exergy and industrial systems - smart-er.eu · 3rd international exergy, life cycle assessment, and...

3rd International Exergy, Life Cycle Assessment, and Sustainability Workshop & Symposium (ELCAS3)

07 -09 July, 2013, NISYROS - GREECE

Page | 243

EEXXEERRGGYY AANNDD IINNDDUUSSTTRRIIAALL SSYYSSTTEEMMSS



Page | 244



Page | 245

ORAL PRESENTATIONS – EXERGY AND INDUSTRIAL SYSTEMS

Title Page

ΙΙ.1 RENEWABLE AND DISTRIBUTED SOURCES WITHIN SMART ENERGY REGIONS 247

ΙΙ.2 MODELING OF CONCRETE ENERGY ACCUMULATOR COOPERATING WITH A VAPOR COMPRESSION HEAT PUMP

257

ΙΙ.3 EXERGY ANALYSIS OF BIOMASS COGENERATION SYSTEMS BASED ON GASIFICATION AND COMBUSTION

269

ΙΙ.4 SIMPLIFIED PERFORMANCE EXERGY ANALYSIS TOOLS FOR THERMAL POWER PLANTS

283

ΙΙ.5 MODELICA-BASED MODELING AND EXERGY ANALYSIS OF A CENTRAL HEATING SYSTEM

297

ΙΙ.6 COMPARATIVE ANALYSIS OF BIOPLASTICS PRODUCTION METHODS 307

ΙΙ.7 TECHNICAL AND SUSTAINABILITY ASSESSMENT OF SMALL SCALE BIOENERGY SYSTEMS

319

ΙΙ.8 TECHNICAL AND SUSTAINABILITY ASSESSMΕNT OF USED TYRES VALORIZATION THROUGH DEPOLYMERISATION

329

ΙΙ.9 CROSS-INDUSTRY INNOVATIONS IN THE RENEWABLE ENERGY INDUSTRY 339

ΙΙ.10 ENVIRONMENTAL PERFORMANCE OF ANTIMICROBIAL PAINTS ENHANCED WITH BIOCIDE SMART RELEASE TECHNOLOGIES

347

ΙΙ.11 ENERGY OPTIMIZATION PRACTICES FOR SUSTAINABLE OPERATION OF MBR WASTEWATER TREATMENT SYSTEMS

357

ΙΙ.12 THE USE OF A LOW ENERGY HYBRID MEMBRANE PROCESS FOR WATER TREATMENT

367

ΙΙ.13 EXERGY ANALYSIS OF THE POWER MODE OPERATION OF A BIMODAL NUCLEAR THERMAL ROCKET FOR THE NERVA CONCEPT

377

ΙΙ.14 LUMINOUS RECYCLING 387

ΙΙ.15 FUTURE OF SOLAR COLLECTORS: TİLT ANGLE OPTİMİZATİON FOR MAXİMUM PERFORMANCE

395

ΙΙ.16 A STATISTICAL COSTING APPROACH FOR EXERGOECONOMIC ANALYSES APPLIED TO A HEAT PUMP

405

ΙΙ.17 EXERGETIC ANALYSIS OF THE ENEXAL BAUXITE RESIDUE TREATMENT ON THE OVERALL RESOURCE EFFICIENCY OF THE PRIMARY ALUMINA REFINING PROCESS

427

ΙΙ.18 ENERGO AND EXERGO-ENVIRONMENTAL ANALYSIS OF A MULTIPURPOSE PROCESS FOR ETHANOL PRODUCTION FROM SUGARCANE AND CORN

437

ΙΙ.19 EXERGY ANALYSIS OF ALUMINUM RECOVERY FROM MUNICIPAL SOLID WASTE INCINERATION

449



Page | 246

ΙΙ.20 ENHANCING THE THERMAL INSULATION OF CLAY BRICKS USING FLY ASH AS PORE-FORMING AGENT

459

ΙΙ.21 MODELING THE GREEK ENERGY SYSTEM: SCENARIOS OF CLEAN ENERGY USE AND THEIR IMPLICATIONS

471

ΙΙ.22 BUILDING INTEGRATED PHOTOVOLTAICS 493

ΙΙ.23 THE USAGE OF ADVANCED EXERGETIC ANALYSIS FOR IMPROVING A GEOTHERMAL DISTRICT HEATING SYSTEM

503

ΙΙ.24 BENCHMARK ON COST AND EMISSION REDUCTIONS OF PIPE INSULATION OPTIMIZATION USING EXERGY ANALYSIS

511

ΙΙ.25 EXERGETIC PERFORMANCE PREDICTION OF GEOTHERMAL DISTRICT HEATING SYSTEM BY USING ANN METHOD

521

ΙΙ.26 FLUIDS IN LOW TEMPERATURE THERMODYNAMIC POWER CYCLES 531

ΙΙ.27 MESO AND MICRO SCALE NUMERICAL MODELLING FOR THE ASSESSMENT OF URBAN WIND AND THERMAL ENVIRONMENT

543

ΙΙ.28 FLOATING MARINE PARKS FOR POWER GENERATION 553

ΙΙ.29 EXERGETIC OPTIMIZATION OF GEOTHERMAL POWER PLANT 565

ΙΙ.30 AUTOMATED PET CLASSIFICATION IN WASTE MANAGEMENT SYSTEMS USING KERNEL PCA

575

ΙΙ.31 DOUBLE-EFFECT DISTILLATION AND THERMAL INTEGRATION APPLIED TO THE ETHANOL PRODUCTION PROCESS FROM SUGAR CANE

587

II.32 OPTIMIZATION OF LOW ENTHALPY GEOTHERMAL ENERGY SYSTEM FOR GREENHOUSE APPLICATIONS

599

ΙΙ.33 IS A NET POSITIVE ENERGY WASTEWATER TREATMENT PLANT FEASIBLE? 609

II.34

EXERGY-BASED METHODS APPLIED TO THE CHAIN “NATURAL GAS – LNG - NATURAL GAS” − LIQUEFACTION USING A SINGLE MIXED-REFRIGERANT PROCESS

615

II.35 SUSTAINABILITY ASSESSMENT OF POWER GENERATION IN COMBINATION WITH LNG EVAPORATION: A COMPARISON OF LCA METHODS AND EXERGY ANALYSIS 623

II.38 EXERGY-BASED METHODS APPLIED TO THE CHAIN “NATURAL GAS – LNG – NATURAL GAS” − REGASIFICATION OF LNG 635

II.36 EXERGY ANALYSIS OF THE JET THERMO-TRANSFORMERS 649



Page | 247

RENEWABLE AND DISTRIBUTED SOURCES WITHIN SMART ENERGY REGIONS

Jovan Todorović

ELEKTROPRENOS BiH a.d. Banja Luka Power Transmission Company of Bosnia and Herzegovina

Elektroprenos BiH, a.d. Banja Luka Marije Bursac 7a

78 000 Banja Luka, Bosnia and Herzegovina Email: [email protected],

Email: [email protected]

Abstract

The concept of smart energy region is supposed to be supplied with renewable sources integrated into smart grid. Connection issues of renewable sources impose new challenges and tasks for the smart/distribution grid regardless of source and connection type within regions. Efficient integration of renewable sources either concentrated or distributed along smart region stand as one of the main tasks for the reliable customer supply. In order to fulfil this requirement, a smart grid is provided with various measurement data generated from many locations within region. This paper presents different approaches in integration of renewable sources. Not predictable and intermittent nature of renewable generation, wind or solar power, influences on voltage fluctuation and consequently on power quality in a smart grid. Vast amount of measurement and forecast data are supposed to inform all "players" in smart grid to use electricity in efficient manner with minimum storage and export to distribution grid. The great potential in solar power can be utilized either for water heating or electricity production what can depend on subsidy policy (feed in tariffs). A household in countries with fair feed in tariff can have benefit of produced electrical energy from photovoltaic cells at annual level.

Keywords: Smart Grid, Distributed Renewable Integration,

1. Introduction

Before introducing a Smart Grid concept, households were consuming electricity in old

fashioned manner. Electricity consumption management was limited to choice to either

higher or lower electricity tariff, only. Power Utilities used this method to initiate users to

consume electrical energy during peak off hours in order to decrease peak in a power

system. The fast development of information and telecommunication technologies (ICT)

along with introduction of renewable sources in the last decade of 20th century has

revealed opportunity to distribution system operators and consumers to use electricity

efficiently. Installation of many metering systems at both distribution grid and consumers

side makes household consumers are being smart users and not only passive load but

also active electricity producers. Transforming the grid into a "Smart Grid" is not only

enabling energy efficiency and deployment of dispersed renewable resources but it

provides to the Distribution System Operators (DSOs) many other opportunities:

improving the business of the outage management, asset management, reducing

operational costs and supplying several services to different vendors of electricity and

aggregators [1]. Technical system transformation into smart grid has to be followed by the

regulatory authorities with proper regulations and subsidies. From the consumer point of



Page | 248

view, Smart Grid concept will be approved if there are power supply improvement and

decrease in electricity bills.

2. Smart Grid

Integration of ICT and renewable sources within power sector impose new challenges to

DSOs and gives more opportunities (services) for the power grid users (consumers and

producers). In such power grid, consumers are supposed to have options to choose

electricity retailer and organize (optimize) their own consumption. Furthermore, passive

consumers are encouraged, in a manner of feed in tariff, not to be passive more but

active. Consequently, participation of market, renewable sources and various metering

data make such power grid much more "complicated" and transform it into "Smart Grid".

Also, it can be explained that besides regular electrical infrastructure (hardware) Smart

Grid comprises and information infrastructure (software) in parallel [2]. Besides new

opportunities, Smart Grid has to preserve main performances, typically divided into two

parts: continuity of supply and voltage quality.

Smart Grid concept is seriously recognized from the highest authorities in European

Commission (EC), so many actions and projects have been supported by EC. One of

these is Smart Grid Task Force (SGTF) - European Task Force for the Implementation of

Smart Grids into the European Internal Market, set up by the Commission at the end of

2009. The ultimate goal of the SGTF Work Programme is to jointly produce a set of

regulatory recommendations and to identify projects of common interest to ensure EU-

wide consistent, cost effective, efficient and fair implementation of Smart Grids, while

achieving the expected services and benefits for all network users. The work and

initiatives on Smart Grids have been growing in number, participants and scope during the

last years in Europe. But the implementation of Smart Grids at a European level has been

fragmented since the beginning of the decade and not accelerated as expected. The main

reasons were/are the uncertainties regarding consumer acceptability, new retail market

models, business models for investors, the global investments needed and, to some

extent, the technology needed [3].

There are many definitions of Smart Grid concept and one of these is that: Smart Grids

can be described as an upgraded electricity network enabling two-way information and

power exchange between suppliers and consumers, thanks to the pervasive incorporation

of intelligent communication monitoring and management systems [4]. Graphical

explanation of such definition is presented (fig. 1.) where Distributed Energy Resources

are added as the additional part of Smart Grid.

Within Smart Grid consumers will be provided by information that helps them to make

more sensible use of energy. If consumers can save energy, they can see the benefit

directly as lower energy bills. The lower energy bills the less energy consumed and the

lower impact on environment.



Page | 249

Fig. 1: Smart Grid conceptual model [4]

Smart Grid concept stands as the one of the cornerstones for accomplishment of Smart

Regions (Smart Buildings, Smart Cities). Smart Grid network, electrical and data network,

are supposed to be key tools for efficient and reliable power supply of all consumers

within Smart Region. All users (fig. 1.) have to be provided by necessary data and be

ready to share data in order to be able to fulfill users' requirements.

3. Distributed generation

Before significant deployment of small scale renewable generation, at the end of last

century, consumers have been supplied by power produced from concentrated large

conventional units, by transmission and distribution grid. There were minor number of

small scale hydro power units, but percent of these was negligible, comparing to

nowadays and planned in future. Rapid development of renewable generation technology,

followed by proper regulatory subsidy policies and electricity price increase, have

motivated households to invest in small scale renewable generation unit, predominantly

solar and wind power. It makes them less dependant on power supply on distribution grid;

electricity produced can be either consumed for their own purposes or sell to distribution

grid. The renewable generation technology is getting improved every year and the pay

back of investment in such facilities is less, so many households've decided to invest in

their own small scale renewable production. This trend influences on transmission and



Page | 250

distribution grid, so a supplied region or city is not passive any more but active.

Consequently, concept of power system is now changed and transmission and distribution

system operators have to adopt new approaches in power system management. In this

scenario, central power stations will continue to exist, but in addition there will be a large

number of smaller, distributed systems. This change in structure demands the

coordination of the operation of a large number of systems and the electricity networks.

Thus, the importance of information and communication technology for energy systems

will further increase [6].

A successful transformation of the current electricity system designed around large-scale

centralized electricity generation towards a future electricity system with a large role of

small-scale distributed electricity generation requires an efficient integration of distributed

generation from two perspectives: market integration and network integration [5].

Generally, variable production from the renewable sources, either small scale distributed

or large scale, requires additional efforts from power system operators to keep power

system in stable operational point. Regardless, distributed generations have some

advantages against large scale production units:

- failure in one small scale production units has less impact on grid stability,

significantly,

- higher electricity supply independence,

- reduced transmission/distribution losses, electricity produced at local level is

consumed in local,

- reduced investment capital risk (no huge investments in large scale units),

- energy management on local/regional level – everyone will be able control your

consumption/production,

- support local industry – components production and maintenance employment,

- less independence on fossil fuels,

- reduced environmental impact of electricity production, etc.

Distributed renewable generation is supposed to be main energy sources within smart

cities/regions. As the part of a smart grid, distributed renewable generation should

improve supply security, reduce consumers' electricity bills and interact with distribution

grid in effective manner. All these tasks and actions assume to be followed by adequate

regulatory activities, such motivating feed in tariff and not complicated permission

documents provision.

4. Renewable distributed sources

Solar and wind power stand as the main renewable sources in smart regions. Depending

on weather conditions, feed in tariff, investment possibilities, urban infrastructure,

electrical connection issues, etc., the source type and size is chosen, usually. The single



Page | 251

unit installed power are 5, 10, 15 kW, what is quite enough to provide electrical power for

single household purposes and export to distribution grid in peak off hours.

2.2. Integration of distributed renewable sources

Integration of small scale renewable sources into electrical grid imposes fewer

requirements comparing to large scale renewable sources integration. Small scale

renewable units are connected to either low voltage (LV) or medium voltage (MV).

Influences on power system stability and power quality are less significantly, since

production variation or trip out of one small scale unit has local impact, only. Also, when

connected within smart region, connection lines length between sources and consumers'

loads are rather short what improves smart grid electrical stability [10]. While, production

variations of large scale unit have impact on whole power system stability and influences

on all consumers. Consequently, there are fewer technical requirements, from regulatory

bodies, and hence the procedure for production permission is much easier comparing to

one for large scale units. Generally, from the supply quality point of view, either for small

or large scale renewable production units, the main disadvantage of renewable production

is power production variations. It is hard to expect constant sun radiation (for PV

production) or constant wind sped (for wind turbine production) so power produced from

such sources are not "clean" like from conventional sources. In order to decrease such

"impurities" in some cases hardware interface is needed when connected to power grid,

i.e. VAR compensators, AC-DC-AC convertors, STATCOM etc

Distributed renewable generation deployment is very dependant on subsidy policy of the

regulatory bodies. Depending on feed in tariff, regulatory bodies can support solar or/and

wind power production. For example in Australia, with abundant solar power, the feed in

tariff for small scale photovoltaic (PV) units is in range 7.7 to 12.9 cents/kWh and average

retail electricity price is 24.4 $/kWh [7]. In Japan, with strong support policy on solar

production from small scale PV (up to 10 kW installed power), electricity paid to small

scale PV is 42 ¥/kWh and retail electricity price is about 23 ¥/kWh [8, 9]. In Bosnia and

Herzegovina, where regulatory agency has recently enacted feed in tariff to support small

scale renewable generation, for PV units up to 50 kW installed power, produced power will

be paid for 0.25 €/kWh and for wind power up to 10 000 kW installed power, produced

power will be paid for 0.083 €/kWh. The average retail electricity price is 0.05 €/kWh.

2.3. Small scale wind power

Serious wind utilization started in 1980s in Denmark. Since then, rapid development of

wind turbine design has started for both small and large scale, in parallel. Nowadays,

there are great offer of small scale turbines with different power, blade designs, for various

installation sites, etc. Usually, small scale wind turbines are up to 50 kW of installed

power. Selection of installation site in urban area is more complicated comparing in rural

area, not just because of lack of proper space but also there is wind wake effect if

installed close to buildings. So, inadequate site can make problems for optimal wind



Page | 252

turbine operation and decrease energy extracted from wind. Urban sites are within built-up

areas. They are likely to be quite close to buildings and other ground features (fig. 2.) [11].

Fig. 2: Example urban site [11]

When deciding install wind turbines within urban sites, there have to be considered

possible visible and sound pollution what could be important factor to choose PV units

instead wind power in urban sites.

2.4. Small scale PV and water heater units

Recent improvements of solar power technology have made PV cells financially affordable

and installation feasible for the small households. Assuming minimum sunny hours and

with adequate feed in tariff one household can deliberate about supplying for its own

purposes and export electricity to distribution grid. Installation of PV units in urban sites is

not such demanding comparing to wind turbines. PV units could be installed on the roofs,

walls, highway noise barriers (fig. 3.), etc. and be as the "natural part of construction".

Also, PV units have minor visible effects on landscape no sound pollution at all, no

rotating mechanical parts so installation and maintenance works is much easier

comparing to wind turbines.

Fig. 3: PV highway noise barriers [12]



Page | 253

PV cells convert sunlight directly into electricity. PV generates DC voltage what has to be

inverted to AC in order to connect to distribution grid. At the connection point with

distribution grid it has to me measurement equipment able to measure electrical energy in

both directions. These components make simplified grid tied PV unit (fig. 4.).

Fig. 4: Grid tied solar power system [12]

Another way to use solar power is solar water heating. More than 50 % of electricity

consumed in households is for water heating, used either for space heating or everyday

domestic purposes. Also, by this way no need to transfer solar power into electrical and

electrical power into thermal when heating water, so no necessary energy transformation

losses. Advantages of this system are very simple technology, much cheaper then PV

units, no need to connect to distribution grid and simple installation. If no economically

reasonable and technically feasible for PV installation, each Smart region/Smart city

should consider installation of such solar hot water system. One such system is presented

in Fig. 5.



Page | 254

Fig. 5: Solar Hot Water System [12]

Solar collector is the part which catches up solar power and has to be exposed to sun

radiation, while other parts can be located in a house. During winter cloudy days, the

boiler with hot water has to be heated up by electric energy in order to have hot water. In

the regions with not enough sunny days and average low temperature during winter this

system might not be used.

5. Conclusion

Integration of renewable sources and ICT within exciting power network is the key factor

for successful transformation of robust and nonflexible existing power grid into flexible and

smart power grid. Within such smart grid, consumers should have possibilities for

improvements of supply quality, lower electrical bills and less environmental impact. In

such grid, consumers have opportunities to install own renewable production distributed

along whole region/city. These renewable distributed renewable sources should be major

power supply for smart grid consumers. When distributed, power productions have many

advantages comparing to concentrate sources. These technical upgrades of power

network have to be followed by proper support from regulatory bodies.

In urban area, solar power utilization has many advantages against wind power utilization.

Solar thermal heater or PV units are easier for installation and maintenance, no sound

pollution, usually less visual impact and no rotating parts. If not feasible to install PV units

Smart region/Smart home should consider installation of solar water heater.

References

[1]. Yves Bamberger, Smart Electricity Distribution Networks, available at http://ec.europa.eu/information_society/activities/sustainable_growth/docs/sb_publications/pub_smart_edn_web.pdf (assessed on 17/01/2013).



Page | 255

[2]. Christine Lins, Needs and perspectives to accelerate the transition to renewable energy, (Freiburg 2012).

[3]. http://ec.europa.eu/energy/gas_electricity/smartgrids/doc/mission_and_workprogramme.pdf (assessed on 28/03/2013).

[4]. Vincenzo Giordano, Steven Bossart, Assessing Smart Grids Benefits and Impacts: EU and U.S. Initiatives, (Joint Report EC JRC – US DOE 2012).

[5]. Jeroen de Joode, Adriaan van der Welle and Jaap Jansen, Distributed generation and the regulation of distribution networks, available at http://cdn.intechopen.com/pdfs/10141/InTech-Distributed_generation_and_the_regulation_of_distribution_networks.pdf (assessed on 01/04/2013).

[6]. http://www.iset.uni-kassel.de/dispower_static/documents/fpr.pdf (assessed on 02/04/2013).

[7]. http://www.ipart.nsw.gov.au/Home/Industries/Electricity/Reviews/Retail_Pricing/Changes_in_regulated_electricity_retail_prices_from_1_July_2012/25_Jun_2012_-_Energy_Australia_-_Approved_annual_pricing_proposals/EnergyAustralia_-_Regulated_Electricity_retail_tariffs_and_charges_for_201213 (assessed on 02/04/2013).

[8]. http://www.polsoz.fu-berlin.de/polwiss/forschung/systeme/ffu/veranstaltungen_ab_2012/pdfs_salzburg/Takehama.pdf (assessed on 02/04/2013).

[9]. http://www.platts.com/RSSFeedDetailedNews/RSSFeed/ElectricPower/8242469 (assessed on 02/04/2013).

[10]. Jovan Todorovic and Mico Gacanovic, Electrical Connection and Stability Issues of the Single Wind Turbine Pilot Project, (International PhD Seminar on Computational electromagnetics and optimization in electrical engineering – CEMOEE 2010 10-13 September, Sofia, Bulgaria).

[11]. http://www.wind-power-program.com/Library/Policy%20and%20planning%20documents/Carbon-Trust-Small-Scale-Wind-Report.pdf (assessed on 03/04/2013).

[12]. Eric Buchanan, University of Minnesota Guidebook to Small-Scale Renewable Energy Systems for Homes and Businesses, (July 2012).



Page | 256



Page | 257

MODELING OF A CONCRETE ENERGY ACCUMULATOR COOPERATION WITH THE COMPRESSION HEAT PUMP

B. Salacinski a, V. Pisarev b & Gorski J. c

a OTTO Engineering Poland Ltd. 3, Wetlińska str., 35-082 Rzeszów, Poland

Email: [email protected],

b Heating and Air Conditioning Department Rzeszów University of Technology,

6, Powstańców Warszawy av., 35-959 Rzeszów, Poland Email: [email protected]

cAGH University of Science and Technology

Faculty of Energy & Fuels 30,Mickiewicza av., 30-059, Cracow, Poland


Abstract

The subject of this paper is an analysis and computer simulation of the concrete heat accumulator performance in cooperation with the compression heat pump. The objective of this study is twofold. On the one hand, it seeks to obtain the unsteady heat gains during daily solar operation and its use for water heating based on piping system embedded inside the wall. On the other hand, it aims to obtain the proper dimensions and the cooling fluid temperature and flows necessary for the heat pump operation. For this purposes, original software Akumulator v.2.0 has been developed (Salacinski, 2012). This program based on two-dimensional finite differences method (FDM) allows simulating unsteady thermal phenomena in the composed concrete baffle walls. For the practical purposes a procedure of calculating monthly and daily values of total solar radiation intensity for Poland’s climate conditions was also developed. The computer simulation of concrete heat absorber in cooperation with compression heat pump was carried out for meteorological conditions corresponding to Rzeszow (= 50,3 oN, L = 22,0 oE) in the daily and each month of the year. Obtained results showed the validity of such solutions for purposes of hot water preparation in the detached housing. Gained experiences also allow to properly choosing values of crucial design parameters of concrete heat accumulator, depending on the required energy demand.

Keywords: heat accumulation, heat pumps, solar energy, passive systems

List of Symbols cp : specific heat (kJ/kgK) : time q : solar radiation (W/m2) Subscripts T : temperature (oC) zew : ambient conditions Ab : radiation absorption coefficient sf : front surface V : volume flow [dm3/s] sg : upper surface B : wall thickness [m] st : back surface L : length [m] rura : pipe H : wall height [m] ochł : cooling dW : internal pipe diameter [mm] śr : average dZ : external pipe diameter [mm] j : per unit v : fluid velocity [m/s] wł : initiation Q : power [W] or energy stream [kW] wył : stoppage : heat transfer coefficient (W/m2K) cz, gl : cooling fluid : growth odp : evaporation : thermal conductivity (W/mK) skr : condensation



Page | 258

1. Introduction

Choice of low temperature heat source gives many possibilities what in turn allows to

widen implementation of systems equipped with compression heat pumps [7]. One of

such sources could be heat accumulating concrete elements that take the form of wall or

building’s fence. Absorption and storage of solar energy is due to the coolant ciculation in

the pipes embedded in a concrete heat accumulator. A comprehensive study of this

problem was extensively treated by D’Antoni and Saro in [1]. Design of such energy

absorber requires detailed specification of few factors such as: thickness and location of

the element, volumetric flow rate of cooling fluid, pipe diameter and pitch, temperature

difference determining the beginning and termination of the heat reception. Performance

analysis of concrete energy storage should also include the impact of changes in total solar radiation and convective heat transfer to the atmosphere [5].

The accumulation and use of energy from low-temperature concrete barriers, which may

take the form of building elements (terrace, garage wall) or which belong to his immediate

environment (fence), needs to equip this item in the effective internal cooling system [1].

Energy gained as a result of absorbers cooling can be used for domestic hot water

preparation in a system based on the compression heat pump (HP) unit. Schematic

diagram of the described system is shown in the Fig.1.

Fig. 1: Compression heat pump cooperation with the concrete accumulator [2]

(Legend: 1- concrete accumulator; 2 - cooling pipes; 3 – evaporator; 4-compressor; 5 – condenser; 6 - expansion valve; 7 - hot water tank; Tzew() – ambient temperature; qsf(), qsg(), qst() – solar fluxes on the front, upper, and back surface; Tcz – average temperature of cooling fluid; Tcz_z, Tcz_p – temperature of cooling fluid at the entrance and exit from evaporator; Todp, Tskr – evaporation and condensation temperature of the refrigerant; Tw_we, Tw_wy – temperature of intermediate fluid at entrance and exit to/from condenser; Twz – cold water temperature at entrance to storage tank; Tc.w.u. – hot water temperature; Ab – radiation absorption coefficient; Lrura – pipes pitch).



Page | 259

More detailed, two-dimensional (2D) model of concrete absorber for assumed conditions

of the heat exchange is shown in the Fig.2. Influence of solar radiation and convection

heat transfer on side surfaces (dimensions L and H) and upper surface (dimensions B and

L) were taken here into account, see, Fig.2. Rationality of an application the described

solution in the climatic conditions of Poland became author’s object of interest, see [2].

Verification of the system behavior from the point of view of functionality and effectiveness

was conducted based on 2D simulations of the unsteady thermal behavior and

temperature distribution in the internally cooled concrete elements.

Fig. 2: Two-dimensional model of the accumulation concrete wall [2]

(Legend: cz – heat transfer coefficient to the cooling fluid, other descriptions as for the Fig.1).

Solar radiation fluxes and ambient temperature were assumed as transient. Calculations

were performed with an original software tool Akumulator v.2.0 [3]. It is based on explicit

forward difference scheme in the Finite Difference Method (FDM). Accuracy of the

program was confirmed inter alia by fulfillment of requirements described in the standard

EN ISO 10211:2008, [4]. Aim of first calculation series was to determine the most

favorable values of accumulator’s design parameters which were assumed to be: cooling

fluid flow Vcz [dm3/s], concrete wall thickness B [m], distance between embedded pipes

Lrura [m], and the temperature difference between beginning and stop of the cooling

process Tochł [oC], (Fig.1). As a consequence of simulations it was established that:

decreasing of cooling fluid (ethylene glycol) velocity through the accumulator

affects an increase of its unit temperature growth Tgl_śr_j [oC/m]. It is crucial from

the point of view of compression heat pump operation, see, Fig.3. Simultaneously

it doesn’t cause a significant loss of power retrieved from an accumulator and has



Page | 260

a additional effect of reducing pressure losses in the pipes. The laminar flow was

assumed (at Vcz =0.2 dm3/s, mean velocity vcz = 0.137 m/s, and pipe internal dia dW

= 41.3 mm),

increase of the power received in the heat pump evaporator connected to the

concrete accumulator should be realized through several parallel absorbers or

embedding coils one next to the other. It provides for the conditions of laminar flow

minimum rise of glycol temperature Tgl_śr (approx. 3oC), that is normally required

by heat pump unit,

a favorable internal tube pitch was determined as Lrura = 0.09 m. This value

corresponds to the glycol temperature growth Tgl_śr, at the flow in the

accumulator, and average useful energy taken from the accumulator Qbet-rura_śr, and

the total pipes length Lrura_total. The proper ratios were calculated with taking into

account values from each current (index “n”) and next step (index “n+1”)

simulation, see, Fig.4. For an analyzed case the optimum tube pitch Lrura = 6 cm,

should be selected,

Fig.3. Increase of glycol temperature at flow in the accumulator Tgl_śr_j and unit average power transported by glycol qbet_rura_śr_j as a function of flow (pipe dia dZ= 48,3x2,6 mm) [2]

Fig.4. Relative change of average power Qbet-rura_śr and total pipes length Lrura_total as a function of tube pitch Lrura [2]



Page | 261

decrease of temperature difference Tochł=Twł -Twył, between initiation (Twł) and

pausing (Twył) of the cooling process has a positive influence on the accumulator

operation. With an early start the cooling process allows to receive a greater

amount of energy gained from the environment by a concrete accumulator, see,

Fig.5,

smaller thickness (B) of the accumulator wall contributes to increase of glycol’s

average temperature growth ∆Tgl_śr, and thereby average heat power gains Qbet-

rura_śr, (Fig.6). When choosing the dimension B, it should be taken into account the

destiny of concrete element forming part of the building structure.

Fig.5. Relative energy effectiveness of the accumulator vs. temperature difference Tochł [2]

Fig.6. Increase of glycol temperature and heat power gains as a function of thickness B [2]



Page | 262

Obtained values of accumulator’s crucial design parameters, i.e. Vcz, Lrura, Tochł, B,

(Figs.3-6), were used for a complete annual analysis of the system that is based on

cooperation of concrete thermal absorber with compression heat pump in the climatic

conditions of Poland (Rzeszow), [2,5].

2. Annual simulation of the system

2.1. Assumptions

In the annual analysis a typical concrete accumulator that takes a form of property’s fence

was taken into account. The following values of necessary parameters have been taken in

a detailed computer simulation:

Concrete properties: specific heat cp_bet=1 kJ/(kg·K), density bet=2300 kg/m3,

thermal conductivity bet=1.7 W/(m·K), radiation absorption coefficient Ab =0.94, [4].

Accumulator’s dimensions and orientation: height H=2.0 m, thickness B=0.12 m,

length L=1.0 m, configuration of the cooling tubes: steel pipes embedded in the

accumulator dZ=48.3x2.6 mm, pich Lrura=0.09 m, no. of pipes ilrur=20, total length

Lrura_total=20 m. The angle deviation from the south direction =0o, deflection from a

surface =90o.

Cooling fluid circulating in the embedded pipes: 36% water solution of ethylene

glycol (specific heat cp_gl =3.53 kJ/(kg·K), heat conductivity gl =0.424 W/(m·K),

density gl =1055 kg/m3 at 0oC). Volume flow of the cooling fluid Vcz =0.2 dm3/s,

and Tochł =1oC).

Convective heat transfer coefficient to the glycol cz calculated based on internal

flow similarity correlations. During lack of cooling flow it was assumed that this

coefficient is equal to cz_0 =200 W/(m2·K).

Heat exchange with an ambient: for surface that contacts ground (dimensions

BL), and both frontal planes (BH), lack of both solar radiation reception and

convection was assumed zew =0 W/(m2·K). For other external surfaces heat

transfer coefficients were instantaneous zew()=var. They were calculated by the

Akumulator v.2.0 with taking into account temporary values of ambient conditions

[6]. In the case of solar radiation flux, that impacts the accumulator performance,

some averaged local data (years 1971-2000) were taken into account and

expressed in [Wh/(m2·month). The data are valid for Rzeszów–Jasionka

(=50,06oN, L=22,03oE) meteo station and published by Ministry of Transport,

Construction and Maritime Economy, [5]. Based on the data and values of

atmosphere clearness coefficient the hourly total solar radiation intensity on

particular accumulator planes (respectively: qsf(), qsg(), qst(), see, Fig.1) were

calculated and used during the simulations. The albedo coefficients responsible for

the reflection of solar radiation are shown in the Table 1.



Page | 263

Table 1: Average monthly values of the albedo coefficient (śr) in Poland [5]

Month Jan. Febr. March April May June July Sept. Aug. Oct. Nov. Dec.

śr 0.43 0.33 0.3 0.18 0.18 0.19 0.2 0.19 0.18 0.17 0.17 0.33

Ambient temperatures Tzew(), in a particular month were determined based on the

Polish standard PN-76/B-03420 [6], and its appropriate values are given in the

Table 2.

Simulation period for the particular month include a time schedule from the first

sunny hour to 21:00. Computational initial average temperature of concrete Tp_bet,

in each month was assumed 2oC lower than ambient temperature for first sunny

hour, Table 2.

Average temperature of the coolant Tcz, at which receives heat from the

accumulator in a particular month was assumed to be approximately 5oC higher

than the average initial wall temperature Tp_bet, (Table 3).

Table 2: Average ambient temperatures in Rzeszow, Tzew() [5]

Month

Hour

5:00 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 19:00 20:00

Jan. - - - -13.2 -11.7 -10.3 -8.6 -7 -5.8 -4.5 -4 -4.5 -5.1 -5.8 -7 -8.6

Febr. - - -8.6 -7.1 -5.7 -4.2 -2.6 -0.9 0.3 1.6 2.1 1.6 0.9 0.3 -0.9 -2.6

March - -0.4 1.1 2.6 4 5.5 7.1 8.8 10 11.3 11.8 11.3 10.6 10 8.8 7.1

April - 7.3 8.8 10.3 11.8 13.2 14.9 16.5 17.8 19 19.5 19 18.4 17.7 16.4 15

May 11.3 12.8 14.3 15.8 17.3 18.7 20.4 22 23.3 24.5 25 24.5 23.9 23.2 21.9 20.5

June 14.5 16 17.5 19 20.5 21.9 23.6 25.2 26.5 27.7 28.2 27.7 27.1 26.4 25.1 23.7

July 16.3 17.8 19.3 20.8 22.3 23.7 25.4 27 28.3 29.5 30 29.5 28.9 28.2 26.9 25.5

Sept. - 17.8 19.3 20.8 22.3 23.7 25.4 27 28.3 29.5 30 29.5 28.9 28.2 26.9 25.5

Aug. - 14.4 15.9 17.4 18.9 20.3 22 23.6 24.9 26.1 26.6 26.1 25.5 24.8 23.5 22.1

Oct. - - 11.5 13 14.5 15.9 17.6 19.2 20.4 21.7 22.2 21.7 21.1 20.4 19.2 17.6

Nov. - - 4.8 6.3 7.8 9.2 10.9 12.5 13.7 15 15.5 15 14.4 13.7 12.5 10.9

Dec. - - - -2.4 -0.9 0.5 2.2 3.8 5 6.3 6.8 6.3 5.7 5 3.8 2.2

Temperature of the cooling fluid at the entrance to evaporator Tcz_z, was assumed

to be 2oC higher than temperature Tcz, see, Table 3. Fluid temperature at the

evaporator exit Tcz_p, was respectively lower by 4oC. Appropriate values are shown

in the Table 4.



Page | 264

Table 3: Average glycol temperatures at flow through the accumulator in a particular months [2]


Tcz, [oC] -10 -5 2 10 14 18 18 18 17 15 8 0

Table 4: Calculated glycol temperatures at the evaporator entrance (Tcz_z) and exit (Tcz_p) [2]


Tcz_z, [oC] -8 -3 4 12 16 20 20 20 19 17 10 2

Tcz_p, [oC] -12 -7 0 8 12 16 16 16 15 13 6 -2

The heat reception from an accumulator was initiated after reaching by it set

average temperature Twł, and terminated after reaching temperature Twył. It was

assumed that the end of heat receiving will take place when accumulator’s

average temperature is 2oC higher than temperature Tcz_z, of the cooling fluid at

the heat absorber exit.

Temperatures of initiation and termination of heat receiving from the accumulator

in the particular months are given in the Table 5.

Table 5: Temperatures of start and stoppage of accumulator cooling in particular months [2]


Twł, [oC] -5 0 7 15 19 23 23 23 22 20 13 5

Twył, [oC] -6 -1 6 14 18 22 22 22 21 19 12 4

In reference to the applied finite differences scheme and the stability conditions it

was assumed a time step = 10 s and the distance between nodes x = y =

0.01 m.

1. Results of the studies

As a result of studies concerned with thermal characteristics of concrete accumulator that

cooperates with the compression heat pump throughout the year, new and original data

have been obtained (Table 6). They are essential from the design practice point of view.



Page | 265

Table 6: Performance characteristic of concrete accumulator that cooperates with a compression

heat pump [2]

Parameter Month

Jan. Febr. March April May June July Aug. Sept. Oct. Nov. Dec.

QS, [kWh] 4.35 5.79 8.05 11.26 13.69 14.25 14.18 12.15 9.06 6.31 4.03 3.52

Q, [kWh] -0.19 -0.25 0.00 -0.96 -1.1 -1.61 -0.83 -0.29 -0.32 -0.38 0.31 1.73

Q [kWh] 4.16 5.54 8.05 10.3 12.59 12.64 13.35 11.86 8.74 5.93 4.34 5.25

Qbet-rura, [kWh] 1.9 3.3 6.0 8.4 10.4 10.5 11.6 10.0 6.7 3.8 1.3 2.2

U=Qbet-rura/Q, [%] 45 59 75 81 82 83 87 85 77 64 30 43

wł, [min] 216 235 234 239 267 272 269 219 262 247 282 189

wył, [min] 356 467 584 669 768 783 855 700 638 501 382 353

ochł, [min] 140 232 350 430 501 511 586 481 376 254 100 164

Tgl_śr, [oC] 1.1 1.2 1.45 1.56 1.65 1.63 1.58 1.66 1.41 1.2 1 1.1

Tgl_śr_j, [oC/(m)] 0.06 0.06 0.07 0.08 0.08 0.08 0.08 0.08 0.07 0.06 0.05 0.06

qbet-rura_śr_j, [Wh/(m) ] 93 164 300 418 518 523 581 502 335 191 65 112

qbet-rura_śr_j, [W/(m)] 40 42 51 58 62 61 59 63 53 45 39 41

Ej, [kWh/(m)] 1.86 3.28 6.0 8.36 10.36 10.46 11.62 10.04 6.7 3.82 1.3 2.24

Vwoda_j, [dm3/m2 Fbocz] 33 55 93 120 144 141 157 136 91 53 19 38

G40, [%/m2 Fbocz] 82.5% 138% 233% 300% 360% 353% 393% 340% 228% 133% 47.5% 95%

Markings in the Table 6: QS – total amount of solar energy that impacts on accumulator,

Q – energy transferred by convection, Q – energy that impacts an accumulator, Qbet-rura –

energy received by cooling fluid, U – efficiency expressed as a ratio of energy taken from

the accumulator by cooling fluid and total energy that impacts absorber, wł – time of

accumulator’s cooling process initiation that corresponds to the time of reaching average

temperature Twł, (Tab.5), wył – time of accumulator’s cooling process termination that

corresponds to reaching average temperature Twył, (Tab.5), ochł=wł-wył – duration of

cooling process, Tgl_śr – average increase of glycol temperature at flow through the

accumulator, Tgl_śr_j – unit increase of glycol temperature at flow inside accumulator,

qbet_rura_śr_j – unit average power transferred from accumulator to the cooling fluid, Ej –

energy received from accumulator per 1 meter of its length, Vwoda_j – amount of water that

can be heated with the heat pump usage (water temperature rise Twoda=42oC) per 1 m2 of

the accumulator front side area, G40 – percentage of covering of hot water demand (with

the consumption Vc.w.u.= 40 dm3/per person and day, [7]) referred to 1 m2 of an

accumulator front side area, see Fig.7.



Page | 266

Fig.7. Covering the hot water demand per unit accumulator front side area (water consumption Vc.w.u.= 40 dm3/(person·day), based on the Table 6) [2]

3. Conclusions

Elements of building’s wall or fences can be used as energy accumulators that allow

receiving and accumulating low temperature heat from an ambient. For domestic

purposes the concrete heat acumulator can complement or replace conventional low-

temperature sink with a shallow heat exchanger system from the ground, practically within

10 months of the year. Appropriately shaped absorbers, equipped with the embedded

elements that allow for internal heat reception, can cooperate with the heat pump systems

to prepare hot water in the detached houses. The system can prepare amount of hot

water that allows fulfilling over daily water demand. Conduct of annual calculation analysis

concerned two-dimensional temperature distribution in internally cooled concrete wall,

which is simultaneously subjected to impact of solar radiation and convective heat

exchange, allowed to determine crucial values of design parameters. They allow to

comprehensive sizing of concrete accumulator depending on desirable amount of

received energy or temperature growth of cooling fluid. Obtained results show that usage

of such heating system in climatic conditions of Poland can bring, in appropriate

configuration, beneficial results. Analysis’s results points that the monovalent system

characterizes daily periodical usage limitations. It results from the fact that analyzed

system is based on accumulation of energy that origins practically completely from solar

radiation (contribution of the convection is marginal). That in turn results directly in

possibility of heat pump’s operation only in particular time scopes that are defined as ochł,

when absorber keeps appropriate average temperature. Functioning of the system in such

configuration allows for its usage for domestic hot water preparation. Accumulated in

concrete, temporarily during day time, energy can be transported to an insulated tank.

This allows its use in period longer that it would result from possibilities of heat pump’s

function. That role can be fulfilled by the hot water tank, (Fig.1). Due to disadvantageous

of the winter season which characterizes itself by long time gaps between successive

cycles that allow a heat pump operation and small values of ochł, the tank should be

appropriately dimensioned. Dimensioning of concrete accumulator should always be

conducted based on the data concerning the most disadvantageous month. Time values



Page | 267

ochł, preclude the use of monovalent type systems in the small buildings, due to the lack

of energy supply continuity. It is necessary to carry out real experimental studies to

compare thermal performance of cooled concrete barriers and verify the results of

numerical simulation.

References

[1]. M. D’Antoni, O. Saro, “Massive Solar-Thermal Collectors: A critical literature review“,

Renewable & Sustainable Energy Reviews, 16 (2012) 3666–3679.

[2]. B. Sałaciński, “Usage of concrete constructions as a heat accumulators for water heating

systems in buildings”, (Ph.D. Thesis, Warsaw University of Technology, 2012, in Polish).

[3]. http://www.poczta.interia.pl, (freeware, login: [email protected]).

[4]. EN ISO 10211:2008 – Thermal Bridges In Building Construction - Heat Flows And Surface

Temperatures - Detailed Calculations.

[5]. https://cms.transport.gov.pl/2-48203f1e24e2f-1787735-p_1.htm, (assessed on 12/10/2012).

[6]. Standard PN-76/B-03420: Wentylacja i klimatyzacja – Parametry obliczeniowe powietrza

zewnętrznego. (in Polish).

[7]. H. Recknagel, et al.: Heating and air conditioning, Omni Scala, Wroclaw 2008.



Page | 268



Page | 269

EXERGY ANALYSIS OF BIOMASS COGENERATION SYSTEMS BASED ON GASIFICATION AND COMBUSTION

J. Taillon a, R. E. Blanchard b

a ANDRITZ, Carbona Inc. Tammasaarenkatu 1,

FI-00180, Helsinki, Finland Email: [email protected]

b Loughborough University

School of Electronic, Electrical and Systems Engineering LE11 3TU, Leicestershire, UK

Email:[email protected]

Abstract

This paper presents detailed exergy analyses of three biomass CHP plants; 1) Skive Denmark, an existing gasification plant with three gas engines (5.5MWe, 11 MWth) 2) IGCC A, a pressurized gasification plant integrated to a combined cycle (15.6 MWe, 7.8 MWth) 3) Plant C, a recently built combustion plant (22.5 MWe, 45 MWth). The paper uses two methods to calculate exergy destructions; exergy flows (for input/output streams and changes in streams chemical composition) and entropy balance. A recently published biomass exergy fuel formula is used. Results show that gasification provides 4 – 5% efficiency benefit over combustion and about 50% of total exergy is destroyed from biomass oxidation reactions. More important is to evaluate the whole process, upstream and downstream, which will use best the remaining 50% of available energy. The exergy destruction from a belt dryer with heat exchangers is 3%. Exergy dryer formulas are presented as they are complex and rarely published in the literature. Overall rational exergy efficiencies are; 32.8%, 31.4% and 31.8% for Skive, IGCC A and Plant C respectively. Published exergy efficiencies should not only show the rational efficiency value but should be compared with its inverse, the overall exergy destructions, so to validate the results. Correspondingly, the difference between the rational efficiency and its inverse for Skive and Plant C gives an acceptable 0.5% and 0.8% respectively. For the IGCC, this could not be compared because the manufacturer does not disclose gas turbine isentropic efficiencies.

Keywords: Exergy; Gasification; Combustion; Biomass; Cogeneration

1. Introduction The following three biomass cogeneration power plants, supplying district heating, are

analyzed;

Skive; existing gasification plant feeding three gas engines, located in Skive,

Denmark.

IGCC A; pressurized air gasification plant integrated to a combined cycle.

Plant C; recently built combustion steam cycle plant equipped with the latest

technology.

The first part of this paper will verify and quantify the claim that product gas from

gasification is more valuable than flue gas from combustion. Although this point was

reported by Prins et al. [1] with simulated data, this paper will use data from an existing

gasification plant (Skive). The next section will present detailed breakdown of exergy

destructions of the power plants.



