process control and protective systems. RP 556 defi nes protective actions as basic process control action, opera-tor action and SIS action, and includes input devices, logic solvers and output devices as components. Compliance with IEC 61511-MOD is recommended for SIF. Specifi c recommendations for safe states and startup and shutdown
sequences are described. This docu-ment, meant to be a recommended practice, is not a prescriptive stand-ard and allows the sophisticated user to determine the best practice for their heater, leaving room for improvement and innovation. RP 556 was updated from an earlier release in May 1997 to refl ect advances in safety automation and design procedures and implemen-tation. The revised edition represents current cumulative best practices and provides a good
Sensor
Logic solver
Final control element
Figure 3 Elements of a safety instrumented system
design specifi cation for the process fi red heater.
The preceding list is not an exhaustive list of standards and practices. There are more available, such as FM 7605, developed by Factory Mutual, which requires that any programmable logic controller (PLC) listed for use in combustion safeguard service meets the SIS requirements contained in IEC 61508. A European standard, EN 50156-1, covers electrical equipment for furnaces and invokes SIS requirements for BMS.
There is no regulation requiring compliance to any specifi c standard or practice so it is left up to the user to sort through the standards and recommended practices, adopt the methods best suited to their needs, and then follow through with those practices. While it may seem overwhelming, the freedom to select a system best suited to a specifi c user’s needs provides an opportunity for innovation and improvement beyond minimum safety requirements.
Safety instrumented systemsA common trend in all of the stand-ards is the use of SIS for protective actions. By defi nition, a SIS is a set of components such as sensors, logic solvers and fi nal control elements arranged for the purpose of taking the process to a safe state. Figure 3 shows the components of a SIS. These are separate from all other control systems such that, in the event of a failure of the control system, the SIS is not prevented from performing the SIF. Certifi ed SIS systems follow a stringent certi-fi cation process and are designed, maintained, inspected and tested per applicable standards and recommended practices. Safety rated hardware is more robust and experiences fewer device failures than traditional hardware. SIS offer the benefi ts of improved safety, increased system availability and compliance with standards and practices. The benefi ts can be extended to operational benefi ts by taking advantage of safety systems that go beyond meeting minimum safety requirements and can actu-ally improve operational effi ciency.
The design and selection of a safety system starts with the evaluation of risk factors and risk tolerance
emerson.indd 3 07/06/2013 18:02
www.eptq.com PTQ Q3 2013 39
startup procedure may, therefore, be a lengthy process to complete.
When the startup procedure is a manual operation, each permissive must be manually verified before proceeding. Let us consider a fired heater with multiple burners and each with the associated gas valves. It could take a significant amount of time for an operator to manually check the valve positions based on the location of the heater and the individual valves. This also assumes the operator can correctly locate each valve and is careful to check them all. This manual confir-mation may take a significant amount of time and has the poten-tial for human error. A BMS with an automated sequence for light-off can save valuable startup time by eliminating the need for operators to manually verify valve position, detect a flame or manually time a purge.
The automated sequence ensures that each step is properly executed and eliminates human error associ-ated with possibly verifying the wrong valve or condition, as the sequence will check for the correct condition at the correct time. Even further, a BMS with a graphical
that a safe state is initiated when unsafe conditions are detected. Example interlocks include loss of flame, loss of combustion air, high or low fuel pressure, and excess process pressure or temperature. Should an interlock condition be detected, a fully automated BMS will initiate a safe state and, if necessary, shut off the fuel supply.
Consistent with the standards and recommended practices, a BMS can be treated as a SIS and the safety lifecycle can be followed. Minimum safety requirements can be satisfied through permissive and interlock conditions in the sequence logic and the associated SIF can be SIL rated. Figure 4 shows an exam-ple of a BMS sequence.
The sequential framework of a fully automated BMS provides additional value beyond meeting minimum safety requirements. One benefit is a significant reduction in startup time. Light-off events for fired heaters may only occur once every two to three years, so the startup procedure may be unfamil-iar to operators. Due to the inherent danger associated with light-off, each step in the process must be carefully executed. A manual
Going beyond minimum safety requirementsA BMS represents a great opportu-nity to go beyond the minimum requirements and can simultane-ously meet safety targets and provide operational benefits. By definition, a BMS is a system to monitor and control fuel burning equipment during all startup, shut-down, operating and transient conditions. They can range from a simple procedure that requires manual verification before proceed-ing or a fully automated system that automatically detects condi-tions and takes action.
A fully automated BMS uses sequence logic designed with a set of states, transitions, outputs and trips. The sequence is only allowed to proceed to the next step and take action if the permissive conditions for the transition are met. Permissive conditions are designed such that a safe environment is confirmed before proceeding. Examples of permissive conditions include fuel block valve positions, flame detection, minimum process flow, purge flow and purge timer. While in any state, interlock or “trip” conditions are designed so
S01
S02S03
S04
S05
S06
S07
S08
S09S10
S13
S12
Shutdown, not ready
Shutdown and ready
Pre-purge in progress Purge
complete
Ignite pilot
Pilot only running
Cold start. Set low fire position
Ignite main with pilotMixed gas Main without pilot.
Not at temperature
Waste gas only
Mixed firing. Set low fire position
Trips from states 5, 6, 7, 8, 9, 10, 12
Start-up failure
Figure 4 Example of a burner management sequence
emerson.indd 4 07/06/2013 18:02
Leading in LNG technology.
Linde Engineering, a leader in cryogenic technology, offers the full scope of engineering and contracting services – from feasibility studies, pre-FEEDs and FEEDs all the way up to full EPC turnkey responsibility. And as a world-class manufacturer of key cryogenic equipment, we are in the unique position to offer both brazed plate-fin heat ex-changers and coil-wound heat exchangers.
The combination of proprietary technology, front-end engineer-ing and project execution capabilities – as well as experience as a manufacturer of key cryogenic equipment – enables Linde Engineering to deliver tailor-made LNG plants to the full satisfaction of clients worldwide.
Linde – ideas become solutions.
Linde AGEngineering Division, Dr.-Carl-von-Linde-Strasse 6–14, 82049 Pullach, GermanyPhone +49.89.7445-0, Fax +49.89.7445-4908, [email protected], www.linde-engineering.com
linde.indd 1 07/06/2013 16:53
www.eptq.com PTQ Q3 2013 41
improvement. Implementing a fully automated BMS as a SIS for burner control and monitoring can simulta-neously meet minimum safety targets, improve system availability and lower costs.
Nikki Bishop is a Senior Application Consultant at Emerson Process Management. With over 12 years in the process control industry, her experience includes automation projects in industrial energy, pharmaceuticals, power generation, pulp and paper, and refi ning. She holds a BSChE degree from Georgia Tech and is a registered professional engineer in the state of Georgia. David Sheppard is Engineering Manager with the Midwest Engineering Center for Emerson Process Management. He has served as Lead SIS Engineer for multiple burner management system and refi nery safety systems projects, including complex multi-burner, multi-fuel burners and has been a Certifi ed Functional Safety Expert (CFSE) since 2007.
specifi c unit. In cases where spuri-ous trips are reduced through integration of device diagnostics in the BMS, the savings are even greater. The savings associated with automating the light-off sequence to reduce the startup time and integrating device diagnostics to reduce spurious trips are signifi -cant and simultaneously ensure a safer facility.
ConclusionSafety systems are required to protect personnel and environment, but can go beyond meeting the minimum requirements and provide fi nancial benefi ts as well. The appli-cable standards and recommended practices provide guidance for safety system design and selection criteria, and also provide an oppor-tunity for innovation and
user interface that is easy to under-stand and clearly indicates status eliminates the need for operators to understand and sort through complex logic diagrams. A clear fi rst out indication should be provided to the operator to ensure they know exactly what is prevent-ing a startup or what caused a trip. This saves many hours’ trouble-shooting when compared to a manual process or PLC-based solu-tion, where inherently dangerous trial and error procedures must be used to determine what condition is preventing startup. Instead, the specifi c trip condition can be addressed quickly. Figure 5 shows an example of a graphical user interface for an automated BMS.
Diagnostic data can also play a role in an automated BMS. The use of smart devices and HART communication for system hard-ware fault identifi cation and fi eld device failure alerts provides continuous monitoring of sensors, logic solvers and fi nal elements so that faults can be diagnosed early. This diagnostic data can be inte-grated with the BMS so that its overall integrity can be maintained, reducing spurious trips that signifi -cantly reduce operations and maintenance cost.
Table 1 shows an example of a cost savings calculation associated with a two-hour reduction in startup time for a typical 200 000 b/d refi nery with typical down-stream processing capacities. The savings calculated are for a single light-off event. For example, a two-hour reduction in startup time for a single heater in a hydroc-racker unit equates to savings of $100 000 and $300 000 if you consider all three heaters in the unit. In the case of the reforming unit, where margins are even higher, the savings equate to over $250 000. A typical refi nery may have three or more hydrotreating units, and thus the $17 000 esti-mated savings would apply to each unit, such as hydrotreating units for naphtha, diesel, heavy oil and possibly jet fuel. The total annual savings will depend on the number of light-off events for each heater and the margin associated with the
Figure 5 Graphical user interface for BMS
Unit Heater Margin $/bbl Two-hour value Unit totalCrude distillation Crude heater $2 $33 334 Prefl ash heater $2 $33 334 $66 668Vacuum distillation Vacuum heater $1 $16 666 $16 666FCC Feed heater $6 $100 000 $100 000Alkylation lsostripper reboiler Heater $4 $66 666 $66 668Hydrotreating Reactor feed heater $1 $8334 Stripper reboiler heater $1 $8334 $16 668Reforming Reactor feed heaters $8 $133 334 Stabiliser reboiler heater $8 $133 334 $266 668Hydrogen plant PSA Reformer reactor furnace $8 $133 334 $133 334Hydrocracking Feed heater $6 $100 000 Fractionator heater $6 $100 000 Stabiliser reboiler heater $6 $100 000 $300 000Coking/thermal cracking Feed heater $8 $133 334 $133 334
Example savings calculation associated with two-hour reduction in unit startup time for a 200 000 bpd refi nery
with typical downstream processing capacities.
Sample cost savings for a fully automated BMS to reduce startup time
Table 1
emerson.indd 5 10/06/2013 14:10
Manufacturing execution systems for the refining industry
Transforming data into mean-ingful business knowledge is vital to optimising production
and maximising commercial poten-tial. Today, a refinery business needs to be more agile and respon-sive to fluctuations within the market. Effective performance management involves integrating planning, scheduling, execution and the ability to respond to change immediately. This article explores the background to manufacturing execution systems (MES) and reveals how MES software technol-ogy delivers more efficient data management, improved production execution and enhanced operational performance, enabling refiners to quickly turn data into profit.
Milestones of MESDelivering products that consist-ently meet customer expectations helps refiners remain competitive and achieve higher profitability. MES provides intelligence for opti-mising operations with rapid, accurate and transparent data in real time. With a better understanding of how their plants are performing in real time, refiners can positively impact the bottom line with timely, informed decisions about produc-tion performance.
There are three key milestones for MES, which form the pillars of profitability: • More efficient data management (MES 1.0: integrating the past)• Improved production execution (MES 2.0: the era of work process automation)• Enhanced performance manage-ment (MES 3.0: technology on the move)
Real-time data and decision support tools provide access to plant information to enable a timely response to production issues
MARTY MORANAspenTech
MES 1.0: integrating the pastMES emerged in the process indus-tries over 30 years ago. Only in the late 1970s did minicomputers become affordable enough to be successfully used in the process industries. The earliest applications were primarily data historians (handling values such as tempera-tures and pressure levels) in the large continuous industries, such as refining and bulk chemicals. The primary need was “historising” time-series data for trending and later analysis. Over time, refiners began to generate significant volumes of data, but struggled to leverage information effectively.
During the era of MES 1.0, other new technologies, such as planning, scheduling and advanced process control (APC), also emerged to further enhance refinery profitabil-ity. For example, APC enables refiners to increase throughput, improve product quality, reduce energy and raw material usage, as well as enhance operational effi-ciency while keeping the process within safe limits of reliable opera-tion. Today, many leading refiners have successfully adopted this tech-nology using aspenONE APC software. Not only can it efficiently scale to any control problem size, it has also been successfully applied to virtually every control problem in refining, chemicals and petrochemi-cals processing. Significantly, this helps improve the financial perfor-mance of the plant, where companies have experienced bene-fits ranging from a 3-5% increase in capacity and a 3-5% reduction in energy usage.
Another significant technology
development in the early to mid-1990s was refinery scheduling. This allowed a refiner to take the leading planning model (Aspen PIMS) and create a time-based schedule for either crude or unit scheduling. Crude scheduling involved informing operations of whatever oil movements were involved in the crude oil portion of the refinery, such as offloading a ship to a specific tank, or deciding what tank should charge the crude unit.
As MES software continued to progress through this period, the foundations to greater manufactur-ing profitability were being laid. During the 1980s and through to the economically challenged early 1990s, an expert engineer was usually required to interface with a MES. However, new technology has emerged in recent years that has made it much easier for casual engi-neers to leverage today’s MES. For example, greater intelligent search functionalities and business intelli-gence (BI), such as Microsoft Suite, make it easier to find and leverage information in a MES. The Aspen InfoPlus.21 family, for example, was initially developed to aggregate process, production and business information into a cohesive context for understanding and improving performance. It is now the founda-tion for integrating and connecting plant control systems and the shop floor with business systems.
MES 2.0: the era of work processautomationThroughout the 1990s and into the new millennium, the petroleum industry began to recognise the
www.eptq.com PTQ Q3 2013 43
aspentech.indd 1 07/06/2013 19:12
44 PTQ Q3 2013 www.eptq.com
historian software technology can provide value for refinery blend engineers by allowing them to compare and contrast past blends to improve blend correlations. The net result is using fewer of the higher valued materials during blending operations while still meeting prod-uct specifications.
The era of automation began to show value as refiners strived for operational excellence through improved monitoring, control and analysis of their operations. MES today enables refiners to quickly identify manufacturing performance problems, assess root causes and take corrective action; aspenONE MES, for example, was developed to provide the tools that enable compa-nies to increase profitability, reduce variability and improve asset utilisa-tion. Crucially, production execution software is tightly integrated with data historian software to improve the manufacturing process, which automatically reaps positive returns on investment.
MES 3.0: technology on the moveIn the 21st century, the world has witnessed a revolution in ground-breaking interactive technology. “Smart” products developed during the past decade have provided greater communication and collabo-ration functionalities, facilitating quicker decision-making while oper-ating on the move. Flexibility, ease of use and real-time data visualis-ation are significant benefits to users. This period of “greater intelli-gence” in technology has opened up new possibilities. Refiners have a greater need today to streamline processes to improve operational performance and manage intelli-gently the huge quantities of data that process plants produce on an hourly and minute-by-minute basis.
BI empowers employees to perform with greater flexibility, as it helps improve access to manufactur-ing data at all organisational levels to drive quicker decisions. Event notifications coupled with mobile analysis tools enable faster adjust-ments to minimise the impact of production issues. This is vital in the process industries because there are many operations-based personnel
importance of automated work processes. Simply relying on a data historian was not enough. During this period, software was introduced that allowed refiners to automate oil movements, order processing and monitoring. Normally, the schedul-ing function sends down orders that certain oil movements need to occur. However, the actual work process of oil movements had many manual steps and effective monitoring of oil movement operations was always a challenge.
Software developed during this period automated oil movement monitoring. It calculated the gross volume in tanks using strapping tables based on real-time estimates of tank level and qualities. It also alerted the operator when a move-ment should occur, when 90% of the movement had been reached, if a tank was in double movement and whether unauthorised movements were taking place. Before the move-ment began, it ensured that there was enough source material and sufficient room at the destination to adequately complete the movement.
In this era, software programs were introduced that allowed for operations reconciliation and accounting, providing an accurate and timely account of liquid flows and inventories on a daily basis and enabling more intelligent business decisions. The automated system could reconcile production on a daily rather than monthly basis, giving refiners improved confidence of actual inventory positions, includ-ing being able to tabulate the exact inventory in each tank.
Oil accounting is a business process of measuring, validating, reconciling and publishing all the work flows on, and inventories within and out of, a refinery. Its practice varies widely. Flows are normally imported from the refinery data historian and tank inventories from the tank gauging system. It boosts profitability by identifying product loss and consistently prob-lematic instrumentation, and provides decision-makers through-out the plant with critical reconciled production data. Proper yield accounting, for example, must also take into account oil movements
that occurred during a particular day (for instance, where automated oil movement software understands the amount of oil moved during a 12am to 12am period). This informa-tion can be used in the yield accounting software to more accu-rately reflect the final daily inventory positions. Companies can now closely track production and take timely corrective actions when deviations occur, ensuring that commitments are met and helping to avoid purchasing components or downgrading shipments.
During the MES 2.0 timeframe, new MES non-work process soft-ware was developed that refiners would ultimately find useful to solve important problems facing the industry. Initially established for the batch-orientated industries, produc-tion record historian technology was developed for production segments with a defined start and end “marker”. Using this technology, engineers were able to review past production runs much more frequently using data from many different sources visually overlaid on top of one another to speed prob-lem analysis. This gave engineers the ability to quickly learn the intri-cacies of their production process and to use the resulting knowledge to improve the process. Prior to this technology, this type of analysis was never undertaken due to the complexity of assembling all the data required.
One example where this technol-ogy could be applied is in refinery product blending, which has a well-defined start and stop marker. A refinery product blend, such as premium gasoline, must meet certain product specifications. For example, minimum gasoline blend requirements are octane and Reid Vapour Pressure, but often there are other specifications, such as sulphur, benzene ethanol or other require-ments. The amount of gasoline blend materials, such as reformate, alkylate, catalytic reformer gasoline and light straight run, are usually pre-determined based on blending correlations. However, these correla-tions are imperfect and often assumed to work across a narrow range. This is where production run
aspentech.indd 2 07/06/2013 19:12
who are not desk-bound and can benefit from access to real-time data, trends and alerts — anytime, anywhere.
Plant managers and production engineers can use mobile BI anytime, anywhere in order to first under-stand an issue and then propose solutions to problems in a time-frame not previously experienced with tradi-tional desktop solutions. The prevalence of mobile devices is transforming the process industries. Mobile solutions empower decision-makers to have immediate access to important data, enabling them to make informed and quick decisions to improve profitability. Easy, digestible analysis of plant information, even in remote locations, helps industry to react to adverse changes and keep the operation performing to targets. The ability to access and analyse real-time plant data has enormous benefits. In the past, users needed to be in the control room or in front of a monitor to track and manage manufacturing performance. Mobile BI has proven to be more effective when users are provided with visualisation tools (charts, graphs, portals and so on).
For today’s engineers, the message is simple: mobile intelligence provides the platform to achieve greater profitability. State-of-the-art mobile software, such as Aspen InfoPlus.21 Mobile, enables faster decision-mak-ing and troubleshooting and displays critical, up-to-date information. The software functionality improves employee efficiencies by simplifying routine engineering analysis tasks, such as examining and comparing process data, reducing root cause analysis time and easily finding KPI data that will enable the engineer to respond to changing process conditions.
ConclusionOver the past 30 years, MES technology has evolved to help refiners survive in highly competitive markets. Real-time data and decision support tools provide access to plant information to allow quick and timely responses to production issues that negatively influ-ence efficiency, quality and regulatory compliance. MES is essentially the nucleus of the operation, which links all capabilities of the business. It is an integrated set of production activity and support software designed to harmonise and optimise the plant.
The bottom line is that effective production drives operational excellence, enabling better and faster deci-sions. Software technology helps refiners achieve consistent performance across all assets. It also defines the importance of real-time business performance management: plan, execute, monitor and respond to change immediately on all time horizons. History has shown that manufacturing execution systems have laid the foundations to help refiners across the globe strengthen their competitiveness and build upon the pillars of profitability.
Marty Moran is the MES Manager at AspenTech. During 30 years in the process industries, he has worked as a consultant in over 65 refineries, chemical, gas plants and other manufacturing environments. He holds a US patent for multivariable control and a degree in chemical engineering from the University of Illinois.
www.eptq.com www.eptq.com
cause sediment formation in the heavy fuel oil. The fouling tendency can be shown to increase exponen-tially with conversion increase (see Figure 1). Therefore, the value benefit of the conversion increase can be lost to fouling and sediments. However, this fouling-to-con-version relationship can also show that there is no major advantage in decreasing severity whenever fouling rates are acceptable or controllable. To do so would simply result in lost conversion without a corresponding value improvement in terms of sediment deposition control and run lengths. This relationship brings about the concept of optimising conversion as a function of the rate of fouling and fuel oil sediment formation.
Fouling problems, their monitoring and control Problem areasThe sections most prone to fouling are the atmospheric column, vacuum column and preheat exchangers. At very high conversion, the reactor and separator may also suffer from high coke generation. Extensive foul-ing of the separators and columns can lead to unplanned shutdowns, downtime and lost production.
The same trends for conversion severity are valid for sediment generation in the fuel oil. Below certain limits of conversion the fuel will be stable, while above certain limits the tendency to generate sediment with time cannot be controlled.
From the above considerations, it is clear how setting the proper operating conditions is important, as this enables the best trade-off between conversion maximi-sation and production of stable fuel and acceptable rates of fouling. Optimal plant management requires continuous control for resid product stability. The stability is related to the tendency to produce fouling deposits and generate sediment.
Optimal severity depends on the properties of the feed being processed. The feed changes whenever the refinery feed quality, residual feed make-up or the plant feed rate changes.1 Feed composition-related factors that may influence severity/conversion are the stability reserve of asphaltenes in the vacuum resid (often reported as p-value), the content of asphaltenes and the intrinsic solubility of these asphaltenes. Low stability reserve, high asphaltene content and poor solubility will all contribute to an increased tendency to generate coke and unstable residuum product and fouling deposits.
The amount of metallic contaminants (especially sodium and iron) is another factor that can impact process performance by affecting catalyst performance and, in some cases, by increasing coking tendency, favouring dehydrogenation and conversion of coke precursors — those less soluble thermally cracked asphaltenes — into coke.
To avoid deposit generation, the LC Finer catalyst plays a major role.3 The type of catalyst utilised in the process can have a great effect. In ebullated-bed reac-tors, the catalyst is changed continuously to maintain catalyst activity and to remove metals from residue oil. The effect of the catalyst operations on the process and fouling can be monitored easily with sophisticated
efficiency trays must be used to reduce reflux ratio and lower the overhead cooling load. Flue gas turbines, energy recovery hydraulic turbines, energy-saving motors, frequency conversion motors and air flow regulation systems of compressors should be employed to recover pressure energy and reduce elec-tric energy consumption. High-efficiency intensified burners need to be used in furnaces to improve effi-ciency. The exhaust temperature of furnaces should be reduced to improve thermal efficiency by 2-3%. New insulation material needs to be employed to reduce the heat loss of equipment and pipelines.
Low-temperature heat should be utilised by apply-ing low-pressure steam generation technology, low-temperature Organic Rankine Cycle (ORC) systems and Kalina Cycle systems.
Energy optimisation of the area should be carried out. Heat integration between the refinery and local co-generation power plant needs to be realised. A large quantity of low-temperature heat in a refinery cannot be recovered because a heat sink is not availa-ble, and this portion of low-temperature heat may be used as a heat source for demineralised water and boiler feed water in a cogeneration power plant. Also, integration with the local chemical plant is important to achieve material exchange and optimisation, as well as optimum energy use. Integrated energy optimisa-tion of the area should be carried out rather than energy optimisation of a single refinery.
ConclusionsThe priority of refining technology development and the configuration of refining units in China have distinct characteristics. During a long period in the future, the FCC unit will still be the main secondary conversion unit for gasoline and diesel production.
Therefore, improving the quality of FCC gasoline and diesel is very important for oil product quality upgrading technology in the future.
With the trend towards poor-quality crude oils, future refineries should further optimise process flows and change unit configuration so as to improve the comprehensive utilisation rate of resources, and to meet the requirements of energy savings and emis-sions reduction.
More detailed classification of crude oil fractions and the consideration of various narrow-cut process-ing technologies will actively promote the low carbon emission of refineries. Narrow-cut processing will become the development trend in the overall process flow optimisation of refineries in the 21st century.
Sun Lili is Vice President of Sinopec Engineering Incorporation in charge of processing solution studies, engineering, construction and startup of grassroots refineries and refinery modification projects. She has over 20 years’ experience in engineering design and startup of hydroprocessing units, has won several Prizes of National Scientific and Technological Progress, and holds a BS in petroleum refining from China University of Petroleum. Email: [email protected]
www.eptq.com
www.dpharp.com
High Performance
Safety
Robustness
A New Standard in Pressure
Measurement
EJA-E-270x86.indd 1 07/09/12 13:50
sinopec.indd 8 10/12/12 13:02:15baker hughes.indd 2 08/03/2013 12:10
aspentech.indd 3 07/06/2013 19:12
Machinery uptime is at your � ngertips
Intelligence on Duty
[email protected] www.prognost.com
Which monitoring system gives you the most time to respond? Maximize lead times with PROGNOST® – the #1 system in early failure detection. You get the early warnings you need to make well-informed decisions and cut maintenance costs.
Trust the #1 online monitoring system for mission-critical equipment
prognost.indd 1 06/06/2013 10:15
Advances in engineering and design technologies
To a professional responsible for the safe and efficient oper-ation of an oil and gas facility,
today’s immersive 3D computer games might seem like only a form of engaging relaxation. But the 3D visualisation pioneered for the games industry is about to play increasingly important roles in the lifecycles of tomorrow’s plants.
Computer processing power can be considered as a sort of digital budget, which has to be appor-tioned by an application to achieve optimum overall performance. In engineering design, 3D visual rendering is less important than more immediately productive features such as responsive posi-tioning of complex objects and sophisticated clash detection. But as processing power has increased, to overlook the potential of realistic 3D representation is to miss an opportunity to increase design productivity and quality.
Recent development at Aveva showed that, for design tasks, only limited visual cues are necessary to create a convincing representation of reality. By incorporating these into its latest design solution, Aveva Everything3D (E3), the company provides designers with intuitive visualisation without slowing system performance. Interactive controls enable a user to adjust three key rendering elements: edge definition, high-lighting and shadowing. The effect is surprising; as one gradually increases the settings, simple geometric shapes quickly assume convincing solid forms and unam-biguous positions in the virtual plant.
3D technology is moving out of the design office to transform the entire asset lifecycle
SIMON BENNETT Aveva
The result is a new level of intui-tive interaction with the design model (see Figure 1). As a designer moves an object, the subtle cues of its highlights and the shadow it casts make its actual 3D location more obvious. The result is a small but valuable improvement in the time and effort required to position an object or route a pipe. Aggregate this across the hundreds of individual positioning operations performed every day during design development and the result is significant. Design productivity increases, saving time and effort through quicker, more accurate positioning and less repositioning. Soft clashes — those between colli-sion spaces around objects — can be avoided almost unconsciously as shadows indicate proximity between adjacent objects.
Upgrading assetsGains in design productivity for new projects are always welcome,
but the same approach can also help with the far more numerous revamp and upgrade projects. Here, the bottleneck has always been in the limitations of available surveying methods. Rapid advances in 3D laser scanning systems, and the soft-ware that exploits the rich data they generate, have not only overcome this, they have unlocked a new level of capability in refurbishing older facilities.
First, 3D scanning captures far more detail at far greater accuracy and with less effort than any other method. Today’s scanners can not only generate accurate, photorealis-tic 3D representations of an in-service facility, they can do so quickly and (usually) without disruption to normal operations. Importantly, these new technolo-gies to capture 3D information are safer than traditional measuring methods, requiring less work at height, and are more cost effective (see Figure 2).
www.eptq.com PTQ Q3 2013 47
Figure 1 Intuitive interaction with the visually realistic design model (right) makes the avoidance of clashes easier, saving valuable design time
aveva.indd 1 07/06/2013 19:16
48 PTQ Q3 2013 www.eptq.com
enabling of lean construction meth-odologies. Casting an envious eye on the many benefits brought by the lean manufacturing revolution, the plant industries have long sought the key to unlock lean construction. That key has now been created. By exploiting the ease and affordability of laser scanning at every stage in the fabrication and construction sequence, and integrating the data with the as-designed model, the feedback loop can be closed between design, fabrication and construction.
In one-off capital projects, if a costly or long-lead item is made incorrectly, the programme impact can be considerable. But if the devi-ation can be identified immediately and in detail, an informed decision can be made to mitigate its impact and protect the programme. For example, suppose a project requires a concrete base with a number of mounting points for key modules. The concrete is poured, but only when the modules are being installed several weeks later is it discovered that some mounting
Second, software advances have brought new ways in which to use 3D surveys. Early developments enabled the accurate but relatively sparse “point cloud” representations of the as-operating plant to be refer-enced within a 3D design system, enabling new design to be aligned accurately with existing construc-tion. This substantially reduced the commercial risk in revamp projects, as new design could be created and fabricated in the confidence that it would fit right first time during on-site installation.
Rapid development has taken this further. The latest software releases enable both design models and laser scan data to be combined in the same 3D design environ-ment. The improved design visualisation described above is matched by high-definition “real-world” laser scans, blurring the visual distinction between design objects and surveyed objects. The designer can work equally intui-tively with both types of information. Now, for the first time,
the real and the virtual worlds can be integrated in a common environ-ment (see Figure 3).
This brings important benefits. One is the ability to efficiently reverse-engineer existing plant construction. Software can now recognise, for example, that a cylin-drical array of 3D scan data points represents a pipe run. By comparing its diameter with available pipes in the system catalogue, it then offers the designer a shortlist of candidate pipe specifications. The correct spec-ification is determined from the P&ID and selected from the short-list, whereupon the software creates a native, intelligent pipe object accu-rately co-aligned with its scan representation. Current capabilities cover pipes, nozzles and steel beams, increasing productivity on some of the most repetitive aspects of reverse engineering.
Lean revolutionHowever, the most far-reaching benefit of integrating the as- designed and the as-built lies in the
©20
13 S
TREA
MLIG
HT, I
NC.
CONNECT WITH US WWW.STREAMLIGHT.COM
When you have a job to do,
you’re not thinking about
your flashlight’s engineer or
the design, functionality or
keen ingenuity that offers
a premium product for the
GREATEST VALUE.
In fact, you don’t have to think
about your flashlight at all.
Because it’s a STREAMLIGHT.
130 LUMENS15,000 CANDELA
POLYSTINGERLED HAZ-LO®
200 LUMENS1,150 CANDELA
KNUCKLEHEAD®FLOOD
80 LUMENS4,000 CANDELA
TRIDENT® LEDHEADLAMP
615 LUMENS4,000 CANDELA
E-FLOOD®LITEBOX®
aveva.indd 2 07/06/2013 19:16
positions are incorrect. Inevitably, recovery incurs cost and schedule overruns.
Now it is possible to survey the foundation as soon as the concrete is cured enough to walk on. An accurate, photo-realistic 3D scan can be immediately sent back to the design office, loaded into the design system and quickly compared with the design model. Immediate, informed action can be taken to recover the situation and protect the project schedule. This might, for example, involve the re-routing of pipes or access struc-tures, or authorising a design modification to the affected plant modules while still in fabrication.
While this is evidently a consider-able benefit when applied to major plant elements, low cost and ease of use enable the process to be applied routinely right down to the level of an individual pipe spool. The commercial incentive on fabricators, suppliers and sub-contractors to supply exactly what is specified to the next customer in the chain can be realigned to reflect their stake in the overall success of the project.
Into operationsIn parallel with the technology evolution, a new generation of
www.eptq.com PTQ Q3 2013 49
industry professionals has grown up in an environment rich in inter-active, highly visual technology. The stage is set for a transformation in the way we manage engineering assets. Where plant operators once regarded 3D as only a design tool, they are now coming to recognise the considerable value of 3D models in operations.
Staff training and procedure planning are obvious applications. People learn most effectively by doing, and understand most easily by seeing. With realistic, immersive visualisation of complex engineer-ing assets, one can learn by doing in a safe environment, just as in a flight simulator. Not surprisingly, the term “industrial gaming” has
Figure 2 Combining a 3D CAD model and an accurate laser scan BubbleView representation of an in-service facility offers a new level of capability for plant revamp projects
Figure 3 Design models and laser scan models can be combined in the same 3D environment
aveva.indd 3 07/06/2013 19:16
• Valve (numerous options)
• Sieve or perforated Bubble cap trays
• Dual fl ow • Baffl e
TRAYS RANDOM PACKING
• Catalyst bed supports • Basket strainers
• Nozzles • Outlet/Inlet baskets • Distributors
Hub and header laterals
WEDGE WIRE SCREEN
Wide variety of random packing types, sizes and materials in stock!
Ask us how our patented SuperBlend™ 2-Pac can increase your capacity
and effi ciency.
DISTRIBUTORS & SUPPORTSManufactured to customer
specifi cations or engineered to meet performance requirements.
COALESCERS• Oil water separations • Haze removal
from fuels • Removal of tower wet refl ux • Caustic treater applications
STRUCTURED PACKINGWoven, sheet metal, and knitted structured packing. Built to spec
or performance requirement.
Need a Mist Eliminator...NOW!Special Alloys? Fast Delivery? Ask AMACS Process Tower Internals about our Fast Track delivery for our full line of mist eliminators. If you need mesh or vane mist eliminators in a particular alloy, AMACS in-house engineering and fabrication can deliver to spec or performance requirement.
Numerous Mesh Alloys and Plastics Vane (Chevron)
Multi pocket vaneInsertion Type Mistfi x®
Visit our new website atwww.amacs.com
www.amacs.com • 24hr EMERGENCY SERVICE • (281) 716-1179
amacs.indd 1 10/06/2013 16:51
www.eptq.com PTQ Q3 2013 51
technology that will trigger the development of lean construction methodologies. It enables existing legacy assets to be brought into the digital environment for more effi -cient management, and it supports highly effi cient, safe and compliant plant operations. The future plant will be both a physical and a digital entity.Simon Bennett is a Senior Product Business Manager with Aveva in Cambridge, UK. He has over 10 years of experience as a software product manager and was responsible for launching Aveva’s Everything3D.
And this can be done even in loca-tions that would be inaccessible on the physical plant.
3D from start to fi nish3D is moving out of the design offi ce to transform the entire asset lifecycle. Combining more powerful design functionalities with the abil-ity to accurately capture the as-built asset and associate both types of information with every other type is fundamentally changing the way we create, operate and maintain plants. It now provides the enabling
been coined to describe the concept. 3D visualisation can be used by new recruits for facility familiarisa-tion, in preparation for visits to remote facilities, or for updating skills and procedures following plant modifi cation. It can cover training in operations or safety procedures, testing the most complex what-if emergency response scenarios, or collaborative planning between multi-site teams.
However, there are even more powerful ways to use it. Information management technologies like Aveva Net enable 3D data — whether a CAD model, a laser scan representation, or both — to be inte-grated and cross-referenced with every other types of engineering or operational data. Navigable 3D models can not only be combined with other information, they can be used as a powerful tool for working with the vast and complex digital plant information asset.
For example, if a leaking valve is reported, an engineer could quickly locate it in the 3D view and view or navigate to all its related informa-tion, such as its location on the P&ID, its full specifi cation, mainte-nance history, spares availability and so on. Importantly, the ability to quickly collate the P&ID, the 3D location of the valve and the HAZOP analysis can help deter-mine the severity of the fault: is the fl uid water or a fl ammable liquid? What is it dripping onto? What else is in the vicinity? Are there any personnel hazards?
Maintenance management is made easier by applications that show the physical locations of current and planned work orders on a 3D representation of the facil-ity. This is a valuable means of avoiding clashes between appar-ently unrelated tasks.
Colour-coding objects in the 3D model view can support tasks such as risk-based inspection (RBI) plan-ning. By colour-coding the various lines according to, say, fl uid carried or the individual plant module, and being able to “fl y” through the virtual plant, it becomes possible to do a virtual walk-down to trace the route of a particular line, checking its proximity to adjacent objects.
Compressed air, gas and vacuum solutions
Energy eff iciency wins!
