journal - dl.uctm.edu · new techniques that are currently being developed include capillary...

VOLUME 48 2013 ISSUE 3

www.uctm.edu ISSN 1311-7471 (print)

ISSN 1314-7978 (on line)

JOURNAL

OF CHEMICAL TECHNOLOGY

AND METALLURGY

SOFIA

Journal of Chemical Technology and Metallurgy

The Journal of Chemical Technology and Metallurgy started originally in 1954 as Annual Journal of the former Higher

Institute of Chemical Technology. It ran in Bulgarian. In 2000 its name was changed to Journal of the University of

Chemical Technology and Metallurgy. It was published quarterly in English. Since 2013 it will be run bimonthly as

Journal of Chemical Technology and Metallurgy.

Honorary Editor

R. Dimitrov

Editor-in-Chief

B. Koumanova

University of Chemical Technology and Metallurgy,

8 Kl. Ohridski blvd., 1756 Sofia, Bulgaria

Tel: (+ 359 2) 8163 302, E-mail: [email protected]

7BEditorial Board

S. J. Allen

Queens University of Belfast, UK

D. Angelova

UCTM, Bulgaria

M. Bojinov UCTM, Bulgaria

J. Carda

University Jaume I, Castellon, Spain

G. Cholakov UCTM, Bulgaria

V. Dimitrov Bulgarian Academy of Sciences, Bulgaria

N. Dishovsky

UCTM, Bulgaria

S.J.C. Feyo de Azevedo Universidade do Porto, Portugal

M. Jitaru

University “Babeş -Bolyai”, Cluj-Napoca, Romania S. Kalcheva

UCTM, Bulgaria

F. Keil Hamburg University of Technology, Germany

T. Konstantinova

UCTM, Bulgaria

M. Kucharski AGH University of Science and Technology, Krakow, Poland

A. Mavrova

UCTM, Bulgaria

D. Mehandjiev Bulgarian Academy of Sciences, Bulgaria

V. Meško

International Balkan University, Skopje, Macedonia

L. Mörl University “Otto-von-Guericke”, Magdeburg, Germany B. Nath

European Centre for Pollution Research, London, UK

D. Pavlov

Bulgarian Academy of Sciences, Bulgaria

L. Petrov Bulgarian Academy of Sciences, Bulgaria

D.C. Shallcross

The University of Melbourne, Australia

M. Simeonova

UCTM, Bulgaria

V. Stefanova

UCTM, Bulgaria

D. Stoilova

Bulgarian Academy of Sciences, Bulgaria

N. Tsarevsky

Southern Methodist University, Dallas, Texas, USA

I. Turunen

Lappeenranta University of Technology, Finland

S. Vassileva

UCTM, Bulgaria

S. Veleva

UCTM, Bulgaria

L.Vezenkov

UCTM, Bulgaria

Ž. Živković

University of Belgrade, Technical Faculty, Bor, Serbia

Technical secretary: S. Georgieva

The annual subscription (for 6 issues) is 300 €, including postage, handling and packaging charge.

©UCTM, 2005. All rights reserved. Reproduction in part or whole without permission is strictly prohibited. Prepress and design: A. Staneva; Printing: S Print Ltd., Sofia

The articles of this journal are indexed and abstracted in EBSCO, SCOPUS, Chemical Abstracts and Referativnii Journal Himii (VINITI).

(www. http://www.uctm.edu)

TABLE OF CONTENTS

Biomimetic nanosensors for determination of toxic compoundsin food and agricultural products L. Yotova, S. Yaneva, D. Marinkova.............................................................................................................................215

New glucose oxidase - PAMAM conjugate as fluorescent biosensor matrixin acetylcellulose membraneD. Staneva, I. Grabchev, L. Yotova, R. Betcheva..........................................................................................................228

Impact of polysaccharides of malt on filterability of beer and possibilities for their reduction by enzymatic additivesG. Jonkova, A. Surleva.................................................................................................................................................234

An attempt for correlation between Mooney viscosity and rheological properties of filled rubber compoundsD. Zheleva....................................................................................................................................................................241

Determination of the effect of gray component replacement level on colorimetric characteristics of color proofI. Spiridonov, M. Shopova............................................................................................................................................247

Rapid spectrophotometric method for determination of hexamethylenetetramine (Urotropine) in foot care products A. Tachev, V. Christova-Bagdassarian, N. Vasileva, A. Dimitrova, M. Atanassova.........................................................254

The effect of introducing copper dopant on the photocatalytic activity of ZnO nanoparticlesK. Milenova, I. Stambolova, V. Blaskov, A. Eliyas, S. Vassilev, M. Shipochka..............................................................259

Ethane production plant for better energy integration and cost reduction in JordanMenwer Attarakih, Mazen Abu-Khader, Tahani Saieq, Hans-Jörg Bart.......................................................................265

Heat saving in evaporative crystallization by introducing a heat pumpZaid Ahmed Al-Anber.................................................................................................................................................277

Synthesis of alumina porous supports via different compaction routes: vibration and pressingMohammad Javad Ghaderi, Mahdi Shafiee Afarani, Ghodratollah Roudini...................................................................289

Optimization of the basic parameters of cathodic deposition of Ce-conversion coatings on D16 AM clad alloyS. Kozhukharov, J.A.P. Ayuso, D.S. Rodríguez, O.F. Acuña, M. Machkova, V. Kozhukharov......................................296

Stationary and pulse electrodeposition of CoNi and CoNiCu coatingsK. Ignatova, Y. Marcheva.............................................................................................................................................308

Crystallization in the three-component systems Rb2SeO4 - MeSeO4 - H2O (Me = Mg, Ni, Cu) at 25°CV. Karadjova..................................................................................................................................................................316

Guide for Authors..........................................................................................................................................................326

L. Yotova, S. Yaneva, D. Marinkova

215

Journal of Chemical Technology and Metallurgy, 48, 3, 2013, 215-227

BIOMIMETIC NANOSENSORS FOR DETERMINATION OF TOXIC COMPOUNDS IN FOOD AND AGRICULTURAL PRODUCTS

(REVIEW)L. Yotova, S. Yaneva, D. Marinkova

University of Chemical Technology and Metallurgy8 Kl. Ohridski, 1756 Sofia, BulgariaE-mail: [email protected], [email protected]

ABSTRACT

In the recent years many types of biosensors have been developed and used in a wide variety of analytical settings with applications in biomedicine, health care, drug design, environmental monitoring, detection of biological, chemical and toxic agents etc.

In the field of sensor research and development, bionanotechnology is poised to make significant contributions and has the potential to radically alter the way sensors are designed, constructed and implemented.

The most recent unrestrictive techniques used in biosensor design are surface plasmon resonance (SPR) and quartz crystal microbalance (QCM), combined with a biological receptor. On that basis several enzymes and microbial cells have been immobilized onto newly synthesized hybrid membranes. They contain particles from SiO2, TiO2, polymers and dendrimers and are attached to the sensor’s surface. They are applied for determination of xenobiotics in crops, milk, nuts and other food and agricultural materials.

The aim of this review is to summarize some of the most significant research achievements in the field of biomimetic nanosensors design based on immobilized tyrosinase for determination of toxic compounds and smart biosensors for determination of Mycotoxines.

Keywords: optical biosensors, tyrosinase, peroxidase, multy enzime systems.

Received 03 March 2013Accepted 05 May 2013

INTRODUCTION

Tyrosinases (monophenol, o-diphenol:oxygen oxido-reductase, EC 1.14.18.1) belong to a larger group of proteins named “type-3 copper proteins”, which includes the catecholoxidases from plants and the oxygen-carrier haemocyanins from molluscs and arthropods [1, 2].

Although tyrosinases are widely distributed in microorganisms, plants and animals, much of the current interest in the development of their biotechnological applications has focused on the use of mushroom tyrosinases. Several aspects of mushroom tyrosinases, such as their biochemical characteristics, their roles in the metabolism of the producing organism and some of their potential biotechnological applications have been

extensively reviewed [3-6].Phenols, due to their toxicity, persistence and common

occurrence in the biosphere, are one of the most important groups of ecotoxins. These compounds, as ingredients (components) and precursors of other chemicals including polymers, solvents, dyes (aminophenols), explosives (nitrophenols), surfactants (alkylphenols) or drugs are in common use [7]. Wastewaters containing phenols and phenolic derivatives are generated by the textile, coal, chemical, petrochemical, mining and paper industries, amongst others [8-10]. Increasingly strict environmental laws are providing an impetus for the development of analytical techniques for fast monitoring of these compounds.

Traditionally, analysis has been based on spectro-

Journal of Chemical Technology and Metallurgy, 48, 3, 2013

216

photopmetric [11] or chromatographic [12] methods. New techniques that are currently being developed include capillary electrophoresis [13], immunoassays [14] and biosensors [15-18]. They potentially provide better specificity, lower costs, faster and simpler sample processing. Biosensors for the detection of phenolic compounds, based on the reaction of these compounds with immobilized mushroom tyrosinase are currently being developed.

In food quality control, biosensors have already confirmed their potential usefulness as tools for the detection of several types of compounds of interest: carbohydrates, alcohols, phenols, carboxylic acids, amino acids, biogenic amines, heterocyclic and inorganic compounds, additives or contaminants. The most common enzyme used for this purpose is tyrosinase. When this enzyme is entrapped in different supports, it is able to react with polyphenols to measure their concentration in samples such as olive extracts, tea [19], wine [20] and beer [21].

Mycotoxins are biological pollutants. They are toxic metabolites produced by several fungi in foods and feeds, and are probably the best known and most intensively examined mycotoxins in the world.

Mycotoxins are non volatile, relatively low-molecular mass secondary metabolic products that may affect exposed persons in a variety of ways. These compounds are considered secondary metabolites because they are not necessary for fungal growth and are simply a product of the primary metabolic processes.

The functions of mycotoxins have not been totally studied, but it is established that they play a key role in the antagonistic processes concerning microorganisms from the same environment. They are also believed to help parasitic fungi invade host tissues. The amount of toxins needed to produce adverse health effects varies widely among toxins, as well as to each person’s immune system.

Some mycotoxins are carcinogenic, some are vasoactive, and some cause damage to the central nerve system. Often, a single mycotoxin can cause more than one kind of toxic effect. More than 240 fungi produce 100 toxic compounds, which cause lancination and chronic diseases called mycotoxicoses [22].

The latest experiments with environmental biosensors are concentrated on molecular affinity reactions such as photosynthetic systems, antibody actions. The antibody-antigen interactions are preferred in low concentration measurements. There are immunochemical biosensors applied for mycotoxin assays. Gaag at al. developed a method based on surface plasmon resonance for measuring of four different mycotoxins – aflatoxin B1, zearalenone, ochratoxin A and fumarotoxin B1 [23]. The scheme below (Fig. 1) presents the main components in biomimetic nanosensors design and application.

Immobilization is essential because it ensures intimate contact between the enzyme and the underlying signal detector and also prevents the enzyme from being washed off the electrode when readings are made in aqueous samples. One of the biggest challenges in

Fig. 1. A schematic view of the basics in biomimetic nanosensors design and application.


217

designing a new enzyme-based biosensor is to find the optimal balance between stability and activity of the enzyme [24]. Surfaces play an important role in biology and medical research, since most biological reactions occur on surfaces and interfaces [25]. Tailored surface properties such as tunable reactivity, biocompatibility or wettability could be obtained by different approaches of surface modification, so that the design of biofunctional surfaces is of great interest in bioanalysis research [26]. Immobilization methods such as cross-link bonding [27], covalent attachment [28], polymer inclusion [29] and sample adsorption [30], have different effectiveness with regard to the stability of the signal generated, because a higher or lower percentage of enzyme immobilized is lost during the measurement. A good combination of support material and immobilization methods is of a fundamental importance to achieve the desired performances from the sensing system [31].

1. Gold nanoparticles for biosensors design Bionanoconjugates of the enzyme tyrosinase

(TYR) and gold nanoparticles (AuNPs) functionalized with a peptide (Cys-Ala-Leu-Asn-Asn) CALNN were produced in solution and characterized. The conjugation of enzymes with AuNPs can lead to the retention or even - to an increase of their biological stability/activity. Electron transfer between the catalytic sites of immobilized enzymes and the electrode materials is facilitated, improving the analytical sensitivity and selectivity of the biosensors and often overcoming the need for enzyme mediators. These properties have led to an intensive study of AuNPs in the construction of electrochemical biosensors with enhanced analytical performance in respect to other biosensor designs. The enzyme activity of the BNCs was stable for 6 weeks, with a loss of activity towards 4-methyl catechol of only 10 %. This indicates that these conjugates present a good level of storage stability for preparation and assembly of the biosensor at a later stage, suggesting that they will also be stable after immobilization onto the electrode surface [32].

An indirect colorimetric method is available for detection of trace amounts of hydroquinone, catechol and pyrogallol. The reduction of AuCl4

− to gold nanoparticles (Au-NPs) by these phenolic compounds in the presence of cetyltrimethylammonium chloride (CTAC) produces very intense surface plasmon resonance peak of Au-NPs. The plasmon absorbance of Au-NPs allows the

quantitative colorimetric detection of the phenolic compounds. The calibration curves, derived from the changes in absorbance at λ = 568 nm, were linear with concentration of hydroquinone, catechol and pyrogallol in the range of 7.0 × 10−7 to 1.0 × 10−4M, 6.0 × 10−6 to 2.0 × 10−4 M and 6.0 × 10−7 to 1.0 × 10−4 M, respectively [33].

A. Carralero and co-authors report for the development of a new tyrosinase biosensor which is based on a construction of graphite–Teflon composite electrode matrix in which the enzyme and colloidal gold nanoparticles are incorporated by simple physical inclusion. The Tyr–Aucoll–graphite–Teflon biosensor exhibited suitable amperometric responses at −0.10 V for the different phenolic compounds tested (catechol; phenol; 3,4-dimethylphenol; 4-chloro-3-methylphenol; 4-chlorophenol; 4-chloro-2-methylphenol; 3-methylphenol and 4-methylphenol). The limits of detection obtained were 3 nM for catechol, 3.3 μM for 4-chloro-2-methylphenol, and approximately 20 nM for the rest of phenolic compounds [34]. A sequential competitive configuration between the analyte and progesterone, labeled with alkaline phosphatase (AP), was used. Phenyl phosphate was employed as the AP-substrate and the enzyme reaction product, phenol, was oxidized by tyrosinase to o-quinone, which is subsequently reduced at −0.1 V at the biocomposite electrode. The response shows a good linearity for low progesterone concentrations (it should be noted that usual progesterone concentration in cow`s milk is around 5 ng mL-1). Conversely, some of the progesterone immunosensors described previously exhibited a hook effect, i.e. a curvature in the calibration graph, for low antigen concentrations. Although this hook effect was not observed for the calibration graph constructed with the immunosensor using a colloidal gold-graphite-Teflon composite electrode, the monitoring of the affinity reaction was accomplished in this case at +0.30 V. On the contrary, the use of a Tyr-Aucoll – graphite – Teflon biosensor as transducer allows a detection potential such as – 0.1 V to be applied, thus minimizing the impact of potential interferences from electrochemically active compounds that may be present in a real sample [35].

Sanz and co-workers report for immobilization of the tyrosinase onto a glassy carbon electrode modified with electrodeposited gold nanoparticles (Tyr-nAu-GCE). The enzyme immobilized by cross-linking with glutaraldehyde retains a high bioactivity on this


218

electrode material. The Tyr-nAu-GCE was applied for the estimation of the phenolic compounds content in red and white wines [36].

Penicillamine (PCA) is a sulfhydryl amino acid with a hydrogen ion in the beta-carbon of cysteine replaced by a methyl group. In addition to chelation of heavy metals, such as copper, it suppresses the cross-linking of collagen by formation of a thiazolidine bond with the aldehyde group of collagen. Gold electrodes were modified with mono layers of 3-mercaptopropionic acid and further reacted with poly-(amidoamine) (PAMAM) dendrimers to obtain thin films. The high affinity of PAMAM dendrimer for nano-Au due to its amine groups is used to ensure the role of nano-Au as an intermediator to immobilize the enzyme of tyrosinase. When penicillamine was added to the solution, it reacted with o-benzoquinone to form the corresponding thioquinone derivatives, which resulted in decrease of the reduction current of o-benzoquinone. Based on this, a new electrochemical sensor for determination of penicillamine was developed. Fig. 2 describes the basic strategy for the preparation of a tyrosinase-modified electrode.

When the tyrosinase was immobilized on the nano-Au surface, the AFM image indicates that the enzyme film is even more uniform and smooth than that of the nano-Au film. Cyclic voltammetry (CV) is

an effective method for probing the feature of surface-modified electrode and testing the kinetic barrier of the interface because the electron transfer between solution species and the electrode must occur by tunneling either through the barrier, or through the defects in the barrier. Therefore, CV has been chosen as a marker to investigate the changes of electrode behavior after each assembly step [37].

2. Nanoparticles and nanomaterials for tyrosi-nase immobilization

The application of Quartz Crystal Microbalance (QCM) techniques in the characterization of the successive immobilization steps is involved in the development of bioanalytical platforms. A particular emphasis is placed on the application of these techniques to the characterization of the immobilization of enzymes on different modified and unmodified surfaces as well as on the study of protein interactions, which is a more recent and less spread application [38].

Authors report herein about a biocompatible interface composed by a beta-nanozeolite three-dimensional architecture on an indium tin oxide (ITO) electrode using a layer-by-layer (LbL) assembly technique. The large surface area and unique surface properties of the nanozeolite matrix resulted in a high enzyme adsorption capacity and the enzyme adsorbed in this film retained its activity to a large extent. By adjusting the nanozeolite-assembled layers, thus regulating the amount of the enzyme immobilized, the biocatalytic property of the enzyme electrode could be easily controlled [39].

ZnO nanoparticles, porous film, nanocombs and nanorods have been developed into biosensors to detect phenolic compounds. Notably, the isoelectric point (IEP) of ZnO is as high as about 9.5. This is suitable for immobilization of biomolecules with low IEP, such as enzymes and proteins, assisted by electrostatic attraction in proper pH value. Using zinc powder directly as a source material, ZnO nanorods were fabricated on the surface of gold wire by hydrothermal reaction, without any other surfactant and stabilizing agent. The gold wire was treated to improve the nucleation for growth of ZnO nanostructures and to further improve the performance of the biosensor, constructed by immobilizing tyrosinase on the ZnO nanorods for phenol detection. Electrochemical measurement, Fourier transform infrared (FT-IR)

Fig. 2. A schematic representation of the surface modification process and fabrication of a tyrosinase according to Li and Kwak [37].

AuMA

Au/MA Au/MA/PAMAM

Au/MA/PAMAM/nano-Au/Tyr

S

OHO

S

OHO

S

OHO

S

OHO

S

HNO

EDS

S

NHO

S

HNO

S

NHO

Au/MA/PAMAM

S

HNO

S

NHO

S

HNO

S

NHO

Nano-Au S

NO

S

NHO

S

NO

S

NHO

Tyr


219

and scanning electron microscopic (SEM) analyses demonstrated that the TYR was stably adsorbed on the ZnO nanorods surface with bioactivity for phenol oxidation. The film assembly process and enzyme adsorption were monitored by Quartz Crystal Microbalance measurements. The nanozeolite film exhibited an amazing adsorption capacity (about 350 mg.g-1) for tyrosinase as a model enzyme. The tyrosinase biosensor showed high sensitivity (400 µAmM-1), short response time (reaching 95 % within 5 s), broad linear response range from 10 nM to 18 µM, very low detection limit (0.5 nM) and high operational and storage stability - more than 2 months [40].

In [41] Dai and co-authors report ed that an electro-chemical biosensor for phenol based on immobilization of tyrosinase-peroxidase on mesoporous silica was constructed. An enhanced sensitivity of the tyrosinase-horseradish peroxidase based biosensor to phenol was observed in comparison with tyrosinase or horseradish peroxidase monoenzyme modified electrodes. The two enzymes retained their enzymatic activities for phenol determination without any mediator. The phenol sensor exhibited a fast response of less than 10 seconds. The sensitivity of the biosensor for phenol was 14 mA mM-1 cm2 with a linear range from 2x107 to 2.3x104 M and a detection limit of 4.1x109 M. MCM-41 has a high surface area, controlled porosity and mechanical resistance, which make it more suitable for enzyme loading to get high sensitivity [42]. Authors in [43]

reported that the activity of tyrosinase attached to acrylic beads and silica gels was low, despite the large protein content. A carbon nanotube matrix which was easy to prepare was developed by Alarcon and co-workers. It ensures a very good entrapment environment for the enzyme, being simpler and cheaper than other reported strategies. In addition, the proposed matrix allows for a very fast operation of the enzyme, which leads to a response time of 15 s for phenol. The biosensor keeps its activity during continuous flow injection analysis (FIA) measurements at room temperature, showing a stable response (RSD 5 %) within a two week working period at room temperature [44].

3. Sol-gel hybrid materialsThe sol-gel process allows for the preparation of

porous glasses at low temperature and high purity of the starting materials, for the production of ceramic materials

by hydrolysis and polycondensation of alkoxides [45]. TEOS (tetraethyl orthosilicate) was chosen as a precursor of sol–gel silicate for immobilization of a MBTH (3-methyl-2-benzothiazolinone hydrazone) reagent.

The properties of the sol-gel silicate matrix such as surface area, pore size and pore distribution can be affected by various factors. Among them, the pH and the Si to water ratios are the most important parameters for sol–gel silicate preparation. Nafion (sulfonated tetrafluoroethylene) has a hydrophobic fluorocarbon backbone but hydrophilic cation-exchange site. Thus, it exhibits moderate hydrophobicity, which can assist in retaining MBTH in the film and reduce leaching. The lowest response of the biosensor is observed when a pure sol-gel silicate matrix is used. This behavior could be due to the hydrophilic nature of TEOS that could not retain much of the hydrophilic MBTH reagent in the immobilization matrix. The hybrid material with a ratio of 1:1 showed the optimum biosensor response. Further increase in nafion content decreased the sensor response. This may attributed to the decrease of porosity of the sol-gel silicate film, as well as to the hydrophobicity of the matrix when the content of nafion in the hybrid materials increases. Thus, it affects the amount of MBTH trapped in the hybrid nafion/sol-gel silicate network since a decrease in porosity of the hybrid material results in a low amount of MBTH trapped in the hybrid film [46]. Fabrication of a test strip for detection of benzoic acid was implemented by immobilizing tyrosinase, phenol and MBTH onto a filter paper, using polystyrene as a polymeric support. In this fabrication of a test strip, an impregnating method was used. It is one of the possible enzyme immobilization procedures. It has shown a highly reproducible measurement of benzoic acid with a calculated RSD of 0.47 % (n = 10). A linear response of the biosensor was obtained in 100 to 700 ppm of benzoic acid with a detection limit (LOD) of 73.6 ppm. The activity of immobilized tyrosinase, phenol and MBTH in the test strip was fairly sustained during 20 days when stored at 3 °C. The developed test strip was used for detection of benzoic acid in food samples and was observed to have comparable results to the HPLC method [47].

Majidi and co-authors present a sol-gel based biosensor for atrazine determination that has been obtained by introducing the enzyme polyphenol oxidase


220

from apple tissue in a sol-gel matrix. The biosensor consisted of 10.3 % (mass/mass) of apple tissue. Atrazine is an inactive compound electrochemically and therefore redox coupling of dopamine was used for studying atrazine behavior [48]. Researchers have demonstrated that the silica sol-gel materials can retain the catalytic activities of enzymes to a large extent. The inorganic silica sol-gel material is biocompatible, has a high thermal stability and chemical inertness, and a negligible swelling in nonaqueous solutions. Titanium isopropoxide is much more active to water than tetraethyl orthosilicate. In case of contact with water, a precipitate of titanium dioxide is formed immediately [49]. Compared to the element of silicon, titanium has a larger covalent radius, which results in larger sol-gel pore size. This is also conducive to a fast equilibrium of the substrate between the organic solvent and titania film. Both factors are favorable to create a faster diffusion of the substrate from bulk solution to the enzyme. This results in a faster response of the sensor. Such a short response time further proves that the vapor deposition derived titania sol-gel material is a promising matrix for the construction of organic-phase biosensors. An amperometric biosensor was constructed from Yu and Ju for the determination of phenols in pure organic phase. The biosensor was fabricated by immobilizing tyrosinase in a titania sol-gel membrane which was obtained with a vapor deposition method. This method was facile and avoided the calcination step needed in conventional titania sol-gel process. The titania sol-gel membrane could effectively retain the essential water layer around the enzyme molecule, needed for maintaining its activity in an organic phase [50].

Our research group reported about the creation of new matrices for diverse applications, particularly in the field of biotechnology and food industry. We reported on the synthesis of new matrices based on a mixture of polymer cellulose acetate butyrate/copolymer polyacrylonitrile acrylamide/TiO2 [51] and silica hybrid membranes [52]. The characterization of the new matrices was performed using IR spectroscopy, QCM technology and SEM. The tests revealed that when polymers were used as carriers, there is a limit of 5 % for the concentration of titanium. Further increase of Ti concentration leads to precipitation processes. The QCM analyses show that a low concentration of Ti(OBu)4 does not influence the viscosity of the obtained sols, but elasticity of matrices changes significantly. The membranes obtained were successfully applied

for biofilm formation of yeast strain Saccharomyces cerevisiae [51] and for biosensors construction [52, 53].

The Sonogel-Carbon electrodes show the general good properties of other CCE’s (Ceramic Carbon Electrodes). Besides, in comparison with other carbon electrodes, they exhibit especially favorable electrochemical properties [54].

Kaoutit and co-authors report the development of a biosensor based on the bi-immobilization of laccase and tyrosinase phenoloxidase enzymes. This biosensor employs as the electrochemical transducer the Sonogel–Carbon, a novel type of electrode developed by this research group. The immobilization step was accomplished by doping the electrode surface with a mixture of the enzymes, glutaric dialdehyde, and Nafion-ion exchange as protective additive. The response of this biosensor, denoted the dual Trametes versicolor laccase (La) and mushroom tyrosinase based Sonogel–Carbon, was optimized directly in real beer samples and its analytical performance with respect to five individual polyphenols was evaluated. The Lac–Ty/Sonogel–Carbon electrode responds to nanomolar concentrations of flavan-3-ols, hydroxycinnamic acids, and hydroxybenzoic acids. The limit of detection, sensitivity and linear range for caffeic acid, taken as an example, were 26 nM, 167.53 nA M-1, and 0.01–2 μM, respectively [55].

Tyrosinase was covalently immobilized onto amino-functionalized carbon felt surface via glutaraldehyde-coupling under ultrasonic treatment for 10 min. The resulting TYR-immobilized carbon felt was used as a working electrode unit of a bioelectrocatalytic flow-through detector for tyrosinase substrates. Cathodic peak currents based on the electroreduction of enzymatically produced o-quinones were detected at –50 mV vs. Ag/AgCl. Compared with previous work [56], in which the enzyme was immobilized onto amino-functionalized carbon felt for 16 h without the ultrasonic treatment, the researchers in [57] succeeded in shortening the enzyme immobilization time from 16 h to 10 min. In the Table below some biosensors constructed by different kind of carriers, for phenolic compounds detection are presented (Table 1).