Page | 270

Units, List of symbols

Symbols Subscripts, acronyms and abbreviations Cp Specific heat (kJ/kg) #0 (superscript) formation

Specific exergy (kJ/kg) #0 (subscript) at dead state Exergy rate (kJ/s) AF as fed

d Exergy destruction (kJ/s) ch chemical exergy e Stream flow exergy (kJ/s) CV control volume

ER Equivalence Ratio db dry basis h Enthalpy (kJ/kg) e exit LHV Lower heating value (kJ/kg) env environmental M Molecular weight (kg/kmol) f stream flow ṁ Flow rate (kg/s) g saturated n Number of moles i inlet or index p Pressure (bara) in input (fuel)

Heat energy rate (kJ/s) j index ′ Exergy heat rate (kJ/s) no env non environmental Universal gas constant L loss

s Entropy (kJ/kg K) ph physical exergy y mole fraction ref reference v specific volume (kg/m3) s source W Work (kJ/s) tot total

moisture (%-weight) u used T Temperature (0C, K) HEX Heat exchanger th Thermal

Greek symbolsσ Entropy production (kJ/kg K) η Energy efficiency (%)

µ Chemical potential (kJ/mol) Ψ Exergy efficiency (%)

Dryer flow sheeta air stream 1 State; between heat exchangers / dryer p wood (product) 2 State; inlet of fuel feeding x air inlet of heat exchangers 3 State; exit of moist air from dryer y water 4 State; exit of fuel from dryer

2. Power plants process description 2.1. Skive

The existing Skive plant (Fig 1) gasifies wood pellets into a product gas which feeds three

gas engines. To assist in the fluidization process, limestone as bed material is fed to the

bubbling fluidized bed (BFB) gasifier with air under sub-stoichiometric conditions.

Unwanted tars and ammonia present in the gas are decomposed at high temperature in

the catalytic reformer. To remove impurities (e.g. alkali metals), the gas is cooled down to

2000C while the sensible heat is recuperated for district heating. The filter removes fly

ashes and condensed harmful elements from the gas. To further remove water soluble

impurities (e.g. NH3, HCl), the gas goes through a wet scrubber [2][3]. Reheated to 400C,

the gas is fed to three low calorific value gas engines (GE Jenbacher JMS620GS, 2 MWe).

Gas engines exhaust heat is recovered to feed the district heating network. The plant is

operated with sliding pressure to maintain stable gas volume flow for various plant

capacities (50-130%) [4]. Skive produces net 5.5MWe and 11.3MWth with an energy input



Page | 271

of 19.5 MWfuel.

Fig 1: Skive flow sheet

2.2. Integrated Gasification Combined Cycle (IGCC) A

This plant gasifies dry wood chips (Fig 2). The pressurized (20 bara) product gas is fed to

an integrated combined power cycle composed of a gas turbine, heat recovery steam

generator (HRSG) and steam turbine. The dryer is of a belt type to provide long residence

time for the wood chips and uses hot water as a low heat source. The 45%-w moisture as

received wood chips are dried to 20%-w. The 8500C gas from the BFB gasifier is cooled

down to 4000C and filtered to remove fine particulates. Air and steam are used as

oxidation agents. The product gas has a low calorific value and for a similar energy input

(or inlet gas turbine temperature), the mass flow with biomass is much higher than with

natural gas. Since the gas is mainly H2 and CO, significantly less air is needed to oxidise

the gas which implies that less air is needed from the turbine compressor. As the gas

turbine is designed to run on natural gas to minimize costs, compressed air is bled off the

compressor. This valuable compressed air is recirculated in the form of a product gas at a

pressure controlled by a booster compressor. The gas turbine is of a standard design

(MAN, THM 1304-14) aside from the modified extraction air and combustion systems. Gas

turbine exhaust heat is converted to steam in the HRSG and fed to the steam turbine as in

conventional combined cycles. The steam turbine can operate in condensing or back

pressure mode. The condenser is cooled by air in the cooling tower and the condensate

system includes preheating (E) where the heat of the gasification air (B) is utilized. The

back pressure heat (D) is sent to the district heating network. The plant produces net

15.6MWe and 7.8MWth with an energy input of 43.5MWfuel



Page | 272

Fig 2: IGCC A flow sheet

2.3. Combustion Plant C

The combustion Plant C is equipped with a biomass BFB boiler and feed a conventional

steam cycle. The plant supplies net 22.5 MWe and 45 MWth of district heating with an

energy input of 76 MWfuel from 100% wood chips.

3. Methodology 3.1. Exergy key equations

The exergy destructions were calculated from exergy flows [5];

∑ 1 ∑ ∑ (1)

(2)

∑ 0 ∑ ∑ . (3)

and in the absence of chemical composition changes, from entropy balancesa;

∑ ∑ (4)

3.1.1. Efficiency calculations

Because exergy is not conserved, exergy analyses do not validate themselves as a Skive exergy destructions were calculated entirely based on exergy flows while for IGCC, it was a combination of both methods; exergy flows and entropy balances.



Page | 273

naturally as energy analyses do. To ensure the amount of destruction is correct, the

rational efficiency equation 5 must match the inverse equation 6 using total exergy

destruction. Equality between these two equations is the proof that exergy destructions

are valid and the rational efficiency equation serves as the reference value.

′ (5)

1 (6)

4. Results & Discussions

Verify and quantify the claim that product gas is more valuable than flue gas.

Using only energy conservation principles Prins et al [1] reported no difference in total

energy content for biomass oxidized based products. Thus for a hot bioliquid from

pyrolysis (ER = 0), a flammable product gas from gasification (ER = 0.25) or a hot flue gas

from combustion (ER = 1.25), regardless of their chemical energy (if any) and sensible

heat content, the overall energy efficiency of the power plant will solely depends on the

other plant sub-processes. However, using exergy concepts, Prins reported that biomass

oxidation at an equivalence ratio associated to gasification is more efficient than

combustion. This section quantifies this claim with data from an existing gasification plant.

Based on Skive gas composition (%v; CO 21%, CO2 11%, H2 17%, H2O(g) 10%, CH4 4%,

N2 38%), the stoichiometric equations for a) Skive and b) for combustion at ER = 1.25b are

derived below;

. . . 0.1 (7)

0.26 3.76 → 0.55 0.28 0.43 0.25 0.1

0.99 0.012 Ash (8)

1.296 3.76 → 0.82 0.1 4.87 Ash (9)

The calculated input fuel chemical formula for Skive ( . . . is similar to other

sources [1][6][7][8]. The resulting exergy content per kg of wood is given by;

(10)

From Fig 3, the product gas exergy content exceeds the hot flue gas by 43% whereas

Prins found an increase of only 23% with fuel CH1.4O0.59. The difference (43% vs. 23%)

comes mainly from the physical exergy content. This paper used the maximum

temperature (13000C) in the furnace of the existing combustion Plant C. It is probable that

Prins used a higher furnace temperature increasing the exergy content of the flue gas.

Simulating these temperatures may be difficult. The highest achievable temperature is

called adiabatic flame temperature and cannot be achieved because at high temperatures

b Boiler typically operates at 25% excess air.



Page | 274

some combustion products may dissociate (e.g. CO2 to CO) and these endothermic

reactions lower the temperature [5].

Fig 3: Exergy content versus ER

Biomass oxidation in gasification processes is done in two steps (gasifier and heat

engines) as opposed to one step in combustion (furnace). As will be shown later, in terms

of fuel input, total exergy destructions from oxidation in Skive, IGCC and Plant C are 49%,

48% and 53% respectively which indicates a 4 - 5% higher exergy destruction for the

combustion case.

Thus the most energetic process occurs at equivalence ratio related to gasification

(excluding pyrolysisc). To quote Prins; “The reason that gasification … is more efficient is

that the exothermic oxidation reactions are effectively coupled with endothermic reactions,

so that the driving force for the overall chemical reaction (difference in chemical potential)

is lower.” [1]. Irreversibilities are null when chemical equilibrium is achieved, that is when

no spontaneous reaction is occurring. Any spontaneous reaction, regardless how long it

takes, always creates irreversibitlities. Since combustion is a combination of highly

spontaneous reactions, for the most part exothermic, it is further away from its own

chemical equilibrium when compared to gasification. The combination of endo- with

exothermic reactions in gasification means that it is closer to its own chemical equilibrium.

Therefore, combustion leads to more irreversibility, thereby decreasing its thermodynamic

efficiency. In thermodynamic terminology, the equilibrium criterion is expressed by [5];

∑ μ 0 ; μ (11)

In other words, chemical equilibrium is achieved when chemical potential (µi) is 0 and µi is

calculated by the Gibbs function, the fundamental equation of exergy. Variable (ni) is non-

zero as it represents the number of moles in ith gas component of the gas mixture.

c For pyrolysis bio liquid, the sensible heat, which gives its exergetic superiority, cannot be stored.



Page | 275

Detailed exergy analysis of Skive power plant

If previous section showed that 50% or so of available energy was lost in the biomass

oxidation reaction itself, this section examines the upstream and downstream processes

of the power plants.

To determine fuel input exergy, papers reviewed in the literature research used a well-

known formula [9] developed by Szargut and Styrylska in 1964. This paper used a

different equation, introduced recently [5] which determines biomass exergy on a dry

basis. The equation is easier to use and its authors claim a good correlation with the

traditional model.

1812.5 295.606 587.354 17.506 17.735 95.615 31.8 (12)

The exergy value on a wet basis uses a similar logic as with the higher heating

value; the latent heat of vaporization is not subtracted because exergy is based on a

reference point (dead state). For a moist fuel, the wet lower heating value

decreases because a portion of the combustion heat is used to evaporate moisture in the

fuel and is not condensed to return the heat back to the system [10].

1 (13)

1 (14)

Table 1: Skive detailed exergy balance

Fuel input data DETAILED kJ/s %

C,w%db H,w%db N,w%db 0,w%db S,w%db ash, w%db moisture, w%

50.20% 5.80% 0.10% 42.70% 0.03% 112% 45%

Initial amount of exergy (= Fuel exergy) Net Power Gross power Power consumption District Heating Flue gas exhaust

21514 5,528 5,904 (376) 1523 738

100% 26% 27.4% -17% 7% 4%

Exergy destruction

Ed phys (kJ/s)

Ed chem (kJ/s)

Ed tot (kJ/s)

Ed/Ein (%)

Gasifier (1,638) 5,777 4,139 19%

Reformer (70) 352 282 1%

Cooling 1,361 0 1,361 6%

SUMMARY Filter 9 (27) -18 0%

TOTALEdestruction (kJ/s) 13,778 Scrubber 119 58 177 1%

Overall exergy efficiency Eout/Ein 32.8% Gas Heater 11 0 11 0%

Overall exergy efficiency 1 -Ed/Ein 32.3% Gas Engines (8,174) 14,639 6,465 30%

Electrical exergy efficiency (net) 25.7% Heat Recovery

1,361 0 1,361 6%

From table 1, Skive has overall exergy efficiency of 32.8% with a 0.5% gap between the

two efficiency equations 5 and 6 which validates the exergy destruction balance. Net

electrical and thermal exergy efficiencies are 25.7% and 7% respectively. For the gasifier,

the ratio of exergy destruction to input fuel exergy is 19% which corresponds closely to

17% obtained in section 5.1. Although the temperature rises in the gas engines, they

generate the highest exergy destruction due to losses in chemical exergy (30%). The



Page | 276

relatively high exergy losses in the cooler and heat recovery are caused by large

temperature differences between the hot and cold streams. An increase in the gas inlet

temperature of the engines would improve the efficiency with less heat transfer

irreversibilities in the cooler and heat recovery but at the same time, decrease the

volumetric efficiency and most probably, the overall output of the gas engines [11].

Detailed exergy analysis of IGCC A power plant

4.3.1. Dryer

Drying is often needed with gasification because fuel moisture should usually not exceed

20%. The energy and exergy models are based on the comprehensive work of Dincer

[12][13][14] with the addition of air to water heat exchangers before the belt dryer.

The mass and energy balance equations and results are shown below and in table 2.

Specific evaporation:

(15)

Wood: 1 1 (16)

Air: (17)

Water (in BDS); (18)

Water (in HEX): (19)

HEX; (20)

Table 2. Dryer input data, mass and energy balances

X yi y2 1 2 4 3 Dead state

Other dead state properties

T °C 10 90 60 85 8 30 44 25 (Xv)0 0.0240

Ρ bara 1.013 3.500 3.500 1.013 1.013 0.982 1.013 (xs) 0 0.0313

ω % 0.53% 0.53% 45% 20% 1.92% 1.53% s (Τ0,Ρ0) kJ/kgK 8.6800

Φ % 70% 1.47% 34% 77% Sf (Τ0,Ρ0) kJ/kgK 0.3673

m total kg/s 110.27 110.27 4.89 3.36 111.80 Pg(T0) bara 0.0317

m fan air kg/s 109.69 109.69 109.69

m water kg/s 49.61 49.61 0.58 2.20 0.67 2.11

m wood db kg/s 2.69 2.69

pg bara 0.0123 0.5787 0.0911 0.0317 Constants

pv bara 0.0086 0.00852 0.03079 0.02433 Specific heats air (Cp)a 1.00

hair kJ/kg 283.15 358.15 317.15 2,546.92 Specific heats vapor (Cp)v 1.87

h water kJ/kg 2,519.5 377.2 251.4 2,660.2 33.7 125.8 2,582.6 104.8 Gas constant air Ra 0.2881

QLOSS kJ/s 50.0 26.00 Gas constant vapor Rv 0.4615

W kW 420.0 Spec. Evap. Energy kJ/kgwater 4,079



Page | 277

A key input variable is the specific evaporation (Evap) of the dryer set at 4,079 kJ/kg of

water evaporated. With above equations, all unkowns can be derived except for the heat

losses (QL1=50 kJ/s, QL3=26 kJ/s) and the fan efficiency (ηfan=76%). The authors believes

these values to be plausible.

The key exergy equations for the drying are listed below;

1 ∑ E ∑ E (21)

∑ (22)

∑ (23)

The as-fed wood chips temperature to the dryer and gasifier are assumed to be the same.

The chemical composition of the wood chips does not change except for the water being

removed. Therefore, the difference in exergy for the solid content in the wood chips (efp2

and efp4) is taken as zero.

For an air + vapor stream, efk, (efx, ef3, ef1) [12];

.

.1.6078 (24)

For the water content in wood, efk , (efw2 and efw4)[12];

(25)

In the above equation, the vapor saturated reference values hg(T0) and sg(T0), proposed

by Dincer [12], is replaced by the liquid saturated values hf(T0) and sf(T0) because the

water content in the wood is in liquid form thus more representative of the actual case.

This is relatively significant as it represents a 3% decrease in the dryer exergy destruction

or 0.1% exergy destruction of the exergy fuel input.

To calculate the exergy value for the heat losses, Tav is needed and is given by;

(26)

For the water, efyk (efy1 and efy2) [12];

(27)

The heat exchanger exergy efficiency is defined as [5];

66% (28)

The dryer exergy efficiency is expressed as [13];

5.9% (29)

Combining equations 28 and 29, the overall drying exergy efficiency is calculated as such;

6.8% (30)



Page | 278

To explain these low efficiency values, removing water from wood is difficult. In wood,

water is contained in three forms [15]. Free water is found in cell cavities or lumens in

liquid form, water vapor is present in the air within the lumens and the cell wall material.

Free water leaves the lumens before bound water. Water can be removed fairly easily up

to the point it reaches its fiber saturation point (FSP). The FSP is the point at which

moisture is saturated within the cell walls and the cell cavities are free of water. FSP for

most wood species falls in the range of 25-35% moisture content. Beyond this point, it

becomes harder to remove water. Still, the exergy destruction for the dryer is relatively

small at 3% (table 3).

4.3.2. IGCC A: Gasification Plant

Only a portion of the product gas exergy is transferred to the turbine itself. The rest is lost

in the combustion chamber as irreversibilities due to the spontaneous thermo-chemical

reaction as shown in Fig 4. The heat input to the turbine is the exergy difference between

the outlet and inlets (air and product gas) of the combustion cha mber.

The exergy destruction in the combustion chamber is calculated by;

(31)

(32)

Input data for the inlet (2) of the combustion chamber are known but for the exit (3),

temperature and pressure are unknownd. Using a weighted average of the mass flows

between the two known inlet pressures of gas and air, an exit pressure was calculated (11

bara). The pressure at state 3 makes almost an isobar with state 2 (11.2 bara), giving an

acceptable 2% pressure loss through the chamber. The value of the temperature has a

significant effect on the efficiency as it determines the heat input to the gas turbine. For

IGCC A, an apparent temperature of 9530C was found such that the two efficiencies

(Ψoverall 1, Ψoverall 2) are equal. The actual combustion temperature is expected to be higher

(about 10400C), but since cooling air internally by-pass the combustion chamber to cool

the blade’s tip in the turbine, this apparent temperature does not represent the true

combustion temperature.

The exergy balance shown in table 3 shows that IGCC A has an overall and electrical

exergy efficiency of 31% and 28.6% respectively. The thermal exergy efficiency (2.7%) is

low because the plant could not sell all of its heat output. Together, irreversibilities from

spontaneous chemical reactions in the gasifier and the combustion chamber account for

48% compared to 30% if the fuel would be natural gas [5][16][17]. Without its combustion

chamber, total exergy destruction in the combined cycle accounts for 13%. The booster

compressor loop with its two heat exchangers has only 1% exergy destructions. Although

the product gas quality could be improved by reducing the fuel moisture further and thus

d Proprietary information from turbine manufacturer



Page | 279

increasing its heating value, it would not change the turbine efficiency. Gas turbines are

limited by the inlet temperature and increasing heat value would result in more air dilution

from the compressor to maintain the inlet temperature under a certain level.

Fig 4; T-s diagram of combined cycle

As can be seen from the logarithmic pressure term of equation 4, the high pressure (20

bara) of IGCC gives a 4% exergy efficiency gain because it reduces changes in entropy.

Finally, the exergy from fly and bottom ashes were neglected because they represent only

46 kJ/s and 27 kJ/s respectively [18].

Table 3; IGCC A Exergy destruction

Fuel Input Data EXERGY DESTRUCTIONS Ed (kJ/s) Ed/Exin (%)]

C,w%db H,w%db N,w%db 0,w%db S,w%db ash, w%db moisture, w%

49.50% 5.60% 0.20% 42.10% 0.10% 2.50% 45%

DRYER 1,802 3%

GASIFIER 11,230 21%

Gasifier (Ed physical) (5,446) -10%

Gasifier (Ed chemical) 16,676 31%

COOLING 1,097 2%

SUBTOTAL 14,128 26% IGCC (TOTAL) 21,877 40%

DETAILED kJ/s % Combustion Chamber 14,737 27%

Fuel input exergy 54,550 100% Compressor 382 1%

Net Power 15,628 29% Gas Turbine 2,078 4%

Gross power 17,730 33% Booster compr. (C) 419 1% Power consumption (2,102) -4% HEX 1 (A) 31 0%

District Heating 1,491 3% HEX 2(B) 33 0% Loss to e nvironment 1,426 3% HRSG 2,425 4%

Flue gas exhaust 637 1% Steam Turbine 1,017 2%

Cond. Cooling tower 351 1% Extraction (D) 417 1% Drying exit moist air 437 1% Condensation (0) 0%

SUMMARY Pumping (0) 0%

TOTAL Edestruction (kJ/s) 37,431 HEX 3(E) 318 1%

Overall exergy efficiency Eout/Ein 31.4% Stream mix (F) 12 0%

Overall exergy efficiency 1-Ed/Ein 31.4% FWT mix (G) 8 0%

|Electrical exergy efficiency (net) 28.6%



Page | 280

Exergy Analysis of Combustion Plant C

For this case, only the following basic data were known; fuel input energy and moisture,

electrical and heat outputs with boiler steam data. To deduct the plant performance, the

following steps are done (table 4); A) The exergy destruction in the furnace (53%) for the

thermo-chemical reaction can be derived because the chemical reaction is known

(equation 9) and is evaluated at a furnace temperature of 1300 0C. B) Keeping the

existing capacity of Plant C with the above 53% ratio, the combustion exergy destruction

becomes 51,977kJ/s based on IGCC A fuel composition at 50% moisture. C) With steam

at 5250C and 117 bara, the exergy output is 38,118 kJ/s giving thermal transfer losses

between the hot flue gas and the heat input to the steam turbine of 8,348 kJ/s. D)

Assuming similar destructions in the steam cycle (17%) and losses to the environment

(1%) as in the IGCC case, the total exergy destruction of the combustion plant represents

69% of the fuel input. The overall exergy efficiency is 31.8% with a gap of 0.8% between

the two efficiency equations 5 and 6 which validates the exergy destruction balance.

Table 4; Combustion exergy destructions

A) Simulated case Fuel input data

C,w%db H,w%db N,w%db O,w%db S,w%db ash, w%db moisture, w%

49.50% 5.60% 0.20% 42.10% 0.10% 2.50% 50%

Fuel exergy INPUT (specific) kJ/kg 18,253

B) Mode Combustion Gasification

Exergy out kJ/kgwood 8,615 15,210

Exergy destruction kJ/kgwood 9,644 3,050

Exergy destruction 53% 16.7%

Exergy efficiency 47.2% 83.3%

Edestr/Ein Simulation 53% 17% Exergy balance Plant C Ex destr (kJ/s)

Ex balance (kJ/s)

Edestr / Ein Skive 19% Exergy IN (fuel) 98,442

Edestr / Ein IGCC 21% Combustion (Ed) 53% 51,977

C) Exergy OUT (steam) ACTUAL PlantC IGCC A Exergy in furnace 46,466

Flow rate m kg/s 26.1 7.6 Exergy OUT (steam) 38,118

Pressure process p bara 117.0 60.0 Thermal transfer (Ed) 8,348

Temperature steam T OC 525 470 Steam cycle (Ed) (17%) 6,642

Exergy OUT (steam) ACTUAL kJ/s 38,118 10,164 Loss to environment (Ed) 984

D) Ex destruction Steam Cycle IGCC A TOTAL (Ed) 67,951

Ex destr in IGCC kJ/s

1,771 Total(Ed) / Exergy IN 69%

Steam turbine Ex INLET kJ/s

10,164 Exergy OUT(useful) 30,491

Edestr/ Ex INLET ST 17% Power (net) 22,536

DH (exergy) 8,802

Exergy efficiency (total) (1-Ed/Edtot) 31.0%

Exergy efficiency (total) (Eout/Ein) 31.8%

Exergy efficiency (net electrical) 22.9%

5. Conclusions With respect to the biomass oxidation reaction itself, calculations show that less

irreversibilities occur with the two oxidation steps of gasification (gasifier with heat engine)

than with the one oxidation step (furnace) of the steam combustion plant. The calculated



Page | 281

exergy efficiency benefit for gasification is 4 to 5% and is explained by the combination of

endo- and exothermic reactions in the gasifier.

Nonetheless, biomass oxidation in one or two steps, consumes about 50% of the fuel

input exergy compare to 30% for gas fired power plants. Therefore, it is important to

evaluate the whole process, upstream and downstream, which will use best the remaining

50% of available energy. This type of detailed exergy analysis provides key information to

engineers involved in design improvement, performance comparisons etc.

The fuel input energy and exergy content with efficiencies summary are shown in table 5;

Table 5: Summary of efficiencies and fuel input energy and exergy content.

Fuel input Efficiencices

Energy MJ/s

Exergy MJ/s

Ψtot %

Ψel% Ψth %

Skive 19.5 21.5 32.8% 25.7% 7.1%

IGCC A 43.5 54.6 31.4% 28.6% 2.7% Plant C 76.0 98.4 31.8% 22.9% 8.9%

Skive presents the highest overall exergy efficiency mainly because pellets are used and

no energy is spent on drying. The IGCC concept gave the highest electrical efficiency

because it relies on a combined cycle at high pressure. However, for this particular case,

the district heating customer could only accept half of the useful heat reducing significantly

its overall exergy efficiency.

The dryer efficiency is relatively low because it is hard to remove water beyond the fiber

saturation point but still, its overall exergy destruction is relatively small at 3% of fuel input.

To validate calculations, conventional analyses use mass and energy balances. Similarly,

this paper recommends that exergy destructions should always be checked against

rational exergy efficiency. For any detailed plant study, it is unlikely that a perfect match

will be found but the comparison is an effective way to validate the total exergy

destructions.

References

1. M. Prins J., K. Ptasinski J. and F. Janssen J.J.G., "Thermodynamics of gas-char reactions: first and second law analysis," Chemical Engineering Science, vol. 58, pp. 1003-1011.

2. M. Ketomäki, "Clean-up and conditioning of biomass derived synthesis gas: Utilization of experimental data in evaluation and optimization of a tar reforming unit.", (M.Sc. Thesis, Aalto University, School of Chemical Technology, 2011).

3. V. Pietarinen, "Optimization of biomass-based synthesis gas quality for chemical production", (M.Sc. Thesis, Tampere University of Technology, Degree Programme in Environmental and Energy Technology, 2007).

4. K. Salo, A. Horvath, J. Patel, “ANDRITZ Carbona gasification process for synthesis gas production”, International Conference on Polygeneration Strategies 2013, September 3-5th 2013, Vienna, Austria



Page | 282

5. M. J. Moran and H. N. Shapiro, Fundamentals of Engineering Thermodynamics. Chichester: John Wiley & Sons, Ltd, 2006.

6. G. H. Song, L. H. Shen and J. Xiao, "Estimating Specific Chemical Exergy of Biomass from Basic Analysis Data," Ind Eng Chem Res, vol. 50, pp. 9758-9766.

7. M. J. Prins, K. J. Ptasinski and F. J. J. G. Janssen, "From coal to biomass gasification: Comparison of thermodynamic efficiency," Energy, vol. 32, pp. 1248-1259.

8. R. Karamarkovic and V. Karamarkovic, "Energy and exergy analysis of biomass gasification at different temperatures," Energy, vol. 35, pp. 537-549, 2, 2010.

9. V. S. Stepanov, "Chemical energies and exergies of fuels," Energy, vol. 20, pp. 235-242. 10. S. Sokhansanj, Oak Ridge National Laboratory, September 2011,

http://cta.ornl.gov/bedb/appendix_a/The_Effect_of_Moisture_on_Heating_Values.pdf 11. Rakopoulos CD. Giakoumis EG. Second-law analsis applied to internal combustion engines

operation. Progress in Energy and Combustion Science 32 (2006) 2-47 12. Dincer and A. Z. Sahin, "A new model for thermodynamic analysis of a drying process," Int. J.

Heat Mass Transfer, vol. 47, pp. 645-652. 13. Dincer, "Exergy as a potential tool for sustainable drying systems," Sustainable Cities and

Society, vol. 1, pp. 91-96. 14. C. Coskun, M. Bayraktar, Z. Oktay and I. Dincer, "Energy and exergy analyses of an industrial

wood chips drying process," International Journal of Low-Carbon Technologies, vol. 4, pp. 224-229.

15. J. Reeb. Drying wood. 2012(August/13), pp. 8. Available: http://www.ca.uky.edu/agc/pubs/for/for55/for55.pdf;.

16. W. R. Dunbar and N. Lior, "Understanding combustion irreversibility," ASME, vol. Industrial and Environmental Applications, pp. 81- 90, 1991.

17. S. K. Som and A. Datta, "Thermodynamic irreversibilities and exergy balance in combustion processes," Progress in Energy and Combustion Science, vol. 34, pp. 351-376.

18. "NIST Chemistry WebBook," vol. 2012.



Page | 283

SIMPLIFIED PERFORMANCE EXERGY ANALYSIS TOOLS FOR THERMAL POWER PLANTS

J. Taillon a, R. E. Blanchard b

a ANDRITZ, Carbona Inc. Tammasaarenkatu 1,

FI-00180, Helsinki, Finland Email: [email protected]

b Loughborough University

School of Electronic, Electrical and Systems Engineering LE11 3TU, Leicestershire, UK

Email:[email protected]

Abstract

There is an unequivocal support for exergy analysis in thermodynamics literature. However, the industry at large relies on energy conservation principles. Exergy barriers are its complexity and seemingly futility when power, heat and fuel are multiplied by their costs rates. This paper demonstrates the weakness of the current industrial approach and introduces novel tools which will facilitate the use of exergy. A closed system steady state power plant model is introduced. New exergy equations are derived with results graphed against parametric curves. Figure 4 combines overall, electrical and thermal exergy efficiencies. Figure 5 splits thermal exergy efficiency into its heat quality (2nd law) and thermal losses (1st law). Figure 6 helps distinguish between superior technical efficiency and advantageous costs rates. The model was tested on 21 existing and design phased thermal power plants based on biomass gasification and combustion. The main results are; (1) a biomass combustion plant, ranked 1st in overall energy efficiency slipped to 16th position in exergetic terms demonstrating that energy efficiency may give erroneous results. (2) condensing power plants can be more efficient than CHP plants, a fact undetected by the current approach (3) low efficiency district heating plants with low quality heat hot water could be more profitable than plants with higher quality heat low pressure steam (4) the figures provide; easy to understand, true and complete measure of performance for power plants.

Keywords: Exergy, Energy, Efficiency, Biomass, Cogeneration

1. Introduction For thermal power plant performance analysis, the current approach by the industry (i.e.

companies, institutions and governments) relies almost exclusively on energy

conservation principles as stated by the 1st law of thermodynamics. This approach is

flawed mainly for two reasons; (a) heat and fuel input energy terms cannot be added to

power and (b) irreversibilities are ignored.

The correct approach must involve an exergy analysis based on the 1st and 2nd law of

thermodynamics. There is a consistent and unequivocal support for exergy analysis in

thermodynamic literature [1][2][3][4] but, as of today, it has not been embraced

“wholeheartedly by the industry” [1]. Exergy barriers are its complexity and seemingly

futility when energy parameters (heat, power and fuel input) are multiplied by their costs

rates. As shown in the operating profitability equations 1 and 2, the costs rates neutralize

these energy parameters into monetary values.

∗€

∗€

∗ €

(1)

€ € € (2)



Page | 284

Units, List of symbols Symbols Subscripts, acronyms and abbreviations

A availability c cold C carbon (%) db dry basis

chemical exergy el electrical fuel input rate (energy) f fuel

′ fuel input rate (exergy) fg saturation ′ exergy destruction rate h hot ′ exergy destruction (total) rate L loss

H hydrogen (%) op operating h enthalpy p product

LHV lower heating value s supplied HHV higher heating value th thermal

mass flow rate tot total N nitrogen (%) O oxygen P price

heat (energy) rate ′ heat (exergy) rate Greek symbols

S sulfur energy efficiency s entropy π operating profit T temperature (K) π’op specific operating profit, exergy based W Work exergy efficiency w humidity ratio ′ exergetic efficiency

As such, these equations are correct and form the basis of any typical feasibility study.

However, some elements might be missing or difficult to recognize from these studies. A

plant may be profitable because it is either technically efficient or has advantageous costs

rates. The main objective of this paper is to provide simple tools based on exergy which

will standardize thermal plant performance analysis on a technical and economical basis.

As a reminder, exergy represents the maximum useful work a system can achieve within

a reference environment (250C, 1atm). The concept utilizes the 1st and 2nd laws and is

particularly suitable for the study of thermo-chemical systems. The 1st law refers to the

conservation of energy while the 2nd law uses the entropy concept to quantify energy that

is not available to a system, referred as irreversibilities.

2. Methodology This paper develops a mathematical model of a closed system steady state power plant.

Exergy equations are derived and results are graphed against parametric curves. The

model is tested on 21 existing and design phased power plants based on biomass

gasification and combustion as well as a natural gas internal combustion engine power

plant as listed in table 1. The key features of the plants are:

- BGGE (Biomass Gasification based on Gas Engines): These plants may include a

dryer (fuel is wood chips, C) or not (fuel is pellets, P). The product gas from the

BFB (bubbling fluidized bed) gasifier feeds gas engines with recovered heat used

for district heating.



Page | 285

- COMB (Combustion steam cycle plant): These plants use the conventional

Rankine cycle built with the latest technologies in boilers and steam turbines.

- IGCC (Integrated Gasification Combined Cycle): These plants include a dryer

which feeds dried biomass to a pressurized BFB gasifier. The product gas from the

gasifier is fed to a combined cycle equipped with a gas turbine, HRSG (heat

recovery steam generator) and steam turbine. The plant works in condensed mode

(CO) with no useful heat or in back pressure mode (BP) where the useful heat is

either district heating hot water or low pressure steam.

- ICE NG (Internal combustion engine based on natural gas): The complete

description of this plant is found in a paper by Badami and Mura. [5]

To reproduce similar graphs, the basic input data are listed in table 2. The primary focus

of this paper is to explain the model and graphs. Reference plants are only used as a

mean to support the model with real data. Plants are unidentified so to not disclose any

proprietary information and only a few specific technical issues are explained in more

details.

Table 1: List of plants by technologies and fuel input

No. Technology Fuel No. Technology Fuel No. Technology Fuel No. Technology Fuel

1 IGCC 1 Wood chips

6 IGCC 5 CO Bagasse 11

IGCC 7a BP Bagasse 16 BGGE2C chips

2 IGCC 2 Wood chips

7 IGCC 6 CO Bagasse 12 IGCC 7b BP Bagasse 17 BGGE 3 P pellets

3 IGCC 3 Wood chips

8 IGCC 7 CO Bagasse 13 BGGE1P pellets 18 COMBI bio

4 IGCC 4 Wood chips

9 IGCC 8 CO Bagasse 14 BGGE 2 P pellets 19 COMB 2 bio

5 ICE NG NG 10 ICE 9 CO Bagasse 15 BGGE1C chips 20 BGGE4P pellets

21 COMB 3 Wood chips

Table 2: Basic required input data

Input data (essential)

Fuel input data Ultimate analysis, moisture, flow rate, energy content

Net power

Heat output (district heating) Temperature in / out, Flow rate

Heat output (steam) Flow rate, Temperature, Pressure

Costs rates Fuel, Heat and Electricity

Exhaust flue gas Temperature

3. Weakness of current approach The cogeneration energy based efficiency is given by;

(3)



Page | 286

As explained by Kanoglu et al. [4], “work and heat have the same units but are

fundamentally difficult to add because they are different, with work being more valuable

than heat”. Numerous publications [6][7][8][9] used this ratio wrongly as a mean to

compare power plant efficiencies. Another common approach is simply to omit the heat

component and use the electrical energy efficiency to gauge cogeneration power plants

performance. The above references point to well-known institutions (IEA, EC, COGEN)

and indicate that this situation affects the entire industry. To propose any changes to such

a widespread practice, might very well be a daunting task. As explained next, two

fundamental steps must be done to correctly convert heat output and fuel energy content

into work equivalent components.

3.1. Conversion of heat and fuel input into equivalent work components

a. Heat must be put on the same basis as work as shown in figure 1. As the fuel input

is, in itself, a heat component, it is converted into work by going through a real

heat engine (e.g. turbines, engines). Similarly, output heat (Qp and QL) must be

converted to work equivalent basis by going through a perfect heat engine. Such

ideal engines, which do not exist in real life, must satisfy four conditions; i)

frictionless, ii) adiabatic, iii) cyclic and iv) heat source and sink are reservoirs. In

doing so, the only irreversibilities arise from the temperature difference. This

concept takes its origin in the well-known Carnot engine.

Figure 1; Conversion of product heats to an equivalent “work component”.

b. For the fuel energy content, exergy values and formulas are found in the literature.

For biomass, the authors propose a recently published formula [10][11];

1812.5 295.606 587.354 17.506 17.735 95.615 31.8 (4)

Ein

Wcycle= Qp ‐ Qc

Heat Source

Cold Sink

Q’p= Qp ‐ Qc

Hot Reservoir

Cold Reservoir

Qp

+

perfect heat engine

real heat engine

QL + Qp Qc



Page | 287

1 (5)

From the above point a) and b), the exergy heat component and fuel exergy content

are plugged in the exergy rationale efficiency equation given by;

(6)

3.2. Include irreversibilities

The second step is to evaluate irreversibilities. As irreversibilities and heat losses to the

environment accumulate, exergy is destroyed. Spontaneous thermo-chemical reactions,

mechanical friction and thermal transfer between finite temperatures are among others,

important sources of irreversibilities in power plants.

Spontaneity makes chemical reactions irreversible [12][13]. Biomass oxidation reaction

destroys 50% of exergy input in either gasification or combustion power plants whereas

natural gas is reported to destroy 30% [11][13][14]. Mechanical friction is unavoidable and

takes the form of heat losses to the environment. Irreversibilities from thermal transfer

through finite temperature occur because warm bodies cool off. To keep constant thermal

transfer, extra energy must be added so to keep the same temperature. In other words,

the heat source is not a reservoir. The above irreversibilities reduce available energy and,

aside from mechanical friction, are ignored by the sole use of energy conservation

principle.

3.3. Weakness 1; Overall energy versus exergy efficiencies

Figure 2 shows weaknesses and potential sources of errors using energy based

efficiencies as expressed by equation 3. Three examples are highlighted;

(A) In terms of overall energy efficiency (horizontal axis), the most efficient plant is COMB

1 at 88.9% but in exergy terms, this plant drops to 16th position with an efficiency of

29.1%, as shown by the arrows. This is explained by its relative large heat output and as it

is not converted, it falsely overestimates its energy efficiency. (B) The most exergy

efficient plant is the ICE NG power plant based on natural gas (39.7%). This makes sense

because chemical exergy destructions are significantly lower with fossil fuels.

Interestingly, this plant ranked 9th in the energy based efficiency (80.1%). (C) The figure

shows also that condensing power plant (IGCC 3, 5CO, 6CO and 7CO) can be more

efficient than cogeneration power plants (COMB 1) regardless of plant capacity (fuel

energy input) as COMB 1 (76MJ/s) is a larger plant than IGCC 3 (45MJ/s) or 7CO

(48MJ/s).

With the conventional approach, these weaknesses cannot be detected and may become

more prominent as different technologies and fuels are introduced e.g. combustion,

gasification, torrefaction, co-firing, etc. A more diversified pattern of efficiencies emerges,

which modifies the conventional order when both power and heat are correctly combined,

affecting the reliability of the conventional approach.



Page | 288

Figure 2; Electrical versus total efficiency based on energy and exergy

3.4. Weakness 2; Electrical energy versus exergy efficiencies

On an exergy basis, electrical efficiencies are lower (figure 2, vertical axis) but rankings

are somewhat similar aside from the following two plants; IGCC 7CO and IGCC 3. On an

energy basis, IGCC 3 is superior at 39.7% while on an exergy basis, it is inferior (31.5%)

to IGCC 7CO. This is explained by the fuel input energy to exergy ratio given by;

1

1 (7)

Table 3: Electrical efficiencies comparison between IGCC 3 and IGCC 7 CO.

IGCC 3 IGCC 7 CO

wood chips, 45%-w bagasse, 10%-w

′ ′ η ′ η 0.397 0.794 0.385 0.838

0.315 0.323

From table 3, the reason for the reversed ranking is due to high moisture of the IGCC3.

This plant feeds on 45%-w moisture wood chips as opposed to 10%-w bagasse for IGCC

7CO. At high moisture content, LHVAR decreases more rapidly than the echAR.

A potential weakness of the current approach is discussed here with the lower heating

value (LHVdb) introduced in equation 7. The fuel input energy to exergy ratio is usually

below 1 for biomass. By using LHVdb, as opposed to HHVdb, one assumes that all water



Page | 289

formed by combustion is vapor and the latent heat of vaporization in the fuel is not

recovered when the water condenses back [13][15]. Since exergy uses 250C as reference

where water is a liquid, the latent heat is considered as available energy. This explains

why specific exergy is higher than LHVdb. Since latent heat is usually not recovered, the

exergic value is higher than what it should be. A similar situation arises when higher

heating values (HHV) are used. In all cases, recovered or not, the exergy destruction will

correctly account for the latent heat as heat loss to the environment (typical) or useful heat

(e.g. by using flue gas condensers). Energy conservation principles can also do the

proper accounting but does not convert this heat, loss or useful, into a work equivalent

component (section 4.1).

This section highlighted three cases which showed that exergy and energy efficiencies

rankings are clearly different. The authors argue that since;

Exergy is based on the 1st and 2nd laws of thermodynamics (unlike the

conventional industrial approach which uses the 1st law onlye),

Exergy has been overwhelmingly approved in the literature for its correctness,

Exergy destructions are correctly accounted for (unlike the conventional method

which assumes only thermal losses),

It must be concluded that energy efficiencies (equation 3); (a) incorrectly combines

power and heat, (b) may lead to erroneous conclusions and (c) should be replaced

by cogeneration exergy rational efficiencies (equation 6).

4. Exergy based power plant model

4.1. Mathematical model

Figure 3 represents a closed system steady state power plant model. The plant is supplied by fuel input and useful outputs are power and heat . The plant is

divided by an imaginary line where energy is transferred from box 1 to box 2. Each

box loses heat to the environment , . Exergy is destroyed in the sub-processes of

box 1 and 2 , .

e The 2nd law is seldom used in the conventional approach except for isentropic efficiencies, Mollier diagram etc. and not used in mass and energy balances.

1 2



Page | 290

Figure 3: Closed system steady state power plant model

Mass, energy and exergy balance equations for box 1 and 2 are;

1 (8)

1 (9)

W 2 (10)

2 (11)

Assuming all sub-processes are accounted for, symbolized by moving the vertical line to

the far right, then all exergy destructions are accounted for by box 1 and expressed by;

0 → (12)

As for heat losses to the environment, the vast portion comes out as flue gas through the

stack. As such, it would be fair to express total heat losses as such;

0 → (13)

To clarify further exergy theory [13], equation 14 differentiates the relationship between

overall exergy destruction in the form of heat losses and irreversibilities .

(14)

Combining the above equations gives;

1 (15)

1 (16)

2 (17)

2 (18)

Combining equation 16 and 18 gives;

(19)

Assuming that the average heat loss boundary temperature is closest to the given

exhaust flue gas temperature gives the following heat loss exergy equation;

1 1 (20)

Similarly, the equation for the supplied heat exergy equation is given by;

1 1 (21)

Heat output exergy values for steam and district heating are given by;

, , (22)

1 (23)



Page | 291

where;

(24)

At this point, all variable can be calculated from given input data aside from the heat

supplied temperature discussed next.

4.2. Average source temperature (Ts)

Combining equation 17 with 21, the average heat supplied temperature gives;

∆ ∆ (25)

The value of gives values slightly above exhaust flue gas temperature which reinforces

the model because box 1, from which originates from, is set at the frontier of box 2. In

addition, the far right term of equation 25 link to entropy values. The authors have not

pursued further the possible meaning, if any, of such equation.

5. Simplified performance exergy analysis tools

5.1. Overall exergy efficiency

Figure 4 combines electrical, thermal and total exergy efficiencies, all in a single chart.

Using equations from section 5.1, the parametric lines represent thermal exergy

efficiencies given by;

1 (26)

For biomass power plants, IGCC plants exhibit the largest electrical efficiencies because

they are based on a combined cycle. Condensing power plants are shown on the 0%

thermal exergy efficiency line. The graph shows that most combined cycle plants running

in condensed mode are more efficient than single cycle plants (BGGE, COMB). With

IGCC 5 and 6, the shift from condensed (CO) to back pressure (BP) increases both

thermal and overall exergy efficiencies but reduces electrical exergy efficiency. The back

pressure cases obtain the highest thermal exergy efficiencies because their heat output is

low pressure steam as opposed to district heating hot water, as explained in next section.



Page | 292

Figure 4: Electrical, thermal and overall exergy efficiencies.

5.2. Thermal exergy efficiency

Figure 5 splits the thermal exergy efficiency into its two fundamental components; an

exergetic efficiency and thermal energy efficiency , by combining equations 14,

15, and 16;

1 (27)

The 1st term represents the quality of the product heat. A smaller gap between source and

product temperatures improves quality and efficiency. The 2nd term represents the

proportion of irreversibilities without thermal losses. Together, these two terms represent

an exergetic efficiency as derived from the 2nd law and defined by a new factor . The

3rd term represents the conventional thermal energy efficiency, as defined by the 1st law.

Thus equation 27 can be simplified by;

(28)

Figure 5 plots these two terms and the parametric curves represent the same thermal

exergy efficiency as in figure 4.

For all plants, the heat output is district heating hot water except for three plants; IGCC 2,

IGCC 5BP and IGCC 6BP which deliver low pressure steam. Steam displays the highest

exergetic efficiency (or quality) because it has the lowest temperature gap between the

product heat and the heat source. The exergetic efficiency of IGCC 7aBP is low

(11.8%) because its average district heating temperature is 740C. Correspondingly, by

raising the average temperature to 980C for the same plant, the efficiency improves as

shown by plant IGCC 7bBP (16.4%). IGCC 1 and IGCC 4 show low thermal energy

efficiency because the plants could not sell all of their heat output. The use of pellets

improves thermal efficiency because drying energy is excluded. From the gap between



Page | 293

BGGE 1P and 1C, the energy efficiency benefit is 8% (horizontal axis) which corresponds

to a 1.6% increase in exergy terms (parametric curves).

Figure 5; Exergy thermal efficiency – exergetic factor vs. thermal energy efficiency

5.3. Profitability versus exergy efficiency

An exercise can be done to derive what would be the price of heat if it would follow the

exergy concept. That is, the revenue from the heat energy output should equal the heat

exergy output multiplied by the electricity price as shown below;

1 (35)

1 (36)

Assuming district heating temperature (hot water) of 900C and electricity price of 112

€/MWh, the equivalent 2nd law district heating price gives 20 €/MWh. In fact, district

heating prices often vary around 35 – 45 €/MWh. For low pressure steam at 1500C, the

equivalent 2nd law price gives 33 €/MWh and correspond more closely to the real costs of

steam in a process plant. Therefore, lower efficiency power plant supplying district heating

will benefit because their heat price will increase profit by a higher margin than what would

be dictated by the 2nd law, as observed in the two graphs of figure 7. In this figure, the 1st

graph is a reproduction of figure 5 and the 2nd graph plots specific operating profit in

exergy terms given by equation 37 versus the thermal exergy efficiency .

€

€

€ (37)



Page | 294

Figure 6; Graph #1 and #2 help differentiate between technical superiority and advantageous

costs rates.

Given the following input prices; electricity (112€/MWh), heat district heating (40€/MWh),

heat low pressure steam (35€/MWh), fuel chips (15€/MWh) and fuel pellets (25€/MWh),

four zones are identified;

Zone 1, in graph #1, (low , low ) generates low heat revenues so that profits are at

their lowest as shown in graph #2.

Zone 2, in graph #1, (low , medium ) earns the highest profit (graph #2) because

district heating hot water is highly priced but not in line with its low quality.

Zone 3, in graph #1, (high , medium ) earns less profit (graph #2) because the

higher quality heat, low pressure steam, is priced less than district heating hot water in

this example.