Change over to Aerzen product solutions and be a winner by reducing your energy costs! The innovative blowers, compressors and turbos from Aerzen will convince you:
• Increased energy efficiency up to 15%
• Lasting reduction of life-cycle-costs
• Guaranteed reliability provided by the leader in technology
Benefit now from the the innovative solutions from Aerzen!
For more information see www.aerzen.com
Aerzener Maschinenfabrik GmbHReherweg 28 . 31855 Aerzen . Phone +49 5154 81-0 . www.aerzen.com . [email protected]
PTQ_120x190.indd 2 15.05.2013 16:27:11 Uhr
aveva.indd 4 10/06/2013 14:15
www.hpa-usa.comwww.hpa.sg
Reactor Turnkey Services HPA offers all reactor turnkey services from blinds to blinds using state of the art equipment. We specialize in inert-entry using the latest life support units. All our catalyst technicians are trained in vessel rescue, fi rst aid, and CPR.
Hydropac Catalyst Dense Loading The Hydropac allows the sprinkling of catalyst in a continually uniform pattern at a rate slow enough to let each particle settle, but fast enough for acceptable loading time. The Hydropac sits just six inches below the trays, is able to rotate both directions, and can load around transfer tubes and other internal obstructions.
Catalyst Unloading Services Our catalyst unloading services make use of powerful vacuum units with 26 inches of vacuum.
Vessel Repairs and Retro-Fits We have our U and R stamp allowing us to do complete vessel welding, repairs, and inspection.
Hydroprocessing Associates is ISO 9001 and OHSAS 18001 accredited, and ISNetworld and PICS compliant.
Hydroprocessing Associates is located in the U.S.A. and Singapore, and can mobilize to any country.Phone: +1 832-794-7942E-mail: [email protected]
hpa.indd 1 23/2/12 12:28:06
A novel approach to cleaning furnace coils
As the refining industry moves towards heavier and dirtier crudes, attention to maintain-
ing longer run lengths for furnaces is increasingly important in reducing downtime. Large numbers of furnaces with different services and types require frequent cleaning due to fouling and coke deposition in the furnace tubes. A typical block flow diagram of refinery units and associ-ated furnaces that require frequent cleaning is shown in Figure 1.
A modified pigging operation aims to greatly reduce the time required for coke removal from furnace coils
RUPALI SAHU, SHYAM KISHORE CHOUDHARY, UGRASEN YADAV and M K E PRASAD Technip KT India
Fouling/coke formation is a func-tion of fluid composition, residence time and temperature. Crude oil’s API value and viscosity play a major role in fouling and coke formation in furnace coils. The sodium, asphaltene, Conradson carbon residue (CCR) and calcium content of the operating fluid enhances fouling/coke formation. Operating parameters including a high furnace outlet temperature, low fluid mass velocity (high film
temperatures), loss of velocity steam, uneven heat distribution (formation of hot spots inside the furnace), and fluid residence time above the cracking threshold result in fouling/coke deposition on the coils inside a furnace. Furnaces dealing with heavier process fluids — crude distillation unit furnaces, vacuum distillation unit furnaces, coker furnaces and visbreaker furnaces — are more susceptible to fouling and coke formation.
www.eptq.com PTQ Q3 2013 53
Crude furnace VGO
furnace
Coker furnace
Visbreaker furnace
Crude oil
Crude distillation
unit
Naphtha hydrotreating
unit
Catalytic reforming unit
LPG treating unit
Gasoline treating unit
VGO hydrotreating
unit
Diesel hydrotreating
unit
Fluidised catalytic
cracking unit
Hydrocracking
Coker unit
Resid processing (Visbreaker)
unit
Vacuum distillation
unit
Products
Vacuum furnace
Figure 1 Furnaces in a refinery that require frequent cleaning
or
technip.indd 1 11/06/2013 12:19
54 PTQ Q3 2013 www.eptq.com
while the heater is on-line. Spalling is carried out in one pass of a multipass heater while the other passes remain in operation. By varying the steam and boiler feed water flow rate on the fouled coil, coke breaks off the coil. This coke is then disposed of to a downstream coke drum. Thus, on-line spalling offers the advantage of allowing the furnace to continue in operation while the furnace tubes are being cleaned and has fewer environmen-tal issues than steam-air decoking. However, on-line spalling may not remove all of the coke from the coils, and other methods such as steam-air decoking and mechanical pigging are still required to bring the furnace back to start-of-run conditions. The other disadvantage of this method is that the coils are susceptible to damage due to contraction and expansion during the spalling process.
Mechanical piggingMechanical pigging eliminates the problems associated with steam-air decoking and on-line spalling, such as venting of waste gases to the atmosphere and the vulnerability of coils to rupturing due to high-tem-perature operation. Mechanical pigging is the process of propelling
The thickness of the coke deposits on the inner wall of a furnace coil can be calculated from the differ-ence between the maximum tube metal temperature (TMT) and bulk fluid temperature based on the following equation from API 530:
Tm = T
b + ∆ T
f + ∆T
f + ∆T
c + ∆T
w
Where Tm = TMTTb = bulk fluid temperature∆Tf = temperature difference across the fluid film ∆Tc = temperature difference across the deposited coke∆Tw = temperature difference across the tube wall
With this method for estimating TMT, and available empirical corre-lations, refiners can plan decoking operations for a given fluid.
Methods of furnace coil cleaningIncreased pressure drop inside the coils and high TMTs are indications of fouling inside the furnace coils. Hence, cleaning of the furnace coils is required in case any one of — or a combination of — the following conditions is encountered:• Increased pressure drop across the coils• Increased TMTs• Increased fuel consumption.
There are three generally accepted industrial practices to remove coke from the coils:• Steam-air decoking• On-line spalling• Mechanical pigging.
Steam-air decokingIn steam-air decoking, a steam and air mixture is passed through the coke deposits inside the coil walls. Shrinkage and cracking of the coke occurs by heating the coils from the outside while steam and air flow through the coils. This results in chemical reactions of hot coke, steam and air to produce CO, CO2 and H2. Although this process is more effective than the on-line spalling process, because these gases are vented to the atmosphere, it is not friendly to the environ-ment. Also, the coils are vulnerable to rupture during this procedure.
On-line spallingThe on-line spalling method was developed to increase the on-stream factors of units that process heavy and dirty feedstocks. On-line spalling is generally performed at pre-planned intervals or when high TMTs are observed in the furnace coils. Coke is removed by deliver-ing thermal shocks to the coils
Main features Steam-air decoking On-line spalling Mechanical piggingFunction Remove coke from the coils Remove coke while the furnace is on-line Remove coke from the coils. Measure coil to prolong the furnace run length prior thickness using “intelligent” pigs to pigging or decoking
Technique employed Coke is burnt off the furnace coils Coke is removed by using a high-velocity Coke is removed by pumping a metal-studded in a controlled decoking process by steam flow while thermally contracting foam or plastic pig with water in the coils, circulating an air-steam mixture at and expanding the coils mainly by scrubbing coke from inside the coils elevated temperatures
Off-line/on-line Performed while furnace is off-line Performed while the furnace is on-line Performed while the furnace is off-lineoperation
Safety concern Potential for coil rupture Potential coil damage due to No damage to coils foreseen expansion/contraction
Effluent generation/ Environmental concern as Coke is collected in coke drums Dirty water needs to be disposed offdisposal air-steam mixture is vented to atmosphere
Operating personnel Performed by operating crew Performed by operating crew using Requires an external agency/vendor to a regular decoking sequence perform pigging operations
Efficiency Removes almost all the coke May not remove all coke; decoking/ Removes all coke from the inside coils from the inside coils pigging still needed Furnace run length Fair run length (shorter than that An intermittent operation that helps in Largest furnace run length is achieved achieved by pigging) is achieved extending the furnace run length prior since efficiency is less than pigging to pigging
Comparison of different methods of coil cleaning
Table 1
technip.indd 2 07/06/2013 19:21
Reliable SwiSS Quality
API 618Rod load up to 1'500 kN/335'000 IbsPower up to 31'000 kW/42'100 HP
Full Range:
You get moRe than Just aPRocess gas comPRessoR
Lubricated up to 1'000 bar, non-lubricated up to 300 bar
For highest availability: We recom-mend our own designed, in-house engineered compressor valves and key compressor components
Designed for easy maintenance
We are the competent partner with the full range of services – worldwide
→ www.recip.com/api618
Reliable SwiSS Quality
API 618Rod load up to 1'500 kN/335'000 IbsPower up to 31'000 kW/42'100 HP
Full Range:
You get moRe than Just aPRocess gas comPRessoR
Lubricated up to 1'000 bar, non-
For highest availability: We recom-mend our own designed, in-house engineered compressor valves and key compressor components
Designed for easy maintenance
We are the competent partner with the full range of services – worldwide
YouR BeneFit:lowest liFe cYcle costs
burckhardt.indd 1 7/3/12 21:39:05
Light, warmth and much more. And our HX-Factor makes it all possible.
Our unique expertise in heat exchange (HX = HEAT EXCHANGE) lets us
keep to performance targets and ensures optimal product benefits for you.
Durability, energy efficiency and sustainability make GEA heat exchange
solutions compelling for reliable energy production or power supply. You
can depend on us day and night. In fact, whenever you switch us on.
engineering for a better world GEA Heat Exchangers
No energy production or power supply without HX-Factor
Heat exchange
with HX-Factor.
Switch us on.
GEA Heat Exchangers GmbHDorstener Str. 484, 44809 Bochum, Germany Phone: +49 234 980 2585, Fax: +49 234 980 34 2585 [email protected], www.gea.com
gea.indd 1 06/06/2013 10:24
a “pig” through a coil with the help of a pig launcher, for the purpose of cleaning or inspection of the coil. A pig is a device inserted into a pipe that travels freely through the pipe, driven by the motive fl uid. The pigging assembly consists of the pig launcher/receiver, pigs, pumps and motive fl uid storage tank. Pig launchers are temporary bores used to push the pig into the coil with the aid of water at a higher pressure. The pig launcher/receiver is placed at grade and the pig is launched into the coil some-where near the pass control valve at grade or at a suitable location at grade. The number of pigging cycles is equal to the number of passes in the furnace. Pigging removes almost all the coke from the coils. It is a faster cleaning process and comparatively longer run lengths are achieved with respect to the other cleaning processes.
Comparison of different methods of furnace coil cleaningThe main features of each cleaning method and a qualitative compari-son with respect to function, safety concerns, effi ciency and so on are shown in Table 1.
Mechanical pigging: a recent trendIn the past, the key method used was steam-air decoking, where coke was burnt off the furnace coils in a controlled decoking process by circulating an air-steam mixture at elevated temperatures. On-line spalling was developed to increase on-stream days in visbreaker and coker units. Nowadays, mechanical pigging is used to remove deposits from the coil surface by pumping a pig with water in the same manner as in offshore and onshore pipeline transportation systems. Pigging has become the preferred method in the refi ning and petrochemical industries to remove coke and scale deposits from the walls of furnace coils because it is more effective and faster than steam-air decoking and on-line spalling.
The type of pig to be selected for a specifi c pigging operation depends on the following factors:
www.eptq.com PTQ Q3 2013 57
• The purpose of pigging ■ Cleaning/decoking of coils (cleaning pigs) ■ Data recording (intelligent pigs)• Characteristics of process fl uid, properties such as CCR, heavy metal content • Driving pressure of motive fl uid• Pig velocity• Coil confi guration ■ Coil diameter ■ Length to be covered ■ Bend radius.
Pigs for cleaningA cleaning pig is a plastic foam cylinder with pins uniformly stud-ded around its surface. When these pins come into contact with the inside wall of a coil they scrape coke and other contaminants off the coil surface. As the pig travels through the coil, dirty water is routed to a collection reservoir. After the clean-ing pig has been received from the coil, the remaining coke is fl ushed out of the coil with a high fl ow of water over several hours until clear water is received from the coil. The appearance of clear water continu-ously from the coils indicates to the operator that the coils are satisfacto-
rily cleaned. As a result, effective cleaning can be done without caus-ing harmful thermal fatigue, which happens with the spalling and steam-air decoking cleaning meth-ods. Figure 2 shows the movement of a cleaning pig inside the bend of a coil during pigging.
Intelligent pigs for coil thicknesssurveyingModern, intelligent pigs are sophis-ticated instruments that vary in their technology and complexity accord-ing to intended use. These are also called smart pigs and are used for health monitoring of the coil by measuring thickness and corrosion along the coil. Cracks and corrosion are often detected using magnetic fl ux leakage pigs. Some intelligent pigs use ultrasonic devices or elec-tromagnetic acoustic transducers to detect coil deformation. These pigs consist of various built-in sensors and electronics that collect and store data while the pig is travelling in the coil. The electronics are sealed to prevent ingress of coil fl uid in the pig. Data are stored on analogue and digital tapes, solid-state memory devices, and so on. These
Foam/plastic/metal pig Studs/grits/protrusions on the pig’s
surface
Figure 2 Cleaning pig through the bend
Battery module
Sensors and recorders
Figure 3 Intelligent pig through the bend
1 2 3 4 n
To convection section
From process unit
Figure 4 n-pass furnace: arrangement of passes at grade
technip.indd 3 07/06/2013 19:21
58 PTQ Q3 2013 www.eptq.com
at the convection inlet and radiant outlet, respectively.
Conventional pigging The conventional pigging process consists of the following steps:• Using a pig launcher, the pig is inserted into one pass of the furnace• The pig travels inside the pass driven by the hydraulic force of motive fl uid• Reverse pigging is carried out using a pump; the pig travels back-wards and is received by the pig receiver• The above steps are repeated for other passes until all n passes are cleaned.
Thus, n passes in the furnace will require n times to connect, launch and receive the pig. Figure 7 shows the conventional pigging process for an n-pass furnace.
Although mechanical pigging is faster than other cleaning methods, there is scope to make it even faster. Considerable time is taken to clean each pass individually. This time taken by conventional pigging can be reduced by modifying the pigging operation as follows.
Modifi ed pigging schemeA reduction in the number of cycles in conventional pigging can be achieved by interconnecting the various passes of the furnace. The convection inlets of various passes at grade are interconnected by temporary spools; similarly, the radiant outlets of various passes are interconnected by temporary spools. This modifi ed arrangement reduces the number of pigging cycles, making this process faster than the conventional pigging process. This arrangement can:• Minimise the number of pigging cycles, in turn reducing the dura-tion of a pigging operation• Reduce the quantity of modifi ca-tion work, especially at height (radiant outlet at top)• Launch and receive pigs at grade for ease of operation• Reduce the number of pig launchers/receivers.
In this modifi ed pigging proce-dure, pigs are launched from the inlet of the convection section of the fi rst pass of a furnace. The
these, n/2 passes enter the furnace through one side of the convection section and the other n/2 passes enter the furnace through the other side. From the convection section, the coils enter the radiant section through internal or external cross-overs. The outlet of the radiant coils is at either side of the furnace. Figure 4 shows the arrangement of n passes at grade. Figures 5 and 6 show the arrangement of n passes
data can be archived for comparison with past and future inspection data. Figure 3 shows the movement of an intelligent pig inside the bend of a coil during pigging.
Pigging techniquesn-pass vertical cylindrical furnacearrangementThe inlet header for an n-pass furnace branches into n separate passes upstream of the furnace. Of
1 2 3 4 n/2
n
To convection section
n/2+1 n/2+2 n/2+3 n/2+4
From process unit
From process unit
Figure 5 n-pass furnace: arrangement of passes at the convection section inlet
1 2 3 4 n/2
n
From radiant section
n/2+1 n/2+2 n/2+3 n/2+4
Heater outlet to process unit
Heater outlet to process unit
Figure 6 n-pass furnace: arrangement of passes at the radiant section outlet
Loop-1
Pass 1
Loop-2
Pass 2
Loop-3
Pass 3
Loop-4
Pass 4
Loop-n
Pass n
Launch pig Receive pig
Figure 7 n-pass furnace: conventional pigging process
technip.indd 4 07/06/2013 19:21
Onis lnc. | One Riverway | Suite 1700 | Houston | Texas 77056 | USA1 (713) 840 6377 | [email protected] | www.onislineblind.com
www.onislineblind.com
onis.indd 1 10/06/2013 13:10
60 PTQ Q3 2013 www.eptq.com
the loop by means of hydraulic force. The pig now travels back-wards and is received from pass 1.
Precautions to be taken during pigging operationsThe furnace coil design pressure is a function of the coil operating pressure. Different furnaces have different coil design pressures. Mechanical pigging incorporates water as the driving medium. Generally, water is pumped using a manually regulated pump. In case the pump shut-off pressure exceeds the furnace coil design pressure, a pressure safety valve should be installed in the line to protect the coil from over-pressure during a pigging operation.
Thermocouples at the furnace outlet are susceptible to damage during pigging and may cause hindrance to the pig. Hence, ther-mocouples should be removed before pigging and placed in their respective positions after pigging is completed.
Incorporation of new piggingtechniques during furnace designThe following points should be taken into consideration during the design phase of a furnace to incor-porate new pigging techniques as well as to ease operations while pigging:
Step 1 Pig 1 is launched through the inlet of pass 1 at grade. The pass 1 inlet at grade should be modified temporarily for pig launchingStep 2 The pig travels the complete length of pass 1 from the inlet at grade to the radiant outlet. At the radiant outlet of pass 1, the pig enters pass 2 through an intercon-nection between the radiant outlets of passes 1 and 2 at the radiant topStep 3 The pig travels the complete length of pass 2 from the radiant outlet at the top to the inlet at grade. At the inlet of pass 2, the pig enters pass 3 through an intercon-nection between the inlets of passes 2 and 3 at gradeStep 4 The pig travels the complete length of pass 3 from the inlet at grade to the radiant outlet at the top. At the radiant outlet of pass 3, the pig enters pass 4 through an interconnection between the radiant outlets of passes 3 and 4 at the radiant topStep 5 The pig travels through a series of passes via various inter-connections at grade and at the radiant top and enters the last pass (pass n/2) in the loopStep 6 The pig travels the complete length of pass n/2 from the radiant outlet at the top to the inlet at grade. This is the end point for the pig. Here, the pig is sent back to
outlet of the radiant section of this first pass will be temporarily connected by a piping spool piece to the second pass. This second pass inlet at grade is connected to a third pass through the pipe spool at grade and so on up to pass n/2 (see Figure 8). With this arrange-ment, the pig that was launched at the inlet of the first pass of the convection section is received at n/2 pass in this loop at grade. A similar arrangement and pigging procedure is performed in the second loop from the n/2 +1 to nth pass. In this way, a pig will go through n/2 passes in a single cycle. Hence, n passes are pigged in two cycles. Figure 8 shows the pigging loops for n passes.
Steps of a modified pigging schemePigging can be carried out in two loops, either sequentially or simul-taneously. In the sequential approach, a single pig and setup is used in sequence for both Loop 1 and Loop 2. In simultaneous pigging, two pigs and their corre-sponding setups are used. The pigging steps for Loop 1 (pass 1 to pass n/2) are shown in Figure 9. The steps for Loop 2 are similar to those for Loop 1 except for the pass numbers (pass n/2+1 to pass n).
A sequential description of this procedure is:
Furnace inlet valve at grade
Radiant outlet at top
Launch pig
Receive pig
Interconnections at grade
Loop-1
Pass 1
Pass 2
Pass 3
Pass 4
Pass n/2
Interconnections at radiant top
Launch pig
Receive pig
Interconnections at grade
Loop-2
Pass n/2+1
Pass n/2+2
Pass n/2+3
Pass n/2+4
Pass n
Interconnections at radiant top
Figure 8 ‘n’ pass furnace: arrangement for modified pigging scheme
technip.indd 5 07/06/2013 19:21
www.eptq.com PTQ Q3 2013 61
3 Katala K A, Karrs M S, Advances in delayed coking heat transfer equipment, Hydrocarbon Processing, Feb 2009.4 Roberts R D, Increased reliability/reduced risk by applying intelligent pigging technology to inspect coils in process heaters, 4th Middle East NDT Conf, 2007.5 Adams J, Coker furnace - online spalling, paper presented at AFM, 2012.6 API 560, Fired heaters for general refi nery services.7 API 530, Calculation of heater tube thickness in petroleum refi neries.
Rupali Sahu is Senior Engineer, Process & Technology Department with Technip KT India Ltd. She holds a bachelor’s degree in chemical engineering from MIET, Meerut, India.Email: [email protected] Kishore Choudhary is Principal Engineer, Process & Technology Department with Technip KT India Ltd. He holds a bachelor’s degree in chemical engineering from BIT, Sindri, India. Email: [email protected] Yadav is Deputy General Manager, Refi nery & Petrochemicals, in the Process & Technology Department of Technip KT India Ltd. He holds a master’s degree in chemical engineering from HBTI, Kanpur, India. Email: [email protected] K E Prasad is Head of the Process and Technology Department with Technip KT India Ltd. He holds a bachelor’s degree in chemical engineering from Osmania University, Hyderabad, India. Email: [email protected]
pigging is the only method by which a furnace coil’s thickness can be measured.
Conventional pigging used in furnaces involves individual clean-ing of each pass. Since the pigging auxiliaries are individually connected to the inlet and exit of each pass, downtime is greater during conventional pigging. The modifi ed pigging scheme elimi-nates this problem as half the total number of passes is pigged in one go by interconnecting various passes. Thus, substantial time and labour are saved as there is no need to connect and receive pigs at each pass. By using a modifi ed pigging scheme, furnace net downtime is reduced by approximately a half to one-quarter of the time taken for conventional pigging. Reducing the cleaning time for furnace coils with a modifi ed pigging scheme increases the availability of corre-sponding process plant.
References1 Jegla Z, Design and operating aspects infl uencing fouling inside radiant coils of fi red heaters operated in crude oil distillation plants, Heat Exchanger Fouling, Jun 2011.2 Conticello R, Bernhagen P, Fired heater design & decoking techniques, NPRA Technology Forum, 2005.
• The furnace should have an even number of passes• Spare interconnection spools should be provided so that there is no need to prepare or procure these while pigging• The thermowell fl anges at the outlet of passes should be equal to the coil size so that they can be used for pass interconnections during pigging• Temperature transmitters should not be head-mounted to thermocou-ples and should be remote-mounted so that they can be removed during pigging• Suffi cient space should be provided on radiant/convection platforms to accommodate person-nel and material for safe pigging.
ConclusionsMechanical pigging is the preferred method for furnace cleaning in view of the operating run length that it provides compared to other cleaning methods. Furnace down-time for cleaning is less during mechanical pigging compared to steam-air decoking. Compared to on-line spalling, the effectiveness of mechanical pigging is higher. Mechanical pigging is the safest method of cleaning. Intelligent
Loop-1
Interconnections at gradeDirty water
Temporary interconnections
at radiant top
Pig launcher
Pig receiver
Water tank
Step 1Step 2, Pass 1Step 3, Pass 2Step 4, Pass 3Step 5Step 6, Pass n/2
Figure 9 Steps of a modifi ed pigging scheme
technip.indd 6 07/06/2013 19:22
enersul.com
enersul.com
enersul.indd 1 12/03/2013 09:45
Extending gasification runtime with antifoulant
Gasification is a mature process for converting hydro-carbons into electric power,
clean synthesis gas, fertilizers, fuels and chemicals with minimum environmental impact. Powerful antifoulants can improve runtime and profitability when fouled syngas coolers or soot scrubbers are causing bottlenecks.
Gasification is an old technology and the process has been develop-ing for over 200 years. It is an exothermic, non-catalytic reaction between hydrocarbons and a limited amount of oxygen in a highly reducing atmosphere. It was primarily used to produce town gas from coal for heating and lighting purposes. Oxygen and steam flowed through a bed of coal. A fraction of the coal was burned, producing heat to maintain the operating temperature. The carbon in the remaining coal formed a mixture of carbon monoxide and hydrogen gases by the so-called steam reactions:
Partial oxidationC
nH
m + n/2 O
2 → n CO + m/2 H
2 + Heat
Steam reactionsC
nH
m + n H
2O → n CO + (n + m/2) H
2 - Heat
CO + H2O → CO
2 + H
2 + Heat
In the 1950s, the importance of coal gasification declined because of natural gas exploration. But there was still a need for synthesis gas, and the demand for ammonia as a nitrogen fertilizer grew expo-nentially. In the late 1940s, the Texaco gasification process was developed and commercialised,
When fouling is observed in the syngas cooler, an antifoulant programme can provide a gasification unit with an extended runtime
BERTHOLD OTZISK Kurita Europe
and in 1950 commissioned for the production of ammonia. The first Texaco gasifier for oil feedstocks was introduced six years later to the market. The Shell gasification process was developed in the early 1950s, and the first gasifier for heavy fuel oil went into operation in 1956.
Meanwhile, there have been a number of technical developments with different designs and a broad range of reactor types. In most cases, the reactor type can be grouped into one of three categories:• Moving-bed gasifiers• Fluid-bed gasifiers• Entrained-flow gasifiers.
The gasification process takes place in a temperature range of 800-1800°C. The exact temperature
depends on process design and the characteristics of the feedstock. The produced gases are called synthesis gas (syngas). The detailed composi-tion of syngas may differ, depending on feedstock qualities and the applied gasification process.
Figure 1 shows the principle of gasification. Coal, natural gas, oil, refinery residual oils, petroleum coke (pet-coke), biomass or munici-pal waste are typical feedstocks for gasification. More than 250 units with the Texaco or Shell gasifica-tion process are installed worldwide. Shell’s gasification process is designed for fluid feed-stock gasifiers and the Shell coal gasification process for solids such as coal, petroleum coke and lignite.
Air-blown or oxygen-blown gasi-fier designs are used to provide oxygen to the partial oxidation process. Air-blown gasifiers are larger in size and the syngas contains significant amounts of nitrogen. Oxygen-blown gasifiers produce syngas with a large proportion of combustible gas. They operate at very high tempera-tures and can produce syngas with high purity.
The cleaned syngas can be supplied to a combined cycle gas
www.eptq.com PTQ Q3 2013 63
Gasification Purification and refining
Typical feedstocksCoal
Natural gasOil
Visbreaker residuePetroleum coke
BiomassRecycled products
Typical productsSyngas (H2 + CO)MethanolAmmoniaLiquid fuelsElectric power
Figure 1 Gasification process
Coal, natural gas, oil, refinery residual oils, pet-coke, biomass or municipal waste are typical feedstocks for gasification
kurita.indd 1 07/06/2013 19:24
64 PTQ Q3 2013 www.eptq.com
differential pressure increase are clear indicators of existing syngas cooler fouling. Typical elements found in the solids are Fe, Ni, V, Na, Ca, K, Mg, Al and Si, because in the high- temperature regions the carbon frac-tion of the particles is gasified, while the inorganic components remain as deposits. The molten V, Al, Mg, Cr and Si do not initiate the typical hydrocarbon fouling on the boiler tube as no liquid hydrocarbon compounds are formed, but they are still fouling contributors by them-selves. Severe fouling of the syngas cooler may result in an unplanned shutdown of the unit. The typical cleaning time is two to four working days, with production losses and loss of high-pressure steam. Beside additional cleaning costs, the shut-downs weaken the metallurgy when hot equipment is frequently cooled down and heated up again.
Fouling inhibitionThe injection of a suitable antifou-lant can help to extend the runtime significantly. The technical diffi-culty in this case is the fact that the antifoulant has to be injected into the feed in front of the gasifier reac-tor. Under these thermal conditions, commonly used antifoulants will decompose at once with no effect. Kurita developed an antifoulant technology for the severe process conditions of a partial oxidation unit. This technology results in an extended runtime with stable outlet temperatures and differential pressures.
Kurita AP-2505 is an antifoulant with an excellent thermal stability, which contains no water. It mini-mises the fouling potential in the waste heat reboilers by softening the deposits. This keeps the parti-cles to be transported with the syngas small, and improves the down-cooling efficiency and oper-ating time. Kurita AP-2505 contains no surface-active ingredients and has no cleaning tendencies, which is why once-formed deposits cannot be removed during normal operation. Based on experience, the antifoulant programme should be used continuously from start-up, directly after cleaning of the syngas cooler.
turbine. Integrated gasification combined cycle (IGCC) technology is one of the most promising tech-nologies to meet the most stringent emissions limits, and provides an alternative way of producing elec-tricity, steam and hydrogen for hydroprocessing facilities. The produced syngas can be used to fuel the gas turbine. This is a very efficient and economical way to produce electricity.
Figure 2 shows a typical flow scheme of the Shell gasification process. The partial oxidation of hydrocarbons takes place in the entrained-flow gasifier reactor. The oxidant is preheated and mixed with steam. The specially designed burner and reactor geometry are constructed so that the oxidant/steam mixture is intensively mixed with the preheated feedstock. The produced raw gas has a tempera-ture of about 1300-1400°C. The reactor effluent is routed to the syngas cooler (proprietary design), where the raw gas is cooled to about 320-340°C by generating high-pressure steam.
The Texaco reactor is an empty, refractory-lined vessel with a water-cooled burner design. Steam and oil are routed through a circu-larly slit surrounding the oxygen pipe. Texaco offers direct quench-ing with water or a syngas cooler to generate steam. The syngas cooler mode is used when a high CO concentration is required. The
quench removes the main part of the solids in the gas, which are extracted as soot-water slurry or “black water”. The gas is scrubbed in a Venturi scrubber and in a packed column to remove traces of soot. The raw gas is then routed to downstream units for CO shift and acid gas removal.
The produced syngas has high levels of purity with a large propor-tion of combustible gas.
Syngas coolingThe hot raw gas leaving the gasifier reactor has to be cooled before the gas can be treated for use. There are a number of syngas cooler designs available for oil gasification. Water-tube boilers or fire-tube boilers are principal designs. Syngas coolers are designed so that formed ash and soot particles are transported with the process stream and do not usually deposit in the cooler. A shorter runtime or a temperature or
Feed
Steam
Oxygen
HP steamSyngas product
BFW Soot water to soot recovery unit
Cycle water from soot recovery unit
Gasifier reactor
Syngas effluent cooler
Soot quench
Soot separator
Soot scrubber
Figure 2 Shell gasification process
Integrated gasification combined cycle technology is one of the most promising technologies to meet the most stringent emissions limits
kurita.indd 2 07/06/2013 19:24
Case study 1In a western European refi nery, a customer experienced a signifi cant outlet temperature increase in its Shell syngas cooler associated with a reduction in runtime. To have a full understanding of the fouling problem, an initial survey with an intensive data collection and infor-mation exchange between the customer and Kurita was carried out. Next, AP-2505 was injected into the feed in front of the gasifi er reactor to stabilise the syngas cooler outlet temperature. The result of the antifoulant treatment was very impressive, leading to stable oper-ating conditions. During the process, the customer asked for the antifoulant dosing pump to be stopped three times to confi rm the effect of the chemical treatment. After each stoppage, the outlet temperature of the syngas cooler increased signifi cantly.
Figure 3 shows the syngas cooler outlet temperature during the trial phase. The average dosage of AP-2505 during the trial phase was 50-100 ppm. Based on positive eval-uation of the test run, the customer decided to continue with the anti-foulant treatment.
Case study 2In another western European refi n-ery, the gasifi cation unit was suffering from short runtimes in the Shell syngas coolers. In this case, the fouling indicators were the syngas cooler’s runtime and differential pressure. Typically, the differential pressure increased exponentially after 800-900 opera-tional hours. After compilation and analysis of the process data, the Kurita AP-2505 antifoulant programme was started. During the fi rst trial phase, the differential pressure increased with a lower gradient. Adjusting and optimising the dosing rate resulted in stable operation conditions. The average dosage was 50-150 ppm.
Figure 4 shows the pressure drop at two conditions: without the anti-foulant programme and with an injection of AP-2505. The pressure drop was normalised against throughput to get comparable results. The results were impressive
www.eptq.com PTQ Q3 2013 65
and the customer decided to run the antifoulant programme continuously.
ConclusionsGasifi cation is a very cost-effective way of producing syngas and elec-tricity from coal, natural gas, oil, refi nery residual oils, petroleum coke, biomass or municipal waste. The resulting syngas can be puri-fi ed and converted into valuable products such as hydrogen, metha-nol or ammonia. When fouling is observed in the syngas cooler, the Kurita AP-2505 antifoulant programme can provide gasifi ca-tion units with an extended runtime and stable outlet tempera-tures and differential pressures.
Further reading1 Higman C, van der Burgt M, Gasifi cation, Gulf Professional Publishing, 2003, ISBN 0-7506-7707-4.2 Higman C, Eppinger M, The Zero-residue refi nery, presented at ACHEMA, Frankfurt, 8 June 1994.3 Texaco Gasifi cation Process, United States Environmental Protection Agency, Offi ce of Research and Development Cincinnati, OH 45268, EPA 540/R-94/514a, April 1995.
Berthold Otzisk is Consulting Engineer in the technical department with Kurita Europe, Viersen, Germany, focusing on refi nery and petrochemical applications. Email: [email protected]
1 7 13 19 25 31 37 43 49 55 61 67 73 79Days
330
370
390
350
310
300Out
let
tem
per
atur
e, º
C
290
Kurita AP-2505 injection occasionally stopped
Syngas cooler 2Syngas cooler 1
Figure 3 Syngas cooler outlet temperature during trial phase
020
040
060
080
010
0012
0014
0016
0018
0020
0022
0024
00
Runtime, h
3
2
1
No
rmal
ised
pre
ssur
e d
rop
, bar
0Very successful Kurita AP-2505 treatmentWithout Kurita AP-2505 treatment
Figure 4 Normalised syngas cooler pressure drop
kurita.indd 3 10/06/2013 14:16
GE Works to redefine pump efficiencies.
In processing industries, GE’s SPS pumping
systems work to increase efficiency, reduce
environmental impact and provide versatile
pumping solutions. For example, our multi-stage
centrifugal SPS pumps are a widely-used, efficient
solution for circulation, transfer boosting and
disposal applications. Our SPS systems provide
enhanced pumping efficiencies by lowering noise
and vibration levels, decreasing construction
lead times and delivering improved reliability and
extended runtimes. With GE, you have a worldwide
support system ready to provide engineering,
field service or sales support.
Take a look at our results, and see how
GE can work for you.
Call +1 713 260 6731 or e-mail [email protected]
www.geoilandgas.com/sps
GE SPS Refining Ad_PTQ_1-18-13.indd 1 1/22/13 9:50 AMge.indd 1 07/03/2013 13:37
Improving reliability and efficiency in centrifugal pumps
Even nowadays, the ubiquitous centrifugal pump — the beat-ing heart of so many processes
— can still be a weak link, where failure can have significant conse-quences. In the oil and gas processing industries, such pump failures can cause expensive repairs, potential loss of product to the atmosphere and possible process losses. One failure mode is seizure of the metal wear rings due to run-dry conditions or other off-design operation (see Figure 1).
To reduce the risk of pump seizure, maintenance shops some-times increase the gap or clearance between the wear rings. While this may minimise contact at the wear rings, it can create a new set of problems. This includes loss of effi-ciency and increased vibration due to reduced hydraulic damping forces from the wear rings, and it can also result in increased shaft deflection. In turn, that places greater wear and tear on moving parts, and can lead to more frequent seal and bearing failures. When you consider that the repair of an API-style pump costs an aver-age of €10 000, not factoring in the extra cost of potential downtime and lost production, increasing the clearance can be a very expensive “fix” to this problem.
Long-term performance incentrifugal pumpsA better solution is to replace metal wear components with advanced, non-metallic parts made of compos-ite materials, permitting much closer clearance between moving parts and enhanced run-dry capabilities should a process upset occur.
Conversion of metal pump wear parts to advanced composite materials can lead to substantial annual cost savings
GEOFF LEWIS DuPont
A well-proven, non-metallic wear component material is DuPont Vespel CR-6100 composite. Parts made from the composite are increasingly used in thousands of pumps and hundreds of refineries and petrochemical plants around the world. Companies using it include BP, ConocoPhillips, ExxonMobil, Lanxess, Shell and Sunoco.
Vespel CR-6100 composite parts
and shapes are specified for wear rings, line shaft bearings, bowl bearings, throat bushings and pressure-reducing bushings in most types of centrifugal pumps (see Table 1). Other uses include compressor valve plates, mixer and agitator bearings, gear pump idler bushings, valve seats, thrust wash-ers, mechanical seal components, wear strips and plates.