Synthetic and natural polymers for biomimetic biosensors construction

Synthetic polymersOne of the most important parameters to be

considered in enzyme immobilization is storage stability.


221

The stabilities of the free and the immobilized tyrosinase preparations were determined after the preparations were stored in phosphate buffer solution (50 mM, pH 6.5) at 4 ºC for a predetermined period. Under the same storage conditions, the activities of the immobilized tyrosinase preparations decreased slower than that of the free tyrosinase. The free enzyme lost all of its activity within 4 weeks. The immobilized tyrosinase preserved its initial activity during a several months storage period [58], which corresponds to the data reported by other authors [59].

Cross-linked enzyme crystals (CLECs) have several

characteristics that confer significant advantages over conventional enzyme immobilization methods, like enhanced temperature stability, absence of an inert support, catalysis under harsh conditions, such as temperature, pH and organic solvents. The insoluble nature of CLECs facilitates easy separation of the enzyme from the reaction mixture, which increases the reuse efficiency of the enzyme. Immobilization/stabilization of tyrosinase by cross-linking crystallized tyrosinase and a bovine serum albumin complex is a method employed to develop a biocatalyst, which transforms L-tyrosine in the presence of ascorbic acid

Table 1. Biosensors based on tyrosinase immobilization for phenolic compounds detection.

Method of sensing Type of membranes Phenolic compounds Limit of detection

(LOD)

Ref.

Optical detection,

surface plasmon

resonance (SPR)

Gold nanoparticles Pyrogallol

Hydroquinone

Catechol

3.2.10-7 (M)

5.3.10-7 (M)

2.5.10-6 (M)

[33]

Amperometric Aucoll-graphite-Teflon Phenol

Catechol

3-4-Dimethyl phenol

4-Chloro-3

methylphenol

4- Chlorophenol

0.020 (μM)

0.003 (μM)

0.011 (μM)

0.012 (μM)

0.019 (μM)

[34]

Amperometric Gold nanoparticles Phenol

Catechol

Caffeic acid

Callic acid

2.1.107 (M)

1.5.107 (M)

6.6.107 (M)

70.107 (M)

[36]

Amperometric Nanozeolite Phenol 0.5 (nM) [39]

Amperometric ZnO nanorods Phenol 0.623 (μM) [40]

Amperometric Mesoporous silica

derivatives

Phenol 4.1.10-9 (M)

[41]

FIA/amperometric Carbon nanotubes Phenol 0.14 (μM) [44]

Optical Chitosan/naflon/sol-gel Catechol

Phenol

m-cresol

0.18 (mg/L)

0.23 (mg/L)

0.43 (mg/L)

[46]

Amperometric Titania sol-gel Catechol

Phenol

p-cresol

6.4.10-7 (mol L-1)

8.0.10-7 (mol L-1)

1.6.10-6 (mol L-1)

[50]

Amperometric Sonogel-carbon Polyphenols

Caffeic acid

Ferulic acid

(+) catechin

(-) epicatechin

19.10-2 (μM)

2.6.10-2 (μM)

6.4.10-2 (μM)

3.4.10-2 (μM)

4.3.10-2 (μM)

[54]

Amperometric Carbon felt Catechol

Phenol

p-cresol

6.5.10-9 (μmol/L)

3.9.10-8 (μmol/L)

1.3.10-8 (μmol/L)

[57]


222

to L-DOPA (L-3,4-dihydroxyphenylalanine). Tyrosinase and bovine serum albumin were co-crystallized by saturated ammonium sulfate solution (65 %) using 20 % polyethylene glycol (PEG) 6000 and n-propanol as co-solvents. The developed biocatalyst could be recycled 6 times without further loss of tyrosinase activity. No loss of activity of cross-linked tyrosinase-bovine serum albumin crystals was observed upon storage of the developed CLECs in a refrigerator for six months [60]. A conducting, polymeric film of poly-(indole-5 carboxylic acid) has been prepared by an electrochemical polymerization for covalent immobilization of tyrosinase. As the polymer contains pendant carboxylic groups, an one-step carbodiimide method was used to immobilize tyrosinase on the polymer matrix. The linear dependence was found to be 15 µM of catechol with sensitivity of 250 mA/M cm2. Biegunski and co-authors describe novel results for immobilization of tyrosinase on the surface of electrochemically polymerized poly-(indole-5 carboxylic acid) (PIn5COOH). The immobilization of tyrosinase on the polymer film was proved by in situ surface enhanced resonance Raman spectra (SERRS) [61].

A research group constructed a novel sensitive amperometric biosensor, based on polyaniline-polyacrylonitrile composite matrix, which was applied for determination of benzoic acid. The inhibiting action of benzoic acid on the polyphenol oxidase electrode was reversible and of the typical competitive type, with an apparent inhibition constant of 38 µM. This biosensor detected levels of benzoic acid as low as 2x10-7 M in solution. Inhibition studies revealed that the proposed electrochemical biosensor was applicable for monitoring of benzoic acid in real samples such as milk, yoghurt, sprite and cola [62].

Acrylic copolymers are especially versatile as a family of carrier materials for enzyme immobilization that can be prepared with a wide variety of properties. Among these, an epoxy group carrying acrylic copolymer exhibited some significant advantages as a potential carrier matrix, i.e., easy and stable covalent linkages with different groups. The covalent bond formation via amino groups of the immobilized tyrosinase might also reduce the conformational flexibility and may result in a higher activation energy for the molecule to reorganize into the proper conformation required for binding to the substrate. One of the main reasons for enzyme immobilization is the anticipated increase in its

stability to various deactivating forces due to restricted conformational mobility of the molecules following immobilization. All immobilization studies published in the literature have been performed under different conditions. Therefore, it is almost impossible to compare immobilization results [63].

Polyamidoamine dendrimersThe synthesis of linear polymers has as a result

final products with heterogeneity that could influence the optical performance of the polymer sensor. Poly (amidoamine) dendrimers (PAMAM) are an interesting class of polymers. They have mono dispersive, well defined and developed three-dimensional structures with functional groups in high concentration. In their core they have amidic groups. Structurally modified PAMAM dendrimers with 1,8-naphthalimide derivatives are fluorescent dendrimers and they can be applied as effective and selective sensors for different metal cations and protons in organic solvents. In [64] we present our first results about the possibility for the incorporation of bio-receptors into some new PAMAM dendrimers. Dendrimers under study contain peripherally bonded 1,8-naphthalimide derivatives. The authors present the result of the preparation of fluorescence PAMAM dendrimer – acetyl cellulose membrane by the spin coating method. The fluorescent PAMAM contained a chemically bonded fluorescent dye. Lipoxygenase, peroxidase and aflatoxine antibodies were covalently immobilized and biosensors for mycotoxines detection were constructed [65].

Natural polymersRecent research performed, using a butyrylcholineste-

rase and choline oxidase enzyme electrode, suggested the validity of the biosensor approach using enzyme inhibition OPEEs (i.e. enzyme electrodes working in organic phase) in the case of organophosphorus and carbamate pesticides, which were poorly soluble in aqueous solutions. Since these pesticides were generally much more soluble in chloroform than in water, the present research aimed at analysing this class of pesticides using a tyrosinase inhibition OPEE operating in water-saturated chloroform medium. The tyrosinase biosensor was assembled using an oxygen amperometric transducer coupled to the tyrosinase enzyme, immobilized in kappa-carrageenan gel.


223

The tyrosinase biosensor, in its inhibition OPEE configuration, allows organophosphorus and carbamate pesticides, as well as the triazine or benzotriazine pesticides to be determined. It was possible to achieve a low detection limit of the order 10−8 to 10−9 mol L-1 [66].

Mushroom tyrosinase was immobilized by adsorption onto the totally cinnamoylated derivative of D-sorbitol. The polymerization and cross-linking of the derivative initially obtained was achieved by irradiation in the ultraviolet region, where this prepolymer shows maximum sensitivity. The immobilized enzyme showed an optimum measuring pH of 3.5 and greater activity at acid and neutral pH values than the soluble enzyme. The stability of immobilized tyrosinase enzyme was evaluated by storing it in distilled water at −18 ºC for 7 days. The optimal immobilization reaction time was 3 h and, at longer times, the amount of enzymatic activity retained on the support remained constant [67]. In spite of the fact that p-nitrocatechol is a very bad substrate for tyrosinase (kcat = 0.0241 ± 0.0020 s-1), it has a great affinity for the same enzyme (Km= 4.27 ± 1.12 μm). Similarly, p-nitrophenol (PNP) is probably a worse substrate than p-nitrocatechol and its affinity for the enzyme is also very high (Ki = 62 ± 6 μm). Maximum enzymatic activity was obtained at 55°C. The optimal reaction temperature was higher than that of the commercial enzymes in their free form or when immobilized on other supports [68,69], which illustrates a substantial degree of enzyme stabilization [70].

The research, described in [71] focuses on the covalent and adsorptive immobilization of tyrosinase from Agaricus bisporus onto cellulose-based carriers using DEAE (diethylaminoethanol) ligands. The effect of carrier anchor groups on the activity and stability of the immobilized tyrosinase was examined by monitoring monophenolase and diphenolase activities using L-tyrosine and L-DOPA as substrates. It should be noted that the two activities can be accurately determined from the amount of dopaquinone spontaneously associating into dark brown pigments. Finally, the stability of the obtained preparation was tested at 55°C to demonstrate the advantages of immobilization. As the immobilized enzyme is about 22 times more stable at 55 ºC, this seems to be a remarkable improvement not often reported before.

The feasibility to build up layer-by-layer (L-b-L) self-assemblies of tyrosinase and quaternized chitosan

(CHI+) on a glassy carbon (GC) rotating disk electrode is reviewed in [72]. This work highlights the promising properties of this modified polysaccharide to immobilize PPO in self-assembled structures under conditions that preserve its catalytic activity. Referring to the low amount of enzyme immobilized in the multilayer structure, the amperometric response of the biosensor reached an excellent sensitivity of about 2000 Amol−1

cm, proving that an important part of the entrapped enzyme is accessible to the substrate molecule while keeping a good level of activity. The authors show that the many amino groups present in chitosan provide a biocompatible environment for enzyme immobilization, and the enzymes retain the essential feature of their native structure in the chitosan, leading to highly sensitive sensors. The adsorption spectrum of PPO exhibits a broad band centered at 278 nm that was utilized to monitor the efficiency of the L-b-L film growth onto the two faces of quartz cells. Using the absorbance values measured at λmax = 278 nm and the corresponding extinction coefficient of the protein (2.39 mg−1 mLcm−1), a PPO surface concentration (Γ E, mol cm−2) has been obtained for each layer (considering a PPO molar mass of 128,000 g mol−1). The molecule of PPO is assumed to be spheric with a volume of 219 nm3 and its estimated projection area is about 67.8 nm2. The surface concentration of the enzyme which should correspond to the saturation of a monolayer on a 3 mm-diameter disk can be estimated as 1.73×10−13 mol.

Research, predicting the 3D structure of tyrosinase from Agaricus bisporus, used a docking algorithm to simulate binding between tyrosinase and oxalic acid, and studied the reversible inhibition of tyrosinase by oxalic acid [73].

CONCLUSIONS

The newly constructed biosensors are characterized by their response time, reproducibility, linear range and operating stability. They demonstrate high stability, shorter response time, high sensitivity and wider linear range than traditional analytical methods. Using nanomaterials allows the construction of biosensors with better parameters like biocompatibility, longer application time and mobility. The results from our research for biosensors design shows good compatibility between membranes and enzymes without a change


224

of the the conformation of the enzyme molecule and binding always takes place outside the enzyme active centres.

AcknowledgementsThe present work was supported by project DUNK

01/3 FUND “Scientific Investigations”, Ministry of Education, Youth and Science, Bulgaria.

REFERENCES

1. S. Halaouli, M. Asther, J. C. Sigoillot, M. Hamdi, A. Lomascolo, Fungal tyrosinases: new prospects in molecular characteristics, bioengeering and biotech-nological applications, J. of Applied Microbiology, 100, 2006, 219-232.

2. Y. J. Kim, H. Uyama, Tyrosinase inhibitors form natural and synthetic sources: structure, inhibition mechanism and perspective, Cell. Mol. Life Sci., 62, 2005, 1707-1723.

3. N. Z. M. Muslim, M. Ahmad, L. Y. Heng, B. Saad, Optical biosensor test strip for the screening and direct determination of L-glutamate in food samples, Sen-sors and Actuators B, 161, 2012, 493-497.

4. S. R. K. R. Sambasiva, K. N. Tripathy, D. R. Srinivasa, S. R. Prakasham, Production, characterization, cata-lytic and inhibitory activities of tyrosinase, Research, Journal of Biotechnology, 8, 1, 2013, 87-99.

5. C. Spahn, S. D. Minteer, Enzyme immobilization in biotechnology, Recent Patents on Engineering, 2, 2008, 195-200.

6. T. Coradin, J. Allouche, M. Boissiere, J. Livage, Sol-gel biopolymer/silica nanocomposites in biotechnol-ogy, Current Nanoscience, 2, 2006, 219-230.

7. J. Michalowicz, W. Duda, Phenols transformations in the environment and living organisms, Current Topics in Biophysics, suppl. A, 30, 2007, 24-36.

8. H. Decker, T. Schweikardt, D. Nillius, U. Salzbrunn, E. Jaenicke, F. Tuczek, Silmilar enzyme activation and catalysis in hemocyanins and tyrosinases, Gene, 398, 2007, 183-191.

9. P. Gautam, P. S. Suniti, K. Amrita, D. Madathil, N. A.N. Brijesh A review on recent advances in biosen-sors for detection of water, Contamination, Int. J. of Env. Sci., 2, 3, 2012, 1565-1574.

10. J. H. Purohit, Biosensors as molecular tools for use in bioremediation, J. of Cleaner Production, 11,

2003, 293–301.11. Y. Fiamegos, C. Stalikas, G. Pilidis, 4-Aminoan-

tipyrine spectrophotometric method of phenol analysis study of the reaction products via liquid chromatography with diode-array and mass spec-trometric detection, Analitica Chimica Acta, 476, 2002, 105-114.

12. M. A. Hossain, S. M. Salehuddin, Quantification of phenol in surface water by gas chromatography and mass spectroscopy, As. J. Energy Env., 10, 20, 2009, 91-98.

13. C. M. Santana, Z. S. Ferrera, J. J. S. Rodriguez, Meth-odologies for extraction of phenolic compounds from environmental samples: new approaches, Molecules, 14, 2009, 298-320.

14. M. A. G. Martinez, R. Puchades, A. Maquiera, Opti-cal immunosensors for environmental monitoring: How far have we come? Anal. Bioanal. Chem., 387, 2007, 2005-2018.

15. F. S. Ligler, Perspective on optical biosensors and integrated sensor systems, Anal. Chem., 81, 2, 2009, 519-526.

16. S. Singh, A. Srivastava, M. H. Oh, Y. C. Ahn , G. G. Choi , R. K. Asthana, Recent trends in development of biosensors for detection of microcystin, Toxicon, 60, 2012, 878–894.

17. S. Otles, B. Yalcin, Review on the application of nanobiosensors in food analysis, Acta Sci. Pol., Technol. Aliment., 22, 2, 2012, 7-18.

18. S. Tembe, M. Karve, S. Inamdar, S. Haram, J. Melo, S. F. D`Souza, Development of electrochemical biosensor based on tyrosinase immobilized in com-posite biopolymeric film, Analitical Biochemistry, 349, 2006, 72-77.

19. M. C. Puig, X. M. Berbel, C. C. Blanchard, J. L. Marty, Diazonium-functionalized tyrosinase-based biosensor for detection of tea polyphenols, Micro-chmica Acta, 171, 1, 2012,187-193.

20. S. I. C. Fernandes, M. J. F. Rebelo, Polyphenolic biosensors: application in red wines, Portugaliae Electrochimica Acta, 27, 4, 2009, 457-462.

21. Y. I. Goryacheva, E. Y. Basova, C. V. Peteghem, A. S. Eremin, L. Pussemier, J. C. Motte, S. D .Saeger, Novel gel based rapid test for non-instrumental detection of ochratoxin A in beer, Anal. Bioanal. Chem., 390, 2008, 723-727.

22. L. Yotova, I. Grabchev, R. Betcheva, D. Marinkova,


225

Detection of Bacteria, Viruses, Parasites and Fungi, In: M. V. Magni, Ed., Springer, Heilderberg, 2010, 389-414.

23. B. Gaag, S. Spath, H. Dietrich, E. Stigter, G. Boon-zaaijer, T. Osenbruggen, K. Koopal, Biosensors and multiple mycotoxin analysis, Food Control, 14, 2001, 251-254.

24. R. O. Faria, V. R. Moure, M. A. L. A. Amazonas, N. Krieger, D. A. Mitchell, The biotechnological potential of mushroom tyrosinases, Food Technol. Biotechnol.,45, 3, 2007, 287-294.

25. A. V. Rao, S. S. Latthe, D. Y. Nadargi, H. Hitashima, V. Ganesan, Preparation of MTMS based transparent super hydrophobic silica films by sol-gel method, Journal of Colloid and Interface Science, 332, 2009, 484-490.

26. M. Przybit, Potentiometric tungsten electrodes with enzymes immobilized by sol-gel technique, Materi-als Science, 21, 4, 2003, 417-429.

27. S. Jus, I. Stachel, M. Fairhead, M. Meyer, L. Thonymeyer, G. M. Guebitz, Enzimatic cross-linking of gelatin with laccase and tyrosinase, Bio-catalysis and Biotransformation, 30, 1, 2012, 86-95.

28. K. Labus, I. Gancarz, J. Bryjak, Immobilization of laccase and tyrosinase on untreated and plasma-treated cellulosic and polyamide membranes, Mate-rials Science and Engineering C, 32, 2012, 228–235.

29. P. Rijiravaich, K. Aoki, J. Chen, W. Surareungchai, M. Somasundrum, Micro-cylinder biosensors for phenol and catechol based on layer-by-layer im-mobilization of tyrosinase on latex particles: Theory and experiment, Journal of Electroanalytical Chem-istry, 589, 2006, 249-258.

30. A. Sassolas, L. J. Blum, B. D. L. Bouvier, Immobi-lization strategies to develop enzymatic biosensors, Biotechnology Advances, 30, 2012, 489-511.

31. M. E. M. Zamora, F. R. Melgarejo, F. G. Canovas, P. A. G. Ruiz, Cinnamic ester of D-sorbitol for im-mobilization of mushroom tyrosinase, J. of Chem. Technol. and Biotechnology, 80, 2005, 1356-1364.

32. J. Cortez, E. Vorobieva, D. Gralheira, I. Osorio, L. Soares, N. Vale, E. Pereira, P. Gomes, R. Franco, Bionanoconjugates of tyrosinase and peptide-deri-vatised gold nanoparticles for biosensing of phenolic compounds, J. Nanopart. Res., 13, 2011, 1101-1113.

33. M. R. H. Nezhad, M. Alimohammadi, J. Tashkou-rian, S. M. Razavian, Optical detection of phenolic

compounds based on the surface Plasmon resonance band of Au nanoparticles, Spectrochimica Acta Parta A, 71, 2008, 199-203.

34. V. Carralero, L. M. Mena, A. G. Cortes, P. Y. Sedeno, J. M. Pingarron, Development of a high analytical performance-tyrosinase biosensor based on a com-posite graphite-Teflon electrode modified with gold nanoparticles, Biosensors and Bioelectronics, 22, 2006, 730-736.

35. V. Carralero, A.G. Cortes, P.Y. Sedeno, J.M. Pingar-ron, Nanostructured progesterone immunosensors using a tyrosinase-colloidal gold-graphite-teflon biosensor as amperometric transducer, Analytica Chim. Acta, 596, 2007, 86-91.

36. V. C. Sanz, M. L. Mena, A. C. Gonazalez, P. Y. Sedeno, J. M. Pingarron, Development of tyrosi-nase biosensor based on gold nanoparticles-modifed glassy carbon electrodes. Application to the meas-urement of a bioelectrochemical polyphenols index in wines, Analytica Chimica Acta, 528, 2005, 1-8.

37. N. B. Li, J. Kwak, A penicillamine biosensor based on tyrosinase immobilized on nano-Au/PAMAM dendrimer modified gold electrode, Electroanalysis, 19, 23, 2007, 2428-2436.

38. E. Casero, L. Vazquez, A. M. P. Alfamba, E. Lorenzo, AFM, SECM and QCM as useful analytical tools in the characterization of enzyme-based bioanalytical platforms, Analyst, 135, 2010, 1878-1903.

39. X. Zhou, T. Yu, Y. Zhang, J. Kong, Y. Tang, J. L. Marty, B. Liu, Nanozeolite-assembled interface towards sensitive biosensing, Electrochemistry Communications, 9, 2007, 1525-1529.

40. B. X. Gu, C. X. Hu, G. P. Zhu, S. Q. Liu, L. Y. Chen, X. S. Li, Tyrosinase immobilization on ZnO nanorods for phenol detection, J. Phys. Chem. B, 113, 2009, 377-381.

41. Z. Dai, X. Xu, L. Wu, H. Ju, Detection of trace phe-nol based on mesoporous silica derived tyrosinase-peroxidase biosensor, Electroanalysis, 17, 17, 2004, 1571-1577.

42. B. J. Melde, B. J. Johnson, P. T. Chrles, Mesoporous silicate materials in sensing, Sensors, 8, 2008, 5202-5228.

43. K. Zynek, J. Bryjak, K. Szymanska, A. B. Jarzebski, Screening of porous and cellular materials for cova-lent immobilization of Agaricus bisporus tyrosinase, Biotechnology and Bioprocess Engineering, 16,


226

2011, 180-189.44. G. Alarcon, M. Guix, A. Ambrosi, M. T. R. Silva,

M. E. P. Pardave, A. Merkoci, Stable and sensitive flow-through monitoring of phenol using a carbon nanotube based screen printed biosensor, Nanotech-nology, 21, 2010, 1-9.

45. Y. A. Shchipunov, T. Y. Karpenko, A. V. Krekoten, Hybrid organic-inorganic nanocomposites fabri-cated with a novel biocompatible precursors using sol-gel processing, Composite Interfaces, 11(8-9), 2005, 587-607.

46. J. Abdullah, M. Ahmad, Y. L. Heng, N. Karuppiah, H. Sidek, Chitosan-based tyrosinase optical phenol bio-sensor employing hybrid naflon/sol-gel silicate for MBTH immobilization, Talanta 70, 2006, 527-532.

47. H. H. Hamaz, N. A. Yousof, A. B. Salehh, F. A. Bakar, An optical test strip for detection of benzoic acid in food, Sensors, 11, 2011, 7302-7313.

48. M. R. Majidi, K. A. Zeynali, S. Gholizadeh, Sol-gel derived biosensor based on plant tissue: The inhibi-tory effect of atrazine on polyphenol oxidase activ-ity for determination of atrazine, J. of the Chinese Chemical Society, 55, 2008, 522-528.

49. H. Zhou, L. Liu, K. Yin, S. Liu, G. Li, Electrochemi-cal investigation on the catalytic ability of tyrosinase with the effect of nanotitanium dioxide, Electro-chemistry Communications, 8, 7, 2006, 1168-1172.

50. J. Yu, H. Ju, Pure organic phase phenol biosensor based on tyrosinase entrapped in a vapor deposited titania sol-gel membrane, Electroanalysis, 16, 16, 2004, 1305-1310.

51. D. Marinkova, D. Danalev, S. Serfaty, L. Yotova, E. Caplain, P. Greismar, Characterization of new tita-nium oxide polymer hybrid membranes for biofilm formation, Phosphorous, Sulfur and Silicon, 187, 2012, 926-936.

52. S.Yaneva, D. Marinkova, L. Yotova, B. Samuneva, Immobilization of biocatalysts and cells on hybrid membranes syntheses on sol-gel method, Int. J. Bioautomation, 8, 2007, 172 – 183.

53. L.Yotova, N. Medhat, Optical biosensor with multi-enzyme system immobilized onto hybrid membrane for pesticides determination, Int. J. Bioautomation, 15, 4, 2011, 267-276.

54. M.E. Kaoutit, I.N.R.K.R. Temsamani, M. Dominguez, J.L.H.H. Cisneros, Structural characteristics and cyclic voltametric behavior of sonogel-carbon ty-

rosinase biosensors. A detailed comparative study of three immobilization matrixes, J. Sol-gel Sci. Technol.,45, 2008, 157-163.

55. M. E. Kaoutit, I. N. Rodriguez, K. R. Temsamani, M. D. Vega, J. L. H. H. Cisneros, Dual laccase-tyrosinase based sonogel – carbon biosensor for monitoring polyphenols in beers, J. Agaric. Food Chem., 55, 2007, 8011-8018.

56. Y. Wang, Y. Hasebe, Carbon felt-based biocatalytic enzymatic flow-through detectors: Chemical modi-fication of tyrosinase onto amino-functionalized carbon felt using various coupling reagents, Talanta, 79, 2009, 1135-1141.

57. Y. Wang, Y. Hasebe, Tyrosinase-modified carbon felt-based flow-biosensors: The role of ultra-sonication in shortening the enzyme immobilization time and improving the sensitivity for p-chlorophenol, J. of Environmental Sciences, 23, 6, 2011, 1038-1043.

58. M. Y. Arica, G. Bayramoglu, N. Bicak, Characteri-zation of tyrosinase immobilized onto spacer-arm attached glycidyl metacrylate-based reactive micor-beads, Process Biochemistry, 39, 2004, 2007-2017.

59. E. Jahangira, R. Agharafeie, H. J. Kaiser, Y. Ta-hambasi, R. L. Legge, K. Haghbeen, Medium engi-neering to enhance mushroom tyrosinase stability, Biochemical Engeneering Journal, 60, 2012, 99-105.

60. D. Norouzian, A. Akbarzadeh, H. M. Abnosi, M. S. Atyabasi, A. Farahngi, R. M. Mehrabi, A, Eslamifar, Stabilization of tyrosinase-bovine serum albumin crystals by glutaraldehyde, Am. J. Biochem. & Biotech., 3, 3, 2007, 110-113.

61. A. T. Biegunski, A. Michota, J. Bukowska, K. Jakowska, Immobilization of tyrosinase on poly(indole-5-5carboxylic acid) evidenced by electrochemical and spectroscopic methods, Bio-electrochemistry, 69, 2006, 41-48.

62. D. Shan, Q. Shi, D. Zhu, H. Xue, Inhibitive detec-tion of benzoic acid using a novel phenols biosensor based on polyaniline-polyacrylonitrile composite matrix, Talanta, 72, 2007, 1767-1772.

63. N. Duran, M. A. Rosa, A. D. Annibale, L. Giandfred, Applications of laccases and tyrosinases (polyphe-noloxidases) immobilized on different supports, En-zyme and Microbial Technology, 31, 2002, 907-931.

64. I. Grabchev, R. Betcheva, L. Yotova, Polyamidoam-ine dendrimers as potential substances for obtaining biosensors, COST Action 868 Biotechnical fiction-


227

alization of renewable polymeric materials, Sitges, Spain, 2007, 27-32.