By setting the price of heat in proportion to its quality, as described in equation 36, it

flattens the profitability curve in graph #2. Zone 2 shifts down to zone 4.

6. Conclusions

This paper showed the weaknesses of the traditional approach to power plant

performance analysis which relies solely on the principle of conservation of energy. Three

examples were highlighted; (1) A combustion plant, which ranked 1st on an energy

efficiency basis because of its high unconverted thermal energy component, slipped to

16th position based on an exergy basis. (2) In exergy terms, the most efficient power plant

is based on natural gas simply because fossil fuels have much lower chemical exergy

destruction but yet, ranked 9th on an energy efficiency basis. (3) Power plants running in

condensing mode can be more efficient than cogeneration plants; a fact undetected by the

current approach and which could influence public support policies.

The principal aim of this paper was to simplify power plant performance analysis by

introducing a novel approach to industry and exergy theory alike. The approach was the

development of a closed system steady state power plant mathematical model based on

energy and exergy balances and led to the formulation of efficiency equations and the

Graph #2 Graph #1



Page | 295

following figures. It is emphasized here that figures 4 and 5 are new to the field of exergy.

Figure 4; Combines electrical, thermal and total exergy efficiencies, all in a single graph.

Figure 5; Splits the thermal exergy efficiency into a thermal exergetic factor and thermal

energy losses. The 1st term relates to the quality of heat as governed by 2nd

law while the 2nd term relates to the well-known thermal energy efficiency (1st

law). The product of these two terms represents the same thermal exergy

efficiency displayed as parametric curves in figure 4.

Exergy analysis may seem obsolete when plant performance is based on a profitability

basis as all terms are on a monetary basis. Thus, an additional graph was used to help

distinguish between superior technical efficiency and advantageous costs rates. This is

important because costs rates tend to be more volatile than technical efficiencies.

Figure 6 showed that low efficiency district heating plants with low quality heat (hot water)

could be more profitable than plants with higher quality heat (low pressure steam) given

certain input costs rates.

This paper proved that heat quality cannot be ignored and thermal power plant

performance analysis must be based on exergy.

To make use of these tools in the industry will be a daunting challenge. However, as

different technologies, processes and fuels are introduced, a more diversified pattern of

efficiencies emerge, affecting the reliability of the conventional approach. A key benefit of

these graphs based on exergy is to provide an easy to understand, true and complete

measure of performance for thermal power plants.

References [1]. Dincer and M. A. Rosen, EXERGY: “Energy, Environment and Sustainable Development.”

Burlington: Elsevier, 2007. [2]. M. Prins J., K. Ptasinski J. and F. Janssen J.J.G., "Thermodynamics of gas-char reactions:

first and second law analysis," Chemical Engineering Science, vol. 58, pp. 1003-1011. [3]. R. Saidur, G. BoroumandJazi, S. Mekhilef and H. A. Mohammed, "A review on exergy analysis

of biomass based fuels," Renewable & Sustainable Energy Reviews, vol. 16, pp. 1217-1222. [4]. M. Kanoglu, I. Dincer and M. A. Rosen, "Understanding energy and exergy efficiencies for

improved energy management in power plants," Energy Policy, vol. 35, pp. 3967-3978, 7, 2007.

[5]. M. Badami and M. Mura, "Exergetic analysis of an innovative small scale combined cycle cogeneration system," Energy, vol. 35, pp. 2535-2543.

[6]. Bauen, G. Berndes, M. Junginger, M. Londo and F. Vuille, "Bioenergy - A sustainable and reliable energy source. A review of status and prospects." IEA Bioenergy, Tech. Rep. IEA Bioenergy: ExCo: 2009:06, 2009.

[7]. European Commission, "Proposal for a Directive of the European Parliament and the Council on Energy Efficiency and repealing Directives 2004/8/EC and 2006/32/EC." vol. 2011/0172 (COD), 2011.

[8]. COGEN Europe, "What is Cogeneration?" vol. 2013. [9]. G. Davies and P. Wood, "The potential and costs of district heating networks," Pöyry Energy

(Oxford) Ltd and Faber Maunsell, 2009. [10]. G. H. Song, L. H. Shen and J. Xiao, "Estimating Specific Chemical Exergy of Biomass from

Basic Analysis Data," Ind Eng Chem Res, vol. 50, pp. 9758-9766.



Page | 296

[11]. J. Taillon, "Comparative exergy analysis of biomass cogeneration systems based on gasification and combustion." (M.Sc. Thesis, CREST, School of Electronic, Electrical and Systems Engineering, Loughborough University, UK, 2012).

[12]. S. Khan, "Gibbs Free Energy and Spontaneity," 2013. [13]. M. J. Moran and H. N. Shapiro, Fundamentals of Engineering Thermodynamics. Chichester:

John Wiley & Sons, Ltd, 2006. [14]. W. R. Dunbar and N. Lior, "Understanding combustion irreversibility," Asme, vol. Industrial

and Environmental Applications, pp. 81- 90, 1991. [15]. Wikipedia, "Heat of combustion," 2013.



Page | 297

MODELICA-BASED MODELING AND EXERGY ANALYSIS OF A CENTRAL HEATING SYSTEM

R. Sangi a, R. Streblow a & D. Müller a

a Institute for Energy Efficient Buildings and Indoor Climate E.ON Energy Research Center

RWTH Aachen University Mathieustr. 10, 52074 Aachen Germany

Email: [email protected],; [email protected]; [email protected]

Abstract

In this study, dynamic modeling and simulation of a conventional central heating system has been

performed. The system contains a boiler that supplies heat for both a hot water storage tank for

domestic use and a room with two radiators. Each radiator has a thermostatic valve that regulates

the mass flow rate to maintain the room temperature at the set point.

The analysis has been carried out by simulation of the system using the object-oriented programing

language Modelica. Dymola, which is a multi-domain modeling and simulation tool, has been used

as simulation environment. The Modelica Standard Library 3.2 and the Modelica libraries for

building simulation developed at the Institute for Energy Efficient Buildings and Indoor Climate,

E.ON Energy Research Center, RWTH University such as BaseLib, Database, Building and HVAC

components libraries have been applied to simulate the hydraulic and thermal behavior of the

system. Detailed sub-models for the hydraulic system have been developed and coupled with a

room model to make a general model for simulating the performance of the whole system.

In the next step, the system has been analyzed form exergy point of view and the simulation results

from the dynamic model have been presented in an exergetic framework.

Keywords: Central heating system; Dynamic modeling and simulation; Dymola; Exergy analysis;

Modelica

List of Symbols

e : Specific flow exergy (J/kg) Greeks

E : Exergy (J) : Efficiency

h : Specific enthalpy (J/kg) Subscripts

m : Mass flow (kg) 0 : Dead state

Q : Heat transfer (J) cv : Control volume

s : Specific entropy (J/kgK) D : Destruction

t : Time (s) e : Outlet

T : Ambient temperature (K) i : Inlet

V : System volume (m³) j : Location on the boundary

W : Work (J)



Page | 298

1. Introduction

The amount of energy used for heating and cooling in the building sector is about one

third of the total energy consumed in the world. The finiteness of natural energy resources

on the one hand, and the ever-increasing demand for energy in the world on the other

hand, necessitate the development of systematic approaches for improving the efficiency

of building energy systems as well as minimizing the usage of primary energy resources

and the damaging impacts and harmful effects on the environment.

Dynamic modeling and simulation of heating systems, which are one of the most energy

consuming parts of buildings, provides not only a convenient and low-cost tool for

evaluating the performance of heating systems, but also makes engineers capable of

developing new control and optimization strategies for such systems.

In this research, a model for a conventional hydronic heating system with thermostatic

radiator valve has been developed and the dynamic performance of the system has been

simulated. To evaluate the performance of a heating system, a model of the entire energy

chain from generation to distribution is needed. Therefore, in this work, detailed sub-

models for the hydraulic system have been developed and coupled with the model of a

room to make a general model for simulating the performance of the whole system.

In the next step, the system has been exergetically analyzed and the simulation results

from the dynamic model are presented in an exergetic framework.

The analysis has been carried out by simulation of the system using the equation-based,

object-oriented programing language Modelica. Models written in the Modelica language

cannot be executed directly, and a simulation environment is needed to translate a

Modelica model into an executable program. In this study, the models have been

developed and simulated in the Modelica modeling and simulation environment Dymola

version 2013 FD01.

2. System description

The simulated space heating system is a model of a conventional central boiler heating

system of a room with two radiators which are served by distribution pipes. The room has

two external walls, one door and two windows. The room dimensions are 6.5 5.62.65 and the dimensions of the door and windows are 2 0.9 and 1.2 1.7 ,

respectively.

The hydraulic system contains a boiler that supplies heat for both a hot water storage tank

for domestic use and two radiators. The heated water is circulated through pipes by a

central circulation pump which is located at the boiler outlet. The temperature of the boiler

is controlled based on the outside temperature. Each radiator has a thermostatic valve

that regulates the mass flow rate to maintain the room temperature at the set point. During

the heating period, which is from 1st to 31th January in this simulation, the room set

temperature is set to 21° and for the day time period (06:00 AM – 08:00 PM). During the



Page | 299

night, the setback temperature is 19° and the room temperature decreases depend on

the outside temperature. The nominal power per meter of the radiators is 1229 / .

The boiler is also supplies heat for a hot water storage tank for domestic use. The heated

water passes through the heating coils of the storage tank where it gives up its energy.

Fig.1 provides a schematic illustration of the system.

Fig. 1: Schematic illustration of the system

3. Modeling

The Modelica Standard Library 3.2 and the Modelica libraries for building simulation

developed at the Institute for Energy Efficient Buildings and Indoor Climate, E.ON Energy

Research Center, RWTH University such as BaseLib, Database, Building and HVAC

components libraries have been applied to simulate the hydraulic and thermal behavior of

the system. The following paragraphs explain the developed sub-models of the general

system in detail.

3.1 Room Model

The components airload, infiltration, wall, window, door, thermal and radiation interfaces,

and solar radiation and wind speed ports have been used to model the room. The airload

model represents a heat capacity consisting of air described by its volume, density and

specific heat capacity. The infiltration model describes heat and mass transport by

infiltration. The thermal and radiation connectors equate the temperatures and balance

the convective and radiative heat flows, respectively.

The boundary conditions of internal walls, floor and ceiling have been considered as

adiabatic. The view of the room model in the graphical model editor of Dymola has been

shown in Fig.2.



Page | 300

Fig. 2: Graphical user interface of the room model in Dymola

3.2 Weather model

To simulate the dynamic performance of the system under real climate conditions, the

developed model has been coupled with a weather model that is capable of interpreting

the weather data and integrating solar radiation, wind velocity, and ambient temperature

into the simulation. In this simulation, the weather data of Aachen has been selected. The

data has been provided by the Federal Institute for Research on Building, Urban Affairs

and Spatial Development (BBSR) [1]. Fig.3 demonstrates the view of the weather model

in Dymola.

Fig. 3: Graphical user interface of the weather model in Dymola



Page | 301

3.3 Hydraulic system model

The components that have been used to model the hydraulic system are boiler, expansion

tank, pipe, valve, t-joint, radiator, pump, hydraulic resistance, storage tank and

thermostatic valve. Sensors for measuring temperature, mass flow rate, enthalpy and

entropy have also been used.

The sensors monitor the thermodynamic properties of the fluid passing their ports while

they do not influence the fluid.

The hydraulic resistance component defines a pressure drop based on the given loss

factor, mass flow rate and diameter of pipe.

The Radiator model represents a heating device. Heat energy taken from the hot water

flow through the device is being emitted via convective and radiative energy transport

connectors. The ratio of convective and radiative energy flows depends on the type of the

heating device.

The boiler is a table-based model whose set point was computed as a function of the

outside temperature, using a heating curve.

The thermostatic valve model with a switch for day/night operation determines the mass

flow rate of the radiator to keep the room temperature at the set point.

The model of the hot water storage tank has heating coils which transfer heat to domestic

hot water (DHW). Heat demand of the storage tank is determined by a daily domestic hot

water profile.

The pump model uses a table for the flow characteristic at which head has been defined

as a function of volume flow rate. All mass flow rates are computed by solving the

pressure distribution in the piping network, which depends on the pump curve and the flow

frictions.

3.4 General model

The hydraulic system model has been coupled with the model of the room in order to

make the general model of the system. The thermal connectors that connect the room

model to the radiators equate the temperatures and balance the convective and radiative

heat flows between radiator and rooms. Also the solar radiation interfaces, thermal

connections and wind ports of the room model have been connected to the weather

model. The view of the general model in Dymola has been illustrated in Fig. 4.



Page | 302

Fig. 4: Graphical user interface of the general model in Dymola

3.5 Exergy meter

In this study, a new component entitled exergy meter has been developed. The exergy meter

provides the possibility of component selection for the user and it is capable of measuring exergy

destruction rate and exergy efficiency of the selected component.

4. Exergy analysis

Exergy is defined as the maximum available work that can be extracted from a system

during a process that brings the system into equilibrium with its environment. Exergy

analysis, which is a powerful thermodynamic technique for optimizing and assessing the

efficiency of complex energy systems, helps to identify which components of the system

are responsible for irreversibility.

The general exergy balance for a control volume can be written as

∑ ∑ ∑ (1)

where the specific flow exergy is evaluated as

(2)

The exergy efficiency is defined as the ratio of the desired exergy output to the exergy

used [2]:

(3)

Details for developing the exergy balance equations for various components have been

provided in references [3,4]. The performance of a boiler system can be analyzed



Page | 303

exergetically based on two methods. The first way is by writing the exergy balance for the

boiler, and the second is by calculating the entropy generation rate in each physical and

chemical process separately. In this study, the second method has been selected and the

boiler has been assessed by considering a thermodynamic average temperature [3,5].

The exergy transfer rates of radiators have been also evaluated using the logarithmic

mean of the inlet and out temperatures.

5. Results and discussion

The developed model has been simulated for a one-month heating period, which is from

1st to 31th January. The results of the last two days of January have been graphically

presented in the followings. Fig. 5 illustrates the flow and return temperatures of the

radiators, respectively. The flow temperature of each radiator depends on its distance

from the center. The return temperature is calculated based on the heat demand of the

room.

Fig. 5: Flow and return temperatures of the radiators

The mass flow rates of the radiators have been shown in Fig. 6.

Fig. 6: Mass flow rates of the radiators



Page | 304

The powers of the radiators required to maintain the room temperature at the set point

have been displayed in Fig. 7.

Fig. 7: Powers of the radiators

It should be mentioned that since the water model used in this study is the medium model

with linear dependency of enthalpy from temperature, a fully dynamic exergy analysis

leads to negative values for exergy destruction at the times that the temperature

difference is not significant. The general model was also simulated with the more complex

models for water available in the Modelica Standard Library, but the results were not

satisfactory, as they make the model unstable. Therefore, the exergy analysis has been

performed based on a point at the outlet of the boiler that has the average value of a one-

month simulation with the simple water model. The thermodynamic properties for water at

the inlet and outlet of the analyzed components have been presented in Table 1.

Table 1: Process data

Component

/

°

°

/

/

/

/

Boiler 0.159 46 66 1 1 0.6516998 0.9058596 192.6950 276.3266

Expansion tank 0.159 66 65.5 1 1 0.9058596 0.8996844 276.3266 274.2338

Pump 0.159 65.4 65.4 1 1.39 0.8984483 0.8984260 273.8153 273.8478

Storage Tank 0.1 65.4 41.9 1.2 1.1 0.8984369 0.5982004 273.8319 175.7295

DHW 0.06 20 60 1 1 0.2964829 0.8311736 84.01181 251.2216

Main flow pipe 0.059 65.4 64.3 1.2 1.19 0.8984369 0.8848175 273.8319 269.2275

Main return pipe 0.059 57.6 56.9 1.05 1.04 0.8009330 0.7920741 241.1883 238.2603

Thermostatic valve1 0.03 64 64 1.18 1.11 0.8810961 0.8811001 267.9712 267.9654

Radiator 1 0.03 64 59.2 1.11 1.1 0.8811001 0.8211125 267.9654 247.8839

Thermostatic valve2 0.029 63.7 63.7 1.18 1.11 0.8773710 0.8773749 266.7159 266.7101

Radiator 2 0.029 63.7 57.5 1.11 1.1 0.8773749 0.799666 266.7101 240.7744



Page | 305

Table 2. presents the exergy efficiency, exergy destruction rate and exergy destruction in

a one-month heating period evaluated for each component. It is evident from Table 2 that

the greatest irreversibility occurs in the boiler. The hot water storage tank has the second

high irreversibility. The amount of exergy destruction in radiators and pumps are not

significant in comparison with the boiler. The expansion tank has the lowest irreversibility

in the system. This irreversibility occurs due to the heat loss from expansion tank to the

room air.

Table 2: Results of the exergy analysis of the simulated system

Component / %

Boiler 2582.976936 1859.743394 45.29

Expansion tank 0.039962063 0.028772685 99.94

Pump 1.234280949 0.888682284 83.26

Storage Tank 371.1213324 267.2073593 75.15

Main flow pipe 0.048967062 0.035256284 99.90

Main return pipe 0.050031158 0.036022434 99.82

Thermostatic valve1 0.1753164 0.126227808 15.26

Radiator 1 0.016990954 0.012233487 99.98

Thermostatic valve2 0.169595617 0.122108844 15.21

Radiator 2 0.007143354 0.005143215 99.98

In general, boiler heating systems are not considered as low-exergy systems, regardless

of the type of the controller. Two key parameters should be taken into account in order to

develop a low-exergy system. Firstly, the reduction of heating load needs to be seriously

considered in design of buildings, which applies to the building’s envelope. Secondly, the

heating demand of buildings should be supplied using systems other than fossil energy

carriers. The temperature of the heat supplied by a renewable energy supply system like a

geothermal heat pump system differs slightly from the room temperature, and as a result,

low-exergy energy is consumed. To sum up, more attention should be paid by design

engineers to the quality of the supplied energy in designing new innovations of energy

systems in buildings.

6. Conclusions

In this paper, the performance of a conventional central heating system was evaluated.

Detailed sub-models for the hydraulic system were developed and coupled with a model

of room to make a general model. The developed model was simulated for a one-month



Page | 306

heating period and the simulation results from the dynamic model were used to determine

the exergy destruction and exergy efficiency of the components of the system. It was

concluded that more attention should be paid to the quality of the supplied energy in

designing buildings energy systems. Furthermore, the necessity of developing a more

accurate and stable Modelica-based model for the medium in order to carry out a fully-

dynamic analysis was recognized. This topic is the subject of future research.

References

[1]. http://www.bbsr.bund.de.

[2]. Kotas, TJ., The exergy method of thermal plant analysis. Tiptree, Essex: Anchor Brendon Ltd.; 1985.

[3]. Bejan, A., Tsatsaronis, G., Moran, M., Thermal design and optimisation. New York: Wiley; 1996.

[4]. Dincer, I., Rosen, M.A., Exergy: energy, environment and sustainable development: Elsevier; 2007.

[5]. Jiang Y.Y., Zhou S.X. Exergy Analysis of Boiler Based on the Temperature Gradient. 2010 Asia-Pacific Power and Energy Engineering Conference (APPEEC), March 2010, Chengdu, China



Page | 307

COMPARATIVE ANALYSIS OF BIOPLASTICS PRODUCTION METHODS

V.Tsamili , E.Angeli, C. Koroneos

a Interdisciplinary Programme of Postgraduate Studies “Environment and Development”

P.O. Box 15780, National Technical University of Athens,

9, Heroon Polytechneiou str., 157 73, Zographou Campus, Greece Email: [email protected]; [email protected]

Abstract The energy produced from biomass is recognized as a renewable energy source with high potential for sustainable development in the near future. Biomass already provides 14% of global primary energy production, however, much of the energy is lost due to inefficient use of unsustainable exploitation. The utilization of the total potential energy provided by biomass requires new approach using modern technologies. Biomass includes all organic material provided, either as energy crops or as organic waste. The use of the organic fraction of waste as a raw material in the production of biomaterial and bioenergy is an innovative process with environmental and economic benefit.At present and in the near future, the characteristics of agricultural and forestry byproducts makes them a major source of bioproducts. In particular, the EU woody byproducts are estimated at 94% of biomass used for energy production. In the medium term, agricultural and forestry residues, will further develop bio-energy and bio-polymer industry. Lignocellulose in the form of forestry, agricultural, and agroindustrial wastes is accumulated in large quantities every year. These materials are mainly composed of three groups of polymers, namely cellulose, hemicellulose, and lignin. Cellulose and hemicellulose are sugar rich fractions of interest for use in fermentation processes, since microorganisms may use the sugars for growth and production of value added compounds. Submerged and solid state fermentation systems have been used to produce compounds of industrial interest from lignocellulose, as an alternative for valorization of these wastes and also to solve environmental problems caused by their disposal. When submerged fermentation systems are used, a previous stage of hydrolysis for separation of the lignocellulose constituents is required. Considering the increasing demands for fuel, energy and chemical products, the need to address the problem of environmental pollution from the large volume of waste and the fact that cellulose is the most abundant organic chemical on earth, this paper attempts to address and compare techniques used in two basic and significant stages of manufacturing process of lignocellulosic based biopolymers and in particular the pre-treatment techniques of lignocellulosic biomass and the biological biomass treatment techniques. Keywords: Lignocellulosic waste; LCF Biorefinery; Pre-treatment methods; Biological treatment

1. Introduction

Shifting society’s dependence from petroleum-based to renewable biomass-based

resources is generally viewed as key to the development of a sustainable industrial

society and to the effective management of greenhouse gas emissions [1], [2], [3]. The future

production of chemical products will increasingly depend on biomass, particularly plant

biomass. The topic of processing bio-sourced materials has come to the forefront of

sustainable engineering research and, in the last 15 years, a number of critical research

reviews have been published that address the topics of lignocellulosic biomass

pretreatment [4], [5], enzymatic and microbial conversion [6], [7], as well as biomass

processing approaches [8], [9] and future biorefining perspectives. Consequently, there has

been an increase in lignocellulosic biomass processing research, focusing particularly on



Page | 308

agricultural and forestry residues, since these are low in cost, abundant, readily available

and renewable [10], [11], [12]. With the abundance of biomass waste available, the

development of new technologies that will make use of biomass for materials production

beyond biofuels represents an important opportunity to fully utilize our resources.

Development of efficient techniques to fractionate lignocellulosic biomass into its core

components will facilitate research on the production of specific biomass-derived sugars,

building block chemicals and ultimately value-added commodity chemicals while

preserving the concept of the biorefinery approach by promoting effective utilization all

feedstock fractions.

2. Pre-treatment techniques of lignocellulosic biomass

The structure of cellulose favours ordering of the polymer chains into tightly packed

arrangements that are water insoluble and resistant to depolymerisation [4]. Thus, it is

imperative that a pretreatment regime alter the structure of biomass to make the cellulose

more accessible to hydrolysis. Crystallinity, accessible specific surface area, particle size,

degree of polymerization of the cellulose, coating of cellulose by both lignin and

hemicellulose and the heterogeneous nature of biomass all have been proclaimed to

contribute to the recalcitrance of lignocellulosic materials to hydrolysis [4], [14], [15]. Over the

years a number of different technologies have been developed for pretreatment of

lignocelluloses. There is an overall consensus that the successful pretreatment should:

maximize the enzymatic convertibility

minimize loss of sugars

maximize the production of other valuable by-products, e.g. lignin

not require the addition of chemicals toxic to the enzymes or the fermenting

microorganisms

minimize the use of energy, chemicals and capital equipment

be scalable to industrial size. [16]

Pretreatment approaches can be classified into two main categories, de-lignification and

fractionation.

2.1. De-lignification pre-treatment techniques

De-lignification pretreatment include physical, chemical, physico-chemical and biological

techniques:

physical pretreatments exhibit comparatively lower performance and higher costs [17], including mechanical comminution and ultrasound [18], [19].

chemical pretreatment regimes show a high degree of selectivity for the biomass

component they degrade, but also involve relatively harsh reaction conditions,

which may not be ideal in a biorefinery scheme due to possible effects on



Page | 309

downstream biological processing. These processes include ozonolysis, dilute and

concentrated acid, alkaline, oxidative H2O2 delignification and organosolv, an

organic or aqueous organic solvent mixed with an inorganic acid catalyst [4], [19].

physico-chemical pretreatment combines chemical and physical treatment options.

Often, milder chemical conditions are used, but under more extreme operational

conditions, typically involving elevated pressures and temperatures, adding to the

cost of implementing these techniques in a biorefinery scheme. Different physico-

chemical pretreatment techniques include liquid hot water (hydrothermolysis,

aqueous or steam/aqueous, uncatalyzed solvolysis aquasolv), steam explosion

(autohydrolysis with and without chemical addition), ammonia fibre explosion

(AFEX), and CO2 explosion [4], [12]. Biological pretreatment techniques offer

advantages such as low chemical and energy use, in addition to mild operational

conditions and likely ease of integration into a consolidating bioprocessing set-up [17], [19].

2.2. Lignocellulose fractionation

Fractionation is a specific class of pretreatment where the biomass feedstock is separated

into its core components (cellulose, hemicelluloses, and lignin) such that each individual

component may be more readily processed or functionalized. The largest obstacle to be

overcome, in terms of technological and economic barriers, to make biomass biorefineries

a reality is the implementation of cost-effective methods to release soluble sugars from

lignocellulosic biomass at an industrial scale [21]. It has been argued that the conversion of

lignocellulosic biomass to higher value products also requires fractionation [22].

2.2.1 Acid based fractionation

It has been shown that cellulose solvents such as concentrated phosphoric acid can

completely dissolve cellulose fibres and disrupt the hydrogen bonds that hold together

crystalline cellulose, which in turn increase the accessibility of cellulose to cellulase

enzymes [23] presented a cellulose solvent-based lignocellulosic fractionation (CSLF)

technique using a cellulose solvent, volatile organic solvent, and water to separate

lignocelluloses at relatively low temperatures and pressures (50 0C and atmospheric

pressure). The process fractionated lignocellulosic biomass components based on the

difference in solubility of cellulose, hemicellulose, and lignin in the cellulose solvent,

organic solvent and water, respectively [23]. The cellulose solvent was also recycled in the

process as a result of the difference in volatility between the cellulose (phosphoric acid)

and the organic (acetone) solvents [21]. The aim of the CSLF technique is the

decrystallization of cellulose fibres, partial removal of lignin and hemicellulose from the

cellulose fraction and relatively modest reaction conditions. Additionally, the CSLF

approach has been reported to work independently of biomass type, which would make it

a relatively generic and powerful fractionation technique. Studies under various CSLF

conditions have demonstrated that the acid concentration is the most important factor,



Page | 310

while temperature is the least important in terms of efficiency. It should be noted,

however, that these technologies are very capital intensive.

2.2.2 Ionic liquid based fractionation

Ionic liquids (ILs) are organic salts that exist as liquids at low temperatures, often well

below 100 0C. They have tuneable physico-chemical properties, negligible (or very low)

vapour pressures, generally good thermal stability and there is nearly a limitless

combination of anions and cations that can be used to synthesize ILs [24], [25]. Recent

studies of particular interest have indicated that both cellulose and lignin can be dissolved

in a variety of ionic liquids, and, perhaps more importantly, easily regenerated from these

solutions [26], [27]. In terms of dissolution of cellulose, one of the most prominent studies

involved the treatment of cellulose with 1-butyl-3-methylimidazolium chloride ([C4mim]Cl),

where dissolving pulp cellulose was dissolved in concentrations of up to 300 g/L without

prior treatment [26], [28]. The structural form of the regenerated cellulose can be altered to

include powders, tubes, beards, fibres, and films by altering the regeneration process. The

regenerated cellulose microstructure can range from amorphous to crystalline based on

the regeneration process [29]. However, it should be noted that the regenerated cellulose

has essentially the same degree of polymerization and polydispersity as the original

cellulose [27]. Typically, regeneration of the solute is achieved by precipitation in the

presence of an anti-solvent (water or methanol) due to preferential solute- displacement.

This anti-solvent can then be

stripped from the IL (for example through flash distillation, evaporation, reverse osmosis

or salting out) and the IL recovered for reuse [29], [30]. Furthermore, 1-butyl-3

methylimidazolium chloride (BMIMCI) has been employed under microwave irradiation

and/or pressure to dissolve lignocellulosic materials [29]. 1-Ethyl-3-methylimidazolium

acetate ([Emim][CH3-COO]) has been used to dissolve lignin from lignocellulosic biomass

with minimal cellulose dissolution. It was observed that the removal of 40% of the original

lignin resulted in higher cellulose digestibility through enzyme hydrolysis (greater than

90%) [25]. Thus, ILs have the potential to be used beyond a lignocelluloses fractionation

regime for the more environmentally benign pretreatment of lignocellulosic biomass.

Preliminary studies have shown that conversion of IL-extracted cellulose to both ethanol

and lactic acid was higher than in samples pretreated by steam explosion and chemical

pretreatment technologies [27], [29]. Regenerated cellulose exhibited a significant

improvement, in terms of hydrolysis rate and glucose formation, over untreated cellulose [30]. The improvement in hydrolysis rates was attributed to the slight decrease in the

degree of polymerization of both cellulose and hemicellulose, the decrease in the

crystallinity of the cellulose, and an increase in accessibility due to lignin separation [31].

Table 1 summarizes the characteristics of the pre-treatment methods.



Page | 311

Table 1. Comparison of pretreatment methods of lignocellulosic biomass

3. Biological biomass treatment techniques in polymer production

3.1. Fermentation

Fermentation has been widely used, both in academic and industrial settings, to produce

some of the most highly sought after building block chemicals. As such its integration into

biorefinery schemes is essential in moving this technology forward. For example, succinic

acid, one of the most in demand building block chemicals outlined by the United States

Department of Energy, can be produced through fermentation [2]. Salts of succinic acid

can be derived through the fermentation of glucose, which fixes CO2 and is thus a green

technology [9], [32]. Since most microorganisms used in industrial fermentation are not

tolerant of acidic conditions, the process is neutralized. These salts then undergo more

conventional chemical processing such as separation and recovery where they are

separated from the microorganisms and then dissolved in an acid solution to form succinic

acid [32].

In nature, 1,3-propanediol is produced through the fermentation of glycerol [33]. Genencor

and DuPont have developed a low-cost route of producing 1,3-propanediol (1,3PDO) by

modifying natural routes. 1,3-Propanediol is a key building block for polypropylene

terephthalate, which is not available from petrochemical sources, in addition to being used

as fibre in the apparel and carpet industries. Itaconic acid, another DOE top 12 building

block chemical, is currently an expensive specialty product with many uses and is

industrially produced from the fermentation of carbohydrates by fungi in small quantities.

Polymerized esters (methyl, ethyl, and vinyl) are used in adhesives and coatings. Itaconic

acid is often found in emulsion paints to aid in polymer adhesion and as a hardening

agent for organosiloxanes which are used in contact lenses. In addition, due to its two

reactive carboxyl groups, itaconic acid has the potential to be incorporated into polymers.

Currently, it is being assessed as a biofriendly substitute for acrylic and methacrylic acid in

polymers and in styrene–butadiene systems [2], [9].



Page | 312

Lactic acid, which can be chemically converted to several important chemicals including

methyl lactate, lactide, and polylactic acid (a biodegradable replacement for polyethylene

terephthalates (PETs)), can be produced through fermentation. Research is being

conducted to convert lactic to acrylic acids, major commodity monomers with an

international volume demand approximately 1.6 billion kg/year in 2007 [34]. The hydroxyl

and carboxyl groups present in lactic acid make it amenable for conversion into a wide

range of value-added products.

Recent research suggests that some organisms express acrylic acid pathways, thus direct

fermentation of biomass to acrylic acid is a possibility [35]. Acrylic acid and its amide and

ester derivatives are key components in polymer manufacturing, appearing in a wide

range of products including surface coatings, textiles, detergents, and absorbent

materials. Ethylene, which can be polymerized into widely used polyethylene, has been

shown to be synthesized directly from biomass hydrolysates using naturally occurring

bacteria in soil and the surface of fruits [36]. Although the possibility of ethylene production

exists from biomass sources, essentially all production of ethylene is derived from steam

cracking in the petrochemical industry, currently resulting in a negligible contribution of

biological production of ethylene in industry. Thus, the widespread potential of

fermentation technologies to add value to lignocellulosic biomass sources make them

integral to development of biorefineries. Table 2 summarizes the positive and negative

characteristics of the fermentation process.

Table 2. Characteristics of the fermentation process



Page | 313

3.2. Direct biological conversion

3.2.1 Extraction

Value-added products have been synthesized within biological species, both plant matter

and microorganisms. In order to benefit from their production, efficient extraction

technologies need to be implemented into biorefineries early on to allow for extraction

before

further processing. For example, commodity chemicals have been directly extracted from

biomass via conventional chemical extraction techniques. One of the most notable

chemicals derived in this manner is ferulic acid, extracted in high percentage yields from

corn fibre. Ferulic acid is a chemical feedstock used in the production of fine chemicals

such as vanillin and guaiacol. DuPont has been able to extract a monomer from tulips,

A-a-methylenebutyrolactone or tulipalin A, which polymerizes in a manner similar to

methyl methacrylate and also has favourable durability and refractive index [9], [37]. This

monomer could replace some petroleum-based methacrylate monomers for products such

as mouldings. Thus, a market for value-added product extraction for specialty chemicals

exists and biorefinery designs should make the most of this opportunity. Table 3

summarizes the characteristics of extraction process.

Table 3. Characteristics of extraction process

3.2.2. Enzymatic transformation

In one of the best illustrations of direct biological conversion, polymers have been

produced completely within microbial cells, most notably the family of

polyhydroxyalkanoates (PHAs). This technology can be incorporated into a biorefinery as

a separate processing stream since the bacteria that produce PHAs have been shown to

be amenable to feeding from a wide range of carbon sources. PHAs are a family of

natural polymers, comprised of over 150 different hydroxyalkanoates, produced by a

range of bacteria (at least 75 genera) for carbon and energy storage [32], [38], [39]. PHAs are

formed as intracellular granules that have been reported at upwards of 90% of dry cell

weight [40]. They exhibit a wide range of properties, enabling them to be viable in a large

fraction of the plastics industry. Typically bacteria synthesize PHA when carbon is

abundant in excess, while one essential growth nutrient such as nitrogen or phosphorous

is limited. It has also been shown that in some organisms, a carbon-limited feeding



Page | 314

strategy can be employed for the production of PHA, with no limiting of essential growth

nutrients [41].

Researchers have also been working on genetically modifying plants for direct PHA

production [32]. Advances in genome sequencing provide the foundation for optimizing

PHA synthesis and for developing the ability to produce PHA polymers with a range of

properties. PHAs are unique in the field of biobased polymers in that they are synthesized

directly as polymers, leading to the potential use of metabolic and genetic engineering

tools to generate new polymers with tailored properties [40]. A wide range of carbon

sources have been used to synthesize both short-chain-length and medium-chain-length

PHAs, which has been summarized in several review papers on the topic [39], [42]. Recently,

it has been reported that PHAs can be synthesized using a forestry-based biorefinery

approach with lignocellulosic process streams, including hemicelluloses hydrolysates,

levulinic acid derived from cellulose, and tall oil fatty acids from kraft pulping, being

employed as the carbon sources for the bacteria Burkholderia cepacia [43]. Other sources

of lignocellulosic by-products have been investigated for PHA production, including the

use of tequila manufacturing bagasse (the rind and fibrovascular bundles dispersed in the

Agave tequilana stalk) [38]. The bacteria used in this study were shown to not only produce

PHAs, but also degraded insoluble cellulose under similar conditions, reducing the

amount of required pretreatment.

Alternatively, instead of using one particular ‘‘super” bacteria, other research has been

focused on the use of microbial consortia to produce PHAs from different carbon sources.

Industrial wastewaters from methanol-enriched pulp and paper mill foul condensate,

fermented municipal primary solids, and biodiesel have been shown to yield PHAs when

passed through batch bioreactors containing a mixed microbial consortium from municipal

activated sludge [44]. Analysis of the system revealed different microbial communities

present, but functional stability was maintained in spite of the contrasting populations. The

wide range of potential carbon sources for PHA production lends itself to the inclusion of

this technology in biorefinery designs since it is not bound by geography. Table 4

summarizes the characteristics of enzymatic transformation process.

Table 4. Characteristics of enzymatic transformation process



Page | 315

4. Conclusion

The biological production of chemicals is not a new technology, though it is one that is

central to the sustained development of biorefining technologies due to the high-value and

relatively low material demands of this industry. However, the economics must be

balanced against the costs of downstream processing, and there is a need to elucidate

the trade-off between operating costs, capital costs, and input (biomass) costs when

investigating a pretreatment regime for a biorefineries [17], [45]. Ιt should be noted that there

can be difficulties in interpreting the effects of pretreatment. When a substrate is exposed

to conditions that alter one of the factors influencing reactivity, other factors that influence

reactivity may also change [7], [14], [46]. Decoupling these effects remains a challenge.

References

[1]. Mabee, W.E., Saddler, J.N., 2010. Bioethanol from lignocellulosics: status and perspectives in Canada. Bioresour. Technol. 101, 4806–4813.

[2]. Pacific Northwest National Laboratory, National Renewable Energy Laboratory, 2004. Top Value Added Chemicals from Biomass: Results of Screening for Potential Candidates from Sugars and Synthesis Gas. Department of Energy, Oak Ridge, TN.

[3]. Ragauskas, A.J., Williams, C.K., Davison, B.H., Britovsek, G., Cairney, J., Eckert, C.A., et al., 2006. The path forward for biofuels and biomaterials. Science 311, 484–489.

[4]. Mosier, N., Wyman, C., Dale, B., Elander, R., Lee, Y.Y., Holtzapple, M., 2005. Features of promising technologies for pretreatment of lignocellulosic biomass. Bioresour. Technol. 96, 673–686.

[5]. Yang, B., Wyman, C., 2008. Pretreatment: the key to unlocking low-cost cellulosic ethanol. Biofuels Bioprod. Biorefin. 2, 26–40.

[6]. Lynd, L.R., Weimer, P.J., van Zyl, W.H., Pretorius, I.S., 2002. Microbial cellulose utilization: fundamentals and biotechnology. Microbiol. Mol. Biol. Rev. 66, 506– 577.

[7]. Zhang, Y.P., Lynd, L.R., 2004. Toward an aggregated understanding of enzymatic hydrolysis of cellulose: noncomplexed cellulase systems. Biotechnol. Bioeng. 88, 797–824.

[8]. Dodds, D.R., Gross, R.A., 2007. CHEMISTRY: chemicals from biomass. Science 318, 1250–1251.

[9]. Lucia, L.A., Argyropoulos, D.S., Adamopoulos, L., Gaspar, A.R., 2006. Chemicals and energy from biomass. Can. J. Chem. 84, 960–970.

[10]. Claassen, P.A.M., van Lier, J.B., Lopez Contreras, A.M., van Niel, E.W.J., Sijtsma, L., Stams, A.J.M., de Vries, S.S., Weusthuis, R.A., 1999. Utilisation of biomass for the supply of energy carriers. Appl. Microbiol. Biotechnol. 52, 0741–0755

[11]. Galbe, M., Zacchi, G., 2002. A review of the production of ethanol from softwood. Appl. Microbiol. Biotechnol. 59, 618–628.

[12]. Pan, X., Arato, C., Gilkes, N., Gregg, D., Mabee, W., Pye, K., et al., 2005. Biorefining of softwoods using ethanol organosolv pulping: preliminary evaluation of process streams for manufacture of fuel-grade ethanol and co-products. Biotechnol. Bioeng. 90, 473–481.

[13]. van Ree, R., & Annevelink, B., 2007. Status Report Biorefinery 2007.

[14]. Chandra, R., Bura, R., Mabee, W., Berlin, A., Pan, X., Saddler, J., 2007. Substrate pretreatment: the key to effective enzymatic hydrolysis of lignocellulosics? Biofuels. Adv. Biochem. Eng. Biotechnol. Springer, pp. 67–93.

[15]. Puri, V.P., 1984. Effect of crystallinity and degree of polymerization of cellulose on enzymatic saccharification. Biotechnol. Bioeng. 26, 1219–1222



Page | 316

[16]. Jørgensen, H., Kristensen, J.B., Felby, C., 2007. Enzymatic conversion of lignocellulose into fermentable sugars: challenges and opportunities. Biofuels, Bioprod. Bioref. 1:119–134

[17]. Yang, B., Wyman, C., 2008. Pretreatment: the key to unlocking low-cost cellulosic ethanol. Biofuels Bioprod. Biorefin. 2, 26–40.

[18]. Khanal, S.K., Montalbo, M., van Leeuwen, J.H., Srinivasan, G., Grewell, D., 2007. Ultrasound enhanced glucose release from corn in ethanol plants. Biotechnol. Bioeng. 98, 978–985.

[19]. Sun, Y., Cheng, J., 2002. Hydrolysis of lignocellulosic materials for ethanol production: a review. Bioresour. Technol. 83, 1–11.

[20]. Teramoto, Y., Tanaka, N., Lee, S., Endo, T., 2008. Pretreatment of eucalyptus wood chips for enzymatic saccharification using combined sulfuric acid-free ethanol cooking and ball milling. Biotechnol. Bioeng. 99, 75–85.

[21]. Moxley, G., Zhu, Z., Zhang, Y.-H.P., 2008. Efficient sugar release by the cellulose solvent-based lignocellulose fractionation technology and enzymatic cellulose hydrolysis. J. Agric. Food Chem. 56, 7885–7890.

[22]. Huang, H.J., Ramaswamy, S., Tschirner, U.W., Ramarao, B.V., 2008. A review of separation technologies in current and future biorefineries. Sep. Purif. Technol. 62, 1–21

[23]. Zhang, Y.P., Ding, S., Mielenz, J.R., Cui, J., Elander, R.T., Laser, M., 2007.Fractionating recalcitrant lignocellulose at modest reaction conditions. Biotechnol. Bioeng. 97, 214–223.

[24]. Fort, D.A., Remsing, R.C., Swatloski, R.P., Moyna, P., Moyna, G., Rogers, R.D., 2007.

[25]. Can ionic liquids dissolve wood? Processing and analysis of lignocellulosic materials with 1-n-butyl-3-methylimidazolium chloride. Green Chem. 9, 63–69.

[26]. Lee, S.H., Doherty, T.V., Linhardt, R.J., Dordick, J.S., 2008. Ionic liquid-mediated selective extraction of lignin from wood leading to enhanced enzymatic cellulose hydrolysis. Biotechnol. Bioeng. 102, 1368–1376.

[27]. Fort, D.A., Remsing, R.C., Swatloski, R.P., Moyna, P., Moyna, G., Rogers, R.D., 2007. Can ionic liquids dissolve wood? Processing and analysis of lignocellulosic materials with 1-n-butyl-3-methylimidazolium chloride. Green Chem. 9, 63–69.

[28]. Zhu, S., 2008. Use of ionic liquids for the efficient utilization of lignocellulosic materials. J. Chem. Technol. Biotechnol. 83, 777–779.

[29]. Swatloski, R.P., Spear, S.K., Holbrey, J.D., Rogers, R.D., 2002. Dissolution of cellulose with ionic liquids. J. Am. Chem. Soc. 124, 4974–4975.

[30]. Zhu, S., Wu, Y., Chen, Q., Yu, Z., Jin, S., 2006. Dissolution of cellulose with ionic liquids and its application: a mini-review. Green Chem. 8, 325–327.

[31]. Dadi, A.P., Varanasi, S., Schall, C.A., 2006. Enhancement of cellulose saccharificationkinetics using an ionic liquid pretreatment step. Biotechnol. Bioeng. 95, 904–910.

[32]. Liu, L., Chen, H., 2006. Enzymatic hydrolysis of cellulose materials treated with ionic liquid [BMIM] Cl. Chin. Sci. Bull. 51, 2432–2436.

[33]. Carole, T.M., Pellegrino, J., Paster, M.D., 2004. Opportunities in the industrial biobased products industry. Appl. Biochem. Biotechnol. 115, 871–885.

[34]. Nakamura, C.E., Whited, G.M., 2003. Metabolic engineering for the microbial production of 1,3 propanediol. Curr. Opin. Biotechnol. 14, 454–459.

[35]. Dodds, D.R., Gross, R.A., 2007. CHEMISTRY: chemicals from biomass. Science 318, 1250–1251.

[36]. Xu, X., Lin, J., Cen, P., 2006. Advances in the research and development of acrylic acid production from biomass. Chin. J. Chem. Eng. 14, 419–427.

[37]. Danner, H., Braun, R., 1999. Biotechnology for the production of commodity chemicals from biomass. Chem. Soc. Rev. 28, 395–405.

[38]. Mullin, R., 2004. Sustainable specialties. Chem. Eng. News 82, 29–37.



Page | 317

[39]. Munoz, L.E.A., Riley, M.R., 2008. Utilization of cellulosic waste from tequila bagasse and production of polyhydroxyalkanoate (PHA) bioplastics by Saccharophagus degradans. Biotechnol. Bioeng. 100, 882–888.

[40]. Sun, Z., Ramsay, J.A., Guay, M., Ramsay, B.A., 2007b. Fermentation process development for the production of medium-chain-length poly-3- hyroxyalkanoates. Appl. Microbiol. Biotechnol. 75, 475–485.

[41]. Lee, S.Y., 2006. Deciphering bioplastic production. Nat. Biotechnol. 24, 1227–1229.

[42]. Sun, Z., Ramsay, J.A., Guay, M., Ramsay, B.A., 2007a. Carbon-limited fed-batch production of medium-chain-length polyhydroxyalkanoates from nonanoic acid by Pseudomonas putida KT2440. Appl. Microbiol. Biotechnol. 74, 69–77.

[43]. Suriyamongkol, P., Weselake, R., Narine, S., Moloney, M., Shah, S., 2007. Biotechnological approaches for the production of polyhydroxyalkanoates in microorganisms and plants – a review. Biotechnol. Adv. 25, 148–175.

[44]. Keenan, T.M., Nakas, J.P., Tanenbaum, S.W., 2006. Polyhydroxyalkanoate copolymers from forest biomass. J. Ind. Microbiol. Biotechnol. 33, 616–626.

[45]. Coats, E.R., Loge, F.J., Smith, W.A., Thompson, D.N., Wolcott, M.P., 2007. Functional stability of a mixed microbial consortium producing PHA from waste carbon sources. Appl. Biochem. Biotechnol. 137, 909–925.