The composite has a range of
www.eptq.com PTQ Q3 2013 67
Figure 1 No-flow operation of a boiler feed water pump caused these metal wear rings to weld together, requiring extensive repairs. Replacement of metal parts with Vespel CR-6100 parts provided better operation in no-flow conditions.
Pump type Vespel CR-6100 partsOverhung and vertical inline (API pumps) Stationary* wear rings and throat bushingsSingle-stage between bearings Stationary wear rings and throat bushingsMulti-stage horizontal Stationary wear rings, throat bushings, inter-stage bushings and pressure reducing bushingsVertical Stationary wear rings, inter-stage bushings, line shaft bearings, and throat bushings
Vespel CR-6100 should be mounted in compression, which in nearly all pumps will be the stationary, case and head rings.
Pump parts convertible to Vespel CR-6100
Table 1
dupont.indd 1 07/06/2013 19:26
68 PTQ Q3 2013 www.eptq.com
CR-6100 has led to inclusion of the material in API 610, 11th edition/ISO 13709:2009, Centrifugal pumps for petroleum, petrochemical and natural gas industries. It is listed in Annex H, Table H.3 of the stand-ard, under the generic description “PFA/CF reinforced composite”.
Reduced clearance In practical terms, this means the clearance of wear ring and bushing components can potentially be reduced to values up to 50% less than the recommended minimums for metal components as specified in API 610/ISO 13709, by replacing the stationary metal wear parts with Vespel CR-6100 parts, with the rotating wear component remaining metal.
Pump efficiency increases when wear ring clearance is reduced. The tighter clearance reduces internal recirculation within the pump, reducing the power required to produce the desired flow rate. The relationship between clearance and pump efficiency is well docu-mented in industry textbooks, and has been demonstrated with field studies and on pump test stands. The conversion of pump wear parts to the composite has been reported to lead to substantial cost savings through increased pump reliability and efficiency (see Figure 2). Conversion cost payback can be less than 12 months.
The efficiency gain achieved is related to the pump’s specific speed. This is a pump design factor that shows the relationship between flow and head. Most industrial process pumps have a specific speed in the range of 10 to 40. For these pumps, an efficiency gain of 2% to 5% can be expected when clearance is reduced by 50%. For a 75 kW pump, a 4% efficiency gain produces annual savings of €2350 (assuming a typical industrial power price of €0.09 EUR/kWh). If clearance can be reduced by more than 50%, efficiency gains and savings can be significantly higher.
Effect on reliability and long-term performanceVespel CR-6100 wear rings with reduced clearance can also help to
properties that are suitable for such applications. Its coefficient of ther-mal expansion is lower than that of steel (see Table 2). The material can withstand condensate, light hydro-carbons, amines, strong acids, sour water, gas liquids, gasoline and other aggressive fluids. Beyond
this, it can operate in hostile service conditions involving high pressures and temperatures from cryogenic up to 260°C — generally beyond the capabilities of polyether ether ketone thermoplastic, which has a glass transition at 143°C.
Industry experience with Vespel
8
6
4
2
0 20 40 60 80
Eff
icie
ncy
gain
, %
Pump specific speed, ISO units
0
Field study B
Textbook gain
Test stand 2Test stand 3
Test stand 1
Field study A
Figure 2 Efficiency gain from 50% reduction in wear ring clearance1
0 0.02˝0.508mm
0.04˝1.016mm
0.06˝1.524mm
0.08˝2.032mm
0.1˝2.54mm
Lom
akin
sti
ffness
coeff
icie
nt
Wear ring clearance
Figure 3 The inverse relationship between wear ring clearance and rotor stiffness2
Thermal property Temperature Test method Direction Units Typical valuesCTE, linear 23-260°C ASTM E-831 x-y 10-6 m/m×°C 5.6 35-149°C z 300 149-204°C z 470 204-260°C z 750Heat deflection temperature at 1.82 MPa ASTM D-648 x-y °C 315 z 120Thermal conductivity 23-149°C Hot wire method W/m×K 0.7
Wear property Condition Test method Direction Units Typical valuesPV limit 1.72 MPa Internal test falex x-y MPa×m/s 9.7
Material properties of Vespel CR-6100 (x-y: plane of major fibre orientation)
Table 2
dupont.indd 2 07/06/2013 19:26
increase a pump’s reliability. As noted above, its non-seizing prop-erties and run-dry capabilities help pumps survive off-design process conditions that could cause metal components to seize.
In addition, reducing the wear ring clearance in a pump increases hydraulic stiffness and damping, resulting in lower vibration levels and less wear and tear on critical pump components such as mechan-ical seals and bearings. This is known as the Lomakin Effect, where the wear rings produce dynamic rotor stiffness in the pump. As the wear ring clearance is cut in half, the stiffness coeffi-cient from the wear rings doubles (see Figure 3).
A study by Boulden International demonstrated the long-term perfor-mance of Vespel composite wear rings monitored closely in 61 centrifugal pumps over a period of five years. In a first study phase, the number of repairs was reduced by 45%, overall vibration by 25%, and leak detection and repair (LDAR) seal leaks by 70%. The long-term study confirmed that these results were maintained for several years of operation (see Figure 4). The key conclusions of that study were: • Repairs, vibration levels, and seal leaks were all substantially reduced• Several pumps that were previ-ously troublesome have run for more than five years with zero failures• Multi-stage pumps and pumps with start/stop operation showed the most significant reliability improvements.
More reliable pumps A large process facility in Alberta, Canada, experienced operating problems with a nine-stage boiler feed water pump heavily depend-ent on 12 chrome metal wear components for rotor stability — the root cause of several pump seizures.
All stationary wear rings, inter-stage case bushings, throat and centre-stage bushings were converted to Vespel CR-6100 parts. The facility reports that the pump now meets full load production
www.eptq.com PTQ Q3 2013 69
demands, vibration and risk of metal-to-metal seizure has been greatly reduced, and amperage draw has fallen by 5%, saving the plant about $6200/y (€4570/y) per year for one pump.
A major pump manufacturer in
Asia reported front and rear side metal wear ring problems in a horizontal single-stage naphtha equivalent liquid transfer pump (see Figure 5). Wear ring clearance had been increased to avoid risk of seizure, but this caused instability, an over current problem and pump inefficiency. Since changing the wear rings to Vespel CR-6100, it has been reported that pump
power consumption was reduced from 3.67kW to 2.91kW, operation is more stable and efficiency has increased by 10.1% compared to previous efficiency data.
These efficiency and energy gains were achieved because of the enhanced dry-run capability of the composite, enabling the company to reduce wear ring clearance, with no loss of resistance to chemicals.
References1 Aronen R, The power of wear rings: efficiency, Pumps & Systems, Cahaba Media Group, USA, März, 2011.2 Aronen R, Boulden B, Russek M, Driving pump reliability forward with advanced composite wear rings, 23rd International Pump Users Symposium, Turbomachinery Laboratorium, TX, USA, S. 15 bis 19.
Geoff Lewis is Segment Lead, Energy & Material Handling Industries — DuPont Vespel parts, for DuPont de Nemours in Europe. He holds a BSc degree in physics from Birmingham University and an MBA.
60
50
40
30
20
10
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Annual re
pair
s
0
Conversions
61 pumps evaluated
Metal repairsComposite repairs
Figure 4 Annual repairs to 61 pumps evaluated before and five years after replacement of metal wear rings with Vespel composite wear rings with clearance reduced by 50%2
Figure 5 Replacing metal front and rear wear rings in a horizontal single-stage naphtha equivalent liquid transfer pump with Vespel CR-6100 parts enabled a major Asian pump manufacturer to reduce output power and increase efficiency
Reducing the wear ring clearance in a pump increases hydraulic stiffness and damping
dupont.indd 3 07/06/2013 19:26
For over 60 years, we’ve been there for our refining and petrochemical customers. Linde Engineering North America Inc. offers single source responsibility for technology, engineering, procurement and construction.
− Selas Fluid refinery and petrochemical fired heaters − Oxidation and incineration technologies − Engineered revamps and rebuilds − Hydrogen and synthesis gas plants − Air separation plants
A new name, a long history. Selas Fluid Processing is nowLinde Engineering North America Inc.
Headquarters:Five Sentry Parkway EastBlue Bell, PA 19422 USA Phone +610-834-0300
Texas Office: 3700 West Sam Houston Pkwy. South, Suite 425 Houston, TX 77042 USA Phone +281-717-9090
Al-Khobar: Linde Arabian ContractingPhone +966-3-887-0133
www.lindeus-engineering.com
linde.indd 1 07/06/2013 16:21
Oil sands-derived feed processing
The processing of oil sands- derived feeds presents inher-ent challenges to refineries.
The very nature of heavy feedstocks such as bitumen, coupled with substantive technology and opera-tional issues, often complicates effective, efficient hydroprocessing. Criterion has collaborated with Canadian operators to develop customised catalyst systems and operating strategies to improve days on stream, generate higher yields and facilitate the processing of more challenging feeds. This article details the growing perva-siveness of oil sands as a viable feedstock for global oil processing, and the lessons, challenges and opportunities gleaned by Criterion through its customised research and technical service.
The realitySynthetic oil continues to grow and supplement current oil supplies. The realities of growing global demand, reduced conventional oil supply, increased competition and tighter environmental controls drive oil producers to adapt and optimise their production sites to process synthetics, including extra-heavy or bitumen-derived crudes. While Canada and the Americas are the most abundant sources of bitumen, the raw material is availa-ble on every continent, says the US Geological Survey’s Energy Resources Program. Today, many full or partially synthetic crude oils are created by Canadian and Venezuelan producers. Canadian production originates from extra-heavy crudes (Cold Lake) and oil sands (Athabasca bitumen).
Processing crudes derived from oil sands requires a good understanding of the facility, and of the sources and processing requirements of the available crudes
MAX OVCHINNIKOV, JOSIANE GINESTRA, DORIAN RAUSCHNING, BILL GILLESPIE and KEVIN CARLSONCriterion Catalysts & Technologies
According to Canada’s Center for Energy, “oil sands deposits are in an area larger than the island of Newfoundland or the state of North Carolina. The Athabasca oil sands area, by far the largest, is the site of all surface mining projects and most in-situ extraction projects.” Venezuelan production originates primarily from extra-heavy crude from the Orinoco region. IHS CERA, an energy consulting firm, projects oil sands production will grow from 1.7 million b/d in 2012 to 3.2 million b/d in 2020. It also says that markets outside of North America, particularly Asia, hold potential for oil sands expansion, with China expected to nearly double its 10 million b/d refining capacity by 2030.
Maximising the opportunities ofheavy oil/bitumenUpgrading these heavy feeds is typically accomplished by employ-ing a combination of hydrogen addition such as hydrocracking, and carbon rejection such as coking,
together with many other subse-quent hydrotreating steps. As production facilities have been developed and expanded with a variety of technologies, the result is synthetic crudes that typically fall into three categories (see Figure 1):• Fully synthetic: ranging from light, bottomless sweet crudes to full-range partially treated blends• Synbit: a combination of synthetic/heavy blends• Dilbit: simply diluted heavy feeds.
Processing heavy oil and bitu-men-derived feeds increases global oil supply, delivers market-ready fuels that are clean and environ-mentally responsible, and delivers a direct economic benefit to produc-ers. Many facilities have increased the use of these types of feeds in order to maximise asset utilisation and facility margins. This often requires increased process and technology capabilities in a number of refining process units, particu-larly hydroprocessing assets. The magnitude of impact depends on the feed processed, the inherent
www.eptq.com PTQ Q3 2013 71
Bitumen and heavy crude production
MiningSAGDHASD
Conventional
Upgrading
H2 additionCarbon rejection
“Synthetic” full range and bottomless
Synthetic blends
Coke and sulphur
Diluted crudeDiluted bitumen
Condensate and diluent production
Recovered diluent
Diluent
Figure 1 Routes to synthetic crude production
cri.indd 1 10/06/2013 16:26
72 PTQ Q3 2013 www.eptq.com
heat exchangers, premature catalyst deactivation and possible complete equipment failure.
Through collaboration with numerous upgrader and refi nery operators, Criterion has deciphered the DNA code of bitumen and how refi neries — with minimal adapta-tion and investment — can complement their existing opera-tions with bitumen feedstock processing capabilities. The key is collaboration.
Development of an improved guard catalyst for arsenic removalArsenic is naturally present in many petroleum feedstocks. Although the concentration of arse-nic is low in most petroleum feeds, some crude oils from Taching (China), Venezuela, certain Russian crudes and Athabasca bitumen contain high levels of arsenic (see Figure 2).1 Some shale oil deposits also exhibit high levels of arsenic, in particular the Green River area of Colorado, Wyoming, and Utah.
Species containing arsenic can be found in a wide distillation range of petroleum feedstocks, but in general the highest concentration of arsenic is found in the naphtha boiling point range. For naphtha feedstocks, it is not uncommon to see arsenic concentration in the 10-50 wppb range, and as high as 600-700 wppb in extreme cases. The predominant types of arsenic- containing species are aryl- and alkyl-partially oxygenated arsines, and the typical structures of these compounds are shown in Figure 3. In some crudes the source of arse-nic is purely inorganic, which is sometimes indirectly evidenced by the absence of zinc in a crude.2 The exact molecular composition of arsenic-containing species is highly dependent on the geographical and geological source of crude oil, and the precise molecular mapping of arsenic compounds is often compli-cated by a low concentration of arsenic compounds with respect to the detection limits of speciation analytical methods.3
Arsenic is the most potent poison for hydrotreating catalysts by a wide margin: the deposition of only 0.1 wt% arsenic on a catalyst
fl exibility of the facility operations and the ability to offset this increased processing requirement.
Maximising for synthetics: facing the challengesProcessing heavy oil is not always easy. When compared to most conventional crude oil processing and conversion, a bitumen-derived crude is a greater challenge, particularly to hydroprocessing unit operations. Diffi culty arises due to inherently higher concentra-tions of sulphur, nitrogen and aromatic contents; high levels of metal contaminants such as nickel and vanadium, which are detri-mental to catalyst performance; and increased levels of asphaltene and arsenic contaminants. Asphaltenes can foul process equipment and lead to an unexpected pressure drop, while arsenic will rapidly deactivate catalysts even if at very low levels.
When it comes to heavy crudes, variations within the crude proper-ties themselves are expansive, usually because of the different technologies and operational
practices applied during produc-tion and upgrading. In some cases, heavy crude properties may vary from shipment to shipment because of blending approaches or opera-tional changes in their production. While these changes may not seem signifi cant on a bulk property basis, they become quite signifi cant —often problematic — for downstream processing units at a facility. Even within the various types of crude — sweet, sour and bottomless — there may be signifi -cant variations, usually caused by how it was produced, blended and upgraded.
Ultimately, one must be aware of the operational impact of these heavy feedstocks. This requires an improved understanding of the streams that are both directly recovered from the fractionation of the refi nery feeds and those from upstream units. Major operational adjustments may be required for a facility to maintain targeted opera-tions. As a result of these numerous issues, hydroprocessing becomes a seemingly complex and expensive endeavour that can result in fouling
Alberta, Canada
Green River shale
Venezuela
US West Coast
WTI
0.001 0.01 0.1 1 10 100
Arsenic, wppm
Figure 2 Examples of arsenic content in various crude oils
As
O
OHOH
As
CH3
CxH2x+1
As
Phenylarsonic acid Methylalkylphenylarsine Triphenylarsine
Figure 3 Structures of typical arsenic-containing compounds found in crudes
cri.indd 2 10/06/2013 16:39
Aim for More Diesel, Better DieselExxonMobil’s MIDWTM* technology is a commercially proven process using an advanced proprietary catalyst for the production of ultra-low sulfur and low cloud-point diesel.
Benefits include: • Low-pressure,fixedbedprocess—improvesproduct yieldandquality
• Increasedsulfurremovalandreducednaphthayields
• Proprietarycatalyst**isomerizeswaxyn-paraffinsin gasoil,toleranttoH2S,ammonia,andorganicSandN
• Stand-aloneorintegratewithothers
* MIDW is a trademark of ExxonMobil Research and Engineering Company** Available from ExxonMobil Catalyst Technologies LLC
ExxonMobilResearchandEngineering
Company’sMIDW*Technology
ImprovesProductYieldandQuality
3225 Gallows Road, Fairfax, Virginia 22037-0001, USAwww.exxonmobil.com/refiningtechnologies • +1-703-846-2568 • fax +1-703-846-3872 • [email protected]
exxon.indd 1 07/06/2013 09:56
typically results in a loss of hydrodesulphurisation (HDS) and hydrodenitrogenation (HDN) activ-ities as high as 50%. The deposition of arsenic on a hydrotreating cata-lyst is a two-step process: organo-arsenic species undergo C-As bond hydrogenolysis followed by the reaction of reduced arsenic intermediates with a promoter metal of a hydrotreating catalyst, nickel or cobalt, to form the corresponding metal arsenide. For instance, nickel arsenide NiAs is the thermodynamically favoura-ble product, although the formation of other stochiometries (Ni11As4, Ni5As2) was documented at lower arsenic concentrations.4
74 PTQ Q3 2013 www.eptq.com
Arsenic reacts with the promoter metal, nickel or cobalt, of the Ni(Co)-Mo-S phase presumably through adsorption at coordinately unsaturated metal sites (see Figure 4). Môssbauer spectroscopy studies suggest that those interactions take place without destruction of the Ni(Co)- Mo-S phase, while altering its electronic structure and, respec-tively, resulting in substantial loss in hydrotreating catalytic activity.5
Subsequently, arsenic reacts with the bulk Ni(Co)S phase to form the aforementioned metal arsenides.
In a commercial reactor, the arse-nic accumulation rate is often diffi cult to determine due to a lack of accurate or consistent arsenic
measurements in the feed. However, arsenic catalyst poison-ing will manifest itself through a signifi cant loss in HDS and HDN activities and a shift in the reactor axial temperature profi le of a reac-tor due to reduced hydrogenation activity of the arsenic-poisoned top layers (see Figure 5).
Poisoning of hydrotreating cata-lysts with arsenic was historically overlooked, in no small part due to the lack of fundamental under-standing, and was accepted as the “normal” part of general catalyst deactivation. Development of high-performance hydrotreating catalysts, driven by more stringent environmental regulations and economic necessity to process diffi -cult crudes, prompted refi ners and catalyst suppliers to look for dedi-cated arsenic abatement solutions in order to protect a main catalyst bed and catalysts in post-treat reac-tors from arsenic poisoning.
To address this challenge, Criterion launched a dedicated arsenic guard catalyst with high HDS and HDN activities, Arsenix, in 2004.6 This catalyst featured high levels of arsenic capacity achieved through a signifi cant increase in reactive nickel sites. Conscious optimisation of the Mo:Ni ratio resulted in initial HDS/HDN activ-ities comparable to conventional NiMo catalysts and lower HDS/HDN activity decline rates. Further incremental improvements to Arsenix led to the commercial launch of the second-generation arsenic guard catalyst, MaxTrap [As], in 2006. This featured broader applicability within refi nery appli-cations ranging from naphtha to heavy vacuum gas oil hydrotreat-ing and fl exible deployment as a top layer trap or full reactor load. Since 2004, both Arsenix and MaxTrap[As] were installed in over 150 commercial applications.
A joint R&D programme with a North American refi ner to develop a new guard catalyst for managing arsenic in naphtha-kero feed blends derived from oil sands has relied heavily on multi-cycle catalyst basket studies in commercial naphtha-kero hydrotreaters complemented by a dedicated
Primary site of arsenic adsorption
Ni
Figure 4 Model representation of Ni-Mo-S phase
Axi
al t
em
pera
ture
0 10 20 30 40 50 60 70 80 90 100
Axial position, % of catalyst bed
1.5
2.5
3.0
2.0
1.0
0.5
As
load
ing,
lb/f
t3
0
As contamination at end of runTemperature at end of runTemperature at start of run
Figure 5 Correlation between arsenic deposition profi le and reactor temperature at the end of run
cri.indd 3 10/06/2013 16:39
ThyssenKrupp Uhde
ThyssenKrupp Uhde –Engineering with ideas.The key to our success is the creativity and resourcefulness of our employees. And it is this that keeps turning major challenges into solutions that are not only brilliant and innovative, but often set the standard for the entire engineering sector.
www.uhde.eu
t krupp.indd 1 06/06/2013 14:41
76 PTQ Q3 2013 www.eptq.com
exhibits better and more stable HDS and HDN activities, as is evident from pilot plant testing using 100% coker naphtha feed-stock (see Figure 7).
MaxTrap[As]syn represents a substantial improvement in arsenic capture activity over previous generations of guard catalysts for a wide range of hydrotreating appli-cations. The first commercial deployment of the catalyst is sched-uled for the first half of 2013.
Low-cost hydrogen meets low-value feeds With natural gas production hitting unprecedented highs, the cost of both natural gas fuel and hydrogen production have dropped to the lowest cost levels in the past decade. Utilising low-cost hydro-gen increases liquid volume yields and increases the production of higher-value products.
Advances in catalyst technology employed during the global clean fuels initiatives have resulted in many “original” hydroprocessing unit designs with under-utilised capabilities, many of which can be exploited with drop-in solutions during a catalyst change-out to achieve improved aromatics satura-tion, selective ring opening (SRO), conversion and cold flow improve-ment benefits (see Figure 8).
Criterion can effectively offer additional reactor volume to provide further hydrogen utilisa-tion upgrades and extend days on stream. Examples of the benefits available include:• Aromatic saturation can further enable density upgrades, smoke point improvements, feed difficulty capability and FCC yield gains. Technologies such as enhanced aromatics saturation (EAS) provide volume expansion with improve-ments in distillate cetane and aromatics contents while maximis-ing distillate yields• SRO will further improve density and volume swell while improving diesel cetane. SRO maximises cetane improvement per unit of hydrogen consumption• Conversion via mild hydrocrack-ing (distillate, VGO, resid) can minimise lower-value fuel yields,
arsenic uptake capacity compared to MaxTrap[As] while providing >50% better utilisation of nickel for arsenic capture, as evidenced from the As:Ni molar ratio in spent cata-lyst shown in Figure 6. This was achieved by a novel catalyst manu-facturing process, which greatly enhances the kinetics of arsenic capture. Moreover, the new catalyst
catalyst pilot plant testing program and comprehensive characterisation studies of fresh and spent catalysts.
This collaboration resulted in the MaxTrap[As]syn catalyst. It is a breakthrough step forwards both in terms of the overall arsenic capacity and utilisation of nickel in the cata-lyst. MaxTrap[As]syn exhibits a step-out 70% better volumetric
Nickel utilisation efficiency for arsenic captureVolumetric arsenic uptake0.5
0.6
0.4
0.3
As:
Ni ra
tio in s
pent
cata
lyst
(n
ickel uti
lisati
on)
0.2
140
180
100
60
Volu
metr
ic a
rsenic
upta
ke,
%
20MaxTrap[As] MaxTrap[As]syn
Figure 6 Arsenic capture and nickel utilisation efficiency
0 100 200 300 400 500
350
370
365
360
355
345
340
335
330
325Tem
pera
ture
requir
ed
for
5ppm
N,
ºC
320
MaxTrap[As]syn
MaxTrap[As]
HDN activity
0 100 200 300 400 500
350
370
365
360
355
345
340
335
330
325Tem
pera
ture
req
uir
ed
for
5p
pm
S,
ºC
320
MaxTrap[As]syn
MaxTrap[As]
HDS activity
Figure 7 Hydrotreating activity of MaxTrap[As]Syn catalyst
cri.indd 4 10/06/2013 16:26
cat tech1.indd 1 23/2/12 12:04:42cat tech.indd 1 10/06/2013 16:35
78 PTQ Q3 2013 www.eptq.com
are in optimal condition, can allow the level of performance to be pushed while maintaining opera-tional reliability and fl exibility.
Determining the optimal targeted operation and process scope requires an evaluation of catalyst system designs with their predicted yields slates and cycle life esti-mates. All of these are largely dependent on the individual unit design and confi guration, feed types and contaminant level, as well as on the scope of the proposed revamp (see Figure 10).
In specifying a MHC catalyst system, the balance of hydrotreat-ing versus cracking catalyst and the potential addition of reactor volume is largely infl uenced by feed qualities and desired level of conversion. As many of the feeds processed are high in contaminant metals, sulphur and nitrogen, the pretreat section is required to remove these contaminants to ensure a suffi cient cycle life can be maintained while both meeting product targets and minimising nitrogen slip into the cracking section of the reactor.
Feed quality, reactor and speci-fi ed catalyst system determine the ultimate S and N removal capabil-ity for a given cycle life; HDS functionality can remain an impor-tant criteria for some MHC units depending on existing product specifi cations that are dependent
the hydrogen utilisation and volume expansion achieved to be tailored to specifi c refi ner capabili-ties and objectives (see Figure 9).
Beyond the gains we may be able to achieve with drop-in catalytic solutions, further capabilities can often be realised by combining these options with other process improve-ments that can fi t within a normal turnaround window. Improving reactor internals and instrumenta-tion/control, and ensuring the recovery and fractionation sections
increase refi nery heavy oil process-ing capacity, and improve product quality and margins. MHC offers the maximum volume expansion opportunity. Implementing MHC into an existing unit requires an evaluation of unit light handling, quench, hydrogen supply and other capabilities to ensure compatibility with conversion operations• Shape-selective paraffi n cracking and isomerisation will improve distillate cold fl ow properties.
The available technologies allow
CompositionCon
di
tions
Chem
istry Cataly
sts
Low cost H2
Selective ring opening
Cracking
Aromatic saturation
Isomerisation
Cetane improvement
Density improvement
Renewable diesel
Kero/jet Smoke/aromatics
Synthetics upgrading
Dewaxing cold flow improvement
Conversion to naphtha
Conversion LVGO to diesel
Latest generation
Latest generation
Increased H2 consumption
Increased H2 consumption
Design catalyst
Design catalyst
Design catalyst
Design catalyst
Latest generation
Aromatic saturation
Selective ring opening
Conversion
Isomerisation
Current system
Future catalyst system
Upgrading opportunities
Figure 8 Hydroprocessing unit designs with under-utilised capabilities can be exploited with drop-in solutions during a catalyst change-out
Diesel
VGO
ULSD
Boiling point conversion
Volume expansion
VGO
EAS
ASAT
SRO
HDW
LHC
MHC
HCU
HT
Aromatic conversion
Aromatic saturation
Selective ring opening
Hydrocracking
R R’
+R’
2H23H2
H2
H2
R
Figure 9 Routes to low-cost hydrogen
cri.indd 5 10/06/2013 16:26
www.eptq.com PTQ Q3 2013 79
course, the possibility to extend operation into very high volume yield ranges if desired. Unlike its predecessor, this very active system
erably extended, while the proper turndown control is maintained.
The product composition is shifted only slightly, with, of
on refinery constraints and capabil-ities. However, HDN capability is often more important as it influences cracking catalyst selec-tion and performance due to remaining nitrogen heteroatoms, reducing cracking reactions.
This case study illustrates how Criterion’s newest generation of HDN and cracking catalysts can be utilised to enhance the performance of an existing LGO hydrotreater that processes oil sands-derived coker LGO. Through the use of Criterion’s Centera and the newest Zeolyst cracking catalyst technolo-gies, the unit’s HDN, aromatics saturation and cracking activity have been increased while lowering unit fill cost.
An extensive pilot plant evalua-tion of catalyst systems was carried out to scope the options and care-fully assess performance benefits and associated changes in product composition. The predicted impact of these changes on downstream equipment was discussed to ensure that smooth operation could be maintained within all the opera-tional constraints (exotherm and quench capabilities, hydrogen consumption, downstream separa-tion and fractionation hydraulic constraints, and so on). The main objective was to provide increased unit flexibility, in terms of feed quality and targeted volume yield, to allow for better overall refinery optimisation.
The selected catalyst system consisted of Centera DN-3630 and a new-generation distillate-selective Zeolyst hydrocracking catalyst, Z-2513. This system exhibits an activity benefit of more than 20°C over the old-generation incumbent system, both in the pilot plant stud-ies and in commercial operation. This huge activity benefit can trans-late into increased run length, but ultimately the incremental activity can be used for more conversion and overall liquid volume yield. Volume yield is improved, even at a constant conversion level, and the amount of gas produced (C1 to C4s), at a constant volume yield, is significantly lower with the new system (see Figure 11). The range of achievable volume yields is consid-
VGO H2
HDN
HDN
HDN
MHC
Diesel
Naphtha
Off gas
HDT VGO
108
112
110
106
104
Volu
me y
ield
, %
102
109
111
107
105
103
0 10 20 30 40 50 60 70 80 90 100
Conversion of the heavy fraction, wt%
DN-3630/Z-2513Incumbent system
6
7
5
4
3
2
1
C1-C
4,
wt%
0100 105 110 115
Volume yield, %
DN-3630/Z-2513Incumbent system
Figure 11 Volume yield is improved and the amount of gas produced is significantly lower
Figure 10 Determining the optimal targeted operation and process scope requires an evaluation of catalyst system designs
cri.indd 6 10/06/2013 16:27
80 PTQ Q3 2013 www.eptq.com
derived LGO feedstock. A small layer of cracking catalyst can also provide a solid volume yield boost in the same application. Such a catalyst system design, using a stack of DN-200 and Z-2513, has also been commercially demon-strated (see Figure 13).
The combination of a Centera catalyst and a distillate-selective cracking catalyst such as Z-2513 is an obvious alternative for such cases. For each unit, the fi nal cata-lyst decision has to be based on a cost/performance analysis after the detailed performance prediction is understood for various catalyst systems.
Increased capability withbitumen-derived feedsCatalysts applied in FCC feed pretreat (FCC PT) service must operate robustly with the wide range of feeds charged to these units. These feed origins include crude atmospheric and vacuum towers, cokers, Rose/deasphalters, fl uid catalytic crackers, lube units and others. FCC PT units often process bitumen-derived feeds. Catalysts that operate in FCC PT service must obtain optimised sulphur, nitrogen and aromatic satu-ration performance to drive the FCC economic performance and be able to operate stably in an environment where signifi cant levels of contaminants such as nickel, vana-dium, silicon, sodium, arsenic and asphaltenes are present. The new-generation Centera technology catalyst DN-3651 combines the commercially demonstrated stability and contaminant-tolerance features of Ascent DN-3551, a NiMo FCC PT catalyst, with Centera active site architecture, allowing for step-out HDS, HDN and aromatics saturation (ASAT) performance, plus high catalyst stability in FCC PT applications.
A signifi cant aspect of the DN-3651 development programme is that the catalyst was developed specifi cally for FCC PT applications utilising high throughput experi-mentation while targeting improvements in challenging synthetic feeds. The high through-put reactor system utilises multiple
The predicted incremental yield associated with this system was realised in the commercial unit.
Opportunities also exist for less powerful units at medium pressure on straight-run or vacuum-derived light gas oils. Centera catalysts on their own offer a volume yield advantage over conventional catalysts, as observed with a straight-run Athabasca bitumen-
allows operation at temperatures below the thermodynamic equilib-rium temperature, with excellent volume yields as well as HDN and HDS conversions. When operated above the equilibrium temperature, a higher level of saturation is main-tained, thanks to the ring opening and cracking activity of Z-2513, which contributes to the excep-tional yields (see Figure 12).
96
100
99
98
97
95
94
93
92
91
Satu
rate
by
SFC
, %
90Temperature
DN-3630/Z-2513Incumbent system
Thermodynamic aromatic saturation equilibrium
Figure 12 Above the equilibrium temperature, a higher level of saturation is main tained
DN200DN200/Z-2513
0.3
0.5
0.6
0.4
0.2
0.1Volu
me y
ield
gai
n,
%
0DN-3110 DN-3630
4.0
5.0
4.5
3.5
3.0
2.5
2.0
1.5
1.0
0.5Rela
tive
liq
uid
volu
me y
ield
gain
s, w
t%
0Temperature
Figure 13 Volume yield advantage with a straight-run Athabasca bitumen-derived LGOfeedstock
cri.indd 7 10/06/2013 16:38
www.eptq.com PTQ Q3 2013 81
customer conveyed the need to process more oil sands-derived coker gas oil. Consequently, the fi rst initiative was to plan a transi-tion to a more metals-tolerant catalyst system, and Ascent DN-3551 catalyst was selected for Cycle 2. This catalyst has been shown to have superior feed poison tolerance and was considered to be better suited to accommodate the expected poisons that would be associated with processing incre-mentally higher percentage of oil sands-derived coker gas oils.
The second initiative included optimisation of the demet catalyst package to enhance catalyst stabil-ity. Prior to Cycle 1’s completion and the DN-3551 loading, we advised the customer that the opera-tion was being subjected to a higher feed endpoint and the catalyst system was at risk of accelerated aging due to feed resid and metals entrainment. Subsequent detailed feed component analysis at Criterion’s Research facility confi rmed elevated feed poisons
by DN-3551. The points enclosed in the highlighted areas labelled DN-3651 represent samples tested as part of scale-up and manufactur-ing optimisation. Use of high throughput experimentation allowed rapid development of a reliable manufacturing process yielding the highest practical cata-lyst performance.
Applying technology improve-ments such as DN-3651 requires a holistic joint approach to ensure that the full benefi ts of the technol-ogy can be taken advantage of by the operator, taking into account the site-specifi c constraints and objectives.
In this example, Criterion utilised a collaborative approach with its customer to improve catalyst performance and facilitate more diffi cult oil sands-derived feed processing capabilities. Three initia-tives were implemented.
The original cycle 1 utilised Criterion’s Centinel DN-3100 cata-lyst as the primary conversion system. During the cycle, the
tubular fl ow reactors with auto-mated process control and sampling. The use of this multi-tube reactor system allows signifi cant accelera-tion of catalyst development relative to conventional testing techniques. Leads generated with the high throughput equipment were confi rmed by conventional-scale pilot plant testing.
The target performance level for DN-3651 was an improvement in HDN activity of at least 10°F (5.5°C) relative to DN-3551 (~20 RVA) with equivalent or better HDS activity with a challenging design feed containing bitumen-derived vacuum gas oil and heavy coker gas oil. Figure 14 illustrates the high throughput experimenta-tion HDN and HDS activity data obtained during the prototype development phase as well as during the scale-up and commer-cialisation phase. The activity as presented in these plots is the decrease in temperature required to achieve the target S or N level in the product relative to that required
Kurita Europe GmbH . Industriering 43 . 41751 Viersen . Germany Phone:+49 2162 9580110 . www.kurita.de
We keep a close eye on your plant efficiency.The trouble-free operation of your plant is the precondition for maximum economic efficiency. Kurita’s patented ACF technology inhibits the formation
of chloride and ammonium salts, thereby preventing fouling and corrosion. This is only one example of how Kurita protects your production unit against
unscheduled shutdowns, extends its runtime and lowers process costs – while at the same time increasing plant and operational safety. Particularly
for refineries and petrochemical plants, Kurita offers innovative and sustainable process technologies as well as a broad portfolio of products for water
management. From consultation through to implementation and maintenance, we support you with a comprehensive range of services.
As you can see, we keep a close eye on the effcient operation of your plant.
Process technologies for refineries and petrochemical plants
Water treatment technologies for cooling water, boiler water, waste water and reverse osmosis
RZ_Anzeige_Quer_24052013_MQ.indd 1 24.05.13 15:24
cri.indd 8 10/06/2013 16:38
82 PTQ Q3 2013 www.eptq.com
whereby feed poisons could be routinely monitored and, in conjunction with catalyst perfor-mance data, projections could be performed on catalyst metals accumulation to plan/manage top-bed skims. The data were also used to further optimise the top-bed demetallisation package. Once again, the collaborative approach facilitated a proactive identification of a catalyst deactiva-tion threat and the implementation of a mitigating strategy.
The third initiative focused on improving SOR catalyst activity in order to increase cycle length and/or facilitate more oil sands-derived coker gas oil processing. A pilot plant programme was developed using the unit’s current feed, as well as an alternate feed with a higher percentage of oil sands- derived coker gas oil, to compare the catalyst activities of the incum-bent DN-3551 and Criterion’s latest-generation Centera DN-3651 catalysts. Activity measurements confirmed that DN-3651 is ~15-20°F more active than DN-3551. In addition, the pilot plant’s feed sensitivity study quantified the incremental temperature requirement if the higher oil sands coker gas oil feed was to be processed. The relative HDS temper-atures are shown in Figure 15.