65. I. Grabchev, R. Betcheva, L. Yotova, Photophysical and biological properties of fluorescent PAMAM dendrimer. Biotechnical fictionalization of renewa-ble polymeric materials, Graz, Austria, 2007, 63-68.

66. L. Campanella, D. Lelo, E. Martini, M. Tomassetti, Organophosphorus and carbamate pesticide analysis using an inhibition tyrosinase organic ohase enzyme sensor: comparison by butyrylcholinesterase and choline oxidase and application to natural waters, Analytica Chimica Acta, 587, 2007, 22-32.

67. M. E. M. Zamora, F. R. Melgarejo, F. G. Canovas, P. A. G. Ruiz, Cinnamic ester of D-sorbitol for im-mobilization of mushroom tyrosinase, J. of Chem. Technol. and Biotechnology, 80, 2005, 1356-1364.

68. T. Sandu, A. Sarbu, F. Constantin, E. Ocnaru, S. Vulpe, A. Dimitru, H. Iovu, Polyphenol oxidase immobilization of functionalized polypyrrole, Rev. Roum. Chim., 56, 9, 2011, 875-882.

69. T. Zhang, B. Tian, J. Kong, P. Yang, A. B. Liu, Sen-

sitive mediator-free tyrosinase biosensor based on an inorganic-organic hybrid titania sol-gel matrix, Analitica Chimica Acta, 489, 2003, 199-206.

70. M. E. M. Zamora, F. R. Melgarejo, F. G. Canovas, A. G. Ruiz, Direct immobilization of tyrosinase enzyme from natural mushrooms (Agaricus bisporus) on D-sorbitol cinnamic ester, Journal of Biotechnology, 126, 2006, 295-303.

71. K. Labus, A. Turek, J. Liesiene, J. Bryjak, Efficient Agaricus bisporus tyrosinase immobilization on cellulose-based carriers, Biochemical Engineering Journal, 56, 2011, 232-240.

72. L. C. Guerente, J. Desbrieres, J. Fatisson, P. Labbe, M. C. Rodriguez, G. Rivas, Physicochemical char-acterization of the layer-by-layer self assembly of polyphenol oxidase and chitosan on glassy carbon electrode, Electrochimica Acta, 50, 2005, 2865-2877.

73. M. K. Sezginiturk, T. Goktug, E. Danckaya, Detec-tion of benzoic acid by an amperometric inhibitor biosensor based on mushroom tissue homogenate, Food Technol. Biotechnol., 43, 4, 2005, 329-334.


228


NEW GLUCOSE OXIDASE - PAMAM CONJUGATE AS FLUORESCENT BIOSENSOR MATRIX IN ACETYLCELLULOSE MEMBRANE

D. Staneva1, I. Grabchev2, L. Yotova1, R. Betcheva1

1 University of Chemical Technology and Metallurgy 8 Kl. Ohridski, 1756 Sofia, Bulgaria E-mail: [email protected] Sofia University ”St. Kliment Ohridski”, Faculty of medicine, 1407 Sofia, Bulgaria

ABSTRACT

For the first time we report the covalent immobilization of glucose oxidase onto a fluorescent PAMAM dendrimer. A stable solid state membrane has been prepared using acetylcellulose and fluorescent enzyme-PAMAM membrane and its functional properties have been investigated. The new hybrid material obtained shows highs enzyme activity. Thus it shows promising features as potential applications to the fabrication of high sensitive biosensors.

Keywords: dendrimer, 1,8-naphthalimide, acetylcellulose, Glucose oxidase, enzyme, biosensor, photophysis.

Received 11 January 2013Accepted 15 May 2013

INTRODUCTION

The rapid detection of biological important com-pounds is of a particular interest for detecting different biologically active substances of metabolic or industrial origin. Fluorescence as a signal for identifying the pres-ence and quality of organic compounds is widely used. For this purpose a number of appropriate sensors has been developed using fluorophores as signalling units [1-3]. Among them polymer sensors are considered as quite promising [4].

Enzymes as biological important substances have been widely used in field of medical, biochemical and food analysis. This is dictated greatest interest to enzymatic sensors in the recent years. In this case the enzyme stability is very important. Immobilization of the enzymes on biocompatible polymers is a good solution because thus they improve their stability, efficiency and sensitivity [5-7].

Dendrimers are a relatively new category of star shaped polymers [3]. A great deal attention has been paid to this class of macromolecules owing to their new form of structure organization, which combines the properties of low and high molecular weight com-

pounds. Poly(amidoamine) dendrimers (PAMAM) are mono disperse, well defined and developed three-dimensional structures comprising functional groups at high concentration in the dendrimer periphery [3]. In the core they possess amidic and tertiary amino functional groups. This makes them suitable materials with excel-lent biocompatibility and high capacity for carrying biomaterials, with potential in drug delivery and sensing applications. Our investigations on the synthesis and functional properties of some new PAMAM dendrimers decorated with 1,8-naphthalmide units in their periph-ery have been published recently [8-17]. Structurally modified PAMAM dendrimers with 1,8-naphthalimide derivatives are fluorescent dendrimers which can be applied as effective and selective sensors for different metal cations and protons in organic solvents [18]. This interesting property can be transferred to other polymer matrixes for example textiles [19].

In this study we reports our first results on the pos-sibility to incorporate of bio receptors on a fluorescent PAMAM dendrimer containing peripherally bonded 4-ethylamino-1,8-naphthalimide as signalling fragment. Glucose oxidase enzyme has been immobilised to the PAMAM dendrimer and acetyl cellulose has been used

D. Staneva, I. Grabchev, L. Yotova, R. Betcheva

229

for the preparation of solid state transparent matrix. The photophysical and biological properties have been investigated and discussed.

EXPERIMENTAL

Materials and methods4-ethylamino-1,8-naphthalimide-labelled PAMAM

dendrimer under study (Scheme 1) has been prepared recently [12]. N,N-dimethylformamide (DMF), ethanol and dichlorometane from Merck were of spectroscopic grade. Cu(NO3)2.3H2O and Co(NO3)2.6H2O, salts were the metal cation sources and used as obtained from Aldrich.

Acetyl cellulose has been used for preparation of membrane with 1% dendrimer by using a spin coating method.

Glucose oxidase [EC 1.1.3.4] from Penicillium chrysogenum was supplied by the plant of “Biovet” , Peshtera, Bulgaria.

UV/vis spectrophotometric investigations were performed using “Thermo Spectronic Unicam UV 500” spectrophotometer. Fluorescence spectra were taken on a “Cary Eclipse” spectrophotometer. All spectra were recorded using 1 cm pathlength synthetic quartz glass cells. at concentrations of 1.10–6 mol l–1. For all fluorescent measurements, samples were exited at its absorption maxima.

The quantum fluorescence yield of dendrimer in all organic solvents has been calculated on the basis of the results obtained from the absorption and fluorescence spectra using equation 1:

2

2u st Du

F stst u Dst

S A nS A n

Φ = Φ (1)

where Fst is the quantum yield of the reference, Ast and Au represent the absorbance of the reference and the sample, respectively, Sst and Su are the integrals of the emission of the reference and the sample respectively, and nDst and nDu are the refractive index of the reference and the sample, respectively. Fluoresein was used as reference (F0 = 0.85).

A solar simulator (Suntest CPS+, HERAEUS), equipped with a 1.5 kW xenon arc lamp, protected with an adequate filter to simulate the solar spectrum between 290 nm and 800 nm, was used and the experiments were carried out in ordinary atmosphere.

RESULTS AND DISCUSSION

Photophysical propertiesThe modified PAMAM dendrimer has the struc-

ture presented in Scheme 1. It is seen that the 4-eth-ylamino-1,8-naphthalimide fluorophore fragments is covalently bonded to the dendrimer molecule. Table 1 presents the spectral characteristics of the 4-ethyl-amino-1,8-naphthalimide-labelled PAMAM dendrimer in tree organic solvents with different polarity (N,N-dimethylformamide, ethanol and dichloromethane) solu-tion: the absorption (lA) and fluorescence (lF) maxima, the extinction coefficient (e), Stokes shift (nA - n F) and quantum fluorescence yield (FF).

In all organic solvents under study the dendrimer has yellow colour absorbing at lA = 434-440 nm and emitted green fluorescence whit the fluorescence maximum at lF = 512-528 nm.

The molar extinction coefficient in the long wave-length band of the absorption spectra is e =191000-198000 l mol-1 cm-1. The extinction coefficient for the monomer 1,8-naphthalimide having an ethylamino group as substituent at C-4 has been determined as 12 000-14000 l mol-1 cm-1 [20,21]. The dendrimer molar extinction coefficient is approximately 16 fold higher than that of the monomeric 1,8-naphthalimide derivative having the same substituent at C-4 position. This allows

O N

N

O N

N

ON

N

O

NN

O

N N

O N

ON

N

N

O

O A

O

O

A

O

O

AO

O

A

ON

N

ON

N

O

NN

N

O

N O

N

N

ON

N

ON

N

O

O

A

A

O

O

O

O

A

OO

A

O N

N

O N

N

O

N N

N

O

NO

N

N

O N

N

ON

N

O

O

A

O

A

O

OO

A

O

O

AON

N

ON

N

ON

N

O

NN

O

NN

ON

ON

N

N

O

O

A

O

OA

OO

A

A O

O

N N

HH

H H

H

H

H

HH

H

H

H

HH

H

H

HH

H

H

H H

H

H

HH

H

H

A = NHCH2CH3Scheme 1. Chemical structure of 4-ethylamino-1,8-naph-thalimide-labelled PAMAM.


230

the suggestion that no ground state interaction occurs between the 1,8-naphthalimide chromophoric units.

The ability of the dendrimer to emit the absorbed light energy is characterized quantitatively by the quantum fluorescent yield FF. It was determined on the basis of the absorption and fluorescence spectra of the dendrimer in organic solvents under study. From the tabulated data in Table 1, it is seen that the quantum fluorescent yields are FF = 0.26-032. They are similar as other dendrimers having alkylamino substituents at C-4 position of the 1,8-naphthalimide fluorophores [8, 9, 13, 17].

The study also covered the spectral characteristics of dendrimer in thin acetyl cellulose membrane as well. In order to evaluate this membrane as substrate for biologically measurement, it is useful to study the functional properties of the material in solid state. In solid state the membrane has yellow-green colour with green fluorescence. Excitation and fluorescence spectra of the membrane are plotted in Fig. 1. The maximum of the excitation spectra is at 442 nm and the respective fluorescence maximum is at 510 nm. The fluorescence

maxima of thin solid film differ significantly from those in polar organic solvents being hypsochromically shifted (DlF =12-18 nm) and is close in value to the non-polar solvent. From the Figure 1 it is also seen than the excita-tion spectrum is a mirror image of the fluorescence one whit small overlap. This is indicative for the preserved planarity of the 1,8-naphthalimide chromophoric struc-ture in the exited state.

Photodegradation of 4-ethylamino-1,8-naphthal-imide-labelled PAMAM dendrimer in acetyl cellulose membrane

Photodegradation of the 4-ethylamino-1,8-naphthal-imide-labelled PAMAM dendrimer has been measured comparing the fluorescence maxima of the acetyl cel-lulose membrane before and after irradiation. In Figure 2 are presented the kinetics of photodegradation of the dendrimer. It is seen then the fluorescence inten-sity decrease during the irradiation. No new emission maxima appear in the spectrum. Also there is not any change in the fluorescence maxima before and after the irradiation. This fact demonstrates that the products of photodestruction neither fluoresce in the spectral region

Solvent λA nm

ε l mol-1 cm-1

λF nm

νA - νF cm-1

ΦF

N,N-dimethylformamide 440 194600 528 3787 0.26 Ethanol 438 191000 524 3747 0.24 Dichlorometane 434 198000 512 3510 0.32

Table 1. Photophysical properties of 4-ethylamino-1,8-naphthalimide-labelled PAMAM dendrimer in organic solvents (see text).

350 400 450 500 550 600 6500,0

0,2

0,4

0,6

0,8

1,0

FE

Norm

alize

d In

tens

ity

Wavelength / nm

Fig. 1. Normalised fluorescent intensity of excitation (E) and fluorescence (F) spectra of the 4-ethylamino-1,8-naphthalimide-labelled PAMAM dendrimer in thin acetyl cellulose membrane.

0 2 4 6 80

20

40

60

80

100

Fluo

resc

ence

Int

ensit

y / A

.U.

Time / Hours

Fig. 2. Photodestruction of 4-ethylamino-1,8-naphthal-imide-labelled PAMAM dendrimer in acetyl cellulose membrane.


231

where the 4-ethylamino-1,8-naphthalimide-labelled PAMAM is photoactive, which is in agreement with the observations on the good photostability of some similar monomeric 1,8-naphthalimide derivatives and dendrimers [9, 12,19, 20].

Biological investigationsGlucose oxidase enzyme is widely applied in bio-

chemistry, analytical chemistry and clinical diagnostic as well as in the development of microreaction biosys-tems. In our study glucose oxidase was used as a model enzyme to test the feasibility of its use in the design of the membrane and construction of the biosensors respectively. This enzyme was covalently immobilized on the dendrimer membrane by preliminary oxidation of carbohydrate residues of the glucose oxidase with pe-riodic acid [22]. It was supposed that the enzyme reacts with amidic functional groups from the dendrimer core. Thus the model enzyme and dendrimer form a stable conjugate with a high activity and were used to create a stable and flexible fluorescent transparent film. In this regard, aacetyl cellulose has been used as solid polymer membrane whit thickness of 30 mm. High relative activ-ity of the immobilized enzyme (84.69 %) at pH = 6.0

has been obtained. The amount of the bonded protein to the dendrimer in the acetyl cellulose membrane was 22.48 mg/g.

In our previous study we obtained that in DMF solution Co2+ metal cations quench the fluorescence intensity of this dendrimer [12]. The quenching effect of the Co2+was 56% at 4x10-5 mol l-1 concentration of Co2+. In this case the coordination occurring in the cores of the dendrimer molecules is followed by a photoinduced electron transfer processes or energy transfer to the pe-riphery of the molecule which quenches the fluorescence without any change in the maxima of the fluorescence and absorption spectra.

The ability of the new membrane to detect biologi-cally important metal cations as Cu2+ and Co2+ has been tested in aqueous solution at pH=6 at metal cations concentration c = 10-5 mol l-1. Fig. 3 plot the change in the fluorescence intensity of the membrane in the pres-ence of Cu2+ and Co2+ metal cations (lex = 442 nm and lem = 510 nm). It is seen than the fluorescence intensity decrease in the presence of these cations. On the other hand the quenching of the florescence emission has weak dependence from the nature of the metal cations. Slightly higher quenching effect is observed in the presence of Co2+ cations. The sensing properties of this system are based on the complexing ability of the PAMAM. These results indicate that during the immobilization of the enzyme on PAMAM dendrimer not all groups in the dendrimer core are reacted with the enzyme. This means that the membrane remain some active centers in the dendrimer structure, which can detect metal ions. The dendrimer at the membrane and the metal cations form a non fluorescent complex containing more than one metal per ligand according to the following relationship:

Dendrimer + n M2+ ⇄ [Dendrimer Mn]n 2+

NON-FLUORESCENT

Evidence comparing these results with results obtained in solution, we can conclude that the dendrimer retains its sensor properties after immobilization of the enzyme.

0 100 200 3000,94

0,96

0,98

1,00

Cu2+

Co2+

Norm

alize

d In

tens

ity

Time / sec

Fig. 3. Dynamic response fl uorescence curves of the mem-ynamic response fluorescence curves of the mem- of the mem-of the mem-mem-brane recorded as a result of exposure to Cu2+ and Co2+ in aqueous solution (c = 10-5 mol l-1).

Enzyme

Amount of bound protein (mg/g carrier)

Specific activity (U/mg)

Relative activity (%)

pH opt

Free - 64.72 - 6.0 Immobilized 22.48 54.63 84.69 6.0

Table 2. Characteristics of immobilized enzyme onto dendrimer membrane.


232

CONCLUSIONS

For the first time we have developed a novel en-zyme-PAMAM fluorescent solid state composite when glucose oxidase and dendrimer form a stable conjugate with a high activity. The photophysical and biological investigations shows that the dendrimer membrane possesses promising properties for applying in biosen-sor constructions with fluorescent detection. Also the results show that the immobilised dendrimer could act as fluorescent sensor for detection of biological important metal cations as Cu2+ and Co2+.

REFERENCES

1. K. Rurack, Flipping the light switch ‘ON’ – the design of sensor molecules that show cation-induced fluores-cence enhancement with heavy and transition metal ions, Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 57, 2001, 2161-2195.

2. V. Bojinov, N. Georgiev, Molecular sensors and molecular logic gates, Journal of the University of Chemical Technology and Metallurgy, 46, 2011, 3-26.

3. G. R. Newkome, C. D. Shreiner, Poly(amidoamine), polypropylenimine, and related dendrimers and den-drons possessing different 1 → 2 branching motifs: An overview of the divergent procedures, Polymer, 49, 2008, 1-173.

4. P. Adhikari, S. Majumbar, Progress in sensor ap-plications, Progress in Polymer Sciences, 29, 2004, 699-766.

5. S.S. Mark, N. Sandhyarani, C.C. Zhu, C. Campagnolo, C.A. Batt, Dendrimer functionalized self-assembled monolayers as a surface plasmon resonance sensor surface. Langmuir 20, 2004, 6808–6817.

6. S. Pathak, A.K. Singh, J.R. McElhanon, P.M. Dentinger, Dendrimer-activated surfaces for high density and high activity protein chip applications. Langmuir 20, 2004, 6075–6079.

7. R. Benters, C.M. Niemeyer, D. Wohrle. Dendrimer-activatedsolid supports for nucleic acid and protein microarrays, Chembiochem, 2, 2001, 686–694.

8. I. Grabchev, X. Qian, V. Bojinov, Y. Xiao, W. Zhang, Synthesis and Photophysical Properties of 1,8-naph-thalimide labelled dendrimers as PET sensors of protons and transition metal ions, Polymer, 43, 2002, 5731-5736.

9. I. Grabchev, V. Bojinov, J.-M. Chovelon, Synthesis, photophysical and photochemical properties of fluorescent PAMAM dendrimers, Polymer, 44, 2003, 4421-4428.

10. I. Grabchev, S. Dumas, J-M. Chovelon, A. Nedelcheva, First generation poly(propyleneimine) dendrimers functionalised with 1,8-naphthalimide units as fluorescence sensors for metal cations and protons, Tetrahedron, 2008, 64, 2113-2119.

11. I. Grabchev, J-M. Chovelon, H. Petkov, An iron (III) selective dendrite chelator based on polyamidoamine dendrimer modified with 4-bromo-1,8-naphthal-imide, Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 69, 2008, 100-104.

12. I. Grabchev, J.-M. Chovelon, V. Bojinov, G. Ivanova, Poly(amidoamine) Dendrimers Peripherally Modified with 4-Ethylamino-1,8-Naphthalimide. Synthesis and Photophysical properties, Tetrahedron 59, 2003, 9591-9598.

13. S. Sali, I. Grabchev, J.-M. Chovelon, , G. IvanovaSelective sensors for Zn2+ cations based on new green fluorescent poly(amidoamine) dendrimers peripherally modified with 1,8-naphthalimides, Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 65, 2006, 591-597.

14. I. Grabchev, J.-P. Soumillion, B. Muls, G. Ivanova, Poly(amidoamine) dendrimer peripherally modi-fied with 4-N,N-dimethylaminoethyleneamino -1,8-naphthalimide as sensor of metal cations and protons, J. Photochem. Photobiol. Science, 3, 2004, 1032-1037.

15. I. Grabchev, J.-M. Chovelon, A. Nedelcheva, Green fluorescence poly(amidoamine) dendrimer func-tionalized with 1,8-naphthalimide units as potential sensor for metal cations, Journal of Photochemistry and Photobiology A: Chemistry, 183, 2006, 9-14.

16. I. Grabchev, D. Staneva, V. Bojinov, R. Betcheva, V. Gregoriou, Spectral investigation of coordina-tion of cuprum cations and protons at PAMAM dendrimer peripherally modified with 1,8-naphthal-imide units, Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 70, 2008, 532-536.

17. I. Grabchev, S. Guittonneau, Sensors for detecting metal ions and protons based on new green fluorescent poly(amidoamine) dendrimers peripherally modified with 1,8 naphthalimides, Journal of Photochemistry and Photobiology A: Chemistry, 179, 2006, 28-34.


233

18. I. Grabchev, D. Staneva, R. Betcheva, Fluorescent dendrimers as sensors for biologically important metal cations, Current Medical Chemistry, 29, 2012, 4976-4983.

19. I. Grabchev, D. Staneva, R. Betcheva, Sensor activity, photodegradation and photostability, of a PAMAM dendrimer comprising 1,8-naphthalimide functional groups in its periphery, Polymer Degradation and Stability, 91, 2006, 2257-2264.

20. I. Grabchev, P. Meallier, T. Konstantinova, M. Popova,

Synthesis of Some Unsaturated 1,8-Naphthalimide Dyes, Dyes and Pigments, 28,1995, 41-46.

21. I. Grabchev, Photophysical characteristics of polym-erizable 1,8-naphthalimide dyes and their copoly-mers with styrene or methylmethacrylate, Dyes and Pigments, 38, 1998, 219-226.

22. L.K. Yotova, I.P. Ivanov, Simultaneous immobiliza-tion of glucose oxidase and peroxidase to urea deriva-tive of regenerated acetylcellulose granules, Applied Biochemistry and Biotechnology 61, 1997, 277-287.


234


IMPACT OF POLYSACCHARIDES OF MALT ON FILTERABILITY OF BEERAND POSSIBILITIES FOR THEIR REDUCTION BY ENZYMATIC ADDITIVES

G. Jonkova, A. Surleva

University of Chemical Technology and Metallurgy8 Kl. Ohridski, 1756 Sofia, BulgariaE-mail: [email protected]

ABSTRACT

Beta-glucans are principal constituents of barley endosperm cell walls. They are linear high-molecular poly-saccharides, which tend to increase the solutions viscosity by forming gels. As β-glucans influence wort and beer viscosities, as well as beer quality, they are ones of the most studied wort and beer components.

This study is aimed at production of wort from malt at different degree of cytolytic modification and estimation of the effect of exogenous β-glucanases addition during wort fermentation. β-Glucans content in poorly- and well-modified malt was 1000 mg L-1 and 384 mg L-1, respectively. The addition of enzymatic preparation with β-glucanases activity during the fermentation of wort, obtained from poorly modified malt, resulted in decreasing of beer viscosity by 5 – 40 % (depending on exogenous enzyme concentration: 1 and 2 BGU/L Finizym® 200 L) and decreasing of β-glucan content by more than 90 %.

Keywords: malt, wort, polysaccharides, β-glucans, exogenous β-glucanases.

Received 20 December 2012Accepted 20 May 2013

INTRODUCTION

Polysaccharides are high-molecular carbohydrates consisted of monosaccharide residues linked by gluco-sidic bounds. According to their chemical composition and structure they are referred as: homopolysaccharides and heteropolysaccharaides [1, 2].

β-Glucans, being the main part of endosperm cell walls of barley grain - about 75 % of all carbohydrates, are ones of the most studied homopolysaccharides in brewing [3]. The inherent tendency of β-glucans to form gels cause increased wort and beer viscosity. At the same time the viscosities of wort and beer influence the brewing process and beer quality in opposite aspects [4]. At one side, beer viscosity positively contributes to its body and head retention of beer foams [5]. At the other side, high viscosities of wort and beer can lower the ef-ficiency of many brewing operations and in large extent the process of beer filtration [6]. β-Glucans content in barley is reported to be between 0.4 and 8.6 % [6]. It

depends on genotype of barley varieties, year, growing sites and environmental conditions [6-10].

Mixed linkage (1-3,1-4)-β-D-glucans, commonly known as β-glucans, are linear homopolymers of D-glucopyranosyl residues linked mostly via two or three consecutive β-(1-4) linkages that are separated by a single β-(1-3) linkage. Less frequent are longer segments of consecutively (1-4)-β-linked glucosil residues with degree of polymerization 5–28. There is no current evi-dence that two or more adjacent (1-3)-β-linkages occur in the β-glucan chains [11, 12]. 70 % of β-(1-4)- and 30 % of β-(1-3)- bounds are estimated in barley β-glucans (Fig. 1) [3,13-15]. Despite the non-random arrangement of individual (1-3) and (1-4)-β-linkages, the glucosil residues are arranged in an essentially independent and random fashion in the β-glucan chain [16,17].

Mixed-linkage β-glucans play mainly structural role in the barley grain. Cell walls of the starchy endosperm of barley consist of approximately 17 % of (1-3, 1-4) -β-D-glucans and 20 % of arabinoxylan, together with


235

smaller amounts of mannose-containing polysaccharides and cellulose [18]. Enzymatic hydrolysis (disintegration) of β-glucans is the most crucial brewing process. These polymers are broken down to various degrees mainly during malting [19]. The process is also known as malt modification. The cell walls are degraded by enzymatic complex referred as “citase”. The complex comprises the enzymes: (1-3)-endo-β-glucanases; (1-4)-endo-β-gluconases; β-glucan-solubilases, and non-specific enzyme (1-3,1-4)-β-glucan endohydrolases also named (1-3,1-4)-β-glucanases; exo-β-glucanases and lamina-rinase [6,20]. As a result of cytolytic break down of cell walls during barley germination, other enzymes efficiently depolymerize the starch and other reserve proteins of endosperm. β-Gluconase is reported to be in very low quantity in barley grains, and their content dramatically increased after malting [10]. However, the endogenous barley β-glucanases synthesized during germination are damaged during mashing where tem-peratures around 50 oC are used. As a result β-glucans content in wort increased [5,21,22]. Thus, both the level of glucan-hydrolysing activities achieved during germination and the amount of β-glucans are important factors in the production of high quality malt [23]. The malt β-glucan content was reported to be more dependent on malt β-glucanase activity than on the original level of β-glucan in grains [10].

Incomplete disintegration of β-glucans has a negative impact on brewing. The presence of non-disintegrated high molecular β-glucans in wort increase viscosity of the wort and the resulting fermented beer, causing difficulties in filtration in the brewery. Reduced filterability of mash and beer has often been attributed to large β-glucans (molecular weight of 31 - 443 KDa) which tend to increase the viscosity of beer by forming inter-molecular hydrogen bonds between sequences of (1-4)-β-linkages [21,24-26]. Residual β-glucans may also play a role in beer maturation, promoting the for-

mation of undesirable precipitate and hazes. β-Glucans content in barley is an indicator of malt modification and quality and illustrates effects of malting technology on quality of malt and wort [7,27]. As it was reported, wort viscosity varies from 1.59 to 5.16 mPa s and beer viscosity – between 1.45 and 1.96 mPa s [27].