[46]. da Costa Sousa, L., Chundawat, S.P., Balan, V., Dale, B.E., 2009. ‘Cradle-to-grave’ assessment of existing lignocellulose pretreatment technologies. Curr. Opin. Biotechnol. 20, 339–347.

[47]. Kumar, R., Mago, G., Balan, V., Wyman, C.E., 2009. Physical and chemical characterizations of corn stover and poplar solids resulting from leading pretreatment technologies. Bioresour. Technol. 100, 3948–3962.



Page | 318



Page | 319

TECHNICAL AND SUSTAINABILITY ASSESSMENT OF SMALL SCALE BIOENERGY SYSTEMS

A. Zabaniotou a, P. Manara a, D. Mertzis b, Z.Samaras b

a Biomass Group, Department of Chemical Engineering, P.O. Box 455,

Aristotle University Thessaloniki, GR-54124, Thessaloniki, Greece

Email: [email protected] Email:[email protected]

b Laboratory of Applied Thermodynamics, Department of Mechanical Engineering,

P.O. Box 455, Aristotle University Thessaloniki, GR-54124, Thessaloniki, Greece

Email:[email protected] Email: [email protected]

Abstract

Small-scale gasification systems coupled with ICEs could be suggested as feasible alternatives for

CHP production utilizing agro-wastes while contributing to waste management and reinforcing

sustainable agriculture. In this study, the region of Western Macedonia, Greece was selected as a

representative rural region of Mediterranean and Balkan area. With regard to the EU “20-20-20”

goals and to increase investments in CHP production, while avoiding biomass logistics, a technical

assessment of a mobile gasification system was conducted considering sustainability issues. The

bioenergy system was designed, manufactured and demonstrated in Greece with EU support. The

unit has a maximum thermal output of 12 kW in addition to approximately 5 kW electrical output.

The unit was tested and operated in different locations close to biomass feedstock origin in the

rural areas of Western Macedonia in Greece. Peach, olive and grape kernels were utilized as

biomass feedstock while the unit is operated constantly on a 24/7 basis overcoming technical

issues. The present research work describes the unit performance in terms of constant operation,

energy output and process efficiency by utilizing different agricultural residues. The unit envisages

the production of cost effective renewable energy for rural areas while it promotes the concept of

mobile energy production units that utilize by-products of agricultural and forestry activities, which

are otherwise treated as waste. The outcomes of the study indicated that these systems could be

used in industrial agro-companies, to cover needs in heating, being a sustainable means to

improve rural biomass utilization towards decentralized CHP production.

Keywords: gasification, CHP, sustainability, biomass

List of symbols CHP: Combined Heat and Power ICE : Internal Combustion Engine EU: European Union HHV: Higher Heating Value BFB: Bubbling Fluidized Bed RES: Renewable Energy Source



Page | 320

1. Introduction

Through the implementation of Action Plans, Protocols and Common Decisions, the EU

aims at increasing biomass for primary energy use, meeting RES utilization targets,

reducing at the same time greenhouse gas emissions. Decentralized electricity production

is likely to play an important role in future energy supply. Facilities utilizing indigenous

renewable sources, and designed to supply local communities and decentralized energy

users, are in the forefront of EU energy and environmental policies [1]. Feedstock supply,

conversion technology and energy allocation are the main pillars in implementing

bioenergy systems. These factors are interactions among social, environmental and

economic outcomes which a sustainable bioenergy system has to meet. Small Scale CHP

bioenergy systems is one of the very challenging option within renewable and if it can be

proven technically and economically viable it will be of a high added technology due to the

fact that it offers a) decentralized and independent energy generation for rural or isolated

areas (remote communities) b) self-generated, cheap fuel (local biomass and wastes) c)

higher energy efficiency, zero logistics, negligible raw material price d) waste

management. Agro-industrial residues/wastes such as olive kernels from olive oil

production units, grape wastes from wineries, fruit stones from fruits treatment industries

etc, are inefficiently utilized or even totally wasted. The residues to energy technology

route can lead to the displacement of fossil energy sources and hence contribute

significantly to the mitigation of greenhouse gases (GHG) emissions.

In the present study, a prototype Small Mobile Agricultural Residue gasification unit for

decentralized CHP production is evaluated in terms of technical performance and

sustainability. The unit was designed and manufactured and finally demonstrated in four

different locations in rural areas of Western Macedonia. Demonstrative operation of the

developed technology was conducted through testing and application in real world

conditions by using peach, olive and grape kernels as feeding biomass. Over this period,

the unit was steadily improved (overcoming technical problems) to meet the operation

target of 240 hours in each location. The study was carried out in order to evaluate the

potential application and challenges related to the utilization of biomass waste for energy

production via an innovative technological scheme appropriate for rural applications. In

particular, the analysis assesses the potential power production via combined heat and

power biomass waste gasification. The analysis is based on the unit’s demonstration

results.

2. Methods and Materials

2.1. The small mobile bioenergy system

The Small Mobile Agricultural Residue gasification unit is presented in fig. 1 and 2. This

unit has a maximum output of 4.7 kWel. The unit combines the technologies of Bubbling

Fluidized Bed (BFB) gasification and Internal Combustion Engine (ICE).



Page | 321

Fig. 1: Small mobile system

BFB gasification is used to convert solid biomass into a gaseous fuel (producer gas)

which is in turn fed to an ICE coupled to an electricity generator. As a result, producer gas

utilization leads to electricity production. Additionally, water is used in heat exchangers in

order to keep the ICE and generator operation temperatures at specified levels. The

heated water can then be utilized by a consumer, thus making CHP a cogeneration unit.

The flow diagram is depicted in fig. 2.

Fig. 2. Flow diagram of the unit

2.3. Demonstration operation

The unit demonstrative operation was greatly improved during its performance and

reached a state mature enough to cope with any future demonstration. A detailed field

research on the availability profile of residual biomass, in the region of Western

Macedonia, Greece was conducted. 3,614,502 tn of agro-residues were estimated to be

generated in 2009, in the prefectures of Kastoria, Grevena, Florina and Kozani,



Page | 322

accounting for a total savings of 1,360,000 TOE. The criteria employed in order to select

olive, peach and grape for the demonstrative operation of the unit were: A) the fuel’s

suitability for gasification performance B) logistic issues, such as biomass availability,

transportation cost, capital and operating costs related to receiving, handling, storing, and

processing of biomass species. Peach, olive and grape kernels were the biomass

feedstocks which were utilized during experiments testing long term operation. The

ultimate goal of each demonstrative action was to operate the unit on a 240 hour basis at

each demonstration location. The producer gas composition was measured online during

each demonstrative operation by means of a portable gas chromatographer. During gas

sampling, several parameters (temperature, fuel flow, air flow) are also measured in order

to facilitate the evaluation of the results. The fuel characteristics (moisture and ash

content) significantly affected the stability of the gasification process and the overall

efficiency. It can be also noticed that the energy output of the unit is practically

independent of the biomass species. Minor fluctuations exist but at this scale, it is obvious

that the same unit can successfully operate on any of the selected biomass feedstock

without any major process parameter modification. Demonstration results showed that the

unit approximately consumes 5kg/hr and thus the resulted energy balances are described

as follows:

The average energy conversion efficiency of the gasification process (including

gas treatment) is about 62%, which reflects the percentage of energy carried in

biomass that is converted to useful producer’s gas energy.

The total average energy efficiency of the CHP unit is 72%. In particular the CHP

unit is characterized by a 20% electrical efficiency and a 52% thermal efficiency.

The system total energy efficiency is 48% (6% electrical efficiency).

It has to be noted that more than 60% of the electrical output returns to the unit for self-

consumption purposes and only the remaining is guided to the electricity grid.

The overall process efficiency of the unit is depicted in fig. 3, where the Sankey diagram

of the combined heat and power production process is presented. A biomass input of 100

kWth is considered as the reference for the calculations. All results, efficiencies, outputs

and additional inputs are scaled to 100 kWth input.



Page | 323

Fig. 3: Sankey diagram of the CHP production process.

3. Sustainability analysis Key sustainability issues for bioenergy production can involve provision energy service to

rural and remote areas, implications for agro-industrial development and job creation,

health implications, implications for the agriculture policy restructuring and food security,

implications for trade and energy security, impact on biodiversity and preservation of

natural resources, management of wastes and climate change [2]. The developed unit is

assessed in this study by using various sustainability indicators and in the context of EU

policy [3-6].

3.1. Metrics and indicators

Bioenergy plans should be promoted and developed according to sustainability criteria.

Key performance indicators are already used by Governments and businesses to assess

the progress or success rate of policies and strategies. Indicators are quantitative or

qualitative factors that provide means to measure the degree of achievement, to reflect

changes, or to assess performance or compliance [7]. There is no single indicator which

can embody all the issues of sustainability. In that respect, many indicators can be

derived. Thus, sustainability indicators can be classified in environmental, economical and

social. Indicators can cover several issues of these sectors. Several research groups have



Page | 324

worked to develop indicators for bioenergy sustainability [8-11]. Building on existing

knowledge, the study recommends sustainability indicators and metrics for the present

bioenergy system in order to guide any analysis undertaken of bioenergy schemes at the

domestic level with a view to informing decision making and facilitating the sustainable

development of bioenergy (Table 1, 2, 3).

For an overall sustainability assessment all steps from feedstock production to the end

use of bioenergy should be taken in consideration. However, considering that

sustainability is contextual, relative and its implications of bioenergy choices are large and

complex, the selection of sustainability indicators related to the bioenergy system of this

study, was based on performance indicators and technical outputs of the unit

demonstration period, aiming to assess the performance. For the purpose of this study,

the assessment is conducted for the biomass conversion step till the end product

(electricity) due to the fact that feedstock production and feedstock logistics are out of the

system boundaries since the raw material are residues and not crops. In addition their

treatment will be done in site by moving the CHP unit, avoiding feedstock logistics.

3.1. Sustainability performance evaluation

The implemented scenario for the assessment assumes that the gasification unit operates

with agro-residues for an overall of 7000 h/y and all heat and power produced are

economically utilized. The cost for raw materials purchase would be determined by the

utilized biomass fuels feeding rates per year multiplied by their price. The demonstrative

operation scheme assumes zero cost for the raw materials, since the implemented

scenario assumes that the unit is operating under the frame of an agro-industry exploiting

the produced agro-wastes. Furthermore, capital and annual operational costs are defined

as follows; the capital expenditure for the unit development consists of the equipment

purchase cost and the installation of the equipment, while the annual operational and

maintenance cost is an objective function of the following sub-costs; the material purchase

cost, the waste disposal cost, labor cost and maintenance cost that are essential to the

unit operation on an annual basis. Revenues result from both electricity and thermal

power sales. Power produced via biomass gasification accounts a feed-in tariff of 220

euros/ ΜWhel, while thermal power price is equal to 50 euros/ ΜWh. Specifically, the

operational cost, at an annual basis, includes energy and material consumption for the

unit operation. Waste management deals with the disposal of ash, accumulated particles

as well as tars downstream of the gas cleaning system. Transport cost and some extra

costs that are essential to the unit operation on an annual basis are included in the

overheads. Extra costs include labor, maintenance and repair of the equipment and the

plant overhead costs. During the plant operation 2 operators are required. The average

salary of an operator is 12,000 €/y. The annual production/operating cost is computed

summarizing all the above sub-costs. To evaluate the profitability of the two schemes we

have set the lifetime of the plant to 20 years. The rate of interest (i) used in this study set

to be 8% and taxes that are charged against the plant’s profits have assumed to have a

rate of 25%.



Page | 325

Table 1. Environmental indicators (quantitative/qualitative/long term)

Environmental Pillar

ISSUES Environmental sustainability is related to energy, resources and emissions savings, “carbon footprint”, waste production, from end of pipe to prevention.

Indicator Indicator description Value Comments

Net Energy Ratio (NER)

Thermodynamic indicators, measure the efficiency of resource and energy utilization indicating indirectly the environmental consequence. Energy intensity metrics are quantified based on the first law of thermodynamics- energy balance. In general, these indicators are defined in units of energy consumed per product unit. Net Energy Ratio (NER) is applied to estimate the net energy process product value and it is defined as the ratio of the energy output to energy input.

48%

NER value, less than one, indicates a net energy loss. However, biomass is a lower quality energy source (per joule) than electricity. Furthermore, the unit utilizes as feedstock for energy production (CHP) waste biomass of zero and even sometimes negative value. CHP production enhances process efficiency.

Reduction in greenhouse gas emissions (GHG)

The indicator estimates the degree to which greenhouse gas (GHG) emission savings can be achieved through bio-energy use in the power sectors. Emissions (per kWh) from lignite power production are compared to the emissions from our bio-energy unit. The assessment was based on the utilization of an emission factor for electricity production that reaches the value of 0.537 kg CO2/kWh [12].

~0.55 kg/kWh

CO2 savings from the implementation of this energy production scheme, in a commercial scale, is appeared to be proportional to the scale of electrical power production. Calculations are based on line emission analysis. Emissions from crop harvesting and soil carbon balances are out of our system boundaries that are defined from waste storage to end use. These emissions are taken into account in the production cycle of the primary product (e.g wine, oil)

Carbon footprint

Power production utilizing biomass fuels is considered by many researchers and analysts to be carbon neutral.The total green house gas emissions produced within our system boundaries over the entire life cycle of the system is the "carbon footprint" (annual basis)

3.500 kg CO2-eq/year

The calculation refers to the unit’s “carbon footprint” at an annual basis and is calculated based on line emission analysis.

Waste product

ion

The indicator refers to the production of tarry by -products, ash and particles (kg/y).

Zero Waste process

Small amounts of tars are produced. Tars could be fed in the gasifier, treated within the waste water treatment plant of an agro-industry or disposed off as waste for offsite controlled treatment. Solid waste (ash, accumulated particles downstream the gas cleaning system) are nutrient sources for soil fertilization (biochar)



Page | 326

Table 2. Social indicators and metrics (quantitative/qualitative/long term)

Social Pillar ISSUES

Social sustainability is related to employment and stability of livelihood in the local communities, whereas from the point of consumers’ view, is related to product quality and public acceptance of biomass activities. Social sustainability indicators are difficult to quantify and are often qualitative.

Indicator name Indicator description Value Comments

Employment creation

A quantitative indicator for social sustainability assessment is the number of jobs per unit investment. Net job creation as a result of bioenergy production and use, is classified as follows: a) skilled/ unskilled personnel, b) temporary/ indefinite jobs. It reflects the economic and social contribution of the bioenergy plan to the local society.

1-4 Jobs/ CHP unit

The created jobs are part or full time, indefinite and appeal for skilled, high level educated personnel. The number of personnel depends on the unit’s automation system, as well as on the scale.

Food and feed security

The indicator is related to the potential implications for food security due to a bioenergy plan.

YES Waste and residual biomass that is used as feed in the unit does not compete with food and feed.

Waste management

The indicator reflects the opportunity that a bioenergy plan offers for waste management

YES Waste management of agro-industrial by-products with energy production.

Incidence of occupational injury, illness and fatalities

Incidence of occupational injury, illness and fatalities in the production of bioenergy in relation to comparable sectors (lignite power production plants).

NO

In general, bioenergy will assist towards the improvement of quality of life in the region of Western Macedonia, Greece by partial replacement of lignite. The incidence of occupational injuries, illness and fatalities is decreased when bioenergy production units are established in comparison with high risk lignite mining activities wich are taking place in the region.

Land uses and its effect on rural livelihoods and vulnerability

The indicator reflects the land competition between bioenergy uses and other land uses

NO

The mobile, small scale bioenergy production unit is implemented in the source of waste biomass production. The unit’s successful operation is not related to land uses for energy crops cultivation.



Page | 327

Table 3. Economic indicators (quantitative/qualitative/long term)

Economic Pillar

ISSUES Economic sustainability can be analyzed either through internal functions of the company or by the external effects on society and environment.

From the internal point of view, the financial performance of a company and capability to manage assets are the most important factors leading to economic sustainability. Analysis of the external implications of economic sustainability management focuses on the company`s influence on the wider economy and

how the company manages social and environmental impacts. Indicator name Indicator description Value Comments Microeconomic Indicators

Investment cost per bioenergy unit

The capital expenditure for the unit development consists of the equipment purchase cost and the installation of the equipment. The indicator is given per kWel produced.

~6,800 €/kWel

The value depends on the number of constructed units (10 units-5kWel). Fixed cost shows a 20% reduction in case of multiple unit construction.

Gross Profit An indicator that measures the difference between output and intermediate consumption.

up to 3,500 €/year

In case of system improvements; automated system performance, 1 part time operator, 10units construction (5kWel installed capacity) - best case scenario-30% state subsidy.

Profitability Level of profit taking into account taxes and depreciation over a period of 20 years

up to 2, 500 €/year

Best case scenario. Stimulating bioenergy development through adequately structured law and policy can potentially provide additional income and employment for small-holders and other agricultural producers.

Macroeconomic indicators

Total value added to the economy

The Total Value Added to the local economy is comprised of the following values; labor income plus taxed profit at an annual basis per bioenergy plant

6,000 €/ unit (annual basis)

The area of activity is the region of Western Macedonia. The total value added reflects the added value per unit due to the additional income from sales and employment

Energy diversity Change in diversity of total primary energy supply due to bioenergy. Bioenergy units per unit investment

5 kWel/ unit investment

Electricity production and district heating from waste biomass.

Business opportunities

New market opportunities YES In the field of bioenergy.

Long-term profitability

Impact to long term sustainability. Economic sustainability is strongly connected to other dimensions of sustainability, and is defined in the context of social and environmental impacts.

YES

The bioenergy system promotes sustainability, making positive contributions to the local community, broader society. The technological scheme excels in the three major sectors (environment, society, economy)



Page | 328

4. Conclusion

Biomass gasification coupled with an ICE engine has the potential to promote bioenergy in

rural areas. The total electrical efficiency of the system reaches the value of 6%. Combined

heat and power production increases the overall energy efficiency. The unit operates having

low greenhouse gas emission profile. Social and economic metrics show the potential gains

from the implementation of the present technological scheme in the local communities. The

innovative proposed technological scheme seems of great potential. However, at the

moment, commercial success depends on capital reduction instruments such as subsidies,

electricity feed in tariff, biomass price, scale.

Acknowledgements

The authors wish to acknowledge co-funding of this research by European Union- European Regional Development Fund and Greek Ministry of Εducation/EYDE-ETAK through program ESPA 2007-2013 / EPAN II / Action “SYNERGASIA” (Project 1165)

References

[1]. P. Carneiro, and P. Ferreira, P., “The economic, environmental and social value of biomass”,

Renew Energ, 44 (2012) 17-22. [2]. Evans, V. Strezov, T. J. Evans, “Sustainability considerations for electricity generation from

biomass”, Renew and Sust Energ Review, 14 (2010)1419–27. [3]. EU, European Union Climate and Energy Package, “Communication from the Commission to the

European Parliament, the Council, the European Economic and Social committee and the Commitee of the Regions”, COM (2010) 265, 26.5.2010

[4]. EU, European Union, Report from the Commission to the Council and the European Parliant, COM (2010)11, 25.2.2011

[5]. Law 3851/2010, “Promotion of Renewable energy sources to eliminate the effect of climate change”.

[6]. Law 4062/2012, “Sustainability criteria for bio-fuels”, in compliance with Directive 2009/30/EU. [7]. FAO, Food and Agriculture Organization of the United Nations, Criteria and Indicators for

Sustainable Forest Management – Annex: 1. PRINCIPLES, CRITERIA, INDICATORS AND VERIFIERS. Rome, 2002

[8]. http://www.biomassboard.gov/index.html (assessed 15.5.2013) [9]. http://rsb.org/ (assessed 15.5.2013) [10]. http://www.csbp.org/ (assessed 15.5.2013) [11]. GBEB, Grobal bioenergy partnership, Full Report, 2011 [12]. Vlysidis, A., Binns, M., Webb, C., Theodoropoulos, C., A techno-economic analysis of biodiesel

biorefineries: Assessment of integrated designs for the co-production of fuels and chemicals, Energy, Vol. 36, pp 671-83, 2011.



Page | 329

TECHNICAL AND SUSTAINABILITY ASSESSMENT OF END OF LIFE TYRES DEPOLYMERISATION

N. Antoniou & A. Zabaniotou

Department Of Chemical Engineering, P.O. Box 455,

Aristotle University Thessaloniki, GR-54124, Thessaloniki, Greece Email: [email protected],


Abstract

This paper presents the essential features of an efficient and environmentally attractive pyrolysis for End of Life Tyres valorization, with energy and materials recovery. The problem of End of Life Tyres management strongly affects not only the environmental protection but even the resources maintenance, since problems related to the depletion of resources, energy demand and waste management, are strictly connected and required an integrated approach. The present study aims at the technical and sustainability assessment of End of Life Tyres pyrolysis units, towards carbonaceous material production. The final destination of the pyrolysis solid residue largely influences the industrial applications of pyrolysis. The implementation of an innovative technique such as pyrolysis in industrial scale, taking under consideration research data and already known methodologies, while obeying to actual environmental regulation can be sustainable if targeted to char valorization for adsorptive materials production. The potential added value activated carbon can be effectively used for water treatment and pesticide removal.

Keywords: ELT valorisation; Used tyres; depolymerisation; carbonaceous materials from waste; sustainability;

List of Symbols

ELT : End of Life Tyres

BR : Butadiene Rubber

TPO : Tyre Pyrolysis Oil

NR :natural rubber (NR),

SBR : styrene–butadiene rubber

LHV : Lower Heating Value (MJ/Nm3)

1. Introduction

Tyre production companies are scattered through the world, generating annually large

quantities of various sizes and purposes, tyres. Statistical data from the previous years show

that in 2011, 14.68 million tons of tyres were produced, 5% more than the previous year [1-

3].

Used tyres or End of Life Tyres are classified as wastes containing principally organic

constituents, which may contain metals and inorganic materials (B3140), or a non hazardous

waste (160103) [4, 5]. Tyres are produced by more than 100 different species and their

composition from different tyre parts (tyre sidewall or the tyre tread) varies due to the



Page | 330

different desired characteristics of product. Generally, tyres are consisted of natural rubber

(NR), styrene–butadiene rubber (SBR) and butadiene rubber (BR), Carbon Black and other

additives [6].

ELT management is of crucial importance in order to maximize the benefits of such an

important raw material. Several models of ELT management are adopted around the world.

More specifically across Europe, Producer Responsibility, Free Market System and Tax

System are proposed, giving the right to each member state to choose the proper for it [7].

The target of the implementation of these ELT management systems, is the increase of

product, material and energy valorization. As science progressed, more alternatives were

added to the already known technologies and progressively reduced the amount of ELTs

dumped in landfills.

Although they constitute a problem, at the same time they offer an opportunity for resources

conservation, since they are a potential source for recovery of raw materials and fuels.

Several techniques and methods like retreating and reclamation were applied in the past.

Recycling and natural degradation are characterized by lack of efficiency, since they are high

cost processes with questionable results. The most immediate thermal application of used

tyres, is the combustion with or without the production of energy, or co-combustion in cement

kilns. As an alternative, pyrolysis and less frequently gasification were proposed and tested.

Depolymerisation via pyrolysis attracted much attention due to the value of by-products.

However, problems concerning the disposal of by-products have made it less attractive than

expected. The problem that mostly occurs though, is the disposal of process products and

especially pyrolysis oil and residual carbon black; pyrolysis gas can be consumed in-plant for

heating the pyrolysis kiln. As far as the above problems remain unsolved, an investment in

tyre pyrolysis should be most probably characterized as non attractive. It gets obvious thus,

that the implementation of pyrolysis, as a means of solving the used tyres management

system, without seeking to produce any marketable product is not sustainable. It is highly

important to search for ways of using products and investigate niche markets that will offer

economic benefits and decrease pyrolysis cost.

An ELT depolymerisation plant could be viable if targeted to the production of carbonaceous

materials with adsorptive characteristics while at the same time uses process’s by-products

(pyrolytic gas and oil) to fulfill the energy needs of the plant. A technical analysis is essential,

since it could highlight the most important parameters that determine the economic viability

of this attempt. The sustainability analysis moreover, as a useful tool, can assist decision-

makers by determining which actions should be taken in an attempt to make the project

sustainable.



Page | 331

2. Depolymerisation process

2.1. Pyrolysis

The main route for ELT depolymerisation is via pyrolysis. During pyrolysis, heat is supplied at

an inert atmosphere, to the tyres. Pyrolysis of ELT is performed in temperatures ranging from

4000C to 6000C under nitrogen or helium atmosphere. Numerous thermal cracking reactions

take place and finally decompose tyres into a series of gas, liquid and solid products.

Pyrolytic gases, according to literature, are found to be of high energetic content (LHV> 35

MJ/Nm3), proving that their use, primary or supplementary, for electricity production in order

to cover plant’s energy needs is possible. They can sufficiently cover needs of the pyrolysis

plant except for the start up period [6].

TPO comprised of the condensable fraction of large hydrocarbons shows also high values of

GCV (>25 MJ/Nm3) [6]. This high calorific value can permit their direct use as liquid fuels or

as a component to mixtures with common Diesel fuel, not creating any seizing or injector

blocking problems during the entire operation of the engine running [8]. Moreover, if any

further processing of the TPO is programmed including (i) moisture removal (ii)

desulphurisation and (iii) vacuum distillation, towards a decrease in emissions and viscosity,

this could help in the more applied use of TPO as a fuel in diesel engines.

The solid product of pyrolysis, also known as char, may be used either as a smokeless fuel, carbon black filler material or as an adsorptive material. However, due to the high calorific

value of these solid residues, they can also be combusted for energy recovery. The chars

obtained in tyre pyrolysis are mainly mesoporous materials with a low surface area. In most

cases, ELT char in order to acquire some specific characteristics it has to be activated. Then,

the end product can be used in several cases with impressive results antagonizing

commercial products. The solid product can be further processed to acquire specific

characteristics, so as to meet specifications for carbon black, or it can be marketed directly

[9].

2.2. Reactors

Literature data has shown that the use of rotary kilns for both the processes of pyrolysis and

activation is dominant for the production of solids. More specifically, in pyrolysis process,

over a large variety of reactors including fluidized, fixed bed, ablative, molten salt or even

more innovative (plasma and microwave), the rotary kilns are selected due to their structural

characteristics, ease of handling, their controllable operation, affordable cost and custom

potentials [6]. The same are valid for activation process, since it is a much more intense

process with higher final temperatures and residence times. Alternatives to rotary kilns for

activation but with limited appliance are the fluidized bed and the multiple hearth.

2.3. Adsorptive material production



Page | 332

The received solid product from ELT pyrolysis is used as a raw material for the activation

process, physical or chemical. The difference between the two processes is the different

activating agent, steam or CO2 or mixtures of them in one hand and KOH, ZnCl2 and H3PO4

on the other. Additionally, higher temperatures are applied to the physical activation (>800 0C).

The end product depending on the process conditions and the type of raw material

(demineralized or not) can show improved textural characteristics. These parameters can

determine the possible uses of the end product which can in turn, also influence

depolymerisation plant’s viability. The already known uses of Activated Carbon are targeted

mainly to gas and liquid phase applications [10].

2.4. The Depotec Project

DEPOTEC (Depolymerisation Technology for Rubber with Energy Optimisation to Produce

Carbon Products) is an EU LIFE + project whose objective is to contribute to the

implementation, updating and development of EU environmental policy with European added

value. The DEPOTEC project proposes a tyres depolymerisation process that will add value

to the waste tyres by producing products that can be used as substitute carbon filler

materials in the rubber manufacturing process or activated carbons, aiming to ultimately lead

to a reduction in stockpiling of tyres, as they will become valuable raw materials for the

production of these products [11]. It will also offer an alternative to burning ELT to produce

tyre-derived fuel (Fig. 1).

Fig. 1: Depotec project target

3. Technical assessment

An ELT depolymerisation plant could be viable if targeted to the production of carbonaceous

materials with adsorptive characteristics while at the same time uses process’s by-products

(pyrolytic gas and oil) to fulfill the energy needs of the plant. In order to achieve this goal, the

process of the adsorptive material production should be performed within the whole

Sustainable

+

Environmental friendly process

Feedstock : End of Life Tyres

Low cost procedure

High added value materials CBp, Activated Carbon, Pesticide adsorber

TARGET



Page | 333

depolymerisation plant; this additional unit should be part of the integrated tyre pyrolysis-

activation production complex. The carbonaceous pyrolytic char can be further upgraded to

activated carbon. In Fig. 2 the three stages of morphological development of ELT particles

are easily distinguished.

Fig. 2: Morphological development of a 10-15 mm tyre particle: a) ETL particle containing linen fibers,

free of steel, b) ELT pyrolysis char (after pyrolysis process-1/3 of the initial mass), c) Activated Carbon

(minimum mechanical process to the desired pulverized form)

The availability of the raw materials, as well as, the potentials of the local market are two

parameters of crucial importance. The selection for use of rotary kilns has proved to be the

best, since it assures fewer problems during operations and a steady level of accepted

quality for the end products. The number of rotary kilns that will be used is dependant of the

above parameters. In either case, steam or even CO2 can be used depending of the end

product’s specifications. In case of a demanding market, two rotary kilns are considered a

logical approach. The first will be dedicated for pyrolysis process and the second for

activation. In a different case, one medium sized rotary kiln reactor seems practical enough

for performing both the pyrolysis and activation process.

In the proposed flowsheet (Fig. 3) two reactors are used, one for each thermal process,

pyrolysis and activation. Heat and energy recovery is achieved by economizer and co-

generation units, valorizing pyrolysis by-products (oil and gas) thus supplying electricity and

steam for the activation process of char pyrolysis. Furthermore, any excess of electricity may

be directed to the local power company yielding to additional financial profits.

By the implementation of this route, a strong economic incentive is gained, since a

marketable, high-added value product is produced in a self-sufficient plant that obeys strict

environmental regulations. Finally, attention should be paid to regulations required for using



Page | 334

by-products as fuels and the implementation of well accredited standards to secure end

products specifications.

Fig. 3: DEPOTEC flowsheet

4. Sustainability assessment

Pyrolysis as a thermal process, requires modest land space in relation to landfills, minimizing

at the same time the dependence on landfills. The process produces marketable by-products

finding numerous appliances and furthermore the energy that can be created is renewable

energy. However, the conservative public perception to innovative technologies can be a

burden in ELT pyrolysis implementation. The public perceives it as being similar to

incineration. Opponents claim that it undermines recycling and composting, it has toxic air

emissions, and it requires carbon-based materials to work effectively. However, it is different

than incineration because the waste is only partially oxidized, making the process much

cleaner. It also can accept more types of waste, and it has more useable by-products. To

adopt this technology, significant consumer education needs to be invested to gain public

support. Other potential disadvantages are that the ash by-product may contain toxic

substances and heavy metals due to the lack of oxygen, and any by-product that cannot be

used (e.g., ash) is typically landfilled. While this technology has been used for other

processes, such as coal power plants, it is still at the development stage for use with solid

waste.

A sustainability assessment (Table 1) was attempted based on Futurity, Equity, Public

participation, Environment and Economic principles and derived indicators [12]. In the

qualitative assessment, depicted in Table 1, indicators weight on measuring sustainability is

marked by high, medium and low.

Tyre pyrolysis plant economics are very sensitive in product yield, since economics are

substantially affected by product’s quality and not on weight [6]. Product price, production



Page | 335

capacity, total production cost, capital investment and the tipping fee are the most important

variables for the economic evaluation of the plant. The wide use of char as a marketable

solid fuel or adsorptive material depends on its structural and surface characteristics. Its

approval by environmental restrictions can further improve its competitiveness among other

adsorbing materials.

The decision for a tyre depolymerisation plant for a waste management company, will

depend on whether the costs of the pyrolysis process are less than that of

combustion/incineration. SWOT analyses for both the cases of pyrolysis and incineration

were performed and their results are depicted in Tables 2 & 3 [7].

Table 5: Sustainable Development Principles for ELT depolymerisation Principle Indicators Sustainability

Environment Decreased Emissions High

Protection from fires/contagious illness

High

Renewable Energy High

Economic Capital Cost Low

Operating Cost Low

Revenue potential High

Feed in Tariff Low

Public Participation Public Scepticism Medium

Futurity Landfilling minimization High

Fossil Fuels preservation High

Additional Fossil Fuel consumption

Low

Equity Protection from disturbing/potentially harmful

odors, noises

High

The type of product of an ELT industrial pyrolysis plant and its desired properties, in a

sustainable and eco-friendly frame that reuses or diminishes any harmful byproducts, is of

critical importance. Secondly, the production of marketable products, with known

characteristics and applications to be used is another key element in the decision making.

The latter is in full agreement with the “End of life tyres” report published in 2010 by

European Tyre & Rubber Manufacturers Association (ETRMA). In this report it is clearly

stated that “the economic viability of this alternative route for high temperature resource

recovery from tyres is hampered by the fact that the prices obtained for the by-products often

fail to justify the process costs. Under current market conditions the economic viability of

these options has yet to be proved (there are few or no large-scale plants currently in

operation) but they have the merit to offer scope for increasing recycling rates” [13].



Page | 336

Table 6: SWOT analysis for the pyrolysis of ELT

Internal

Strengths Weaknesses

1. Reduced air emissions 2. High efficiency and energy self-

sufficiency 3. Full utilization of products

4. Green energy 5. Zero Waste

1. High investment and operating costs 2. New Technology, a limited number of

commercial applications 3. Lack of products’ standardization

4. Viable for capacity> 20.000 tns / yr

External

Opportunities Threats

1. Sufficient expertise (research level) 2. Saving resources

3. Reduction / elimination of the exploitation abroad

4. Funding opportunities from technology innovation programs

5. New market development 6. Zero Waste

1. Shortage of BAT (Best Available Techniques)

2. Unstable economic environment 3. Uncertain markets for products

4. Confusing legal framework

Table 7: SWOT analysis for the incineration of ELT

Internal Strengths Weaknesses

1. Known technology 2. Existing expertise.

3. Efficient control systems 4. Legal framework

1. Expensive clean system emissions 2. Production of waste (ash)

3. Problematic by-product (ash) 4. Low material recovery 5. Low energy efficiency

External Opportunities Threats

1. Alternative fuel

2. Existing infrastructure (cement kilns) 3. Converting waste tyres into complementary

fuel

1. Low social acceptance of the local community

2. Unstable economic environment 3. Harmful emissions to the environment

5. Conclusion

Pyrolysis of ELTs is an alternative, efficient and environmental friendly valorization process,

in comparison to widely used combustion–incineration. ELT Pyrolysis economics can

become considerably attractive, if the end product becomes, via activation, competitive to

commercial products by proper standardization. Furthermore, by the effective valorization of

the liquid and gaseous pyrolysis products towards heat and energy recovery, the operating

cost could be minimized.

Acknowledgements

The project is funded by LIFE10 /ENV/IE/00695 DEPOTEC project.



Page | 337

References

[1]. www.etrma.com (assessed on 4/4/2013) [2]. http://www.jatma.or.jp/english/about/ (assessed on 4/4/2013) [3]. http://www.rma.org/ (assessed on 4/4/2013) [4]. http://archive.basel.int/index.html (assessed on 4/4/2013) [5]. http://www.environ.ie/en/Publications/Environment/Waste/WEEE/FileDownLoad,1343,en.pdf

(assessed on 4/4/2013) [6]. N. Antoniou & A. Zabaniotou, "Features of an efficient and environmentally attractive used tyres

pyrolysis with energy and material recovery", Renewable and Sustainable Energy Reviews, 20 (2013) 539-558.

[7]. M.C. Samolada & A.A. Zabaniotou, "Potential application of pyrolysis for the effective valorisation of the end of life tires in Greece", Environmental Development, 4 (2012) 73-87.

[8]. S. Murugan, M.C. Ramaswamy & G. Nagarajan, "The use of tyre pyrolysis oil in diesel engines", Waste Management, 28 (2008) 2743-2749.

[9]. R. Murillo, E. Aylón, M.V. Navarro, M.S. Callén, A. Aranda & A.M. Mastral, "The application of thermal processes to valorise waste tyre", Fuel Processing Technology, 87 (2006) 143-147.

[10]. H. Marsh & F.R. Reinoso, Activated Carbon, (Elsevier Science, Great Britain, 2006). [11]. http://www.depotec.eu (assessed on 4/4/2013) [12]. S.R. Bahia, "Sustainability Indicators for a Waste Management Approach", (MSc Research

Project Report on Environmental Pollution Control, University of Leeds, Leeds, 1995). [13]. www.etrma.com (assessed on 4/4/2013)



Page | 338



Page | 339

CROSS-INDUSTRY INNOVATIONS IN THE RENEWABLE ENERGY

INDUSTRY

M. Kloibera & R. Priewasserb

aInstitute for Environmental Management in Companies and Regions

Johannes Kepler University, Altenbergerstrasse 69, A-4040, Linz, Austria


bInstitute for Environmental Management in Companies and Regions Johannes Kepler University,

Altenbergerstrasse 69, A-4040, Linz, Austria

Email: [email protected] Abstract

As in many other industries the products, services and business models of the renewable energy industry are largely shaped by the mindset of their own economic sector. On the other hand analogical problem solving and search for technological solutions in distant industries can open up interesting new perspectives and could be a significant source for radical eco-innovations. Those cross-industry innovations lead to lower development time, lower project risks and higher growth rates and margins due to radical innovations. Companies implement cross-industry innovation strategies mostly unsystematically because there is a lack of existing process models and methods. Therefore, the purpose of this article is first to analyse the value of cross-industry innovations contributing to the creation of innovations focussed on the renewable energy industry. In a second step a step-by-step strategy is presented which should help companies and especially innovation managers to implement systematically eco-innovations across industries in the fuzzy front-end of the innovation process.

Keywords: Cross-Industry Innovation, Radical eco-innovation, Renewable Energy Industry, Innovation Process

1. Introduction

Global trade, rapid technology change and short product life cycles have led to increasing

competition over the past decades. This development forces companies to rethink their

traditional internal innovation strategy in order to differentiate themselves from their

competitors [1,2]. Thereby, open innovation has been proposed as a new paradigm for the

management of innovation [3,4]. Empirical research demonstrates that innovation

collaborations across organizational boundaries - by using a wide range of external actors

and sources - have a positive impact on a firm’s innovation performance [5].

Creating innovations with companies of distant industries is a new phenomenon for theory

and practice in respect of an open innovation approach. These so-called cross-industry

innovations can be technologies, patents, specific knowledge, business processes, or whole

business models [6]. Imitation and Retranslation of already existing solutions from other

industries could contribute significantly to the development of breakthrough innovations in the

renewable energy industry. Highly novel innovations lead to lower development time, lower

project risks and higher growth rates and margins. Despite these advantages companies

implement the Cross-Industry Innovation approach very rarely or unsystematically [7].



Page | 340

Therefore, the research question of this paper is: ‘How could companies of the renewable

energy industry implement Cross-Industry Innovations systematically in the fuzzy front-end of

the innovation process?’

This article is structured as follows: First, we review current literature on open innovation and

cross-industry innovations and we illustrate two examples of firms which developed cross-

industry innovation successfully with analogical thinking. We critically discuss the potential of

the Cross-Industry Innovation approach for the renewable energy industry. Next, we present

a step-by-step strategy which should help companies and especially innovation managers to

implement systematically eco-innovations across industries in the fuzzy front-end of the

innovation process.

2. State-of-the-Art 2.1. Open Innovation

In the past, most of the companies were focused on their internal Research and

Development (R&D) activities and their internally developed products which are distributed

by the firm. This innovation strategy with limited interaction with external partners is defined

as closed innovation strategy. In contrast to this closed strategy a new innovation framework

has evolved. The open innovation paradigm could be seen as an open system – innovative

firms use a wide range of external actors and sources to help them create innovations

[5,8,9]. Chesbrough (2006a) defines open innovation as ‘the use of purposive inflows and

outflows of knowledge to accelerate internal innovation, and expand the markets for external

use of innovation, respectively’ [8,9].

There are diverse potentials of open innovation. On the one hand it could lead to shorter

innovation cycles, lower development costs, opening up of new markets and to the reduction

of market and technological uncertainty in the innovation process through external

knowledge transfer into the company. On the other hand, profits can be generated by

licensing IP and/or multiplying technology through the transfer of internal ideas to other

companies [10, 11]. These numerous potentials signify different open innovation strategies.

Gassmann and Enkel (2004) identified three core open innovation processes:

(1) Outside-in process: the aim is to enhance the companies knowledge base by

integrating the innovative knowledge of external actors.

(2) Inside-out process: earning profits by transferring ideas to the outside environment.

(3) Coupled process: linking outside-in and inside-out processes by working in alliances

with complementary companies.

Recent empirical studies demonstrate that an open innovation strategy provides ideas and

resources that help organizations gain and exploit innovative opportunities [5].

Companies can interact with a range of different external actors. They can transfer external

knowledge from customers, suppliers, competitors, research institutes and universities.

Analogical problem solutions and novel ideas could be also found successfully among distant

industries [6,12].



Page | 341

2.2. Cross-Industry Innovation

Innovations are usually the outcome of the linkage of different pieces of knowledge that have

not been connected before. Searching for analogies is a promising methodology to create

new combinations of knowledge. Analogies are characterised through similar aspects of two

objects of different domains. Drawing analogies beyond the borders of one’s own industry

can evolve Cross-Industry Innovations. Here, a solution is found in one industry and applied

to solve a problem in another industry. On the one hand, the creation of Cross-Industry

Innovations may reduce uncertainty as potential solutions have already proved to function in

a similar context. On the other hand, analogies applied across industries may entail

breakthrough innovations because different pieces of knowledge are combined [7,13,14,].

There are a multitude of examples of technology spillovers from other industries. The path-

breaking interface iDrive from the BMW Group is based on the tried-and-tested technology of

the joystick of the video game industry. Another famous example is the Aeorccino from

Nespresso. The company adapted the established stir principle used in labs, which uses a

contact-free driven beater with magnetic torque transmission. A milk creamer that is easy to

clean was the innovative result [6,7].

Based on a multiple case study, Gassmann and Zeschky (2008) proposed a model for the

development of product innovations by means of analogical thinking (see Fig. 1). It is

targeting the early innovation challenges in how to find highly novel solutions.

Fig. 1: Opening up the Solution Space by Abstraction from the Underlying Problem [14].

This Cross-Industry Innovation model includes three major steps: (1) Abstraction, (2)

Analogy and (3) Adaptation. First, a problem has to be analysed in detail and key terms have

to be abstracted. Second, Analogies in different industries has to be found. In the last phase,

the relevant knowledge technology has to be transferred and adapted.

This Cross-Industry Innovation model is a rare example of existing frameworks. The three

developed steps are crucial for each CII-process; however, the analogy search is not



Page | 342

described in detail. Therefore, the purpose of this paper is to present a Cross-Industry

Innovation process which should help companies of the renewable energy industry to find

promising analogies for their technological problems and to implement them in the

developing phase.

3. Potentials of Cross-Industry Innovations for the Renewable Energy Industry

As in many other industries the products, services and business models of the renewable

energy industry are largely shaped by the mindset of their own economic sector. On the

other hand analogical problem solving and search for technological solutions in distant

industries can open up interesting new perspectives and could be a significant source for

radical eco-innovations. Following list presents different advantages for the renewable

energy industry:

The time-to-market is shorter and the project risk is lower because of the integration

of an already tested and utilised technology in other industries.

Analogies from another industry can generally be utilised without competitive

conflicts.

Stronger differentiation of the product in comparison to the competitors leads to

higher growth rates and margins.

Knowledge transfer across industries can help a company to improve its own powers

of innovation.

4. Systematic Cross-Industry Innovation Implementation

Current empirical studies demonstrate that managers implement normally Cross-Industry

Innovations very unsystematic because they lack practical advice. Extending the Cross-

Industry Innovation model of Gassmann and Zeschky (2008), we present a process model

which should help companies of the renewable energy industry to implement Cross-Industry

Innovations with the outside-in dimension of open innovation (see Fig 2).



Page | 343

Fig. 2: Cross-Industry Innovation process model.

The process model includes five major steps:

4.1. Strategy Analysis:

The Cross-Industry Innovation Strategy paradigm with its open and transparent character

could clash with the existing closed corporate strategy. Therefore, it is very important to

analyse the internal corporate strategy and its potentials in detail. The management should

foster open mindset. Employees should be allowed to question own products and



Page | 344

technologies and to be open for external developments and innovations in other industries

[14,15].

The competency analysis is a crucial part of the strategy analysis. The companies have to

analyse their existing technological knowledge base and competencies in order to enhance

and extend them with new solution inputs from outside the company.

4.2. Problem Abstraction:

To find adequate solutions across organisational boundaries the problem has to be analysed

in detail. This problem abstraction is one of the greatest challenges because the employees

of the company have to break down the problem into sub-problems and different functions.

This method only works if they are able to rethink their expertise and open their mindset.

Integrating the views and needs of the customers into the problem definition is also an

important success factor.

There are different abstraction methods. The functional modelling method for instance helps

to analyse and abstract the functions of a product or a problem in order to find technologies

of other industries which have similar functions.

4.3. Search for Analogies:

The search for analogies should start internally. The employees can find analogies to past

projects or experiences from hobbies, education or other areas during brainstorming

sessions. To generate radical innovations such as Cross-Industry Innovations, the

knowledge of the internal employees is most of the time not sufficient [16]. There are

numerous methods which help to create Cross-Industry Innovations. With focus on the

renewable energy industry, following three concepts seem to be appropriated - TRIZ-

database, knowledge broker and expert workshops/lead user approach.

TRIZ is a database which evaluated 2.5 million patents. Here, companies could search for

analogue solutions for their technological problems. Organisations could also work together

with knowledge broker. Knowledge broker are consulting firms and organisations which

cooperate with different industries. The knowledge base and network of these intermediaries

helps companies to get in contact with interesting partners of different industries. Companies

could also organise a workshop with experts of different industries and lead userf to search

together for analogies. A company does not have to choose one method; it could also

combine these three methods to get successful ideas.

With help of the Cross-Industry Innovation methods the company should be able to identify

an industry and in addition a partner who would like to cooperate with them. The potential

candidates should have similar strategic goals and organizational structures like the seeking

company, and the technological distance should be within acceptable boundaries. Cultural

and language issues should be also keep in mind.

f Lead users = progressive users that have high motivation to obtain a solution to their so far unmet needs.