Given the unit’s existing activity constraints and the desire to increase the oil sands-derived coker gas oil content, the customer elected to utilise DN-3651 as part of its next catalyst change-out along with the further optimised demet catalyst package. The trend of proactive collaboration between customer and supplier demon-strates how Criterion’s technical service and catalyst developments can be leveraged to facilitate improved hydroprocessing opera-tion with oil sands-derived feedstocks.
Forward momentum continuesMany refiners have upgraded their capability to reliably and efficiently process and utilise these often difficult-to-treat feeds while main-taining plant reliability and overall operational objectives. This often
During Cycle 2’s DN-3551 opera-tion, Criterion provided further assistance in assessing the hydro-processing unit’s feed poison content to ensure catalyst stability is not compromised by undetected events. A rigorous Cycle 1 DN-3100 spent catalyst analysis programme was proposed and subsequent anal-ysis confirmed the presence of feed poisons. The analysis helped vali-date the conclusions that the prior cycle DN-3100 catalyst system was, indeed, subjected to resid entrain-ment and contained the expected poisons. However, other catalyst contaminants were also detected that could not be accounted for by just having resid entrainment. Further analysis suggested that the contaminant was likely a part of the regular feed diet. As such, a comprehensive feed poison analyti-cal programme was developed,
and that the magnitude of change was greater than what would be expected with just incrementally more coker gas oil. As such, and prior to the end of Cycle 1, the subsequent DN-3551 cycle’s demet catalyst package was modified to further enhance the reactor’s toler-ance for feed metals ingress and potential resid entrainment. The collaborative approach used during Cycle 1’s DN-3100 loading facili-tated proactive identification of a catalyst risk and the implementation of a drop-in catalyst solution to miti-gate future cycle deactivation risk. The enhanced demet package, along with the DN-3551 catalyst system, was loaded and delivered a cycle that has met the minimum cycle length target, but with an average cycle lower feed API, higher feed sulphur and lower product sulphur content.
10
30
25
20
15
5
0
−5
−10
−15
Acti
vity
im
pro
vem
ent
rela
tive
to D
N-3
551,º
F
−20
Acti
vity
im
pro
vem
ent
rela
tive
to D
N-3
551,º
C
1
91113
753
−1−3−5−7−9
−110 50 100 150 200 250
Catalyst sample number
Approach 4Approach 1 Approach 2
Approach 5Approach 3
DN-3651 development samples DN-3651
10
30
25
20
15
5
0
−5
−10
−15
Acti
vity
im
pro
vem
ent
rela
tive
to D
N-3
551,º
F
−20
1
91113
753
−1−3−5−7−9
Acti
vity
im
pro
vem
ent
rela
tive
to D
N-3
551,º
C
−110 50 100 150 200 250
Catalyst sample number
DN-3651
Approach 4Approach 1 Approach 2
Approach 5Approach 3
DN-3651 development samples
Figure 14 DN-3651 HDN & HDS activity improvement: high throughput experimentation
cri.indd 9 10/06/2013 16:38
Max Ovchinnikov is a Senior Research Chemist with Criterion Catalysts and Technologies based in Houston, Texas. He is primarily engaged in the research and development of catalysts for hydroprocessing applications and has 12 years of experience in heterogeneous catalysis and refining technologies. He has co-authored over 20 technical publications and holds a PhD degree in organic chemistry from Iowa State University.
www.eptq.com PTQ Q3 2013 83
2 Dekkers C, Daane R, Oil & Gas J., 1999, 97, 145.3 Puri B K, Irgolic K J, Environ. Geochem. Health, 1989, 11, 95.4 Nielse B, Villadsen, Appl. Cat., 1984, 11, 123.5 (a) Internal communication, Criterion Catalysts & Technologies; (b) Merryfield R N, Gardner L E, Parks G D, Catalyst Characterization Science, ACS Symposium Series 1985, 1.6 Bhan O K, Arsenic removal catalyst and method for making same, US Patent 6759364.
requires increased process unit capabilities in a number of refining process units, particularity hydro-processing assets. The magnitude of impact, of course, depends on the particular feed processed, the inherent flexibility of the facility’s operations and the ability to offset this increased processing requirement.
In making this move, proper selection requires a good under-standing and modelling of the facility as well as a strong knowl-edge of the sources and processing requirements of the available crudes. With the continued devel-opment and increased availability of such feeds, this knowledge needs to be continuously updated to ensure minimal operational surprises. Reviewing TBP curves and bulk properties of whole crude or even the individual cut-point ranges does not necessarily characterise the feed difficulty or indicate the full impact on the individual processing units.
References1 Henke K R, Arsenic, Environmental Chemistry, Health Threats and Waste Treatment, Wiley, 2009, 186.
30
50
60
40
20
10
Norm
alis
ed
HD
S t
em
pera
ture
, ºF
Base
Alternative feed Current feed
Temp. 1 Temp. 2 Temp. 3Temp. 3Temp. 2 Temp. 1 check-back
Temp. 1
Catalyst age
DN-3651 alternative feed DN-3551 alternative feedDN-3651 current GO DN-3551 current GO
Figure 15 DN-3651 and DN-3551 normalised HDS temperature
increased water in the sulphur plant feed and thus more condensa-tion in the sulphur plant feed NH3 gas knock-out drum (and thus water that has to be recycled back to the sour water stripper).
The design shown in Figure 2 was common in the 1960s. However, it also suffers from the same heat balance drawbacks and needless complications as seen in
www.eptq.com PTQ Q2 2013 101
Figure 1. In other words, a lot of equipment is added to generate refl ux when no fractionation is required between the feed and overhead product. Again, the only purpose of the tower is to strip out the NH3 and the H2S.
Correct stripper designIn 1969, while working for the now vanished Amoco International Oil
(1)
(8)
(15)
Sour water feed
90ºF
Low NH3 water to desalters
Steam
250ºF
E-1
P-1
13 PSIG
Ammonia gas to sulphur plant
190ºF
Medium content NH3 water to hydrotreaters
Figure 4 Two-stage sour water stripper design without feed preheat
Company in the UK, I designed a sour water stripper that eliminated the unnecessary features of the unit shown in Figure 2.
Figure 3 shows the essentials of a correct sour water stripper design. Feed is brought in at ambient conditions (70-100°F, 21-38°C) from the sour water feed tank. To heat the feed from 90°F (32°C) to 250°F (120°C) requires about 16 wt% steam fl ow, or about 1.3-1.4lb of steam per gallon of stripper bottoms, which is close to a typical design stripping steam ratio for sour water strippers. The E-1 feed preheater, refl ux pump (P-2) and the refl ux cooler (E-2) shown in Figure 2 are all eliminated. How, then, does one know that the design shown in Figure 3 will work? Because it was built this way (at the Amoco refi nery in Milford Haven, Wales, UK) in 1970, where it worked just as well as the conventional design shown in Figure 2.
Two-stage sour water stripperFigure 4 shows a sour water strip-per with a side draw-off. The partly stripped sour water is extracted from tray 8 and directed to the hydrotreaters for use as make-up water in the salt (NH4HS) removal step of the reactor effl uent. Completely stripped water from the sour water stripper bottoms is sent to the crude desalter. While
102 PTQ Q1 2013 www.eptq.com
communicate where the module will be installed on the plot plan. Connections between the modules are designed to be similar in config-uration so that construction is relatively straightforward. Ventech estimates that, with modularisation, approximately 70% of a project is already complete even before the modules are shipped from their facility. This greatly decreases field construction time to deliver an operational facility (see Figure 1).
These methods also facilitate easy disassembly and relocation, if necessary, at some point in the future. For example, a remotely located gas processing facility could be easily taken apart and moved to a new natural gas source if an exist-ing supply was depleted in its current location.
Applying modularisation to refin-ery construction has advantages with regard to productivity, prod-uct quality and ensuring the safety of construction personnel. Since the modules are built in a well-lit, climate-controlled environment, work can continue around the clock regardless of weather conditions, for greater productivity and easier quality control. Since module height is restricted, safety is enhanced, as workers build at limited heights within the fabrication facility.
Modularising GTLThe same advantages of modular
construction of refineries are being applied to the construction of distributed GTL plants. The GTL process involves two operations: the conversion of natural gas to a mixture of carbon monoxide (CO) and hydrogen (H2), known as syngas, followed by a Fischer-Tropsch (FT) process to convert the syngas into paraffinic hydrocarbons that can be further refined to produce a wide range of hydrocarbon-based products, includ-ing clean-burning, sulphur-free diesel and jet fuel. Speciality prod-ucts including food-grade waxes, solvents and lubricants can also be produced from the paraffinic hydrocarbons.
Large, commercial-scale GTL plants, including the Sasol Oryx and the Shell Pearl plants (both located in Qatar), have been built at enor-mous capital cost. The Oryx plant, designed for production levels of 34 000 b/d, cost around $1.5 billion to build. The Shell Pearl plant, with an ultimate design capacity of 140 000 b/d of GTL products and 120 b/d of natural gas liquids, cost around $18-19 billion. Conventional GTL plant designs rely on econo-mies of scale to drive positive financial returns and are viable only where there are large supplies of low-priced natural gas.
However, another option being developed — smaller-sized and distributed GTL plants — shows
promise for deriving value from smaller accumulations of unconven-tional gas that would otherwise be left underground, such as shale gas, tight gas, coal bed methane and stranded gas (gas fields located too far from existing pipeline infra-structure). A small, modularised GTL plant has the flexibility to be installed close to the trapped resource and then used to process that resource locally. Associated gas (gas produced along with oil) is another area of opportunity for modularised GTL plants. This gas is typically disposed of either by re-injection, at considerable expense, back into the reservoir or by the wasteful and environmentally damaging practice of flaring, which is subject to increasing regulation. Modularised GTL plants enable this otherwise wasted gas to be converted into additional revenue.
In the larger economic picture, a modular GTL capability can be the key factor that enables the construc-tion of upstream projects that would otherwise be cancelled because of poor results derived from economic models. For exam-ple, some shale gas discoveries are being hampered by high develop-ment costs, which result in marginal economics due to gas prices that are often low. These projects can be enhanced by converting the gas to higher-value clean fuels produced in the GTL process.
ITWInnovative
TechnologiesWorldwide
ITW is a fast growing Company, marketing and implementing unique and patented Production Units Online Cleaning, Tank Cleaning, Decontamination and Reclamation technologies, along with Specialty Chemicals. Given the considerable success, ITW is expanding its markets and activities and is looking for experienced professionals worldwide. The candidates should have minimum a 5 years experience in at least one of these fields:
refining/petrochemicals process specialty chemicals sales refining/petrochemicals process technology, operations, maintenance, turnaround
and should be Having a technical degree and good market knowledge is also required. The available positions will cover: technical sales, implementation of ITW technologies on the field, sales and operations management.
Interesting compensation plans will be given along with serious career possibilities. Please contact:
ITW S.r.l.— S.Cusumano 96011 Augusta - Italy [email protected] www.itwtechnologies.com
HIRING EXPERIENCED PROFESSIONALS WORLDWIDE
velocys.indd 2 10/12/12 14:27:56lieberman.indd 3 08/03/2013 16:02
cri.indd 10 10/06/2013 16:38
sabin.indd 1 27/02/2013 15:01
Predicting catalyst lifetime
The catalytic reforming of heavy naphtha (heavy straight-run gasoline, or
HSRG) is a favoured process in petroleum refineries for producing high-octane gasoline. From the viewpoint of process operation, there are three kinds of widely used catalytic reforming units: semi-regenerative, continuous cata-lyst regeneration (CCR) and cyclic regenerative processes. Semi-regenerative catalytic naphtha reforming is the oldest type and it is generally carried out in three or four fixed-bed reactors in series with intermediate preheaters oper-ated adiabatically at temperatures between 450°C and 520°C, total pressure between 10 and 35 bar, and molar hydrogen-to-hydrocar-bon ratios between 3 and 8. Among all commercial technologies for the catalytic reforming process, those using fixed-bed reactors in series are the most common configura-tions due to the relative simplicity of scaling up an operation. However, the main disadvantages of fixed-bed reactors are coke formation, active phase sintering and poisoning during the catalyst’s life cycle. Consequently, the main problem with fixed-bed reactors is the loss of catalyst activity over time, which reduces the length of continuous operation.
The catalyst’s life cycle is further complicated by numerous technical, environmental and organisational issues. In principle, different companies can be involved in each of the life cycle steps. The life cycle of a catalyst starts with the initial production of fresh catalyst oxide, which is pre-sulphided prior to
A simulation model compares accurately with actual operating data to provide reliable predictions of catalyst lifetime in naphtha reforming
S REZA SEIF MOHADDECY and SEPEHR SADIGHIRIPI
being used in the refinery process. During its use when the catalyst deactivates and it does not meet performance targets within the limits of the reactor’s operating conditions, the reactor is shut down. Therefore, the catalyst’s life cycle typically involves a long chain of operations, normally performed by different specialised companies.1,2
Catalyst life is dependent essen-tially on process conditions and also, to a large extent, on the distri-bution of species such as sulphur and nitrogen present in the oil, both of which can change during a typical commercial run. Moreover, depending on the process used, the catalyst’s life cycle may vary from a few seconds, as in fluid catalytic cracking (FCC), to several years, as in ammonia synthesis; therefore, a reliable prediction of catalyst life has been extremely difficult.
Before the catalyst is loaded in the reactor, many laboratory exper-iments are carried out to identify its operability. An accelerated method for a one-day laboratory test is proposed here for the prognosis of activities and an estimation of the lifetime of the catalyst. The method is based on knowledge of the deac-tivation behaviour of catalysts under different reaction conditions, including extreme conditions at high liquid hourly space velocities (LHSV). Hence, the method enables prediction of lifetime and perfor-mance using data obtained from a one-day test in laboratory condi-tions. The curve of LHSV versus lifetime is an individual property of a given catalyst, which has to be established experimentally. Any
point from this curve can be used for an estimation of lifetime.3,4 Since deactivation of the catalyst takes place over a long time scale, it is difficult to mimic this in a pilot plant or at lab scale. Consequently, any study on an industrial scale to predict the life cycle of a catalyst plays an important role regarding economic, operational and environ-mental issues.5,6
Works on the life cycle of an industrial-scale catalytic naphtha reforming unit are scarce. In our present work, the semi-regenerative catalytic reforming process of a commercial-scale oil refinery was simulated using the Petro-Sim simulator (KBC Profimatics, 2009). After validating the simulation, the lifetime of the next cycle was predicted by the previous simu-lated cycle lives, and the results were compared with actual test runs of the plant at the end of the cycle.
Process descriptionA commercial fixed-bed catalytic naphtha reforming unit called Platformer, licensed by Chevron and with a nominal capacity of 16 500 b/d, was chosen as a case study. The feed to the plant, prior to entering the catalytic reformer, should undergo hydrodesulphuri-sation (HDS) in the hydrotreatment unit. Then, the produced naphtha, called Platcharge, is introduced to the catalytic reforming process. The most commonly used types of cata-lytic reforming units have three or four reactors, wherein each has a fixed catalytic bed. For such a unit, the activity of the catalyst reduces during operation due to deposited
www.eptq.com PTQ Q3 2013 85
ripi.indd 1 07/06/2013 19:54
86 PTQ Q3 2013 www.eptq.com
tions, creates challenges for modelling the process. To reduce these complications, reactants in the mixture are classifi ed in certain and limited groups called pseudo- components. Additionally, Arhenius and Langmuir-Hinshelwood kinet-ics are widely used for kinetic-based catalyst modelling and simulation of the catalytic naphtha reforming process.9-13 Petro-Sim software is a simulator capable of simulating commer-cial-scale catalytic reforming units,15 and so enables us to simulate reac-tors with different catalyst weights and sizes.16,17 In this research, Petro-Sim has been used to simulate and analyse a catalytic reforming unit.
To simulate a catalytic naphtha reforming process using Petro-Sim, external calibration for determining the kinetic parameters is needed. For this unit, the Ref-Sim module is provided for the task. For genera-tion of the required parameters, such as frequency factors,
coke and loss of chloride. Hopefully, the activity of the cata-lyst can be periodically regenerated or restored using in situ high-temperature oxidation of the coke followed by a chlorination process. Therefore, the catalyst of a semi-regenerative catalytic reformer is regenerated during routine shut-downs of the process every 18-24 months. Normally, the catalyst can be regenerated three or four times. Then, it must be returned to the manufacturer for reclamation of platinum and/or rhenium.
In Figure 1, Platcharge is preheated by the fi rst furnace (H-1). It then enters the fi rst reactor (R-1), where naphthenes are dehydrogen-ated to aromatics. Next, the product stream from the fi rst reactor passes through the second reactor (R-2), and the outlet stream of that enters the third reactor (R-3). Similarly, the product stream from the third reactor enters the fourth reactor (R-4). The overall reforming reac-tions are endothermic, so a preheater (H-1, H-2, H-3 and H-4)
is provided ahead of each reform-ing reactor.
Next, the product stream from the fourth reactor enters a separa-tor, V-1, in which hydrogen produced during the reforming process (the gas stream) is recycled, then mixed with Platcharge. Finally, the liquid product leaving the separator is introduced to the gasoline stabiliser, in which LPG and light gases are separated from gasoline. The vapour pressure of the gasoline can be set according to market requirements.
The distribution of catalyst in the reactors is shown in Table 1. The normal operating conditions of the unit are shown in Table 2.
Process simulation and validationCatalytic reforming is often modelled and simulated based on the number of reactive species and the type of kinetic model used.7,8
The presence of many components, either as reactants or as intermedi-ate products in the reactive mixture, with resulting new reac-
Table 1
Platcharge
R
Purge
Gas to LPG
LPG
Reformate
Recycle
H-1 H-2 H-3 H-4
E-3E-2E-1
C-1
R-1R-2
R-3R-4
V-1
Vapour
Stabiliser column
Air cooler Water cooler
Figure 1 Block fl ow diagram of the catalytic reforming unit of the target oil refi nery
1st reactor 2nd reactor 3rd reactor 4th reactorCatalyst weight, kg 5077.25 7615.87 12 693.13 25 386.25Catalyst distribution, wt% 10 15 25 50
Catalyst distribution in reforming reactors
Table 1
Process variable ValueInlet temperature, °C 490-515Hydrogen/hydrocarbon ratio, mol/mol 3-7LHSV, h-1 1-2Yield, vol% 70-85
Operating conditions in the catalytic reforming of the target oil refi nery
Table 2
ripi.indd 2 07/06/2013 19:54
Value adding catalysts, absorbents, additives and process technology for oil refining processes.
www.jmcatalysts.com/refineries
UKTel +44 (0)1642 553601
USATel +1 630 268 6300Tel +1 732 223 4644
Naphtha HDS
Diesel HDS
Hydrotreating catalysts
Vacuum tower
Hydrogen catalysts
FCC unit
FCC additivesSOx NOx removal
Light olefin productionBottoms conversion
Metals trapsActivity boosters
CO oxidation
Gas / liquid purification
S, Hg, Cl removal
Crude unit
VGO HDS
Hydrogen plant
Process diagnostics
Untitled-1 1 31/07/2012 17:18jm.indd 1 26/02/2013 17:04
ProMax is a powerful and versatile process simulation package that is used worldwide to design and optimize gas processing, re ning and chemical facilities. Totally integrated with Microsoft Visio®, Excel® and Word®,ProMax is the best simulation tool for designing gas plants and predicting performance under varying process conditions.
Discover why engineers around the world depend on ProMax!
2
2
Process Focus: Gas Processing
ProMax® with TSWEET® & PROSIM®
Process Simulation Software
bre.indd 1 8/3/12 10:09:31
activation energies and decay constants, by Ref-Sim, actual test runs from the unit under study should be gathered. These parame-ters mainly consist of inlet and outlet reactor temperatures, catalyst weight, specifi cation of feed and products, operating pressures, fl ow rates of makeup hydrogen, and recycle rates and fl ow rates of all gaseous and liquid products.14 After preparing a test run and running Ref-Sim, tuning parameters are sent to the Petro-Sim environment to simulate and optimise the operating conditions. Moreover, the life of the catalyst can be predicted in the Petro-Sim environment. It is obvious that after simulating catalytic naph-tha reforming (see Figure 2), the simulated variables should be compared against actual data to validate the process simulation; otherwise, the accuracy of the predictions can be compromised.
Since the catalyst in catalytic reforming loses activity over time, predicting the life cycle of a commercial catalyst must take into account catalyst production, sulphi-dation, usage and oxidative regeneration. The rate of decay in the Petro-Sim environment is predominantly a function of reformer type and severity of oper-ation. The latter is determined by the feed — the percentage amounts of naphthenic, aromatic and paraffi nic hydrocarbons — there-fore, during operation, reactor temperatures should be raised to achieve the same reformate octane. To represent this behaviour, Petro-Sim tracks catalyst decay by predicting decay factors through each reactor. This calculated decay information forms part of the model and is available on the cali-bration factor tab for predicting the life cycle of the catalyst.
Results and discussionFrom the start of run to the end of run, actual test runs were gathered from the target reforming unit. Since, for a specifi ed catalyst life, the simulator was used for predict-ing the steady-state condition of the Platformer, the data used for cali-bration should be selected from the normal condition when no
www.eptq.com PTQ Q3 2013 89
abnormalities, such as tower fl ood-ing, emergency depressurisation and pump or compressor shut-down, were happening during the operation. Before using these data to estimate the tuning parameters, it was necessary to validate them. If a reasonable overall mass balance (±5%) cannot be calculated, the validity of the test run is compro-mised. After validating the test runs, generating the calibration factors and sending them to the
Petro-Sim environment, the comparison of the simulated outlet temperatures of reforming reactors for the fi rst, second and third cycles are shown in Tables 3 to 6. From these tables, it can be concluded that Petro-Sim can simulate reactor temperatures with an absolute average deviation percentage (AAD%) of less than 4%.
To gain better insight, the most important process variables of Platformer — RON of product,
Platcharge
Gas to LPG
Liquid to LPG
Stabiliser column
FurnaceR
CRU(R-1 to R-4) H2 product
Reformate (gasoline)
Air cooler Water cooler
Figure 2 Simulation of catalytic naphtha reforming in the target refi nery
Variable Unit Test run 1 Test run 2 Test run 3 AAD% Simulation Actual Simulation Actual Simulation Actual 1st cycle °C 441.8 446 441 447 440 449 1.442nd cycle °C 466 447 471.5 440 473.2 448 5.683rd cycle °C 452.1 451 446 430 458.1 431 3.42 * Absolute average deviation per cent
Comparison of outlet temperature of fi rst reactor for previous cycles (fi rst, second and third cycles) of catalytic reforming of the target oil refi nery
Table 3
Variable Unit Test run 1 Test run 2 Test run 3 AAD% Actual Simulation Actual Simulation Actual Simulation 1st cycle ° C 486.3 487.2 488 486 489 483.6 0.572nd cycle ° C 488 488.2 482 486.5 493.5 488 0.643rd cycle ° C 497 466.8 465 474.1 468 474.5 3.14
Comparison of outlet temperature of second reactor for previous cycles (fi rst, second and third cycles) of catalytic reforming of the target oil refi nery
Table 4
ripi.indd 3 11/06/2013 09:53
KALDAIR
JZ-0182 PTQ-297mm x 210mm-Reliability Ad-July 2013 Q3.indd 1 4/26/13 12:08 PMj zink.indd 1 06/06/2013 10:21
www.eptq.com PTQ Q3 2013 91
was catastrophic for the catalyst. As a result, the control room shut down the unit to decide whether to regenerate or replace the catalyst. By using the developed simulation program, the life cycle of the
510°C to keep the RON of the prod-uct at the desired value (>94). As was expected, on day 719, the temperature of the reactor needed to be set at values higher than 510°C to preserve the RON; this
yield of gasoline and life of catalyst — were also simulated. The AAD% of each of the simulation results (for the first, second and third cycles) is shown in Tables 7, 8 and 9. It is supposed that the main source of deviation was the possi-bility of errors arising from measurements in gathering data obtained from faults such as signal transmission, calibration and power fluctuation of instruments, which could not be excluded from the actual data. However, from the presented simulation results, it can be concluded that the developed simulation program was reliable enough to be applied for predicting the behaviour of the catalytic reforming unit under study.
After evaluating and validating the simulation, the process varia-bles were predicted for the fourth cycle. After starting up the process, predicted results from the simula-tor were compared against actual results versus days on stream. Figures 3 to 8 show comparisons between the reactor temperatures, RON and yields of products against the actual values. As these figures show, there are close mappings between the measured and the predicted process variables. It should be mentioned that the AAD% of predictions for the first, second, third and fourth outlet reactor temperatures were 0.75, 0.58, 0.54 and 0.61, respectively. Moreover, the AAD% of the predic-tion for product RON and product volume yield were 1.45 and 0.03, respectively.
The most important parameter of a catalytic reforming unit — the life of the next cycle — was also stud-ied by the validated simulation program. Before starting up the next cycle, the life of the cycle (end of run) was predicted by Petro-Sim for different days on stream (see Table 10). From this table, it can be seen that the average life of the catalyst for the fourth cycle is about 679 days. To validate this result, after starting up the fourth cycle, the operation of Platformer was accurately monitored. After 710 days, it was found that the control room had to raise the inlet temper-atures of the reactors to around
Table 7
Table 8
Variable Unit Testrun1Testrun2Testrun3 AAD% Actual Simulation Actual Simulation Actual Simulation1st cycle ° C 500 510.8 502 509.8 502.5 507.8 1.592nd cycle ° C 502 503.4 498 502 508 505.5 0.543rd cycle ° C 511 483 480 496.9 481 490.5 3.66
Comparisonofoutlettemperatureofthirdreactorforpreviouscycles(first,secondandthirdcycles)ofcatalyticreformingofthetargetoilrefinery
Table5
Variable UnitTestrun1Testrun2 Testrun3 ADD% Actual Simulation Actual Simulation Actual Simulation1st cycle ° C 509.9 510 511.8 509 511.6 507.1 0.482nd cycle ° C 511.8 509.4 507.5 509.4 517.8 511.5 0.693rd cycle ° C 520.1 494 490.2 503.3 493 499.7 3.02
Comparisonofoutlettemperatureoffourthreactorforpreviouscycles(first,secondandthirdcycles)ofcatalyticreformingofthetargetoilrefinery
Table6
Variable UnitTestrun1 Testrun2 Testrun3 AAD% Actual Simulation Actual Simulation Actual Simulation1st cycle - 97.3 97.47 97.2 97.25 96.6 96.9 0.182nd cycle - 94.2 94.55 92.2 92.22 93.9 92.7 0.293rd cycle - 92.3 92.91 95.5 94.9 93.6 93.8 0.08
ComparisonofproductRONforpreviouscycles(first,secondandthirdcycles)ofcatalyticreformingofthetargetoilrefinery
Table7
Variable UnitTestrun1 Testrun2Testrun3 AAD% Actual Simulation Actual Simulation Actual Simulation1st cycle vol% 75.8 74.33 75.5 74.8 74 75.3 1.52nd cycle vol% 80.2 76.2 80.4 79.8 81.1 79.1 2.83rd cycle vol% 79.5 79.22 80.2 76.2 78.8 77.9 2.25
Comparisonofproductyieldforpreviouscycles(first,secondandthirdcycles)ofcatalyticreformingofthetargetoilrefinery
Table8
Variable UnitTestrun1 Testrun2 Testrun3 AAD% Actual Simulation Actual Simulation Actual Simulation1st cycle days 920 895.6 920 889.6 920 904 2.632nd cycle days 891 859.6 891 820.2 891 856 5.453rd cycle days 591 568.7 591 569.2 591 600 3.08
Comparisonoftotallifetimepredictionforpreviouscycles(first,secondandthirdcycles)ofcatalyticreformingofthetargetoilrefinery
Table9
ripi.indd 4 11/06/2013 09:54
92 PTQ Q3 2013 www.eptq.com
Data gathered from the fi rst three cycles were chosen to calibrate the simulator, then the plant was simu-lated using the Petro-Sim process simulator. The signifi cant process variables — outlet temperatures of the fi rst, second, third and fourth reactors, product RON and volume yield — were simulated. After comparing results against actual data, the accuracy of the simulator was validated due to the low devia-tion of the simulated process variables.
In the next step, the simulator was used to predict the process variables corresponding to the next cycle. Comparing the results with the real outputs showed that the per cent of absolute average devia-tion (AAD%) of the main parameters were about 0.75%, 0.58%, 0.54%, 0.61%, 1.45% and 0.03%, respectively. Additionally, before starting the cycle, the simu-lated life of the catalyst was estimated at about 679 days and recorded at about 719 days at the end of the cycle under study.
References1 Eijsbouts S, A A, van Leerdam G C, Life cycle of hydroprocessing catalysts and total catalyst management, Catalysis Today, 2008, 130, 361-373.2 Girardier F, Meerstadt R P, Oude Groeniger H, Docter S, Akzo Nobel Catalyst Symposium, 1998.3 Petrov L, Vladov Ch, Neshev N, Bonev Ch, Prahov L, Kirkov N, Vasileva M, Filkova D, Dancheva S, Method for Forecasting Lifetime of Industrial Catalysts for Hydrogenation, Bulgarian Patent 41960, 1986.4 Petrov L, Vladov Ch, Bonev Ch, Prahov L, Kirkov N, Eliyas A, Neshev N, Filkova D, Dancheva S, Prognosis of the lifetime of industrial hydrogenation and oxidation catalysts, EuropaCat-2, Book of Abstracts, Maastricht, The Netherlands, 1995, 646.5 Moulijn J A, Van Diepen A E, Kapteijn F, Catalyst deactivation: is it predictable? What to do?, Applied Catalysis A: General, 2001, 212, 3-16.6 Parrott S L, Adarme O, Lin F, Catalyst Life Prediction in Hydrodesulphurization, US Patent 5341313, 1994.7 Shakoor Z M, Catalytic reforming of heavy naphtha, analysis and simulation, Diyala Journal of Engineering Sciences, 2011, 4, No. 02.
and life of the catalyst in a commer-cial process is vital. To perform such a task, a catalytic naphtha reforming unit with a nominal capacity of 16 500 b/d was studied.
reforming unit can be estimated with acceptable accuracy.
ConclusionPrediction of the process variables
450
460
455
445
440
Tem
pera
ture
, ºC
4350 100 200 300 400 500 600
Days on stream
ActualSimulation
Figure 3 Comparison of outlet temperature of fi rst reactor for the current cycle
480
490
485
475
470
Tem
pera
ture
, ºC
4650 100 200 300 400 500 600
Days on stream
ActualSimulation
Figure 4 Comparison of outlet temperature of second reactor for the current cycle
495
505
500
490
485
Tem
pera
ture
, ºC
4800 100 200 300 400 500 600
Days on stream
ActualSimulation
Figure 5 Comparison of outlet temperature of third reactor for the current cycle
ripi.indd 5 07/06/2013 19:55
www.eptq.com PTQ Q3 2013 93
Research Institute of Petroleum Industry (RIPI). He holds a PhD in chemical engineering from Universiti Teknologi Malaysia. Email: Sadighis @ripi.ir
Technology. Email: Seifsr @ripi.irSepehr Sadighi is a Project Manager in the Catalysis and Nanotechnology Division, Catalytic Reaction Engineering Department,
8 Arani H M, Shirvani M, Safdarian K, Dorostkar E, Lumping procedure for a kinetic model of catalytic naphtha reforming, Brazilian Journal of Chemical Engineering, 2009, 26(4), 723-732.9 Mohadesi M, Mousavi H S, Kinetic investigation of catalyst deactivation in catalytic reforming of naphtha, Chemical Product and Process Modeling, 2012, 7.10 Saidi M, Mostoufi N, SotudehGharebagh R, Modeling and simulation of continuous catalytic regeneration (CCR) process, International Journal of Applied Engineering Research, Dindigul, 2011, 2(1).11 Arani H M, Shokri S, Shirvani M,Dynamic modeling and simulation of catalytic naphtha reforming, International Journal of Chemical Engineering and Applications, 2010, 1(2).12 Hou Weifeng, Hongye Su, Yongyou Hu, Jian Chu, Modeling, simulation and optimization of a whole industrial catalytic naphtha reforming process on Aspen Plus Platform, Chinese Journal of Chemical Engineering, 2006, 14(5), 584-591.13 Rahimpour M R, Esmaili S, Bagheri Yazdi S A, A kinetic and deactivation model for industrial catalytic naphtha reforming, Iranian Journal of Science and Technology, 2003.14 Liang Ke-min, Guo Hai-yan, Pan Shie-wei, A study on naphtha catalytic reforming reactor simulation and analysis, Journal of Zhejiang University Science, 2005, 6B(6), 590-596.15 Petro-Sim User Guide, KBC Advanced Technologist, KBC Profimatic, 2005.16 Mohaddecy S R, Sadighi S, Ghabouli O, Lumping kinetic model and simulation of catalytic naphtha reforming process, Chemical Technology: An Indian Journal, 6(4), 2011.17 Mohaddecy S R, Sadighi S, Optimising a reactor’s catalyst distribution, PTQ, Q3 2008, 101-107.
S Reza Seif Mohaddecy is a Project Manager in the Catalysis and Nanotechnology Division, Catalytic Reaction Engineering Department, Research Institute of Petroleum Industry (RIPI), Tehran, Iran. He holds a MS in chemical engineering from Sharif University of
500
510
505
495
490
Tem
pera
ture
, ºC
4850 100 200 300 400 500 600
Days on stream
ActualSimulation
Figure 6 Comparison of outlet temperature of fourth reactor for the current cycle
96
100
98
94
92
RO
N
900 100 200 300 400 500 600
Days on stream
ActualSimulation
Figure 7 Comparison of product RON for the current cycle
76
80
78
84
82
74
72
Volu
me y
ield
, %
700 100 200 300 400 500 600
Days on stream
ActualSimulation
Figure 8 Comparison of product volume yield for the current cycle
Life cycle, days Days on stream, days646 108707 222657 288740 382635 443684 503
Total catalyst life prediction for the current cycle
Table 10
ripi.indd 6 07/06/2013 19:55
JOHNSON SCREENS® GLOBAL FIELD SERVICE______
WE APPLY OUR EXPERTISE FROM INITIAL PRODUCT DESIGN THROUGH PROPRIETARY FABRICATION, INSPECTION, TESTING, INSTALLATION AND START-UP. OUR ACQUIRED EXPERIENCE GIVES US A WEALTH OF KNOWLEDGE WHICH WE MAKE AVAILABLE TO YOU THROUGH OUR FIELD SERVICE TEAMS AND IN-HOUSE TECHNICAL SUPPORT GROUPS.
• Project engineering assistance for revamping or upgrading• Technical assistance when damage is found or assumed on reactor internals• Internal expertise after catalyst unloading • On-site installation of Johnson Screens® equipment• Supervision of contractor’s installation
______
BILFINGER WATER TECHNOLOGIESwww.water.bilfinger.com
AUSTRALIA - ASIA PACIFIC Phone +61 7 3867 [email protected]
EUROPE - MIDDLE EAST - AFRICA Phone +33 5 4902 [email protected]
JAPAN Phone +81 55 997 [email protected]
NORTH, SOUTH & CENTRAL AMERICA Phone +1 651 636 [email protected]
PTQ Q3 [July]_Field Service.indd 1 07/06/2013 09:30:06j screens.indd 1 07/06/2013 10:04
Additives provide flexibility for FCC units and delayed cokers
Flexibility is a primary advan-tage in the ever-changing refining industry. Successful
refiners achieve profitability by quickly reacting to market condi-tions. And catalyst additives have long been one method refiners have turned to when reacting to volatile market dynamics. Albemarle has long been a manufacturer of FCC additives, which are used for envi-ronmental compliance, propylene maximisation and overall yield improvement. Recent opportunities in this market have highlighted the benefits of using its Bottoms Cracking and Metals Tolerance (BCMT) additives to drive up FCC unit profitability. The company has expanded its focus on refinery additives by partnering with OptiFuel Technology Group (OFTG) to develop a technology utilising a proprietary additive supplied by Albemarle that increases the liquid yield from delayed coker units.1 The benefit can be around 3.7 $/bbl to refiners that implement this process, with additional profit possible in cases where increased unit throughput results.