Exogenous enzymes are used to supplement the malt’s own enzymes in order to prevent filtration prob-lems and β-glucan hazes [25, 28]. Addition of exogenous β-glucanases could be made during mashing-in or during wort fermentation [5, 23]. The commercial enzymatic preparations with β-glucanase activity are obtained by deep cultivation of special strains. Usually bacte-rial and fungal β-glucanases are used. The research on the application of commercial enzymatic preparations (exogenous enzymes) in brewing is focused mainly on the mashing and very scarcely on beer fermentation [5, 29-32].The reported technological effects of using commercial enzymatic preparations can be summarised as: reduced β-glucans content, better wort lauterring, increased extract yield and lowered beer haze.

This report presents the results from studying the effect of the addition of a commercial enzymatic prepara-tion with β-glucanase activity during the fermentation of the wort, obtained from poorly modified malt.

EXPERIMENTAL

MaterialsThe malt used in this study is produced by two

breweries A and B. The malts differed from each other in their degree of modification.

The commercial enzymatic preparation used was Fi-nizym® 200L (Novo Nordisk, Denmark) – a fungal beta glucanase produced by a selected strain of Aspergillus niger. The enzymatic complex hydrolysis (disintegrated) barley β-glucans to oligosaccharides. The preparation is recommended for use during beer fermentation and

Fig. 1. β-Glucans structure.


236

maturing [33]. Tap water with residual alkalinity of 0.75 pH was used in brewing.

MethodsTen wort samples were produced from different

Bulgarian brewers. The worts were analysed and the viscosity and β-glucans content were determined.

Two malts obtained from the brewers A and B were used to obtain worts in lab scale experiments. The worts were fermented to obtain beer. The enzymatic prepara-tion Finizym® 200 L was added only to worts, obtained from poorly modified malt.

Two lab-scale experiments were performed:1) characterisation of beer obtained from malt at

different extent of modificationThe well modified malt supplied from brewery A

was used as a control sample. The poorly modified malt supplied from brewery B was denoted as an experimental sample. 500 g of preliminary milled malt was mixed with water at 45oC and mashed following infusion method (Fig. 2).

At the end of the mashing, the weight of the mash was adjusted to 4500 g with distilled water and well homogenized. The obtained wort was separated by lautering. The wort was hopped with 80 mg L-1 α-bitter acids by boiling for 90 min. Yeast strain Saccharomyces carlsbergensis (4 g L-1) was used in the fermentation at temperature 9-10 oC. At 10 % difference between “final apparent” and “apparent” degrees of fermentation, the obtained young beer was set to maturation at 4 oC for 20 days.

2) Study of the effect of enzymatic preparation Finizym® 200 L on brewing

Wort without enzymatic preparation was used as a control. The brewing process follows the same schema

described above (section 1). The influence of quantity of added enzymatic preparation was studied at 1 and 2 BGU L-1 of enzyme. The Finizym® 200L was added to the wort in the start of fermentation.

Analysis Standard analytical methods for beer quality as-

sessment were used according to the European Brewing Convention [34]. β-Glucans were determined by spec-trometric anthrone method for quantification of total carbohydrates [35].


Estimation of the wort samples from different breweries in Bulgaria

Ten randomly chosen worts produced in Bulgaria were taken for the study of the range of wort β-glucan content and viscosity. The extract content in all studied Fig. 2. Temperature program of infusion mashing.

Fig. 3. Minimal, mean and maximal values of wort viscosity and wort β-glucans content of 10 wort samples, obtained from different breweries.


237

worts was 10.5 %. The results are presented in Fig. 3. The wort viscosity ranged between 1.67 mPa s and

2.05 mPa s and the mean was 1.82 mPa s. The β-glucan content in wort was from 115 mg L-1 to 998 mg L-1 and mean was 394 mg L-1. The β-glucans ratio calculated as a ratio of β-glucan content in well-modified malt to β-glucan content in poorly-modified malt was 1 to 2.7. Our results were lower from the reported in the litera-ture ratio from 1 to 6 [36]. As it was mentioned in the introduction, quantity of released (solubilised) β-glucans during mashing depends on barley variety and malt modification during germination (β-glucan content and

β-glucanases activity). As the enzymes responsible for β-glucan hydroly-

sis were temperature depended [22], modifications of known mashing methods were not very efficient for lowering β-glucans content in wort. Malt quality, rather than the mashing method, was dominating factor deter-mining the wort β-glucan level [5].

Evaluation of the malts samplesQuality of malts, produced by two breweries A and

B, used for mashing in this study is presented in Table 1. As can be seen from the results, the malts differed from each other in their degree of modification. The viscosity, extract difference and β-glucan content of malt B were higher compared to the malt A. The value of viscosity of 1.63 mPa s (8.6 % Congress wort) was widely accepted as a reference for well-modified malt. Hence, the malt from brewery B was classified as “poorly modified” and the malts from brewery A - as “well modified”.

Quality of beer obtained from poorly and well modified malt

The β-glucan content and viscosity of wort and beer, obtained from the above presented malts (A and B) are illustrated in Figure 4. As it was reported in the literature, an intensive cytolytic break down of β-glucans by the endogenous enzymes was observed in the temperature interval 35-51oC [5]. As can be seen form the results, β-glucans content in wort produced from poorly modi-fied malt (B) was 1050 mg L-1 and decreased in beer down to 540 mg L-1, whereas the β-glucans content in wort obtained from well-modified malt (A) and in produced beer was 116 mg L-1 and 25 mg L-1, re-spectively. The β-glucans ratio in wort obtained from well- and poorly-modified malt was 1 to 9, higher than the β-glucan ratio in malts. The difference in β-glucan content in well- and poorly-modified malt couldn’t be changed during mashing and the trend was the same. The viscosity of wort obtained from malt A and B was 1.71 and 1.85 mPa s, respectively, following the same trend as the β-glucans content: the higher viscosity was observed

Fig. 4. Viscosity and β-glucan content in wort and beer obtained from: sample A – well modified malt; and sample B – poorly modified malt.

malt A malt B extract difference, % 1.0 2.9 viscosity, mPa s 1.58 1.70 β-glucan, mg L-1 384 1040

Table 1. Quality of malts used for mashing.


238

in beer obtained from poorly modified malt (Fig. 4). The results coincided with the observation already made by other authors that the β-glucan content in wort depended mainly on the quality of malt [5]. For comparison we present practical criteria for good filterability of wort reported in the literature: viscosity below 1.65 mPa s and β-glucan content below 200 mg L-1 [7].

Quality of beer obtained with addition of exoge-nous enzymes

The addition of enzyme preparation Finizym® 200L at the beginning of the fermentation process was studied. The effect was estimated at two concentration levels of exogenous enzymes by comparing viscosity, β-glucan content and apparent degree of fermentation with control beer, produced without addition of enzymes. The quality of beer obtained with and without addition of enzymatic preparation Finizym® 200L is compared in Fig. 5. The obtained results showed that the addition of exogenous β-glucanases affected positively beer vis-cosity. The effect was more pronounced at 2 BGU L-1 of

enzymatic preparation – around 40 %, compared to 5 % decreasing at 1 BGU L-1 level. Moreover, the addition of Finizym® 200L at concentrations 1 and 2 BGU L-1 increased apparent degree of fermentation up to 4.4 % and 5.2 %, respectively, compared to the control sample. The results indicate an improved fermentation of wort.

The most pronouncing effect of the addition of ex-ogenous enzyme was observed on β-glucans content. It decreased by more than 90 % compared to the control beer. The results indicated a high degree of β-glucans break down. The highest activity of the exogenous enzyme Finizym® 200L was at the conditions of low temperatures and pH created during wort fermentation.

The obtained beer samples were also analysed to estimate the haze. Both samples B1 and B2 showed almost equals haze values, but 2 times lower than the control beer.

As the results showed, the beer filterability could be improved by addition of enzymatic preparation Finizym® 200L during fermentation of wort. Addition-

Fig. 5. Quality of beer obtained by addition of exogenous enzymes: K – control sample without exogenous enzimes; sample B1- with 1 BGU L-1 Finizym® 200 L; sample B2 - with 2 BGU L-1 Finizym® 200 L.


239

ally, the volume of the beer filtered by one pre-coat of kieselguhr filter can be increased.

CONCLUSIONS

The quality of ten batches of 10.5 %-wort produced by Bulgarian breweries was studied and the results showed that the wort viscosity and β-glucans content varied from 1.67 mPa s to 2.05 mPa s and from 115 to 998 mg L-1, respectively. Two malts were produced in an industrial scale by two breweries. The cytolytic modification of the malt was characterized by viscosity, extract difference and β-glucan content. Based on their values the studied malts were classified as well- and poorly-modified. Both malts were used to obtain wort. A difference in the extent of β-glucans break down was observed. The β-glucan content in wort obtained from well- and poorly-modified malts was 116 and 1050 mg L-1, respectively. The same trend was observed in the final beer. After addition of 1 or 2 BGU L-1 of enzymatic preparation Finizym® 200 L during wort fermentation, β-glucan content decreased by 90 %. The results showed that the addition of enzymatic preparation leads to high degree of enzymatic hydrolysis of biopolymers at the conditions of wort fermentation: low temperature and slightly acidic medium (pH around 5.0). Moreover, the addition of enzymatic preparation had a positive effect on brewing process lowering beer viscosity, improving beer fermentation and its colloidal stability.

AcknowledgementsThe authors wish to gratefully acknowledge the

University of Chemical Technology and Metallurgy, Sofia, Bulgaria, for the financial support of this research by Grant 10867/2011 from the Science and Research Program.

REFERENCES

1. H.D. Belitz, W. Grosch, Lehrbuch der Lebensmittelchemie, Springer-Lehrbuch, Berlin, 1992, p. 226.

2. H. Scherz, G. Bomm, Analytical chemistry of car-bohydrates, Thieme organic chemistry monograph series, Stuttgart, 1998, p.17.

3. M.S. Izydorczyk, J.E. Dexter, Barley β-glucans and arabinoxylans: Molecular structure, physicochemical properties, and uses in food products–a review, Food

Research Intern., 41, 2008, 850–868. 4. J.S. Hough, D.E. Briggs, R. Stevens, T.W. Young,

Beer flavour and quality, in: Malting and brewing science, Vol. II. Hopped wort and beer, Chapman and Hall Ltd., New York, 1982, p. 839-876.

5. Y.L. Jin, R. Speers, A. Paulson, R. Stewart, Barley β-Glucans and their degradation dyring malting and brewing, TQ MBAA, 41, 2004, 231-240.

6. L. Narziss, β-Glucan und fitrierbarkeit, Brauwelt, 37, 1996, 1696-1706.

7. V. Psota , J. Ehrenbergerovа, P. Havlovа, J. Hartmann, Beta-glucan content in caryopses, malt and wort of the selected spring barley varieties, Monatsschrift fur Brauwissenschaft, 55, 2002, 10-14.

8. A.M. Pérez-Vendrell, J. Brufau, J.L. Molina-Cano, M. Francesch, J. Guasch, Effects of cultivar and environment on β-(1,3)-(1,4)-D-glucan content and acid extract viscosity of spanish barleys, J. Cereal Science, 23, 1996, 285-292.

9. R. M. de Sa, G.H. Palmer, Assessment of Enzymatic Endosperm Modification of Malting Barley Using Individual Grain Analyses, J. Inst. Brew., 110, 1, 2004, 43–50.

10. J. Wang, G. Zhang, J. Chen, F. Wu, The changes of β-glucan content and β-glucanase activity in barley before and after malting and their relationships to malt qualities, Food Chem., 86, 2004, 223-228.

11. W. Cui, P.J. Wood, B. Blackwell, J. Nikiforuk, Physico-chemical properties and structural characterization by two-dimensional NMR spectroscopy of wheat β-D-glucans – Comparison with other cereal β-D-glucans, Carbohydrate Polymers, 41, 3, 2000, 249–258.

12. P. Dais, A.S. Perlin, High-field, C-N.M.R. spectros-copy of β-D-glucans, amylopectin, and glycogen, Carbohydrate Research, 100, 1, 1982, 103–116.

13. J.R. Woodward, G.B. Fincher, B.A. Stone, Water-soluble (1→3), (1→4)-β-D-glucans from barley (Hordeum vulgare) endosperm. II. Fine structure, carbohydrate polymers, 3, 1983, 207-225.

14. B.V. McCleary, М. Glonnie-Holmcs, Enzymic quantification of (1-3)(1-4)-β-D-glucan in barley and malt, J. Inst. Brew., 91,1985, 285-295.

15. P. Aman, K. Hesselman, An enzymic method for analysis of total mixed-linkage β-glucans in cereal grains, J. Cereal Science, 3, 1985, 231-237.

16. G.S. Buliga, D.A. Brandt, G.B. Fincher, The se-


240

quence statistics and solution conformation of a barley (1-3, 1-4)-D-glucan. Carbohydrate Research, 157, 1, 1986, 139–156.

17. R.G. Staudte, J.R. Woodward, G.B. Fincher, B.A. Stone, Water-soluble (1→3), (1→4)-β-d-glucans from barley (Hordeum vulgare) endosperm. III. Distribution of cellotriosyl and cellotetraosyl resi- of cellotriosyl and cellotetraosyl resi-of cellotriosyl and cellotetraosyl resi- cellotriosyl and cellotetraosyl resi-cellotriosyl and cellotetraosyl resi- and cellotetraosyl resi-and cellotetraosyl resi- cellotetraosyl resi-cellotetraosyl resi- resi-resi-dues, Carbohydrate Polymers, 3, 1983, 299-312.

18. G.B. Fincher, Morphology and chemical composi-tion of barley endosperm cell-walls, J. Inst. Brew., 81, 1975, 116-122.

19. S. Aastrup, K. Erdal, Quantitative determination of endosperm modification and its relationship to the content of 1,3:1,4-glucans during malting of barley, Carlsberg Res. Commun., 45, 1980, 369–379.

20. G. Kabaktschieva, W. Atanasov, T. Dimitrova, J. Platikanov, A. Atev, P. Peev, Production of beer using cytolitic enzymatic compelx, Food Industry, 41, 1992, 31-38 (in Bulgarian).

21. R. Buchow, V. Heinz, D. Knorr, Effect of high hydrostatic pressure-temperature combinations on the activity of from barley malt, J. Inst. Brew., 11, 3, 2005, 282-289.

22. R. Muller, Factors influencing the stability of barley malt β-gluconase during malting, J. Am. Soc. Brew. Chem., 53, 3, 1995, 136-140.

23. K.R.S. Celesatino, R.B. Cunha, C.R. Felix, Characterization of a β-glucanase produced by Rhizopus microspores var. microspores, and its potential for application in the brewing industry, BMC Biochemistry, 7, 23, 2006, 1-9.

24. L. Narzis, Die technologie der Malzbereitung, Enke Ferlag Stuttgart, 1999, 157-163.

25. Y.L. Jin, A. Speers, A. Paulson, R. J. Stewart, Effects of β-glucans and environmental factors on the viscosities of wort and beer, J. Inst. Brew., 110,

2, 2004, 104–116. 26. C. Bamforth, Beta glucan and beta glucanases in

malting and brewing: Practical aspects, Brew. Dig., 69, 5, 1994, 12-16.

27. R. Oonsivilai, The effect of β-glucan polymers on the rheological and filtration properties of wort, M.Sc. Thesis, Dalhousie University, Halifax, NS, 2000.

28. S. Aastrup, H.S. Olsen, Enzymes in brewing, 2, 2008, www.biokkemi.org/biozoom/issues/522/article2368

29. G. Kabaktschieva, W. Atanasov, T. Dimitrova, A. Atev, P. Peev, Biotechnologische und technologische untersuchungen mit dem bulgarischen Hydrolase- Enzympraparat “Trichozease SU-2”, Die Nahrung, 54, 1990, 407-413.

30. G. Kabaktschieva, W. Atanasov, J. Platikanov, Verbesserung der ablauterung und der bierfiltration durch enzympraparate, Brauindustrie, 78, 1993, 1159-1160.

31. S. Maltfliet, L. De Cooman, G.Aerts, Application of thermostable xylanases at mashing and during germination, Brewing Science, 63, 2010, 133-141.

32. A. Scheffler, C. Bamfoth, Exogenous β-Glucanases and pentosanases and their impact on mashing, Enzyme Microbial Technol., 36, 2005, 813-817.

33. Finizim® 200 L, Novo Nordisk, Denmark, 2005, data booklet.

34. European Brewery Convention, Analytica EBC, Brauerei und Getränke Rundschau, 5th ed., CH-8047, Zürich, Switzerland, 1997.

35. K. Erdal, P. Gyerksen, β-Glucans in malting and brewing, Proc. Eur. Brew. Conv. Cong., Elsevier, Madrid, 1967, 295-302.

36. L. Narziss, β-Glucan and filterability, Brauwelt Int., 5, 1993, 435–442.

D. Zheleva

241


AN ATTEMPT FOR CORRELATION BETWEEN MOONEY VISCOSITY AND

RHEOLOGICAL PROPERTIES OF FILLED RUBBER COMPOUNDS

D. Zheleva


ABSTRACT

For the characterization of rubber compounds from the point of view of their processing capability in the rub-ber industry is widely used a technological indicator – the Mooney viscosity. The Mooney viscosity obtained gives information just for one point of the flow curve of the respective rubber or rubber compound. This is the reason that no full rheological characteristics of the elastomeric material are obtained.

The aim of this study is to find a correlation between the Mooney viscosity and the torque on a Plasticorder for rubber compounds.

A correlation was settled between the Mooney viscosity ML and the torque MB and the corresponding equations for this dependency were derived for rubber compounds based on SBR (styrene-butadiene rubber) containing different types of carbon black and for compounds with one and the same type of carbon black, but of different level of filling.

Keywords: rheology, rubber compounds, Mooney viscosity, torque, correlation.

Received 11 November 2012Accepted 10 May 2013

INTRODUCTION

In order to describe rubber compounds from the point of view of their processing capability different methods are used [1]. In this respect full characterization is achieved by analyzing the flow curves of rubbers and rubber compounds [2]. The flow curves can be obtained by different viscometers. In the rubber industry is widely used (over 100 years) a technological indicator – the Mooney viscosity. It serves for a comparison and con-trol. It is measured relatively quickly with the respective viscometer according to the International standard ISO 289-1:2002 [3]. The Mooney viscosity obtained gives information just for one point of the flow curve of the rubber or rubber compound and this is the reason that no full rheological characteristics of the elastomeric mate-rial are obtained [4]. In order to describe the processing capability of rubber compounds by means of Brabender Plasticorder, the measurements are usually performed at relatively high rotation speed – at about 20 min-1. When these revolutions are retained the shear conditions during

the production process are mostly similar to the processing conditions, but are very different from the shear condi-tions during the measurement of the Mooney viscosity.

Our previous investigations were carried out on a Plasticorder [5] at low rotor revolutions of 1, 2, 5 and 10 min-1. This investigation includes an attempt to calculate the average shear rate

•

γ based on the experimentally obtained data at different rotor revolutions. It was found through extrapolation of the linear relationship shear rate

•

γ v/s rotor speed, that •

γ values in the range 1-2 s-1 correspond to the rotor revolutions 1 to 5 min-1. Taking into account that

•

γ in the Mooney viscometer is at about 1-1,5 s-1, we tried to find out a correlation between the values of the Mooney viscosity and the results from the measurements carried out on the Plasticorder.

In [6] a correlation between the Mooney viscos-ity and the effective viscosity, obtained by means of a capillary viscometer, is found. The investigations were carried out on the both rheometers with 6 types of rub-bers, namely NR crepe sheets (natural rubber), SMR-20 (Standard Malaysian Rubber, i.e. type natural rubber),


242

SBR (styrene-butadiene rubber), NBR (acrylonitrile-butadiene rubber), MQ (silicone rubber), IIR 268 R (butyl rubber) and their compounds. Following equation is obtained:

10 (lg )PL tη= +where is: PL – Mooney viscosity, Mooney units;η – effective viscosity, Pa.s;t – duration for flow out of material in capillary viscom-eter, s.

According to the authors [6], the error does not exceed 2 %.

Other researchers [7] found a correlation between rheological behavior of filled rubber blends and their properties/structure. They suggested that a thorough understanding of the characteristic rheological response to the morphology/structure evolution of multiphase/multi-component polymers facilitates researchers’ opti-mizing the morphology/structure and ultimate mechani-cal properties of polymer materials.

Similar investigations aimed to find out a correla-tion between the Mooney viscosity and the torque were carried out for NR (natural rubber) and BR (butadiene rubber) compounds on a Plasticorder in the monography [8]. According to the author, the deformation conditions strongly influence the compounds structure, respectively their effective viscosity. This is the reason to find just qualitative dependences property/composition.

The aim of the present study is to find a correla-tion between the Mooney viscosity and the torque on

a Plasticorder Brabender for rubber compounds based on SBR (styrene-butadiene rubber), filled with different types of carbon black or with one type of carbon black with different filling degree.

EXPERIMENTAL

Objects of investigation Styrene-butadiene rubber SBR (Bulex 1500) with

molecular weight ~ 200000 and density ρ=0,93 g/cm3.Carbon black

The properties of carbon black used are listed in Table 1 [9].Composition of rubber compounds

Investigated rubber compounds have following composition:SBR /Bulex 1500/ 100 phr, Carbon black 50 phr, ZnO 5 phr, Stearic acid 1,5 phr.

Methods for investigation- Mooney viscosity – ISO 289-1:2002 [3].- Determination of the torque by Brabender Plas-ticorder

The investigated rubber sample is loaded in the Plas-ticorder camera at low rotors revolutions (i.e. 1-2 min-1). After filling the camera revolutions increase to 30 min-1. The temperature of the compound is followed until it reaches 100°C. Then the rotors revolutions are reduced

Types of carbon black

Diameter of the carbon black particles [nm]

Specific geometric surface [m2/g]

Specific adsorption surface [m2/g]

DBP adsorption [cm3/100g]

Iodine index [mg/g]

Bulk mass [g/1000cm3]

Density [g/ cm3]

PM-15

155-210 12-18 10-20 85-105 20-30 250 1,82

PGM-33

85-100 32-38 32-38 85-105 20-40 400 1,82

N-550 50-65

45-60 40-55

96-100 40-50 330 1,84

N-330 38-42

75-82

75-85 95-105 80-100 330 1,86

N-220 29-32 96-105 90-100 97-105 100-130 330 1,86

Table 1. Characteristics of investigated types of carbon black.

D. Zheleva

243

to 2, 3 or 4 min-1 and the torque values are read at the respective temperature. A diagram “torque vs. tempera-ture” is plotted and the experimental points define а band of this dependence. Torque at 100°C is determined by interpolation in the middle of the band obtained for this temperature.


Influence of the type of carbon black on the correla-tion between the Mooney viscosity and the torque on a Plasticorder Brabender of a rubber compound

Experimental results of the dependence “torque MB vs. temperature” of the investigated compounds at given revolutions of the rotors are shown in figures 1÷3. In these figures are presented the bands lining between the curves of highest and lowest torque values resulting from multiple measurements for each of the compounds.

For each of the revolutions defined at 100оС the average torque value of the corresponding compound is defined as “х”. In some cases this band is quite wide (Fig. 1a), in others it is reduced almost to one curve (Fig. 2a). However it is impossible to establish any conformity neither with the dependence from the type of the carbon black, nor from the number of the rotor revolutions.

Fig. 4 presents the diagrams of Mooney viscosity val-ues and of torque values, for compounds containing differ-ent types of carbon black at different rotors revolutions.

As it can be seen, the Mooney viscosity increases with increasing the activity of the investigated carbon black. The tendency of changing of the torque MB is different in dependence on the activity of carbon black. The better coincidence between the torque and the Mooney viscosity, by their absolute values, is observed at rotor revolutions of 3 min-1 for carbon black PM-15, PGM-33 and N-550, the coincidence for type carbon black N-330 occurs at 2 min-1, and the dispersing is the highest for carbon black N-220.

80 85 90 95 100 105 110 115 120 125 13020

25

30

35

40

45

50

55

60

x

2 min-1

MB, [N.m]

t, 0C

85 90 95 100 105 110 115 120 125 130

25

30

35

40

45

x

3 min-1

MB, [N.m]

t, 0C

Fig. 1. Dependence of the torque MB on the temperature for the SBR compound filled with 50 phr carbon black PM-15 at different rotors revolutions on Plasticorder.

90 95 100 105 110 115 120 125 13020

25

30

35

40

x

2 min-1

MB, [N.m]

t, 0C

90 95 100 105 110 115 120 125 13030

35

40

45

50

x

3 min-1

MB, [N.m]

t, 0C

Fig. 2. Dependence of the torque MB on the temperature for the SBR compound filled with 50 phr carbon black PGM-33 at 2 min-1 (a) and carbon black N-550 at 3 min-1 (b).

a) a)

b) b)


244

Influence of the filling level of rubber compounds containing the same type of carbon black on the cor-relation between Mooney viscosity and the torque

Fig. 5 shows the dependence of the Mooney vis-cosity on the carbon black N-550 content in rubber

compounds, respectively 30, 40, 50, 60 and 70 phr carbon black.

Fig. 6 and Fig.7 demonstrate the experimental data for the dependence of the torque vs. temperature of the investigated compounds at different rotors revolutions of the plasticorder.

90 95 100 105 110 115 120 125 13040

45

50

55

60

x

3 min-1

MB, [N.m]

t, 0C

95 100 105 110 115 120 125 13035

40

45

50

55

60

4 min-1

MB, [N.m]

t, 0C

Fig. 3. Dependence of the torque MB on the temperature for the SBR compound filled with 50 phr carbon black N-330 at 3 min-1 (a) and carbon black N-220 at 4 min-1 (b).

0

10

20

30

40

50

60

PM-15 PGM-33 N-550 N-330 N-220

Moony viscosity2 minˉ¹3 minˉ¹4 minˉ¹

Mb,N.m

Fig. 4. Values of Mooney viscosity ML and torque MB for the SBR compounds filled with 50 phr of different types of carbon black (PM-15, PGM-33, N-550, N-330 и N-220) at different rotors revolutions (2, 3, 4 min-1).

20 30 40 50 60 70 8020

30

40

50

60

70Moony viscosity

ML

phr carbon black N-550

Fig. 5. Dependence of the Mooney viscosity on the carbon black N-550 content in SBR based compounds.

90 95 100 105 110 115 120 125 13030

32

34

36

38

40

42

44

x

2 min-1

MB,[N.m]

t, 0C

90 95 100 105 110 115 120 125 13034

36

38

40

42

44

46

48

50

x

4 min-1

MB,[N.m]

t, 0C

Fig. 6. Dependence of the torque MB on the temperature for the SBR compound filled with 30 phr carbon black N-550 at 2 min-1 (a) and 40 phr carbon black N-550 at 4 min-1 (b).

a)

b)

a)

b)

D. Zheleva

245

In the Fig. 6 and Fig. 7 are depleted the bands lining between the curves of highest and lowest torque values obtained for each of the revolutions defined at 100оС. It is clear that at low level of filling the band width is smaller (fig.6a) and at high filling level it is greater (Fig. 7a,b).

Following the results obtained for the torque at 100оС and Mooney viscosity in dependence on the car-bon black content in the rubber compounds are shown in Fig. 8.

It is settled that for carbon black contents of 30, 40 and 50 phr the Mooney viscosity is lower in comparison with the respective torque values, and at 60 and 70 phr the Mooney viscosity values are close to the highest torque values.