Page | 345

In a next step (Cross-Enterprise-Idea-Development) a workshop should be conducted where

the both parties - the idea seeker and the solution provider - could generate ideas for novel

solution principles in an interactive manner.

4.4. Assessment:

It is very important to understand the structure and the functions of the analogies in order to

evaluate it. After this verification, the analogies have to be evaluated concerning their

transferability to the original problem as well as technical and commercial success factors.

The objective of this step is to receive a pool of assessed ideas where the company could

filter the knowledge pieces most relevant from them.

4.5. Adaptation:

After the Assessment phase the best identified analogous solution could be transferred to the

knowledge base and the problem of the company.

5. Conclusion

This article presents a process model which should help managers to search and implement

Cross-Industry Innovation in a systematic way. Thus, we aim to extend the model of

Gassmann and Zeschky (2008) and add a detailed search strategy. In the future this process

will be applied to companies of the renewable energy industry to get further insights of the

process and we will develop it if necessary. Authors are welcome to test this theoretical

model in other industries.

References

[1]. Lichtenthaler, U. “Open Innovation: Past research, current debates, and future directions“. Academy of Management Perspectives, 25(1), (2011) 75-93.

[2]. Stoetzel, M. et al. “Key Differentiators of Open Innovation Platforms – A Market-oriented Perspective“. Wirtschaftinformatik Proceeding. Paper 60, (2011).

[3]. Chesbrough, H. “Open Innovation: The New Imperative for Creating and Profiting from Technology“ (2003). Harvard Business School Press, Boston, MA.

[4]. Gassmann, O. “Opening up the innovation process: towards an agenda“. R&D Management 36(3), (2006) 223–228.

[5]. Laursen, K. & Slater, A. “Open for Innovation: The role of openness in explaining innovation performance among U.K. manufacturing firms“. Strategic Management Journal, 27(2) (2006) 131-150.

[6]. Enkel, E. & Gassmann, O. “Creative Imitation. Exploring the Case of Cross-Industry Innovation“. In: R&D Management 40 (2010) 256-270.

[7]. Enkel, E. & Dürmüller, C. “Cross-Industry-Innovation: Der Blick über den Gartenzaun“. In: Gassmann, O./Sutter, P. Praxiswissen Innovationsmanagement. Von der Idee zum Markterfolg. (2011) Carl Hanser Verlag: München.

[8]. Chesbrough, H. W. Open Innovation. “The new Imperative for Creating and Profiting from Technology“. (2006a) Boston MA.

[9]. Chesbrough, H. W. “Open Innovation: A New Paradigm for Understanding Industrial Innovation“. In: Chesbrough, H. W./Vanhaverbeke, W/West, J. (ed.): Open Innovation: Researching a New Paradigm. (2006b) Oxford: Oxford Univ. Press, 1-12.



Page | 346

[10]. Gassmann, O. & Enkel, E. “Towards a Theory of Open Innovation: Three Core Process Archetypes“. 2004. - R&D Management Conference (RADMA). (2004) Lisabon, Portugal.

[11]. Leimeister et al. “Theses about managing open innovation processes“. In: Jacobsen, H./Schallock, B. (ed.): Innovationsstrategien jenseits traditionellen Managements. (2010) Fraunhofer Verlag.

[12]. Reichwald, R. & Piller, F. “Interaktive Wertschöpfung: Open Innovation, Individualisierung und neue Formen der Arbeitsteilung“. (2006) Wiesbaden.

[13]. Kalogerakis, K. et al. “Generating Innovations Through Analogies. An Empirical Investigation of Knowledge Brokers“. Working paper 33 (2005) Hamburg-Harburg.

[14]. Gassmann, O. & Zeschky, M. “Opening up the Solution Space: The Role of Analogical Thinking for Breakthrough Product Innovation“. In: Creativity & Innovation Management 17 (2008) 97-106.

[15]. Ertl, M. “Strategiebildung für die Umsetzung von Open Innovation. In: Ili, S. (ed.): Open Innovation umsetzen. Prozesse, Methoden, Systeme, Kultur“. (2010) Düsseldorf.

[16]. Schild, K. et al. “How to use analogies for breakthrough innovations“. Working paper 24 (2004).



Page | 347

ENVIRONMENTAL PERFORMANCE OF ANTIMICROBIAL PAINTS ENHANCED WITH BIOCIDE SMART RELEASE TECHNOLOGIES

M. Taxiarchou a, A. Peppas a, M. Krokida b & Kyriakopoulou K. b

a National Technical University of Athens Section of Metallurgy and Materials Technology

9, Heroon Polytechneiou str., 157 80, Zographou Campus, Greece Email: [email protected] Email: [email protected]

b National Technical University of Athens

Section of Analysis, Design and Development of Processes and Systems 9, Heroon Polytechneiou str., 157 80, Zographou Campus, Greece

Email: [email protected] Email: [email protected]

Abstract

The growing demand for surface protection technologies that meet the requirements of material durability and environmental sustainability has led to the introduction of antimicrobial paints in the market. These coatings integrate a passive and uncontrolled release of biocides and fungicides, resulting in biocidal functionality between for a time period between a half and two years. The limited bio-resistance of the materials allows the growth of micro-organisms, which negatively affect health, leading to hypersensitivity and allergic reactions such as rhinitis and asthma. In this study, the environmental impacts associated with the manufacturing and the use of enhanced antimicrobial outdoor coating materials were investigated, using a Life Cycle Impact Assessment (LCIA) methodology, based on the ISO 14040 standard. A comparative analysis was carried out between conventional paints and antimicrobial paints adapting smart release mechanisms of eco-acceptable biocide affecting coating durability and prolonging service-life. The integration of a smart biocide release mechanism in surface coatings results to significant environmental benefits by prolonging the materials’ bio-protection and improving human health and quality of life.

Keywords: Anti-fungal paint, Biocide control release; Comparative Analysis; Environmental Impact Assessment; Life Cycle Assessment

1. Introduction

Weathering and deterioration of building materials can occur due to the growth of

microorganisms, such as algae, fungi, etc. The determining factors which influence the

growth of mold on building materials have proven to be moisture as well as temperature and

time (1). Therefore, the development of new durable coating materials proper for application

in humid condition is necessary. Bio-resistancy of building and finishing materials usually

requires addition of dedicated bioactive chemicals, so-called biocides. These chemicals offer

protection by providing antimicrobial action, hence reducing the micro-organisms population,

but also negatively affecting the durability of material, i.e. induce biodegradation. In addition,

biocides are also used to prevent the growth of micro-organisms (algae, fungi and bacteria)

that create unhealthy indoor environmental conditions or affect the aesthetic value.

Traditionally, the action of biocides in materials, (e.g. coatings, plasters) is based on a

passive and uncontrolled release principle, i.e. molecular dispersion of the active ingredients

in the material matrix. As a consequence these bio-active agents have a high and inherent

mobility in the matrix, which causes an initial boost in biocide activity and a steep decrease



Page | 348

when time proceeds.(2) Consequently, coatings in a built environment usually exhibit biocidal

functionality between 0.5 and 2 years, whereas the desired service life in building practice is

at least 10 years. Application of an increased biocide concentration only results in a minor

prolongation of the material service-life. Besides the environmental impact of early

replacement of such functional coatings, an inherent disadvantage of the inefficiency of the

traditional release method is the emission of a relatively large amount of biocide molecules

into the environment. In addition, the limited bio-resistance of the material will allow the

growth of micro-organisms, which negatively affect health, leading to hypersensitivity and

allergic reactions such as rhinitis and asthma. This is a prevalent problem in homes and

occupational buildings worldwide (3, 4). Moreover, growing ecological demands and

international environmental legislation increase even more the pressure on the materials’

performance. First, the application of active agents for materials’ bioresistance is regulated

by the Biocidal Product Directive 98/08/EC, which sets strict conditions with respect to the

use of bio-active agents (5). And second, the industrial trend towards eco-friendlier building

products is often accompanied with an increase in the biodegradability, requiring even more

biocides to compensate. In order the paint industry to provide a prediction on the

environmental impact of controlled release technology, a life cycle analysis was performed,

comparing different conventional and antimicrobial coating products. Life Cycle Assessment

(LCA) methodology is a suitable and valuable tool to assess the environmental impact of

materials, products and services during their life cycle and it should be part of the decision-

making process towards sustainability (6).

The objectives of this paper is to examine the environmental performance of conventional

and biocide containing materials. Several case studies have been developed in order to

evaluate the finishing materials, which implement a “smart” release mechanism consisting of

an induced response on external stimuli in finishing materials, and have an extended service-

life. The selected smart release mechanism is based on functionalized mineral (nano-clay)

particles, which reduce the amount of biocides required considerably (at least a factor of 5

compared to ordinary materials) and tune the release systems to “environmental friendlier”

biocides. As a result, the life cycle analysis of the different antimicrobial paint model systems

investigated allows the evaluation of their sustainability in relation with the achievable

environmental benefits, as a measure of ecological quality, based on our knowledge of the

influence on the environment.

2. Goal and Scope definition

4.1. Goal definition

The goal of the present LCA study is to determine the environmental footprint associated

with the production and use of antimicrobial paint, to compare conventional antimicrobial

paints in comparison with paints adopting smart release mechanisms of eco-acceptable

biocide and finally to evaluate the effect of coating durability and prolong service-life.

4.2. Scope definition



Page | 349

2.2.1 System definition, calculation and functional unit

In the studied case, the system is the set of processes that constitute the production, the

transportation between the factory and the residence and the application of the coating

material. This system includes different subsystems such as the individual activities of

gathering raw materials or the process for the production of the paint. The selected

calculation unit is one kg of coating produced in the factory, equal to the paint needed to

cover 10 m2 wall surface which is considered as the functional unit of the system.

2.2.2 System boundaries

The system includes the stages from the extraction of raw materials needed for the

production to the application of the antimicrobial paint. Secondary processes have not been

considered due to the fact that their environmental loads represent a small percentage of the

total load associated to the product under study. However, the impacts associated with the

energy production have been considered.

Fig. 1: Impact 2000+ Impact categories.



Page | 350

2.2.3 Selected impact categories, Impact assessment and methodology

The Impact Assessment methodology used in this LCA study is the Impact 2002+. According

to this methodology, the Impact Categories which are being taken into consideration are

allocated to four Damage Categories: Human Health, Ecosystem Quality, Climate Change

and Resources. In Figure 1 the Impact Categories related to each Damage Category are

presented.

2.2.4 Assumptions

For the studied cases of antimicrobial paint the following assumptions have been made:

- Water based paint was selected for this study

-The reference antimicrobial paint is considered to contain 2000 ppm biocide per 1kg of paint

(0.2% w/w) and this product is estimated to have a lifespan of 5 years.

-The antimicrobial agent used, is considered to be a non-VOC broad spectrum, highly

effective antimicrobial agent hence their emissions in the production phase of the

antimicrobial paint were considered negligible.

-In order to achieve the desired controlled release of the biocide, nanoclay particles were

considered as an encapsulant. The diffusion rate of the antimicrobial agent is depended on

its encapsulation. Therefore, paint with no biocide encapsulation is assumed to release the

whole amount of the biocide within the products life span. On the other hand, the slow

release rate provided by the nano-clay is assumed to reduce the amount of biocide released

up to 70% within the life cycle of paint (7, 8).

-Humidity and rainfall were considered to transfer the biocide to the aquifer, while an amount

of biocide is assumed to be retained by the soil.

-External walls of the building are considered to be on humid conditions several days a year,

therefore are more susceptible to mold growth.

-A typical lifespan of a building is considered to be 60 years.

-The packaging of the paint (PP pot) was taken into consideration in the use phase.

3. Life cycle inventory analysis

A product’s life cycle assessment is a framework that can be used to identify inputs, outputs

and impacts within the life time of a product. In this stage, all data which are either entering

the system (like raw materials, water and other natural inputs) or coming out to the

environment (like products, air emissions, by-products and wastes) are recorded and

quantified, based on mass balances. Therefore, the inventory includes the following steps: a)

development of the alternative paint and application case studies and mapping of the inputs

and outputs in each process and case as a whole; b) collection of the data required for the

calculation from the data inventory and c) development of an appropriate mass calculation



Page | 351

model based on the assumptions made and the data collected. The life cycle of antimicrobial

paint is pictured in Figure 2.

In this study different formulations of antimicrobial paint have been evaluated by taking into

consideration their application, their life span and their environmental impact. Therefore, a

comparative analysis was carried out between conventional paints and antimicrobial paints

adopting smart release mechanisms. In Table 1 the alternative case studies for conventional

and antimicrobial paint are presented and information about the presence of a control

release mechanism as well as the life span of each product is given. An antimicrobial paint’s

use phase describes the application, the maintenance and the replacement that take place

during its operational phase.

Fig. 2: Life cycle of paint enriched with biocide.



Page | 352

Table 1: Case studies of water based-paint.

Case study

Nanoclay concentration

on paint (% w/w) Biocide release factor (%)

Times of repainting

Product lifespan (years)

0 - - 12 5

1 - 100 12 5

2 1 30 12 5

3 0,33 100% free on paint 30% reserved on nanoclay

12 5

4 1 30 4 15

5 0,33 100% free on paint 30% reserved on nanoclay

4 15

4. Impact assessment

In order to understand the environmental importance and to estimate the possible

environmental impacts of antimicrobial paints adopting smart release mechanisms, the

effects of the environmental burdens identified in the Inventory Analysis stage have been

assessed. In this stage a connection between the recorded inputs and outputs and the

environmental impact of such antimicrobial paint has been examined. A comparison between

the studied products is made based on selected indicators. By comparing the impact

between a conventional and an antimicrobial paint, estimation on the impact of the biocide

leaching can be made. Table 2 presents the impact of conventional paint used for covering

10 m2 of wall (CS 0). The results of the assessment are being shown as the accumulated

contributions of paint application and the use and disposal to various environmental impact

categories over the life time of the building. The case studies start with the application of the

paint and as the time passes, every environmental impact corresponds to the specific time

period, is added. A 60 year time phase is chosen because it is long enough to show the long-

term consequences of paint choices, while being a realistic operation period for a typical

external wall.

Table 2: Impact of conventional paint

Damage category (Pt) CS 0 Human health 2,47E-03 Ecosystem quality 2,47E-04 Climate change 2,57E-03 Resources 4,32E-03 Total 9,61E-03

4.1. Comparison of conventional paint to antimicrobial paint

In the frameworks of this study passive mechanisms of slow-release of eco-acceptable

biocides in order the service life of finishing materials to be substantially extended, are being

examined. The slow release concept refers to slowing down the diffusion rate of the biocide.



Page | 353

In order to achieve this, a modified nano-clay surface carrying the antimicrobial agent, has

been used. The nano-clay technique is more sophisticated and aims to be built in a retarding

step before the actual diffusion of the biocide to the surface. A rapid decrease of the biocide

concentration due to fast leaching can then be avoided with the condition that at the same

initial biocide concentration the release of biocide will remain longer above the lowest

effective dose. Usually, the active compounds are bound in a poor permeable carrier system

with a high specific surface area, such as porous systems or (nano-)clay systems. This

comprises a big step forward compared to the use of regular, non encapsulated, biocide

systems. The comparison between encapsulated or not biocide systems has made the

following life cycle assessment in order to provide the necessary information for the impact

resulting from the use of such products (Table 3).

Table 3: Impact assessment of antimicrobial paint free and control release.

By comparing CS 0 to CS 1, the amount of biocide leached to the environment affects mainly

the Aquatic and Terrestrial ecotoxicity and therefore the ecosystem quality. By comparing CS

1 of free release with CS 2 of control release an improvement up to 69% in ecosystem

quality is observed due to the reduced amount of biocide leached. When comparing the

control release with a partially control release, the environmental impact is proportional to the

total biocide leached. Free release is considered to be unpredictable. Encapsulation

provides a potential environmentally acceptable and controlled mean, which prolongs

biocidal activity in coatings (9, 10). According to literature, controlled release of certain

biocides has resulted in a reduction in the threshold levels of biocides to prevent

microbiological attack. Therefore slow or controlled diffusion of the active agents results to

lower environmental impact.

4.2. Comparison of antimicrobial paints with different release rate and prolonged

lifetime

In order to provide information on how the diffusion rate of the biocide through paint lifetime

prolongation affects the environment, two alternative scenarios of CS 2 and CS 3 were

devised. By reducing the release rate of the biocide to the environment by 1/3, it is assumed

that the lifetime of the antimicrobial paint is significantly extended.. The lifespan of this paint

is considered to be 15 years, reducing the times of repaint needed. The total amount of paint

needed throughout the lifetime of the building is considered to be approximately 4.2 kg. The

characterization values of these case studies in each impact category are shown in Table 4.

Damage category (Pt) CS 1 CS 2 CS 3 %Improvement*

Human health 2,43E-03 2,41E-03 2,43E-03 0,70%

Ecosystem quality 1,98E-02 6,10E-03 1,53E-02 69,15%

Climate change 2,51E-03 2,49E-03 2,50E-03 0,84%

Resources 4,28E-03 4,25E-03 4,27E-03 0,87%

Total 2,90E-02 1,52E-02 2,45E-02 47,42%



Page | 354

All impacts were reduced thus resulting to more environmental friendly products even if the

biocide is partial encapsulated.

Table 4: Impact assessment of controlled release antimicrobial paint with extended life time.

Damage category (Pt) CS 4 CS 5

Human health 8,05E-04 8,08E-04

Ecosystem quality 2,03E-03 5,09E-03

Climate change 8,29E-04 8,34E-04

Resources 1,42E-03 1,42E-03

Total 5,08E-03 8,15E-03

Fig. 3: Total impact assessment per damage category using Impact 2002+.

4.3. Comparison of different release rates of the antimicrobial agent

The encapsulation of the biocide plays an important role on the release rate of the chemical

agent to the environment. In order to evaluate the impact of alternative impregnation

effectiveness of the materials different case studies of release rates were created. Each

case study represents a different release rate and the overall impact in the main four damage

categories is shown in figure 4. By controlling the biocide release the effects on ecosystem

quality can be reduced dramatically. Therefore the production of new impregnation materials

that can encapsulate the biocide and contribute to its slow release is essential. The reduction

of the environmental impact by controlling the release rate of the biocide combined with

increased paint durability is considered to be the key technology for the production of

environmental friendly antimicrobial paints.



Page | 355

Fig. 4: Total impact assessment of different release rates of the biocide per damage category using

Impact 2002+.

5. Conclusions

The LCA methodology can be a proper tool to achieve the study’s objectives, which are the

analysis, quantification and assessment of the environmental impact of various water based

coatings. The overall environmental performances of conventional and antimicrobial paint

have shown that the quantity of the biocide leached to the environment affects mainly the

ecosystem quality and more specifically the aquatic and the terrestrial ecotoxicity category.

In addition, the emissions compartment of the biocide plays an important role on its overall

impact and solid wastes are considered to be the most detrimental. Antimicrobial paints

adopting smart release mechanisms of eco-acceptable biocide are considered to have

reduced diffusion rate, therefore affecting coating durability and prolonging its service-life.

The integration of a smart biocide release mechanism in surface coatings results in

significant environmental benefits. Τhe impact of the biocide release can be reduced up to 3

times by prolonging the materials bio-protection and therefore minimizing the times of

repaint. Free biocide release has presented the highest impact in all damage categories,

while control release, resulting in an expanded lifespan of the paint, has shown less total

impact than conventional paint.

6. Acknowledgements

The authors gratefully acknowledge the financial support of the European Commission,

within the frame of the Seventh Framework Programme (project contract CP-TP 228504-2,

Smart release of biocides in finishing materials for the sector of construction, AXIOMA).



Page | 356

References

[1]. K.F. Nielsen, P.A. Nielsen, G. Holm. Growth of moulds on building materials under different humidities. Healthy Buildings, 3 (2000), pp. 283–288

[2]. Description of Work, AXIOMA project. Smart release of biocides in finishing materials for the sector of construction. Version: 23-09-2009.

[3]. F. Wu, D. Jacobs, C. Mitchell, J.D. Miller, M. Karol, Environmental Health Perspectives, 115 (2007) 953-957;

[4]. O.C.G. Adan, et al., Environmental Health Perspectives, 115 (2007) 983-988. [5]. Biocidal Product Directive 98/8/EC issued on 16 February 1998 concerning the placing of biocidal

products on the market. [6]. Baumann, H. and Tillman, A.-M. (2005) The Hitch Hiker’s Guide to LCA. An orientation in life

cycle assessment methodology and application. Studentlitteratur. [7]. Nordstierna, L., A. A. Abdalla, et al. (2010). "Molecular release from painted surfaces: Free and

encapsulated biocides." Progress in Organic Coatings 69(1): 45-48. [8]. Coutu, S., C. Rota, et al. (2012). "Modelling city-scale facade leaching of biocide by rainfall."

Water Research 46(11): 3525-3534. [9]. Jämsä, S., R. Mahlberg, et al. (2013). "Slow release of a biocidal agent from polymeric

microcapsules for preventing biodeterioration." Progress in Organic Coatings 76(1): 269-276. [10]. Sørensen, G., A. L. Nielsen, et al. (2010). "Controlled release of biocide from silica microparticles

in wood paint." Progress in Organic Coatings 68(4): 299-306.



Page | 357

ENERGY OPTIMIZATION PRACTICES FOR SUSTAINABLE OPERATION OF MBR WASTEWATER TREATMENT SYSTEMS

Tolkou Athanasia1, Zouboulis Anastasios1* and Samaras Petros2

1Laboratory of General and Inorganic Chemical Technology,

Department of Chemistry, Aristotle University of Thessaloniki, 54006 Thessaloniki, Greece

Email: [email protected] Email: [email protected]

2Department of Food Technology,

Alexander Technological Education Institute of Thessaloniki, 57400 Thessaloniki, GREECE Email: [email protected]

Abstract

Membrane bioreactors (MBRs) represent wastewater treatment systems integrating biological degradation and membrane filtration. Although these systems have broad benefits, their disadvantages are mainly associated to high cost involving capital cost for membrane units and operation cost corresponding to energy consumption for aeration and for high pressure gradient between the raw influent and the treated effluent. MBR energy requirements are about twice the conventional treatment methods. MBR capital cost became competitive to conventional treatment systems, due to the market availability of low cost membrane modules; however, limited efforts have been made toward reduction of operation costs. Potential processes for the reduction of energy demand in MBRs include application of primary clarification ahead of the MBR, flow equalization, solids adjustment between the aeration and the membrane basins, and pump configuration. In addition, the implementation of certain operation modes, such as air scouring of membranes, and fouling control, might contribute to low power consumption.

Keywords: Energy optimization, Sustainability, MBRs, Wastewater treatment, Membrane fouling

List of Acronyms

MBR Membrane Bioreactor CASP Conventional Activated Sludge Process

SAF-MBR Staged Anaerobic Fluidized Membrane Bioreactor System

MMV Magnetically induced Membrane Vibration

RTMBR Rotating Tubular Membrane Bioreactor

SADp Special Aeration Demand Permeate

AFMBR Anaerobic Fluidized-bed Membrane Bioreactor

COD Chemical Oxygen Demand

MFC Microbial Fuel Cells DO Diluted Oxygen

1. Introduction Membrane bioreactor systems have become a promising wastewater treatment technique

combining activated sludge and membrane separation; it is a process resulting in a high

quality effluent independent of settling characteristics of the biomass. Membrane bioreactors

have several advantages over the conventional activated sludge systems, including stable

and high effluent quality, easy operation and complete removal of bacteria. However,

membrane bioreactors may have several problems due to membrane fouling, which

therefore result to high operation and maintenance costs. Membrane fouling reduces



Page | 358

membrane permeability and therefore increases the energy consumption in a membrane

bioreactor [1].

2. Energy Consumption Energy requirements are of primary interest in MBRs and include influent supply; retentate

recycling; permeate (effluent) withdrawal and aeration. As illustrated in Figure 1, the primary

energy requirements are related to aeration (66%), while pumping is a far second energy

component (14%). Therefore, the key measures for energy reduction are focused on

aeration; however, all energy related elements should be considered in a well designed

system. In order to provide the most cost effective and energy efficient system, critical factors

should be considered during the whole life time of the system i.e. in design, operation, and

equipment [3].

Fig. 1: Energy consumption in the various processes of an MBR system [3].

Power consumption for the aeration of an MBR consists in energy for oxygen supply to the

activated sludge microorganisms and energy for membrane scouring aiming to fouling

control. Efforts toward reduction of air supply to the microorganisms are limited, as this

component is directly related to the activated sludge microfauna activities [2]. The

characteristic MBR configurations i.e. the immersed type and the side-stream configuration

may have substantial differences in aeration. Aeration in the latter case is given by fine

bubble aerators of high oxygen efficiency. However, turbulent aeration mode is achieved in

the immersed MBR systems, with significant cross-flow of the mixed liquor, resulting to

membrane surface scouring. Aeration cost in this configuration represents about 90% of the

total cost, whereas the corresponding percentage in side-stream MBRs is about 20%.

However, total energy consumption of the side-stream system is usually two orders of

magnitude higher than that of submerged systems [4].

The energy consumption by an MBR may reach up to 8 kWh/m3 although values as low as

0.14 kWh/m3 have been reported [4]; MBR energy demand for treatment of municipal

wastewater may be 2–4 times higher than the conventional activated sludge process (CASP)

[5].



Page | 359

3. Process Energy Optimization Practices Reduction of energy requirements may be achieved by adjustment of the aeration: it can be

adjusted to the minimum level required for complete nitrification. Therefore, less aeration

results to energy savings while the development of anoxic micro zones is promoted, leading

to higher nitrogen removal rates. In addition, oxygen transfer to the anoxic zone should be

negligible, reducing thus the anoxic reactor volume. As a result, these adjustments contribute

to lower equipment and energy demands, as the aerobic/anoxic sludge recirculation loop is

not required anymore.

Nitrification and denitrification processes should be implemented in one tank frequently

aerated providing aerobic and anoxic time phases, rather than using two separate tanks,. In

these systems, nitrogen reduction is achieved by aeration control reaching up to 90% with

appropriate control. The concept of the so-called intermittent denitrification has been applied

in a number of MBR installations. [6]

The emergence of submerged MBRs that utilize fairly economical polymer-based

membranes and require less energy than external MBRs has tremendous potential in large

scale-high volume throughput municipal wastewater treatment plants worldwid. The potential

of on-site reuse of the MBR effluent for washing or transport purposes offers several cost

benefits such as reduced fresh water requirements, lower sewer costs, and potential for

direct discharge to surface water. [7]

Commercial Membrane types

MBR operation is usually time-based, with constant aeration and fixed filtration sequences

(cycles), which are generally proposed by the membrane suppliers or selected according to

the operator’s experience [8]. Commercial membranes types are offered today by several

manufacturers, such as:

1. The MemPulse™ Membrane Bioreactor (MBR) System from Siemens Water

Technologies. A mechanical device is used that supplies irregular pulses of air to the

MBR module. This increases scouring effectiveness, decreases operation and

maintenance costs, and reduces energy consumption (from 5367 kWh/day in a

traditional MBR system to 2783 kWh/day). The system can be used in with a wide range

of municipal and industrial wastewater treatment applications [2]. With the Siemens

system, a combination of air and water is used to scour the membranes [3].

2. The Kubota flat sheet MBR where continuous aeration is used and the volume of air is

based on the flux, e.g., lower air scour rates are used with lower flux [3].

3. The Zenon hollow-fiber MBR, Zenon holds patents for “cyclic” air scour which cycled air

on and off in 10 second intervals. The change in scour air operation reduced their

energy requirements in the membrane tank to 0.2 kWh/m3 [3].



Page | 360

Design elements to reduce Energy

There are several locations offering the opportunity for a cost effective design such as use of

primary clarification ahead of the MBR, use of flow equalization, adjusting the balance of the

solids between the aeration basin and the membrane basin, and pump configuration (Table

1).

Table 1: Design elements to reduce Energy [3]

Primary Clarification

(1) Reduce the power requirements associated with aeration (a combination of process air and membrane scour air, with the volume of scour air often equal to or exceeding the process air requirement), and

(2) Reduce the biological tank volume.

The decrease of the organic loading of the MBR provides the potential of operating at lower MLSS concentration for a given flow rate, resulting to: (a) Decreased membrane fouling tendency, leading to longer cleaning intervals and membrane life, and (b) Increased oxygen transfer efficiency, leading to lower blower power consumption.

Flow Equalization

The combination of a reduction in the membrane surface area and the lower air scour rate results to significant energy reduction.

Balance of Solids

The MBR systems:

(1) Have been designed to operate at similar MLSS concentrations in both the aeration basins and the membrane tank.

(2) Tend to be designed using smaller process volumes and higher MLSS concentrations than conventional biological processes.

The energy reduction is twofold: (a) Reduction in pumping and (b) Potential increase in aeration, improving oxygen transfer efficiency.

Pump Configurations

The three key pumping requirements for an MBR include solids return, nutrient recycle, and permeate withdrawal. Innovative plant configurations using in-wall pumps or low head submersible pumps can minimize the energy requirements for the nutrient recycle pumps. Permeate from the membranes may be pumped or flow by gravity depending on the membrane configuration and hydraulic constraints. The optimum configuration to minimize energy is to gravity flow.

Operational elements to reduce Energy

There are various operational elements that influence the overall energy efficiency of the

MBR design. Currently the single largest energy cost is aeration – both for the biology and

for the maintenance of the membranes. Hence, opportunities to reduce aeration have the

potential to significantly reduce the overall energy requirements (Table 2). [3]



Page | 361

Table 2: Operation elements to reduce Energy.

Membrane Air Scour

Air scour represents almost the highest energy demanding process. The following techniques are used to minimize energy consumption:

(1) Intermittent air scour - based on the rotation of the membrane panels through the aerated part of the membrane tank. A combination of air and water my be used to scour the membranes, which results in a significant variation in the energy demand associated with membrane maintenance. (2) Lower air scour flow rates at lower flux - the decreased scour air operation may decrease the energy requirements in the membrane tank to 0.2 kWh/m3. Energy saving may be achieved by allowing longer rest periods between aeration periods when the flux is below the average design condition or by using continuous aeration where the volume of air is a function of the flux [3, 4].

Flux Enhancers

The addition of flux enhancers allows a wider flux operating range and has been used to demonstrate performance benefits:

(1) When the membrane quantity is driven by peak flow, the flux enhancer allows operation at a higher flux than traditionally accepted, without excessive or rapid fouling, which results in both an initial cost reduction based on the quantity of membranes installed as well as energy savings based on the reduction in overall air scour requirements. (2) When the membrane quantity is based on minimum temperature which reduces the design flux, the addition of a polymer based flux enhancer supports the operation at a more aggressive flux at a lower temperature without adverse impact on the membrane performance. By operating at a higher flux, the membrane quantity and the associated energy requirements can be reduced. [2]

Optimize Membranes in Service

Matching the number of membrane trains in service with the plant flow is an operating strategy that can reduce energy, as the membranes which are not in service do not require the same degree of air scour as those in service. Consequently, taking membrane tanks out of service when flow is low provides the opportunity to reduce the air scour requirements during the rest period. [2]

Optimize Dissolved Oxygen (DO) within the Bio-Process

Reduction of the total aeration demand in the aeration basins may be accomplished by:

(1) Operate at the minimum DO required to achieve complete treatment, and

(2) Return the solids from the membrane tank to the oxic part of the biological basins to utilize the elevated DO which can occur within the membrane tank from the air scour.

Consequently, aerobic basins could be operated with a residual DO of 1 mg/L, or less, in order to reduce aeration demands [2].

Recent models for Energy Optimization of MBRs

The technological improvements of membrane modules resulted in the production of

membranes with less energy requirements. A list of the more recent models used for energy

optimization of MBR systems are given below.

Integrated system of MFC and MBR

Microbial fuel cells (MFCs) are devices that use bacteria as catalysts to oxidize various

substrates and recover electricity. One approach to reduce the barriers and improve its

applicability is to incorporate MFC into existing wastewater treatment processes (Figure 2).

The MFC may partially offset the energy consumption in MBR process by generating

electricity, and thus enables a more sustainable wastewater treatment process. In addition,



Page | 362

MBR is more suitable to be coupled with MFC than SBR or other processes, due to the

continuous - flow operating mode. [9]

Fig. 2: Schematic of the MFC–MBR integrated system. [9]

Staged anaerobic fluidized membrane bioreactor (SAF-MBR) system

In order to reduce energy costs for membrane fouling control, a staged anaerobic fluidized

membrane bioreactor (SAF-MBR) system has been proposed, consisting in an anaerobic

fluidized-bed reactor (AFBR) followed by an anaerobic fluidized-bed membrane bioreactor

(AFMBR), as shown in Figure 3. The primary energy requirement is dedicated for recycling

the reactor liquid to fluidize the GAC (0.011 and 0.036 kWh/m3 for the AFBR and AFMBR,

respectively, resulting in a total power energy requirement of 0.047 kWh/m3. Electric energy

can be produced by combustion of the produced methane, and the net energy available for

system operation is then 0.082 kWh/m3. [10]

Fig. 3: Schematic diagram of the SAF-MBR system. [10]

Automatic control system

Automatic control system is an innovative process where the desired aeration rate is

estimated and adjusted accordingly by the information from process instrumentation. The

membrane-performance-based control system was validated at semi-industrial pilot scale

with different membrane configurations achieving a maximum energy saving of 21%, with



Page | 363

respect to the minimum aeration recommended by membrane suppliers, without visibly

interfering on membranes fouling and without affecting the biological nutrient removal [11].

Magnetically induced membrane vibration (MMV) system

A novel magnetically induced membrane vibration (MMV) system is proposed as an

alternative shear enhancement device for fouling control in MBRs. In the MMV system, a

magnetically induced vibration of the membrane is applied in order to provide shear at the

liquid membrane interface (Figure 4). The module consists in one or more membranes that

are integrated in the MMV module. The system includes a vibration driver, an electric wire, a

vibration engine and the actual vibrating module. As the vibrating device is integrated into the

membrane module, while the movement is magnetically induced, it is expected to experience

less friction, to consume less energy and to have a very flexible vibration control. The

movement orientation of the vibrating part faces the narrow face of the module in order to

both prevent the bumping of liquid onto the membrane and minimize the associated energy

loss [12].

Fig. 4: Schematic diagram of the (a) HT-MBR setup equipped with the MMV system, (b) MMV module

in front view, and (c) MMV module in side view, showing the parallel position on the multiple

membranes mounted [12].

Other Methods to Control Fouling

Fouling control that inevitably occurs in MBR operation may be achieved by implementation

of appropriate measures for the adjustment of several key parameters. The most important

strategies are concentration polarization suppression, optimization of physical and chemical

cleaning protocols, pre-treatment of feed wastewater, and mixed-liquor modification.

Fouling related to concentration polarization can be reduced either by promoting turbulence

or by reducing flux. High shear stress over the membrane surface is required for prevention



Page | 364

of fouling due to concentration polarization. However, increased membrane aeration rate is

usually expensive.

Since membrane aeration contributes significantly to the energy demand, efforts have been

focused on reducing aeration whilst maintaining membrane permeability. Progress has been

achieved in aeration efficiency by the use of new jet aeration and cyclic aeration systems. In

practice different aeration systems for biological system and for membrane fouling control

are used, aiming to efficient energy utilization for both processes [4]. The reduction of

permeate flux is always associated to low fouling rate, although more membrane modules

have to be installed resulting thus to high capital cost. Slug bubbling in flat sheet MBRs is an

energy saving bubbling regime to replace free bubbles where SADp (special aeration

demand permeate) values are reduced significantly [13].

A novel rotating tubular membrane bioreactor (RTMBR) has been employed to achieve

shear-enhanced membrane filtration. Nevertheless, the analysis of energy consumption

revealed that by increasing rotary speed to mitigate membrane fouling was much more

energy saving and efficient than increasing aeration rate. When only rotary speed was

modified to reduce membrane fouling rate (from 0 to and 10 rpm), the energy consumption

increased from 1.2 to 2.1 and to 3.0 kWh/m3 permeate respectively, and membrane fouling

rate reduced by 9.56% and 19.03%, respectively. However, when aeration rate was

increased in order to achieve same reduction in membrane fouling rate, the energy

consumption increased from 1.2 to 5.4 and to 9.6 kWh/m3 permeate, respectively. Therefore,

it can be concluded that, when an equal reduction in membrane fouling rate is achieved, the

used energy is much higher by employing aeration than rotation, suggesting that rotation is

much more efficient than aeration. However, the comparison of total energy demand in the

RTMBR and commercially available MBRs (the energy consumption of which can be as low

as 0.4 kWh/m3 product water) reveals that RTMBR does not have any advantage over the

latter systems [14].

The use of flocculants and coagulants such as aluminum or ferric chloride has been

investigated for fouling. control Furthermore, the addition of adsorbent reagents such as

powdered activated carbon (PAC) has been found to improve the membrane performance by

decreasing the level of organic compounds responsible for membrane fouling.

The cleaning protocol is mainly dictated by the desired operation net flux. Usually the

protocol suggested by the manufacturer is followed as a guideline, and the existing plants

usually work in the sub-critical regime. However, cleaning protocol has been studied

intensively by many researchers where the key parameters of interest are duration and

frequency of the cleaning and the back-flush flux. [5]

4. Conclusions The MBR technology has rapidly gained acceptance as an attractive and flexible solution to

plant expansion/enhancement as well as for greenfield facilities. Although capital costs of

MBRs have become fairly competitive to conventional treatment systems, the operating



Page | 365

costs, especially those related to energy consumption, require additional focus. The total

energy consumption by MBRs can in some cases reach values between 6 and 8 kWh/m3. In

order to provide the most cost effective and energy efficient system, it is important to explore

opportunities related to design, operations, and equipment.

There are several processes within the design of an MBR plant challenging toward a cost

effective design. These include use of primary clarification ahead of the MBR, use of flow

equalization, adjusting the balance of the solids between the aeration basin and the

membrane basins, and pump configuration.

Hand in hand with the design elements are the various operation elements that influence the

overall energy efficiency of the MBR design. Currently, the single largest energy demanding

step is aeration – both for the maintenance of healthy microfauna and for the operation of the

membranes. Hence, opportunities to reduce aeration have the potential to reduce the overall

energy requirements significantly. Key areas of focus with respect to energy reduction

include membrane scour air operation strategies, the use of flux enhancers to allow a wider

flux operating range, optimization of the number membranes in service and the oxic

operating conditions within the biological basins. Along with the operation strategies, energy

efficient equipment, specifically the aeration equipment, the blowers and the mixers must be

selected.

Fouling control in MBR operation may take place by the adjustment of several key

parameters. The most important strategies are concentration polarization suppression,

optimization of physical and chemical cleaning protocols, pre-treatment of feed wastewater,

and mixed-liquor modification.

Finally, from the standpoint of the more recent models used for energy optimization of MBRs,

the SAF-MBR system has excellent potential as a low-energy high efficiency cost-effective

wastewater treatment system. A novel magnetically induced membrane vibration (MMV)

system is proposed as an alternative shear enhancement device for fouling control in MBRs

while the MFC is a promising approach to partially offset the energy consumption in MBR

process by generating electricity, and thus enabling a more sustainable wastewater

treatment process.

Acknowledgements

The financial support through the co - Financed by the European Union and the Greek State

Program EPAN-II (OPC-II)/ ESPA (NSRF): 'SYNERGASIA II', Project (FOUL-MEM) - "New

processes for fouling control in membrane bioreactors (11SYN-8-1084), is gratefully

appreciated.



Page | 366

References

[1]. K.-G. Song, Y. Kim, K.-H. Ahn, “Effect of coagulant addition on membrane fouling and nutrient removal in a submerged membrane bioreactor”, Desalination 221 (2008) 467–474

[2]. V. Kippax: “MemPulse MBR system vs Traditional MBR system”, Asia/Pasific region, for Siemens Ltd. Astralia’s Water Technologies Business Unit, June 2011

[3]. C.L. Wallis-Lage, S.D. Levesque: “Cost Effective & Energy Efficient MBR Systems”, Black & Veatch, USA, 2007

[4]. J.A. Gil , L. Túa, A. Rueda, B. Montaño, M. Rodríguez, D. Prats, “Monitoring and analysis of the energy cost of an MBR”, Desalination 250 (2010) 997–1001

[5]. J. Radjenović, M. Matošić, I. Mijatović, M. Petrović , D. Barceló: “Membrane Bioreactor (MBR) as an Advanced Wastewater Treatment Technology”, Hdb Env. Chem Vol. 5, Part S/2 (2008): 37–101 DOI 10.1007/698_5_093, Springer-Verlag Berlin Heidelberg

[6]. M. Kraume, U. Bracklow, M. Vocks, A. Drews: “Nutrients Removal in MBRs for Municipal Wastewater Treatment”, Wat. Sci. Tech. 51 (2005), 391-402, presented at IWA Spec. Conference WEMT 2004 June 7-10

[7]. N. Cicek: “A review of membrane bioreactors and their potential application in the treatment of agricultural wastewater”, Biosystems Engineering, University of Manitoba, Winnipeg, Manitoba, Canada R3T 5V6

[8]. G. Ferrero, I. Rodrıguez-Roda, J. Comas, “Automatic control systems for submerged membrane bioreactors: A state-of-the-art review”, Water research 46 (2012) 3421-3433

[9]. Y.P. Wang, X.W. Liu, W.W. Li, F. Li, Y.K. Wang, G.P. Sheng, R. J. Zeng, H. Q. Yu, “A microbial fuel cell–membrane bioreactor integrated system for cost-effective wastewater treatment”, Applied Energy 98 (2012) 230–235

[10]. R. Yoo, J. Kim, P. L. McCarty, J. Bae, “Anaerobic treatment of municipal wastewater with a staged anaerobic fluidized membrane bioreactor (SAF-MBR) system”, Bio resource Technology 120 (2012) 133–139

[11]. G. Ferrero, H. Monclús, G. Buttiglieri, J. Comas, I. Rodriguez-Roda, “Automatic control system for energy optimization in membrane bioreactors”, Desalination 268 (2011) 276–280

[12]. M. R. Bilad, G. Mezohegyi, P. Declerck, I.F.J. Vankelecom, “Novel magnetically induced membrane vibration (MMV) for fouling control in membrane bioreactors”, Water research 46 (2012) 63-72

[13]. K. Zhang, P. Wei, M. Yao, R. W. Field, Z. Cui, “Effect of the bubbling regimes on the performance and energy cost of flat sheet MBRs”, Desalination 283 (2011) 221–226

[14]. T. Jianga, H. Zhanga, D. Gaob, F. Donga, J. Gaoa, F. Yang, “Fouling characteristics of a novel rotating tubular membrane bioreactor”, Chemical Engineering and Processing 62 (2012) 39–46.



Page | 367

THE USE OF A LOW ENERGY HYBRID MEMBRANE PROCESS FOR WATER TREATMENT

V. K. Patroklos1, E. Tsirekas2, S. Stylianou3, A. I. Zouboulis3 and P. Samaras1

1Department of Food Technology,

Alexander Technological Education Institute of Thessaloniki, 57400 Thessaloniki, GREECE Email: [email protected] Email: [email protected]

2Ergon Energia,

Technopolis ICT Business Park, Building C2, PO Box 60756, 57001 Thessaloniki, Greece


3Laboratory of General and Inorganic Chemical Technology, Department of Chemistry, Aristotle University of Thessaloniki,

54006 Thessaloniki, Greece Email: [email protected]

Abstract

As a result of the demand for high quality water there is a need for more effective, economical and energy efficient processes for the treatment of surface and groundwater, especially when water is supplied from contaminated sources. Two of the most common technologies used for water treatment are ozonation and membrane filtration. The first one, however, has high operation costs due to excess ozone required, while increased costs are required by membrane units for the achievement of appropriate pressure difference. In this paper, a hybrid ozonation - membrane filtration process is presented consisting in a membrane module for ozone supply followed by a membrane filtration unit. The effectiveness of ozonation is increased by a higher contact surface area of ozone through the generation of smaller bubbles, while preventing fouling of the membranes due to the oxidation of pollutants. Characteristic advantages of this hybrid system are the increased efficiency by decreasing the amount of required ozone for treatment, the decreased fouling and the low operational pressure (~ 3 bars). Since sustainability is a key issue of all modern operations, the energy requirements for this hybrid unit are estimated and a potential method providing power using solar panels is presented.

Keywords: Ozonation; membrane filtration; fouling; hybrid water treatment process

1. Introduction

As an essential resource for life, sustainable growth and healthy ecosystems, water has

been an important task on the European research agenda since the early years of the

Community's research, technological development Framework Programmes.

A survey of the current situation in Greece, as well as in Europe, suggests there is a real

need to find solutions to improve the management of contaminated waters. This need

becomes greater because of the tendency to decrease the statutory limits of pollutants’

concentration in the European and Greek legislation. Furthermore, there is always the need

to find innovative processing solutions, which will be more effective, and economic to operate

[1]. These solutions could be applied, where appropriate, to:

Processing of groundwater to meet the requirements of the final recipient;

Additional treatment of drinking water;



Page | 368

Water production industries;

Wastewater from industries that do not belong to industrial areas with a central

wastewater treatment facility.

Additionally, in many cases, such as irrigation water, water used by industries, landfills and

wastewater treatment plants, problems arise in the presence of the following pollutants:

Organic matter and especially persistent organic pollutants (POPs);

Suspended solids;

Metals.

Today the use of membranes in water and wastewater treatment receives special attention.

However, the fouling of the membrane pores and the resulting reduced permeability are the

biggest obstacles for a wide scale implementation of this technology. Various techniques

have been employed in order to overcome this disadvantage [1]. The most effective solution

seems to be the pretreatment of the feed in order to remove or destroy the compounds that

cause fouling.

Several scientific studies have reported methods for the pretreatment of the feed by

flocculation with FeCl3, and activated carbon adsorption [2]. Schlichter showed that a hybrid

process combining ozonation and membrane separation to remove humic acids caused a

drastic reduction in clogging of the membrane pores caused by adsorption on the surface of

organic compounds [3]. Significant research has been conducted related to the use of ozone,

not only to combat fouling, but also to improve the quality of the filtrate. Ozone allows the

destruction of persistent organic pollutants such as pesticides, and at the same time solves

the problem of residue disposal, since the pollutants are directly oxidized and not just

removed [4].