Fluid cracking catalyst additivesRefiners have leveraged the benefits of BCMT additives in many ways. For example, some have used it to process less expensive opportunity feeds. Others have lowered total catalyst expenditures by purchasing fresh distressed FCC catalyst stocks at a discount from closed refineries then enhancing that catalyst with this additive in order to make it usable in their refinery. Several are using BCMT to correct for non-
A proprietary additive applied to a novel coker process increases the liquid yield from delayed coking
ALAN KRAMER and RAUL ARRIAGAAlbemarle Corporation
optimised base catalyst formulations because the refiner is either locked into a supply agreement with a supplier that cannot formulate a catalyst with enough bottoms crack-ing activity to meet their needs, or the immediate economics provide an opportunity to increase profita-bility by changing the product mix temporarily.
Table 1 shows recent use of BCMT-500 and the lower rare-earth BCMT-500-LRT2 at well over 30 refineries globally.
BCMT additives are built upon Albemarle’s Topaz alumina-gel, high-accessibility catalyst technol-ogy.3 However, these additives differ from cracking catalysts because they have been formulated
www.eptq.com PTQ Q3 2013 95
Region Unit design Additive technologyAsia KelloggOrthoflow BCMT-500Asia ExxonModelIV BCMT-500Asia UOPSBS BCMT-500Asia S&WR2R BCMT-500Asia KelloggDualRiser BCMT-500Asia S&WR2R BCMT-500Asia ShellResid BCMT-500Asia UOPRCC BCMT-500Asia S&WR2R BCMT-500Asia Notdisclosed BCMT-500LRTAsia UOPRFCC BCMT-500Europe&Africa UOPSBS BCMT-500Europe&Africa ExxonModelIV BCMT-500Europe&Africa Shell BCMT-500Europe&Africa UOPSBS BCMT-500Europe&Africa UOPSBS BCMT-500-LRTEurope&Africa UOPSBS BCMT-500Europe&Africa UOPHE BCMT-500-LRTEurope&Africa ExxonModelIV BCMT-500Europe&Africa ExxonFlexicracker BCMT-500Europe&Africa UOPSBS BCMT-500Europe&Africa UOPSBS BCMT-500Europe&Africa RussianSBS BCMT-500MiddleEast&India UOP BCMT-500MiddleEast&India UOP BCMT-500LRTMiddleEast&India S&WR2R BCMT-500MiddleEast&India ExxonFlexicracker BCMT-500MiddleEast&India UOPSBS BCMT-500LRTMiddleEast&India S&WR2R BCMT-500MiddleEast&India S&WR2R BCMT-500MiddleEast&India S&WR2R BCMT-500Americas UOPSBS BCMT-500Americas UOPHE BCMT-500Americas Notdisclosed BCMT-500Americas Notdisclosed BCMT-500Americas Notdisclosed BCMT-500Americas Notdisclosed BCMT-500
Recent users of BCMT-500 and BCMT-500-LRT additives
Table 1
albemarle.indd 1 07/06/2013 20:03
96 PTQ Q3 2013 www.eptq.com
ing conditions to maximise LCO yield during BCMT usage, Albemarle normalised the raw oper-ating data from the trial period back to the baseline conditions using a FCC simulator model. At normal-ised conditions of constant riser and feed preheat temperatures, the regenerator temperature is 13°C lower. This improvement in delta coke is due in part to the metals passivation activity of BCMT-500.Unit B is a larger FCC unit process-ing a partially hydrotreated VGO feed with 0.2% Conradson carbon residue. Metals on the equilibrium catalyst are moderate to low at 900 ppm nickel and 300 ppm vanadium. The major driver for BCMT-500 usage at this refinery was to tempo-rarily increase heavy naphtha and light cycle oil yield without reformu-lating the base catalyst, which was geared towards maximum gasoline yield. The additive was introduced into the unit at a 20% addition rate, replacing an equal amount of base catalysts. The unit reached 11% BCMT in inventory before ceasing additions and reverting back to maximum gasoline operations. The riser temperature remained constant during the period of usage at 510°C and the regenerator temperature dropped an average of 2°C at the peak of BCMT-500 turnover.
The yield shifts with 11% additive determined from the refinery oper-ating data have been summarised in the middle column of Table 2. The right-hand column shows model corrected yield shifts, which are more telling of the true performance of the additive because the as- produced gasoline required cut-point adjustments.
As can be seen from the data, the introduction of 11% additive into the FCC unit greatly increased LCO yield by increasing bottoms activity significantly and adjusting the unit’s selectivities towards the desired yield pattern.Unit C is a small, independent refiner struggling to process a tough 6 Conradson carbon residue feed that contains up to 16 ppm nickel and 6.5 ppm vanadium. The feed also contains significant amounts of calcium and iron that together deposit an additional 1.2 wt% on the
to provide extreme accessibility and metals tolerance in a highly concentrated form since the additive makes up only a small fraction of the FCC unit’s circulating inventory.
The mechanism behind BCMT has been discussed in detail,4,5,6,7,8
but can be summarised as follows. These additives enhance the diffu-sional architecture of the catalyst system, thereby improving accessi-bility and allowing feed molecules to selectively pre-crack before entering the zeolite pores. They will also trap deleterious metals, such as nickel, reducing hydrogen make while improving coke selec-tivity and overall catalyst performance. Vanadium poisoning is effectively dealt with through the use of vanadium- tolerant zeolites and matrix bottoms cracking components. The additive also improves overall performance in units with high iron and calcium in part via its high accessibility, as will be shown in one of the following case studies.
FCC additive case studiesThe following are recent examples of BCMT’s performance.Unit A is a FCC unit that processes a near equal mix of hydrotreated VGO and atmospheric resid. The combined feed has a Conradson carbon residue of 2.5 and gravity of 24 °API. Equilibrium catalyst metals range from 3000-4000 ppm nickel and 5000-7000 ppm vanadium. The unit is limited by regenerator temperature, and the refiner has a desire to maximise LCO production. This refinery currently adds BCMT-500 at 20% of the total catalyst addition rate.
Figure 1 shows that the LCO/bottoms ratio increased by over 30% at constant relative conversion with 20% BCMT at steady state. These results become more impressive when one takes into account that, for the period shown, the fraction of atmospheric resid in the feed was 20 relative per cent higher than the baseline and the riser temperature was on average 9°C lower.
Since this refiner adjusted operat-
0.9
1.0
0.8
0.7
0.6
0.5
0.4LC
O/B
ott
om
s, w
t%/w
t%
0.3–6 –4 –2 0 2 4 6
Relative conversion, wt%
BaselineBCMT-500
Figure 1 LCO/bottoms ratio versus relative conversion for Unit A
Yield Yield shift with BCMT-500, wt% Refinery data Model correctedDry gas +0.1 0.0LPG -0.8 -1.0Light gasoline -1.5 Heavy gasoline +1.9 LCO +2.4 Gasoline (C
5-221°C) -2.0
LCO (221-360°C) +5.0HCO -2.1 -2.0Coke 0.0 0.0
Yield shifts in wt% with 11% BCMT-500 in inventory at Unit B
Table 2
albemarle.indd 2 10/06/2013 14:51
C O M P R E S S O R S n T U R B I N E S n G L O B A L S E R V I C E
EBARA CORPORATION
www.elliott-turbo.com
n Customers: Global oil and gas producers.
n Challenge: Changing compression requirements as fields mature and production peaks.
n Result: Flexible, reliable compressor designs for extended, cost-efficient operations.
They turned to Elliottfor reliable compression solutions.
From Aberdeen to Rio de Janeiro, Calgary to Jakarta, Elliott compressors, turbines, and expanders deliver the reliable, efficient performance that producers and processors require. And every piece of Elliott equipment is backed by our unmatched global service network. Customers throughout the world turn to Elliott for critical turbomachinery and service because our resources are global and our execution is local. Who will you turn to?
The world turns to Elliott.
elliott.indd 1 12/03/2013 14:20
equilibrium catalyst. Organic iron from the feed is known to deposit on the surface of cracking catalyst, creating a barrier that prevents the diffusion of large feed molecules into the catalyst.9 Together, iron and calcium can act to form low melting point eutectic compounds, further reducing catalyst accessibility.10 The deleterious effects of metals coupled with low catalyst replacement rates resulted in this refiner’s baseline e-cat activity ranging in the upper 50s to low 60s.
This refiner began adding BCMT-500-LRT, which utilises Albemarle’s Low Rare Earth Technologies (LRT), to achieve a performance similar to BCMT-500’s but with only 25% of the rare earth content.11,12 The LRT formulation provided the best projected benefits per unit cost for this refiner of all the bottoms crack-ing additives considered. The goals
98 PTQ Q3 2013 www.eptq.com
set forth by the refiner were to increase bottoms cracking and improve (lower) the bottoms API gravity, increase total liquid prod-uct, while maintaining or improving coke selectivity. The addition rate of additive constituted 25% of the total daily catalyst addition rate.
A comparison of the base case yields to those achieved with BCMT-500-LRT is given in Table 3. The yield shifts have also been corrected to the base case via a FCC simulator in a similar method to that used in the previous two exam-ples. This accounts for variations that were experienced in feed qual-ity and recycle rates between the trial and base periods. The riser temperature was maintained at a rather cold 495°C for a majority of the data in both the base and usage periods. The regenerator tempera-ture was lower by an average of 5°C
once the target concentration of additive was achieved, again demonstrating that BCMT delivers low delta coke with superior bottoms conversion.
BCMT summaryThese examples show how refiners have been able to utilise FCC addi-tives to react flexibly to opportunities by cracking more bottoms, optimising their product selectivity and mitigating contami-nant metals. The refiners who have recently chosen BCMT to improve profitability have been satisfied with the performance and value of these products.
OptiFuel technology and coker additivesIntroduction and technology For many years, delayed coking has played an important role in oil refin-eries as one of the most cost-effective processes for converting vacuum residue into more valuable prod-ucts. Delayed coking yields are the result of thermal reactions, and are mainly a function of feed properties and operating conditions such as furnace outlet temperature and drum pressure. However, this process results in coke yields from 20-40% of feed.
Albemarle and OFTG have devel-oped and optimised a technology, called OptiFuel Technology, licensed by Albemarle and employing an innovative additive to reduce coke yield and increase liquid yields in delayed cokers. This patented tech-nology and additive result not only in improved performance and prof-itability, but also in increased flexibility that can be used to reduce bottlenecks, increase throughput and produce more valuable prod-ucts. Figure 2 shows a schematic of the technology with the main process components.
The process, invented by Roger Etter of OFTG, consists of injection of an additive into the vapour region at the top of a coker drum, with nozzles specifically designed for optimum distribution and mini-mum carry-over. The additive mixture comprises a liquid portion (the carrier) and a proprietary solid additive supplied by Albemarle.
Yield Yield shift with BCMT-500-LRT, wt% Refinery data Model correctedDry gas 0.0 0.0LPG +1.9 +1.0Gasoline +4.3 +2.1LCO -2.7 -1.5HCO -3.5 -1.6Coke 0.0 0.0
Yield shifts in wt% with 25% BCMT-500-LRT in inventory at Unit C
Table 3
Solid additive
Carrier oil
Products
Coker feed
Additive tank and mixing
system
Steam jacketed tank and mixer
Rotary pump
Coke drum
Coke drum
Decoking cycle
Coking cycle
Figure 2 General schematic of the OptiFuel Technology coker process
albemarle.indd 3 07/06/2013 20:03
www.eptq.com PTQ Q3 2013 99
cracking reactions over the strictly thermal coking reactions that occur in the traditional delayed coker operation. During development, it has been observed that OptiFuel Technology shifts the delayed coker yields towards more valuable products, with reduced amounts of dry gas and coke.
Pilot plant verification Albemarle and OFTG have conducted a series of pilot plant runs at Penn State University (PSU). The scope of these studies has been to quantify the benefits of the tech-nology as a function of additive composition, feed properties and operating conditions.
vessels needed for additive mixing and storage.
The yield improvements seen with this technology are hypothesised to be the result of reactions in both the liquid and vapour phases, which are directly influenced by the additive. The active sites of the additive are intended to preferentially catalyse
Albemarle has designed an addi-tion system to ensure proper mixing of the solid and liquid portions and to avoid solids settling. The supply consists of a liquid carrier and the additive supplied to the refinery in bags or bulk shipments for mixing on-site. The additive injection system design minimises the size of
6
10
8
4
2
0
-2
–4
Abso
lute
delt
as,
wt%
of
feed
–6A1
–4.32
8.83
–4.51
–4.01
8.77
–4.76
–3.38
6.52
–3.14
–4.43
3.00
1.42
–1.44
0.44
0.99
–1.57
–0.67
2.25
–2.28
0.43
1.85
–1.49
–0.30
1.79
Coke, %
C3 + liquid, %
Fuel gas, %
A2 A3 A4 B1 B2 B3 B4
Feed and formulation
Figure 3 Changes in pilot plant yields with application of OptiFuel Technology: pilot plant runs at Penn State University
Base PredictedCoke 32.2% 28.5%Dry gas 5.6% 4.4%C
3+ liquid yields 62.2% 67.1%
Projected yields for commercial application of OptiFuel Technology
Table 4
www.eptq.com PTQ Q2 2013 43
Very Heavy Crude Upgrading – Long Term R&D Opportunities, 1994.2 Yui S, Chung K H, Syncrude upgrader revamp improves product quality, Oil Gas J, 2007, Vol. 105, 46, 52. 3 Chrones J, Germain R R, Bitumen and heavy oil upgrading in Canada, Fuel Sci Tech Int, 1989, 7, 783. 4 Rana M S, Samano V, Ancheyta J, Diaz J A I, A review of recent advances on process technologies for upgrading of heavy oils and residua, Fuel, 2007, 86, 1216. 5 Speight J G, The Chemistry and Technology of Petroleum, 2007, 4th ed, CRC Press/Taylor & Francis, Boca Raton, FL. 6 Sayles S, Romero S, Understand differences between thermal and hydrocracking, Hydrocarbon Process, 2011, Sept, 37. 7 Martinez J, Sanchez J L, Ancheyta J, Ruiz R S, A review of process aspects and modeling of ebullated bed reactors for hydrocracking of heavy oils, Catal Rev Sci Eng, 2010, 52, 60.8 Yui S, Producing quality synthetic crude oil from Canadian oil sands bitumen, J Jpn Petrol Inst, 2008, 51, 1. 9 Yui S, Athabasca oil sands produce quality diesel and jet fuels, Oil Gas J, 2000, Vol. 98, 47, 58. 10 Yui S, Chung K H, Processing oil sands bitumen is syncrude’s R&D focus, Oil Gas J, 2001, Vol. 99, 17, 46.
Variante 4a
A Member of Unitherm-Cemcon Group
CS Combustion Solutions GmbHSimmeringer Hauptstrasse 24A-1110 ViennaAustriaT+43 1 740 40 2812F +43 1 740 40 [email protected]
your expertsn ACID/TAIL GASn LIQUID SULFURn SPENT ACIDn CHEMICAL/PETROCHEMICAL RESIDUES
The team of CS Combustion Solutions is experienced in engineering, supply and commissioning of vertical and hori zontal burners and combustors for more than 20 years.
11 Wadsworth D, LC-Fining options for heavy oil upgrading, Proceedings of the NPRA Annual Meeting, San Diego, CA, 9-11 March 2008.12 Ordorica-Garcia G, Croiset E, Douglas P, Elkamel A, Gupta M, Modeling the energy demands and greenhouse gas emissions of the Canadian oil sands industry, Energy Fuels, 2007, 21, 2098. 13 Morawski I, Mosio-Mosiewski J, Effects of parameters in Ni-Mo catalysed hydrocracking of vacuum residue on composition and quality of obtained products, Fuel Process Technol, 2006, 87, 659. 14 Danial-Fortain P, Gauthier T, Merdrignac I, Budzinski H, Reactivity study of Athabasca vacuum residue in hydroconversion conditions, Catal Today, 2010, 150, 255. 15 Ding F, Ng S H, Xu C, Yui S, Reduction of light cycle oil in catalytic cracking of bitumen-derived crude HGOs through catalyst selection, Fuel Process Technol, 2007, 88, 833. 16 Botchwey C, Dalai A K, Adjaye J, Kinetics of bitumen-derived gas oil upgrading using a commercial NiMo/Al
2O
3 catalyst, Can J Chem
Eng, 2004, 82, 478. 17 Yui S, Sanford E, Kinetics of aromatics hydrogenation of bitumen-derived gas oils, Can J Chem Eng, 1991, 69, 1087. 18 Yui S, Sanford E, Mild hydrocracking of bitumen-derived coker and hydrocracker heavy gas oils: kinetics, product yields, and product
properties, Ind Eng Chem Res, 1989, 28, 1278. 19 Yui S, Removing diolefins from coker naphtha necessary before hydrotreating, Oil Gas J, 1999, 97, 36. 20 Chang A-F, Liu Y A, Predictive modeling of large-scale integrated refinery reaction and fractionation systems from plant data. Part 1: hydrocracking processes, Energy Fuels, 2011, 25, 5264.
Anton Alvarez-Majmutov is an NSERC Visiting Fellow at CanmetENERGY working on bitumen upgrading process modelling and simulation. He holds a PhD from Mexican Institute of Petroleum (IMP).
Jinwen Chen is a Senior Research Scientist and Group Leader at CanmetENERGY. He holds a PhD in chemical engineering from Tianjin University.
Mugurel Munteanu is a Lead Process Engineer at CoSyn Technology, a division of WorleyParsons, in Edmonton, Canada. He holds a PhD in chemical engineering from Laval University, Canada.
canmet.indd 6 08/03/2013 13:04albemarle.indd 4 07/06/2013 20:04
100 PTQ Q3 2013 www.eptq.com
Tulsa, Oklahoma, utilising a single heavy feed and employing further variations of the additives identified in the PSU study. A sampling of the results is shown in Figure 4. Analysis of this data is not yet complete, but the preliminary results support the conclusion that the OptiFuel Technology with coker additives reduces coke and improves liquid yields.
Economic benefits The results of these studies have been used to develop a model to project the potential yield improve-ments with the application of OptiFuel Technology with Albemarle’s coker additives. Table 4 shows the potential yield shifts for application to a 20 000 b/d delayed coking unit processing a heavy feed with a 22 wt% carbon content. The coke yield is projected to decline by as much as 3.7% absolute with a decrease in dry gas and increase in liquid yields.
The liquid products from a delayed coker require downstream processing before use in saleable products, so internal refinery trans-fer economics have been used to estimate the potential economic benefit from the improved yields realised with OptiFuel Technology. For the 20 000 b/d unit in this exam-ple, the potential economic benefit is just under $3.7/bbl and translates to a yearly increase in profit of $27 million. If this delayed coking unit was feed rate limited due to coke capacity, and was now able to increase feed rates to achieve the original coke yield, it could then realise an increase in profit of $53 million per year. Albemarle and OFTG are preparing for the first commercial application, which is expected to occur during 2013.
Computational fluid dynamic (CFD) modelling of additive injectionIn collaboration with a third-party consultant with expertise in CFD modelling, Albemarle has evaluated the potential impact of injecting a coker additive in a commercial delayed coker unit. Modelling of additive distribution, additive carry-over and thermal impacts were conducted.
producing a valuable liquid yield with a reduction in low-value prod-ucts, such as coke and dry gas.
Statistical analysis of the data generated in this testing indicates the coke reduction is much greater than the standard deviation of the coke yield and gives a high confi-dence level in the coke reduction resulting from the application of the coker additive. Specifically, the null hypothesis to be rejected was that “coke yield delta is higher or equal to zero”. The conclusion from this analysis is that there is a high confi-dence level (99.8%) that the negative coke deltas measured in the pilot plant are statistically significant and not merely the product of normal yield variations.1
Additional testing has recently been completed at the University of
The testing at PSU included evalu-ation with two feeds and eight additive formulations, (Feed A: API: 4.8, Concarbon: 28, Feed B, API: 7.1, Concarbon: 15.7). The conclusion from was that with OptiFuel Technology there was a reduction in coke yield in every case, with increased C3+ liquid yields. The yield variations are attributed to the differences in additive formulations, with a different set of coker operat-ing conditions associated with each of the respective feeds.
Figure 3 shows the absolute changes in pilot coker yields compared to a base case without the coker additive for each coker feed. The first four cases were run with feed A and the rest with feed B. The first three additive formulations were identified as optimum for
0
Abso
lute
del
ta, w
t% o
f fe
ed
TA1 TA2 TA3 TA4
CokeC3 + yieldDry gas
Figure 4 Changes in pilot plant yields with application of OptiFuel Technology: sample results, University of Tulsa, Oklahoma
1
2
3
4
X Z
Y
820°F
705°F
590°F
Figure 5 Simulated temperatures of a commercial coke drum during application of OptiFuel Technology
albemarle.indd 5 07/06/2013 20:04
www.eptq.com
The results from this study indicate that with the proper configuration, the additive can be injected with the desired level of distribution and limit the solids carry-over to less than 1 wt% of the additive injected into the unit. The modelling of thermal impacts indicates there is a negligible effect on liquid layer temperatures, but some vapour quenching occurs, which may create a further benefit from the reduction of necessary overhead quench. Additional studies are being conducted to analyse the benefits of further improved nozzle designs and configu-rations. Figure 5 illustrates the modelled temperature gradient of the vapour phase in a coke drum with the application of OptiFuel Technology. This diagram shows that the cooling effect due to the additive mixture is local-ised and in the upper vapour phase, minimising the impact on the liquid layer temperature and overall unit heat balance.
SummaryOptiFuel Technology with coker additives demonstrated in pilot plant testing the consistent ability to reduce coke and increase liquid products yield. The potential economic benefits of the yield shifts alone are estimated to be $3.7/bbl, with additional profit possible in cases where increased unit throughput results. Further improvement in this potential economic benefit is expected as development of the technology continues.
BCMT is a trademark of Albemarle Corporation. OptiFuel is a registered trademark of OptiFuel Technology Group.
References1 Arriaga R, Nickell R, Lane P, Etter R, Improve your delayed coker performance and operating flexibility with Albemarle’s new Optifuel coker additives, AFPM AM, AM-13-67, 2013.2 Arriaga R, Bruno K, Yung K Y, Applying low rare earth in demanding FCC operations, AFPM AM, AM-12-28, 2012.3 Bruno K, Humphries A, Pinto R, Yanik S, Fletcher R, The future has arrived in FCC: Jade and Topaz Technology, Akzo Nobel Technical Seminar, Laguna Cliffs California, 2002.4 Kramer A, Bottom of the barrel upgrading with FCC additive solutions, ERTC Annual Conference, Istanbul ,Turkey, 2010.5 Yung K Y, Jonker R J, Meijerink B, A novel and fast method to quantify FCC catalyst accessibility, American Chemical Symposium, Petroleum Chemistry Division Reprints 2002, 47, 3, 270-280, 2002.6 Kramer A, Bottom of the barrel upgrading, PTQ, Q1, 2011.7 Hakuli A, Bruno K, Imhof P, Fletcher F, FCC catalyst selection in diffusion limited operating regimes, NPRA AM-03-58, 2003.8 Fosket S J, Rautiainen E P H, Control iron contamination in resid FCC, Hydrocarbon Processing, 71-77, Nov 2001.9 Hodgson M C J, Looi C K, Yanik S J, Avoid excessive RFCCU catalyst deactivation: improve catalyst accessibility, Catalysts Courier, 35, 2-5, 1999.10 Vreugdenhil W, Mao M, Calcium contamination in FCC catalysts, Catalysts Courier, 37, 1999.11 Yung K Y, Bruno K, Low rare earth catalysts for FCC operations, PTQ, Q1, 71-79, 2012.12 Arriaga R, Bruno K, Low rare earth in FCC catalysts market and technology update, AFPM Cat Cracker Conference, Houston Texas, 2012.
Alan Kramer is a Global FCC Modeling Specialist with Albemarle Corporation. He holds a BS degree in chemical engineering from Johns Hopkins University. Email [email protected] Arriaga is a Global Applications Technology Specialist at Albemarle. He holds a BS degree in chemical engineering from Simon Bolivar University.
SAMSON AG · MESS- UND REGELTECHNIK Weismüllerstraße 3 60314 Frankfurt am Main · GermanyPhone: +49 69 4009-0 · Fax: +49 69 4009-1507 E-mail: [email protected] Internet: www.samson.deSAMSON GROUP · www.samsongroup.net
A01
120E
N
Partner with the Best
With over 50 independent subsidiar-ies and more than 220 engineering and sales offi ces spread across the world, SAMSON ensures the safety and environmental compatibility of your plants on any continent.
To offer the full range of high-quality control equipment used in industrial processes, SAMSON has brought together highly specialized compa-nies to form the SAMSON GROUP.
albemarle.indd 6 10/06/2013 14:51
Sponsored by:
For further information please contact us at:
www.europetro.com + 359 2 859 41 57 [email protected]
MOSCOW REFINING,GAS & PETROCHEMICALS WEEKSupported by the Association of Russian Refiners
16 - 20 September 2013, Lotte Hotel, Moscow
Over 500 Decision Makers - regional and international
Over 75 speakers included in strong technical programmes, comprising:
- Industry Outlook from Leading Consultants & Specialists
- Latest Innovations and Proven Technologies by World Class Providers
- Insights on Major Recent Projects and Future Plans of the Main Oil & Gas Companies
Over 50 Exhibition Stands
ONE WEEK OF QUALITY NETWORKING!
RRTC13th Russia & CIS Refining Technology Conference
IGTC5th International Gas Technology Conference
RPTC12th Russia & CIS PetrochemicalsTechnology Conference
epc.indd 1 10/06/2013 11:17
Preventing ingress of HCN into amine systems
Hydrogen cyanide (HCN) has far-reaching effects on amine system performance. After
hydrocarbon contamination, its presence is one of the primary reasons refinery amine systems suffer from accelerated corrosion, and operability and reliability prob-lems. When HCN enters the amine system, its hydrolysis produces ammonia and formate, a heat stable salt (HSS). Reaction of HCN with oxygen and H2S generates another HSS, thiocyanate. HSSs are known to chelate iron, and the subsequent accelerated corrosion leads to faster formation of particulate iron sulphide, which, in turn, leads to filter element plugging, fouled equipment, lower capacity and more stable foams. Mass transfer rate-based simulation is used in this article to investigate the prevention of HCN ingress into amine systems via water washing.
Background and contextOver any substantial period of operation, amine systems operated in refineries can be expected to show an increasing build-up of HSSs. This is especially pronounced for units that scrub gases primarily from coking and catalytic cracking units, where the HSSs consist primarily of formate (HCOO-) and thiocyanate (SCN-). Amine systems treating gases from various sources experience different rates of HSS accumulation. The actual rate of HSS anion build-up depends upon the incursion rate of HCN into the amine system. The primary sources of HSSs are summarised in Table 1, from which a direct chemistry link between HCN and the common
Simulation studies provide guidelines on how to avoid contamination of refinery amine systems by hydrogen cyanide
RALPH WEILAND, NATHAN HATCHER and CLAYTON JONES Optimized Gas Treating, Inc
HSS anions formate and thiocy-anate is apparent.
If left unchecked, the build-up of HSSs eventually neutralises permanently part of the amine by protonation and results in loss of treating solution capacity. This is the primary acute symptom. HSSs are also known to complex iron
ion and accelerate corrosion in the hot, lean section of the amine unit. When the complexed iron contacts higher concentrations of H2S in the absorber, iron sulphide precipi-tates. These particles can foul equipment and stabilise foam, leading to a loss of hydraulic capacity, so the operator usually
resorts to trading these costs for the cost of replacing filter elements. The costs of equipment failure vary with the failure mode, are highly site specific and the timing of the lost production can matter significantly. In general, lost profit from equipment failure can be expected to dwarf filter costs. It is unfortunate, however, that costs such as these are often ignored from the economic plan-ning process because they are difficult to quantify with certainty. The industry is generally not forth-coming with reporting minor mishaps because there is little business incentive to be open about potentially embarrassing operating episodes or revealing mistakes that can amount to signif-icant competitive advantages when solved.
Hydrogen cyanide is a byproduct of cracking the heavier fractions of crude oil in a refinery (either ther-mally as in a coker, or catalytically as in a fluid catalytic cracker). The gas oil (boiling point 750°F/399°C+) and heavier fractions tend to have greater concentrations of nitrogen
www.eptq.com PTQ Q3 2013 103
Table 1
Cyanides from treating cracked stocks (FCC units and cokers)
Oxygen incursion (FCC units and vacuum tower off-gas)
SO2 breakthroughs (Claus TGUs)
Sources of HSS incursion
Table 1
If left unchecked, the build-up of HSSs eventually neutralises permanently part of the amine by protonation
ogrt.indd 1 07/06/2013 20:14
104 PTQ Q3 2013 www.eptq.com
Some processes are high producers; others do not seem to produce HCN at all. Once produced, HCN finds its way into the amine system with the H2S-containing gases. HCN forms in various processes within a refinery, whereas HSSs form in the amine system. Once in the amine system, various conditions and the presence of
than the diesel and lighter fractions. These processes break up the larger nitrogen-containing molecules at high temperatures and low hydro-gen partial pressure conditions that may not be as conducive to
complete conversion of byproduct molecules such as HCN to ammo-nia as in a high-pressure hydrotreater or hydrocracker.
Thus, HCN occurs quite naturally in refineries and has many sources.
Wash column
Mixer
Tear stream
Raw gas
Make up
To amine unit
24
4
2
27
31
30
1
29Pump Divider
Purge to SWS
3
Figure 1 Water washing with recirculation
Temperature, °F 100Pressure, psig 180Flow, MMscfd 55
Composition H
2S, mol% 12.36
HCN, ppmv 100NH
3, ppmv 1000
Methane, mol% 36.14Ethane, mol% 9.27Propene, mol% 2.06Propane, mol% 2.061-Butene, mol% 1.03n-Butane, mol% 1.03Hydrogen, mol% 36.04
Raw gas conditions and analysis
Table 2
www.eptq.com PTQ Q2 2013 61
equipment. During shutdown, all fuel inputs to the furnace, except for pilots, must be isolated. Pilots are kept alight as a safety precau-tion so that fuel does not accumulate and lead to an explo-sion. Low process fl ow through the tubes is a good example of “heat off” shutdown.
The very poor condition of the furnace refractory has resulted in
and shutdown. In order to protect equipment, operational procedures for startup and shutdown should be followed completely, and a purge sequence should be carried out to remove fuel from the fi rebox before a burner is ignited.
Safe operationThe primary consideration is safety for operating personnel and
essential for optimum furnace effi -ciency. If necessary, a new analyser should be installed.
The main hazard for furnaces is gas accumulation in the fi rebox and explosion. Explosion will occur if a source of ignition is introduced into a fi rebox containing a fl ammable mixture of fuel and air. The most hazardous periods during the oper-ation of furnaces are during startup
Figure 6 Percentage of oxygen analyser after installation of furnace control
Figure 7 Tag plot for total temperature of furnaces
Triple eccentric
Butterfly Valves
Check Valves
Double Block and Bleed
ESD Valves
www.zwick-armaturen.de
Double Block and Bleed
de
tupras.indd 4 08/03/2013 13:30ogrt.indd 2 10/06/2013 17:52
other contaminants allow the HCN to be converted into HSS anions. An ounce of prevention is worth a pound of cure: HCN is a precursor to HSS formation and its removal from the raw gas entering the amine unit would virtually elimi-nate all the problems associated with HSSs. In other words, the most obvious way to prevent HSS anion formation in amine systems is to prevent the ingress of HCN in the first instance. This thinking has led refiners to try various schemes to remove HCN before it gets into the amine unit. One approach that has been tried repeatedly by many refiners is water washing the raw gas and sending the wash water to a sour water stripper (SWS). Despite best efforts, however, this approach seems invariably to fail because the amine system continues to experience the build-up of HSS anions in operation.
Water washing refinery gas toremove HCNMost applications of water wash-ing appear to be founded upon the thinking that only a relatively small flow rate of wash water ought to be sufficient to remove HCN from a large gas flow. However, to ensure realistic tray or packing hydraulics, the tendency has been to use a large flow of wash water recirculating through the wash column, with a small addition of fresh water and a corresponding small purge of water from the tower bottoms stream. Figure 1 shows this scheme with its small addition, small purge and large recirculation. Using a small purge flow rate, which is added to the regular sour water flow to the SWS, prevents the sour water system from being overloaded.
The specific example for this study uses a typical gas intended to be treated for H2S removal in an amine system. Table 2 shows the raw gas analysis and its condi-tions of temperature, pressure and flow rate. The water wash column contains 20 valve trays sized for 70% of jet and down-comer flood.
Apart from water and H2S, the components HCN and ammonia are of most interest. They are all modelled by the ProTreat simula-tor based on their mass transfer rates, not solely on assumptions about equilibrium between phases. This approach to column simula-tion accounts not just for vapour-liquid equilibrium, but also for the effects of chemical reaction rates and the mass trans-
fer characteristics of the tower internals on the rate of separation of all the components whose pres-ence is important to the gas purification process. In the present context, HCN absorption into the wash water is of particular concern.
There are two distinct cases to consider: in the first, the total water recycle flow rate (Stream 31 in Figure 1) is kept constant at 100
Fresh makeup water flow, gal/min % HCN removed % NH3 removed % H
2S removed
5 3.87 79.8 0.6610 7.05 92.6 0.8820 13.3 97.8 0.9350 30.4 99.6 1.18100 51.9 99.9 1.56250 81.7 100.0 2.59
Impurities removal with fresh water added to 100 gal/min recirculating flow
Table 3
Fresh water flow, gal/min % removal ppmw in sour water HCN NH
3 H
2S HCN NH
3 H
2S
20 13.3 100 0.95 206 9730 2280050 32.3 100 1.09 202 3950 10700100 63.2 100 1.42 199 1980 6940150 88.0 100 1.76 185 1320 5770200 97.2 100 2.10 153 990 5140250 99.8 100 2.40 126 95 4780
Impurities removal by washing with once-through water
Table 4
Fresh water flow, gal/min 1000 ppmvNH3 No NH
3 in gas
HCN H2S HCN H
2S
5 3.87 0.66 3.14 0.03910 7.05 0.88 6.27 0.07820 13.3 0.93 12.5 0.15850 30.4 1.18 30.0 0.393100 51.9 1.56 51.7 0.774250 81.7 2.59 81.8 1.86
Effect of ammonia on % HCN and % H2S removal by recycled water wash
Table 5
Fresh water flow, gal/min 1000 ppmv NH3 No NH
3 in gas
HCN H2S HCN H
2S
20 13.3 0.95 12.6 0.15850 32.3 1.09 31.4 0.394100 63.2 1.42 62.7 0.788150 88.0 1.76 88.4 1.18200 97.2 2.10 97.2 1.52250 99.8 2.40 100 1.86
Effect of ammonia on % HCN and % H2S removal without water recycle
Table 6
www.eptq.com PTQ Q3 2013 105
ogrt.indd 3 10/06/2013 17:52
ammonia removal can be achieved, completely preventing ingress of ammonia into the amine-treating unit. This will also result in correspondingly less ammonia contamination of the amine acid gas feeding a sulphur recovery unit downstream. In contrast to the recycle case, the same 250 gal/min of water used in a fresh-water-only case will remove virtually all the HCN and leave H2S as the only impurity entering the amine system.
The decision about what fresh water flow rate to use depends on:• The ability of the sour water system to handle the additional sour water load• Whether there is a minimum acceptable acid gas content of the sour water• Whether, below certain acid gas levels, the water can be used else-where before being added to the refinery’s sour water system
poor unless very high water makeup rates are used, and even then performance is marginal. The next question then is whether using scrubbing with fresh water alone (no water recycle) can achieve a better result.