Fig. 9 shows the correlation between Mooney vis-cosity ML and torque MB.

On this graphic the dotted line presents the real correlation. This line is described by the following equations:

1.8 45.2BML M= − or ( 45.2) 1.8BM ML= + The torque could be calculated on the base of the

cited equations if the Mooney viscosity is available and

90 95 100 105 110 115 120 125 13036

38

40

42

44

46

48

50

52

54

56

58

60

x

4 min-1MB,

[N.m]

t, 0C

90 95 100 105 110 115 120 125 13040

42

44

46

48

50

52

54

56

58

60

62

x

3 min-1

MB,[N.m]

t, 0C

Fig. 7. Dependence of the torque MB on the temperature for the SBR compound filled with 50 phr carbon black N-550 at 4 min-1 (a) and 60 phr carbon black N-550 at 3 min-1 (b).

30 40 50 60 7025

30

35

40

45

50

55

60

65

7070

65

60

55

50

45

40

35

30

ML MB,[N.m]

phr carbon black

4 min-1

3 min-1

2 min-1

Moony viscosity ML

Fig. 8. Dependence of the Mooney viscosity ML and torque МB on the concentration of carbon black N-550 at different rotors revolutions.

25 30 35 40 45 50 55 60 65 7025

30

35

40

45

50

55

60

65

70MB

ML

4 min-1

3 min-1

2 min-1

Fig. 9. Correlation between Mooney viscosity ML and torque МB at different rotors revolutions on Plasticorder Brabender.

vice versa the Mooney viscosity could be calculated if the torque is available for the investigated compounds. Such calculation could be of practical importance if it serves as a base for obtaining of at least approximated values for rheological properties of carbon black filled SBR compounds. This very useful taking into account the following: The Ostwald de Waele equation is stating [2,4], that the effective viscosity of polymers can be cal-culated, if the consistency index K and the rheological flow index n are available. The rheology stated that, if K in the exponential rheological equation represents the shear stress at shear rates of

•

γ =1, i.e. the torque occurs just during these conditions. Furthermore, it is known that the flow index n for filled rubber compounds is usu-ally between 0,2 to 0,4 and, it is possible to calculate approximately the effective viscosity of the investigated rubber compounds on the base of the results obtained. Thus, starting with measurements of the Mooney visco-sity, it becomes possible to obtain the very useful from practical point of view results concerning the rheological properties of carbon black filled SBR rubber compounds. As concerning the reliability of the results mentioned

a)

b)


246

above it is necessary to do the same measurements with a capillary viscometer which will be an object of our further investigation.

CONCLUSIONS

The hypothesis that it is possible to investigate the torque MB for rubber compounds at a given temperature without preliminary tempering of the mixing chamber is confirmed.

A method for torque measurements of rubber com-pounds by means of a Plasticorder Brabender at minimal rotor revolutions in the mixing camera is developed.

The correlation between Mooney viscosity ML and torque MB by means of a Plasticorder is settled, and this correlation is revealed by a complete coincidence of their values for rubber compounds containing 50 phr carbon black of the types PM-15, PGM-33 and N-550 at 3 min-1, and for N-330 at 2 min-1. Concerning N-220 carbon black no correlation can be founded.

A correlation is revealed between ML and MB for SBR rubber compounds with N-550 carbon black at the levels of 30, 40, 50, 60 and 70 phr and an equation for this dependency is worked out:

1.8 45.2BML M= −or

( 45.2) 1.8BM ML= + .

REFERENCES

1. J. Mark, B.Erman, F.Eirich, Science and Technology of Rubber, Elsevier 3, Ed., 2005, 237

2. Е.Severs, Rheology of polymers, Chemistry, Moscow, 1966, 35, (in Russian).

3. Standard method ISO 289-1:2002. Uncured rub-ber. Determination by disk viscometer. Part 1. Determination of Mooney viscosity.

4. R.Chhabra, J.Richardson, Non-Newtonian Flow in the Process Industries, 1999, 3.

5. E. Djagarova, D. Jeleva, Z. Zdravkov, Une possibilité d‘élargir l‘information obtenue par le Plasticorder Brabender, J. Univ. Chem. Technol. Met. (Sofia), 37, 5, 2002, 71.

6. R.S. Popovic, M. Plavsic, R.G. Popovic, N. Ilic, Correlation between viscosity measured by capil-lary rheometer and Mooney viscometer on testing rubbers and rubber mixes, Kautschuk und Gummi Kunststoffe, 44, 7, 1991, 702.

7. Zuo M., Zheng Q., Correlation between rheological behavior and structure of multi-component polymer systems, Science in China Series B: Chemistry, 51,1, 2008, 1.

8. V. Kuleznev, Polymers blends (book), Chemistry Ed., Moscow, 1996, (in Russian).

9. N. Dishovsky, G. Tzenkov, Hadbook of Rubber, Sofia, 2006, 163, (in Bulgarian).

I. Spiridonov, M. Shopova

247


DETERMINATION OF THE EFFECT OF GRAY COMPONENT REPLACEMENT LEVEL ON COLORIMETRIC CHARACTERISTICS OF COLOR PROOF



ABSTRACT

The main goal of this study is determination the effect of GCR (gray component replacement) levels on colori-metric characteristics of color proof. To determine the effect of GCR levels on colors of color proofs, a comparison of 2D and 3D color gamuts depending on GCR level have been performed. In addition to obtain better assessment of the effect of GCR levels on color gamut were calculate volumes of 3D color gamuts and 2D surface areas. In order to determine the effect of GCR on colorimetric characteristics is performed colorimetric evaluation, expressed as color difference ΔE*

ab.

Keywords: gray component replacement, color gamut, color proof.

Received 20 December 2012Accepted 15 May 2013

INTRODUCTION

The main purpose of using a digital color proof-ing press is to simulate the visual characteristics of the finished production prints as closely as possible. In our study the aim is to simulate four color sheet-fed offset press with specification FOGRA39.

In conventional multicolor printing, the chromatic inks cyan (C), magenta (M) and yellow (Y) reproduce the color shades, and black ink is used to increase im-age quality and save inks. Increasing of image quality is expressed as increasing the gamut size, improving the details, obtaining more dark colors and making gray balance more stable [1-3].

An important characteristic of an output device is its color gamut, or the range of its reproducible colors. This range of colors can be thought of as a volume in 3D color space. The gamut is usually specified in a colorimetric or visually based space such as CIE L*a*b*, where L* correspond to lightness, a* to red-green color (+a* - red, -a* - green), and b* to yellow-blue color (+b* - yellow, -b* - blue). Knowledge of the size and shape of color

gamut surface is useful for determination how colors outside the color gamut should be reproduced [4-7].

There are several methods for generation and con-trolling the amount of substitution of chromatic colors – cyan, magenta and yellow (CMY) with black ink. The most commonly implemented method in practice for generation of achromatic composition is GCR (gray component replacement). Many CMY combinations contain certain amount of “gray component”. The “gray component” is a combination of inks which, if printed alone, will produce a neutral gray. The replacement of gray component with black ink reduce the total amount of ink without changes in colors [8, 9].

Theoretically the substitution of chromatic inks with a black one can be easily performed, but in practice there are certain inaccuracies, especially in neutral colors. For more than 50% reduction of chromatic inks (50% GCR), differences occur. Using the higher level of reduction of the chromatic inks lead to lowest optical density and lightness in shadow tones. Improper settings of GCR level can cause significant color deviations in the im-age, because black ink cannot replace the colorfulness


248

of chromatic inks [10, 11]. The main goal of this study is determination the

effect of GCR (gray component replacement) levels on colorimetric characteristics of color proof. To determine the effect of GCR levels on colors of color proofs, a comparison of 2D and 3D color gamuts depending on GCR level have been performed. In addition to obtain better assessment of the effect of GCR levels on color gamut were calculate volumes of 3D color gamuts and 2D surface areas. In order to determine the effect of GCR on colorimetric characteristics is performed colorimetric evaluation, expressed as color difference ΔE*ab. The color difference was calculated after conversion of spectral data to CIE L*a*b* color coordinates.

EXPERIMENTAL

A special test form that contain different control strips and elements, test chart ECI 2002 CMYK ran-dom, test images has been designed. The test chart ECI 2002 CMYK random contains 1485 color patches with different combinations of cyan, magenta, yellow and black, which are used to evaluate the effect of GCR level. The test form was printed on Proofing System Epson Stylus Pro 9900, color proofing device, certified for color conformation of offset lithography printing ISO 12647-2 [12, 13].

The used paper and inks are in accordance to ISO 12647-7 [14].

In order to visually match a specific printing condi-tion, proofing processes requires a set of parameters to be specified that are not necessarily identical to those listed in ISO 12647-1 or another part of ISO 12647. This is caused by differences in colorant spectra or phenomena such as gloss, light scatter (within the print substrate or the colorant), and transparency. Therefore a spectrophotometer/ densitometer SpectroEye of X-Rite has been used for measuring a color characteristics in the CIE L*a*b* color space. All measurements are in accordance with ISO standards: D50 illuminant, 2° observer, 0/45 or 45/0 geometry, black backing. The spectral measurements shows, that the digital color proofing system is simulating correctly the colors ac-cording to ISO 12647-7.

A spectral measurements have been performed for the printed test charts using spectrophotometer X-Rite i1Pro and automatic scanning device i1i0, considering

values in the wavelength range of 380 to 730 nm with a step of 10 nm (internal step 5 nm). The measurements are performed according to ISO 13656 [15].

Four ICC color profiles have been created by Profile Maker, X-Rite, with GCR minimum level (signed below as min GCR), with medium GCR (signed below as GCR 1), with heavy GCR (signed below as GCR 2) and with maximum GCR (signed below as max GCR). Each of these four ICC profiles was applied to the test form and was printed under the same print conditions. The visualization of color gamuts and calculation of color gamut volumes were performed with Color Think Pro, Chromix and Surfer, Golden Software Inc.

A series of spectral measurements (at the conditions listed above) with different GCR levels have been made for each printing sheet. The spectral data were converted to tristimulus values CIE XYZ. [16, 17] CIE L*a*b* values were calculated from the tristimulus values. The color difference ΔE*ab was calculated (the calculations for ΔE*ab were performed to GCR minimal value as a reference).


3D and 2D presentation of color gamuts gives pre-cise and comprehensive information of colors, that can be reproduced in the specific conditions. Presentation of 3D color gamut provides general information and shows the shape of color body. 2D gamut presentation at different cross-section of CIE L* coordinate, gives more detailed information for analyses and comparison. Therefore it is very important to determine changes in color gamuts in dependence of gray component replace-ment level.

At Fig. 1 are presented 3D color gamuts depending on GCR level viewed from different angles.

The comparison of 3D color gamuts (Fig. 1) shows that GCR maximal level has generally the biggest color gamut. It is clearly visible that using minimal level of GCR lead to lower color gamut.

2D gamuts at different cross-section of CIE L* coordinate for dark, middle and highlight tones are presented on Fig. 2. These cross-sections of 3D hull of color gamut are chosen, because the human eye has a different sensitivity in dark, middle and highlight tones.

Fig. 2 shows that in the dark tones (Fig. 2 a and 2b) the color gamut at minimal level of GCR is consider-


249

ably smaller than the others. The biggest color gamut is obtained at maximal level of GCR. The color gamuts at maximal level of GCR and level 2 of GCR are similar. In yellow-green area at L= 22 (figure 2a) the color gam-uts are similar, except color gamut at minimal level of GCR, and in blue-violet area at L = 30 (Fig. 2b) all color gamuts are similar. In middle tones (Fig. 2c) generally all color gamuts are similar, but only in one small part in yellow area difference occur. In highlight tones (Fig. 2d and 2e) the biggest color gamut is obtained at level 2 of GCR, and the smallest color gamut, at maximal level of GCR. In yellow-green and yellow-red areas the color gamuts are similar.

In addition to graphical comparison of color gamuts we have calculated color gamut volumes. The obtained results are given in Table 1.

According to the results shown in Table 1, the big-gest color gamut volume is obtained at maximal level of GCR and the smallest – at minimal level of GCR. The highest difference in volumes of color gamuts is only 2%, and the lowest – 0.58%. A difference in color gamut volumes with such magnitude is negligible.

GCR maximal level has generally the biggest color gamut and GCR minimal level has the smallest color gamut, which can be seen from 3D color gamuts (Fig. 1) and from their volumes (Table 1).

For better assessment of changes in color gamuts the surface areas of 2D cross-sections of 3D color gamuts were calculated depending on GCR levels in highlights, middle and dark tones. Calculated surface areas are presented in Table 2.

The obtained results shows that the values of surface areas at maximal level of GCR are considerable bigger than these for the others GCR levels. It is clearly visible that at CIE L* 17 to 60, the highest values of surface areas are obtained at maximal level of GCR, and the lowest values are obtained at minimal level of GCR. After that, CIE L* > 60, the highest values of surface areas are obtained at level 2 of GCR, and the lowest

Fig. 1. Comparison of 3D color gamuts for different GCR levels in CIE L*a*b*.

GCR level Color gamut volume, ∆E3

min 333966

1 338799

2 334590

max 340762

Table 1. Color gamut volumes depending on GCR levels.


250

values - at maximal level of GCR. In order to determine the effect of GCR on colo-

rimetric characteristics of color proof is performed colorimetric evaluation, expressed as color difference ΔE*ab. The color difference was calculated using GCR minimal level as a reference.

The evaluation has been made for patches in high-lights, middle and dark tones. Detailed information for these patches is represented in Table 3. These color patches have been chosen, because they represent some of most important tones and shades near the gray axis in CIE L*a*b* system. The obtained results for ΔE*ab in highlight tones are shown on figure 3, for ΔE*ab in middle tones on figure 4 and for ΔE*ab in dark tones on Fig. 5.

The graph (Fig. 3) shows considerable small color difference, by human perception point of view. The minimal value of color difference is 0.12 units, and the maximal is 0.78 units, both obtained at GCR maximal level. It means that in tones and colors, which are close to neutral, the GCR level does not have a big impact for

Fig. 2. 2D color gamut of colors under different GCR levels in CIE L*a*b*: а/ by L* = 22 /dark tones/; b/ by L* = 30 /dark tones/; c/ by L* = 50 /middle tones/; d/ by L*= 75 /highlight tones/; e/ by L*=80 /highlight tones/.

At CIE L* Surface areas, ∆E2

GCR level

min 1 2 max

17 56.45 121.96 202.28 254.48

22 207.77 547.57 542.16 827.94

25 724.96 1122.27 1450.23 1602.88

30 2133.55 2518.84 2829.92 3001.90

40 5191.19 5349.24 5469.56 5665.99

50 8010.02 8119.10 8124.80 8205.77

60 5219.62 5356.13 5474.69 5682.50

70 4595.20 4653.20 4705.93 4498.62

75 3439.24 3507.28 3603.24 3041.13

80 1982.88 2052.91 2383.85 1740.75

86 395.31 448.60 537.48 388.51

Table 2. Surface areas depending on GCR levels.


251

color accuracy in tones near to gray axis. There is a big difference in values at patch 2L3

on Fig. 4. The highest value of ΔE*ab is 1.89 units (at maximal level of GCR) and the lowest value is 0.42 units. A difference with such magnitude is observed only for this patch.

The graph on Fig. 5 shows that the highest values of color difference are obtained for solid patches. The lowest value of color difference is 0.41 units, and the highest – 1.59 units.

For more precise evaluation of the effect of GCR level color accuracy, it have been calculated the average color difference. The average color difference provide valuable information about differences in colors by hu-man perception point of view. It is important because it shows the difference in all parts of spectral data, all shades and colors. The obtained results for average, minimal and maximal color difference for all 1485 patches are given in Table 4.

The average color difference for all 1485 patches is

Patch ID Cyan, %

Magenta, %

Yellow, %

2O32 3 3 3

Hig

hlig

ht to

nes

2P32 5 3 3 W32 7 7 7 2R1 10 6 6 G17 10 10 10 A28 20 10 10 2S2 20 12 12 2G6 20 20 10 2S23 20 20 20 H26 30 20 20

2A16 40 30 30

Mid

dle

tone

s M20 40 40 40 2L3 55 30 30 R20 55 40 40 F3 55 55 55

2U4 60 45 45 M19 70 40 40 E25 70 55 55

Dar

k to

nes

2G3 70 70 70 C32 85 55 55 A2 80 65 65

2H29 85 70 70 W31 100 85 85 O20 100 100 100

Table 3. Tone value information for chromatic patches in highlight, middle and dark tones.

Fig. 3. ΔE*ab for printed colors in highlight tones depending

on GCR level.

Fig. 4. ΔE*ab for printed colors in middle tones depending

on GCR level.

Fig. 5. ΔE*ab for printed colors in dark tones depending on

GCR level.

calculated by equation 1. 1 2 1485/ min / min / min...

1485

Field Field FieldSample GCR Sample GCR Sample GCR

AVERAGE

E E EE

∆ + ∆ + + ∆∆ =

(1)where, ∆EAVERAGE – mean arithmetic colour differ-ence of 1485 measured fields between the specific GCR level sample and the GCR minimum level, Field

GCRSampleE min/∆


252

– colour difference between a specific sample color field with different GCR level and the same field with minimum GCR level.

According to the results listed in Table 4, the big-gest average color difference is obtained for the GCR level 1. The smallest value of average color difference is obtained at GCR level 2. There is a small difference in values of average, minimal and maximal color dif-ference (0.33 units ∆E*ab, 0.04 units ∆E*ab, and 0.53 units ∆E*ab, respectively). It means that the difference remain relatively constant.

CONCLUSIONS

The comparison of 3D color gamuts shows that GCR maximal level has generally the biggest color gamut. It must be noted that using minimal level of GCR lead to lower color gamut.

According to the results from comparison of 2D gamuts at cross-section of CIE L* coordinate in dark and middle tones the biggest color gamut is obtained at maximal level of GCR, and the smallest at minimal level of GCR. In highlight tones the biggest color gamut is obtained at level 2 of GCR, and the smallest color gamut, at maximal level of GCR.

The results of calculated color gamut volumes shows that the biggest color gamut volume is obtained at maxi-mal level of GCR and the smallest – at minimal level of GCR. The highest difference in volumes of color gamuts is only 2 %, and the lowest – 0.58 %. A difference in color gamut volumes with such magnitude is negligible.

The obtained results for surface areas of 2D cross-sections of 3D color gamuts shows that the values of surface areas at maximal level of GCR are consider-able bigger than these for the others GCR levels. It is clearly visible that at CIE L* 17 to 60, the highest values of surface areas are obtained at maximal level

GCR level ∆E*ab,AVERAGE ∆E*

ab,MIN ∆E*ab,MAX

1 0.93 0.03 5.54

2 0.60 0.06 5.17

max 0.77 0.02 5.70

Table 4. Average, maximal and minimal color difference. of GCR, and the lowest values are obtained at minimal level of GCR. After that, CIE L* higher than 60 units, the highest values of surface areas are obtained at level 2 of GCR, and the lowest values - at maximal level of GCR. There is a big differences between surface areas up to 77 % at cross-sections in dark tones. This big difference decrease to 8 % at cross-sections in middle tones and to 27 % at cross-sections in highlight tones. This phenomenon is very important of practical point of view, because the volume of colors is one of the most important factors, that impact on human perception and therefore on print quality. Therefore it is very important to determine changes in color gamuts in dependence of gray component replacement level.

GCR level 2 generally has color gamut similar to those at GCR maximal level in dark tones, meanwhile has the biggest color gamut in highlight tones. This is confirmed by the graphical comparison of color gamuts and by their volumes and surface areas.

According to the results from colorimetric evalu-ation for colors in highlight tones, expressed as color difference ΔE*ab, there is considerable small color difference, by human perception point of view. The minimal value of color difference is 0.12 units, and the maximal is 0.78 units, both obtained at GCR maximal level. It means that in tones and colors, which are close to neutral, the GCR level does not have a big impact for color accuracy in tones near to gray axis. The color difference for printed colors in middle tones is about 1.89 units, and in dark tones – 1.59 units. From human perception point of view that is considerable difference.

For more precise evaluation of the effect of GCR level color accuracy, the average color difference have been calculated. According to the results, the biggest average color difference is obtained for the GCR level 1, and the smallest value is obtained at GCR level 2. There is a small difference in values of average, minimal and maximal color difference (0.33 units ∆E*ab, 0.04 units ∆E*ab, and 0.53 units ∆E*ab, respectively). It means that the difference remain relatively constant.

It must be noted that in dark tones GCR maximal level has the biggest color gamut, while in highlight tones, it has the smallest color gamut. The color gamuts at maximal level of GCR and level 2 of GCR are similar in dark tones, but in highlight tones GCR level 2 has the biggest color gamut. According to the colorimetric evaluation, GCR level 2 has the smallest maximal and


253

average color difference.A research study and implementation of methodol-

ogy from this research should be performed for running an experiment in conditions of sheetfed offset and web offset printing for different printing substrates. The re-sults obtained from real production conditions should be compared with digital color proofs systems.

In future, by collected data from this research, it could be developed mathematical model describing relationship between ink quantity of process colors - C, M, Y, K, GCR levels and color reproduction accuracy. Certainly it will be very useful for predicting of correct color reproduction and choosing the correct level of GCR in dependence of printing conditions.

AcknowledgementsThis study was funded by the Bulgarian Science

Fund (DMU 03/69/2011).

REFERENCES

1. G. Sharma, Digital Color Imaging Handbook, CPC Press LLC, Boca Raton, FL, 2002.

2. H. Kipphan, Handbook of Print Media, Technologies and Production Methods, Springer-Verlag Heidelberg, Berlin, 2001.

3. B.-H. Kang, H.-K. Choh, C.-Y. Kim, Black color replacement using gamut extension method, NIP21: International Conference on Digital Printing Technologies, September 2005, 21, 384-386.

4. R. Balasubramanian, E. Dalal, A method for quantify-ing the color gamut of an output device, Proc. SPIE, 3018, 1997, 110–116.

5. T.J. Cholewo, S. Love, Gamut boundary deter-mination using alpha-shapes, Proc. 7th Color Imaging Conference: Color Science, Systems and Applications, 7, 1999, 200–204.

6. I. Farup, J.Y. Hardeberg, A.M. Bakke, S. Kopperud, A Rindal, Visualization and interactive manipulation of color gamuts, Proc. IS&T and SID’s, 10, 2002,

250–255.7. I. Spiridonov, M. Shopova, R. Boeva, M.Nikolov,

The effect of different standard illumination condi-tions on color balance failure in offset printed images on glossy coated paper expressed by color difference, Phys. Scr., T149, 2012, 014019.

8. E. Neumann, M. Bohan, Ink Optimization: An Evaluation of the Different Strategies, GATFWorld, 20, 2, 2008, 46-48.

9. R. de Queiroz, K. Braun, R. Loce, Detecting spatially varying gray component replacement with application in watermarking printed images, JEI-14-033016, 14, 3, 2005.

10. D. Agić, M. Gojo, M. Strgar-Kurečić, Determination of equivalent-density domain in black compensa-tion implementation for selected profile, Technical Gazette, 18, 1, 2011, 63-68.

11.T. Costa, Effect of GCR and TAC in Color Gamut Volume, Test Targets 4.0, Advanced Color Manage-ment RIT School of Print Media, Rochester, New York, USA, 2004.

12. ISO 12647-2, Graphic technology - Process control for the production of half-tone colour separations, proof and production prints, Part 2: Offset litho-graphic processes, 2004.

13. ISO 12647-2/Amd.1, Graphic technology - Process control for the production of half-tone colour sepa-rations, proof and production prints, Part 2: Offset lithographic processes, 2007.

14. ISO 12647-7, Graphic technology - Process control for the production of half-tone colour separations, proof and production prints, Part 7: Proofing pro-cesses working directly from digital data, 2007.

15. ISO 13656, Graphic technology - Application of reflection densitometry and colorimetry to process control or evaluation of prints and proofs, 2000.

16. ISO 11664-1 (CIE S 014-1/E:2006), Colorimetry, Part 1: CIE standard colorimetric observers, 2007.

17. CIE 15, Technical Report Colorimetry, 3rd Edition, 2004.


254


RAPID SPECTROPHOTOMETRIC METHOD FOR DETERMINATION OF HEXAMETHYLENETETRAMINE (UROTROPINE) IN FOOT CARE PRODUCTS

A. Tachev1, V. Christova-Bagdassarian1, N. Vasileva1,

A. Dimitrova2, M. Atanassova3

1 National Centre of Public Heath and Analysis, Ministry of Health, Sofia, 15, Akad. Iv. Ev. Geshov Blv. 1431 Sofia, Bulgaria Е-mail: a.tachev@ncphа.government.bg2 Technical University of Sofia, Faculty of Industrial Technology, Sofia, Bulgaria3 Metallotechnika Ltd.

ABSTRACT

Hexamethylenetetramine (urotropine) is a compound that is often used in foot care products. Urotropine (hexamethylenetetramine) has a broad antibacterial effect (due to release of formaldehyde), which ensures the elimination of already occurring microorganisms and prevent the recurrence of infections, so commonly is used in cosmetic products for skin care of the feet (creams, lotions, deodorants, etc.). In cosmetic products are usually dosed at concentrations of 0.05 % to 0.15 %. Formaldehyde is known for its toxic and allergic effect on skin. There is evidence for its possible carcinogenic effects. An easily accessible spectrophotometric method for determination of urotropine in cosmetic products was developed and validated. The method is based on the interaction of urotropine with chromotropic acid in sulfuric acid medium, forming a characteristic colored compound with a maximum light absorption at a wavelength of 570 nm. The LOD is 0.01 %. The LOQ is 0.02 %. Repeatability (SD) at urotropine concentration 0.10 % in basic foot care product with a homogeneous viscous consistency is 0.0022 %. Reproducibility at urotropine concentration 0,10 % in basic skincare product, expressed by RSD, is 1.51 %. The recovery at urotropine concentration 0.10 % in a cosmetic product is 97.17 % (90.00 % - 110.00 %) and at urotropine concentration 0.15 % is 98.33 % (95.00 % - 100.00 %). The developed method was applied for the analysis of five samples of cosmetic products for the hygiene of the feet, containing urotropine, as a preservative. The measured concentrations are within concentration interval permitted by law: from 0.11 % to 0.14 %.The validated spectrophotometric method for urotropine determination in foot care products is rapid, with sufficient sensitivity, accuracy and reproducibility and does not require complicated and expensive equipment. It is easily applied in control of cosmetic products containing the preservative urotropine.

Keywords: urotropine, foot care products, spectrophotometric method.

Received 20 June 2012Accepted 15 April 2013

INTRODUCTION

Hexamethylenetetramine (urotropine) is a compound that is often used in foot care products. The pure substance is a white crystalline powder obtained by reaction of 6 mol of formaldehyde and 4 mol of ammonia. The particle size ranges between 80 and 800 µm. Melting point: 270°C. Urotropine can be absorbed through the skin and it may cause allergic reactions, usually occurring in the form of rashes.