The addition of ozone for the improvement of membrane filtration process has been

investigated by certain researchers. However, as ozone is a highly reactive compound,

during the combination of ozone and membrane processes, it deteriorates the surface of

polymeric membranes and only chemically inert membranes, such as ceramic membranes,

should be applied. As most of the commercially available ceramic membranes are made of

metal oxides like alumina, titania or silica, they can offer a higher chemical, thermal and

mechanical stability than polymeric membranes. On the other hand, conventional methods of

ozone gas addition to the water phase are taking place by using bubble columns, diffusion

heads or spargers. These devices produce gas bubbles of different sizes from which ozone

is transferred into the aqueous phase. However, due to the relatively low contact surface,

only part of ozone gas introduced to the system reacts with contaminants in the water while

the non-reacted ozone is removed in the process off gas and has to be further treated to

avoid secondary pollution, resulting in addition to increased operation cost of the unit. An

alternative approach for the addition of ozone gas might be the use of a membrane

contactor, providing a high contact surface due to the formation of small gas bubbles and

favoring the complete reaction and utilization of reactants.



Page | 369

A new, laboratory-scale processing module was designed and built in order to study the

possibility of processing contaminated water through a hybrid unit, by combining two

successful processes: ozonation and microfiltration / nanofiltration. Furthermore, the ability to

use ceramic membranes was investigated aiming to ozone supply at exactly the required

dosages through very small bubbles in order to improve the process efficiency.

Operation of the hybrid system was tested in simulated groundwater and surface water

contaminated with inorganic and organic pollutants, which according to recent studies were

identified in alarming high concentrations in Greece [5]. Nevertheless, humic acids were

chosen as a representative of natural organic matter in surface waters while kaolin was used

as a source of turbidity in surface water. In addition, arsenic was used as an inorganic

pollutant.

2. Experimental setup of the ozonation-micro/nano-filtration unit

The flowchart of the hybrid unit is shown in Figure 1.The hybrid unit consists in two modules

that contain ceramic membranes connected in series. In the first module, ozone is

transferred through the small pores of the ceramic membrane to the polluted water, resulting

in the formation of micro-bubbles. In this way, the formation of larger bubbles is avoided

(bubble-less ozonation). Such a configuration is expected to result in almost complete

consumption of the ozone, without the need for destruction of any unreacted gas. Then, the

ozonated water passes through the second module where microfiltration / nanofiltration takes

place. There, the pressure difference is reversed, resulting in the removal of suspended

particles, macromolecules, organic pollutants and metal ions.

Al2O3 tubular ceramic membranes are used in each module with a pore size of 0.1 microns

and 3-5 nm. Plexiglas containers were manufactured for waterproofing the membranes and

the instrumentation. A peristaltic pump is used to transport the contaminated water under

atmospheric pressure and at various flow rates to the first module, where the ozonation takes

place. An ozone flow meter, a pressure transducer and a needle valve are used to regulate

the pressure and the flow rate of ozone produced in the ozone generator. After the first

module, an ozone analyzer is used for monitoring the dissolved ozone concentration, while

any quantity of unreacted gas is led into a trap where ozone concentration is calculated.

Finally, a membrane progressive cavity pump is used to transfer ozonated water into the

second module, where microfiltration / nanofiltration takes place. Samples of the filtrate and

the retentate were collected and analyzed for the determination of the pollutants.



Page | 370

Fig.1: Flow chart of the hybrid unit of ozonation and membrane filtration used for water treatment.

The feed water used in the treatment experiments was a simulated contaminated surface

water of medium turbidity, containing 25 mg/l of humic acid and 25 mg/L kaolin in tap water,

or a model dispersion containing the same amounts of humic acid and kaolin in deionized

water. A fresh sample was prepared before each experiment from stock solutions of kaolin

(1000 mg/L), humic acid (1000 mg/L) and tap water or deionized water. Solid humic acid

reagent was obtained from Sigma-Aldrich, whereas the kaolin (clay) was a typical

commercial available kaolin powder, giving rather stable dispersions. The rate of humic acid

removal was determined by measuring the UV absorbance of the sample at 254 nm, which is

used as an indication of organic molecules concentration; absorbance measurements took

place on a Hitachi UV-Vis spectrophotometer. pH was measured by a pH meter (Jenway,

model 3540), while turbidity was measured by a Hach Ratio/XR turbidity meter. Ozone

concentrations in the gas phase were determined according to the potassium iodide standard

method. TOC measurements were performed by a TOC-VCSH Total Organic Carbon

Analyzer (Shimadzu).



Page | 371

3. Results

3.1. Contact time of ozone and contaminated water

Preliminary experiments were conducted in order to determine the optimum contact time

between ozone and contaminated water. In the case where ozonation was followed by

nanofiltration, the optimum time was 2 min, where ozone consumption reached 97% and the

reduction of absorption at 254 nm was 100%, while TOC (total organic carbon) concentration

reduction was 35% (data not shown).

In the case where ozonation was followed by microfiltration, the optimum time was 1 min,

because further ozonation of humic acid molecules was leading to the brake down of

molecules to such a degree that microfiltration membrane was unable to further remove

TOC.

3.2. Ozonation / Nanofiltration of distilled and tap water contaminated with humic

acid and kaolin

The advantages of ozonation in combination with nanofiltration were more prominent

compared to ozonation/microfiltration (data not shown). In the hybrid system, the fouling of

the membrane pores took place at very slow rate as shown in the following Figure 2.

Fig. 2: Changes in the flow of the membrane during nanofiltration experiments.

As shown in Figure 3, the use of the hybrid unit resulted in an almost complete removal of

humic acids, expressed as the % reduction of absorbance at 254 nm. However, the removal

of Non-Purgeable Organic Carbon, was slightly decreased when the hybrid unit was used as

shown in Figure 4. It is assumed that the breakdown of large molecules into smaller allows

their passage through the pores of the membrane.



Page | 372

Fig. 3: Removal UV254 absorbance during single filtration and the hybrid process.

Fig. 4: Non-Purgeable Organic Carbon (NPOC) removal during single filtration and the hybrid

process.

3.3. Ozonation / Nanofiltration of contaminated with As(III) tap water

The hybrid unit was examined for the removal of As (III). Ozonation of the contaminated

water oxidizes As (III) to As (V), which is then easier to be removed by the nanofiltration

membrane. This was confirmed by the results shown in Figure 5. The removal of arsenic by

the hybrid unit reached 50%, while by single nanofiltration was just 10%.



Page | 373

Fig. 5: Arsenic removal during single filtration and the hybrid process.

3.4. Energy provision and consumption

The technical features of the electric appliances and devices of the hybrid water treatment

unit for the assessment of its energy consumption include:

1. Influent water feed pump

In order to operate properly, the unit requires that the velocity of water is at least 6 m/s, a set

value which is achieved by the circulation pump. The circulation pump is a stainless steel

screw pump, which nominally consumes 1,1 kW.

Moreover, there is an indicator light bulb whose consumption amounts to 0,01 kW

2. Purification of the hybrid unit membranes

The purification of the membranes is achieved by a back-flush mechanism, which utilizes

previously ozonated water in conjunction with air supply. The electricity consumption of the

system amounts to 0,3 kW.

Moreover, there is an indicator light bulb whose consumption amounts to 0,01 kW

3. Ozone generator

The generator’s energy consumption amounts to 0.135 kW.

4. Auxiliary consumption: 0.5 kW.

Therefore, the total electricity consumption of the hybrid unit is estimated to about 1,92 kW

(permeate flux is 3*10-3 m3/h for microfiltration and 6*10-4 m3/h for nanofiltration).

The estimation of the cost due to energy consumption from a low-voltage electrical grid, is

based on the following calculations:

0

10

20

30

40

50

60

70

80

90

100

0 50 100 150

% A

s re

mo

val

time (min)

10ml/min, 4 bar

10ml/min, 4 bar Hybrid



Page | 374

Unit’s operational time (h): 24

Total energy consumption per day (kWh): 24 x 1,92 = 46,08

Monthly operational time (days): 30

Average price (including all taxes and fees) (€ / kWh): 0,14

Monthly energy cost (€): 30 x 46,08 x 0.14 = 193.5

Therefore, the total power cost for this unit is estimated to 193.5 € while the corresponding

cost per m3 of the produced water is calculated to 0.73 €/m3

The energy demand of the hybrid unit can be met by energy produced from renewable

sources. These sources include photovoltaic arrays (PV), wind generators, geothermal

sources, biomass plants, etc. In order to meet the energy requirements of the hybrid unit by

an autonomous PV system, the panels’ power should be 1.52 kW. Such a system should be

accompanied by the following devices:

Regulator Charge: 30A / 24 V

Batteries: 14 x 200 Ah / 12V

Power Converter: 2500 W / 24V

4. Conclusions

Different approaches for the combination of ceramic membrane microfiltration and ozonation

in a hybrid process for surface water treatment have been presented and the following

conclusions can be drawn.

Ozonation, when used in higher concentrations, or for a longer period of time, may

mitigate the impact of membrane fouling efficiently. The combination of ozonation

with membrane filtration significantly decreases the degree of fouling (25-30%).

Ozonation is effective when used for the treatment of surface water contaminated by

humic acids and As(III).

The hybrid unit successfully increased the removal of As, by oxidizing As(III) to As(V).

Acknowledgments

The financial support through the co- Financed by the European Union and the Greek State

programme EPAN-II/ ESPA: 'SYNERGASIA' Project NanoMemWater, (09SYN-42-440) is

gratefully appreciated.

References

[1]. Peleka, E. N., Mavros, P., Zouboulis, A. I., Matis, K. A. (2009), A hybrid flotation-microfiltation cell for effluent treatment. Desalination 248 (2009) 881-890.

[2]. Hilal N., Ogunbiyi O., Miles N. & Nigmatullin R., Separation Sci. and Technol.(2005) 1957-2005.



Page | 375

[3]. Shon H.K, Vigneswaran I.S, Kim I.S, Cho and Ngo H.H, Membrane Science, 234 (2004) 111-120.

[4]. Schlichter B.V, and Mavrov H.C. Desalination, 156 (2003) 256-257 .

[5]. Benitez J., Acero J, Real F and Garcia C., 2009. “Combination of chemical oxidation-membrane filtration processes for the elimination of phenyl-ureas in watermatrices”. Wiley Interscience, 84 1883–1893

[6]. National Technical University of Athens (NTUA), School of Civil Engineering, Department of Water Resources and Environment, National Program Management and Protection of Water Resources, supported the training of National Program Management and Protection of Water Resources, Athens, 2008.



Page | 376



Page | 377

EXERGY ANALYSIS OF THE POWER MODE OPERATION OF A BIMODAL NUCLEAR THERMAL ROCKET FOR THE NERVA

CONCEPT

E. Skordilis a, L. Benos a, M.M. Ioanniti a, K. Pavlou a & Stamatis A. a

a University of Thessaly

Department of Mechanical Engineering, Leoforos Athinon, Pedion Areos, 38334 Volos, Greece

Email: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]

Abstract

In 2004, NASA announced a rocket program named NERVA (Nuclear Engine for Rocket Vehicle Application) in order to develop a thermal nuclear propulsion system for use on long-range manned space missions to Mars and other destinations. Apart from crew`s safety, the development of the power and propulsion systems is considered as the most significant feature of these models. A power system that exploits the same nuclear reactor for both propulsion and generation of consistent electric power could weigh less than a separate one. Therefore, a Bimodal Nuclear Thermal Rocket (BNTR) system is regarded to be the most suitable option. The purpose of this paper is to make an exergy analysis of a closed Brayton cycle (CBC) which is chosen for the power mode of the BNTR. The model describes entirely the closed Brayton cycle from a component-level operation, where each component is presented along with each of the thermodynamical components’ properties. More specifically, the entire Brayton cycle was separated into different sections, where each section represents a state. At these states, calculations so as to determine some thermodynamical properties such as pressure, temperature, enthalpy, entropy and physical and chemical exergies were conducted. These properties were then used for determining the total output of the power system. Furthermore, exergy destruction is exhibited for each component of the aforementioned cycle. Finally, the exergy efficiencies for all the components of the CBC are calculated and presented.

Keywords: Exergy Analysis; Exergy Destruction; Exergy efficiency;

List of Symbols

Prt: Turbine pressure ratio xHe: Percentage of Helium

Prc: Compressor pressure ratio xXe: Percentage of Xenon

Cp: Specific heat capacity (kJ/kgK) si: Entropy of ith state Phex_i: Physical exergy of ith state (kW) pi: Entropy of ith state

Chex: Chemical exergy (kJ/kg) im : Mass flow of ith state (kg/s)

Extotal i: Total exergy of ith state (kW) Ti: Temperature of ith state (K) Chex_He: Molar chemical exergy of Helium (kJ/mol) Tsink: Sink temperature (K)

Chex_Xe: Molar chemical exergy of Xenon (kJ/mol) si: Entropy of ith state

Ex_destj : Exergy destruction for the jth component (kW)

h: Enthalpy of ith state (kJ/kg)

Pj: Power of the jth component (kW) Qj: Heat flux of the jth component (kW)

0: reference state index, i.e. state 8 Greek symbols

γ: The heat capacity ratio Cp/Cv ηalt: Alternator efficiency

ηex-j: Exergy efficiency for the jth component ABHe: Atomic mass of Helium (kg/mol) ηt: Turbine efficiency ABHe: Atomic mass of Xenon (kg/mol) ηc: Compressor efficiency



Page | 378

Subscripts He: Helium i, j: states of the cycle Xe: Xenon

c: compressor ex: Exergy t: turbine

1. Introduction Since the last Moon landing, over 40 years have passed and many people have been

disappointed with the lack of progress in continuing the Apollo journey. To address this, and

the apparent lack of direction with NASA, the U.S. Government launched the Vision for

Space Exploration (VSE). This plan set out a 2020 timetable for the return of humans back to

the Moon. The most exciting aspect of the Vision, however, is the planning of landing

humans on Mars by the year 2030.

The use of nuclear energy for spacecraft propulsion was first proposed in 1945 by Von

Karmon. Several individual investigations were carried out throughout United States of

America that eventually lead to the government sponsored Rover and NERVA Projects [1, 2]

in 1955. However, in 1958, the newly created NASA inherited the Air Force responsibilities

and the proposed engine became slated for use in advanced, long-haul space missions to

the Moon and Mars. Between 1959 and 1972, the Space Nuclear Propulsion Office oversaw

23 reactor tests. In the late 1960's and early 1970's, the Nixon Administration cut Rover and

NERVA funding dramatically. Eventually, the NERVA and Rover programs lost their funding,

and were terminated on January 5, 1973. An elaborate literature review of these projects was

made by [2]. Interest and research in NTR propulsion were continued to increase since the

early 1990's based on the technical success that the previous programs had.

Nuclear fission in the form of nuclear thermal propulsion provides twice the fuel efficiency of

chemical propulsion, which can lead to lighter and less expensive vehicles, because of the

lower fuel/oxidizer mass ratio. A second and very important feature of the nuclear thermal

propulsion is that the spacecraft’s power system utilizes the same nuclear reactor as the

propulsion system, which can provide consistent electric power regardless of the distance

from the Sun, thus reducing the weight of a separate, independent power system. For these

reasons, a system which uses a common reactor for propulsion and power generation,

formally known as a Bimodal Nuclear Thermal Rocket (BNTR) system, is considered to be

the most suitable engine for manned Mars exploration.

The objective of this work is an elaborate study of the exergy of the present BNTR power

system, i.e. an attempt to find out the maximum work which is available in this reversible

thermodynamic process, if a material changes from an initial state to a secondary state that

is in equilibrium with the environment. Throughout the research regarding the power

production at nuclear powered space vehicles and stations, a wealth of information and

studies revealed the extended work done by many researchers around the world either at

universities or at space agencies (NASA, ESA, RSA and so on). However, there was virtually

no research considering thermodynamics and exergy analysis, thus leaving a scientific gap

to this area. If the BNTR is to be utilized in the near future, then the results of this paper may

be useful to demonstrate a solid approach to the exergy analysis of such systems.



Page | 379

2. Bimodal Nuclear Thermal Rocket

BNTR systems take advantage of the nuclear reactor throughout the entire mission by

operating it in two separate modes: a high-power propulsion mode and a secondary, low-

power, power generation mode. This work focuses on the second mode which consists of a

closed loop on which a reactor serves as a heat source (in other words as a furnace). In

addition, the inert gas mixture He-Xe is pumped through the reactor’s fuel elements, which

are called tie tubes, flowing through this closed loop Brayton cycle. Helium and Xenon is

used instead of Hydrogen, in order to prevent the corrosion of the reactor. In a nutshell, the

components of the so-called Brayton Cycle are briefly described next.

First of all, the heat of the reactor comes from the fission of U-235, which is contained in the

tie tubes. These are made from a composite of UC-ZrC and graphite. The reactor contains

over 500 tie tubes, which corresponds to approximately 10,000 channels. In other words, the

heat transfer to the propellant is very effective [3].

On the other hand, there is a turbine that powers the compressor and the alternator (which

produces the electrical power) and a recuperator which has been added in order to improve

the cycle efficiency by pre-heating the reactor inlet flow with the turbine exhaust. Moreover, a

heat pipe manifold exists so as to exhaust the excess heat from the Brayton cycle to an

external radiator, which radiates the excess heat into space. This is accomplished through a

loop that uses NaK, a liquid metal fluid.

Finally, there are seven ducts that connect the aforementioned components. DUCT1

connects the reactor with the turbine. DUCT2 represents the duct that links the turbine with

the recuperator (hot stream). DUCT3 connects recuperator and heat pipe manifold. DUCT4

links the heat pipe manifold with the compressor. DUCT6 connects the compressor and the

recuperator (cold stream). Last but not least, DUCT7 closes the He-Xe loop by connecting

the turbine exhaust to the reactor. An extra DUCT5 was added to the output of the

compressor and before the decrease of mass flow due to lubrication of the shaft that

connects the compressor and the turbine. The analysis of DUCT5 was neglected because of

its minor effect in the final results. For the purpose of calculating the pressure drop across

the above mentioned ducts we consider that it is proportional to the square power of the flow

rate for each one of them (Eq. 1). In addition to this, taking into account the estimations of

the PhD thesis of J. Clough [4], Tout was approximated from Tin from the Eq. 2, where the

indices “in” and “out” are referred to the input and output of each duct respectively.

2

pP K m

(1)

out T inT K T

(2)

Briefly, the He-Xe working fluid is heated in the channels which are located in the reactor.

Then, it is led towards a turbine where it is expanded in order to enter a recuperator and

eventually to be cooled through the heat pipe manifold. Afterwards, the air mixture moves

into a compressor so as to be compressed and to be driven again to the recuperator in which



Page | 380

it is heated up and then re-enter the reactor. The above process is illustrated in Figure 1.

Note that there is an increase and a decrease on the mass flow before the input and after the

output of the compressor, respectively. On the other hand, as far as the turbine is concerned,

there is an increase of the mass flow in the output. The reason of this fluctuation is both the

gas bearing lubrication and shaft and alternator cooling. The other part, which deals with the

propulsion of the rocket, is beyond the scope of this work but someone who is interested in it,

could refer to [4] and several related papers such as [5] in the literature.

Fig. 5: Schematic of the closed Brayton cycle of a BNTR [5]

3. Exergy Analysis

According to popular belief, all real processes involve energy losses. However, such an

interpretation is not satisfactory from the logical point of view, since energy is bound by the

law of conservation, so it cannot be destroyed. A deeper analysis reveals that in real

processes energy is not destroyed, but rather transformed into other forms, less suitable for

feeding and driving real processes. Hence, besides energy, another physical quantity should

be introduced to characterize the quality of the kind of energy under consideration. The

ability to perform useful work in a natural environment has been suggested and investigated

as a measure of energy quality by Gibbs, A. Stodola, G. Gouy, J. Keenan, F. Bosnjakovic

and many other researchers. Z. Rant in 1956 proposed the term exergy, which has since

been broadly accepted. Exergy analysis is based upon the second law of thermodynamics,

which stipulates that all macroscopic processes are irreversible. Every such irreversible

process entails a non-recoverable loss of exergy, expressed as the product of the ambient

temperature and the entropy generated (the sum of the values of the entropy increase for all

the bodies taking part in the process). Some of the components of entropy generation can be



Page | 381

negative, but the sum is always positive. The analysis of exergy losses identifies possibilities

for improving thermal processes and helps select a rational scheme for thermal systems.

3.1 System Modeling and mathematical formulation

At first glance, we put as known variables the values of temperature, pressure and mass flow in the entrance of the reactor, T8, P8 and 8m , similar to J. Clough’s work [4]. In addition, the

ambient temperature, Tsink, was considered 200 K and the efficiencies of the turbine,

compressor and alternator 0.9081, 0.8201 and 0.95, respectively. The inert gas mixture He-

Xe relative rate is in 1:1 and their molar chemical exergy was 30.31 and 40.27 KJ/mol,

respectively [7]. Finally, the specific heat capacity of the working fluid was 0.34 kJ/kgK. The

model was worked out in MATLAB programming environment.

The physical exergy of each state and the chemical exergy of the He-Xe mixture were

calculated as demonstrated in equations 3a and 3b, while the total one as the summation of

physical and chemical exergy (Eq. 4).

_ 0 0 0ex i i i iPh m h h T s s (3a)

_ sin_ _ ( log( ) log( ))He He Xe Xe He Xe k He He Xe Xeex

He He Xe Xe

x Ch ex x Ch ex R T x x x xCh

x AB x AB

(3b)

_ _ _total i ex i i ex iEx Ph m Ch (4)

Enthalpy and entropy were deduced from equations 5 and 6, respectively. The values of Cp

and RHe-Xe were taken from [6].

0 0i p ih h C T T (5)

0 _0 0

log logi ii p He Xe

T ps s C RT p

(6)

The temperature of the turbine output was derived from equations 7 and 8. T`output represents

the temperature in the case of an isentropic procedure. The turbine pressure ratio was

initially regarded as 1.812, i.e. the value in the aforementioned PhD [4].

1

1 1` `1

Pr

`

Pr

output output ouput

input input t input

inputoutput

t

p T T

p T T

TT

(7)

``

input outputt output input t input output

input output

T TT T T T

T T

(8)



Page | 382

On the other hand, as far as the compressor is concerned, the isentropic and the compressor

output temperatures were estimated from equations 9 and 10, respectively.

1

` Proutput input cT T

(9)

`` output inputoutput inputc output input

output input c

T TT TT T

T T

(10)

As for the recuperator, the output pressures were again proportional to the squared power

of mass flow. Thus, Toutput(hot) was deduced from Eq. 11. The heat pipe manifold was also

considered as a heat exchanger, so the output temperature was deduced from equation

12.

( ) ( )_ _

( ) ( )

( )recuperator output hot input hotT hot T hot

p input hot p input hot

Q UA T TK K

m C T m C T

(11)

_ _

( )out inT heatpipemanifold T heatpipemanifold

in in

Q m Cp T TK K

m Cp T m Cp T

(12)

The nuclear reactor of the system was regarded as a combustion chamber. Therefore, Eq.

13 shows the output temperature of the working fluid, while the output pressure was directly

proportional to input pressure.

_ _

( )p output inputreactorT reactor T reactor

p input p input

m C T TQK K

m C T m C T

(13)

3.2 Post-processing

The post-processing of the results presented above lead to the calculation of the works on

the compressor, turbine and power generator from the equations 14, 15 and 16, respectively.

compressor cP m h

(14)

turbine tP m h

(15)

total compressor turbine altP P P

(16)

The values of the exergy destruction for each individual process of the BNTR are presented

below. Equations 17 to 22 show the exergy destruction at the compressor, recuperator,



Page | 383

reactor, turbine, heat pipe manifold and each duct respectively. The total exergy destruction

is the summation of the aforementioned values.

_ _ _compressor total input total output compressor cEx dest Ex Ex P (17)

( ) ( ) ( ) ( )_ _ _ _ _recuperator total input cold total input hot total output cold total output hotEx dest Ex Ex Ex Ex

(18)

0_ _ _ (1 )reactor total input total output reactorout

TEx dest Ex Ex QT (19)

_ _ _turbine total input total output turbine tEx dest Ex Ex P (20)

0_ _ _ (1 )heat pipemanifold total input total output heat pipemanifoldout

TEx dest Ex Ex QT (21)

_ _ _duct total input total outputEx dest Ex Ex (22)

Finally, the exergy efficiency of each component and the total exergy efficiency of the system

are calculated from equations 23 to 28.

_ __

total output total inputexergy compressor

compressor

Ex Ex

P

(23)

__ _

turbineexergy turbine

total input total output

P

Ex Ex

(24)

_ ( ) _ ( )_

_ ( ) _ ( )

total input cold total output coldexergy recuperator

total input hot total output hot

Ex Ex

Ex Ex

(25)

__

_

total outputexergy reactor

total input reactor

Ex

Ex Q

(26)

__

_

total output heatpipemanifoldexergy heatpipemanifold

total input

Ex Q

Ex

(27)

_total

exergy totalreactor

P

Q (28)

4. Results and discussion

In this section, the results aroused from this analysis are represented. The results include the

values of all the variables encountered in the previous chapter.



Page | 384

The values of temperature, pressure and mass flow for each state are summarized on Table

1. It is indicated that the highest temperature values are in the input and output of the

reactor, whereas the lowest ones located in the input and output of the heat pipe manifold, as

expected. The relative error related to J. Clough’s results [4] is below 0.04% and below 0.4%

concerning the values of temperature and pressure, respectively. Thus, overall, the obtained

results were in good agreement with those of [4]. The values of the mass flow were regarded

to be the same as [4].

Table 1: Temperature, Pressure and Mass Flow values for each state

State Temperature [K] Pressure [kPa] Mass Flow [kg/s]

Present Study J. Clough [4]

Error [%] Present Study J. Clough [4]

Error [%]

8 911.810 911.810 0 1023.970 1023.970 0 1.201 9 1284.000 1284.000 0 1018.290 1018.290 0 1.201

10 1280.280 1280.280 0 1015.060 1015.060 0 1.201 12 1034.250 1034.170 -0.008 560.188 560.160 -0.005 1.224 13 1022.177 1022.100 -0.008 557.949 557.920 -0.005 1.224 14 691.046 690.890 -0.023 557.949 557.900 -0.009 1.224 15 691.046 690.890 -0.023 557.919 555.870 -0.368 1.224 16 426.156 426.000 -0.037 546.806 544.750 -0.377 1.224 1 426.156 426.000 -0.037 544.997 542.940 -0.379 1.256

17 426.156 426.000 -0.037 544.998 542.940 -0.379 1.224 5 578.260 578.090 -0.024 1035.495 1031.580 -0.379 1.201 6 578.260 578.090 -0.024 1033.305 1029.390 -0.380 1.201 7 915.81 915.540 -0.021 1030.915 1027.000 -0.381 1.201

Furthermore, Table 2 shows the values for the exergy destruction on each component. The

maximum exergy destruction appeared to the heat pipe manifold, as expected, because of

the significant gradients of the temperature due to the loss of the abundant heat to outer

space. That is to say that the entropy generation increases and thus the available work

decreases.

Table 2: Exergy destruction and Percentage on each component

Component Exergy destruction [kW] Exergy destruction percentage [%]Compressor 12.408 7.323

Turbine 11.222 6.623 Recuperator 4.976 2.936

Heat Pipe Manifold 129.131 76.208 Reactor 4.464 2.634 DUCT1 1.385 0.817 DUCT2 4.178 2.465 DUCT3 0.002 0.001 DUCT4 0.110 0.065 DUCT5 0.000 0.000 DUCT6 0.069 0.040 DUCT7 1.496 0.883

Total exergy destruction 169.445 100.00



Page | 385

5. Conclusions A study was presented of the exergy analysis of the power mode of a Bimodal Nuclear

Thermal Rocket. Specifically, the power mode was demonstrated as a closed Brayton cycle,

which uses the nuclear reactor of the rocket as a heat source. The exergy analysis has

shown that the exergy destruction on the compressor and the turbine is noteworthy, but this

is prudent since there is a significant drop and increase of the temperature at the turbine and

the compressor respectively. There is also considerable exergy destruction at DUCT 2,

compared to the other ducts, again due to the drop of temperature along it, which suggests

that this is the longer and more flexuous duct in the cycle. Finally, as mentioned above, the

exergy destruction is remarkably greater at the heat pipe manifold, which constitutes the 76

% of the total exergy destruction of the cycle.

References

[1]. [1]. D. R. Koenig. “Gained from the Space Nuclear Rocket Program (Rover)”, Tech. Rep. LA-10062-H. Los Alamos National Laboratory, Los Alamos, (1986).

[2]. [2]. R. W. Bussard. “Powered Rockets: A Historical Survey and Literature Summary”, The LASL Nuclear Rocket Propulsion Program (edited by R. E. Schreiber), Los Alamos Scientific Laboratory, (1956), 32-49.

[3]. [3]. F. P. Durham. “Nuclear Engine Definition Study Preliminary Report - Engine Description”, Tech. Rep. LA-5044-MS, Los Alamos Scientific Laboratory, Los Alamos, (1972), vol. 1.

[4]. [4]. J. Clough. “Integrated Propulsion and Power Modeling for Bimodal Nuclear Thermal Rockets”, (PhD Thesis, University of Maryland, U.S.A, 2007).

[5]. [5]. S. P Fusselman, S. K. Borowski, P. E. Frye, S. V. Gunn & C. Q. Morrison, “NERVA-Derived Concept for a Bimodal Nuclear Thermal Rocket”, Space Technology and Applications International Forum, (2005).

[6]. [6]. J. Tarlecki, N. Lior & N. Zhang, ”Analysis of thermal cycles and working fluids for power generation in space”, Energy Conversion & Management, 48 (2007) 2864–2878.

[7]. [7]. R. Rivero & M. Garfias, “Standard Chemical Exergy Of Elements updated”, Energy Vol. 31, 15 (2006) 3310–3326.



Page | 386



Page | 387

LUMINOUS RECYCLING

Rodrigo Antonio Vasquez Paredes a, Fidel Franco b, Jaume Roset b

a Universidad Politécnica de Cataluña (UPC)

Departamento de Construcciones Arquitectónicas I Universidad Politécnica de Cataluña

Diagonal 649 08028 Barcelona


b, Universidad Politécnica de Cataluña (UPC) Departamento de Física Aplicada ETS Arquitectura de Barcelona

Diagonal 649 08028 Barcelona

Email: [email protected]; [email protected]

Abstract

Nowadays architecture considers daylight as an element of spatial perception, but not much as an evaluable luminous energy, however, usual software allow to calculate with enough precision illuminances at interior spaces.

Using calculation, we establish compatibility between daylight and artificial lighting reducing energetic expenses. Taking calculations of daylight as a basic energy, we redesign electric systems and we propose possible architectural solutions making artificial lighting the precise complement.

As practical example, we have studied, in Barcelona, a building at UPC where we have checked the feasibility of recycling the present building by using daylight as luminous energy and diminishing about 30% expenses (consumption, maintenance and luminaries).

Even with the important variability of solar energy in Mediterranean countries, calculations allow determining the worst points along to the year and to work in order not to let never interior rooms below normative illuminations.

Keywords: Luminous Recycling, Energetic Compatibility, Energy Savings

1. Introduction

In order to make comprehensive the concept of luminous recycling in a building, we have

done a comparative between a nowadays state at the level of natural illumination

(daylighting) and artificial illumination in buildings.

We have proceeded to a winter analysis considering a cloudy day with partial rain where

natural illumination is low. As far as we know, this problematic situation has not been

analyzed before.

2. Objectives

We shall develop a detailed calculation in order to establish more specific ranks, later we will

propose solutions in order to save energy in such spaces, spaces have chosen with the most

possible formal diversity in our University (UPC). Specific objectives are to calculate daylight

and artificial light for each space at all the seasons in the year. In this way, we will be able to



Page | 388

pose some hypothesis of energetic compatibility in order to optimize energetic consumption

for lighting.

3. Methodology

Illuminances (luxes) of the spaces have been 1 meter above the terrain (fig.1), with a

multimeter (fig.2)

(A) By the façade

(B) 1 meter from the façade.

(C) 3 meters from the façade.

Fig.1 Graphic scheme of measurements Fig.2 Multimeter used for measuring luxes

Measures are taken with artificial light on and off. We show photographies taken in-situ

3.1. Case 1 Building 1 (+41° 23' 0.55", +2° 6' 56.42")

Measures are taken with a grill closed (fig.3) and with the same grill open (fig.4)

(A) 50 lx. / 269 lx.

(B) 38 lx. / 103 lx.

(C) 82 lx. / 89 lx.

12:30 a.m. / 14.02.2011

Fig.3-4 Interior views

3.2. Case 2 Building 1

Two measures with the grill closed (fig.5-6)

(A) 54 lx. / 227 lx.

(B) 40 lx. / 167 lx.

(C) 35 lx. / 91 lx.

12: 41 a.m. / 14.02.2011

Fig.5 Detail of window with exterior grill. Fig.6 Interior view



Page | 389

For thermal reasons, window is also sealed. Luminous difference is important.

3.3. Case 3 Building 2 (+41° 23' 3.99", +2° 6' 49.60")

Space has an exterior grill that blocks direct sun light entrance.

(A) 263 lx. / 268 lx.

(B) 220 lx. / 200 lx.

(C) 174 lx. / 138 lx.

11: 14 a.m./ 14.02.2011

Fig.7 Interior view

3.4. Case 4 Building 3 (+41° 23' 6.07", +2° 6' 55.99")

We can appreciate a very little variation in results.

(A) 781 lx. / 681 lx.

(B) 301 lx. / 290 lx.

(C) 402 lx. / 360 lx.

12:12 a.m./ 14.02.2011

Fig.8 Interior view

3.5. Case 5 Building 4 4 (+41° 23' 1.53", +2° 6' 46.20")

(A) 2499 lx. / 2430 lx.

(B) 825 lx. / 806 lx.

(C) 200 lx. / 170 lx.

12:02 a.m./ 14.02.2011

Fig.9 Interior view

3.6. Case 6 Building 4

(A) 2499 lx. / 2430 lx.

(B) 825 lx. / 806 lx.

(C) 200 lx. / 170 lx.

12:02 a.m./14.02.2011

Fig.10 Interior view



Page | 390

Due to different orientation, we can observe changes, in these two last cases, respect to the

previous example. So, as each façade has a different performance, we are forced to

particularize our calculations.

3.7. Measurements Analysis

From previous measurements we can deduce that daylight is bigger than artificial light for all

chosen cases.

These types of solutions vary following space typology; it can be seen in previous cases.

Some existent architectural elements are revealed as obsolete: Instead of contributing to

luminous comfort, they do not and provoke a growing of energetic consumption.

4. Building 5 (+41° 23' 21.74", +2° 6' 47.79")

Building 5 has a multiplicity of illumination manners (Fig.11-14)

Fig.11,12,14 Render: Isometric views: West, East and South. Fig.13 Photo: Isometric view: North

4.1 Daylight calculations

All rooms of the building are calculated (see 3rd stage in figures 15 to 19)

Fig.15 Isolines Fig.16 Graphic of values



Page | 391

Fig.17 Render of the plan Fig.18 and 19 Render with False Colors and Table.

Artificial lighting in each room of building 5 is also calculated. Measures are shown in lux at a

work plan height of 0.85 m

4.2 Daylight calculations during one year

Same calculation at a height of 0.85 m.

Fig.20 Isolines Fig.21 Graphic of values



Page | 392

Fig.22 Render of the plan Fig.23 and 24 Render with False Colors and Tablea.

To obtain more precise results, we have calculated this building at the four seasons (spring,

summer, fall and Winter) trying to ascertain the answer of the different spaces and actual

possibilities of compatibility for each space. Calculations have done at 12:00 hrs for all the

days in the year.

4.3 Daylight calculations during working time

Use of the building is very complex. A multiplicity of activities is developed in: administrative,

docent, research, and so on.

Each space has, in fact, its own timetable.

Fig.25 Isolines at 8:00 a.m. Fig.26 Graphic of values at 8:00 a.m.

Calculation was done at the considered worst day (in work timetable) of the year. Work

timetable was estimated between 8.00 a.m. and 17.00 p.m.

4.4 Electric consumption at Building 5



Page | 393

Fig.27 Yearly consumption Fig.28 Historical consumption

Compared with other similar buildings, we have less consumption: 74 kWh/m2 instead of 114

kWh/m2 as a mean for UPC

Implementing this study, exact values for diminution of artificial lighting consumption can be

determined for building 5. This system does not consider implementing nor luminous sensors

neither other technologies, so we want to remind that personal and social factors have

always an important degree of variability

5. Conclusions

Results of done calculations show the feasibility of reducing electric consumption by

compatibility of daylight with artificial light.

This methodology can be implemented to any other building.

Energy savings can be between 15% and 40% of total yearly luminous energy consumption.

5.1 Applications

Implementation of the system is rather economic and can diminish in a significant manner

environmental impact. Its main characteristics are:

Make the calculations by qualified staff

Minimal modification of electric circuit systems, using new interrupters

Capacitating the staff.

All these expenses could be absorbed in a reasonable amount of time from the moment of

the starting up of the system.

One of the main factors is capacitating all the staff because are ALL the workers that have to

be concerned in the matter of saving as much energy as possible.

In case of not capacitating executive staff, the rest of staff could have some conflicts in the

orders received giving as a result that the system will be not efficiently incorporated.

Façades modification, provided we do not alter the architectural language of the building,

would improve more the entrance of daylight to the different rooms.

5.2 Social Benefits

The most important social benefit of luminous recycling is that it can be applied to any social

actual circumstances and all types of buildings (dwellings, offices, hospitals, museums,

schools, etc.). As all type of buildings are involved, we can talk about global benefit.



Page | 394

One successful actual implementation in a building will bring a chain reaction that would

produce a general reduction in energetic consumption for existent and future buildings.

References

[1]. Energy Efficient Lighting in Offices. Commission of the European Communities.

[2]. Guía de Iluminación Interior. Informe técnico. Publicación de la CIE nº 29.2 (1986)

[3]. La iluminación en los lugares de trabajo. INSHT (1994).

[4]. Guía de iluminación de interiores. Publicación CIE nº 29.2, 1975.

[5]. Norma ISO 8995, 1989. Principios de ergonomía visual. La iluminación en los sistemas de trabajo en interiores.

[6]. Norma UNE 72-112-85. Tareas visuales, clasificación. Asociación Española de Normalización.

[7]. Norma UNE 72-163-84. Niveles de iluminación. Asignación a tareas visuales. Asociación Española de Normalización.

[8]. http://www.upc.edu/sirena, Website about energy consumption at UPC

[9]. http://www.dial.de/DIAL/, Website of the software used in calculations.



Page | 395

FUTURE OF SOLAR COLLECTORS: TILT ANGLE OPTIMIZATION FOR MAXIMUM PERFORMANCE

Abdulkadir Kocera & Afsin Gungorb

aAkdeniz University Vocational School of Technical Sciences,

07058 Antalya, Turkey Email: [email protected]

bAkdeniz University

Faculty of Engineering Department of Mechanical Engineering, Corresponding author.

Tel.: +90 532 397 30 88; Fax: +90 242 310 63 06.

E-mail address: [email protected] Abstract

Solar energy technologies offer a clean, renewable and domestic energy source, and are essential components of a sustainable energy future. One of the most important parameters that affect the performance of a solar collector is its tilt angle with the horizontal. In this study, the optimum tilt angle for the south facing single axis solar collector has been determined in order to maximize the system performance. The results show that the optimum tilt angle changed between 1 (June) and 65 (December) throughout year in Antalya, Turkey. The loss in the amount of collected energy is around 1 % if the angle of tilt is adjusted seasonally instead of using βopt for each month of the year. The loss of energy when using the yearly average fixed angle is around 7 % compared with the monthly optimum tilt.

Keywords: Solar energy; Alternative energy; Tilt angle; South facing

List of Symbols

Ø Latitude

β Tilt angle

δ Declination of the sun

H Radiation on a horizontal surface

Ho Extraterrestrial radiation

HD Daily diffuse radiation

z Altitude (42 m for Antalya)

ω Sunrise or sunset angle

Gsc Solar constant (1367 W/m2)

HT Radiation on tilted surface

ρ Ground reflectance (≈ 0.2)

Rb Ratio of extraterrestrial radiation

1. Introduction

The sun’s total energy output is 3.8×1020 MW which is equal to 63 MW m2 of the sun’s

surface. This energy radiates outwards in all directions. Only a tiny fraction, 1.7×1014 kW of

the total radiation emitted is intercepted by the earth [1]. However, even with this small

fraction it is estimated that 30 min of solar radiation falling on earth is equal to the world

energy demand for one year [2]. Solar radiation data is usually measured in the form of

global and diffuse radiation on a horizontal surface at the latitude of interest. Flat-plate solar

collectors are tilted so that they capture the maximum radiation and the problem of

calculating solar radiation on a tilted surface is in determining the relative amount of beam



Page | 396

and diffuse radiation contained in the measured horizontal global radiation. The optimum tilt

of solar collector with respect to user is an important subject from application of thermal /

electrical energy point of view. By utilizing maximum solar energy through the optimum tilt,

we are able to harness the energy needed without polluting our environment. It reduces the

CO2 emissions in the atmosphere which is a major culprit for Global warming. By reducing

CO2 emissions in the atmosphere, carbon credit can also be earned which is an international

issue now a day [3].

The performance of a solar collector is highly influenced by its orientation and its angle of tilt

with the horizon. This is due to the fact that both the orientation and tilt angle change the

solar radiation reaching the surface of the collector. A lot of empirical correlations for

estimating the optimal tilt-angle are available in the literature [4–8]. It is reported in the

literature that the optimum orientation of the PV array should be directly towards the equator,

facing south in the northern hemisphere and the optimum tilt angle depends only on the latitude ( ). For example, Lunde [9] and Garge [10] βopt = ± 15°, Duffie and Beckman [11]

suggested βopt = ( ± 15°) ± 15° and Heywood [12] concluded that βopt = ( ± 15°), where

latitude of the location and where plus, and minus signs is used in winter and summer

respectively [2]. There are many papers in the literature which make different

recommendations for the optimum tilt, based only on the latitude [8, 13]. In practice, the

collector plate is usually oriented south facing and at a fixed tilt which is set to maximize the

average energy collected over a year [3]. Monthly averages of the daily solar radiation

incident upon a horizontal surface are available in the literature for many locations [4, 14-16].

However, radiation data on tilted surfaces is not generally available. From this point of view,

in this study, the optimum tilt angle for the south facing single axis solar collector has been

determined in order to maximize the system performance in Antalya, Turkey. Total solar

radiation on the solar collector surface with an optimum tilt angle is computed and analyzed for specific periods (monthly, seasonal, semi-annual, annual, ± 15° and ).

2. Theoretical analysis

The monthly average values of solar radiation incident on surfaces of various orientations are

required for solar energy applications. The monthly averages of the daily solar radiation

incident upon a horizontal surface are available for many locations. However, radiation data

on tilted surfaces are generally not available. A simple method to estimate the average daily

radiation for each calendar month on surfaces facing directly towards the equator has been

developed by Liu and Jordan [17].

The tilt angle(β) of any collector is defined as the angle between the plane of the collector

surface and the horizon. When β is positive, the orientation of the surface is toward the

equator, and when negative, it is toward the pole.

The earth’s axis is tilted approximately 23.45o with respect to the earth’s orbit around the sun.

As the earth moves around the sun, the axis is fixed if viewed from space. The declination of

the sun is the angle between a plane perpendicular to a line between the earth and the sun

and the earth’s axis. An approximate formula for the declination of the sun is given as

follows [11];



Page | 397

36023.45 sin (284 )

365n

(1)

where n is the number of the day of year starting from the first of January (n=1 on January 1st

and n=365 on December 31st , February 29th is ignored).

Sunrise and sunset occur when the sun is at the horizon and hence the cosine of the zenith

angle is zero. Setting the cosine of the zenith angle to zero in the relation, we get the

following equation; 1cos ( tan tan ) (2)

The monthly average daily radiation on a horizontal surface (H), the fraction of the mean

daily extraterrestrial radiation (H0), the monthly average daily diffuse radiation (HD);

0

24 360(1 0.033 cos( )(cos cos sin sin sin )

365 180sc

nH G

(3)

where Gsc is the solar constant (1367 W/m2), is the latitude of the Antalya.

Solar radiation incident outside the earth’s atmosphere is called extraterrestrial radiation. On

average the extraterrestrial irradiance is 1367 W/m2. The monthly average daily solar

radiation on tilted surface (HT), may be expressed as follow (Liu and Jordan, 1960) [17];

( ) (1 cos ) (1 cos )2 2

DT D b

H HH H H R

(4)

where ρ is ground reflectance (≈0.2).

Liu and Jordan (1960), have suggested that can be estimated to be the ratio of

extraterrestrial radiation on the tilted surface to that on a horizontal surface for each month.

For a surface facing directly towards the equator; cos( )cos sin ' ( /180) 'sin( )sin

cos cos sin ( /180) sin sinbR

(5)

where is the sunset hour angle for the tilted surface given by 1

1

cos ( tan tan )' min

cos ( tan( ) tan )

(6)

where “min” means the smaller of the two items in the bracket [2].

3. Methodology

The equations which calculate total solar radiation falling on tilted surface for optimum tilt

angle the monthly, the seasonally, the semi-annually, the annually, latitude ± 15°, and

latitude are solved with a computer code which is written in MATLAB and should be modular

to allow users to update component modules easily as new findings become available. The

calculations begin with measured hourly global and diffuse radiation received on a horizontal

surface. These quantities are then transposed onto an inclined plane by a mathematical

procedure. The optimum tilt angle was computed by searching for the values for which the

total radiation on the collector surface is a maximum for a particular day or a specific period.

In this regard, the calculations were made for a south facing solar collector for 365 days. The

tilt angle is changed from 0° to 90°. The solar reflectivity () was assumed to be 0.2. The βopt

obtained for a specific period allows us to collect the maximum solar energy for Antalya,

Turkey.



Page | 398

4. Result and discussion

The main objective of this study is to determine and analyze the optimum tilt angle for solar

collectors in Antalya, which is located in the southern part of Turkey and is the seventh

biggest city in the country by population which is a main touristic attraction point by the south

coast facing the Mediterranean.