Table 4 shows the performance of the wash column when scrubbing is done using fresh water alone — no water recycle. One advantage to using just fresh water is immedi-ately obvious: virtually 100%
gal/min, to which 5 gal/min of fresh water is added as makeup and later purged. The other case involves no wash water recycle at all — only fresh water is used to feed the wash column. Table 3 shows the effect of adding various fresh water flows to the 100 gal/min recycled flow on HCN and ammonia removal. Table 4 shows the effect of eliminating water recy-cle altogether and pretreating the gas with various flows of fresh wash water only.
At fresh water rates of more than 50 gal/min added to the basic 100 gal/min of water recycle, the use of recycle at all is of questionable hydraulic performance benefit because it certainly is not needed to keep trays functioning properly from a hydraulic standpoint. Above 20 gal/min water makeup, ammo-nia removal is hardly affected and H2S removal is rather minimal in any case. HCN removal is rather
◗ www.digitalrefi ning.com is a new, easy-to-use website for the refi ning, gas and petrochemical processing industries.
◗ A simple search facility enables you to fi nd and display relevant information across the whole spectrum of articles, literature, news, events and companies.
◗ Readers can download technical articles and other information from over 40 categories.
◗ Find relevant industry events worldwide.
◗ Read news items focusing on contracts, company news and new products.
dig ref half page ad.indd 1 8/3/12 09:42:49
Prewashing with water is an excellent way of removing ammonia from the gas before it enters the amine system
106 PTQ Q3 2013 www.eptq.com
ogrt.indd 4 10/06/2013 17:52
www.eptq.com PTQ Q3 2013 107
corresponding impacts on the amine system. Mass transfer rate-based simulation both confirmed and quantified previously anecdotal observations that recir-culating-type water wash systems are, at best, only marginally effec-tive in removing HCN as a feed contaminant. For a significant reduction in HCN ingress into refinery amine systems, fresh makeup water is required, and a lot of it. An economic trade-off between sour water stripping capacity and energy usage versus amine system operating costs (HSS clean-up, filtration, capacity and reliability) will ultimately dictate
whether water washing is an appropriate choice for a specific refinery system
Ralph Weiland is a co-founder of Optimized Gas Treating in Clarita, Oklahoma, US. He holds BASc, MASc and PhD degrees in chemical engineering from the University of Toronto, then spent two years as a post-doctoral fellow in applied mathematics at the University of Western Australia. Email: [email protected]
Nathan Hatcher joined Optimized Gas Treating, Buda, Texas, as Vice-President, Technology Development, in 2009. He holds a BS in chemical engineering from the University of Kansas and is currently a member of the Amine Best Practices Group. Email: [email protected]
Clayton Jones joined Optimized Gas Treating, Inc as a Software Development Engineer in 2012. He holds a BS in chemical engineering from McNeese State University and a MS in chemical engineering from the University of New Mexico.
• The acceptable build-up rate of HSS anions in the amine system.
Regardless of the extent of HCN removal, prewashing with water is an excellent way of removing ammonia from the gas before it enters the amine system. But, more importantly, this approach can be used to remove as much of the HCN as the SWS is capable of handling within battery limits. Gas washing will also reduce other impurities such as chlorides that manage to enter with the raw gas, although this is beyond the scope of the present study.
Since ammonia is a weak base and HCN a weak acid, there is a question as to the role ammonia might play in HCN removal. Table 5 compares the extent of both HCN and H2S removal with and without ammonia in the sour feed gas going to the wash column. This case is with 100 gpm of recy-cle water flow. Table 6 refers to the same situation when there is scrubbing using once-through water. It is readily apparent that regardless of whether there is wash water recycle or not, the presence of ammonia only very marginally improves HCN removal, but it significantly increases H2S pickup in the wash water. H2S is quite a bit stronger acid than HCN, and it preferen-tially reacts with ammonia — HCN seems hardly to react at all.
Gas prewash using only fresh water appears viable for removing contaminants whose presence in the amine system is undesirable. However, using water wash with recirculation and only relatively modest blow-down does not appear to be a viable strategy. This is in agreement with the experience of many refiners who report that high levels of HSS anions continue to build up in the amine system even after imple-menting the water wash with recycle.
Summary and conclusionsThe production of HCN in refiner-ies has been reviewed with an emphasis on HCN’s role in HSS anion formation and the
Gas prewash using only fresh water appears viable for removing contaminants whose presence in the amine system is undesirable
Our knockout drum needs a mist eliminator!
Can we add one without overhauling the vessel?
IT’S A MORE COMMON problem than you might think. A vertical knockout drum removes free liquid from a certain gas stream. But at the time the plant was built, a mist eliminator was not considered necessary. Now mist is carrying over and causing trouble downstream. There is no manway, so adding a conventional mist eliminator would require cutting the vessel open.
Instead, if the fl anged exit nozzle is at least six inches in diameter, you can simply hang a patented MistFix® mist eliminator inside, securing it between the fl anges. Even when there is a manway, MistFix avoids entry, welding, and recertifi cation in retrofi ts. For new and old vessels, it drastically cuts installation cost and downtime.
Phone: 713-434-0934 • Fax: [email protected]
24-hr Emergency Service: 281-716-1179
MistFix® insertion mist eliminator
Visit our new website atwww.amacs.com
ogrt.indd 5 10/06/2013 17:53
GET A RETURN ON INVESTMENT WITHIN 24 HOURS
Using ITW Online Cleaning on a proactive way will help improve TURNAROUND OPERATIONS and OPERATIONAL EXCELLENCE An entire Process Unit can be ITW Online Cleaned in as low as 24 hours on an oil-to-oil basis. Our cleaning technology will clean all of the Unit’s equipment quickly, safely and environmental friendly. To recover a Units’ performance, ITW Online Cleaning can be applied in a Turnaround during shutdown operations, to reduce/eliminate the need for mechanical cleaning, or during the run. Furthermore, use of our cleaning technology eliminates the need to open or enter hazardous process equipment.
ITW Online Cleaning is the only technology which has an immediate Return On Investment (ROI).
By using ITW Online Cleaning during the run, to recover Units’ performance, at Unit start-up (just after 24h, ITW Technology has been applied), all the benefits are immediately monetized in terms of:- improved throughput- improved energy efficiency- improved product quality- reduced emission taxes
Regular application of ITW Online Cleaning will target an increased run length under clean conditions, with a far higher ROI.
For Turnaround applications, a further ROI increase can be added by applying ITW Improved Degassing/Decontamination to achieve quick and effective safe entry conditions.
EVALUATE ITW ONLINE CLEANING TODAY TO IMPROVE YOUR PLANT’S PROFITABILITY !
For more informations contact : ITW S.r.l.- C.da S.Cusumano - 96011 Augusta - Italy Tel. +39 (0931) 766011 E-mail: [email protected] www.itwtechnologies.com
Join ITW Team worldwide and send your Curriculum Vitae to : [email protected]
ITWInnovative
TechnologiesWorldwide
Now hiring Professionals Worldwide
itw.indd 1 11/9/12 11:45:38
Optimised hydrogen production by steam reforming: part 2
Stringent emission control legis-lations and a well-informed market have compelled refiners
to build highly complex refineries for producing cleaner and more effi-cient fuels. With crude slates getting heavier and sour, the demand for hydrogen is rapidly increasing and refiners are looking into new reforming technologies and schemes for optimum hydrogen production. In a previous article (PTQ, Q1 2012), we identified and optimised process parameters that affect the energy performance in hydrogen plants with conventional steam methane reforming (SMR) technology. Autothermal reforming (ATR) and gas heating reforming (GHR) are among the new technologies availa-ble at present, and future developments are focused on economical and environmentally friendly plant operations in a
Emerging technologies for hydrogen production from steam reforming are modelled
KEdar V PaTWardHaN, SaNKE raJYaLaKSHMI and P BaLaraMaKrISHNa Larsen & Toubro
sustainable manner. Process schemes such as GHR in series or parallel combination with SMR and ATR are gaining prominence, as these are expected to have lower capital and operating costs and plot space requirements, as well as a reduction in flue gas emissions.
In the present study, a mathemat-ical model is developed for a greater understanding of the func-tioning of GHR. A simulation model is then developed to opti-mise the performance of a SMR/ATR + GHR combination and the results are validated with reference data. The model is used to study the effect of operating parameters such as S/C and O2/C ratios, the extent of feed preheating and the natural gas feed split on specific energy consumption and export steam production. Performance comparisons indicate that a saving
in natural gas consumption and a reduction in flue gas emissions of 6-7% is possible with new reform-ing technologies and process schemes when compared to a conventional SMR configuration for refinery hydrogen generation.
Increased hydrogen demandIn the last decade, the worldwide refining industry has been impacted by several trends that have increased hydrogen demand significantly. First, in aggregate, crude oil has been getting heavier and contains more sulphur and nitrogen; second, a decreasing heavy fuel oil demand requires more bottoms upgrading; and, third, increasingly stringent envi-ronmental regulations require cleaner transportation fuels produc-tion. These factors have led to a substantial increase in hydrogen
www.eptq.com PTQ Q3 2013 109
Feed pretreatment ATR CO-shift
converterH2
purification
Process condensate
stripper
NG from BL
Steam export
BFW from BL
Air
Fired heater(feed preheating)
O2/air
H2 to BL
WHR-I WHR-II
PSA off-gases
Figure 1 Schematic of hydrogen production by autothermal reforming
l & T.indd 1 11/06/2013 10:11
110 PTQ Q3 2013 www.eptq.com
allows for adjustment of the operat-ing parameters to obtain the perfect balance between the size of GHR and combined outlet composition.
Process scheme: GHR with ATRThe purpose is to increase the ratio between steam reforming and partial oxidation so that the synthe-sis gas will have a higher hydrogen-to-carbon monoxide (H2/CO) ratio than ATR alone. This will also result in a reduced high-cost oxygen consumption and load on the shift section. In a series arrange-ment, all gas passes through the steam reforming unit and then through ATR. This will mean that the reforming catalyst may set a lower limit for the S/C ratio. In a parallel arrangement, the two reformers are fed independently, giving the freedom to optimise the S/C ratio individually. However, the gas heated reformer must oper-ate at a higher temperature than in a series arrangement in order to obtain a low methane concentration in the synthesis gas. The GHR arrangement in parallel with SMR/ATR is shown in Figure 2.
A comparison between various reforming schemes is shown in Table 1.
Basis of simulationModel development and process simulation are carried out on the following basis: • Impurities in the feed are effi -ciently removed in the feed treatment stage and hence do not infl uence reforming reactions• The calorifi c value of natural gas available at the battery limit is 11 211 kcal/Nm3 with a methane
consumption in a refi nery complex. Generally, complex refi neries source 30-60% of their total hydro-gen requirements from on-purpose hydrogen capacity. Overall, approx-imately 95% of the on-purpose hydrogen is supplied by steam reforming of light hydrocarbons.
The hydrocarbons such as natural gas (mainly methane) up to naph-tha and refi nery off-gas can be converted into hydrogen by either steam reforming technology or through partial oxidation and a combination thereof. The future advances in reforming technology are focused primarily on reducing energy consumption and stack fl ue gas emissions. This article evaluates various reforming technology options available on an industrial scale, compares the performance parameters based on a simulation model and suggests an optimal confi guration.
Autothermal reformingThe ATR unit is a refractory-lined pressure vessel containing a burner, a combustion chamber and a cata-lyst bed. The hydrocarbon feedstock is mixed with steam and pure oxygen, enriched air and airat the top of the reactor. In the combustion chamber, partial
oxidation reactions take place and the generated heat is utilised within for endothermic steam reforming reactions. In the lower section of the reactor (loaded with reforming catalyst), the steam reforming and shift conversion reactions occur as the gas passes through the fi xed bed, generating a gas mixture of H2 and CO. A general schematic of ATR is shown in Figure 1.
Gas-heated reformerGHR uses the heat available in the process gas at the reformer exit for steam reforming in a heat exchanger type of reactor. This scheme is avail-able in series or parallel combination with SMR or ATR. In a parallel combination with the available heat, up to 20% of the feed can be split and taken to GHR. Normally, export steam production is minimum in these confi gurations.
Process scheme: GHR with SMRSince the outlet temperature of GHR is less than the SMR outlet tempera-ture, the methane slip is higher in GHR. The higher methane slip can be counteracted by adjusting the steam-to-carbon (S/C) ratio and the inlet temperature to GHR. This option is available in the case of a parallel arrangement only, as it
ATR + GHR SMR + GHR
ATR SMRGas heated reformer
Gas heated reformer
Oxygen
Natural gas
Steam
Syngas
Natural gas
Steam
Syngas
Figure 2 GHR arrangement in parallel with ATR/SMR
Parameters SMR ATR SMR + GHR ATR + GHRSteam export High High Some or none Some or noneOxygen/enriched air No Yes No YesNatural gas consumption High High Low MediumRelative stack emission High High Medium MediumPlot space required High Low High Medium
Comparison of various reforming schemes
Table 1
l & T.indd 2 11/06/2013 10:11
content of 72% and a pressure of 37.5 bar• The hydrogen product flow raterequirement is 100 kNm3/hrwithapurity of >99.5%• The kinetics of the reforming and shift reactions is taken from the open literature and is not specifi-cally applicable to any particular catalyst• The final purification is achievedbypressureswingadsorptionoper-ating with a working efficiency of89%• A HT shift and LT shift reactor configuration is considered for thewatergasshiftreaction.
For the SMR scheme: • Ten per cent excess air is consid-ered in the reformer burner• The reformer outlet temperature isfixedat920°C• S/Cratioisconsideredas2.5• PSA off-gas is recycled back as supplementary fuel for the reform-ing section.
For the ATR scheme: • Oxygen purity is considered as >99%• The reformer outlet temperature isfixedat1050°C• S/CandO2/Cratio iskeptmini-mum and adjusted to match the required reformer outlet temperature• PSAoff-gasisfiredinaheaterforthe feed preheating requirement.
For the GHR scheme: • The natural gas feed is split (based on the heat content of the reformer exit gas)• The GHR outlet temperature for thecombinedgasisfixedat820°C• The S/C ratio is maintained at3.0.A simulation model is developed
for the hydrogen plant flow sheetfor four process schemes (SMR, ATR, GHR in parallel combination withATRandSMR)usingcommer-cial steady-statesimulationsoftwareAspen Hysys 7.3. Model develop-ment of ATR and validation withreferencedataisdescribedbelow.
Simulation model for ATR ATRismodelledas tworeactors inseries. In the first reactor, themixture of hydrocarbons is partially burnedasper the followingchemi-cal reaction. In addition to that,
www.eptq.com PTQ Q3 2013 111
complete oxidation of CO and H2 willcompete:
CmH
n + (m + n/
2)O
2 ⇔ mCO + (n/
2) H
2O
CO + (½) O2 ⇔ CO
2
H2 + (½) O
2 ⇔ H
2O
In the second reactor, leftoverhydrocarbon after partial oxidation is reformed and a shift reaction takesplaceover thecatalyst,asperthefollowingchemicalreaction:
CmH
n + mH
2O ⇔ mCO + (m + n/
2) H
2
CO + H2O ⇔ CO
2 + H
2
The exit temperature of the ATR reactor is maintained at a constant value, thereby fixing methane slipby adjusting the inputs of steam and oxygen. Using the simulation model, the reactor exit conditions in terms of composition and temperature are predicted and compared, with reference dataavailable in the literature. Thecomparison details are given inTable 2. The marginal deviation inthe simulated results may be due to the kinetic data considered for
31500
32000
31000
30500
NG
consu
mpti
on,
Nm
3/h
r
30000400 450 500 550 600 650
Feed preheat temperature, ºC
32000
36000
40000
28000
24000 Ste
am
exp
ort
, kg/h
r
20000
Steam exportNG consumption
Figure 3 Effect of feed preheat on natural gas consumption in ATR
4
6
5
3
2
H2C
O
10.5 1.0 1.5 2.0 2.5 3.0 3.5
S/C
O2/C = 0.45
O2/C = 0.65
O2/C = 0.50O2/C = 0.55O2/C = 0.60
Figure 4 Optimum O2/C ratio for a given S/C ratio
Reference Simulation Input Natural gas feed, Nm3/h 20 000P
exit, bar 23.8
Texit
, °C 1050ResultsH
2O/C 0.60 0.64
O2/C 0.64 0.54
H2/CO 2.4 2.1
H2 + CO, dry mol% 89.6 93.5
CO2 + CH
4, dry mol% 6.6 6.0
Comparison between reference and simulation
Table 2
l & T.indd 3 11/06/2013 10:11
112 PTQ Q3 2013 www.eptq.com
which, in turn, increases the load on the purification stage. From an oper-ational point of view, in order to reduce the possibilities of soot formation, the S/C ratio is kept high for a higher O2/C ratio.
Parametric analysis for a GHR + SMRcombinationFeed split ratioThe higher feed split means that more natural gas will bypass the SMR unit. This has two impacts: the bypassed gas does not require fuel to be fired, so there is a saving in terms of natural gas as fuel, but, at the same time, energy in the effluent of the reformer is not avail-able for heat recovery and for steam production. Hence, with an increase in the GHR feed ratio, natural gas consumption as fuel goes down along with steam production from the plant.
Results and discussionReforming is a highly endothermic process and needs natural gas as feed as well as fuel for the produc-tion of hydrogen. An approximately 90% contribution to total specific energy consumption is by natural gas alone. Credit is given for export from the hydrogen plant to the high-pressure steam generated within the plant and off-gases if these are not utilised within the plant. The simulation results for the various reformer configurations are analysed in terms of product composition, total natural gas consumption and steam produc-tion, and the details are given in Table 3.
Natural gas consumptionThe hydrocarbon feed requirement will be identical for the same production of hydrogen, while the amount of hydrocarbon required as fuel will vary, depending upon the choice of reforming technology. In schemes with GHR, the fuel saved is directly proportional to the amount of feed that has bypassed SMR/ATR. Hence, a considerable amount of natural gas can be saved. The maximum amount of feed that can be taken to GHR depends upon the heat content in the reformer exit gas, and it is possible to take
catalyst zone. Thus, a higher outlet temperature and lower methane slip is obtained at the outlet of the ATR
unit. Higher O2/C ratio operation consumes more natural gas for the same quantity of hydrogen produc-tion. Also, more combustion increases the proportion of CO2,
partial oxidation and catalytic reforming reactions.
Parametric analysis for ATRPreheat temperatureHigh preheat temperatures reduce oxygen consumption, thereby lowering the content of CO2 in the product gas and increasing the H2/CO ratio. This will reduce natural gas consumption and steam export. The effect of feed split to GHR on natural gas consumption and steam export is also analysed.
S/C and O2/C ratios
The higher the O2/C ratio, the more energy will be produced in the ATR unit, which in turn increases the extent of steam reforming in the
30000
31000
29000
28000
NG
consu
mpti
on,
Nm
3/h
r
270000 5 10 15 20
GHR feed ratio, %
32000
36000
40000
36000
40000
28000
24000 Ste
am
exp
ort
, kg/h
r
20000
Steam exportNG consumption
Figure 5 Effect of GHR/SMR feed ratio on natural gas consumption and steam production
Process parameters SMR ATR SMR + GHR ATR + GHRHydrogen production, KNm3/h 100 Oxygen requirement, TPD - 722 - 549Natural gas feed split to GHR, % - - 20 20Natural gas consumption, Nm3/h Natural gas feed 25 200 30 000 24 000 28 500Natural gas fuel 5100 - 4150 -Total natural gas 30 300 30 000 28 150 28 500Relative total natural gas consumption, % 100 99 93 94Synthesis gas composition at reformer exit, dry mole basisH
2/CO ratio 4.0 2.3 3.9 2.6
H2 + CO,% 88.8 92.1 89.9 91.4
CO2 + CH
4, % 10.8 7.4 9.8 8.2
Steam balance mass flow, kg/hSteam flow at reformer inlet 70 985 34 101 70 868 45 315Steam added at shift reactor inlet 0 28 131 0 16 879HP steam export 33 221 33 000 0 3000Steam consumption - turbines 15 438 25 730 12 141 19 285Total steam production 119 644 120 962 83 009 84 479 Relative flue gas emission, CO
2%
100 100 93.7 95.3
Process parameters comparison
Table 3
Higher O2/C ratio
operation consumes more natural gas for the same quantity of hydrogen production
l & T.indd 4 11/06/2013 12:21
www.eptq.com PTQ Q3 2013 113
natural gas feed up to 20% for both ATR + GHR and SMR + GHR schemes, restricting the combined methane slip at the exit to 3%.
Synthesis gas compositionIn the ATR unit, the amount of CO produced due to the partial oxida-tion of hydrocarbons is higher than that formed due to reforming reac-tions. Hence, the H2/CO ratio achieved in ATR is 2.3-2.6 compared to a H2/CO ratio of ~4 in SMR. This results in an additional load on shift reactors for hydrogen generation. The composition of CO2 is less in the ATR outlet stream as compared to SMR. This is because a higher operating temperature in ATR restricts the exothermic water shift reaction. In ATR, the amount of oxygen fed stoichiometrically consumes the hydrogen part of the feed hydrocarbon not contributing to hydrogen production. Even though the process steam required for reforming at the inlet of the ATR unit is less when compared to SMR, additional steam is needed at the inlet of the shift section for a similar CO slip.
Steam production Ideally, the supply of energy should be just sufficient for fulfil-ment of the energy requirement of the reforming reaction. Due to system limitation, higher energy must be supplied and this is recov-ered through steam generation in the plant. Steam is generated in the heat recovery sections between reforming and shift and in between shift reactors. This steam is inter-nally consumed in stripping dissolved gases from generated process condensate, steam turbine drives and process steam.
Distinctively, the choice of reforming technology depends upon the steam requirement of the entire refinery complex. Even under highly optimised conditions, it is difficult to reduce the steam production below a certain amount for SMR/ATR schemes. The excess steam produced within the hydro-gen plant battery limit is at the expense of natural gas as fuel. Steam production can be reduced by adding GHR, where a larger
Process parameters SMR ATR SMR + GHR ATR + GHRRelative total natural gas consumption,% 100 88 91 84Relative CO
2 emissions (feed preheating only), % 100 62 81 58
Relative natural gas consumption: PSA off-gas export
Table 4
portion of energy in the reformer product stream is utilised for reforming. This reforming is achieved without firing natural gas as fuel.
PSA off-gas exportPSA off-gas is used as a fuel for feed preheating in a reformer or in a separate fired heater in ATR. If PSA off-gas is not utilised in the plant, total natural gas consump-tion will increase to account for the feed preheating requirement. The details of relative total natural gas consumption and CO2 emissions for feed preheating only are shown in Table 4.
ConclusionSMR is a well-proven technology conventionally used in refineries for hydrogen generation. With rapidly increasing hydrogen demand and volatile natural gas prices, refineries are looking into alternative technologies for opti-mum hydrogen production. New process schemes such as the auto-thermal reformer and the gas-heated reformer in series or parallel combination with SMR or ATR are gaining prominence, as they offer potential in reducing energy consumption and flue gas emissions.
ATR is simpler and more compact than steam reforming.
However, the cost of oxygen supply makes it less attractive than SMR even for large-scale plants. The H2/CO ratio achieved is ideally suited for methanol and Fischer-Tropsch synthesis.
GHR utilises the energy in reformer effluent for hydrogen production, thus saving on natural gas as fuel at the expense of steam production. It is ideally suited to projects where no steam export is required.
A simulation model is developed for the hydrogen plant flow sheet with various reformer process schemes, and it is observed that savings in total natural gas consumption in the range of 6-7% are possible with GHR schemes at the expense of export steam production.
The scheme with a gas-heated reformer in parallel combination with SMR is a good option for new refinery hydrogen plants. The scheme of GHR with ATR may be considered in future when a low-cost oxygen source is available and sufficient operational experi-ence is gained in ATR technology.
Kedar Patwardhan is a Senior Process Engineer specialising in simulation,modelling, optimisation and process design with the Chemical Engineering Group of Larsen & Toubro. He holds a PhD from the Institute of Chemical Technology, Mumbai. Email: [email protected] Sanke Rajyalakshmi is a Process Engineer specialising in simulation with the Chemical Engineering Group of Larsen & Toubro, Powai, India. He holds a degree in chemical engineering from the National Institute of Technology Warangal. Email: [email protected] P V Balaramakrishna is Head of the Chemical Engineering Group of Larsen & Toubro, specialising in process design, advanced process control, commissioning, troubleshooting and optimisation of process plants. He holds a master’s degree in chemical engineering from the Indian Institute of Technology, Kanpur. Email: [email protected]
The scheme with a gas-heated reformer in parallel combination with SMR is a good option for new refinery hydrogen plants
l & T.indd 5 11/06/2013 10:12
Treating SolutionsTreating SolutionsTreating SolutionsProvenProven
Results
Merichem Company provides a diversi�ed portfolio of products and services to a wide range of
operations, including the re�ning and petrochemicals industries. With more than 1,000 licensed
process units worldwide, Merichem is the leader in licensing patented process technologies and
supplying proprietary equipment to provide hydrocarbon treating solutions and by-product
management services that remove sulfur and other impurities from hydrocarbon liquids and gases.
SCAN THE QR CODE WITH YOUR MOBILE DEVICE TO SEE OUR COMPLETE PORTFOLIO OF TREATING SOLUTIONS.
C
M
Y
CM
MY
CY
CMY
K
Merichem Company.pdf 1 5/22/2013 8:09:34 AM
merichem.indd 1 06/06/2013 10:23
Corrosion control with high-acid crudes
Refinery cost pressures and the increasing availability of heavy, high-acid crudes at
preferential discounts are causing substantial changes in global refin-ing practices. Heavy crude production is increasing and high-acid crudes are a growing proportion of the total. This is not a short-term trend. Refineries are being built near the sources of these crudes with conversion capabilities designed to maximise the produc-tion of attractive liquid fractions.
This article explores high-acid crude corrosion mechanisms, compares strategies to combat corrosion, and concludes that moni-toring and chemical treatment with an appropriately chosen high- temperature corrosion inhibitor (HTCI) can be a cost-effective tool.
Growing importance of high-acid crudesIncreasing volumes of heavy crudes are being produced globally, and high-acid crudes are a growing proportion of the total. Their acid-ity has influenced refinery design and construction at a number of locations where bottom-of-the- barrel conversion is central to refin-ery operation. Figure 1 shows growth projections for heavy crude in key producing regions.
A study by Purvin & Gertz found that high total acid number crude oil supplies, which are included in the estimates of total heavy crude oil, will see a near-term surge as new projects come on stream. Much of this increase will be from Africa, particularly Angola and Sudan. Production increases from Brazil and China will also contribute to
Appropriate corrosion control strategies help refineries to profit from discounted high-acid crudes
INDIA NAGI-HANSPAL, MAHESH SUBRAMANIYAM and PARAG SHAH Dorf Ketal Chemicals
increases in high-TAN output in coming years. These anticipated changes in incremental crude oil supply origination and quality volumes will affect trade flows throughout the world.1
Table 1 contains selected proper-ties of various high-acid crudes from across the globe. Many of these crudes are included in regular
refinery crude diets due to their positive price differential in comparison to premium crudes.
The ability to process these lower-cost, lower-quality crudes can improve refinery profitability to such a significant extent that refin-ers are investing in advanced metallurgy, expanding their blend slates, and evaluating chemical
www.eptq.com PTQ Q3 2013 115
Table 1
Middle EastNorth AmericaAfrica
Latin AmericaRest of world
15
20
10
5
Mill
ion b
/d
02010 2015 2020
Figure 1 Heavy crude oil production1
Characteristics Country API gravity, Total sulphur, TAN, °API wt% mgKOH/gmDoba Chad 20.9 0.127 4.35Dar Sudan 24.54 0.114 4.10Roncador Heavy Brazil 18.4 0.665 2.51Heidrun Norway 26.5 0.5 2.5Albacora Brazil 19.4 0.55 2.37Lokele Cameroon 20.6 0.43 2.21Baobab Gabon 22.5 0.386 2.19Grane Norway 19.0 0.86 2.1Kuito Angola 21.6 0.71 1.66Dalia Angola 23.0 0.51 1.57Merey 16 Venezuela 17.9 2.694 1.51PDVSA 22 Venezuela 21.7 2.15 1.46
Example high-acid crudes and their properties
Table 1
dorfketal.indd 1 10/06/2013 16:23
116 PTQ Q3 2013 www.eptq.com
1920s, and the earliest research into effective solutions did not take place until about 30 years later, when it was discovered that oils containing signifi cant levels of naphthenic acid caused direct attacks on system metallurgy at moderate to high temperatures in non-aqueous media. Indirect attacks were also observed from low molecular weight acids generated through thermal degradation.
The main corrosion problems from direct attacks by naphthenic acids were observed in the atmos-pheric distillation and vacuum distillation units. Atmospheric furnace curves and expansion zones sustained damage, as did portions of the atmospheric tower heavy diesel and atmospheric resi-due zones. Damage from the indirect attacks by acetic, butyric and propionic acid were observed in overhead system aqueous media.
From this and subsequent research, several rules of thumb became well accepted. Generally speaking, when TAN falls between 0.3 and 0.5, minimal naphthenic acid corrosion is expected, but when TAN exceeds 0.5, naphthenic acids begin to cause problems. Since they are soluble in the oil phase, naphthenic acids tend to be carried along with the crude and cuts to downstream units that can be protected with high-temperature corrosion inhibitors.
There are more than 1500 types of naphthenic acids in a typical crude oil, with average molecular weights ranging from about 200 to approximately 400. Most are believed to have the chemical formula R(CH2)nCOOH, where R is a cyclopentane ring and n is typi-cally greater than 12.
Naphthenic acids are most active at their boiling points, and the most severe corrosion occurs on conden-sation. For this reason, naphthenic acid corrosion tends to be minimal below 200°C, peaks between 260°C and 350°C, and declines above 400°C.
At suitable temperatures, naph-thenic acids attack metallic surfaces to form iron naphthenate, which is oil soluble:
treatment to manage corrosion and protect key systems. The growing impact of this trend is evident in increasing cracking-to-CDU ratios industry-wide (see Figure 2).
High-acid crude defi nedHigh-acid crudes are traditionally characterised by their total acid number, or TAN, the mg of potas-sium hydroxide (KOH) required to neutralise 1 gram of crude. Crudes with a TAN above 0.5 (and cuts with a TAN above 1.5) are consid-ered potentially corrosive at temperatures ranging from 230°C to 400°C.
The inclusion of temperature in that statement suggests that TAN alone can be misleading when eval-uating the potential for corrosion and, in fact, temperature is just one of many factors that must be taken into account when evaluating the corrosive impact of crude TAN. Others include the presence of vari-ous naturally occurring components, including naphthenic acids (the generic name for all organic acids present in crude oil).
Among these compounds, naph-thenic acids are typically the most signifi cant because they are the primary cause of corrosion at high temperatures. The types of naph-thenic acids in the crude and their molecular weights infl uence corro-sivity, too.
Sulphur content is another complicating factor because it tends to increase TAN, but it does not always contribute to the corro-sive behaviour of the feed. Other
infl uences on TAN include phenols, carbon dioxide, hydrogen sulphide, mercaptans, mineral acids and various other acidic compounds.
For all of these reasons, there is considerable disagreement about the reliability of TAN as a determinant of appropriate crude discounts. In other words, a variety of factors can make two different crudes with a similar TAN differ greatly in corro-sivity during processing.
Experience and research indicate that a combination of TAN and the factors shown in Table 2 are better predictors of crude oil corrosivity than TAN alone.
Naphthenic acids characterisedNaphthenic acid corrosion prob-lems were fi rst identifi ed in the
Table 2
30
35
25
20
15
10
5
Mill
ion b
/d
0
45
40
35
30
25
%
20
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
CDU capacityCracking-to-CDU ratio (RHS)
Figure 2 Cracking-to-CDU ratios2
Rn
– (CH2)m – COOH
R = small aliphaticn = 1 to 5m > 1
Figure 3 Naphthenic acid structure
• Naphthenic acid content• Molecular weights of naphthenic species • Temperature• Velocity• Shear rates• Reactive sulphur
Key predictors of crude corrosivity
Table 2
dorfketal.indd 2 10/06/2013 16:23
EDV® Technology provides refiners with an economic and reliable means for controlling
emissions of pollutants (PM, PM10, SOx, HAPs, NOx ) from FCCUs and other stationary sources.
Write, call or fax Belco Technologies Corporation.
TM © Belco Technologies Corporation
Reducing refinery emissions has never been so easy.
9 Entin Road, Parsippany, New Jersey 07054, U.S.A. • Phone: 1-973-884-4700 • Fax: 1-973-884-4775 • www.belcotech.com
Copyright © 2010 DuPont. The DuPont Oval Logo, The miracles of science ™, and BELCO® are registered trademarks or trademarks of E.I. du Pont de Nemours and Company or its affiliates. All rights reserved.
Belco_Ad_7_2010_210x297mm.indd 1 9/29/10 4:58:27 PMbelco.indd 1 06/06/2013 10:16
Fe + 2 Z.COOH = Fe (Z.COO)2 + H
2 (1)
Sulphur content can provide limited corrosion protection. As the corrosion cell in Figure 4 shows, reactive sulphur can be present in a variety of forms that adhere to metal surfaces (represented by Fe in the fi gure). Hydrogen sulphide and reactive sulphur species that form hydrogen sulphide in the presence of heat can create a semi-protective iron sulphide scale. The protection this provides is limited because naphthenic acids react with the scale chemistry to form corrosion products, exposing fresh metal and beginning the cycle again.
Reactive sulphur has disadvan-tages, too. In some refi nery systems, increased surface interaction with reactive sulphur can actually increase potential corrosion rates.
Corrosion management optionsThere are four primary options for managing corrosion in refi nery systems. Over the long term, upgrading materials of construction with chromium-molybdenum alloys is the best approach. Obviously, this option is not availa-ble to all refi neries and it comes at considerable cost.
Blending high-TAN crudes with low-TAN crudes can help by reduc-ing overall acidity, but suitable low-TAN crudes are not always available at an acceptable cost. Depending on the crudes used, changes in blend viscosities will sometimes alter shear stress param-eters enough to help reduce corrosive action.
If high-sulphur crudes are availa-ble at an acceptable cost, addition to the blend at carefully chosen levels can form iron sulphide fi lms on metal surfaces (sulphidisation) that partially inhibit corrosion. Careful monitoring is essential. Blending these crudes can be tricky due to reactive sulphur’s potential impact on certain side cuts, and because interactions between vary-ing levels of reactive sulphur and naphthenic acids can cause dramatic changes in blend corrosiv-ity. Additionally, increasingly strict environmental limits on the
118 PTQ Q3 2013 www.eptq.com
sulphur content of fi nished fuels are beginning to foreclose this option, and hydrodesulphurisation systems are costly to install and operate.
Process control changes may also help by reducing charge rates and temperatures, although such options are usually constrained by the need to maintain throughput and meet product volume targets.
Corrosion inhibitors offer a number of advantages to the other alternatives. They can be employed selectively to protect specifi c frac-tions where corrosion is particularly severe. Also, when the formulation is selected carefully and results are monitored closely, treatment can provide highly effective corrosion protection at reasonable cost, allow-ing the refi nery to safely and profi tably exploit the economic benefi ts of high-TAN feeds.
Factors infl uencing corrosion in high-acid crudesMaximising corrosion protection requires a thorough understanding of the interactions between several key variables in the corrosion “equation” other than TAN, includ-ing the effects of temperature, velocity and shear stress, materials of construction, and sulphur content.
Temperature Naphthenic acid corrosion increases with temperature on all distillation unit alloys. It occurs primarily between 220°C and 400°C and is especially severe in high-velocity
streams. When temperatures exceed 400°C, corrosion potential declines as naphthenic acids begin to decompose, reducing their impact, and coke deposits form that help shield metal surfaces from contact with the acids.
Velocity and shear stress The relationship between corrosion potential and velocity is complex. High-velocity streams can remove protective iron sulphide fi lms, and high acidity exacerbates the result-ing corrosion. Corrosion potential is directly proportional to shear stress in moving fl uids near metal surfaces, and this shear stress often is even more important than veloc-ity alone, especially at high temperatures. At high temperatures and high velocities, even very low levels of naphthenic acid can produce very high corrosion rates. Even so, low-velocity zones cannot be overlooked because turbulence in such areas can cause simultaneous vaporisation and condensation, ideal conditions for corrosive action.