Common complaint is excessive sweating, often accompanied by unpleasant odor, which increases with increased physical activity, in stress and at high temperature during summer months The excessive sweating is often localized in the feet, underarms and palms. Moisture, increased in the skin, creates favourable conditions for the emergence and growth of bacteria and fungi that can give rise to some serious diseases and complications. Odours are resulting from the waste products of vital functions of micro organisms. Most

A. Tachev, V. Christova-Bagdassarian, N. Vasileva, A. Dimitrova, M. Atanassova

255

unpleasant is the excessive sweating of the feet, which is always accompanied by a particularly unpleasant odour. Respect for personal hygiene is one of the most necessary, but unfortunately, not sufficient condition to tackle this unpleasant condition. Therefore, in most cases it is essential to effective antiperspirants. Urotropine (Hexamethylenetetramine) shows a broad-spectrum antibacterial activity (release of formaldehyde), which ensures the elimination of already occurring micro organisms and prevention the recurrence of infections, so commonly used in cosmetic foot care products (creams, lotions, deodorants and others. ). In cosmetic products for the hygiene of the feet urotropine is usually dosed at concentrations of 0.05 % to 0.15 %. Formaldehyde is known for its toxic and allergic effect on skin [1]. When uroptopine is used as a preservative at concentrations up to 0.15 % in non-aerosol products, in the literature does not mention its irritant and sensitizer effect on people. This content urotropine release formaldehyde at concentrations less than 0.2 %. But if urotropine is used as the active ingredient in concentrations higher than 0.15 % , in the literature are available data on the sensitizing effect at concentrations equal to or above 0.25 %. There have been studies on the teratogenic and carcinogenic effects of urotropine, but this effects are not proved so far [1-3].

Urotropine and content of formaldehyde in cosmetic products have strict standards in the European and in the harmonized with it Bulgarian legislation. The content of urotropine in preservatives is allowed to 0.15 %, the formaldehyde content in products not intended for oral hygiene – to 0.2 % and in products for oral hygiene -to 0.1 % [4, 5]. In the published methods reference methods for checking composition of cosmetic products (Annex 10 of the Ordinance № 36) [4] no method is mentioned for determination of urotropine. In the Ordinance № 36 and in the literature [4-8] appear sophisticated techniques for formaldehyde determination, when used alone or with other preservatives that are not sources of formaldehyde or in a substance, which is a source of formaldehyde. These methods require complicated and expensive equipment, reagents and materials.

The aim of this study was the development of an easily accessible spectrophotometric method for determination of urotropine in foot care products. The method is based on the interaction of urotropine with chromotropic acid in sulfuric acid medium, forming a

characteristic colored compound with a maximum light absorption at a wavelength of 570 nm.

EXPERIMENTAL

Material and methods As base was used a cosmetic product not containing

urotropine, with homogeneous viscous texture and a certified reference substance urotropine. Samples were prepared in the laboratory by mixing of basic cosmetic product with urotropine and subsequent homogenization.

The development of the method is carried out according to BDS EN ISO / IEC 17025:2006 [9] requirements for the parameters: limit of detection (LOD), limit of quantification (LOQ), repeatability, reproducibility, recovery.

Limit of detection (LOD) and Limit of quantification (LOQ) are defined as follows:

3bl blLOD C SD= ± × ,where:Cbl is the concentration of urotropine in the blank,SDbl is standard deviation of the blank samples (n=10). and

6bl blLOQ C SD= ± × .To determine the values of LOD and LOQ were

developed n number of samples with basic cosmetic product free of urotropine in it. Demonstration of repeatability and recovery was performed using basic cosmetic product free of urotropine content.

To determine the repeatability, expressed by the standard deviation (SD), samples were tested on different days, at the level of concentrations of urotropine 0.15 %. Reproducibility was expressed by relative standard deviation (RSD).

To determine the recovery were prepared samples contained basic cosmetic product inforced by urotropine as active ingredient at two concentrations - 0.10% and 0.15 %. The concentration is chosen in accordance with the quantities used in cosmetic products.

Equipment, chemicals, standards and reagents - A UV-VIS Spectrophotometer CARY Varian

(Varian Australian Pry.Ltd.) was used with cells of 1 mm thickness

- sulfuric acid, diluted with distilled water (1: 2)- chromotropic acid (0.1 g of chromotropic acid

dissolved in sulfuric acid - 1:2 solution and diluted to


256

the mark in a volumetric flask of 100.0 cm3)- urotropine standard solution in distilled water at a

concentration 1.0 mg/cm3

Sampling and Sample preparation In an Erlenmeyer flask 200.0 cm3 were weighed

0.5 g ± 0.0001 grams of basic cosmetic product on an analytical balance. Then 25.0 cm3 chromotropic acid solution and 50.0 cm3 of the sulfuric acid solution (1:2) were added. In another Erlenmeyer flask of 200.0 cm3 were introduced 6.00 cm3 of urotropine standard solution, 25.0 cm3 of chromotropic acid solution and 50.0 cm3 of sulfuric acid solution (1:2). In the third Erlenmeyer flask of 200.0 cm3 (blank) were introduced 75.0 cm3 of sulfuric acid solution (1:2).

The three flasks are placed in a boiling water bath for 30 minutes, then immediately cooled to room temperature, transferred into three volumetric flasks of 100.00 cm3 and completed to the mark with sulfuric acid solution (1:2).

Then the concentration of sample solution at a wavelength (λ = 570 ± 2 nm) was measured against developed the same reference sample of known concentration (6.0 mg, which sets the camera).


The contents of urotropine (X) in cosmetic products, in % is given by:

mCX

××

=1000

100,

where:С – the concentration of urotropine in mg;m – mass of sample taken for analysis in g.

The results were processed statistically using statistical software package SPSS 10.0.

To calculate the limit of detection (LOD) and the limit of determination (LOQ) of the method ware prepared using basic cosmetic product containing no urotropine. The results of LOD and LOQ of urotropine are presented in Table 1.

To demonstrate the repeatability of the method are prepared samples of basic foot care product with a homogeneous viscous consistency with the spike at 0.10 % urotropine. Tests were conducted on three consecutive days (Table 2). The good repeatability of the results can be seen from the calculated standard deviation (SD), in terms of repeatability – 0.0022 %.

Number of

samples

n

Concentration (C) LOD LOQ

_ X, %

SD

%

%

10 0.0098 0.0010 0.01 0.02

Table 1. Limit of detection (LOD) and limit of determination (LOQ) of urotropine in basic cosmetic product.

Spike level for

urotropine,

%

Concentration, founded in spike samples (C), %

Number

of samples

n

X, %

SD, %

RSD,

%

0.10

6

0.0952

0.0022

1.51

Table 2. Repeatability of the results of determining the concentration (C %) at a rate of 0.10 % urotropine in basic cosmetic product.

A. Tachev, V. Christova-Bagdassarian, N. Vasileva, A. Dimitrova, M. Atanassova

257

The reproducibility is expressed by the relative standard deviation (RSD) under repeatability conditions: 1.51 %.

The recovery was evaluated at concentrations of urotropine in basic cosmetic 0.10 % and 0.15 % (total for both concentrations 12 samples) and data are presented in Table 3. The resulting recovery was satisfactory and at a concentration of 0.10 % was ranged from 90.00 % to 110.00 % and 0.15 % concentration – was ranged from 95.00 % to 100.00 %.

The developed method was applied to the analysis of five samples of foot care cosmetic products, containing urotropine, as a preservative. The measured concentrations are within permitted by law - from 0.11 % to 0.14 % [4, 5]. The results obtained from tests of commercial cosmetic products are presented in Table 4.

CONCLUSIONS

A spectrophotometric method for determination of urotropine in foot care cosmetic products has been developed and described. The method is based on the

interaction of urotropine with chromotropic acid in sulfuric acid medium, forming a colored compound with a maximum light absorption at a wavelength of 570 nm.

The method is validated based on 28 trials and evaluated its parameters. The limit of detection (LOD) is 0.01 %. The limit of determination (LOQ) is 0.02 %. The repeatability (SD) is 0.0022 % at a concentration of urotropine in basic skincare product with a homogeneous viscous consistency 0.10 %. The reproducibility at a concentration of urotropine 0.10 % in basic foot care product with a homogeneous viscous consistency, presented by RSD, in repeatability conditions is 1.51%. The recovery at urotropine concentration of 0.10 % in a cosmetic product is 97.17 % (90.00 % - 110.00 %) and at urotropine concentration of 0.15 % is 98.33 % (95.00% - 100.00 %).

Validated spectrophotometric method for urotropine determination in cosmetic foot care products described here is rapid, with sufficient sensitivity, accuracy and reproducibility and does not require complicated and expensive equipment It is easily applied in control of cosmetic products containing the preservative urotropine.

Spike level for urotropine,

%

Number of

samples n

Concentration, founded in spike samples (C), %

Recovery, %

SD,

%

RSD,

% min

max

X

min

max

X

0.10 6 0.09 0.10 0.0971 90.00 100.00 97.17 7.1671 7.36

0.15

6 0.14 0.15 0.1466 93.33 100.00 98.33 4.0825 4.15

Table 3. Recovery for urotropine at levels 0.10 % and 0.15 % in basic cosmetic product.

Code of cosmetic product

Number of measurements

(n)

Average results of measurements,

%

Standard deviation

SD, %

Relative standard deviation

RSD, %

1 6 0.11 0.0020 1.38 2 6 0.12 0.0024 1.44 3 6 0.12 0.0038 1.66 4 6 0.11 0.0046 1.77 5 6 0.14 0.0051 2.03

Table 4. Results of determination of urotropine in commercial cosmetic products for foot care.


258

REFERENCES

1. A. De Groot, I.R. White, M.A. Flwholm, G. Larsen, P.J. Coenraads. Formaldehyde-releasers in cosmetics: relationship to formaldehyde contact allergy. Part 2. Patch test relationship to formaldehyde contact allergy, experimental provocation tests, amount of formaldehyde released, and assessment of risk to con-sumers allergic to formaldehyde, Contact Dermatitis, 62, 1, 2010, 18-31.

2. CIR Reaches Verdich on Buthyl Myristate, HС Red №3, 2008, Cosmetics Toiletries Science Applied, 2008.

3. Environmental Working Group, 2012. Headquarters, 1436 U Street NW, Suite 100, Washington, DC 20009, 2012.

4. Ordinance № 36 of 30 November 2005 on the require-ments for cosmetic products (SG. 101/2005 amended & suppl. in SG. 44/2006, amended in SG.75 /2006, amended in SG 39/2007, amended. & suppl. in SG 106/2007, amended &suppl. in SG 80/2008, amended. & suppl. in SG 35 of 2009, amended. and suppl. in SG 2/2010, amended. & suppl. in SG 62/2010, amended. & suppl. in 53/2011). 5. DIR. 76/768/ EEC of 27 July 1976 On the Approximation of the Laws of the

Member States.5. Relating to Cosmetic Products, and its successive

Amendments and adaptations. Available from: http://europa.eu.int/comm/enterprise/cosmetics/ html/con-solidated_dir.htm>

6. M.A.H. De Kruijf, L.A. Rijk, Pranoto Soctarhi, A. Schouten. Determination of preservatives in cosmetic products: I Thin-layer chromatographic procedure for the identification of preservatives in cosmetic products, Journal of Chromatography A, 410, 1987, 395-411.

7. E. Lucas, E. Kiss, M. Kwast, M. Malanowska, B. Smietanka. Free formaldehyde determination in cos-metic products by the HPLC method. Rocz Panstw Zakl. Hig., 49, 4, 1998, 463-468.

8. Z. Urban-Morlan, , R. Castro-Rios, A. Chavez-Montes, L.M. Melgoza-Contreras, E.Pinon-Segundo, A.Ganem-Quintanar-Guerero. Determination of methenamine, methenamine mandelate and methe-namine hipputate in pharmaceutical preparations using – exchange HPLC, Journal of Pharmaceutical and Biomedical analysis, 40, 5, 2006, 1243-1248.

9. BDS EN ISO / IEC 17025/2006, General require-ments for the competence of testing and calibration, 2006.

K. Milenova, I. Stambolova, V. Blaskov, A. Eliyas, S. Vassilev, M. Shipochka

259


THE EFFECT OF INTRODUCING COPPER DOPANT ON THE PHOTOCATALYTIC ACTIVITY OF ZnO NANOPARTICLES

K. Milenova1, I. Stambolova2, V. Blaskov2, A. Eliyas1, S. Vassilev3, M. Shipochka2

1Institute of Catalysis, Bulgarian Academy of Sciences, Acad. G. Bonchev Str., Bl. 11, 1113 Sofia, Bulgaria E-mail: [email protected], 0897 401 5522 Institute of General and Inorganic Chemistry, Bulgarian Academy of Sciences, Acad. G. Bonchev Str., Bl. 11, 1113 Sofia, Bulgaria3 Institute of Electrochemistry and Energy Systems, Bulgarian Academy of Sciences, Acad. G. Bonchev bl. 10, 1113, Sofia, Bulgaria

ABSTRACT

Copper-doped ZnO nanoparticles have been prepared by the precipitation method. The dopant contents in the samples were 0.24, 0.35 and 1.07 at.%. A set of techniques including XRD, XPS, TG – DTA, EPR and BET analysis has been ap-plied to characterize Cu-doped ZnO samples. The results showed that the crystallite sizes of ZnO and Cu-doped ZnO nanoparticles were within the range of 45 ÷ 49 nm. The dopant exists in the form of isolated Cu2+ ions. According to the XPS analysis the copper ions are located mainly on the surface of the ZnO particles. The photocatalytic activity has been tested in the reaction of Reactive Black 5 discoloration under UV irradiation. Among all the investigated samples pure ZnO samples showed the best photocatalytic properties.

Keywords: ZnO, Cu-doped ZnO, photocatalysis, ultraviolet light, azo dye.

Received 11 January 2013Accepted 05 May 2013

IINTRODUCTION

ZnO is an unexpensive, n-type semiconductor with a wide band gap having optical transparency in the visible range. It crystallizes in a hexagonal wurtzite structure (zincite) with the following lattice parameters: c = 5.205 Å, a = 3.249 Å. The n-type semiconductor behavior is due to the ionization of excess zinc atoms in interstitial positions and the oxygen vacancies [1]. Surface defects play an important role in the photocatalytic activities of metal oxides as they increase the number of the active sites [2, 3]. For this reason it is interesting to study the effect of ZnO doping by transition metals on its photo-catalytic properties.

Data are reported in the current literature about the influence of copper dopant in ZnO powders and thin films on the photocatalytic behavior [4-6]. Various

techniques for the preparation of ZnO nanopowders have been applied: sol–gel method [4, 7], “soak-deoxidize-air oxidation” [5], co-precipitation method [6, 8] etc. Among the different methods, the co-precipitation ap-pears to be one of the most promising methods to prepare nanopowders. Some of the most important advantages of the precipitation method are: easiness of the synthe-sis, low temperature of decomposition and control on the chemical composition. These advantages make the precipitation technique a very attractive preparation method, especially in the case of photocatalytically ac-tive ZnO powders [6].

The photocatalytic activities of pure and Cu-doped ZnO powders have been evaluated by measuring the degradation of organic dye Methylene Blue [3] and Methyl Orange [4] in aqueous solutions under the UV-light irradiation. The photocatalytic activities of ZnO/


260

Cu2O composite material have been evaluated using Methyl Orange in visible light region [5]. The avail-able data about the influence of copper doping on the photocatalytic activity of ZnO are contradictory. It has been found out that the photocatalytic properties of ZnO/Cu2O composite, compared with pure ZnO, were improved greatly [5]. The photocatalytic activity of pure ZnO is nearly the same as that of CuO/ZnO powders [6]. Donkova et al. have found a decrease in the photocata-lytic performance of ZnO powders after doping with Cu in regard to degradation of Methylene Blue dye [9].

During textile colouring large quantities of toxic azo dyes with intensive colour are disposed of as waste waters and they appear respectively in the water-ways. It has been reported that some of the dyes are toxic and carcinogenic [10]. Reactive Black 5 (RB5) azo dye is commonly used in the textile industry and may cause serious environmental problems. Motivated by this fact, we focused this study of ours on copper doped ZnO powders, obtained by precipitation, and on investigation of their photocatalytic behavior towards degradation of textile dye Reactive Black 5.

EXPERIMENTAL

Synthesis of the samples. Undoped and doped with Cu zinc oxide samples have been prepared by precipi-tation method. Analytical grade of purity zinc sulfate (heptahydrate) ZnSO4.7H2O, sodium carbonate Na2CO3 and copper sulfate CuSO4.H2O were used as starting materials. In a typical experiment we dissolved 90 g of Na2CO3 in 850 ml H2O under heating and continuous stirring (solution 1). An amount of 20 g ZnSO4.7H2O was dissolved in 140 ml H2O under heating and continuous stirring (solution 2). Different calculated quantities of copper sulfate were added to the solution 2 such to obtain mixture of ZnO doped with 0.24, 0.35 and 1.07 at.% Cu. This mixed solution of ZnSO4.7H2O and CuSO4.H2O was added drop by drop to the solution 1. After adjusting the pH value to 11, the final mixture solution was stirred under heating for 10 minutes. The precipitate was sepa-rated by filtration, washed several times with distilled water to pH value 7 and dried in air. For preparation of the final catalyst samples, the corresponding precursors were heated for 3 h at 500 ºC in air.

The final samples doped with 0.24, 0.35 and 1.07 at. % Cu were denoted ZC1, ZC2 and ZC3, respectively.

Sample characterization ААS analysis. The chemical composition of the

samples has been determined using Atomic Absorption Analysis on FAAS - SOLAAR M5 spectrometer. For the preparation of standard solutions “Titrisol” standards produced by Merck (Germany) were used, the concentra-tion of reference metal content was 1000 ppm.

X-ray diffraction (XRD) analysis. The XRD patterns have been recorded using TUR M62 diffractometer with CoKα radiation. The observed patterns were cross-matched with those available in the JCPDS database. The particle size was determined by Scherrer’s formula. X-ray photoelectron spectroscopy (XPS).

The X-ray photoelectron spectroscopy (XPS) studies were performed in a VG Escalab II electron spectrom-eter using AlKα radiation with energy of 1486.6eV. The residual gas pressure in the analysis high vacuum chamber was 10-7 Pa.

Electron paramagnetic resonance (EPR). The elec-tron paramagnetic resonance (EPR) spectra have been registered as a first derivative of the absorption signal within the temperature interval of 100 - 400 K using an ERS 220/Q instrument.

The DTA and TG curves have been recorded on a LABSYSTM EVO apparatus SETARAM (France), at heating rates in air (10º/min) from 25ºC to 600ºC.

Аdsorption – texture analysis. The determination of the specific surface areas of the samples was carried out by nitrogen adsorption at the boiling temperature of liquid nitrogen (77.4 K) using a conventional volumetric apparatus. Before the measurement of the surface areas the samples were degassed at 423 K until the residual pressure became lower than 1.333.10-2 Pa. The nitro-gen (N2) adsorption-desorption isotherms were used to calculate the specific surface areas (ABET) using the BET equation.

Catalytic tests. The photocatalytic degree of dis-colouring of RB5 was determined using 150 ml of dye aqueous solution with 20 ppm initial concentration. The photocatalytic activity tests have been carried out using polychromatic UV-lamp (Sylvania BLB, 18 W), with wavelength range 315-400 nm. The light power density on the sample position was 0.66 mW.cm-2. The process of discolouring has been monitored by UV-Vis absorb-ance spectrophotometer BOECO S26 in the wavelength range from 200 to 800 nm. All photocatalytic activity tests have been carried out at a constant stirring rate (400


261

rpm) under ambient conditions. The samples reach ad-sorption-desorption equilibrium in the dark within about 30 min before switching on the illumination. To test the photocatalytic activity of ZnO powders, sample aliquots of the suspension have been taken out of the reaction vessel after regular time intervals. The powder was then separated from the aliquot solution by centrifugation before the UV–Vis spectrophotometric measurement of dye concentration. After that, the aliquot solution, together with the photocatalyst powder, were returned back into the reaction vessel. The degree of degradation of the dye was evaluated using the following equation:

100 Co CDecoloration xCo− =

[%] (1)

The degree of decoloration is expressed as C/Co (where Co and C are initial absorbance before switching on the illumination on and absorbance of the solution after 120 minutes of illumination, respectively at 599 nm corresponding to the peak of the diazo bond (-N=N-).


The X-ray diffraction analyses of the samples (Fig. 1) give evidence for the formation of wurtzite phase ZnO (JCPDS 36-1451). The copper dopant concentrations in samples are 0.24, 0.35 and 1.07 at. % (Table 1). Accord-ing to Fernandes et al. [7] the ideal maximal Cu2+ con-centration to obtain well-crystallized ZnO phase is less than 1 wt.%, while according to Fu et al. [4] the optimal doping concentration has been found to be 0.5 wt.%.

The lattice parameters of the undoped ZnO crystal are very close to those obtained for zincite structure (JCPDS 36-1451). The introduction of Cu into the ZnO samples leads to shrinking of the unit cell (Table 1). A similar results for Cu doped powders and films has been obtained by Belini and Bahsi [1, 11]. It is well known that in Cu doped ZnO, the Cu2+ ions (atomic radii 0,057nm) substitute the Zn2+ ions (atomic radii 0,060 nm) (see Table 1). The diffusion process during the sintering may lead to defect formation, in which Cu2+ ions substitute

Fig. 1. XRD spectra of: (a) ZnO; (b) ZC1; (c) ZC2; (d) ZC3.

Samples Cu concentrations at.%

(wt. %)

ABET [m2.g-1]

Lattice constants a,c (Å)

Crystallite sizes [nm]

a ± 0.002 c ± 0.002 Ref. JCPDS 36-1451 - - 3.244 5.205 -

ZnO - 22 3.24946 5.20689 45

ZC1 0.24 (0.18 )

23 3.25042 5.20158 46

ZC2 0.35 (0.28)

23 3.25010 5.20385 48

ZC3 1.07 (0.8)

23 3.25014 5.20363 49

Table 1. Specific surface areas, copper concentrations and crystallite sizes of the prepared samples.

30 32 34 36 38 40θ

(a)

(b)

(c)

(d)

Inten

sity(a

.u.)

1020 1035 1050

(a)

(b)

Zn2p

Binding Energy (eV)

520 530 540 550

(b)

O1s

Binding Energy (eV)

(a)

969 952 935 918

(b)

(a)

Binding Energy (eV)

Cu 2p

Fig. 2. XPS spectra of Zn 2p, O 1s and Cu 2p core levels for samples: ( a) ZC1; (b) ZC2.


262

Zn2+ cations in the wurtzite unit cell of ZnO and appear-ance of complex defects:

[Cuzn-Zni]x [11].

The compositions and the specific surface areas (ABET) of the pure and the Cu-doped zinc oxide samples are represented in Table 1. It shows that the specific sur-face areas of all the samples do not differ significantly. The calculated mean size of the crystallites, determined by XRD, was within the range of 45 ÷ 49 nm. Compared with the pure ZnO, Cu doping in ZnO samples results in a little increase in the ZnO crystallite size.

The surface composition and chemical state of the Cu doped ZnO powders have been investigated by XPS. Figure 2 shows the Zn 2p, O1s and Cu2p photoelectron spectra. The Zn 2p3/2 signals of ZnO for both samples are similar and have a maximum at 1021.7 eV, typical of ZnO. The O1s peaks are located at 530.3 eV – they are attributed to O2− ions in ZnO crystal lattice. Table 2 shows the surface chemical composition of the ZC1 and ZC2 samples. It can be seen in the table that zinc-copper ratios are much higher than that in the initial composition. This is an indication of segregation of the doping element on the surface of the ZnO particles. This result is in accordance with the studies of Bellini et al. [11]. Figure 3 represents the EPR spectrum of the Cu/ZnO (ZC1) catalyst sample. The analysis confirms the presence of isolated Cu2+ ions [12, 13].

A strong endothermal peak due to the zinc hydroxide precursor decomposition has been registered on the DTA curve (Fig. 4). The precursor decomposition begins at 230oC and it is completed at 270oC.

The degrees of discolouring of the dye solution in the cases of undoped and Cu doped ZnO catalysts after 120 minutes of irradiation are shown in Fig. 5. As it can be seen for catalyst charge amount of 0.08 g, the decrease of the Cu content leads to enhancement of the photocata-lytic activity. Similar behavior has been registered with 0.3 g catalyst amount. The ZnO samples (0.3 g) exhibit higher activities than those of the samples containing 0.08 g catalyst. Among all the investigated samples the pure ZnO samples showed the best photocatalytic properties (Figs. 5 and 6). Liu et al. [6] reported that the

Samples O1s, at. % Zn2p, at. % Cu2p, at. % ZC1 (0.24 at.% Cu) 53.59 45.33 1.08 ZC2 (0.35 at.% Cu) 63.64 35.24 1.12

Table 2. Surface compositions of the Cu/ZnO samples, determined by XPS analyses.

Fig. 3. EPR spectrum of sample ZC1.

1100 2200 3300 4400 5500

dP /

dB ,

a.u.

B , G

113 K

Fig. 4. DTA of zinc carbonate dried at 120◦C (a) and TG curve (b).

0 100 200 300 400 500

-0,5-0,4-0,3-0,2-0,10,00,10,20,30,4

Temperature [°C]

TG |s

c [m

g]

(a)

0 100 200 300 400 500

-4,5-4,0-3,5-3,0-2,5-2,0-1,5-1,0-0,5

Heat

Flo

w |s

c [µ

V]

Temperature [°C]

(b)


263

photocatalytic degree of degradation of Methyl Orange on pure ZnO has nearly the same value as that of CuO/ZnO, calcined at 450o C. Donkova et al. [9] have also found a decrease in the photocatalytic conversion degree on ZnO powder samples after doping them with Cu in

the reaction of Methylene Blue dye degradation because of increased charge carriers recombination rate.

It is also possible that the absorption capacity of the Cu-doped samples decreases upon increasing the Cu contents, which is in accordance with the results, obtained by Muthukumaran et al. [8].

CONCLUSIONS

Copper doped (0.24÷1.07 at.%) ZnO wurtzite nanoparticles, prepared by the precipitation method, have been tested in the reaction of Reactive Black 5 decoloration under UV irradiation. The average sizes of the crystallites of doped and undoped ZnO are below 49 nm. The XRD results proved that the increase in copper concentration causes a slight decrease in the degree of crystallinity. TG-DTA data for zinc hydroxide precursor registered strong endothermic peak at 250oC. The EPR spectra registered isolated Cu2+ ions. The XPS analysis showed that zinc-copper ratios are much higher than those in the experimentally evaluated compositions of the samples. This result is an indication that segregation of doping element is occurring on the surface of the ZnO particles. The decrease in the Cu content in ZnO samples leads to enhancement of their photocatalytic activities. The pure ZnO samples exhibit the best photocatalytic properties.

Acknowledgements

The authors are grateful to Projects (No. 02-066/17.12.2009) and DVU -02/36 (2010) financial support.

REFERENCES

1. Z. Bahsi, A. Oral, Effects of Mn and Cu Doping on the Microstructures and Optical Properties of Sol-gel Derived ZnO Thin Films, Opt. Mater., 29, 2007, 672–678.

2. K. Rekha, M. Nirmala, M. Nair, A. Anukaliani, Structural, optical, photocatalytic and antibacterial activity of zinc oxide and manganese doped zinc oxide nanoparticles, Physica B: Condensed Matter , 405, 2010, 3180–3185.