Turkey lies in a sunny belt between 36° and 42°N latitudes and is geographically well

situated with respect to solar energy potential. Turkey’s yearly average total sunshine

duration is 2640 h and the yearly average solar radiation is 1311 kWh/m2 yr (3.6kWh/m2 day)

[18]. Antalya has a high potential for solar energy production (Fig. 1) and the daily sunshine

duration is very long. Fig. 2 show sunshine duration throughout the year in Antalya.

Fig. 1: Solar Maps of Antalya

Fig. 2: Sunshine duration throughout the year

Table 1 and Fig 3 show optimum tilt angles and Table 2 and Fig 4 show calculated solar

radiation on tilted surface for optimum tilt angles. The optimum angle of tilt of a flat-plate

collector in January is 63° and the total monthly solar irradiation falling on the collector

surface at this tilt is 529.11 MJ/m2-month. The optimum tilt angle in July is 1° and the total

monthly solar radiation at this angle is 701.05 MJ/m2-month. The optimum tilt angle increases

during the winter months and reaches a maximum of 65° in December which collects 517.75

MJ/m2-month of solar energy.

Du

rati

on

(h

ou

r)



Page | 399

Table 1: Monthly, seasonally, semi-annually, annually, latitude±15 and latitude optimum tilt angles

Months Monthly Seasonally Semi-annually

Annually ∅±15 ∅

Jan 63 61 54

33

51,89

36,89

Feb 54

Mar 39

22

13 21,89

Apr 21

May 5

Jun 1

4 Jul 1

Aug 15

Sep 33

48 54

51,89 Oct 50

Nov 61

Dec 65 61

Table 2: Solar radiations on tilted surface for optimum tilt angles (MJ)

Months Monthly SeasonallySemi-annually

Annually ∅ ∅±15

Jan 529,11 528,89 523,66 467,77 482,49 520,74 Feb 503,55 500,46 503,55 474,94 484,45 503,24 Mar 589,77 568,62 541,05 587,05 589,41 568,35 Apr 614,37 614,27 610,11 603,91 596,23 614,29 May 687,87 667,37 683,19 633,07 617,17 667,63 Jun 694,99 692,44 678,16 610,15 591,40 654,01 Jul 701,05 700,07 690,07 629,37 611,66 669,57 Aug 647,92 640,07 647,76 623,85 612,72 644,13 Sep 577,99 561,41 546,23 577,99 576,68 552,11 Oct 562,17 562,00 560,82 542,56 550,64 561,82 Nov 516,12 504,93 512,95 464,22 477,52 510,68 Dec 517,75 516,54 509,05 447,27 462,96 505,47 Total 7142,66 7057,06 7006,61 6662,16 6559,32 6972,04

Fig 3: Monthly, seasonally, semi-annually, annually, latitude±15 and latitude optimum tilt angles



Page | 400

Fig 4: Solar radiations on tilted surface for optimum tilt angles

Fig 5. shows the average seasonally total radiation on a south facing surface on optimum tilt

angle. The seasonal average was calculated by finding the average value of the tilt angle for

each season and the implementation of this requires the collector tilt to be changed four

times a year. In spring the tilt should be 22°, in summer 4°, in autumn 48° and in winter 61°.

It is clear from the figures that a unique optimum tilt angle exists for each month of the

season that corresponds to the maximum point of each curve.

Fig 6. shows the average semi-annually total radiation on a south facing surface on optimum

tilt angle. It is clear from the figures that a unique optimum tilt angle exists for each 6 months

of the year that corresponds to the maximum point of each curve.

Fig 7. shows the average annually total radiation on a south facing surface on optimum tilt

angle. The yearly average tilt was calculated by finding the average value of the tilt angles for

all months of the year. The yearly-average optimum tilt angle was found to be 33° and the

yearly collected solar energy was 6662.16 MJ/m2-year for a south facing solar collector which

nearly corresponding to the latitude of Antalya (36.89°). This is in agreement with the results

of many other researchers. It is clear from the figure that a unique optimum tilt angle exists

for each month of the annually that corresponds to the maximum point of each curve.

Fig 5: Seasonally average solar radiation availability of tilted surfaces.



Page | 401

Fig 6: Semi-annually average solar radiation availability of tilted surfaces.

Fig 7: Annually average solar radiation availability of tilted surfaces.

Fig 8 shows the average monthly total radiation at Antalya on a south facing surfaces for

latitude angle and latitude angle±15 of panel tilt. It is clear from this graph that a unique tilt

angle exists for each month of the year for which the solar radiation is at a peak for the given

month.

Fig 9 shows the results obtained that variation of monthly optimum tilt angle for Antalya.

Fig 8: Monthly average solar radiation for latitude angle and latitude angle±15



Page | 402

Fig 9: The variation of monthly optimum tilt angle for Antalya

When the monthly optimum tilt angle was used, the yearly collected solar energy was

7142.66 MJ/m2-year. The solar collector should be mounted at the monthly tilt angle and the

slope adjusted every month. This would allow an increase in the efficiency of the solar

collector more than 7% over that of a similar fixed solar collector at the optimum annual tilt

angle in Antalya.

When the seasonally optimum tilt angle was used, the yearly collected solar energy was

7057.07 MJ/m2 -year. The solar collector would be mounted at the seasonally tilt angle and

the slope adjusted every season. This will allow an increase in the efficiency of the solar

collector more than 5% over that of a similar fixed solar collector at the optimum annual tilt

angle in Antalya.

5. Conclusion

The optimum tilt angle is different for each month of the year. The collected solar energy will

be greater if we choose the optimum tilt angle for each month. Also it has been found that the

optimum tilt angle in June and July becomes 1°. The optimum tilt angle then increases during

the winter months and reaches a maximum of 65° in December. The results show that the

average optimum tilt angle for the summer months is 4° and for the winter months 61°.

Finally, the yearly-average optimum tilt angle found to be 33° and the yearly collected solar

energy was 6662.16 MJ/m2-year for a south facing solar collector which nearly corresponding

to the latitude of Antalya (36.89°). This is in agreement with the results of many other

researchers.

This study shows the importance of accurate slope angle and orientation. The position of the

solar collectors can be easily adjusted when the supporting structure is designed

accordingly. It can be concluded that a yearly average fixed tilt can be used in many general

applications (e.g. domestic water heating) in order to keep the manufacturing and installation

costs of collectors low. For higher efficiency, the collector should be designed such that the

angle of tilt can easily be changed at least on a seasonal basis, if not monthly. Alternatively,

solar tracking systems can be used in industrial installations where higher efficiency is

required.



Page | 403

References

[1]. Manes A, Lanetz A. “On the optimum exposure of flat-plate fixed solar PV arrays”, Solar Energy (1983), 31(1) 65–73.

[2]. Kumar, A., et al., “Optimization of tilt angle for photovoltaic array”. International Journal of Engineering Science and Technology, (2011) 3- 4.

[3]. Ahmad M.J., Tiwari G.N., “Optimization of Tilt Angle for Solar Collector to Receive Maximum Radiation”, The Open Renewable Energy Journal, (2009) 2 19-24.

[4]. Zuhairy AA, Sayigh AAM., “Simulation and modeling of solar radiation in Saudi Arabia”, Renewable Energy, (1995) 6(2) 107–18.

[5]. Mehleri ED, Zervas PL, Sarimveis H, Palyvos JA, Markatos NC., “Determination of the optimal tilt angle and orientation for solar photovoltaic arrays”, Renewable Energy, (2010) 35(11) 2468–75.

[6]. Moghadam H, Tabrizi FF, Sharak AZ., “Optimization of solar flat collector inclination”, Desalination, (2011) 265(1) 107–11.

[7]. Maatallah T, El Alimi S, Nassrallah SB., “Performance modeling and investigation of fixed, single and dual-axis tracking photovoltaic panel in Monastir city, Tunisia”, Renewable and Sustainable Energy Reviews, (2011) 15(8) 4053–66.

[8]. Benghanem M., “Optimization of tilt angle for solar panel: case study for Madinah, Saudi Arabia”, Applied Energy, (2011) 88(4) 1427–33.

[9]. Lunde PJ. “Solar Thermal Engineering”. (New York: Wiley, 1980).

[10]. Garg HP. “Treative on Solar Energy”,Fundamentals of Solar Energy, (New York: Wiley,1982).

[11]. Duffie JA, Beckman WA. Solar Engineering of Thermal Processes. (New York: Wiley, 1982).

[12]. Heywood H. Operational experience with solar water heating. J Inst Heat Vent Energy, (1971) 39 63–9.

[13]. Khademi, M., Jafarkazemi, F., Saadabadi, S. A., & Ghazi, E., “Optimizing the Tilt Angle of Solar Panels by SQP Algorithm”. Applied Mechanics and Materials, (2013) 253 766-771.

[14]. Tang Runsheng, Wu Tong., “Optimal tilt-angles for solar collectors used in China”, Appl Energy (2004)79 239–48.

[15]. Ulgen K, Hepbasli A., “Comparison of the diffuse fraction of daily and monthly global radiation for Izmir, Turkey”, Energy Sources (2003) 25 637–49.

[16]. Ghosh HR, Bhowmik NC, Hussain M., “Determining seasonal optimum tilt angles, solar radiations on variously oriented, single and double axis tracking surfaces at Dhaka”, Renewable Energy (2010) 35(6) 1292–7.

[17]. Liu B, Jordan R., “Daily insolation on surfaces tilted towards the equator”, Trans ASHRAE (1962) 67.

[18]. Gunerhan H, Hepbasli A., “Determination of the optimum tilt angle of solar collectors for building applications”, Building and Environment (2007) 42 779–83.



Page | 404



Page | 405

A STATISTICAL COSTING APPROACH FOR EXERGOECONOMIC ANALYSES APPLIED TO A HEAT PUMP

A. Dietrich a,b, F. Dammel a,b & P. Stephan a,b

a Institute of Technical Thermodynamics

Technische Universität Darmstadt, Petersenstraße 17, 64287 Darmstadt, Germany

b Darmstadt Graduate School of Energy Science and Engineering

Technische Universität Darmstadt, Petersenstraße 32, 64287 Darmstadt, Germany

Email: [email protected]; [email protected]; [email protected]

Abstract

Exergoeconomic analyses combine an exergy analysis and an economic analysis providing a powerful tool to assess and optimize energy conversion systems. Thermodynamic models used for the exergy analysis usually can be set up with relatively small uncertainties. However, economic models needed to relate the thermodynamic data to capital costs are difficult to develop and contain relatively large uncertainties. In this paper a statistical costing approach suitable for exergoeconomic analyses is presented. Based on market prices probability distributions for the relevant capital investment costs are determined. Applying a Monte Carlo simulation, exergoeconomic parameters relevant for system analysis can be calculated. This method is applied to an air-water vapor compression heat pump system for building heating. For the desired output variables probability distributions and confidence intervals are derived through the statistical costing approach. The results are relatively precise compared to the confidence intervals of the input variables. Based on the most probable values of the output variables, components are identified for optimization.

Keywords: Exergoeconomic analysis; Costing considerations; Monte Carlo simulation; Heat pump

List of Symbols Roman letters Greek lettersahd J annual heat demand CD - condenser efficiency

aot s annual operation time mech - mechanical efficiency

A -1a€ annual constant annuity isen - isentropic efficiency

c -1J€ specific costs per unit of exergy

- exergy efficiency

Fc -1J€ average unit cost of fuel -1s annual operation and maintenance costs factor

Pc -1J€ average unit cost of product

€ mean

C -1s€ cost flow ln - distribution parameter of lognormal distribution

CRF -1a capital recovery factor € standard deviation

e -1kJ g specific exergy ln - distribution parameter of lognormal distribution

E -1J s exergy flow

DE -1J s exergy destruction (flow) Superscripts / subscripts / abbreviations

LE -1J s exergy loss (flow) 0 reference state



Page | 406

1. Introduction

In times of rising energy prices and short running resources, energy conversion systems

should operate as efficient as possible. To achieve this goal, appropriate analysis and

optimization tools are required. Based on the first law of thermodynamics, energy analysis is

widely used to assess energy conversion systems. As energy cannot be destroyed, the first

law of thermodynamics does not distinguish between energy forms of different

thermodynamic quality. Consequently, the results gained by an energy analysis provide only

a limited assessment about the thermodynamic performance of the system. The second law

of thermodynamics, however, accounts for the different thermodynamic quality of energy

forms and the physical limits of energy conversion. Being based on the second law of

thermodynamics, exergy analysis should therefore be applied to assess the efficiency of

energy conversion systems. A basic introduction to the method of exergy analysis can be

found in [1-4].

Considering only thermodynamic methods to assess energy conversion systems is not

sufficient and the gained results may be misleading when it comes to system design or

system optimization. Capital costs have to be taken into consideration when building or

altering a system. A thermodynamically optimized system is useless, as long as it exists

solely on paper due to the huge capital costs tied to its realization. Consequently, the exergy

analysis should be combined with an economic analysis. The resulting method is known as

exergoeconomic analysis, thermoeconomic analysis, or exergy costing. A detailed review of

its fundamentals can be found in [1, 5, 6], a short introduction will be presented in section 2.

For an exergoeconomic analysis, the system to be analyzed needs to be modeled

thermodynamically and economically. The thermodynamic model used for the exergy

h -1kJ g specific enthalpy CD condenser

ir - interest rate CP compressor

M -1skg mass flow CU control unit

n a number of years of operation

D destruction

p -2N m pressure EV evaporator

P -1J s electrical power F fuel

)P( X - probability of X in inflow

Q -1J s heat flow IS installation costs

r - relative cost difference k component R € present value L loss

s -1)Kkg(J specific entropy NE non-exergy-related

Dy - exergy destruction ratio out outflow

Z € total capital costs P product CIZ € capital investment costs R refrigerant MZ € maintenance costs tot total system

OZ € operation costs (excluding fuel)

TV throttle valve

NEZ € non-exergy-related costs W water



Page | 407

analysis usually can be set up with relatively small uncertainties. However, the economic

model necessary to relate the thermodynamic data to capital costs is difficult to develop and

usually contains relatively large uncertainties. Szargut and Tsatsaronis [7, 5, 6] propose an

approximation equation assuming that the capital investment costs scale with component

efficiency and size. Given the appropriate thermodynamic and financial data to set up a

reliable model, this approach is especially useful for system optimization. However, the

required data is usually not available in the necessary quality and quantity.

Although exergoeconomic analyses focus primarily on large power plants and chemical

plants because they are more capital intensive than medium and low-size thermal utilities [8],

exergoeconomic analyses of heat pump systems are also available in literature. However,

only a few different approaches for the economic model can be found. In [9] capital

investment costs are not considered at all, only costs for exergy streams entering the system

are defined. In [10] capital investment costs are considered, but its determination is not

specified. Several approaches [11, 12] can be found that take real component costs into

consideration. These costs have been determined based on actual investments; a general

approach for assigning costs is not presented. In the work of [13, 14] it is assumed that

capital investment costs are proportional to characteristic component properties. For

example, the capital investment costs of a heat exchanger are assumed to be proportional to

the heat transfer area; the capital investment costs of a compressor are assumed to be

proportional to the compression capacities. Although this approach has a good foundation, it

lacks reasonable definitions for the proportionality factors. Also, the approximation equation

introduced by Szargut and Tsatsaronis [7, 5, 6] is used in the economic model of a heat

pump system [15].

As a result, a new approach for the economic model is necessary in order to base the

exergoeconomic analysis on a reliable economic foundation. In this paper a statistical costing

approach is presented. Based on market prices probability distributions for the relevant

capital investment costs are determined. Applying a Monte Carlo simulation,

exergoeconomic parameters relevant for system analysis are calculated.

The statistical costing approach is different from the one taken by Marletta [8] that uses a

Monte Carlo simulation for an exergoeconomic optimization. The costing considerations

applied in this study are also based on the capital cost approximation equation introduced by

Szargut and Tsatsaronis [7, 5, 6].

Before presenting the statistical costing approach in section 3, basic principles of an

exergoeconomic analysis are introduced in section 2. In section 4 the application of the

statistical costing approach in an exergoeconomic analysis of a heat pump system is

exemplified and approaches for system optimization are discussed. A conclusion can be

found in the last section.



Page | 408

2. Exergoeconomic analysis

For an exergoeconomic analysis an exergoeconomic model, a combination of

thermodynamic model and economic model, needs to be derived. This section introduces the

basic principles of each model as well as useful coefficients for system analysis and

optimization. For further details [1, 5, 6] are recommended.

2.5. Thermodynamic model

The thermodynamic model is used to describe the physics of the examined system. To

increase the model accuracy, the system should be divided into subsystems, each of which

is modeled separately. Usually, the system components make good entities for subsystems.

Figure 1 shows an exemplary system component k and the entering and exiting exergy

flows, as well as exergy destruction within the component. In contrast to energy, exergy is no conserved quantity and the exergy destruction kE D,

can be calculated as expressed in

equation (1).

i

iki

ikk EEE ,out,,in,D, (1)

Fig. 1: Exemplary system component with entering and exiting exergy flows and exergy destruction

within the component.

The entering and exiting exergy flows can be calculated using the thermodynamic

information provided by the physical model of the component. The physical model of the

component will not be described here, as it depends on the type and duty of the component.

However, in section 4.1 the physical model of a vapor compression heat pump system is

presented.

As depicted in figure 2 exergy flows crossing the system boundary can be classified into product kEP,

, fuel kEF, and loss kEL,

. “The product is the useful commodity produced by the

component and the fuel is the resource provided for this aim” [8]. Therefore, product and fuel

definitions depend on the type of component and are usually not equal to the sum of exiting

and entering exergy flows, respectively. Several examples can be found in [1, 5].

Fig. 2: Exemplary system component with exergy flows of fuel, product, and loss as well as exergy

destruction.



Page | 409

The exergy loss is an exergy flow rejected to the surroundings not being used by other

components. Similar to equation (1), the exergy destruction can be expressed in terms of

exergy flow of fuel, product and loss:

kkkk EEEE L,P,F,D, . (2)

Important exergetic parameters are the exergy efficiency k and the exergy destruction ratio

kyD, [1]. Definitions are given in equations (3) and (4), respectively. Due to the existence of

exergy destruction and exergy loss in every real system, the exergy efficiency is smaller than

one. The exergy destruction ratio considered in this work relates the sum of exergy

destruction and exergy loss of the thk component to the exergy of fuel entering the overall

system [1].

k

kk E

E

F,

P,

(3)

totF,

L,D,D, E

EEy kk

k

(4)

2.6. Economic model

The economic model is necessary to relate the thermodynamic performance of the

components to total capital costs. In the following subsections, the general model approach

will be described followed by an introduction of a widely applied method to calculate capital

investment costs.

2.6.1 General considerations

As given in equation (5), the component total capital costs kZ are composed of capital

investment costs CIkZ , operating costs O

kZ (excluding fuel costs) and maintenance costs MkZ

[5].



Page | 410

MOCIkkkk ZZZZ (5)

Several approaches exist in determining the different cost components [5]. Here, the

approach by [6] is applied, assuming that the annual operation and maintenance costs are

proportional to the capital investment:

CIMOkkk ZZZ . (6)

To properly account for the total capital costs of a component in an exergoeconomic

analysis, these costs need to be expressed as a cost flow per period of plant operation.

Since the capital investment usually occurs at the beginning of the projected total operation

time, the capital investment costs need to be levelized. The capital recovery factor CRF, as

given in equation (7), is used to calculate “the equal amounts A of a series of n money

transactions, the present value of which is R“ [1]. The total operation time in years and the

interest rate are expressed by n and ir, respectively.

11

1

n

n

ir

irir

R

ACRF

(7)

Finally, the component total capital costs per year become

CIkk ZCRFZ . (8)

2.6.2 Capital investment costs

To evaluate equation (8) the capital investment costs need to be determined. In an often

adopted approach by Szargut and Tsatsaronis [5, 6, 7] equation (9) is proposed to calculate

the capital investment costs CIkZ of the thk system component.

k

km

k

l

k

kkk EBZ P,

CI

1

(9)

Equation (9) is an approximation of the capital investment costs assuming that these costs are proportional to component efficiency and size. The constants kB , kl and km need to be

determined using financial and thermodynamic data. Thus, to obtain a reliable economic

model, a substantial amount of market prices in combination with suitable performance data

needs to be collected. Having the appropriate data to calculate reliable coefficients, this

approach is especially useful for system optimization, since the capital investment costs are

a function of component performance. However, the required data is usually not available in

the necessary quality and quantity.

Other economic models suitable for practical exergoeconomic analyses are currently not

available. To determine the capital investment costs, a novel costing approach based on

statistical probability distributions is presented in section 3.



Page | 411

2.7. Exergoeconomic model

In the exergoeconomic model costs are assigned to the exergy flows calculated in the

thermodynamic model. As introduced by [5, 1] and expressed in equation (10) the product of

exergy flow iE and specific costs per unit of exergy ic yields the cost flow iC .

iii CEc (10)

Recalling the thk system component introduced in section 2.1 the cost balance depicted in

figure 3 and equation (11) can be deduced.

Fig. 3: Cost flows through the exemplary system component.

i

ikiki

kikik EcZEc ,out,,out,,in,,in,

i

iki

kik CZC ,out,,in,

(11)

All cost flows i

ikC ,out, associated with the exiting exergy streams have to equal the sum of

the cost flows i

ikC ,in, associated with the entering exergy streams and the cost flow

resulting from the component total capital costs kZ . For the entering exergy streams, the

cost per unit of exergy ikc ,in, is usually known as a result of the exergoeconomic analysis of

the preceding components. Exergy streams entering the total system usually have to be

purchased and the corresponding costs can be derived from market prices. Examples are

electricity costs or costs for fossil fuels. For the exiting exergy streams the cost per unit of exergy ikc ,out, has to be calculated with the help of equation (11) and auxiliary equations.

Since the auxiliary equations depend on the type and duty of the component, they are not

reviewed here and the reader is referred to [5, 1]. However, in section 4.3 the auxiliary

equations used for the exergoeconomic model of a heat pump system are presented.

Applying the definition of product and fuel as described in section 2.1 results in the cost

balance depicted in figure 4 and equation (12). In order to account for the burden of the exergy loss kEL,

, the average unit cost of exergy loss kcL, is set equal to zero. As a result,

the cost flow of exergy loss kCL, is also zero.

Fig. 4: Product and fuel cost flows though the exemplary system component.



Page | 412

kkkkkkkkk EcEcEcZEc P,P,L,L,P,P,F,F,

kkkkk CCCZC P,L,P,F, (12)

In combination with auxiliary equations, equation (12) can be used to calculate the average unit costs of product kcP, and fuel kcF, . However, if the cost flows of the entering and exiting

exergy streams (i

ikC ,in, and

iikC ,out,

) have already been determined with the help of

equation (11), a computationally less intensive approach should be applied. In this case,

equations (13) and (14) [5, 1] should be used, in combination with the component specific product and fuel definitions, to calculate the average unit costs of product kcP, and fuel kcF, .

Again, examples can be found in section 4.3.

k

kk E

Cc

,P

,PP,

(13)

k

kk E

Cc

,F

,FF,

(14)

The average unit costs of product and fuel can also be determined for the entire system.

When non-exergy-related costs NEZ have to be assigned to the overall system, these costs

have to be considered in the average unit costs of product totP,c and equation (15), instead

of equation (13), has to be applied. Non-exergy-related costs are all costs that have to be

assigned to components which do not release any exergy flows of product (e.g. costs of

control units, overall installation costs).

totP,

NEtotP,totP,

E

ZCc

(15)

Introduced by [5, 1], the relative cost difference kr given in equation (16) is a useful

parameter for system analysis. It describes the relative increase in the costs per unit of

exergy between fuel and product [1]. When comparing different components k, the relative cost differences kr should be used rather than the unit costs per unit of product exergy kcP, .

k

kkk c

ccr

F,

F,P, (16)

3. Statistical costing approach

The economic model is an important part of the exergoeconomic analysis. Assumptions and

decisions taken in the economic model directly influence analysis and optimization results.

The general considerations presented in section 2.2.1 are useful and practical to capture the

influences of the operation and maintenance costs and to allocate the total capital costs to

the total operation time. However, current methods to determine the capital investment costs,

as introduced in section 2.2.2, require economical and technical data that is usually not

available in the necessary quality and quantity.



Page | 413

Therefore, a statistical costing approach is suggested that is based on capital investment

costs derived from statistical evaluations. Firstly, market prices of required components have

to be determined. The larger the amount of data, the more accurate the exergoeconomic

calculations will be. Based on these market prices, probability distributions for the component

capital investment costs are derived, that best resemble the collected data. Next, a Monte

Carlo method is applied in order to calculate the exergoeconomic output variables. Hence, in

repetitive calculations the exergoeconomic output variables are determined using capital

investment costs randomly selected based on the derived probability distributions. Finally,

the output variables are given in form of probability distributions, too. The most probable

results can be determined and probabilities for specific ranges of output variables can be

calculated.

Usually, the determination of component capital investment costs based on market prices is

fairly difficult. If the amount of data is very small the probability distribution should be defined

based on estimates for the most probable value and on estimates for the standard deviation

of an appropriate probability distribution. It is suggested to choose a lognormal distribution for

the distribution type. In contrast to the normal distribution, random variables can only take

positive values. This is important as capital investment costs are approximated which have to

be greater than zero. Moreover, the lognormal distribution is often used in economics to

model the probability of incomes and sales volumes [16]. Consequently, this probability

distribution is considered to be suitable for the approximation of cost probabilities.

For skew distribution functions, as the lognormal distribution, the mean does not coincide

with the mode. Therefore, the most probable value of the component capital investment

costs determined in the market evaluations should be the mode of the chosen distribution

function, and not the mean.

3.1 Advantages

- The statistical costing approach can be applied, even if the amount of data on market

prices is small.

- Different uncertainties can be chosen for estimations based on cost data of different

quantity and quality.

- Based on the input data, this method also provides information about the probability

of calculation results.

- The technical characteristics of the components need to be considered when

determining market prices. However, as specific information as the component

exergy efficiency and the exergy flow of product, required for the application of the

approach by Szargut and Tsatsaronis (see section 2.2.2), are not necessary for the

application of the statistical costing approach.



Page | 414

3.2 Disadvantages

- The determination of applicable parameters of the distribution functions might be

difficult.

- An exergoeconomic optimization based on the statistical costing approach is more

elaborate than an optimization based on the approach by Szargut and Tsatsaronis

(see section 2.2.2), as capital costs are not directly linked to component performance.

However, a sensitivity analysis can be used to identify components having a

reasonable influence on system efficiency.

4. Model application

In the following section the statistical costing approach will be applied to an exergoeconomic

analysis of an air-water vapor compression heat pump system for building heating. Following

the structure of section 2, the thermodynamic model, the statistical costing approach and the

exergoeconomic model will be described successively. Finally, the results are presented and

discussed.

4.1 Thermodynamic model – heat pump system

The schematic of the vapor compression heat pump system working with the refrigerant

R410A is depicted in figure 5. The heat source of the evaporator is the ambient air. The heat

sink of the condenser is a water stream. Reference and boundary conditions as well as

efficiencies used in the thermodynamic model are given in table 1. Component specific

product and fuel definitions of the exergy flows, as introduced in section 2.1, are denoted in

table 2. These definitions are of importance for the calculation of exergetic and

exergoeconomic parameters.

Fig. 5: Schematic of the vapor compression heat pump system consisting of Evaporator (EV),

compressor (CP), condenser (CD) and throttle valve (TV).



Page | 415

Table 1: Reference and boundary conditions as well as efficiencies used in the thermodynamic model.

Component Property Value Component Property Value

Ent. System 0T C8.15- Compressor lowp bar28.4

0p bar1 highp bar37.21

minT C5 isen 0.80

RM 13 skg102.38 mech 0.95

Evaporator min01 TTT C13.15- Condenser WM 13 skg10200

inW,T C30

CD 0.90

Table 2: Product and fuel definitions of the exergy flows for the different heat pump components.

Ent. System Evaporator Compressor Condenser Throttle valve

Product CD,QE 1E 12 EE CD,QE

4E

Fuel CPP 4EV, EEQ

CPP 32 EE 3E

As shown in equations (17) and (18) the compressor model is based on the mechanical efficiency mech and the isentropic efficiency isen . The mechanical efficiency accounts for

losses in the compressor drive.

1isen

1isen,22 h

hhh

(17)

12Rmech

CP1

hhMP

(18)

The model equations of the condenser, evaporator and throttle valve are given in equations (19) to (21), respectively. The condenser efficiency CD accounts for a heat loss; not all heat

delivered by the refrigerant is absorbed by the water stream.

inW,outW,32RCDCD hhMhhMQ W (19)

41REV hhMQ (20)

43 hh (21)

The physical properties of refrigerant and water have been calculated by Refprop [17]. For

calculation of exergy flows, thermal and mechanical exergies have been considered:

iiiiii ssThhME 0,00, . (22)

4.2 Statistical costing approach – heat pump system

For the vapor compression heat pump, a market research was conducted to determine the

component capital investment costs. The results can be found in table 3. The italicized table

entries are non-exergy-related costs, as introduced in section 2.3.



Page | 416

Table 3: Component capital investment costs determined by a market research.

Component

Capital investment

costs ICZ in €

Source Component

Capital investment

costs ICZ in €

Source

Evaporator 3100 [18] Throttle

valve 70 [21]

Compressor 1300 [19] Control unit 750 [22]

Condenser 1290 [20] Installation 2500 [23]

As can be seen in table 3, the amount of cost data per component is small. Although this is

not optimal for the exergoeconomic analysis, it is a good setup to demonstrate the

application of the statistical costing approach. In accordance with section 3, lognormal

distributions have been selected for the probability distributions of the capital investment

costs. The probability distribution function of a lognormal distribution is defined as

0,0

0,2

1)(

2

ln

lnln

2

1

lnx

xex

xpdf

x

, (23)

with ln and ln being the characteristic distribution parameters [16].

For the calculation of the distribution parameters, the most probable value, the mode of the

distribution, was set equal to the investment costs given in table 3:

ICICmode, kk ZZ . (24)

The distribution parameter kln, was then determined so that the probability P of the capital

investment costs being in the range of plus-minus 50% of the most probable value ICmode,kZ

equals 90%:

9.0)5.15.0(P IC,mode

ICIC,mode kkk ZZZ . (25)

Considering the mode of the distribution, the parameter kln, was determined in accordance

with equation (26).

IC,mode

2ln,ln, ln kkk Z (26)

The distribution of the compressor capital investment costs is used to illustrate the

determination of the distribution parameters. Figure 6 depicts the probability distribution

functions of a normal distribution and two lognormal distributions. The corresponding

distribution parameters and other characteristic values are given in table 4.



Page | 417

Fig. 6: Probability distribution functions of a normal distribution and two lognormal distributions

representing the capital investment costs of the compressor.

Table 4: Parameters of the distributions shown in figure 6.

#1: normal

distribution

#2: lognormal

distribution

#3: lognormal

distribution

Mean in € 1300 1484 1440

Mode in € 1300 1300 1300

Standard deviation in € 395.2 395.2 382.5

Distribution parameters 1300 242.7ln 238.7ln

2.395 2680.0ln 2611.0ln

)5.15.0(P ICmode

ICICmode ZZZ 0.9000 0.8912 0.9000

The normal distribution function (#1) is not suitable to represent the capital investment costs,

since negative capital investment costs would be possible. Providing the same standard

deviation, the smaller the mean of the normal distribution function, the larger is the probability

of negative capital investment costs.

The parameters of the lognormal distribution #2 have been determined in order to result in

the same mode and standard deviation as the normal distribution. As shown in figure 6,

negative capital investment costs are not possible. However, equation (25) is not fulfilled.

Consequently, the distribution parameters have been adjusted to fulfill equations (25) and

(26) resulting in distribution #3. The distribution parameters of the capital investment costs

for all heat pump components are summarized in table 5.



Page | 418

Table 5: Parameters of the lognormal distributions representing the capital investment costs of all

components.

EV CP CD TV CU IS ICmeanZ in € 3434 1440 1429 77.54 830.8 2769 ICmodeZ in € 3100 1300 1290 70.00 750.0 2500

std in € 912.11 382.5 379.6 20.60 220.7 735.6

ln 8.107 7.238 7.231 4.317 6.688 7.892

ln 0.2611 0.2611 0.2611 0.2611 0.2611 0.2611

Further parameters necessary for the economic model are given in table 6. The cost per

exergy flow of ambient air EV,Qc entering the evaporator is zero, since the supply of ambient

air is free of charge.

Table 6: Parameters of the economic model.

Property Value Property Value

ir 09.0 ahd kWh11650

n a20 aot h1530

CRF -1a0.10185 CPc -1kWhct23

-1a015.0 EV,Qc 0

With the annual heat demand ahd (taken from a low-energy house [24]) and the heating

capacity of the heat pump calculated by the thermodynamic model (see section 4.4) the

annual operation time aot was determined in accordance with equation (27). Changing

reference and boundary conditions as well as partial load operations have not been

considered.

ahd

Qaot CD

(27)

4.3 Exergoeconomic model – heat pump system

In this section the component specific cost balances forming the exergoeconomic model are

presented. The cost balances for the evaporator, compressor, condenser and throttle valve

are given in equation (28) to (31), respectively. They are derived with respect to equation

(11). Based on the product and fuel definitions given in table 2, the cost flows associated with

the exergy flow of fuel are grouped on the left side of the equal sign and the cost flows

associated with the exergy flow of product are grouped on the right side.



Page | 419

EV: 11EV,EV,44 EcZEcEc EVQQ

(28)

CP: 1122CPCPCP EcEcZPc (29)

CD: DQDQ EcZEcEc C,C,CD3322 (30)

TV: 44CD33 EcZEc (31)

The specific costs per unit of exergy are known for the heat flow entering the evaporator (

EV,Qc ) and for the electric power entering the compressor ( CPc ), as given in table 6.

Consequently, the refrigerant specific costs per unit of exergy 1c to 4c and the specific cost per unit of exergy

DQc C, of the heat leaving the heat pump are unknown. The required

auxiliary equation is given in equation (32). In the condenser, no exergy is added to the

working fluid, therefore the refrigerant specific cost per unit of exergy remains constant, as

suggested by [1].

32 cc (32)

4.4 Results and discussion

The results of the exergy analysis based on the thermodynamic model are given in tables 7

and 8. A T-s-diagram of the refrigerant cycle is depicted in figure 7.

Table 7: Results of the exergy analysis: properties of refrigerant and water as well as input/output

quantities of the heat pump system.

Refrigerant Water kWinCPP 2.355

1 2 3 4 in out kWinEVQ 6.221

Cin T -13.15 75.24 34.85 -18.21 29.85 38.95 kWinEV,QE 0

11 KkgkJin s 1.867 1.901 1.192 1.228 0.4346 0.5583 kWinCDQ 7.613

1kgkJin h 419.76 478.31 256.93 256.93 125.29 163.35 kWinCD,QE 1.059

1kgkJin e 42.081 91.59 58.099 48.374 9.9019 15.195 %in 44.96 1skJin E 1.6078 3.4996 2.2198 1.8482 1.9804 3.039 COP 3.233

Table 8: Results of the exergy analysis: exergetic parameters for component and system evaluation.

EV CP CD TV Entire system

WinLD EE 240.44 463.01 221.14 371.59 1296.18

%in 86.99 80.34 82.72 83.26 44.96 %inDy 10.21 19.66 9.39 15.78 55.04



Page | 420

Fig. 7: T-s-diagram of the R410A refrigerant cycle, based on [17].

The coefficient of performance COP and the exergetic efficiency of the entire system are

in the expected range [10, 15]. Of all components, the compressor has the lowest exergetic efficiency and the highest exergy destruction ratio Dy . Although having the highest exergetic

efficiency of all components, the evaporator does not have the smallest exergy destruction

ratio. It is important to notice that the value of the exergetic efficiency depends on the

component specific product and fuel definitions (see equation (3) and table 2). Therefore, the

product of the component specific exergetic efficiencies does not result in the exergetic

efficiency of the entire system.

In the following, the results of the exergoeconomic analysis based on the statistical costing

approach are presented. Several Monte Carlo Simulations with 610 iterations each have

been conducted. The probability distributions of corresponding output variables are identical

in all simulations (based on a reasonable accuracy compared to the input variables), so the

amount of iterations is sufficient to produce representative results. If not stated differently, the

presented values are the values with the highest probability based on the corresponding

probability distribution function.

The cost flow through the heat pump is depicted in figure 8. With the electric power CPP the

cost flow CPC enters the system. Following the refrigerant cycle, the cost flow through the

system develops. The component total capital costs kZ act as cost sources and are

responsible for the cost increase. The cost flow CDC leaving the refrigerant with the ejected

heat flow needs to be burdened with the non-exergy-related capital costs NEZ . This results in

the total cost flow totP,C associated with the exergy flow emitted by the heat pump. Since the

values depicted in figure 8 are the most probable values for the specific component or flow,

these values do not fulfill equations (28) to (31). These equations are only fulfilled, if mean

values for the specific component or flow are considered.



Page | 421

Fig. 8: Cost flows through the heat pump.

In table 9 specific costs and cost flows of the refrigerant cycle are shown. Component

specific costs and cost flows are given in table 10. This information is useful to assess the

exergoeconomic performance of the heat pump. Compared to the other components, the

compressor has a relatively high relative cost difference r . Considering that the compressor

has the largest exergy destruction rate, this indicates an optimization potential. Although

having the second largest exergy destruction rate (refer to table 8), the throttle valve has the

lowest relative cost difference and should therefore not be chosen for optimization. However,

the largest relative cost increase needs to be accounted to the non-exergy-related costs.

Before optimizing the compressor, possible savings on these costs should be considered.

Table 9: Results of the exergoeconomic analysis: specific costs per unit of exergy and cost flows of

the refrigerant cycle.

1 2 3 4 1MJ€in c 0.3223 0.2006 0.2006 0.2420

1h€in C 1.866 2.527 1.603 1.610

Table 10: Results of the exergoeconomic analysis: exergoeconomic parameters for component and

system analysis.

EV CP CD TV Entire system 1h€in Z 0.2553 0.1067 0.1041 0.005739 0.4718

1NE h€in Z - - - - 0.2660

1F MJ€in c 0.2420 0.06389 0.2006 0.2005 0.06389

1P MJ€in c 0.3223 0.09512 0.2710 0.2420 0.3479

%inr 33.52 49.00 34.22 20.53 444.46

The overall economic operation conditions are summarized in table 11. The annual electricity

costs represent only 40% of the annual total costs. Consequently, 60% of the annual total



Page | 422

costs result from capital costs showing the importance of a thorough consideration of capital

costs.

Table 11: Economic operation conditions of the heat pump system.

total cost flow of fuel 1

CPtotF, h€in CC 0.5416

total cost flow of product1

totP, h€in C 1.3257

annual operation time hinaot 1530

annual electricity costs €intotF, aotC 829

annual total costs €intotP, aotC 2029

The statistical costing approach does not only provide the most probable values of the

desired exergoeconomic parameters. It can also be used to determine the probability of a

value range. In figures 9 and 10 the probability distribution function and the cumulative

density function are depicted for the annual total costs aotC totP, and the annual costs

aotC CD without the non-exergy-related costs.

In Table 12 the 90% confidence intervals of the cost distributions shown in figures 9 and 10 are compared to the 90% confidence intervals of the input parameters IC

kZ . It can be seen

that the 90% confidence intervals of the results are small compared to the intervals of the

input parameters. Although being based on a broad range of input parameters, the statistical

costing approach provides relatively precise results.

Table 12: Economic operation conditions of the heat pump system.

mode

in €

90% confidence

interval in €

90% confidence interval

(relative to mode)

min max min max

aotC totP, 2029 1825 2359 0.90 1.16

aotC CD 1581 1429 1860 0.90 1.18

ICkZ 3100 2159 5097 0.70 1.64



Page | 423

Fig. 9: pdf and cdf of the annual total costs aotC totP, .

Fig. 10: pdf and cdf of the annual costs aotC CD without the non-exergy-related costs.

5. Conclusion

A statistical costing approach for exergoeconomic analyses was presented. This approach is

especially useful when the amount of data specifying the required economic input is scarce,

since uncertainties can be accounted for when defining the probability distributions of the

input parameters. As one of the major advantages, also the results are provided in form of



Page | 424

probability distributions. Consequently, information about the precision of the results is

available and can be considered in the system analysis.

The statistical costing approach was successfully applied to a heat pump system and its

ability to generate reliable results based on input data with uncertainties is shown. In further

studies the combination of the statistical costing approach with optimization techniques will

be examined.

References

[1] A. Bejan, G. Tsatsaronis & M. Moran, Thermal Design & Optimization, (John Wiley & Sons, Inc., 1996).

[2] G. Tsatsaronis, “Strengths and Limitations of Exergy Analysis” in Thermodynamic Optimization of Complex Energy Systems, (eds.) A. Bejan, E. Mamut, (Kluwer Academic Publishers, 1999) 93-100.

[3] M. Moran, “Fundamentals of Exergy Analysis and Exergy-Aided Thermal Systems Design” in Thermodynamic Optimization of Complex Energy Systems, (eds.) A. Bejan, E. Mamut, (Kluwer Academic Publishers, 1999) 73-92.

[4] A. Bejan, “Fundamentals of exergy analysis, entropy generation minimization, and the generation of flow architecture”, Int. J. Energ. Res., 7 (2002) 0-43.

[5] G. Tsatsaronis, “Thermoeconomic analysis and optimization of energy systems”, Prog. Energ. Combust., 3 (1993) 227-257.

[6] G. Tsatsaronis, “Exergoeconomic evaluation and optimization of energy systems — application to the CGAM problem”, Energy, 3 (1994) 287-321.

[7] J. Szargut, “Anwendung der Exergie zur angenäherten wirtschaftlichen Optimierung”, Brennst-Wärme-Kraft, 23 (1971) 516-519.

[8] L. Marletta, “A Comparison of Methods for Optimizing Air-Conditioning Systems According to the Exergonomic Approach”, J. Energ. Resour.-ASME, 4 (2001) 304-310.

[9] M. D. d'Accadia & M. Sasso, “Exergetic cost and exergoeconomic evaluation of vapour-compression heat pumps”, Energy, 11 (1998) 937-942.

[10] A. Ucar & M. Inalli, “Exergoeconomic analysis and optimization of a solar-assisted heating system for residential buildings”, Build. Environ., 11 (2006) 1551-1556.

[11] O. Ozgener, A. Hepbasli & L. Ozgener, “A parametric study on the exergoeconomic assessment of a vertical ground-coupled (geothermal) heat pump system”, Build. Environ., 3 (2007) 1503-1509.

[12] O. Ozgener & A. Hepbasli, “Exergoeconomic analysis of a solar assisted ground-source heat pump greenhouse heating system”, Appl. Therm. Eng., 10 (2005) 1459-1471.

[13] A. Kodal, B. Sahin & A. S. Oktem, “Performance analysis of two stage combined heat pump system based on thermoeconomic optimization criterion”, Energ. Convers. Manage.,18 (2000) 1989-1998.

[14] A. Kodal, B. Sahin & A. Erdil, “Performance analysis of a two-stage irreversible heat pump under maximum heating load per unit total cost conditions”, Exergy Int. J., 3 (2002) 159-166.

[15] Z. Oktay & I. Dincer, “Exergoeconomic analysis of the Gonen geothermal district heating system for buildings”, Energ. Buildings, 2 (2009) 154-163.

[16] L. Sachs & J. Hedderich, Angewandte Statistik, (Springer-Verlag, 2006).

[17] Refprop - Reference Fluid Thermodynamic and Traansport Properties, NIST Standard Reference Database 23, Version 9.0 (2010).



Page | 425

[18] Ochsner, Split-Außenteil Standard MSV 14 (OCHS290237), http://www.sanitaerprofi.at/_DIVERSES_/Waermepumpen/Ochsner-Split-Aussenteil-Standard-MSV-14-OCHS290237::1638214.html (accessed April 2013).

[19] Danfoss, Scroll compressor HRH051U5LP6, http://www.danfoss.com/Pacific/BusinessAreas/Refrigeration+and+Air+Conditioning/Products/ (accessed April 2013).

[20] Alfa Laval, Heat exchanger ACH70X-50H, http://www.klimacorner.de/zubehoer/waermepumpen-zubehoer/waermetauscher/ (accessed April 2013).

[21] Sanhua, Expansion valve DPF(T01)2.0C-03, https://www.sanhuaeurope.com/uk/de/shop-online/elektronische-expansionsventile (accessed April 2013).

[22] Stiebel-Eltron, Heat pump control unit WPMS 2.1, http://www.wolf-online-shop.de/HEIZUNG/Waermepumpe-und-Zubehoer/Stiebel-Eltron/Regelgeraete (accessed April 2013).

[23] Waermepumpen.info, http://www.waermepumpen.info/sole-wasser/kosten-preise (accessed April 2013).

[24] Bastian Ziegler, Energy-Plus-Home in Mühltal near Darmstadt (2011).



Page | 426



Page | 427

EXERGETIC ANALYSIS OF THE ENEXAL BAUXITE RESIDUE TREATMENT ON THE OVERALL RESOURCE EFFICIENCY OF THE

PRIMARY ALUMINA REFINING PROCESS

E. Balomenosa, D. Paniasa, I. Paspaliarisa, D. Kastritisb, D. Boufounosb

a National Technical University of Athens

Laboratory of Metallurgy 9, Heroon Polytechneiou str., 157 73, Zographou Campus, Greece

Email: [email protected]; [email protected]; [email protected]

b ALOUMINION S.A. Agios Nikolaos, 320 03 Viotia, Greece,

Email: [email protected]; [email protected] Abstract

In the framework of the ENEXAL FP7 project, a novel process for treating bauxite residues has been developed and demonstrated in industrial scale. The novel process developed, fully converts the bauxite residues in to marketable products such as pig iron and mineral wool fibers. In this work a detailed exergy analysis of the new process is presented and coupled with the flow-sheet of the primary alumina refining process (Bayer), in order to deduce its effect on the overall resource efficiency and exergy utilization. A methodology for evaluating the chemical exergy of mineral phases and solid solutions is also discussed.

Keywords: Exergy Analysis; Red Mud; Bauxite residues; Alumina refinery; Bayer process

1. Introduction

The primary aluminium production industry is the world’s larger industrial consumer of energy

and is ranked among the most CO2 intensive industries. The industry is separated into two

types of plants: alumina refiniries where bauxite ore is refined to metallurgical alumina (Al2O3)

according to the Bayer process and aluminum smelters where metallurgical alumina is

electrolytically reduced into metallic aluminum according to the Hall-Heroult process. A mass

and energy flow sheet of both processes is presented in figure 1, based on the operations of

the vertical refiner - smelter plant of ALOUMINION S.A. in Greece (ALSA).