Materials of construction Metallurgy is always an important consideration in any evaluation of corrosion potential. When the potential for corrosion is moderate, alloys containing 9% Cr and 1% Mo can be quite effective, although 2% Mo (316L) is typically a better option. More recent alloy develop-ments such as 316Nb (2.5% minimum Mo) have shown promis-ing results. When chloride stress corrosion cracking (CISCC) is a factor, 2205 and 2507 alloys are good choices. For more complex situations involving high-acid crude corrosion accompanied by CISCC, 625 alloy is recommended.
Sulphur and sulphur compounds For any given TAN, low sulphur typically results in more naphthenic acid corrosion, but the reactivity of the sulphur compounds in the process stream is also important. Lighter sulphur compounds tend to be more reactive and do a better job of depositing protective FeS fi lms, but the presence of hydrogen sulphide can regenerate naphthenic acids, increasing corrosion risks.
RCOOHR-S-R
R-SHH2S(RCOO)2Fe(FeS)x
Corrosion products
Naphthenic acids
Reactive sulphur
Scale
Fe
Figure 4 Naphthenic acid corrosion cell
dorfketal.indd 3 10/06/2013 16:23
www.eptq.com PTQ Q3 2013 119
widen the acceptable “crude window” while protecting corrosion-prone systems, signifi-cantly improving high-acid crude profit potential. Planning can also facilitate timely processing of spot cargoes, adding valuable process-ing flexibility.
Dorf Ketal HTCI functionalityExtensive experience and testing demonstrate that high-temperature corrosion inhibitors are the best alternative to metallurgy upgrades and crude blending for successful HAC processing, but traditional formulations have a number of disadvantages.
Traditional HTCI formulations are based on mono-, di and triphos-phate esters that work by decomposing into phosphoric acid and reacting with metal surfaces to create a protective layer of iron phosphates. Since this reaction is activated by temperature, perfor-mance suffers at lower temperatures. Passivation is required to protect refinery systems, and this often
most severe where condensation occurs at suitably high tempera-tures. The furnace, HVGO pumparound and other areas where temperatures are below 400°C are at greatest risk.
Importance of processing andcorrosion-potential analysisTo avoid compromising operational reliability, it is essential to plan ahead before processing high-acid crudes. With proper planning and risk assessment by qualified techni-cal personnel, the refinery can often
Refinery units at high risk of HAC corrosionThe crude distillation unit is a primary target for HAC corrosion. Furnaces, transfer lines, bottom pumparounds and column inter-nals where temperatures exceed 200°C are of particular concern because high temperatures increase the corrosive potential of even small amounts of naphthenic acid. Areas carrying fluids at high veloci-ties should get special attention, as should pipe bends, thermowells and pumps where shear stress causes two-phase flow. Areas where vaporisation and condensa-tion occur are also at risk. Figure 5 shows an example of naphthenic corrosion in a heat exchanger.
High-acid crudes also cause frequent corrosion problems in vacuum distillation units. Although velocities are low, preferential vaporisation and condensation of naphthenic acids increases the TAN of the condensates in the vacuum column. Since corrosion takes place in the liquid phase, it is typically
Figure 5 Naphthenic acid corrosion in heat exchanger (5% Cr 0.5%Mo) — side cut at 343°C (650°F)
www.eptq.com Catalysis 2013 27
troubleshooting and process simulation. He has worked for BASF for four years and previously worked for INEOS and BP Refi ning & Marketing at the Lavera Refi nery.Stefano Riva is Technical Service Manager for Europe, Middle East & Africa with BASF Corporation, Refi ning Catalysts. He has over 20 years’ FCC experience, with Engelhard and BASF in technical sales and technical service, with Tamoil, and at ExxonMobil’s Trecate refi nery, Italy. Vasileios Komvokis is Technology Manager for Europe, Middle East & Africa with BASF Corporation, Refi ning Catalysts. Prior to joining BASF, he was a Research Professor at the Chemical Engineering Department of the University of South Carolina. He holds a BS and MS in chemistry and a PhD in chemical engineering from Aristotle University of Thessaloniki, Greece. Parallel to his studies, he worked for six years as a Researcher at CPERI Institute of Thessaloniki.Stephen D Challis is a Senior FCC Consultant. With over 32 years’ experience, he has been a consultant to BASF for about three years and previously worked for ExxonMobil for 29 years, with 22 years spent as a FCC process specialist providing operations and technical support to EMEA FCC units on optimisation, troubleshooting, project development, design and startups.
References1 McLean J B, The role of FCC catalyst technology in maximising diesel production, NPRA AM-09-34.2 Kraus M, Kiser N, Fu Q, Yang J, Thornton O, Finch J, Stamina – new FCC catalyst for maximum distilate yield demonstrated in Big West’s Salt Lake City refi nery, NPRA AM-10-171.3 Xu M, Liu X, Madon R J, Pathways to Y zeolite destruction: the role of sodium and vanadium, Journal of Catalysis, 207, 2002, 237-246.
Carl Keeley is Marketing Manager for Europe, Middle East & Africa with BASF Corporation, Refi ning Catalysts. He has over 12 years’ experience in the chemical and hydrocarbon industries, specialising in providing technical support for operations and licensing for FCC, and feed and product treating. He holds a MEng in chemical engineering and applied chemistry from Aston University, UK, and is a professional engineer (CEng) and a member of the Institute of Chemical Engineers in the UK.Jeremy Mayol is Technical Account Manager with BASF Corporation, Refi ning Catalysts. With over 15 years’ experience in hydrocarbon processing, he is a recognised technical specialist in FCC unit operations,
expectations. The Stamina catalyst provided an improved yield struc-ture at a similar fresh catalyst addition rate, at a similar rare earth level, with higher added active matrix level and improved zeolite surface area stability:• Net LPG and gasoline yield decreased by about 7 wt%• LCO yield increased by 5 wt% and was even higher at about +8 wt% when compared to the base competitor maximum conversion catalyst• Dry gas and coke yield were similar, despite the signifi cant increase in matrix surface area from 50 to 70 m2/g (the base competitor maximum conversion catalyst had a matrix surface area of 35 m2/g)• Using standard feed and product prices, typical for an FCC unit in Europe, Middle East and Africa, the estimated profi t improvement was almost 1 $/bbl of fresh feed delivered by the improved yield structure towards maximising LCO.
Stamina is a trademark of BASF.
MINIVAP ON-LINE
Phone +43-1-282 16 27-0 | Fax +43-1-280 73 34 | [email protected]
25 years excellencein VaPOr PressUre TesTinG
Vapor Pressure of Gasoline, Crude Oil and LPG•According to the Latest ASTM & ISO Standards•Unparalleled Measuring Accuracy•Up to 2 Sample Streams•Automatic Calibration•Fast and Easy Maintenance•
ON-LINE half page kette.indd 1 05.06.2012 14:52:27
basf.indd 5 27/02/2013 12:31dorfketal.indd 4 10/06/2013 16:23
120 PTQ Q3 2013 www.eptq.com
is a requirement for best results.Monitoring should provide
timely information about tempera-tures, velocities and vaporisation rates in operating units susceptible to corrosion from high-acid crudes. Analytical testing should be used regularly to provide alerts when system changes occur that may influence treatment rates. At a minimum, such testing should include acid characterisation and identification of sulphur compounds in the crude and cuts, along with Fe/Ni ratio analysis and TAN analysis.
Real-time corrosion monitoring can improve success by providing valuable alerts when changes in the treatment regimen may be neces-sary. Several options are available, at least two of which should be used to ensure reliability of results. Regardless of the methodologies selected, high-temperature ultra-sonic testing or a thickness survey should also be conducted:• The field signature method (FSM) measures changes in current flow caused by metal loss. FSM sensors are located on exterior surfaces of monitored systems. Trends in the data indicate the extent and loca-tion of metal loss due to corrosion or erosion• Ceion probes measure changes in resistance/conductance of a metal specimen and convert the informa-tion into corrosion rates. They do not require the presence of a continuous liquid phase. Although these probes provide instantaneous corrosion rate measurements, they cannot detect localised corrosion. • Corrosion coupons can be installed at key points and then removed periodically for evaluation• Ion Science Hydrosteel systems monitor hydrogen flux non- intrusively to provide real-time indications of active corrosion• Corrosion probes (ER, or electri-cal resistance, probes) provide corrosion information continuously once they have been installed, so they need not be removed and evaluated to provide corrosion data• Thickness surveys of product lines and return lines use ultrason-ics and other techniques to measure wall thickness, which decreases as
than 2% phosphorus. All have negligible acidity.
Their sulphur and phosphorus content provides enhanced inhibi-tion without the negative side effects of traditional HTCIs. They adhere by chemisorption rather than by acidic reactions with metal surfaces and therefore do not produce insoluble iron phosphates, which reduces fouling tendencies. These products have been success-fully tested and proven effective under severe operating conditions involving high-TAN crudes with a low sulphur content.
Importance of monitoringCorrosion is a dynamic process that must be monitored for depend-able control. Analytical testing
creates fouling problems by convert-ing iron sulphides into iron phosphates and counteracting any protective benefit that sulphidisation might otherwise provide.
Injection systems used with tradi-tional esters must be managed carefully because the formulations are acidic. Significant fouling prob-lems can arise because high levels of insoluble hydrocarbon residues often form when traditional formu-lations are exposed to high temperatures.
Dorf Ketal has developed a range of proprietary polymeric phospho-rus-sulphur inhibitors that overcome these problems. All of these formulations are thermally stable and contain less than 5% phosphorus. Several contain less
6
8
7
5
4
3
2
1
Corr
osi
on r
ate
, m
py
0
1/11
/12
to 3
0/11
/12
15/1
1/12
to 1
5/12
/12
15/1
1/12
to 2
2/12
/12
15/1
1/12
to 5
/1/1
3
15/1
1/12
to 2
5/1/
13
15/1
1/12
to 2
/2/1
3
15/1
1/12
to 9
/2/1
3
Figure 6 FSM trends for HVGO loop with Dorf Ketal HTCI
30
40
35
25
20
15
10
5
Corr
osi
on r
ate
, m
py
0SA387-5CS516-70 410SS/12cr
Dorf Ketal HTCI average corrosionBlank average corrosion
Figure 7 Comparative corrosion rates of blank coupon, 5.0% KOH/mg TAN
dorfketal.indd 5 10/06/2013 16:24
www.eptq.com PTQ Q3 2013 121
India Nagi-Hanspal is Lead Refinery Engineer: Technical Services with Dorf Ketal Chemicals, Mumbai, India, and is primarily responsible for providing refinery process improvement strategies through unit monitoring. She provides chemical solutions to Dorf Ketal site teams globally and holds a MEng degree in chemical engineering from Imperial College London. Email: [email protected]
Mahesh Subramaniyam is Director of Research & Development with Dorf Ketal Chemicals and leads the company’s chemical developments. He holds a number of patents for oil treatment chemicals and engineers products to address emerging issues in the hydrocarbon industry. He holds a PhD in chemistry from Indian Institute of Technology, Mumbai. Email: [email protected]
Parag Shah works in Global Refinery Technical Services with Dorf Ketal Chemicals, where he is instrumental in software development for desalter adequacy testing and monitoring fouling in preheat exchanger trains. His experience includes more than eight years in the world’s largest grassroots refinery and he holds a BEng in chemical engineering from Mumbai University. Email: [email protected]
corrosion also declined substan-tially, illustrating the importance of metallurgy selection for corrosion control.
ConclusionWith proper planning, high-acid crudes can be processed success-fully if appropriate strategies are implemented before adding high-acid crudes to the feed diet. Success requires a thorough understanding of corrosion processes, identifica-tion of susceptible systems and consistent monitoring to adapt treatment as system dynamics and crude diets change.
CEION is a mark of Teledyne Cormon and Hydrosteel is a mark of Ion Science.
References1 Houlton G, Crude demand to increase, feed-quality changes in store, Oil and Gas Journal, 2010.2 Zhang L, Shortage of crudes, not products, to drive Asian refining market, Oil and Gas Journal, 2011.
corrosion occurs. Once sensors are installed, the system collects data and reports trends over time.
HTCI case studyOne of the world’s largest refineries processed blends with an average TAN of 1.31 mgKOH/g and a sulphur content of 1.28 wt%. The heavy vacuum gas oil loop was monitored closely for corrosion. TAN in the stream was approxi-mately 1.79 mgKOH/g with 1.19 wt% sulphur. The key performance indicator for the trial was corrosion control at a maximum of 8 mpy. Dorf Ketal HTCI treatment results are shown in Figure 6.
Testing by an independent third-party laboratory confirmed the effectiveness of the formulation. As can be seen in Figure 7, HTCI reduced corrosion rates by more than 80% in a laboratory blend with a TAN of 5.0 mgKOH/g. Although the reduction in corro-sion on carbon steel is clearly the most dramatic, 410 stainless steel
82 PTQ Q2 2013 www.eptq.com
content of 5-8 wt% largely present in the unstripped products.
The levels of sulphur and nitrogen in coke are much higher than in the feed. In the regenerator, carbona-ceous coke and unstripped hydrocarbons are combusted to CO2, CO and H2O. Sulphur is oxidised to SO3, SO2, COS and H2S. Nitrogen behaves differently; as oxygen reacts with the coke matrix to form CO2, CO and H2O, much of the nitrogen may initially form HCN.
This is similar to the chemistry observed in coal combustion. HCN is a thermodynamically unstable species under FCC regenerator conditions and, given sufficient time and temperature, HCN will be fully converted and no trace of HCN will be found in the regenerator. Under FCC regenerator conditions, N2 is the most stable nitrogen species, followed by NO, which at thermodynamic equilibrium should reach a stable concentration of about 10 ppm. Evidently, nitro-gen chemistry is not at thermodynamic equilibrium, but is under kinetic control.
Investigation of coke combustionTemperature programmed oxidation (TPO) may be used to study coke combustion. In the TPO experiment, a sample of coked catalyst (“spent catalyst”) is heated in an inert atmosphere up to a chosen initial temperature (say, 150°C). The gas supply is then switched to the combustion gas-containing oxygen (and optionally other selected gases). The temperature is ramped (at, say, 15°C/min) to 730°C, then held isothermally. Combustion flue gases are continuously monitored by infrared and mass spectroscopy.
The first nitrogen species evolved (at the lowest temperature) is HCN (see Figure 1). HCN generation starts at very low temperatures, already around 450°C (780°F). At this low temperature, the rate of conversion of HCN to thermodynamically more stable NO or N2 is relatively slow. NO is not observed until the flue gas temperature reaches about 600°C (1100°F). As the temperature is increased further, the HCN concentra-tion declines as NO increases. The HCN is no longer stable enough to survive so it and subsequent N in coke are converted into NO or N2. Despite being ther-modynamically unstable under FCC regenerator conditions, N2O is also observed at similar tempera-tures to NO. This indicates that N2O may also be a reactive intermediate in the formation of NO.
Increasing the oxygen content of the combustion gas from 3 vol% to 4 vol% increases the rate of HCN conversion substantially. The first traces of NO in flue gas form at a lower temperature of about 550°C (1000°F).
These simple experiments show that HCN is readily formed during coke combustion at low temperatures, especially if oxygen concentrations are high at these low temperatures. Once the temperature is high enough, the HCN becomes unstable and is readily converted to NOx or N2.
The implication of this is that the points of greatest HCN formation in the FCC regenerator are likely to be
Product Nitrogen, wt%FFFuel gas NH
3/HCN 5-15
Gasoline 1-5LCO/diesel 10-20HCO/bottoms 25-35Coke 35-60
Typical nitrogen balance: wt% feed nitrogen to FCC products
Table 1
compounds present in FCC feedstocks; these are meas-ured and reported as total nitrogen and basic nitrogen. Typically, about 30-50% of the feed nitrogen is basic. These nitrogen species strongly adsorb on acid sites on the catalyst and are thereby transported with the cata-lyst into the regenerator, where they are combusted together with the coke. As a rule of thumb, about half of the feed total nitrogen ends up being combusted in the FCC regenerator. Table 1 shows a typical FCC nitrogen balance.
Coke composition varies, depending on feed prop-erties and stripper efficiency. Coke consists of carbon-rich polycyclic aromatic structures containing heteroatoms and contaminant metals as well as unstripped hydrocarbon products (for example, 10-30% of coke may be gasoline, diesel and fuel oil range products that could not be stripped from the pores of the catalyst). Typically, coke has a hydrogen
The Premium Alternativein Process Simulation
j matthey.indd 2 11/03/2013 13:02
8 PTQ Q3 2012 www.eptq.com
before etherification. A TAME unit is an opportunity to reach a higher RON pool. Oligomerisation of C5 into distillate (PolyFuel process) is an opportunity to switch gasoline into diesel, particularly in Europe, where refiners face surplus gasoline production.
Q What arrangement of trays and packing designs will give me the best efficiency of fractionation in my crude and vacuum distillation units?
A Celso Pajaro, Manager of Refinery Applications, Sulzer Chemtech USA, [email protected] is no specific arrangement of packing and trays for getting the best efficiency of fractionation. The choice in column internals is a function of the type of crude being processed and the column diameter available.
Structured packing has been used in every section of crude columns, often to substantially increase efficiency. In many cases, the efficiency within a given section can be increased by two to three theoretical stages each. However, its use in the top section can be a challenge with particularly corrosive crudes. Monel and AL6XN materials have been used successfully in these applica-tions, but care needs to be taken to properly match the correct material with the level and type of corrosivity.
In such corrosive sections, trays have an innate advantage due to their higher thickness. Some heavy,
high-sulphur crudes also have issues with chloride salt deposition, which can plug packing and trays, so specially designed fouling-resistant trays such as Sulzer’s VG AF have been used to increase column run length and maintain tray performance.
If corrosion is not an issue, entrainment-resistant trays, such as Sulzer’s Umbrella Floating Mini (UFM) valve trays will give the highest efficiency over the widest operating range. The downward vapour flow from these valves improves mixing at the tray deck level and minimises entrainment, maintaining the high-est possible efficiency from turndown to full capacity.
For the kero through diesel fractionation sections, the UFM valve trays are the best economical option. Also, structured packing can be used to improve diesel- AGO fractionation.
The wash oil section is characterised by low liquid rates coupled with high vapour rates. This is ideal for structured packings. With its high efficiency, a well-designed structured packing wash section can produce high yields of high-quality AGO under a wide variety of conditions. Sulzer MellapakPlus high-performance structured packing provides a high combination of effi-ciency and capacity.
The stripping section is typically designed with trays with a variable open area to maximise their efficiency. Structured packing has been used in this section with some success; its low pressure drop allows for better stripping. However, plugging has been experienced with some difficult crude types.
For vacuum columns, structured packing is the clear choice due to its lower pressure drop that allows mini-mum pressure at the flash zone and maximum feed vapourisation. For columns with a steam stripping section, trays are the typical solution, although struc-tured packing has also been used due to the lower pressure drop that maximises stripping. There are several vacuum columns successfully operating with structured packing in their stripping section. As is the case with crude columns, structured packing can be subject to fouling when processing difficult crudes.
Q Is there a case for hydrogen production via coke gasification over steam methane reforming?
A Girish K Chitnis, Licensing Director, Technology Sales and Licensing, ExxonMobil Research and Engineering Company, [email protected] demand in modern refineries continues to increase, with additional hydrotreating and hydro-cracking units being deployed to meet the various clean fuels mandates and product demands with heavier crudes and additional conversion capacity.
Historically in low-conversion refineries, byproduct hydrogen from catalytic reforming of naphtha for motor gasoline production was principally used to meet hydro-gen demand. Although catalytic reforming technology has advanced over the years, resulting in increased hydrogen yield due to lower operating pressures,1,2 it does not produce enough hydrogen to
want to keep the FCC at full feed rate. An option would be toadd partial conversion hydrocracking before the FCC. Thefeed rate to the hydrocracker (HC) can be fixed so that theFCC can be kept full, but there is substantial additional dieselproduced. Of course, some additional VGO is required, eitherpurchased or through additional crude runs.
The pressure of the partial-conversion HC needs to beoptimised. A high-pressure unit will be able to make ultra-low-sulphur diesel (ULSD) directly. However, capital cost ishigh. A less capital-intensive option is to use a lower-pressureHC, with some additional post-treat of the diesel integratedwith the HC. If there are other difficult diesel stocks that needto be treated in the refinery, such as FCC light cycle oil (LCO),it may also be possible to economically utilise the post-treatunit for them.
With regard to FCC LCO, this stream is an inferiorpotential diesel component with its low cetane value, evenafter hydrotreating. Partial hydrocracking of LCO (atmoderate pressure) can be considered to upgrade thematerial. While some diesel molecules will be lost to lightermaterial, the diesel retained is upgraded to a higher-qualitydiesel blending stock.
❝What are some of the most attractive isomerisationconfigurations and catalysts available to meet the
growing demand for light paraffin isomerisation? Whatcan be done to lower the equipment cost, such as therecycle hydrogen compressor? ❞Bruno Domergue, product line manager, Axens, [email protected]: Gasoline specifications are gettingtighter when it comes to octane, sulphur, olefins and
aromatics, which is why interest in isomerisation will growaccordingly. With its octane number up to 25 numbers higherthan the C5/C6 cut and the total absence of sulphur, benzeneand olefins, isomerate is the ideal gasoline blendstock. Froma processing scheme standpoint, several isomerisationoptions are available, which include the “once-through” (themost simple) and the deisohexaniser recycle (the currentfavourite), ranging up to complete iso-normal paraffinsseparation and recycle (a must in the near future for somegeographic areas). Of the various types of catalyst available,chlorinated alumina is the most profitable because it givesthe highest possible RON, being by far the most activecatalyst. It also has the lowest operating and investmentcosts, owing to low operating temperatures, low recycle flows(for recycle units) and no need for a recycle compressor.
Margaret Stine, marketing manager, process technology &equipment refining market, UOP LLC, [email protected]: The UOP Penex and Par-Isom processes are themost common processes chosen to meet the worldwide needsfor light paraffin isomerisation. Penex uses the aluminumchlorided catalysts I-8 Plus and I-82. The UOP catalystsprocess all feedstocks, including those with high C7
hydrocarbons or benzene content. Penex is frequentlycoupled with deisopentaising and/or deisohexanising toreach high isomerate product octane. The process is veryefficient and does not require a recycle compressor.
Par-Isom has been widely accepted as the best means ofconverting an idle process unit, such as a fixed-bed reformingunit, to light paraffin isomerisation. Little capital investmentis required, because the idle reforming unit contains all themajor equipment required for Par-Isom operation. Noadditional feed treatment equipment is necessary beyondtypical hydrotreating. The newest Par-Isom catalyst, PI-242, isthe highest-activity non-chlorided catalyst available on themarket today. It is contaminant-tolerant and can recoverfrom water or sulphur upsets.
Stephan Zuijdendorp, technical sales manager,isomerisation, Albemarle Catalyst Company BV, [email protected]: With the latest generation ofchlorinated alumina isomerisation catalysts, refiners save onplatinum investments and gain from long lifecycles andhigh conversion. Albemarle delivers such catalysts with theATIS range. In units of IFP-Axens and ABB Lummus Globaldesign, and others, ATIS-1L and ATIS-2L bring suchimprovements, which finally result in a significant costreduction. Depending on the conditions per unit, a technicaland economic evaluation will show if revamping fromhydrogen recycle to hydrogen once-through will be ofinterest. The lower conversion in the liquid phase might becompensated by more active catalysts and reduced yield lossby cracking.
❝With the growing demand for isobutane as anoctane-enhancing refinery feedstock, as well as a
feedstock for the petrochemical route to propylene oxide,what opportunities are available for increasing thecapacity and efficiency of existing deisobutaniserssimilar to Figure 1? ❞
QUESTIONS & ANSWERS
P TQ Q2 2006
18
Kidextractor
Quarter page
After you have usedit the first time, theKid Extractor willbecome your trustyhydraulic-tube-bundle-extractor
KIDExtractor Ltd.P.O. Box 11, ZebbugMALTATel. 00356-21-462891Fax. 00356-21-462755MOBILE: 00356-94-20596E-mail: [email protected]: www.kidextractor.com
“Of the various types of catalyst available,chlorinated alumina is the most profitablebecause it gives the highest possible RON,being by far the most active catalyst”
ENQUIRY NO 214www.ptqenquiry.com
PTQ&A copy 1 8/3/06 21:12 Page 3
Q&A copy 11.indd 3 1/6/12 12:56:32dorfketal.indd 6 10/06/2013 16:24
The Analytical X-ray Company
Fast and fl exible oil analysisby X-ray fl uorescence (XRF)
COMPLIANCE AND BEYOND
Reduce your total cost of ownership
• Pre-calibrated, calibrations last for months
• Minimal chemical waste production
• No sample preparation
For more information please contact your local sales representative or visit www panalytical.com/oilanalysis
PANalytical PO Box 13T +31 (0)546 534 444F +31 (0)546 534 5987600 AA Almelo [email protected] www.panalytical.com
Compliant with relevant standards:
ASTM D2622 ASTM D4294ASTM D4927 ASTM D6443ASTM D6481 ASTM D7751
PN8818.indd 1 04/06/2013 14:05panalytical.indd 1 06/06/2013 14:42
Chemical analysis in amine system operations
Chemical analysis opens the door to understanding, responding to and preventing
alkanolamine (amine) system oper-ational difficulties. Unfortunately, the link between the analysis report and the amine system operation is often unclear. The uncertainty results from several factors that can lead to misunderstandings, misin-terpretations, frustrations and mistrust of the lab. This article attempts to provide clear defini-tions of analytical terms, link them to their operational importance, translate terminology from differ-ent sources to a uniform set of terms, and expose analytical meth-ods that can mislead you. This article also provides questions you can ask your chemist, lab analyst or amine vendor to avoid operational pitfalls and get the information you need from the analysis reports.
An alkanolamine acid gas scrub-bing system is a simple concept: a solution of water and alkanolamine absorbs acid gases from petroleum gas or liquid and is pumped to a heated regenerator that releases the acid gases, and then the amine is cooled as it returns to the absorber. The amine solution can continu-ously circulate. The simple acid base chemistry of the process can be monitored by a few analytical titrations (see Table 1), and opera-tors need only to monitor temperatures, pressures and flow rates to balance the amine absorb-ing capacity with the acid gas removal requirement of the incom-ing petroleum gas or liquid. (The acid gas content of the sweetened product is, of course, the ultimate control measure, but is beyond the
The array of contaminants that can disrupt the operation of alkanolamine systems needs to be precisely characterised and analysed
SCOTT WAITE, ARTHUR CUMMINGS and GLEN SMITHMPR Services
scope of this article.) This simpli-fied view of amine systems and operational conditions is the basis on which the most common analyt-ical methods were developed.
If no contaminants accumulated in the amine system, this could be the extent of the analytical informa-tion required to operate. Unfortunately, contaminants do accumulate in amine systems and affect equipment longevity as well as the success of the operation of the amine system. More unfortu-nate is the fact that the contaminants can affect the results
of the fundamental analytical meth-ods, misleading the operator, and yet the operator continues to rely on these few simple tests for day-to-day operations.
Increased awareness of the effects of contaminants on operations has led to an increased understanding of the variety and identity of contaminants that exist in amine systems. Common analytical meth-ods have been adapted, modified and sometimes misapplied to contaminant analysis. Contaminant-specific analytical methods have multiplied. The amine system
www.eptq.com PTQ Q3 2013 123
2.5
4.5
4.0
3.5
3.0
2.0
1.5
1.0
0.5
0
Err
or,
wt%
fre
e a
min
e
–0.53.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5
pH at endpoint
1% formate, 1% acetate0% formate, 0% acetate
Bromphenol blue indicator range
Methyl purple indicator range
Parameter Analyte DescriptionAmine strength Free amine (FA) Amine available for acid gas absorptionAcid gas loading (AGL) Rich loading (RL) Acid gas (H
2S & CO
2) in the solution exiting the absorber
Lean loading (LL) Acid gas (H2S & CO
2) in the solution exiting the
regeneratorWater Water The remainder of the solution (if no contaminants)
Fundamental analytical parameters of amine solutions
Table 1
Figure 1 Endpoint pH affects accuracy of free amine titration
mpr.indd 1 07/06/2013 20:35
124 PTQ Q3 2013 www.eptq.com
often described in terms that lead to an incorrect view of amine inter-actions with acid gases and contaminants. Consequently, the understanding of analytical results becomes more difficult.
Alkanolamines are bases. Bases react with acids to make salts. Acid gases are absorbed and held in amine solutions because the amine forms a salt with the acid gas. The acid gas becomes an anion (nega-tively charged ion) and is associated with an amine cation (positively charged ion). The acid gas ion is no longer in a gaseous state and it cannot leave the amine solution as long as it remains as an anion.
The regenerable salts in Table 3 and the HSS in Table 4 are written to emphasis the separateness of the ions. A cation is a positively charged molecule that is physically discon-nected from its neighbouring anion, which is an independent molecule with a negative charge. Anions and cations must be in equal numbers and uniformly distributed through-out the solution, but they are continually changing partners. This view of ions is critical to the under-standing of amine acid gas absorption and regeneration.
For example, H2S absorbed in an amine solution is not bound to the amine. Rather, the amine has taken a hydrogen ion (H+) from the H2S, creating an AmH+ cation and an HS- anion, which cannot escape from the solution. The amine is bound to the H+, and does not readily release it. The only way for the HS- to escape the solution is to take an H+ from an AmH+, thereby recreating H2S, which has low solu-bility and high volatility and will exit the solution, unless another amine molecule reacts with it and removes one of its H+ to form another salt.
Acid gases are readily released from the thin films of liquid amine solution in an amine system regen-erator, not because temperature “breaks the salts” but because at higher temperatures and lower pres-sures AmH+ more readily releases its H+, and the anions of the acid gases readily take the H+, creating gases that are less soluble and more volatile at the elevated
operator can now be confronted with a maze of analytical parame-ters, a blur of analytical results, and a host of analyte names and acro-nyms that can be ambiguous, confusing and even misleading.
Better analytical methods neededA prime example of misleading results is illustrated in Figure 1 and Table 2. When weak acids (such as formic acid, acetic acid, and so on) have accumulated in the amine solution, forming heat stable salts (HSS), the titration to determine free amine can also respond to the weak acid anions. The choice of pH for the endpoint of the titration determines whether the free amine titration is accurate or over- estimates the amine strength. A pH or colour indicator that provides accurate amine strength in a clean amine solution can grossly over- estimate the amine strength of a solution that contains weak acid anions (HSS or LL). Note, for exam-ple, in Figure 1, that Methyl purple and Bromothymol blue are both acceptable indicators for the titra-tion of clean amine solutions, but fail miserably if the solution contains significant weak acid HSS. The same is true for a pH “dead-stop” titration. The most common amine strength titration methods were developed for amine solutions
with no contaminants. Thus, contaminants may cause errors.
Understanding analysesWith knowledge of the preceding information, the engineer responsi-ble for the amine system can now ask the analytical chemist or opera-tor, “What endpoint indicator do you use for the amine strength titration?” If the chemist responds “Bromophenol blue” or “pH 4.5”, the engineer knows that amine strength results are probably higher than their actual level, unless there are no weak acids in the amine solution.
The better titration methods require the tracking of pH or conductivity during the titration and determination of the endpoints by inflexion points in the first or second derivative, respectively. Both pH and conductivity can provide accurate free amine results, but conductivity is preferred because it also provides clear endpoints for the weak acids.1,2
Before we seek to understand analytical methods, let us clarify our understanding of the amine solvent itself.
Amine system is supposed to make saltsThe chemistry of alkanolamine solutions is quite simple, but is
Acetate, Formate, Sodium, Actual free Found free amine, wt% at endpoint pH wt% wt% wt% amine, wt% 4.5 3.5 3.2 0 0 0 45.0 45.0 45.1 45.21 1 0 40.3 42.0 44.0 44.51 0 0 43.0 44.3 45.0 45.11 1 0.5 42.9 44.6 46.6 47.11 0 0.5 45.0 46.9 47.6 47.7
Error of fixed pH endpoint titration 45 wt% MDEA
Table 2
Acid+vase→ cation+anion
Acid Base Salt Salt nameH
2S Freeamine(“Am”) AmH++HS- Aminiumbisulphide
CO2+H
2O→H
2CO
3 Am AmH++HCO
3- Aminiumbicarbonate
HCO3
- Am 2AmH++CO3
= AminiumcarbonateCO
2 Am+Am AmH++AmCO
2- Aminiumaminecarbamate
Regenerable salts from acids gases in amine solutions
Table 3
mpr.indd 2 07/06/2013 20:36
42nd Turbomachinery29th Pump SYMPOSIA
GEORGE R. BROWN CONVENTION CENTER
9.30 – 10.3.2013
/TurbolabatTAMU @PumpTurbo
turbolab.tamu.edu
turbo lab.indd 1 06/06/2013 10:26
temperature and decreased pressure of the regenerator.
HSS, such as those in Table 4, do not leave the amine solution in the amine system regenerator. The acids of these salts are generally stronger acids than the acid gases (thus they do not as readily accept the H+
offered by the AmH+), and are more soluble and less volatile than the
126 PTQ Q3 2013 www.eptq.com
acid gases (thus they do not leave the solution). The HSS anions can be classed as weak acid anions (WAA) and strong acid anions (SAA). The functional distinction between WAA and SAA is seen in titration with acid in water:
WAA- + H+ → HWAA; SAA- + H+ → SAA- + H+
In words: WAA consume H+
(making an acid molecule); SAA do not.
The last four entries in Table 4 exemplify non-amine HSS: salts whose cation is a strong cation such as sodium or potassium. These are often called inorganic HSS (IHSS) and are formed in amine solutions from the addition of caustics (sodium or potassium hydroxide or carbonate). The last two entries illustrate that an excessive concen-tration of strong cations makes HSS of the anions of acid gases, causing the lean loading of the regenerated amine to rise. Strong cations must never be allowed to rise to the level that any acid gas becomes heat stable. This starts to occur when the molar ratio of strong cations to HSS anions reaches approximately 0.66.
Acid gas regeneration is aided by a low concentration of amine HSS, because they increase the popula-tion of AmH+ by decreasing the pH. The anions of acid gases then have a higher probability of obtain-ing H+ to convert back to the
Table 4
Acid + base → cation + anion
Acid Base Salt Salt nameFormic Am AmH+ + formate- Aminium formateAcetic Am AmH+ + acetate- Aminium acetate Propionic Am AmH+ + propionate- Aminium propionateThiocyanic Am AmH+ + SCN- Aminium thiocyanateThiosulphuric Am + Am AmH+ + S
2O
3= + AmH+ Aminium thiosulphate
Hydrochloric (HCl) Am AmH+ + Cl- Aminium chlorideSulphuric Am + Am AmH+ + SO
4= + AmH+ Aminium sulphate
Formic NaOH Na+ + formate- Sodium formateAcetate NaOH Na+ + acetate- Sodium acetate
Acid gases are heat stable with caustics; for example:H
2S NaOH Na+ + HS- + H
2O Sodium bisulphide
HCO3
- KOH K+ + CO3
= + H2O Potassium carbonate
HSS from acids in amine solutions
To keep pace with the demanding quality requirements of modern fuels, advanced, precise and easy to use analytical technology is required. With a complete range of XRF and ICP spectrometers, SPECTRO’s unique solutions for at-line and laboratory elemental analysis are capable of meeting the most demanding product specification testing requirements.
Elemental Analysisof Fuels and Oils
demanding product specification testing requirements.testing requirements.
Discover more exciting details atwww.spectro.com/fuels
Determination of Sulfur and other elements at-line and in the laboratory
mpr.indd 3 07/06/2013 20:36
www.eptq.com PTQ Q3 2013 127
gaseous state and escape the solu-tion of amine.
Acid gas regeneration can be hampered by strong cations, because they have no H+ to give. Anions of acid gases are attracted to the strong cations because of their positive charge, and then must remain in the amine solution because they cannot obtain an H+
from the strong cation. HSS reduce amine strength,
increase corrosivity, increase viscos-ity, increase density and displace water from the amine solution.3,4 All these have deleterious effects on amine system operations. It is thus very important to know what and how much salt is accumulating in your amine system.