3. R. Ullah, J. Dutta, Photocatalytic degradation of or-ganic dyes with manganese-doped ZnO nanoparticles,

0 20 40 60 80 100 1200,0

0,2

0,4

0,6

0,8

1,0

C/Co

Time[min]

ZnO-0.08g ZC1-0.08g ZC2-0.08g ZC3-0.08g ZnO-0.3g ZC3-0.3g

Fig. 5. Degree of decoloration of the RB5 dye (concentra-tion 20 ppm), based on changes in the intensity of the peak, corresponding to azo bond (-N=N-) with the course of time.

0

20

40

60

80

100

ZC3ZC2ZC1

ZnO

a)0.08g

Deco

lora

tion

[%]

Catalysts

0

20

40

60

80

100b)

0.03gZC3

ZC2ZC1

ZnO

Deco

lora

tion

[%]

Catalysts

Fig. 6. Degree of decoloration of the dye after 120 minutes of illumination using ZnO, ZC1; ZC2; ZC3; and UV light photocatalyst content (a) 0.08 g and (b) 0.3 g.


264

J.Hazard. Mater., 156, 2008, 194–200.4. M. Fu, Y. Li, Siwei wu, P. Lu, J. Liu, F. Dong, Sol–gel

preparation and enhanced photocatalytic performance of Cu-doped ZnO nanoparticles, Appl. Surf. Sci., 258, 4, 2011, 1587-1591.

5. C. Xu, L. Cao, G. Su, W. Liu, H. Liu, Y. Yu, X. Qu, Preparation of ZnO/Cu2O compound photocatalyst and application in treating organic dyes, J.Hazard. Mater., 176, 2010, 807–813

6. Z. Liu, J. Deng, J. Deng, F. Li, Fabrication and pho-tocatalysis of CuO/ZnO nano-composites via a new method, Mater. Sci. Eng. B, 150, 2008, 99–104.

7. D. Fernandes, R. Silva, A. Winkler Hechenleitner, E. Radovanovic , M. Custуdio Melo, E. Pineda, Synthesis and characterization of ZnO, CuO and a mixed Zn and Cu oxide, Mater.Chem. Phys., 115, 2009, 110–115.

8. S. Muthukumaran, R. Gopalakrishnan, Structural, FTIR and photoluminescence studies of Cu doped ZnO nanopowders by co-precipitation method, Opt.

Mater., 34, 2012, 1946–1953.9. B. Donkova, D. Dimitrov, M. Kostadinov, E. Mitkova,

D. Mehandjiev, Catalytic and photocatalytic activity of lightly doped catalysts M:ZnO (M = Cu, Mn), Mater.Chem. Phys., 123, 2010, 563–568.

10. S. Oh, M. Kang, C. Cho, M. Lee, Detection of car-cinogenic amines from dye stuffs or dyed substrates, Dyes Pigments, 33, 1997, 119-135.

11. J. Bellini, M. Morelli, R. Kiminami, Electrical properties of polycrystalline ZnO + copper (II) acetate powders, J. Mater. Sci. Mater. Electron., 8, 2002, 485-489.

12. D. Bogomolova, A. Jachkin, A. Krasil‘nikova, L. Bogdanov, B. Fedorushkova, D. Khalilev, EPR of transition metals in fluoroaluminate glasses, J Non-Crystall Solids, 125, 1-2, 1990, 32-39.

13. R. Elilarassi, G. Chandrasekaran, Structural, optical and magnetic characterization of Cu-doped ZnO nano-particles synthesized using solid state reaction method, J. Mater. Sci. Mater. Electron., 21, 11, 2010, 1168–1173.

Menwer Attarakih, Mazen Abu-Khader, Tahani Saieq, Hans-Jörg Bart

265


ETHANE PRODUCTION PLANT FOR BETTER ENERGY INTEGRATION AND COST REDUCTION IN JORDAN

Menwer Attarakih1,4, Mazen Abu-Khader2, Tahani Saieq3, Hans-Jörg Bart4,5

1 The university of Jordan, Faculty of Eng. & Tech. Chem. Eng. Dept., 11942 Amman, Jordan2Al-balqa applied university, Faculty of Eng. Tech. Chem. Eng. Dept., POB 15008, 11134 Amman, Jordan3Al-balqa applied university, Faculty of Eng. Tech. Dept. of Humanities, POB 15008, 11134 Amman, Jordan4 Chair of separation science and technology, TU Kaiserslautern5Centre of mathematical and computational modelling, TU Kaiserslautern, Germany E-mail:[email protected], [email protected]

ABSTRACT

The recovery of ethane from natural gas is considered to be more economically feasible due to shift in ethane prices and advances in process modifications. Process design and retrofitting of ethane recovery technologies are reviewed. A viable process design for the production of ethane using Qatari natural gas as feedstock in Jordan is proposed. The plant capacity is determined to be 8.83*109 scf per year at purity of 98.6 percent. Process design and energy integration are performed using CHEMCAD and HENSAD simulation software. The optimal process flow diagram (in terms of energy integration) consists of three major sections: Pretreatment section, liquefaction and cryogenic separation of methane and ethane. By using parametric studies, the major operating and design variables are optimized. The simulation results show that the ethane recovery plant proved be able to produce ethane and methane as a byproduct at a purity up to 98.6 and 99.7, respectively. Moreover, preliminary economic evaluation (profitability analysis) based on discounted and not discounted methods is performed. It results in an average ROROI of 16 percent value, and a nondiscounted payback period of 4.5 years.

Keywords: ethane recovery, natural gas, process synthesis, process integration, computer-aided flowsheeting, economic evaluation.

Received 11 March 2012Accepted 20 April 2013

INTRODUCTION

Jordan has very limited energy resources of fossil fuels. Iraq supplies Jordan with 10,000 barrels of crude oil a day, but still not enough. Jordan’s government forecasts that domestic fuel consumption will increase to 10 million metric tons per year of oil equivalent by 2020, up one third from today’s 7.5 million tons. The

government expects electricity consumption to almost double within 10 years to 5 200 megawatts. Natural gas is considered another viable energy option and there are two available sources; (a) Al-Risha well near the Jordan-Iraq common border, with ethane content of only 1 %, so the low capacity of this well, results in excluding this source, (b) the Egyptian natural gas, which contains 4% ethane and provides Jordan with most of the energy


266

it needs for the last couple of years. The Arab Gas Pipeline is a natural gas pipeline in the Middle East and delivers Egyptian natural gas to Jordan, Syria, and Lebanon. The pipeline has a total length of 1200 km. But due to the present Arab Spring which started in December 2010 and the Egyptian revolution, this natural gas pipeline in the Sinai Peninsula was bombed more than fifteen times which affected the stability of natural gas supplies to Jordan. These actions had increased Jordan‘s energy bill which hit the economy badly. Therefore serious thinking of replacing this option to a more stable energy source and explore other long-term alternatives is an obvious need. The Qatari natural gas seems to be the best alternative for Jordan as a profitable project since it is rich in ethane with a content of about 13 %, more over the negotiations and studies to import the gas from Qatar are being seriously taken into account. The high ethane percentage in the natural gas is a crucial factor in any plant feasibility analysis. Even with an acceptable margin of profit, the production of Ethane in Jordan will have direct impact on the country’s economy through ensuring stable energy policies for the industry and on both the strategic and political levels where the plant will contribute towards minimizing the high rate of unemployment.

Natural gas releases a high amount of energy upon combustion with few emissions. Therefore, natural gas is an indispensable source of energy for most industries, and electricity generation, and has played a major role in the industrial revolution and in raising the life standards. The composition of natural gas varies depending on the geology of the ground, the location of the well and its depth. The Qatari gas has the composition shown in Table 1.

Recently software simulators have played an important role in synthesis and design of natural gas plants. Also, they are used in technically assessing various overall design scenarios and optimizing operating conditions to reduce the overall cost [2-5]. Konukman and Akkaya [6], minimized the exergy destruction for a turboexpander plant by integrating Aspen Plus process simulator and the optimization-related features of MATLAB. Whereas, Tirandazi et al., [7], simulated an ethane recovery unit with its refrigeration cycle and the exergetic efficiency of the refrigeration cycle is determined to be 43.45 %, indicating a great potential for improvements. Most of the ongoing research is concentrated on efficient utilization of Liquefied Natural Gas (LNG) in a cryogenic process and to separate light hydrocarbons (C2+) from LNG with low power consumption [8]. The capital cost and potential revenue from LNG extraction can have a significant impact on the overall economics of an LNG project. An economy based model uses several online parameters from upstream and downstream plants to minimize C2+ loses and recover C3+ NGL Plants are described [9].

The main aims of this work are to review the available ethane extraction technologies and address various process design scenarios for an ethane production plant from natural gas. An optimized design would be selected which will be located in the Hashemite Kingdom of Jordan. The selected process design should satisfy the targets of maximum heat integration, low capital cost with the country’s needs and suitability. This is carried out through process synthesis and analysis, computer-aided flow sheeting, parametric studies of the basic flowsheet and feasibility studies. The use of simulation

Component Formula Composition (%) Methane CH4 76.6 Ethane C2H6 12.59 Propane C3H8 2.38 i-Butane C4H10 0.11 n-Butane C4H10 0.21 Pentane C5H12 0.02 Nitrogen N2 0.24

Hydrogen Sulfide H2S 1.02 Water H2O 6.83

Table 1. Qatari Natural Gas Compositions [1].


267

software to assess the design of the ethane recovery process is highlighted. The Qatari gas, which is rich in ethane 13 % is chosen as a feed stock for the proposed plant capacity of 8.83 x109 scf per year ethane with purity about 98.6 %.

ETHANE EXTRACTION TECHNOLOGIES

Ethane was typically not separated from the methane component of natural gas, but simply burnt along with the methane as a fuel. With technology advancement [10], a widespread industrial production of ethane, which is used as a chemical feed stock to produce ethy-lene, created a high and continuous demand on ethane as a raw material.

The common ethane recovery technologies are summarized as follows: (a) Steam cracking is one of the well-established technologies to produce ethane from naphtha and heavy fuel. It consists of a pyrolysis furnace, followed by fractionation and compression stages and finally - the product recovery stage [11]. (b) The adsorption process offers another available alternative. In this process, ethane is extracted from a natural gas mixture at 25oC and 101.3 kPa using a packed bed adsorber. Ethane is completely separated from the mixture, however, after a prolonged period. The adsorbent must be large-pored and much larger than the kinetic diameters of ethane (4.44 Å). The advantage of using adsorption to recover ethane from natural gas is that the separation conditions are likely to be not extreme [12]. The main disadvantage is the low recovery of ethane (50 %) if carbon is used as the adsorbent. (c) Membrane technology is applied to separation of ethane on the basis of the difference in its rate of permeation through the membrane. The recovered ethane (the permeate) is never 100 % pure because of finite partial pressure difference. So ethane can be further separated by multiple stages. The membrane permeation has many advantages, which include low capital investment, ease of operation, good weight and space efficiency, no moving parts flexibility, low environmental impact, ease of installation and reliability. The principle disadvantage is that a clean feed is required, leading to the use of filtration to remove particles down to one micron in size [13]. (d) Cryogenic processes generally involve cooling a natural gas stream to temperatures near -85oC. These low temperature requirements have

high associated energy consumption costs [14]. Ethane is most efficiently separated from methane by liquefying at cryogenic temperatures. The most economical process presently in wide use employs turbo expansion which can recover over 90 % of the ethane in natural gas [15-17]. There are certain guidelines that should be taken into consideration when choosing between cryogenics or absorption gas processing systems [18]. A pilot plant technology based on the adiabatic cooling of swirling gas flow in a supersonic nozzle which is effective in separating and processing natural gas components was developed [19]. To be competitive with liquid based crackers, ethane must be extracted at the lowest marginal cost. Barthe, and Gahier [20] presented a discussion, which covers the new processes that are being applied in different gas and LNG projects.

Recently, retrofitting an existing NLG plant to increase ethane recovery is considered [21-25]. Retrofitting typically requires modifications to the process equipment and expansion of the existing refrigeration system [26]. The operating conditions and the effect of column pressures, minimum approach temperature in the LNG exchanger, and split fraction have direct impact on maximizing ethane yield [27].

SYNTHESIS OF ETHANE RECOVERY PROCESS

Process synthesis and design can be categorized into two groups: Optimization based and knowledge-based methods. In optimization based methods an explicit or implicit flowsheet superstructure is required. On the other hand, knowledge-based (heuristic) methods concentrate on the representation and knowledge organization of the design problem. In the synthesis of ethane recovery process, we used an heuristic-based approach rather than rigorous structural optimization. Therefore, we followed the same lines of the Onion model [28,29]. In ethane recovery from natural gas, the synthesis of the basic flowsheet consists of three major stages: (a) The pretreatment “purification” step in which the gas pretreatment involves the removal of hydrogen sulfide (H2S), carbon dioxide (CO2), water, nitrogen and some other possible impurities to sufficiently low levels to meet the specifications and permit additional processing in the plant without any problems, (b) The liquefaction step-natural gas liquefaction is required whenever cryogenic separation is applied. Separation


268

of hydrocarbons generally occurs by the creation of a second phase. Since hydrocarbons (in natural gas) are gases at normal conditions and their boiling points are sufficiently low, a second phase (liquid phase) is created by liquefying part of the gas. Cryogenic separation relies on the fact that heavier components tend to be in the liquid phase more than lighter ones and so liquefy more rapidly, and finally (c) the separation step-the separation can be achieved by different techniques, such as lean oil absorption, and cryogenic fractionation. In natural gas industry, lean oil absorption is not widely employed, while cryogenic fractionation is the most common technique because of its high separation efficiency especially when high recovery rates are required, as in the present case [14]. Cryogenic fractionation simply employs a series of distillation columns that operate at very low temperatures to achieve the desired separation. The synthesis of the required process is based on the nature of the feed stock, which is the Qatari natural gas that contains the highest ethane content among other natural gases.

As mentioned above, a basic process flow diagram is developed using the onion model procedure. At

each layer synthesis decisions may require analysis on the individual unit level and/or the flowsheet as a whole. These analyses are performed using a process simulation tool. The simulation software “CHEMCAD” was used in the synthesis and analysis of the ethane recovery process. It helped in optimizing the flowsheet configuration, by eliminating infeasible options. The “HENSAD” Software [28] was used to improve the flowsheet economics through implementing heat exchangers network synthesis, analysis and design and heat integration based on pinch analysis.

Computer-Aided Flowsheeting and Sensitivity AnalysisSensitivity analysis is used to estimate optimum

process conditions, which result in optimum design in terms of desired product yield, and by minimizing utilities (steam & cooling water) consumption. The sensitivity analysis is performed by changing critical process parameters (input conditions), which have major impact on process output variables. With the use of the computer-aided flow sheeting CHEMCAD package, the possibility to test different process configurations, different thermodynamics packages and equipment

Stream No.

Mass F low

(kg/s)

Cp (kJ/kgK)

Temp In (K)

Temp Out (K)

Stream Enthalpy

(kW)

Film Transfer

Coefficient (W/m2/K)

22 19.75 4.0 473.0 313.1 12628 850.0 24 48.3 2.60 321.0 297.0 3013 30.0 27 48.3 2.75 297.0 244.1 7019 30.0

31 48.3 4.54 248.1 222.1 5701 30.0 48.3 4.0 227.0 217.0 1932 850.0

34 34.25 2.49 349.0 340.0 767.5 30.0 Available Cumulative Hot Stream Energy = 31062.7 kW

Table 2. Hot streams extracted from the ethane recovery process (base case).

Stream No. Mass

Flow (kg/s)

Cp (kJ/kgK)

Temp In (K)

Temp Out (K)

Stream Enthalpy

(kW)

Film Transfer Coefficient (W/m2/K)

45 34.24 4.8 151.0 161.0 -1643 850.0 34.24 4.8 161.0 303.0 -21694 30.0

42 33.27 3.0 207.0 227.0 -1996 1140 33.27 3.0 227.0 280.0 -5289 30.0

54 4.94 3.17 312.1 322.1 -156.5 1140 Available Cumulative Cold Stream Energy = -30780.7 kW

Table 3. Cold streams extracted from the ethane recovery process (base case).


269

models can be easily performed. In the case of recovery of ethane from natural gas, the final ethane purity, which is produced from the Deethanizer, is optimized. The amine package is the default thermodynamics model, which is used in the simulation of the recovery of ethane process. This is because the present process contains a gas sweetening unit in which diethanolamine (DEA) is used to absorb H2S gas. As an alternative to the amine thermodynamics package, the UNIFAC package and the SRK equation of state for the estimation of the k-values were employed for the enthalpy model. The SRK is known to predict accurately enthalpy changes and can quantitatively describe the phase equilibrium especially for light hydrocarbons [28].

The flowsheet is simulated following the process design hierarchy, where the recycle loops are first cut and initialized to recover the sequential modular approach. Once all the recycle calculations converged, sensitivity analyses were carried out on the flowsheet level. The results from the integrated process flowsheet are shown in Fig. 1. The energy integrated flowsheet was generated following the heat exchanger network design and analysis using the HENSAD software [28].

Firstly, the De-ethanizer is analyzed by investigating the effects of feed pressure and temperature on the ethane recovery. The first case is studied by varying the feed temperature of the Deethanizer at a given feed pressure. It is clear that as the feed temperature increases, the ethane purity increases too. At very low cryogenic temperatures, the purity of ethane is not affected at feed pressures in the range of 10 to 14 bar. On the other hand, at high feed pressures (18, 22, 26, 30 bars), the purity of ethane is found more sensitive to temperature changes. The purity of ethane in the overhead product increases by decreasing the feed pressure at constant temperature. On the other hand, the ethane purity decreases when the temperature is decreased at constant pressure. This can be attributed to the increase of the vapor fraction of the feed, where at some pressure and temperature, the feed composition is equal to that at the feed tray. Therefore, at moderate feed pressure and temperature an optimum ethane purity can be obtained. Accordingly, the best operating conditions used in the final simulation flowsheet, shown in Fig. 1, are 14 bar and −17oC . The temperature −17 oC is chosen for further heating the process stream in E-105 that is imposed by the demethanizer bottom stream

temperature, which is −30oC. On the other hand, the pressure of the feed is found to affect the condenser duty, where the latter has a strong impact on the fixed capital cost of the plant (which increases the area of the condenser), and utility costs of the process. This is because the condenser needs a refrigeration cycle which is known to be a very expensive utility [28]. It is found that the minimum condenser duty is achieved at 14 bar. This results in a minimum fixed capital and operating costs, when compared to other operating pressures. The use of the amine fluid package has a substantial impact on the simulated results. This behavior is reported in many simulation case studies, where using different fluid packages can lead to different results [28].Synthesis of the Heat Exchanger Network

The objective of the heat integration is to avoid the expensive refrigeration cycles that are required in the feed preparation of the separation section (liquefaction section), and the bottom stream of the DEA regeneration distillation column. Moreover, heat integration can help in the elimination of hot utility (steam) by replacing the usage of the low pressure steam in the deethanizer reboiler. This is done through the pinch analysis, which is an integral part of the overall strategy for process development, design and optimization (process synthesis) [29-32].

The heat integration and analysis were carried out using "HENSED" software [28]. The heat integration of the ethane production process is done systematically. Firstly, the hot and cold streams in the process are identified. Table 2 shows the hot streams in the process along with the nesecary information that should be provided to the HENSAD software. Table 3 shows the cold streams in the process that is used to eliminate the cooling utilities (refrigerants, cooling water). A phase change occurs in streams (45 and 42 of Fig. 1), where a (10oC) of sensible heat to take into account the latent heat of the phase change in the streams, is used [28].

As an initial guess, the minimum approach temperature can be taken as 10oC for fluids, and 5oC for refrigerants [28]. In the present case, we selected ∆Tmin = 10oC, as a compromise between fixed capital investment and utility costs [28]. The composite curve for the ethane recovery process is given in Fig. 2. It shows that the integrated ethane extraction process (Fig. 1) is a threshold problem from the hot utility side. At the design level, it was necessary to screen and eliminate some streams from


270

Fig.

1.

Ener

gy in

tegr

ated

pro

cess

flow

dia

gram

for t

he e

than

e re

cove

ry p

roce

ss.

CW

CWR

hps

V-10

1

V-10

2

T-10

1T-

102

T-10

3A/

B

T-10

4T-

105

V-10

4

V-10

6

V-10

3 V-

105

V-

107

V-10

8

E-10

1

E-10

2E-

103

E-10

4E-

105

E-10

6

E-10

7

E-10

8E-

109

1 7

2

5

6

7

9

10

11

13

14

16

17

19

20

24

26

28

31

32

33

35 4

37

38

22

42

43

44

48

P-10

4 A

/B

P-10

5 A/B

P-10

7 A/B

P-10

9 A

/B

V-10

1Fl

ash

drum

T-

101

H2S

Abs

orbe

r

V-10

2 Fl

ash

drum

T-

102

DEA

distill

ation

co

lumn

E-10

1

Cond

ense

r V-

103

Reflu

x dru

m P-

104 A

/BRe

flux

pum

p

C-10

1

C-10

2

Te-1

01

P-11

0

A/B

E-10

2 Re

boile

rP-

105 A

/BRe

cycle

d DE

A pu

mp

T-10

3A/B

Ad

sorp

tion

towe

r

T-10

4De

met

haniz

erTe

-101

Tu

rbo

expa

nder

V-10

4 A/B

Flas

h dr

um a

t hig

h pr

essu

re

C-10

2 co

mpr

esso

rC-

101

com

pres

sor

E-10

3Co

oler

E-10

5Co

oler

E-10

4 Co

oler

E-10

6Co

oler

E-10

8

Cond

ense

r V-

105

Reflu

x dru

m

P-10

8 A/B

Rebo

iler p

ump

V-10

6Fl

ash

drum

15

21

34

40

27

41

29

50

49

51

P-10

1 A/B

Recy

cle

pum

p

P-10

3 A/B

Cent

rifug

al pu

mp

P-10

7 A/B

Reflu

xl pu

mp

P-10

6 A/B

Dem

etha

nizer

l pu

mp

P-10

9 A/B

Deet

haniz

er

feed

pum

p

52

25

36 4

30

46

47

53

54

55

56

E-11

0

D-10

1

D-10

1Cy

clone

P-10

1A/B

P-10

3 A/

B

39

P-10

8A/

B

P-10

6 A/B

45

P-11

1 A

/B

23

18

8

3

P-35

6

4

E-11

0Co

oler

P-11

0 A/B

Deet

haniz

er

pum

p

E-10

9

Cond

ense

r V-

107

Reflu

x dru

m

P-11

1 A/B

Rebo

iler p

ump

V-10

8Fl

ash

drum

T-

105

Deet

haniz

er


271

the pinch analysis of the process. Some of the major streams that are eliminated from the pinch analysis are:

Stream (19) of the integrated PFD due to its low flow rate, and its high content of H2S, which is considered to be a highly corrosive material.

Stream (23) is also eliminated from the pinch analysis, where the stream temperature has to be reduced from 106 to 48oC. In order to achieve this temperature reduction, this stream needs about 7376 kW. The optimum way to do this is to use a high flow rate of cooling water with inlet temperature of 30oC and return temperature of 40oC, rather than cooling this stream with the available cold streams. This is because these streams are used for cooling other process streams, where cooling water is not possible.

The synthesis of the Heat Exchanger Network (HEN) based on the temperature interval diagram is performed using the HENSAD software. The results on Fig. 3 shows that the heat integrated process has only five heat exchangers (E-104, E-105, E-106, E-107, E-110) using a minimum approach temperature of 10°C.

The minimum approach temperature has a direct effect on the operating and capital cost of the heat exchanger network. An optimal minimum approach temperature is expected to minimize the total annualized cost. This is represented by the total Equivalent Annualized Operating Cost (EAOC) as function of the minimum approach temperature, which is shown in Fig. 4. It is clear that the EAOC curve as function of the minimum approach temperature is flat around the optimal value, which is around 25°C. At these optimal conditions, the operating cost of the heat exchanger network is found equal to 10.86 $ million/year, and the total operating cost of the process is 42.14 $ million/year (see the section for cost estimation). The heat integration achieved about 26 % savings with respect to the total operating cost of the plant, which is a very important step in the phase of process synthesis and design.

The energy recovery is considered a crucial issue, which is achieved through heat integration, because of the rapid increase in the fuel costs, and the energy crisis. The heat integration results in a reduction of energy requirements by about 319.4 MW or 967980 GJ/year. This high energy comes from burning of fuels to generate steam for heating, and in the refrigeration cycles that consumes a huge amount of electricity, generated in gas turbines to drive the compressors of

Fig. 2. The composite temperature-enthalpy diagram of the heat exchanger network for the ethane recovery process using HENSAD software [28].

Fig. 3. The optimized heat exchangers network for ethane recovery process shown in Fig. 1 using HENSAD software [28].

Fig. 4. EAOC for the heat exchanger network as function of the minimum approach temperature for the ethane recovery process using HENSAD software [28].


272

the refrigeration cycles. On the other hand, the heat integration made the process more complex, since it created a more interactions between the process streams and major pieces of equipment. This complicates the control strategy and process start up and shutdown [29]. Process Description

Fig. 1 shows the energy integrated PFD for the ethane production process. The process shows that the sour natural gas in stream <1> is fed to the packed absorption tower (T-101), which contains Pall Rings as packing. The gas enters from the bottom at 30.7 bar and 35oC. Hydrogen sulfide (H2S) is absorbed using a 30 mass % (6.8 mol %) aqueous solution of Di-Ethanol Amine (DEA). The solvent, stream <2>, flows at of 3,257 kmol/h. The solvent passes through the flash drum (V-102) where hydrocarbon traces are recovered in stream <13> and mixed with the methane product in stream <26>. The solvent stream <15> is regenerated in the stripping distillation column (T-102). The regenerated solvent becomes concentrated due to water losses, and hence a makeup water in stream <45> is used.

The sweet natural gas in stream <10> is dehydrated in the adsorption towers (T-103 A/B), which contain molecular sieves of average pore size equal to 4Å. After that, the dehydrated sweet gas in stream <21> enters the separation feed preparation stage where it is compressed, using the compressor (C-101) to 65 bar. This stream is then cooled in a heat exchangers network consisting of four heat exchangers in series (E-103, E-104, E-105, and E-106) to -51oC. The cooled gas in stream <35> enters the low-temperature separator (V-104), where it is split into two streams: the overhead vapor stream <36> and the liquid (condensed gas) stream <37>. The vapor stream is expanded in a turbo-expander (D-101) where its pressure is reduced to 14 bar. The liquid stream pressure is also reduced to 14 bars using a throttling valve. The two cold streams, <38> and <40>, are then sent to the Demethanizer (T-104) with two feeds, introduced at trays 4 and 15, respectively. Liquid methane is recovered as an overhead product in stream <45> and used as refrigerant for cooling the process gas in (E-106). This stream is then used to cool the regenerated solvent in (E-107), and the process gas in (E-104), where it is mixed with the hydrocarbon recovered in (V-102) and compressed to 30 bar using the compressor (C-102). This stream is then cooled to 76oC using the heat exchanger (E-110), before it is sent

to the final storage. The remaining hydrocarbons (C2+), which leave the bottom of the Demethanizer in stream <42> are partially reboiled in (E-105) and recycled back for flashing in V-106. The vaporized portion returns to the Demethanizer at the last bottom tray, and the liquid portion is pumped to the Deethanizer (T-105) in stream <47>. Ethane (the major product) is recovered in the Deethanizer overhead stream <52> at purity of 98.6 % and then sent storage tanks. The remaining hydrocarbons (C3+) leave the bottom of the Deethanizer in stream <49> and are partially reboiled in (E-110). This stream is then flashed in vessel (V-108).