On average the Bayer process for the production of metallurgical alumina requires 2.65 kg of

bauxite ore to produce 1 kg of alumina, while the slurry containing the remaining bauxite ore,

which is removed from the thickeners during the liquor clarification stage, is by far its greatest

environmental problem. This by-product, called bauxite residue (also known as “red mud”),

on a dry basis is produced in almost a 1 to 1 mass ratio to alumina and consists from various

metal oxides of Fe, Al, Ti, Si, Na, V (depending on the initial chemical composition of the

bauxite ore) along with inclusions of unwashed sodium aluminate solution.

Bauxite residues are classified by EC as a non hazardous waste (Commission Decision

2000/532/EC), however their small particle size (dust-like, mean particle size 0.49μm), high

alkalinity and large amounts (100 to 120 million tons per year on a dry basis worldwide)

makes their disposal a significant problem. Today, the residues are disposed into sealed or

unsealed artificial impoundments, leading to important environmental issues (e.g.



Page | 428

groundwater pH change, leakage, overflow, air pollution by dust) and substantial land use.

The catastrophic red mud spill in Hungary in October 2010 is indicative of the magnitude of

the residue disposal problem. Till this day, due mainly to high costs and low yields, no

industrial application of bauxite residues is in effect.

Fig. 1: The mass and energy flow sheet of ALOUMINION S.A. plant

In the framework of the EC funded ENEXAL project [1] a novel process for the treatment of

the bauxite residues and its transformation into valuable products has been developed and

demonstrated in industrial scale at ALSA. In this work the exergy analysis of the novel

process is presented and compared with the established industrial practice.

2. The Bayer Process

The Bayer process is essentially a cyclic process designed to extract the alumina from the

bauxite ore through high temperature caustic leaching and controlled precipitation, according

to the simplified reaction scheme:

Al2O3.3H2O[bauxite] + 2NaOH[aq] 2NaAlO2[aq] + 4H2O(l) Al2O3

.3H2O[s] + 2NaOH[aq]

A detailed energy and exergy analysis of the process has been presented by the authors in

[2] based on an average industrial flow sheet. As discussed, the Bayer process in total is

characterized by very low exergy efficiency as large amounts of exergy are spent solely to

achieve the chemical separation of alumina from the bauxite “solid- solution”. Based on the

data given in Fig 1, the mass and exergy analysis of the plant is given in table 1 and Fig 2

(details on the chemical exergy calculations of the species are presented in the appendix).



Page | 429

Fig. 2: The exergy analysis of the ALSA alumina refinery

Table 1: Alumina Refinery Plant (exergy of all species calculated at 25oC, 1 atm)

Mass Exergy INPUT Diaspore ore 1.644 kg 0.205 MJ Tropic ore 0.600 kg 0.073 MJ Limestone 0.149 kg 0.029 MJ Caustic Soda (NaOH) 0.039 kg 0.076 MJ Water (H2O) 2.385 kg 0.112 MJ TOTAL INPUT 4.817 kg 0.494 MJ UTILITIES Electricity 0.613 MJ Coal based electricity 0.056 kg 1.856 MJ Heat demand 5.400 MJ Diesel (Heat production) 0.237 kg 10.010 MJ

TOTAL UTILITIES 0.293 kg 11.866 MJ PRODUCT Alumina 1.000 kg 0.417 MJ

TOTAL PRODUCT 1.000 kg 0.417 MJ WASTE Bauxite Residues 0.821 kg 0.160 MJ CO2 from process 0.066 kg 0.030 MJ H2O from process 2.916 kg 1.521 MJ CO2 from Diesel 0.729 kg 0.328 MJ H2O from Diesel 0.291 kg 0.152 MJ CO2 from Electricity 0.162 kg 0.073 MJ

TOTAL WASTE 4.984 kg 2.263 MJ

WASTE HEAT (EXERGY LOSSES) 9.680 MJ Exergy Efficiency of process (without electricity production) 3.75%

Exergy Efficiency of process (coal based electricity production) 3.37%

Total CO2 emissions (direct and indirect) 0.96 kg



Page | 430

Electricity used in the process is considered to be produced through coal burning and is

evaluated in the process as the worst case scenario (including indirect CO2 emissions). The

overall efficiency of the alumina refining plant in ALSA is 3.37% and in terms of chemical

exergy, the bauxite residue embodies 58% of the total chemical exergy of the initial bauxite

ore and is therefore clear that these residues should be treated as a resource rather than as

a waste.

3. The novel bauxite residue treatment

The novel bauxite residue treatment comprises of four stages as shown schematically in

Figure 3. The first stage is the residue drying stage, as even red mud dewatered in filter

presses (current Best Available Technology for bauxite residue handling) contains significant

amounts of moisture (up 25% w/w). This stage can take place in a double skin rotary kiln,

utilising the heat content of the hot off-gases from the EAF. In the next stage of the process

the material feed of the EAF is prepared by mixing the dry red mud, coke fines and

appropriate fluxes to adjust the properties of the produced slag. This mixture is fed into the

EAF where the raw materials undergo reductive smelting and are transformed in three

distinct fluid phases: liquid slag, liquid pig iron and off-gases. The off-gases after heat

exchange in the red mud dryer are sent in a bag-house unit to remove dust particles prior to

releasing them to the atmosphere. The dust collected is recycled in the feed material. The

liquid pig iron and slag phases are separated by sequential pouring (or by tapping in a

continuous process) and the slag is driven directly to the final stage of the process, where

the liquid slag is fiberised to produce inorganic fibers and mineral wool products.

Fig. 3: The envisioned process

In the framework of the ENEXAL project, a 1MW Dust treating EAF was set up in the ALSA

plant and treated in batch mode 1 ton of dried bauxite residues along with 200 kg of coke



Page | 431

fines, 180 kg of silica sand and 150 kg burnt lime The furnace operated at 1600oC, and

produced 230 kg of pig iron and 784 kg of glassy slag. On average the power consumption of

the pilot plant was 2000 kWh per ton of bauxite residue treated (including energy

consumption for plant utilities such as cooling water circulation and off-gasses suction

pumps); the chemical analysis of the products is presented in Table 2. Additionally Table 2

presents empirical chemical composition indexes used in the mineral wool production

industry to evaluate the “fiberise-ability” of a melt [3]. As seen from these indexes the

produced melt is within most of the empirical limits.g

Table 2: Experimental Results

Pig iron %wt

Fe C S P Si Ti V Cr

95.47 3.36 0.26 0.08 0.03 0.00 0.18 0.82

Pig iron phase weight 230 kg Fe recovery in pig iron 70%

Slag %wt Al2O3 SiO2 CaO TiO2 MgO Fe2O3 Na2O V2O5

20.90 27.60 31.19 5.63 8.89 5.01 2.45 0.20

Slag phase weight 784 kg Slag Basicity Ratio 1.24

Empirical indexes

A [<1.8] P [<15] k2 [0.8-1] SHG [1.3

- 1.4] KNB [30-

40] N [<5%] F [>5%]

1.58 9.34 1.06 1.07 36.64 2.45 5.01

Empirical Indexes (oxides in wt%) A = (SiO2 +Al2O3 + TiO2)/(CaO+MgO); N = 4.9/[(MgO+CaO+Fe2O3 +Na2O + TiO2)/(SiO2 +Al2O3)] -0.45; k2 = [100 – (SiO2 +Al2O3)]/( SiO2 +Al2O3); SHG = (SiO2 +Al2O3)/(1.4 MgO+ 0.4 Fe2O3 + CaO + TiO2; KNB = Na2O + MgO + CaO; N = Na2O; F = Fe2O3.

Based on the chemical analysis of the feed and the thermodynamic model of the process the

off-gases of the process at 25oC (at thermodynamic equilibrium with the atmosphere) are 689

kg of CO2 and 151kg of H2O. Table 3 presents the mass and exergy analysis of the new

process, based on the above data.

g The actual fiberization of the slag is not examined here, due to lack of experimental data. Yet this has little effect on the conclusions of the study as the process for melt fiberization is hardly energy or resource intensive.



Page | 432

Table 3: Bauxite Residue treatment (exergy of all species calculated at 25oC, 1 atm)

Mass Exergy

INPUT Bauxite Residues 1.000 kg 0.194 MJ Coke fines 0.200 kg 6.640 MJ Silica Sand 0.180 kg 0.004 MJ Burnt lime 0.150 kg 0.348 MJ

TOTAL INPUT 1.530 kg 7.187 MJ

UTILITIES Electricity 7.200 MJ Coal based electricity 0.655 kg 21.800 MJ

TOTAL UTILITIES 0.655 kg 21.800 MJ

PRODUCT Pig iron 0.230 kg 1.556 MJ Mineral wool 0.784 kg 3.033 MJ

TOTAL PRODUCT 1.014 kg 4.589 MJ

WASTE CO2 from process 0.689 kg 0.310 MJ H2O from process 0.151 kg 0.079 MJ CO2 from Electricity 1.90 kg 0.855 MJ

TOTAL WASTE 2.741 kg 1.244 MJ

WASTE HEAT (EXERGY LOSSES) 23.153 MJ Exergy Efficiency of process (without electricity production) 31.90% Exergy Efficiency of process (coal based electricity production) 15.83%


4. The Novel alumina refinery

The integration of the bauxite residue treatment in the alumina refinery would produce a new

plant with complete bauxite ore exploitation and zero solid/liquid wastes. The mass exergy

analysis of the new plant is presented in table 4 and Fig 4. Burnt lime (CaO) used in the

process is consider to be produced from limestone burning within the plant.

Fig. 4: The exergy analysis of the new alumina refinery



Page | 433

Table 4: New Alumina Plant (exergy of all species calculated at 25oC, 1 atm)

Mass Exergy INPUT Diaspore ore 1.644 kg 0.205 MJ Tropic ore 0.600 kg 0.073 MJ Limestone 0.369 kg 0.073 MJ Caustic Soda (NaOH) 0.039 kg 0.076 MJ Water (H2O) 2.385 kg 0.112 MJ Coke fines 0.164 kg 5.454 MJ Silica Sand 0.148 kg 0.003 MJ TOTAL INPUT 5.349 kg 5.995 MJ UTILITIES Electricity 6.528 MJ Coal based electricity 0.594 kg 19.764 MJ Heat demand 0.000 MJ Diesel 0.258 kg 10.901 MJ

TOTAL UTILITIES 0.852 kg 30.665 MJ PRODUCT Alumina 1.000 kg 0.417 MJ Pig iron 0.189 kg 1.278 MJ Mineral wool 0.644 kg 2.492 MJ

TOTAL PRODUCT 1.833 kg 4.187 MJ WASTE CO2 from process 1.229 kg 0.553 MJ H2O from process 3.039 kg 1.586 MJ CO2 from Diesel 0.729 kg 0.328 MJ H2O from Diesel 0.317 kg 0.166 MJ CO2 from Electricity 1.723 kg 0.775 MJ

TOTAL WASTE 7.037 kg 3.408 MJ WASTE HEAT (EXERGY LOSSES) 29.065 MJ Exergy Efficiency of process (without electricity production) 17.87% Exergy Efficiency of process (coal based electricity production) 11.42%


5. Conclusions

This work presented a detailed mass and exergy analysis of an actual alumina refinery along

with results from industrial demonstrations of a novel process to treat bauxite residues. When

this novel process is intergrated in the alumina refinery plant the overall increase in the

exergy efficiency of the alumina refinery plant is greater than 8 percentile points, marking a

significant improvement in the resource efficiency of the plant, as the bauxite ore is exploited

in full and three marketable products are produced instead of one.



Page | 434

Acknowledgements

The research leading to these results has received funding from the European Union

Seventh Framework Programme ([FP7/2007-2013]) under grant agreement n°

ENER/FP7EN/249710/ENEXAL.

APPENDIX – Chemical Exergy of complex metallurgical products

In general the chemical exergy (ex) of matter defines the maximum work obtainable by the

chemical interaction (reaction) of matter with its environment.

In cases of fuel, chemical exergy is measured by the heating value of the fuel [4]. For diesel

used in this study the HHV was 42.20 MJ/kg; its emissions were 2.96 kg CO2 and 1.23 kg

H2O per kg of diesel burned. To produce 1MJ of electrical exergy through coal burning, 3.03

MJ of coal exergy was assumed to be consumed (the higher heating value, HHV, of coal

used was 34.1 MJ/kg of coal [5]).

For matter that is not fuel, the standard chemical exergy 0xe can be calculated from its

theoretical reaction of formation at the environmental standard state (T0,P0)

a x yaA xX yY A X Y (I-1)

according to the relationship

0 0 0, , 0a x y a x y ix A X Y f A X Y i

i

e G T v e (I-2)

where 0, 0a x yf A X YG T is chemical free energy of formation of the substance, iv is the

stoichiometric coefficient and 0ie is the standard chemical energies of element i. The

standard chemical exergy of elements is related to reference substances found more

commonly in the environment and are given in literature ([4], [6], [7]). In the present study the

standard chemical exergy values of De Meester [7] were used.

The chemical exergy of solution containing n chemical species at the environmental state

(T0,P0 ) is

0, , 0 lnx i x i i x i i i i

i i i

e n e n e RT n x (I-3)

Equation (I-3) can be applied with approximation for solid solutions where all mineral phases

can be considered in thermodynamic equilibrium and the activity coefficient of all species is

considered to be unitary. However in more complex phases resulting from metallurgical

processing the above calculation is not as easily performed. Phases formed under non-

equilibrium conditions or phases with complicated mineralogical compositions, for which

thermodynamic data may not be available, represent a problem for the metallurgical exergy

process analysis.



Page | 435

In this study the use of thermodynamic calculation software is proposed. To define the

mineralogical phases of bauxite residues, the equilibrium module of FACTSAGE was used,

in order to conclude the most probable distribution of the elements in various

chemical/mineralogical species. In this way the red mud with the chemical analysis shown in

table I1, was reduced to a list of mineralogical species with known thermodynamic data.

Through use of equation (I-3) the chemical exergy of red mud was then calculated at 0.19

MJ/kg.

Table I1: Red Mud predicted mineralogical species

Chemical analysis %wt Mineralogical Species %wt

Al2O3 16.22% Fe2O3 hematite 47.74%

Fe2O3 47.74% Al2O3*3H2O gibbsite 17.83%

SiO2 6.09% CaTiO3 perovskite-a 10.10%

TiO2 5.93% NaAlSiO4 nepheline 9.36%

Na2O 2.51% Ca3Al2Si3O12 grossularite 5.31%

V2O5 0.21% CaCO3 aragonite 3.70%

CaO 8.39% Ca(OH)2 3.27%

CaSO4 1.02% Na2SO4(H2O)10 2.18%

CaCO3 3.70% (CaO)3(V2O5) 0.40%

H2O(cry) 8.19% Na2SO4 0.10%

TOTAL 100% TOTAL 100%

Taking this approach one step further, the slag produced from the red mud smelting process,

which due to rapid cooling, is an amorphous (glassy) solution of Al, Si, Ca, Ti and other

oxides, was modeled using FACTSAGE’s “FT-oxid SLAG” liquid solution phase. The later

represents an oxidic solution model capable of predicting the thermodynamic properties of

slag melts. Based on the slag chemical analysis presented in table 2, FACTSAGE’s excess

Gibbs energy prediction for this phase extrapolated at 250C was used directly as the excess

mixing term in (I-3), thus allowing the calculation of the chemical exergy of the glassy slag at

3.87 MJ/kg. This higher value compared to red mud, is justified as the slag represents a

“frozen” non-equilibrium phase with highly “reactive” components like CaO (2.32 MJ/kg) and

Na2O (4.82 MJ/kg).

References

[1]. ENEXAL: Novel Technologies For Enhanced Energy And Exergy Efficiencies In Primary Aluminium Production Industry , http://www.labmet.ntua.gr/ENEXAL/

[2]. E. Balomenos, D. Panias, I. Paspaliaris, Mineral Processing & Extractive Metall. Rev., 32: 69–89, 2011



Page | 436

[3]. E. Balomenos, D. Panias, Proceedings of the 3rd International Slag Valorisation Symposium - Leuven 161- 172, 2013

[4]. Brodyansky V.M. et al.,The efficiency of industrial processes: Exergy analysis and optimization, Elsevier Science B.V., 1994

[5]. Bossel U., Well-to-Wheel Studies, Heating Values, and the energy conservation Principle, European Fuel Cell Forum, 2003

[6]. Szargut J., International Progress in second law analysis, Energy Vol 5, pp. 709-718, 1980

[7]. De Meester B. et al, An improved calculation of the exergy of natural resources for exergetic life cycle assessment, Envirol Sci. Technol. 40, pp. 6844-6851, 2006



Page | 437

ENERGO AND EXERGO-ENVIRONMENTAL ANALYSIS OF A MULTIPURPOSE PROCESS FOR ETHANOL PRODUCTION FROM

SUGARCANE AND CORN

A.C.G. Donke a, M.I.S.F. Matsuura a, E.T. Sugawara b, L. Kulay b

a Brazilian Agricultural Research Corporation - EMBRAPA Environment Division

SP 340 Rd, Km 127.5, 13820-000, Jaguariúna, Brazil Email: [email protected]; [email protected]

b Chemical Engineering Department

Polytechnic School – University of Sao Paulo 580, Prof. Lineu Prestes, Ave, Bloc 18, 05508-900, Sao Paulo, Brazil

Email: [email protected]; [email protected]

Abstract

This study carry out a diagnosis of the energy and environmental performances of ethanol processing under conditions that it occurs in a multipurpose facility located on Brazilian Northwest region. Four productive scenarios were selected to a comparison based on the agricultural assets and energetic utilities. A method of work combining standard environmental impact indicator (ReCiPe) with energetic and exergetic analysis (CED and CExD) was selected to perform the evaluation. For the situation that all the approaches are evaluated simultaneously, scenario II - ethanol from corn and cogeneration - presents the most balanced Energo and Exergo-environmental performances. Scenario III was deeply influenced by the effects provided by the Brazilian grid. This dependence became unfeasible the alternative. The performance of the scenario IV that combines sugarcane and corn was strictly dependent on the performance of ethanol production from sugarcane. It is recommended that additional tests of proportioning between agricultural assets may be carried out

Keywords: Energy Analysis; Exergy Analysis, Life cycle Assessment; ethanol production

1. Introduction

Ethanol has established as fuel for light vehicles in Brazil. Its production occurs from

sugarcane grown in the central-south of the country. Sugarcane also supports domestic and

export markets of sugar. This condition make economically vulnerable and instable both

segments. The situation has stimulated investments in corn production to provide an

alternative to decentralization of ethanol production. Brazilian northwestern region stands to

harvest two harvests per year of corn with high agricultural productivity. This reason has

enabled the deployment of a distillery in the zone with potential for producing up to 100,000-

m3 ethanol/year from both sugar cane and corn. The product aims to meet the needs of the

international market. Therefore, besides economically competitive fuel must demonstrate

appropriate environmental behavior. This study made a diagnosis of the energy and

environmental performances of ethanol processing under conditions that it occurs in the

region. A “cradle-to-gate” approach was considered in order to observe both agricultural and

industrial productions. The results provided by the initiative, in association with those

generated by an economic analysis of the same plant will support the sizing and

implementation of actions to improve the units involved.

The scenario currently practiced in the unit - that ethanol is derived from sugar cane - was

compared with three other possibilities. Among these is the situation most favorable from the



Page | 438

standpoint of business for the organization that manages the plant, in which fuel is produced

for almost twelve months from sugarcane and corn. The evaluation of energy performance

occurred from the determination of energy and exergy demands for every situation. The

environmental assessment takes place by the application of the technique of Life Cycle

Assessment (LCA).

Energo and Exergo-environmental analysis have being performed by the academia in order

to provide evaluations for fuels and energy systems that are more complete and accurate.

Boyano et al. (2012) [1] applied Exergo-environmental method to assess the environmental

aspects of a technology of steam methane reforming. To perform the LCA, it was adopted

Eco-indicator 99 as method of impact assessment. This work emphasized the importance of

thermodynamic efficiency in reducing environmental impacts from the process of energy

conversion. Similar approach have used Peiró et. al (2011) [2] to assess the life cycle of

biodiesel obtained from methyl transesterification of used cooking oil. The process was

evaluated by LCA to account environmental impacts and by an Exergetic Life Cycle

Assessment (ELCA) to measure the exergy input to the system. The results shown that

transesterification is the most impacting stage of the system and the major exergy inputs are

associated to uranium and natural gas for electricity production.

Ozbilen et al. (2012) [3] applied Exergetic Life Cycle Assessment (ExLCA) with LCA to a

hydrogen production process. Exergy efficiencies and air pollution emissions were evaluated

for all process steps. The parametric studies demonstrate that the effect of plant lifetime on

environmental impact per kg hydrogen production diminishes at large-scale production

capacities. The author concluded that ExLCA is a beneficial addition to LCA for introducing

thermodynamic analysis throughout the life cycle analysis of the system.

Koroneos & Tsarouhis (2012) [4] combined methods of LCA and exergy analysis to evaluate

the performance of solar systems for space heating, cooling and hot water production in

domestic scale. The exergy analysis was applied to improve the efficiency of the system

components, as well as reduce costs. The authors highlight the versatility of the analysis,

which can be applied into different areas if the energy needs are set and heatstroke local be

considered.

Banerjee & Tierney (2011) [5] have addressed the environmental impact assessment of

different energy systems for rural communities in developing countries. Ten systems are

proposed, modelled, and assessed by five Exergo-environmental methods. It was found that

the method combining a standard environmental impact indicator – the ReCiPe – with

exergetic analysis is most effective because it uses a well-established and updated

environmental impact metric. The results generated from the ReCiPe and waste exergy

methods were most sensitive to changes in system manufacture and emission levels.

This method also allowed the user to decide how important environmental impact was in the

overall system selection process.

2. Method of Analysis

The analysis strategy applied in this study follows the same conceptual philosophy used by

Banerjee & Tierney (2011) [5] in its study. This occurred because the specificities of the



Page | 439

processing unit of ethanol. In this context, the working method comprised four steps: a)

Setting of ethanol production scenarios based on alternatives available of agricultural raw

materials, thermal energy and electricity; b) Energy modeling of each scenario; c)

Environmental modeling of scenarios from the LCA technique; and d) Combined analysis of

performances Energo and Exergo-environmental of the process.

The plant was designed to produce anhydrous ethanol from both sugarcane and corn. If

processing occurs with sugarcane, thermal and electrical energy are produced by

cogeneration in Rankine cycle from burning bagasse. The surplus bagasse is marketed. If

the unit operates with corn, wood chips meet the energy demand. As an alternative to the

cogeneration, it is possible to purchase electricity from the concessionaire. The option is

used in situations of discontinuing supply of chips. The operating status more attractive for

the managing company of the plant comprises an associated production of ethyl alcohol from

sugarcane and corn (85:15). The alternative makes the unit operative at full capacity for

eleven months of the year. Four scenarios were established for the study according to these

conditions. They are indicated in Table 1.

Table 1: Scenarios for ethanol production in terms of raw materials and energy utilities.

Case scenario Description

I Ethanol from sugarcane + steam production by burning bagasse + electricity by Rankine cycle

II Ethanol from corn + steam generated from burning wood chips + electricity via Rankine cycle

III Ethanol from corn + steam obtained from burning wood chips + electricity purchased from the concessionaire

IV Sugarcane ethanol followed by corn ethanol + steam from burning bagasse and wood chips + electricity via Rankine cycle

Cumulative Energy Demand (CED) aims to investigate the energy use throughout the life

cycle of a good or a service. This includes the direct uses as well as the indirect or grey

consumption of energy due to the use (e.g. construction materials, raw materials, etc.) [6, 7].

CED can also be used as a screening indicator for environmental impacts [8].

Due to the existence of diverging concepts and of the absence of a clear basis for the

characterization of the different primary energy carriers, CED-indicators are split up into two

categories: renewable and non-renewable resources. Renewable resources are organized

into the subcategories: fossil, nuclear and primary forest. Non-renewable resources are

divided into biomass, wind, solar, geothermal and water. Common to all categories is the

thesis that all energy carriers have an intrinsic value. This intrinsic value is determined by the

amount of energy withdrawn from nature.

However, the intrinsic value of energy resources need not be comparable across the

subcategories. CED is calculated per unit process by the expression presented of (Eq. 1).

i

ikCED iEn (1)

CED = cumulative energy demand per unit process (MJeq) mi = amount of material resource i (kg; m3; MJ) En,i = intrinsic energy per amount (kg; m3

; MJ) of substance i (MJeq/(kg; m3; MJ))



Page | 440

Classical analysis treat all forms of energy as equivalent, without distinguishing between the

various types of energy crossing the system boundary. These approaches are based on the

First Law of Thermodynamics. However, the usable work potential supplied to the system –

exergy – that has been consumed or destroyed due to irreversibility during the process is not

considered [9]. Combining the 1st and 2nd Laws of Thermodynamics the exergy analysis

identifies the distribution of the irreversibility of a plant among all the components that make

up the process. It is able to identify the most significant contribution to the overall inefficiency

of a unit. Exergy analysis provides a Thermodynamic diagnosis from which can be applied

engineering actions to improve the overall efficiency of the process [9].

The indicator Cumulative Exergy Demand (CExD) amounts the life cycle exergy demand of

product or process [10]. Exergy is stored in resources as chemical, thermal, kinetic, potential,

nuclear and radiative energy. The assignment of the adequate type of exergy depends on

resource use [11]. CExD assess the total exergy requirement of a product by the sum of

exergy of all resources used to its obtaining (Eq.2). CExD is calculated by adding up the total

exergy requirement of a process over a time-period. The exergy requirement of one unit of

process output was then obtained by dividing the total exergy requirement by the number of

unit outputs during this time-period [12].

j

),,,,(ji

),(i nmCExD jtnpkeexich rEx (2)

CExD = cumulative exergy demand per unit process (MJeq) mi = mass of material resource i (kg) Ex(ch),i = exergy per kg of substance i (MJeq/kg) nj = amount of energy from energy carrier j (MJ) rex – e(k,p,n,r,t),I = exergy to energy ratio of energy carrier j (MJeq/MJ) ch = chemical k = kinetic p = potential n = nuclear r = radiative t = thermal exergy

The emergence of life cycle databases enables and facilitates a product-specific approach,

since such databases provide the resource demand for each unit process. Hence, improved

CExD scores are calculated to indicate the exergy demand of a single product directly.

CExD is specified in MJ equivalents (MJeq) to emphasize that it is an impact assessment

indicator and not an inventory elementary flow [10].

CED and CExD are regarded as safe approaches in terms of cycle life, but they do not

replace an assessment with the help of comprehensive impact assessment methods [8]. In

order to evaluate the environmental performance of each scenario it was carried out an LCA

study. The product system comprised agricultural and industrial stages. The Life Cycle

Impact Assessment (LCIA) was conducted by method ReCiPe. Its application occurred both

according a problem-oriented approach – midpoint – and by a damage-oriented approach –

endpoint [13]. Other technical conditions of the LCA approach are presented later.



Page | 441

The final step of the method consists of the Energo and Exergo-environmental performance

analysis of the process, which was developed in two level. In the first level, we proceeded to

a broad discussion involving simultaneously the three dimensions selected. The second level

of evaluation covered each approach separately, deepening the identification of potential

causes for energetic and exergetic demands and for the environmental impacts identified.

3. System description

The systems of production of ethanol from sugarcane and corn are depicted in Figures 1 and

2. The agricultural models follow the technological and operational procedures practiced in

the Brazilian state of Mato Grosso. The prevalent biome in that region is the Brazilian

Savanah and the cultivation and both sugarcane and corn corps advanced on natural forest.

The average productivities of sugarcane and corn in 2012 are respectively of 65ton/ha and

6ton/ha.

Fig.1: Ethanol production from sugarcane Fig. 2: Ethanol production from corn

Soil fertilization involves the application of sources of nitrogen and potassium. For purposes

of nutrient balance, it was taken into account in sugarcane cultivation the reuse of filter cake

and vinasse. Boiler ashes were accounted for both cases.

In addition, it is necessary to adjust soil acidity. Plagues control occurs in a rigorous way for

sugarcane and corn because of the diversity of predatory species in the region. Agricultural

machines are used in operations of land preparation, cultivation and harvesting. The

calculation adopted in the simulations of scenarios involving sugarcane ethanol were based

on an autonomous distillery of 2.2Mton of crushing capacity for 180 days. The industrial



Page | 442

productivity in this case is about 83.3 L ethanol/ton sugarcane. When using corn, the plant

processing capacity reaches 400 ton/day with an average yield 359 L ethanol/ton corn.

Traditionally, cogeneration systems employed in Brazilian sugarcane mills are based on the

Rankine Cycle [14, 15, 16, 17]. Sugarcane bagasse is used as fuel to supply thermal

mechanical and electrical demand of the sugar and ethanol production process. These plants

used to operate with steam at low levels of pressure and temperature (20 bar and 300C)

and back pressure steam turbines, resulting in low energetic efficiency, high bagasse

consumption, low bagasse surplus and a small electricity surplus, or even none. Table 2

shows a basic configuration of the Rankine Cycle used in cogeneration systems employed in

Brazil. The simulation of the scenarios considered the same parameters.

Table 2. Main parameters adopted in simulation of Rankine Cycle

Ethanol source Parameters Value

Sug

arca

ne

Sugarcane processed (ton. /h) [17] 500 Anhydrous ethanol production 99.5% (L /ton) [18] 83.3 Bagasse content (ton. /ton.) [14, 16, 19] 0.285 Moisture of bagasse (%) [17, 18, 19] 50 Lower heating value of bagasse (kJ/kg bagasse) [17, 18] 7565 Electric power demand in the process (kWh/ton.) [18, 19] 28

Cor

n

Corn processed (ton/h) [20] 25 Anhydrous ethanol production 99.5% (L/ton.) [20, 21, 22] 360 DDGS production (kg/L C2H6O) [20, 22] 1.82 Moisture of wood chips (%) [22] 28 Lower heating value of wood chip (kJ/kg wood chips) [22] 14400 Electric power demand in the process (kWh/ton.) [22] 125

Cog

ener

atio

n

syst

em

First law efficiency of boiler (%) [15, 22] 75 Isentropic efficiency of condensation turbine (%) [15, 18] 70 Isentropic efficiency of mechanical drive turbine (%) [16, 17] 55 Efficiency of electrical generator (%) [16, 17] 98 High pressure steam (bar) [15, 17, 22] 22 Extracted steam (bar) [18, 19] 6 Low pressure steam (bar) [18, 19] 2.5 Condensate losses (%) [16] 9 Boiler purge (%) [16] 3

4. Conceptual bases for LCA

The LCA study was carried out according from the theoretical registration described by ISO

14044 (2006) [23]. In terms of Objectives Definition, the initiative proposes to carry out an

environmental analysis for production of anhydrous ethanol and considering the scenarios

above established. Regarding Scope Definition, it was established as Functional Unit: “to

produce 1.0 liter of anhydrous ethanol in an autonomous distillery”. The product system

comprised agricultural stages - production of cane sugar, corn, and wood for energy - and

industry - the sugarcane crushing, cooking corn, fermentation, distillation and dehydration of

ethanol.

All the units of the cogeneration system were also included. The modeling of all agricultural

stages and part of the industrial production was carried out with primary data. Secondary

data were used to fulfill some lack of data or in order to evaluate the consistence of the Life

Cycle Inventory (LCI). The Temporal Coverage comprised the time-period of 2011-2012. It



Page | 443

was defined the State of Mato Grosso as the Geographical Coverage. Finally, the technical

limitations and assumptions presented in the system description defined Technological

Coverage. Infrastructure processes were excluded but long-term emission were considered

at the analysis. The single allocation made in the study – between ethanol and bagasse –

was held in terms of energy content criterion.

As mentioned before for Life Cycle Impact Assessment (LCIA) was realized by applying the

method ReCiPe – version 1.07 both for midpoint and endpoint indicators [13]. All of its impact

categories were observed in order to determine of the environmental profiles. All of ReCiPe

impact categories were took into account, with the exception of Ozone Depletion,

Photochemical oxidant formation, Ionization Radiation, Marine Eutrophication and Marine

Ecotoxicity, Urban Soil Occupation and Metal Depletion whose contributions are not relevant

for the purpose of diagnosis formulated by this study.

5. Results and Discussion

The overall results of the Energo and Exergo-environmental analysis for the alternative of

ethanol production, expressed in terms of single scores are presented in Figure 3.

Fig.3. Composition of Energy and Environmental analysis for the scenarios under study

The production of ethanol from sugar cane – scenario I – demonstrates the worst

environmental performance (0.26 Pt /L) from among the options evaluated. Regarding CED

this scenario proves to be more appropriate than any of options in which corn is used for

obtaining the fuel – scenarios II and III. Sugarcane ethanol also shows unsatisfactory results

as CExD (12.8 MJeq/L). Its score only exceeds the 19.2 MJeq/L of corn ethanol whose

electricity demand is supplied by the Brazilian grid. Should be noted the influence of ethanol

from sugarcane in the performance scenario IV – in which ethanol is obtained from both

agricultural raw materials – in every dimensions analyzed. The effect is justifies by the

sugarcane participation in the arrangement (85%) and the small amount of surplus bagasse

that is used to produce energy during corn processing. Structural changes in the power



Page | 444

supply system – such as the production of steam at pressures of 90–100bar – associated

with larger volumes of crushed sugar cane – upper than 4.0Mton/crop – could lead to

improve both environmental and energy performances of this scenario.

For the situation that all the approaches are evaluated simultaneously, scenario II presents

the most balanced Energo and Exergo-environmental performances. Although its CED

(1.16MJ/ L) is among the highest, the environmental (0.147 Pt/ L) and exergetic scores

(7.02MJeq/ L) are quite favorable.

5.1 Performances in terms of Cumulative Energy Demand (CED)

Table 3 presents performances discretized by CED indicator for each analyzed scenario.

From these results, it can be seen that the most important contributions in all scenarios

occurred respectively in demands of fossil resources and renewable biomass and water.

In scenario I, the contribution of fossil fuels accounted for 85% of total energy demand. This

is justified due to diesel consumption transport and agricultural machinery. Following is the

demand of renewable water, whose participation (of 13.3%) is due to the contribution of

hydropower in the Brazilian energy mix. According to the Brazilian Energy Balance 2012, the

modal hydropower contributes 81.9% of the energy consumed in the country [24]. For

scenario I this contribution occurs in the form of electrical consumption for the production of

agricultural chemical additives – fertilizers and pesticides – and industrial products – such as

sulfuric acid and vitamins – used in the fermentation process.

Table 3. Energy performance of the analyzed scenarios: CED – Single score – v. 1.08

As expected, stands out for scenario II the demand for renewable biomass (84.5%) due to

the source of energy in the ethanol plant to be wood chips. To this, add up 12.5% of

contributions arising from fossil fuels, enhanced by consumption of diesel in agricultural

machinery for corn cultivations and timber. The option for providing electricity from the

Brazilian energy mix resulted in a change of profile of energy demand modals to scenario III

when it compares to scenario II. Thus, the contribution in terms of renewable water, that

relative to scenario II was 2.81%, achieved 10.2% in scenario III due to the previously

mentioned share of hydroelectricity in the local array. Thus, non-renewable fossils and

biomass - whose absolute values remained constants - had amended their relative

contributions to 10.2% and 78.2%. Scenario IV is highly influenced by the participation of

cane sugar and corn on the arrangement established by the company managing the plant to

produce ethanol. As pointed out before, this weighting is favorably compensated by reusing



Page | 445

surplus bagasse. However, the amount of this energy input originated by the production

model currently practiced – of 10.2kg/ton sugarcane – is not sufficient to influence the overall

result. Anyway, scenario IV had the best score as CED in comparison to the other conditions.

5.2 Performances in terms of Cumulative Exergy Demand (CExD)

Table 4 compare the assessed scenarios regarding to CExD single scores. Preliminarily it is

important to highlight the surplus of accumulated energy demand afforded due to the

irreversibility – i.e. the Exergy Demand – of the various processing that make up each

product system as compared with the cumulative energy demand. The energy surplus can

be calculated by the ratio [CExD / CED] earned in each case.

Table 4. Energy performance of the analyzed scenarios: CExD – Single score – v. 1.02

The best ratio is the scenario II for which [CExD / CED] = 6.1. Already the most critical cases

refer to scenarios I and IV, whose Exergy Demand surpasses the Energy Demand

respectively in 25.8 and 23.0 times. CExD quite significant values for all scenarios under

consideration should be the cumulative contributions of renewable water, which appear in all

product systems every time the Brazilian energy grid is somehow involved throughout the life

cycle. The indicator Cumulative Exergy Demand – v 1.02 assigns equivalence factor f = 50,

for ‘water use in turbine’ – that is the specific case of hydropower plants – regardless of the

provenance of the resource [8, 10]. As other environmental burdens considered in LCIs

whose equivalence factors could be larger than ‘water used in turbine’ has contributions less

significant, its effect on the overall result of the analysis stands out. Therefore, the

contributions of renewable water indicator for scenarios I through IV are respectively 95.9%,

82.7%, 92.4% and 95.4%. This percentage associated with contribution of 6% of renewable

biomass - used for the generation of thermal energy into the distillery - explain why scenario

III expresses the worst performance concerning CExD.

5.3 Environmental Performances (ReCiPe)

The environmental performance of the alternative scenarios for ethanol production was

evaluated by method ReCiPe – v.1.07 of LCIA. To allow a more detailed discussion on



Page | 446

contributions identified in each situation Table 5 shows the results of applying the method on

the midpoint level (i.e., a problem-oriented approach).

Table 5. Environmental performance of the analyzed scenarios: ReCiPe midpoint (M) – v. 1.07

As discussed previously ethanol obtaining from sugar cane – scenario I – results in the most

adverse effects on the environment. Moreover, scenarios II and III achieved the best

performance. Scenario I got the worst results as compared with the others into five of the

categories evaluated. They are Terrestrial Acidification (TA), Human toxicity (HT), Particulate

matter formation (PMF), Water depletion (WD) and Fossil depletion (FD). Moreover, scenario

III showed the worst performance in terms of Freshwater eutrophication (FEu), Ecotoxicity

Terrestrial (ET), Freshwater ecotoxicity (FEc), Agricultural land occupation (ALO) and Natural

land transformation (NLT).

Regarding Climate Changes (CC) the difference between the best and the worst results was

2.47%. The range was also considered of little significance, which makes equivalent the four

scenarios evaluated in terms of performance. Specifically for CC, it were the predominant

emissions of CO2 and N2O occurred respectively during sugarcane and corn cultivation. The

increase in short-term CO2 emissions was due to the suppression Savanah native to the

deployment of sugarcane crop in the region.

Impact on TA derive from emissions of NOx and NH3 from soil fertilization. Contrary to what

one might imagine the use of N fertilizers for the production of energy wood did not provide

significant environmental burdens. Thus, scenarios II and III were benefited.

Pesticides used for plague control causes impacts related to FEC and TE. For sugarcane

these impacts occur from losses of diuron, which contribution for FEC and TE are 38.2% and

79.5%. For corn, there are significant contributions (93% and 96%) due to emissions of

atrazine in fresh water and soil. Again in the case of HT it was observed that the more

effective contributions in terms of impact occurs on the agricultural stage due to the emission

of metals such as Zn and Hg introduced into the biotic system from pesticides. However, the

estimates made to evaluate these losses need to be deepened so that an accurate diagnosis

could be made. Emission of PM associated with sugarcane ethanol derived from biomass

burning for land clearing. For corn ethanol this impact arises from the atmospheric emissions

of ash in the boiler. WD impacts are associated with the consumption of water process and



Page | 447

irrigation. FD concentrates on crude oil depletion to produce diesel used in transport and

agricultural machinery.

Finally, it should be evident that the contributions for all scenarios in terms of FEu, ALO, and

NLT – of about micro-parts / L. ethanol – were considered less significant. Anyway, the

explanation for the fact that the progress of cultivation areas on Savannah native had not

been reversed in drastic environmental consequences is because the average rate of land

transformation – expressed per amount of ethanol – for sugarcane and corn are of just

2.64m2 / L and 4.65m2 / L.

6. Conclusions

The method of work that combines a standard environmental impact indicator (ReCiPe) with

energetic and exergetic analysis allowed that a precise evaluation of the scenarios under

study was performed. For the situation that all the approaches are evaluated simultaneously,

scenario II – ethanol from corn and cogeneration – presents the most balanced Energo and

Exergo-environmental performances. Scenario III was deeply influenced by the effects

provided by the Brazilian energy grid. This dependence became unfeasible the alternative.

The performance of scenario IV that combines sugarcane and corn was strictly dependent on

the performance of ethanol production from sugarcane. It is recommended that additional

tests of proportioning between agricultural assets may be carried out.

References

[1]. A. Boyano, T. Morosuk, A.M. Blanco-Marigorta, G. Tsatsaronis (2012). Conventional and advanced exergoenvironmental analysis of a steam methane-reforming reactor for hydrogen production. In: J Cleaner Production, 20 (1), pp.152–160.

[2]. L. Talens Peiró, L. Lombardi, G. Villalba Méndez, X. Gabarrell i Durany (2010). Life cycle assessment (LCA) and exergetic life cycle assessment (ELCA) of the production of biodiesel from used cooking oil (UCO). In: Energy, 35 (2), pp. 889–893.

[3]. A. Ozbilen, I. Dincer, M. A. Rosen (2012). Exergetic life cycle assessment of a hydrogen production process. Int. J Hydrogen energy, 37 (7), pp. 5665–5675.

[4]. C. Koroneos, M. Tsarouhis (2012). Exergy analysis and life cycle assessment of solar heating and cooling systems in the building environment. In: J Cleaner Production 32 (4), pp. 52–60.

[5]. A. Banerjee, M. Tierney (2011). Comparison of five exergoenvironmental methods applied to candidate energy systems for rural villages in developing countries. In: Energy, 36 (5), pp. 2650–2661.

[6]. I. Boustead and G.F. Hancock (1979). Handbook of Industrial Energy Analysis. Ellis Harwood Ltd.

[7]. D. Pimentel (1973). Food Production and the Energy Crisis. In: Science, 182(4111) p. 443-9.

[8]. R. Frischknecht R., N. Jungbluth N., et.al. (2007). Implementation of Life Cycle Impact Assessment Methods. Final report Ecoinvent v2. Swiss Centre for LCI. Dübendorf, Switzerland, pp. 31-38.

[9]. T.J. Kotas (1985). The Exergy Method of Thermal Plant Analysis, Butterworth, London.

[10]. M. Bösch, S. Hellweg, M. Huijbregts, R. Frischknecht (2007). Applying Cumulative Exergy Demand (CExD) Indicators to the Ecoinvent Database. In: Int J LCA, 12 (3), pp. 181–190.

[11]. J. Szargut (2005). Exergy method: Technical and ecological applications. WIT Press, Southampton.



Page | 448

[12]. J. Szargut, D.R. Morris, F.R. Steward (1988). Exergy analysis of thermal, chemical and metalurgical processes. Hemisphere Publishing Corporation, NewYork.

[13]. M. Goedkoop, R. Heijungs, M. Huijbregts, A.De Schryver, J. Struijs “Description of the ReCiPe methodology for Life Assessment Impact Assessment”, http://www.lcia-recipe.net retrieved mach. 2013.

[14]. M.O.S. Dias, M. Modesto, A.V. Ensinas, S.A. Nebra, C.E.V. Rossell (2011). Improving bioethanol production from sugarcane: evaluation of distillation, thermal integration and cogeneration systems. In: Energy 36 (6), pp. 3691-3703.

[15]. A.V. Ensinas, S.A. Nebra, M.A. Lozano, L.M. Serra (2007). Analysis of process steam demand reduction and electricity generation in sugar and ethanol production from sugarcane En. Conv. Manag., 48 (11), pp. 2978–2987.

[16]. M.O.S. Diasa,M. P. Cunha, C.D.F. Jesus, M.I.G. Scandiffioa, C.E.V. Rossell, R. Maciel Filho, A. Bonomi (2010). Simulation of ethanol production from sugarcane in Brazil: economic study of an autonomous distillery. In: 20th European Symposium on Computer Aided Process Engineering – ESCAPE20. S. Pierucci and G. Buzzi Ferraris (Editors). Elsevier.

[17]. M.O.S. Dias, A.V. Ensinas, S.A. Nebra, R. Maciel Filho, C.E.V. Rossell, M.R.W. Maciel, (2009). Production of bioethanol and other bio-based materials from sugarcane bagasse: Integration to conventional bioethanol production process. In: Chem. Eng. Res. Des., 87 (9), pp. 1206-1216.

[18]. J.C.E. Palacio. Análise Termoeconômica Integrada das Plantas de Cogeração e Etanol em uma Destilaria Autônoma. PhD Thesis. (Energy Conversion). School of Mechanical Engineering, Federal University of Itajubá (in Portuguese), 2010.

[19]. R.A.Romão Júnior. Análise da Viabilidade do Aproveitamento da Palha da Cana de Açúcar para Cogeração de Energia numa Usina Sucroalcooleira. Master Thesis. (Thermal Sciences). School of Mechanical Engineering, Paulista State University – UNESP (in Portuguese), 2009.

[20]. R. K. Perrin, N. F. Fretes, J.P. Sesmero (2009). Efficiency in Midwest US Corn Ethanol Plants: A Plant Survey. In: Energy Policy 37 (4), pp. 1309–1316.

[21]. P. McKendry (2002). Energy production from biomass (part 1): overview of biomass. In: Bioresource Technology 83 (3), pp. 37–46.

[22]. D. Pimentel, T. W. Patzek (2007). Ethanol Production Using Corn, Switchgrass, and Wood; Biodiesel Production Using Soybean and Sunflower. In: Natural Resources Research, 14 (1), pp. 65-76.

[23]. International Organization for Standardization. ISO 14044, Environmental management -- Life cycle assessment -- Requirements and guidelines. Genève. (2006) 52 p.

[24]. Brazilian Energy Balance 2012 Year 2011. Empresa de Pesquisa Energética: EPE, 2011. Rio de Janeiro, 282p.

exergy and industrial systems - smart-er.eu · 3rd international exergy, life cycle assessment, and...

Documents