What analyses are needed?The first-tier parameters affecting amine system performance are free amine and water content. The free amine represents the carrying capacity of the solution, and the water content affects the acid gas absorption and desorption rates. The amine solvent supplier, amine system designer or the amine tech-nical support person typically provides operators with recom-mended ranges of free amine and water for the particular application. If there are no contaminants in the amine system, the Table 1 analyses would be sufficient — water could even be determined by difference. However, contaminants are common in amines systems and a more detailed analysis is required. Table 5 shows several parameters requiring analysis for the proper operation of clean amine systems.
The trends in the amine system should be monitored on a regular basis. When BA rises or TA plus water departs significantly from near 100%, the amine system could be becoming contaminated and more information is needed. The second-tier analyses — for dissolved substances — are listed in Table 6.
The numerous analytes listed in Table 6 can be daunting. The several analytes may each show a different result, or overlap, or even disagree with one another, requir-ing informed interpretation of the respective results.
Type Analytes DescriptionandroleAcidgas(AG) Richloading(RL) Acidgas(H
2S&CO
2)inthesolutionexitingtheabsorber
Leanloading(LL) Acidgas(H2S&CO
2)inthesolutionexitingthe
regenerator H
2S,CO
2, Theacidgasesmeasuredseparately;inaminesolution
acidgasloading(AGL) areionic:HS-,HCO3
-,CO3
2-andcarbamateions
Acids-organic HSS;totalHSS; Acidsthataccumulateintheaminesolutionbyabsorptionandinorganic, BA;HSAS;IHSS; andchemicalreactionswithintheaminesolution.Theyalsocalledheat individualanions*; reduceaminecapacity,contributetocorrosion,increasestablesalts(HSS) ATB-RFB;conductivity densityandviscosity,andconductivity.Twogeneraltypes: weakacidanion(WAA)andstrongacidanions(SAA)
Acids-amino Bicine;totalamino Aminoacidderivativesofaminesbyreactionswithinthe(AA) acids;BA aminesolution;maybeincludedamongHSS;aminoacids titrateasBAandasTA.Aminoacidsstronglyenhance corrosion
Cations, Strongcations(SC), Alkalicationsthataccumulateintheaminesolutionsinorganic,also eg,K+,Na+; fromcoolingwaterleaksordeliberatecausticaddition;calledstrong corrosionmetalions corrosionmetalsresultfromaggressiveacidgases,cations(SC) (Fe,Cr,Ni,etc) erosion,enhancedbyHSSandaminoacids inorganicHSS(IHSS)
Basic Totalbase(TB);total Aminesandaminedimersthatresultfromchemicaldegradation aminebase(TAB); degradationofthesolventamine;havesomeacidproducts(BDP) actualtotalbase(ATB); gasabsorbingcapacity.Aminedimersenhancecorrosion totalnitrogen(TN); TB-ATB;dimers (THEED,HEED,HEP, etc),ureas
Neutral Amides,formylamines, Non-ionresultsofchemicaldegradationofthesolventdegradation oxazolidones; amine;amidesareinactivereversibleequilibriumwithproducts(NDP) hydrolysables theircorrespondingorganicacidHSSanion totalnitrogen(TN); TN-TB
*HSSanionsmostcommonlydeterminedindividuallyincludeformate,acetate,propionate,glycolate,oxalate,chloride,
thiocyanate,thiosulphate,sulphateandsulphite.
Dissolved substances in amine solutions (other than solvent amine)
Table 6
Parameter Analyte Description and roleAminestrength Freeamine(FA); Amineavailableforacidgasabsorption regenerablefreebase(RFB)Reactedamine Boundamine(BA) Aminethathasreactedwithanacid andnowcarriesahydrogenion:AmH+.Totalamine Totalamine(TA)=FA+BA; actualtotalbase(ATB) Alltheamineinthesolution
Activators Proprietaryadditivesthatenhancethe aminesolution’sabsorptionrateor selectivity.GenerallyareincludedinFA, BAandTAresultsWater Water Waterisessentialforacidgasabsorption
Physicalproperties Density, Physicalpropertiesofthesolutionprovide viscosity, monitorsofcompositionalchangesinthe conductivity, aminesolutionthataffectabsorption, surfacetension, regeneration,heatdemand,corrosivityand appearance equipmentlongevity.
Solvent parameters of amine solutions
Table 5
mpr.indd 4 07/06/2013 20:36
128 PTQ Q3 2013 www.eptq.com
foaming and inhibited processing rates. Given the importance of the short- and long-term characteristic effects of these contaminants, it is very important to have regular analytical reports. But how do you make sense of them?
Reconciling various analytical methodsWhen you compare reports from two different labs, there may appear to be inexplicable differ-ences in apparently the same analyte. Some of the confusion results from differences in termi-nology and some from differences in lab procedural definition of analytes of the same name. For example, the question “What is the HSS content of this sample” could have the following correct answers:
HSS anions 14 609 ppm(m)Total HSS 3.13% as DEAHSAS 1.49% as DEAIHSS 1.63% as DEATotal HSS/total amine 11.14% of DEA 0.1114 mol/mol
6 and 7 have their effects on amine solvent performance and/or the amine system hardware. Contaminants can change the phys-ical properties (density, viscosity, surface tension, thermal conductiv-ity, electrical conductivity and foaming tendency) and the water content, which can influence film absorption and desorption rates. Longer-term effects include corro-sion, plugging, flow restriction,
Many of the dissolved substances cause corrosion or enhance corro-sion by attacking the protective iron sulphide layer. Iron carbonate can form protective layers, but most often creates suspended solids. Corrosion products create solids in the amine solution, and along with other insoluble or semi-soluble substances create the third tier of analytical problems listed in Table 7.
All of the contaminants in Tables
Type Analytes Description and roleSolids Total suspended solids (TSS) From corrosion products, chemical and physical; TSS contribute to further corrosion, flow restrictions, and plugging. TSS stabilise foamHydrocarbons Total petroleum hydrocarbons TPH carryover from contact with petroleum gas (TPH) or liquid. TPH may contribute to or inhibit foamSurfactants Foaming tendency, break time, Polar/non-polar molecules that cause foaming. foaming potential; surface typically in very low concentrations but have large tension physical effects inhibiting acid gas absorption and stripping Surfactants cause system upsets and solvent lossesAntifoam Si (confirm presence of silicone Antifoam is deliberately added to reduce foaming; antifoams only); no direct note: excess antifoam can cause foam analyses
Insoluble, partially soluble or dispersed substances in amine solutions
Table 7
Table 5
Today’s Application:LNG-PROCESSING, HANDLING AND STORAGE
HERMETIC design features
■ 100 % leakage free ■ Low life-cycle-costs ■ Low noise level ■ High reliability ■ Customized design – adapted to your
process requirements
Customer‘s technical specifi cation■ Capacity: 120 m³/h ■ Head: 1400 m ■ Pressure rating: PN 100 ■ Motor power: 370 kW
H E R M E T I C A L L Y S E A L E D C E N T R I F U G A L P U M P S
HERMETIC-Pumpen [email protected] · www.hermetic-pumpen.comHERMETIC-Pump Type CAMTV 52/6+6
mpr.indd 5 10/06/2013 14:56
www.eptq.com PTQ Q3 2013 129
HSS. Such methods estimate HSAS, not total HSS.
Strong cations can enter amine systems unintentionally through heat exchanger leaks, from the treated product or intentionally by the addition of caustic to neutralise HSS. Proper analysis is important to make sure they do not contribute to high lean loading and stripping difficulties or contribute to viscosity and corrosion problems. There are several methods of analysis for strong cations, and the following questions should be asked:• How are strong cations (sodium, potassium and calcium, for instance) determined? • If by IC or ICP: ■ Was the signal within the cali-bration range? ■ Are there any false highs due to interferences (such as amine with K+ in IC)• If by titration: ■ Was a sufficient strong-base anion exchange resin used to absorb all the LL and HSS anions (at least two times the equivalent)? ■ Has the titration been checked against ICP or IC for this amine system? (It is a good check on endpoint determination technique. Rarely, but it has happened, the titration method will grossly exceed the actual strong cations concentra-tion for unknown reasons) ■ Only strong alkali and alkaline earth metals are detected by titration.
It is important to measure metals in order to have an early detection system for possible operational difficulties due to corrosion prod-ucts or loss of metal integrity in the unit. Metals are measured by ICP and AA, and there are several rele-vant questions you should ask and items you should consider:• Was the sample filtered or digested? (Dissolved or total metals?) High dissolved metals in lean amine samples indicate the presence of HSS, amino acids and/or polymeric amines (BDP)• Sample preparation of ICP samples, especially for solutions that contain solids, will affect the results. If the sample is just filtered before the digestion of the liquid sample, the measurement is
• Is the endpoint determined by “dead-stop” at a specified pH? If yes, worry about accuracy• Spiking with acetic acid before titration can help get a more accu-rate endpoint for FA.
An important parameter for oper-ational control is the total concentration of the amine in the unit. The questions below should be asked so you can understand how the total level of amine was determined: • How is the total amine test done (GC, IC, HPLC, titration or other)?• If GC, IC or HPLC, what amines are quantified and are there note-worthy not quantified/unidentified peaks?• If titration, are other basic species such as amino acids or piperazine are subtracted from the reported total amine number? If the number is reported as titrated, the total amine number is actually a total base number.
One of the most important analy-ses in amine systems is the measurement of HSS. A number of methods can be used for this analy-sis, and it is recommended that more than one method be used and the results compared to make sure accurate results are obtained. These questions are useful in obtaining a good understanding of the results:• How is the HSS measurement made (IC or titration)?• If by IC, the individual analysis of species should be available. See if the common anions are analysed for formate, acetate, propionate, glycolate, oxalate, chloride, thiocyanate, thiosulphate, sulphate and nitrate• If by titration: ■ Is a direct titration of the amine sample with base? If so, the result is BA. ■ How is lean loading (AGL) removed? If via boiling of a direct sample, worry about false high results ■ How are strong cations removed before analysis or how is their effect accounted for in the final HSS result? Caution: some acidification and back titration methods detect WAA HSS (not SAA HSS) and detect only the strong cations, which exceed the equivalent of the total
From 1.5 to 11%! No wonder we are confused. Notice the impor-tance of the prepositions “as” and “of”, which are sometimes not included in conversation or reports. We prefer the first two in the above list, because they are direct expressions of the two parts of the total HSS: the anions and the amine equivalent cations. The others have good utility, however, and need to be understood.
As can be seen in Table 8, there are a large number of parameters that are measured by a variety of techniques, and a thorough under-standing of the terminology is required to correctly interpret amine analysis reports and to take action based on the analysis.
Questions to ensure measurements are clear and understoodProper sampling is essential to achieving good analysis results, and understanding how the sample was taken is essential in helping you interpret the results. The following questions should be suit-ably answered before the analyst or operator takes a sample: • Was the dead leg flushed before taking the sampling?• Were the operating conditions of the unit recorded when the samples were taken? Is there a plant upset or unique operating conditions?• Was the reason for sample collec-tion recorded (routine or non-routine troubleshooting)?• Where was the sample taken? Upstream or downstream of the filters and carbon beds?
You should make sure that the measurements for free amine (FA) or amine strength are correct so you know how much amine is available for acid gas adsorption. The following questions are helpful to interpret the results:• Is it a titration with acid? If yes, good.• Is the endpoint of the titration determined by inflection points in a conductivity or pH curve? If yes, good• Is the endpoint determined by colour indicator? If yes, worry about accuracy
mpr.indd 6 07/06/2013 20:36
130 PTQ Q3 2013 www.eptq.com
the complicated array of contami-nants that pollute amine solutions and cause problems for amine system function and longevity. The even larger array of analytical methods employed to determine contaminant concentrations in amine solutions can lead to confu-sion if the analytical reports are not properly understood and inter-preted. While there is some duplication among the tables, each provides a different perspective and thus can possibly provide aid to different people in the variety of situations amine systems present to focus on the analytical methods and data that are most instructive for the need at hand.
The fundamental focus for amine analysis must be on the following, which can be tracked with relatively few, well-selected analytical meth-ods: free amine (FA), total amine (TA), bound amine (BA), H2S and CO2, total HSS, including WAA and SAA estimates from conductivity titrations, strong cations (SC), chlo-ride, thiocyanate, amino acids,
If your lab does not have the capability of running all the tests, do not hesitate to use an outside lab to complement your in-house testing. It is also a good idea to have a complete analysis done by an outside lab once a year to check your in-house analysis results. If contaminants are growing or changing, you may be advised to have outside analysis done on a more frequent basis. Remember that HSS, amino acids and BDP contribute to corrosion and can cause severe damage to the amine treating unit.
What information is revealed byanalysis of contaminants?By closely monitoring trends in the analytical data, you can detect possible operational problems before they become a big problem for the unit. Table 9 provides some examples.
SummaryWe have discussed both the simplicity of the amine system and
of only the dissolved metals• If the samples are not filtereduntil after acid digestion, the metals in the solids (total metals) are included in the metals report• Generally, it is more instructive to determine the metals in the filtered solids separately from theliquid, so solids and dissolved solidsprofilescanbedetermined.
Performing a mass balance or “residue calculation” can be instruc-tive. It requires an accurate water analysis by Karl Fisher titration, an accurate total amine measurement, an accurate knowledge of the amine molecular weight (which may be difficult if it is a blended or formu-lated amine), and an accurate analysis of strong cations, HSS, amino acids, AGL, BDP, NDP and amides. A mass balance of less than 100% can indicate that there are contaminants present that have not been analysed. A mass balance of greater or less than 100% can also indicate errors in the analysis meth-ods, especially for total amine and water.
ABB Consulting. Safe, reliable, efficient operations.
Operators aiming to achieve operational excellence in the face of increasing competitive, legislative and economic pressures, often require expert assistance. ABB Consulting assists customers in the global process industries. We offer expertise in; inspection, specialist functional engineering, integrity management, process safety, projects services, technical software and technical training. We benefit customers by reducing risk, optimising cost and improving manufacturing efficiency. www.abb.com/consulting
ABB ConsultingTel: +44 (0)1925 741111E-mail: [email protected]
PTQ Q3 2013 (125x178) Jun 2013.indd 1 07/06/2013 11:56:41
mpr.indd 7 07/06/2013 20:37
www.eptq.com PTQ Q3 2013 131
Nevada, Reno, and holds a BS in chemistry from the University of Arizona and a PhD in physical/analytical chemistry from the University of Utah.
Arthur Cummings retired from MPR Services in July 2012. During his 21 years as Vice President of Technology, MPR helped hundreds of refineries, gas plants, ammonia plants, syngas plants, chemical plants and steel mills improve the operation of their amine units. He holds a PhD in analytical-physical chemistry from Brigham Young University.
Glen Smith is Technical Manager with MPR Services. He managed MPR’s laboratory from 2005 through 20l2, and helped develop and improve methods of analysis for amines and glycols. He holds a PhD in inorganic chemistry from Purdue University; his post-doctoral and professional activities include work in electronic materials, polymers and petrochemicals support.
part one: amine strength, Brimstone Sulphur Symposia, Vail, Colorado, 2004.2 Tunnell D, Asquith J, Bela F, Buzuik F, Eguren R, Cosma G, Hatcher N, Keller A, Kennedy B, Schendel R, Smith C, Stern L, Welch B, Zacher M, Choices for determining amine strength: effects on operations and optimization, Brimstone Sulphur Symposia, Vail, Colorado, 2005.3 Rooney P, Bacon T, Dupart M, Effect of heat stable salts on MDEA solution corrosivity: part 1, Hydrocarbon Processing, Mar 1996, 95-103.4 Rooney P, Bacon T, Dupart M, Effect of heat stable salts on MDEA solution corrosivity: part 2, Hydrocarbon Processing, Apr 1997, 65-71.
Scott Waite is Vice President of Technology with MPR Services, Inc., Dickinson, Texas. He has managed the R&D process to develop new and improved methods for amine and glycol reclamation, and treatment of produced and flow backwater. He is currently an Adjunct Professor of Chemistry at University of
amides and water. With the excep-tion of CO2, these can all be done with an autotitrator.
A mass balance or residue calcu-lation using the analysis in the previous paragraph yields an esti-mate of basic degradation products (BDP) and neutral degradation products (NDP).
Instrumental methods can be added to measure total nitrogen (TN), amine species by GC or IC, and HSS speciation by IC. Then, by “amine balance” and “nitrogen balance”, BDP and NDP estimates might be segregated. These can be supported by HPLC and GC identi-fication of specific BDP and NDP species.
Other important measurements that can provide information on the health of the amine include total suspended solids (TSS), total petro-leum hydrocarbons (TPH), foaming tendency, foam break time, density, pH, conductivity, viscosity and surface tension.
Tracking and graphing various data over long periods of time makes it easier to spot trends, to gain a sense for normal analytical/sampling variability, to recognise possible errors, to key in on unex-pected results, and to track cause and effect on unit operations.
A plant needs to have locally only an automatic titrator, conduc-tivity meter and probe, pH meter and probe, sulphide ion-selective electrode and reference electrode, CO2 analyser, and Karl Fischer titrator in order to provide the most important analyses (free amine, total amine, bound amine, H2S, CO2, total HSS, strong cations, chlo-ride, thiocyanate and water). The other analytes can be done monthly, quarterly or as needed by your company’s central analytical lab, amine vendor or reclaimer.
A further table (Table 8) providing comprehensive reconciliation of analytical methods and their reports is included in the web version of this article. Go to www.eptq.com
References1 Tunnell D, Asquith J, Bela F, Buzuik F, Eguren R, Cosma G, Hatcher N, Keller A, Kennedy B, Schendel R, Smith C, Stern L, Welch B, Zacher M, Evaluation of amine analytical methods
Symptom Possible InterpretationsFA decreasing - Amine losses (carryover, entrainment, valve error, etc) - Water purge out of balance (total solution volume increasing) - HSS ingress - NDP forming: amides, if primary or secondary amine; oxazolidones if in high-pressure CO
2 service
Total HSS increasing - HSS ingress - Water content decreasing - Corrosivity increasing - Need HSS removedFA increasing while BA increasing - FA analysis error: FA report includes weak acid HSSLL rise with no change in operating - Greater AG load in absorbersconditions - Strong cations ingress into solventTotal HSS > BA - Caustic ingress or addition - Cooling water leak into amine solutionBA > total HSS - Amino acids present (not included in total HSS) - BA includes high LLThiosulphate HSS - Oxygen ingress - SO
2 breakthrough (Claus tail gas)
Bicine (and other amino acids) - Oxygen ingress (gas plants) - O
2, SO
2 and/or S
2O
3= attack (refinery)
Strong cations - Caustic addition - “Amine slops” added to amine and included some caustic - Cooling water leak into amine - “Hidden” HSS - Foaming tendency increasing (soaps)Formate HSS - Caused cyanide or carbon monoxide ingress - If primary or secondary amine, amides increase as formate increases, so total amine and free amine reduced => apparent amine loss, but it can be recovered (do not throw it away)Acetate HSS - Acetonitrile ingress - Oxygen ingressThiocyanate (SCN-) HSS - Cyanide reacting with H
2S
Chloride HSS - Cooling water leak - HCl ingressSolids (TSS) rising - Corrosion - Filters inadequate or spentTPH - Hydrocarbon carryover into amine: inadequate knock-out drum“Residue” increasing, but BA not - Caustic ingress or addition - Cooling water leak into amine solution - Amine chemical degradation
Analytical symptoms of some amine system problems
Table 9
mpr.indd 8 11/06/2013 12:22
Join colleagues and customers at AFPM’s industry-leading conference.
Sheraton Dallas HotelDallas, TexasOctober 7-9, 2013
AFPM’s Q&A and Technology Forum is the industry’s most comprehensive meeting on refinery operations and process technology. This event assembles representatives from throughout the industry – including refiners, petrochemical producers, catalyst and chemicals suppliers, plant automation specialists and engineering firms – and gives them the opportunity to address
real problems refiners face today. Attendees will gain from the chance to discuss their challenges and find solutions from the expert Q&A panelists and other attendees.
Today’s competitive refining environment requires attention to plant safety, superior technology, innovation and excellence in operations. Benefit from the shared knowledge of experts throughout the industry – the Q&A and Technology Forum will keep you moving in the right direction as you optimize your plant’s operation.
Register today at www.afpm.org.
PROVIDING SOLUTIONS The 2013 AFPM Q&A AND TechNOLOGy FORUM
PROVIDING SOLUTIONS
afpm.org
afpm.indd 1 06/06/2013 10:14
JV signed for Middle East FCC catalysts plant W R Grace and local partner Al Dahra Agricultural Company have completed a joint venture agree-ment to build the first plant in the Middle East producing FCC cata-lysts and additives. A logistics hub to serve the Abu Dhabi plant is scheduled for completion in Q1 2014. Construction of the plant itself, in the Khalifa Industrial Zone (Kizad), outside Abu Dhabi city, is expected to begin late in 2013, and first production is slated for Q4 2015.
At a briefing in Abu Dhabi before ceremonies to mark the agreement, Grace Catalysts Technologies’ presi-dent, Shawn Abrams, told PTQ that the new plant is a response to growth in the market for FCC cata-lysts and additives, in which the supply side has tightened. While Grace is not disclosing the produc-tion capacity of the $200 million project, Abrams said that it would be a “Middle East-oriented plant” capable of meeting combined demand throughout the Gulf Cooperation Council (GCC) states, which include the major refining nations Saudi Arabia, the UAE and Kuwait.
Grace will own 70% of the plant and 30% will be owned by Al Dahra, which Abrams described as a good local partner, particularly in view of its ability to deal with the logistical side of the project.
Off-take agreements for the new catalyst plant are, for the most part, still to be concluded. Ortic of Oman has signed up to take its products, but local major Abu Dhabi Refining Company (Takreer) has not. Takreer spoke at an event to mark the formation of the JV, under the aegis of UAE Energy Minister Suhail Mohammed Al Mazrouei, and is currently undergoing major expan-sion at its Ruwais and Abu Dhabi sites. Its FCC catalyst supply is currently subject to an agreement signed in 2012 with Albemarle
Corporation. The Grace-Al Dahra project’s site at a world-scale free trade zone and port would position it well for exports to Asia.
The Kizad plant will be Grace’s second foray into refining catalysts production in the region. Through its Advanced Refining Technologies JV with Chevron, the company has a shareholding in Kuwait Catalyst Company, which produces hydro-processing and hydrocracking catalysts under licence for supply within the GCC region.
Shell, Aramco agree on resid gasification schemeShell Global Solutions has signed a licensing agreement with Saudi Aramco for what would be the largest residue gasification unit built so far, generating 2400 MWe.
The Jazan Integrated Gasification Combined Cycle Project (IGCC) agreement includes the licensing of
Shell gasification and acid gas removal technologies and the supply of engineering services. Shell’s CRI/Criterion catalysts and a sulphur recovery unit (SRU) will also treat the off-gases from the acid gas removal unit.
Jazan is to be developed as a new economic city in the Kingdom of Saudi Arabia. As part of the devel-opment, a new refinery and IGCC is to be built 70 km north of the city. The IGCC would enable the gasification of low-value residue feedstocks to produce syngas for power generation. The power generated would provide electricity
Industry News
for the Jazan refinery and other consumers within the kingdom.
When it starts operating, Saudi Aramco’s Jazan refinery will process 400 000 b/d of Arabian Heavy and Arabian Medium crude to produce gasoline, ultra-low- sulphur diesel, benzene and parax-ylene. A marine terminal on the Red Sea coast will accommodate very large crude carriers for the supply of crude oil to the refinery.
BASF expands catalyst testingBASF is to open a testing and research laboratory for FCC cata-lysts at the hte Aktiengesellschaft (hte) site in Heidelberg, Germany. hte is a wholly owned subsidiary of BASF SE and provides high throughput technology and services for enhancing research and development.
The new laboratory will begin operating in the first half of 2014, and will be used for analysis and characterisation of FCC equilibrium catalysts, for process methodology testing and related research for customers in Europe, the Middle East and Africa.
BASF says that the opening of the Heidelberg laboratory will increase the service offered by its FCC cata-lysts laboratory in Iselin, New Jersey, and will enable the company to better serve customers across the world.
Foster Wheeler to engineer Chilean SNG plantPecket Energy has awarded Foster Wheeler a contract for feasibility, conceptual and basic engineering studies to develop a cost estimate for a substitute natural gas (SNG) plant planned to be built near Punta Arenas, Chile. The main objective of the project is to produce syngas that will be used as a feedstock for the production of SNG, which will be distributed to the existing grid in the region of Magallanes, Chile, for domestic or industrial use.
www.eptq.com PTQ Q3 2013 133
The IGCC would enable the gasification of low-value residue feedstocks to produce syngas for power generation
ind news copy 14.indd 1 10/06/2013 14:58
The new plant’s units would include air separation, partial oxidation for syngas production, syngas treatment, acid gas removal, methanation and solid sulphur production. Foster Wheeler’s scope of work is expected to be completed by mid-2013. Pecket Energy is a Chilean company specialising in coal-based energy.
O2 supply for Reliance gasifi ers
The Linde Group has been awarded a contract by Reliance Industries (RIL) to build plants to generate and purify gases in Jamnagar, India. Linde’s Engineering Division will supply four large air separation units (ASUs) for the production of oxygen for its proposed petroleum coke and coal gasifi cation facilities. To treat the synthesis gas gener-ated during this gasifi cation process, Linde will also deliver two Rectisol acid gas removal units and will be supplying the licence, process design, detail engi-neering and procurement services
for this project. Two additional ASUs will supply oxygen to RIL’s ethylene glycol facilities in Jamnagar.
Petrobras trialling wireless technologyYokogawa’s South American subsidiary and Brazilian refi ner Petrobras are evaluating wireless technology based on ISA100.11a, a wireless communication standard for the industrial automation indus-try. Yokogawa will provide Petrobras with technical support on the use of fi eld wireless communi-cations and demonstrate its fi eld wireless products at a test plant.
Petrobras will use the project to gather more data about the use of wireless technology in its processes from the standpoints of robustness and performance. By being among the fi rst to introduce such technol-ogy, the company seeks to reduce complexity at its plants and acceler-ate project implementation at certain production sites to improve productivity.
For Yokogawa, the project is an opportunity to develop its Wireless Anywhere concept and broaden the use of ISA100.11a-compliant products and services in plant-wide applications.
A wireless network can be used to connect a control system with fi eld devices that are distributed throughout a plant. Such networks have advantages that include a reduction in cabling and other installation costs, and allowing the installation of equipment in less accessible locations.
The ISA100.11a standard is designed to achieve high reliability and fl exibility in wireless systems, along with the ability to expand networks and to be compatible with wired communication stand-ards such as Foundation Fieldbus, HART and Profi bus.
HPCL plans for more refi ning capacityHindustan Petroleum (HPCL) wants to reinstate a $10 billion refi nery and petrochemical plant
134 PTQ Q3 2013 www.eptq.com
Part of
Visit www.gasasiasummit.com
Creating Integrated and Sustainable Gas Markets to Fuel Asia
Align your company with the only regionally focused gas and LNG
conference at the SIEW, for bespoke sponsorship packages
please contact Shunker Goel on +44 (0) 203 615 3224 or
email [email protected]
Distinguished Speakers include:
Rajeev Mathur Executive
Director – Marketing GAIL (India) Ltd
Duncan van BergenConference Chairman
General Manager Global Gas & LNG
Market DevelopmentShell Upstream International
Silvia MarucciLiquid Bulk Segment,
Office of Market Research & Analysis
Panama Canal Authority
Neil McGregor Chairman
Singapore LNG
Seah Moon Ming Chief Executive Officer Pavilion Energy Pte Ltd
Anthony Barker General Manager – Asia
BG Group
Steering Committee Members include:
Anil JoshiSenior Vice
President – LNGGujarat State
Petroleum Limited
Yulong LiExecutive Director
National Energy R&D Center for LNG Technology
Director of Licensing & Technology Innovation DivisionChina HuanQiu Contracting &
Engineering Corporation (HQC)
John SheffieldSteering Committee
Chairman Consultant & Instructor John M. Campbell & Co.
Yasushi SakakibaraDeputy General Manager
LNG Contracts, Gas Resources Department
Tokyo Gas Company Ltd
An Zhong GuSecretary General
China LNG Association
Amit SinghVice President Global Trade
GAIL
Examining the Gas Infrastructure and Target Model to Ensure Security of Supply in Singapore
Assessing the Dynamics and Export Capacity of United States LNG to Asia
LNG & Pipeline Gas Imports and Consumption in China
Determining the Impact of Long Term Gas Contracts Versus Spot Trade in 2013 and Beyond
Tapping into the On and Offshore Reserves By Developing Small and Mid Scale LNG: Opportunities and Challenges in Asia
Balancing Growing Domestic Needs with Commitment to Exports
Key topics include:
Recreate PMS(from pdf )
Silver Sponsor
Industry Supporters
125x178 3mmbleed.indd 1 10/06/2013 15:55
ind news copy 14.indd 2 10/06/2013 16:12
www.eptq.com PTQ Q3 2013 135
equipment, including large tanks, for liquefied sulphur.
The contract includes modernis-ing sulphur platform and navigation systems at Mina Al-Ahmadi, Kuwait’s main crude oil exporting terminal. Once completed in 2015, the project will raise Kuwait’s annual sulphur output to 2 million tonnes from a current 850 000 tonnes.
Kuwait has three refineries with a daily production capacity of around 930 000 b/d, almost half of which comes from the Mina Al-Ahmadi refinery.
China funding for Costa Rica refineryChina is to provide Costa Rica with a $900 million line of credit to finance the expansion and remod-elling of the Puerto Limon refinery on the Caribbean coast. The refinery processes about 18 000 b/d of imported crude. The project could boost production to 65 000 b/d, with China sharing the profits.
The original plan was for an exports-only refinery to target demand in South East Asia and the Middle East. The consortium signed a memorandum of under-standing and carried out pre-feasibility and demand studies, but the project was put on hold in 2010.
The refinery was to be built to process sour and heavy crudes, while the petrochemical plant would process naphtha produced in the refinery.
Kuwait ups refinery sulphur outputKuwait National Petroleum Company has signed a $500 million contract with South Korean engi-neering firm Daelim Industrial for the repair and expansion of sulphur processing facilities at the Mina Al-Ahmadi refinery to the South of Kuwait City on the Arabian Gulf.
Daelim will carry out design, engineering and construction, as well as the test run of the new facil-ities. It will also install new
project near Vizag in Andhra Pradesh to meet local fuel demand. The state-owned company has reportedly spoken to Total and BP to join the project.
The 15 million t/y refinery and petrochemicals plant would be built about 70 km away from the company’s existing Vizag refinery, according to HPCL. A consortium led by HPCL and including steel-maker Mittal, Total, Oil India and gas company Gail in 2009 put the project on hold as petrochemical demand then was seen as too weak to justify the investment.
HPCL currently owns a 6.5 million t/y refinery at Mumbai and a 8.3 million t/y plant at Vizag. The Vizag plant is being expanded to 15 million t/y and HPCL is also building a 9 million t/y refinery at Barmer in Rajasthan. The company reckons that its petroleum product sales will be 50 million tonnes in 2020, while its refining capacity would be around 42 million tonnes with-out the new Vizag plant.
Speak at the Gastech ExhibitionCalling Experts to Present at the Centres of Technical Excellence (CoTEs)Gastech is calling on leading technical experts to submit presentations on the relevant streams at Gastech 2014’s CoTEs.
With highly-focussed sessions, provide your industry peers with a comprehensive outlook on:
Technological innovation Recent case studies Key advances in research and development
If you are interested in speaking at the CoTEs as part of a sponsorship or exhibition package please contact us at [email protected] or contact Michelle Thorby on +44 (0) 203 615 2876.
HOSTED BY
SUPPORTED BY
Deadline for presentation submissions is Friday 19 July 2013Visit www.gastechkorea.com/ptq
Natural Gas Vehicles (NGV)
Gas Monetisation
NGL/GTL
LNG as a Shipping
Fuel
Power Generation
HSSE LiquefactionOffshore & Subsea
Technology
Pipeline Infrastructure
LNG & Gas Carrier
Shipbuilding
CoTE Streams for 2014
Submit Your
Presentation Today
GASTECH14-COTE-125x178.indd 1 06/06/2013 16:50
ind news copy 14.indd 3 07/06/2013 20:25
42nd Turbomachinery & Pump Symposia 125
ABB Consulting 130
Aerzener GmbH 51
AFPM 2013 Q&A 132
Air Liquide Global E&C Solutions IFC
AMACS Process Tower Internals 50, 107
Axens OBC
BASF Corporation, Catalysts Division 28
BETE Fog 42
Bilfinger Water Technologies 94
Bryan Research & Engineering 88
Burckhardt Compression 55
Cat-Tec (Europe) 77
CB&I 7
Chevron Lummus Global 13
Criterion Catalyst & Technologies 4
CS Combustion Solutions 99
Curtiss Wright Flow Control 19
DigitalRefining 106
Downstream Asia 2013 IBC
Dupont Belco Clean Air Technologies 117
DuPont Sustainable Solutions 14
Elliott Group 97
Enersul 62
ExxonMobil Research and Engineering Company 73
Flexim 18
Foster Wheeler 34
Gas Asia Summit 134
Gastech 135
GE Oil & Gas, Surface Pumping Systems 66
GEA Heat Exchangers 56
Grabner Instruments 119
Hermetic-Pumpen 128
Hoerbiger Kompressortechnik Holding 27
Hydrocarbon Publishing Company 26
Hydroprocessing Associates 52
ITW Technologies 83, 108
John Zink Hamworthy Combustion 90
Johnson Matthey 87
KBC Advanced Technologies 2
KBR 17
KIDExtractor 21
Kurita Europe 81
Linde AG 40
Linde Engineering North America 70
MAN Diesel & Turbo 46
Merichem Company 114
Moscow Refining, Gas & Petrochemicals Week 102
OHL Guttermuth Industrial Valves 23
Onis International 59
Optimised Gas Treating 33
Panalytical 122
Process Consulting Services 8, 10
Prognost Systems 37
Prosim 121
Rentech Boiler Systems 32
Sabin Metal Corporation 84
Samson 101
Spectro Analytical Instruments 126
Streamlight 48
ThyssenKrupp Uhde 75
UOP 20
UOP Callidus Technologies 25
Yokogawa Europe 45
Zwick Armaturen 104
Alphabetical list of advertisers
136 PTQ Q3 2013 www.eptq.com
For more information on these advertisers, go to www.ptqenquiry.com
ad index.indd 1 10/06/2013 18:05
KEY REASONS TO ATTEND DOWNSTREAM ASIA
40+EXPERT SPEAKERS
40+INTERNATIONAL EXHIBITORS
40+CUSTOM MADE CONFERENCE SESSIONS
500+KEY DOWNSTREAM PROFESSIONALS
30 – 31 October 2013 • Sands Expo & Convention Centre, Singapore
Enhancing refineries and petrochemical firms’ competitiveness through energy efficiency, reliability & cost control
W: www.downstream-asia.com E: [email protected] T: +65 6590 3970 F: +65 6223 9196
Sign up as an exhibitor by15th July to enjoy a 10% DISCOUNT!
To redeem, simply quote the discount code: “DA10_PTQ”
Book your conference seat before15th July to enjoy a 10% DISCOUNT!
To redeem, simply quote the discount code: “DAEB_PTQ”
EARLY BIRD OFFER
REGISTERONLINE FOR YOUR COMPLIMENTARY
VISITOR PASS!Maximise your time by
talking to all your potential partners and suppliers over
2 days in 1convenient location
Part of:
Media Partners:
Supported by:
Silver Sponsor: Bronze Sponsor:
IHC Sealing Solutions
Organised by:
downstream.indd 1 07/06/2013 10:39
Stimulate the heart ofyour hydroprocessing unit
www.axens.net
ImpulseTM, the catalyst technology that combines the stability you recognize with the activity you need
Imp_PTQ.indd 1 04/12/2012 14:07:08axens.indd 1 6/12/12 13:17:48