ECONOMIC & PROFITABILITY ANALYSESIn process synthesis and design, economic analysis

is essential to decide whether the project is feasible or not. The feasibility study is carried out using the preliminary estimation method that has an uncertainty of ± 20 % [28].

Economic EvaluationThe total Fixed Capital Investment (FCI) estimate

is based on the bare module cost algorithm, presented in Turton et al. [28]. The estimated capital investment is found to be $ 125 942 528, which includes all direct costs, indirect costs, contingency and fee, and auxiliary facilities. The Working Capital is estimated as 15 % of the FCI, and is found to be $ 18 891 379. Therefore, the total Capital Investment for the ethane recovery process is $ 144 833 907. The costs of manufacturing, or the operating costs, are estimated from the following cost items [30]:

l Fixed Capital Investment (FCI = $ 125 942 528).l Cost of Operating Labor (COL)The number of operators required per shift depends

Fig. 5. Distribution of utilities costs for the ethane recovery process.


273

on the major equipment in the plant. Then the number of hired operators can be taken as 4.5 times the required number to cover all shifts [28]. Based on the method of Turton et al. [28], the average payment for each employee is estimated to be 46 800 $/year in 1996 and 71 800 $/year in 2010. The number of operators required in the ethane recovery plant per shift is 3.55 operators. therefore, 16 operators are to be employed, with total annual operating labor cost equals to $ 1 148 500 per year.

Cost of Utilities (CUT)The utilities requirements for the integrated ethane

recovery process and their costs are based on a stream factor of 0.95, which amounts to annual costs of 30 199 053. The distribution of the utility costs is shown in Fig. 5. It is clear that the steam contribution is around 50 % of the total utilities costs, followed by refrigeration, which is about 32 %. On the other hand, the cooling water has the lowest share that is about 0.6 %. This means that the energy integration has reduced the annual utility cost by 28.4 %. Cost of Waste Treatment (CWT)

The only stream that may be considered to contain waste is the overhead stream coming from the stripper, which contains hydrogen sulfide and water in the vapor phase. This stream is assumed to go directly to the flaring system. Therefore, no costs are associated with treating any type of wastes in this plant.Cost of Raw Materials (CRM)

The main raw material for the ethane recovery process is the natural gas. Recent prices of natural gas were obtained through the Chemical Market Associates, Inc. (CMAI). The price of natural gas is 4.26 $/MMBtu. The amount of natural gas flowing into the plant is 418 325 lb/h. The lower heating value (LHV) of natural gas is 19 800 Btu/lb. And thus, the total cost of raw material (based on stream factor 0.95) is calculated based on

the LHV of natural gas, which is $ 293 640 747 per year. Based on these estimated values, the Cost Of Manufacturing (COM) is estimated using the following formula [28]:COM = 0.304FCI+2.73COL+1.23(CUT+CWT+CRM) (1)

In the above equation a depreciation of 10 % is assumed. On the other hand, the COM without depreciation (COMd) is:COMd = 0.180FCI + 2.73COL + 1.23(CUT + CWT + CRM ) (2)

Therefore, based on equations (1) and (2) , the total costs of manufacturing with capital depreciation is COM = 439 739 952 $/year and that without depreciation is COMd = 424 123 079 $/year. To find the Equivalent Annualized Operating Cost (EAOC), the FCI is annualized using the capital recovery factor (A/P, i, n):

(1 )EAOC = FCI COM(1 ) 1

n

n

i ii

++ + −

(3)

Assuming an interest rate of 8 % and a project life of 10 years, the EAOC is found to be 458 509 103 $/year.

The distribution of the EAOC is shown in Fig. 6, where it is clear that the raw materials contribute to about 79 % of the total EAOC. This value is an expected result,

Table 4. Product Prices in Ethane Recovery Process.

Fig. 6. Distribution of Equivalent Annualized Operating Cost (EAOC) for the ethane recovery process.

Product Production rate Price

Annual revenue $/yr

Ethane 678 656 gal/day 0.8 $/gallon 188 259 285

Methane 271 821 lb/h 4.26 $/MMBtu 207 541 774

Propane (87 %) 153 730 gal/day 1.4745 $/gallon 78 599 721


274

since the cost of raw materials has a major impact on the total production cost, which ranges from 30 to 85 % [28]. The plant is expected to make revenue from selling three products; Ethane, Methane, and C3+ (containing 87 % propane). Recent prices for these three products were obtained through CMAI. The revenues are shown in Table 4.

Note that the methane revenue is calculated based on the LHV of 21 537 Btu/lb of methane. On the other hand, ethane and propane revenues are calculated based on the price per volume ($/gallon). Accordingly, the total revenue is equal to $ 474 400 780 per year.Profitability Analysis

Profitability analysis is performed for the ethane recovery process based on three criteria: time, cash, and interest rate. For each of these bases, discounted and non-discounted techniques are used. In order to develop a cash flow diagram, the following assumptions are made [30]:

l Straight line method of depreciation over the project life (10 years).

l Since industrial plants are not constructed to be sold at the end of the project life, a zero salvage value is assumed (only land price is retrieved).

l Income taxes: 10 %.l A typical interest rate of 8 % is assumed.l Land costs 1.5 % of the total capital investment.Based on the above assumptions, a discrete after tax

cash flow diagram is shown in Fig. 7. A period of two years for constructions and a project life of ten years are assumed. Fig. 7 shows that the FCI is spent evenly during the construction period (-2 to 0). The plant start-up is

assumed after year (0) and the net cash flow is reduced in the first year due to the deduction of the working capital. A constant net cash flow of $ 32 454 MM/yr from years (2 to 9) is achieved before it jumps in the last year when the working capital and the land price are retrieved. The non-discounted time criterion gives a good indication for the project’s profitability. The term used for this criterion is the payback period (PBP) and it is found to be 4.5 years. On the other hand, the discounted PBP can also be used in time profitability analysis. It is estimated by converting all non-discounted cash flows to the discounted form, then reproducing the cumulative cash flow diagram based on the new discounted values. The discounted PBP for the ethane recovery process is found equal to 6.0 years after start-up. The Net Present Value (NPV) for the ethane recovery plant is calculated with the aid of the NPV built-in function in Microsoft Excel at (i = 8 %) and found to be $ 60 738 824. Whereas, Cumulative Cash Position (CCP), is simply the cumulative non-discounted cash flow at the end of the project life. For this project, the CCP is equal to $198 594 000. The Discounted Cash Flow Rate of Return (DCFROR) represents the highest after tax interest rate, at which the project can just break even. This rate is also called the Internal Rate of Return (IRR), and when computed, is compared with the Minimum Attractive Rate of Return (MARR). IRR is calculated by trial and error after setting the NPV equal to zero. The IRR for the ethane recovery project is calculated also using the IRR built-in function in Microsoft Excel and found to be 17 %. When it is compared to the common MARR of 10-12 % for chemical industries, it is considered very satisfying. An industrial project with such an IRR value is considered to be a low

Fig. 7. Discrete cash flow diagram for the ethane recovery process.


275

risk investment [29]. Also, the Rate of Return on Investment (ROROI) is used for non-discounted interest rate profitability analysis. It represents the non-discounted rate at which money is made from the fixed capital investment. The ROROI is calculated using equation (4) and found to be equal to 16 %.

average annual net profit 1RORI = FCI n

− (4)

The major sources of uncertainty, which will have an effect on the cost economics and savings, come directly from the fluctuating cost price of the raw material and indirectly from the equipment design equations. The uncertainty may reach up to 15 % error [28, 29].

PLANT LOCATION

The plant site should be ideally located where the cost of production and distribution can be at a minimum level, with a good scope for plant expansion, suitable environment and safe living conditions for easy plant operation. Therefore, the Aqaba Special Economic Zone (ASEZ) in the southern part of Jordan is selected. It is directly adjacent to the site where the new Aqaba seaport will be built over the coming five to seven years [31]. The South Industrial Zone (SIZ) is a principal investment opportunity within ASEZ.

The natural gas and ethane storage tanks are located outside the battery limits of the plant. A thirty day capacity for the raw materials and products storage tanks is recommended [28]. The raw material flowrate is 173 983 m3/day, and hence the volume of natural gas needed is 5 219 500 m3. This volume should be divided into 131 floating head tanks with each 40 000 m3. This is because the maximum volume of storage tanks available from vendors is 40 000 m3 with a diameter of 42 m. The main product which is ethane has a flowrate of 2 060 m3/day, and the storage capacity needed is 61 870 m3. This volume can be divided into 2 huge cylindrical tanks of 20 623 m3 volume, 5 m length and 17 m diameter.

CONCLUSIONS

Natural Qatari gas, with 13 percent ethane content is considered a feasible feed stock for the production of ethane in Jordan. In spite of cryogenic conditions, the proposed process design is found to be feasible with an

efficient process at high recovery rates. As expected, the cryogenic liquefaction section consumes most of the energy required in the ethane recovery process. The feed preparation through dehydration of natural gas is a critical step, and is found to affect the performance of down processing steps, due to freezing of water under sever cryogenic conditions prevailing in the separation section. The desulfurization of natural gas is crucial, since sulfur compounds are poisonous and corrosive by their nature, and high precautions should be taken into account while dealing with them. Accordingly, H2S is removed early in the process in the feed preparation section.

With the aid of computer-aided flowsheeting, the sensitivity analysis shows that an operating pressure of 14 bar for the demethanizer yields minimum losses of ethane in the overhead product. Pinch analysis is used to design the optimal heat exchangers network, and the use of liquid methane stream as a refrigerant, has led to significant annual money savings; however at the expense of flowsheet complexity structure. The Equivalent Annualized Operating Cost (EAOC) is found to be around 78 %. The profitability analysis shows that the ethane recovery process has an Internal Rate of Return (IRR) of 17 %. Thus, such an investment is considered to be a safe investment.

To further assess the feasibility of ethane recovery from natural gas, it is recommended to consider other natural gas feed stocks with different ethane contents such as the Jordanian or Egyptian natural gas.

REFERENCES

1. Qatar business and investment opportunity yearbook, International business publications USA, 2006.

2. A. Baldonedo, Maximizing NGL recovery by re-frigeration optimization, GPA Annual Convention Proceedings, 1999, 201-211.

3. B. Hua, Y. Q. Xiong, Y. J. Li, X. M. Yang, Simulation and optimization of the process of light hydrocarbon recovery from LNG, Nat. Gas Ind., 26, 5, 2006, 127-129.

4. R. Chebbi, A. S. Al-Qaydi, A. O. Al-Amery, N. S. Al-Zaabi, H. A. Al-Mansouri, Simulation study compares ethane recovery in turboexpander processes, Oil and Gas Journal, 102, 4, 2004, 64-67.

5. S. Diaz, E. A. Brignole, A. Bandoni, Flexibility study


276

on a dual mode natural gas plant in operation, Chemical Engineering Communications, 189, 5, 2002, 623-641.

6. A. E. Konukman, V. R. Akkaya, Minimisation of exergy destruction in an ethane recovery turboex-pander process using aspen plus process simulator, International Journal of Energy, 8, 4, 2011, 425-446.

7. B. Tirandazi, M. Mehrpooya, A. Vatani, S. Moosavian, Exergy analysis of C2 + recovery plants refrigeration cycles, Chemical Engineering Research and Design 89, 6, 2011, 676-689.

8. T. Gao, W. Lin, A. Gu, Improved processes of light hydrocarbon separation from LNG with its cryogenic energy utilized, Energy Conversion and Management, 52, 6, 2011, 2401-2404.

9. A. Al-Harbi, Ethane demand vs. C3+ NGL recov-ery economic model, GPA Annual Convention Proceedings, 1, 2008, 287-298.

10. M. Economides, Advanced Natural Gas Engineering, 1st Ed., Gulf Publishing Company, Houston, Texas, 2009.

11. K. Blok, M. Patel, T. Ren, Olefins from conven-tional and heavy feed stocks: Energy in steam crack-ing and alternative processes, Faculty of Chemistry, Utrecht University, 2006.

12. N. Magnowski, A. Avila, C. Lin, M. Shi, S. Kuznicki, Extraction of ethane from natural gas by adsorption on modified ETS-10. Chemical Engineering Science, 66, 2011, 2991-2996.

13. A. Kolh, R. Nielsen, Gas purification, Gulf Publishing Company, Houston, 1997.

14. W. Parrish, A. Kidnay, Fundamentals of natural gas processing, 2nd Ed., Taylor & Francis Group, Boca Raton, London, New York, 2006.

15. J. George, I. Montgomery, Ethane recovery system, US4851020 (patent),1989.

16. T. Joe, P. Lynch, How to compare cryogenic process design alternatives for a new project, 86th annual con-vention of the gas processors association, Texas, 2007.

17. W. Ng, K. Kolmetz, S. Lee, Energy optimization of cryogenic distillation, AIChE, Atlanta, Georgia, 2005.

18. Y. Mehra T. K. Gaskin, Guidelines offered for choosing cryogenics or absorption for gas process-ing, Oil and Gas Journal, 97, 9, 1999, 62-69.

19. V. Alfyorov, L. Bagirov, L. Dmitriev, V. Feygin, S. Imayev, J. Lacey, Supersonic nozzle efficiently

separates natural gas components, Oil and Gas Journal, 103, 20, 2005, 53-58.

20. L. Barthe, V. Gahier, Ethane recovery processes evolve to meet market needs, International Gas Union World Gas Conference Papers, 6, 2009, 4944-4954.

21. J. Lynch, R. N. Pitman, Texas plant retrofit improves throughput, C2 recovery, Oil and Gas Journal, 94, 23, 1996, 41-48.

22. D. Schulte, J. Mak, R. Nielsen, C. Graham, M. Woiemberghe, An optimized revamp of an NGL plant to enhance ethane recovery, GPA Annual Convention Proceedings, 1, 2006, 361-373.

23. S. Diaz, L. Urlic, A. Bandoni, E. Brignole, Retrofit of natural gas turboexpansion plants to work in dual mode, Latin American Applied Research, 31, 4, 2001, 325-330.

24. L. Bai, R. Chen, J. Yao, D. Elliot, Retrofit for NGL recovery performance using a novel stripping gas refrigeration scheme, GPA Annual Convention Proceedings, 1, 2006, 300-314.

25. H. Omori, LNG production by retrofitting NGL extraction plants, Proceedings, Annual Convention - Gas Processors Association, 2, 2007, 606-620.

26. S. Golikeri, J. Yu, et al., Retrofit of EnCana’s Fort Lupton Plant to increase capacity and ethane recov-ery. GPA Annual Convention Proceedings, 2011, 479-488.

27. S. Ghassemian, F. Farhadi, Optimization and en-ergy analysis of an ethane recovery plant, 8th World Congress of Chemical Engineering: Incorporating the 59th Canadian Chemical Engineering Conference and the 24th Interamerican Congress of Chemical Engineering, 2009.

28. R. Turton, R. Bailie, W. Whiting, J. Shaeiwitz, analysis synthesis and design of chemical process, 3rd Ed., Prentice Hall, New Jersey, 2009.

29. R. Smith, Chemical Process: Design and Integration, 2nd edition, John Wiley, 2005.

30. M. Peters, K. Timmerhaus, W. Ronald, Plant design and economics for chemical engineers, 5th Ed., McGraw-Hill International, New York, 2003.

31. Aqaba development corporation, Fertilizer chemi-cal cluster investment and development prospect, Jordan, 2006.

Zaid Ahmed Al-Anber

277


HEAT SAVING IN EVAPORATIVE CRYSTALLIZATION

BY INTRODUCING A HEAT PUMP

Zaid Ahmed Al-Anber

Department of Chemical Engineering, Faculty of Engineering Technology, Al Balqa’ Applied University, Amman, Jordan E-mail: [email protected]

ABSTRACT

In this work theoretical calculations had been made for suggested methods of heating mother liquor in a crystallizer for both kinds of evaporative crystallization process. These suggested designs aim to reduce and save the consumption of heat which leads to reducing operational costs for the two kinds of crystallization. It was found that in direct evaporative crystallization: when a heat pump was added to heat the medium (hot air), it had saved energy consumption range between 81 % to 93 % at different values of coefficient of performance (COP). In the second type - indirect evaporation, when a heat pump was added on line containing the mixture of vapor from crystallizer and steam from jacket outlet, this mixture becomes a heat source to the heat pump in order to preheat the inlet steam to the jacket. Calculations of this suggested design showed that the saved energy consumption was 8 % to 26 % at different COP.

Keywords: evaporative crystallization, energy saving, heat pumps.

Received 11 September 2012Accepted 20 April 2013

INTRODUCTION

Crystallization is a solid-liquid separation process in which mass transfer occurs of a solute from the liq-uid solution to a pure solid crystalline phase (process where solid particles are formed from a homogeneous phase) [1]. It is an important operation in the chemical industry as a method for purification and a method for providing crystalline materials in the desired size range. In an energy-conscious environment, crystallization can offer substantial saving as a method of separation when compared with distillation, though it must be recognized that it is more costly to effect cooling than providing heating [2]. In crystallization; equilibrium is attained when the solution or mother liquor is saturated [3]. Crystallizers can be conveniently classified in terms of the method used to obtain deposition of particles: by cooling a concentrated hot solution; by evaporating a solution; by adiabatic evaporation cooling [2-5].

The concerns of energy consumption and envi-

ronmental pollution urge researchers to work on the development of clean energy and the utilization of waste energy. Sorts of novel technologies were developed and the achievements were patented. Among these patents, the heat pump system is one important topic. It utilizes the transformation between potential energy and thermal power energy to realize the performance of heat pumping and refrigeration. Usually in such processes, two zones are present: a cooling zone, where heat must be removed from the process, and a heating zone, where heat must be added to the process.

The utilization of heat pumps in various chemical technological processes is considered one of the promis-ing methods for energy saving [6, 7]. To our knowledge, the utilization of a heat pump in a crystallizer has not been reported in details yet for the purpose of energy saving. This work presents a theoretical analysis of the technical feasibility and the potential use of a heat pump in the process of direct and indirect contact evaporative crystallization. Principle schemes of the process with


278

insulation of the heat pump and a throttling valve, are pro-posed and the corresponding calculations are performed.

THEORY

Yield and heat effects in a crystallization process In most cases, the process of crystallization is slow

and the final mother liquor is in contact with a suf-ficiently large crystal surface so that the concentration of the mother liquor is substantially that of a saturated solution at the final temperature in the process.

In such cases, it is normal to calculate the yield from the initial solution composition and the solubility of the material at the final temperature. If evaporative crystal-lization is involved, the solvent removed must be taken into account in determining the final yield.

The actual yield may be obtained from algebraic calculations or trial-and-error calculations when the heat effects in the process and any resultant evaporation are used to correct the initial assumptions on the calculated yield. The heat effects in a crystallization process can be computed by one of the following methods: A heat bal-ance can be made in which individual heat effects such as sensible heats, latent heats, and the heat of crystal-lization can be combined into an equation for total heat effects. The second method is an enthalpy balance. The method can be realised when the total enthalpy of all leaving streams minus the total enthalpy of all enter-ing streams is equal to the heat absorbed from external sources by the process.

In using the heat-balance method, it is necessary to make a corresponding mass balance, since the heat effects are related to the quantities of solids produced through the heat of crystallization. The advantage of the enthalpy-concentration-diagram method is that both heat and mass effects are taken into account simultaneously. This method has limited use because of the difficulty in obtaining enthalpy concentration data. This information has been published for only a few systems [3].

Direct and Indirect Contact Evaporative Crystal-lization

The two processes of direct and direct evaporation crystallization are used basically to separate mineral salts from their water solution, by partial evaporation using heating agents (hot air in the direct method and

steam in the indirect one) to reach over saturation state as shown in Figs. 1 and 2. When the crystals appear in the crystallizer, the formed slurry concentrated solution (mother liquor with crystals) is removed and fed to a separator to isolate the crystals [3]. The vapor out from the crystallizer goes to the atmosphere.

This work is concentrated on studying of the fea-sibility of possible reduction in the heat demand for the process of direct and indirect contact evaporative crys-tallization by introducing a heat pump in the process. For both processes, a model of sodium chloride water solution was used as a feed solution to the proposed process with the aim of extraction of salts minerals and

Fig. 1. Principle scheme of direct contact evaporative crystallization.

Fig. 2. Principle scheme of indirect contact evaporative crystallization process.

Zaid Ahmed Al-Anber

279

production of pure desalinated water.The introduction of a heat pump (closed cycle) is

shown in Fig. 3. The initial air with a temperature (To = 25oC) is heated to a desired temperature (T1) by directing the air through the condenser of the heat pump which is the heater of the crystallizer in our case (heater). At the outlet of the bubble column crystallizer the moisture con-tent is removed by condensing through passing it over the cooled surface of the heat pump evaporator (cooler). As shown in Fig. 4, we use a heat pump system to achieve a coefficient of performance (COP) which enables us to have specific operating conditions for the process.

Material Balance for Direct Contact Evaporative Crystallization Process with Heat Pump

Fig. 1 shows a flow diagram of a direct contact evaporative crystallization process using hot air as con-tact phase with the solution. The process is carried out

in a bubble column crystallizer. Air is heated usually by an electrical heater or by contacting with other heating agents. Hot air of a flow rate (L) is brought into contact with solution (F) with a temperature and concentration TF, XF, respectively, in a bubble column causing the air to cool and some of the liquid to evaporate. The air temperature will decrease from T1 at the inlet of the column toT2 at the outlet. Consequently, the absolute and relative humidity will increase from αA1, αR1 to αA2

, αR2 respectively. The enthalpy of air may remain con-stant if the contact in the bubble column is going under adiabatic conditions or it may change from H1 at the inlet to H2 at the outlet, if other cooling conditions are applied. Distilled water of amount (W) is obtained by dehumidifying the air in a condenser (the evaporator of heat pump). The formed slurry concentrated solution (D) is separated to crystal phase (K) and mother liquor (M).

The overall material balance is:

)( 21 AALMKF αα −++= (1)

The amount of evaporated solvent (water) is deter-mined according to the difference of moisture content (air humidity) between the inlet and outlet as:

)( 21 AALW αα −= (2) Substituting Eq. (2) into Eq. (1):

WMKF ++= (3)

Also from

MKD += (4)

Then Eq. 4 becomes

WDF += (5)

The balance on solute can be expressed as:

DF WFX DX WX= + (6)

where WDF XXX ,, are the salt concentrations in the feed, slurry concentration solution and condensate phases, respectively. Taking into account that evaporated water contains zero amount of solute (XW = 0), then equation

Fig. 3. The process of direct contact crystallization with a heat pump.

Fig. 4. The operating conditions of the system and COP of the heat pump.


280

(6) becomes:

F DFX DX= (7)

An expression for the amount of crystal phase as slurry can be obtained by solving Eq. (7)

FD

FXXD

= (8)

The amount of crystals formed can be calculated by knowing the efficiency of the separator. Heat Balance of Direct Contact Evaporative Crystal-lization with Heat Pump

The proposed process consists of direct contact evaporative crystallization with a heat pump as shown in Fig. 3. The initial air with a temperature (T1) is heated to a desired temperature (T2) by directing the air through the condenser (heater). At the outlet of the bubble col-umn crystallizer the moisture content is removed by condensing it when passing over the cooled surface of the evaporator. The amount of heat that must be added to the air is the same or less than the amount of heat that must be delivered by the heat pump in the condenser and can be calculated as:

)( 1 oH HHLQ −= (9)

The evaporator energy balance is:

3 2 3 2( ) ( )L pairQ L H H LC T T= − = − (10)

The measure of performance of a heat pump is ex-pressed in terms of coefficient of performance COPHP, defined as

HHP

in

QCOPW

= (11)

By applying energy parlance on the heat pump

in H LW Q Q= − (12)or

LinH QWQ += (13)

An expression for the applied work can be obtained by solving Eq. (11) and Eq. (12) together:

( 1)L

inHP

QWCOP

=− (14)

Economic Analysis for a Direct Contact Evaporative Crystallization with a heat pump

To evaluate the feasibility of introducing a heat pump into direct contact evaporative crystallization, it is required to estimate the cost of heating for an ordi-nary system and compare it with a system that includes a heat pump. Depending on the operating conditions, i.e. the source of heating medium used in the heater to heat up the initial air, the total operating cost ($/hr) can be estimated as:

heaterSHS EDQE +×= (15)

where DS is the price of heating source ($/kJ) in case of using steam, Eheater ($/h) is the cost of heating agent used in the heater

EHE DQE ×= (16)

where DE is the price of the heating source ($/kWh) in case of using electricity.

The cost of the process with a heat pump consists only from the cost of the electrical power supplied to the compressor which is estimated as:

η×= ina WP (17)

where Pa is the actual electrical power required for a compressor and η is the overall efficiency of the com-pressor which ranges from 65 to 75 % [8]. The operating cost ($/hr) of the system with a heat pump is:

The resulting saving factor by introducing a heat pump to a system of direct contact evaporative crystallization is defined by:

,

(1 ) 100%HE

S E

ESE

= − × (19)

where EH is the total operating cost of the heat pump in ($/hr) and ES,E in case of using steam or electricity, respectively.

Zaid Ahmed Al-Anber

281

Energy Recycling and Recovery in Case of Indirect Contact Evaporative Crystallization

For an indirect evaporative process we studied the feasibility to energy recovery process i.e. the outlet steam that exited from crystallizer and jacket that would normally be wasted, converting it into thermal energy by addition of a heat pump on outlet vapor line and steam in three designs:

The inlet to the evaporator is outlet vapor from the crystallizer (Fig. 5).

The inlet to the evaporator is outlet steam from the jacket (Fig. 6).

The inlet to the evaporator is a mixture of the outlet vapor from crystallizer and steam outlet from the jacket (Fig. 7).

In the three cases, the heat pump makes a closed system, the condenser of the heat pump is connected on the inlet steam line to make a preheating of the steam inlet of the jacket, then the steam completeed the heat-ing to the required temperature using heater with low pressure steam.

The recovery energy of indirect contact evaporative crystallization with a heat pump first suggested system for energy is shown in Fig. 5. The superheated steam with temperature (T2 = 150oC) is fed to the jacket to be cooled and to give the evaporator crystallizer the required heat to realize the crystallization state. The satu-rated steam out from jacket is at temperature T=25oC. The vapor resulting by the crystallization process out of the crystallizer is at temperature almost 70oC. The outlet vapor from the crystallizer is drawn to the evaporator of the heat pump. The condenser of heat pump acts as a preheating medium for the inlet of the jacket.

The recovery energy of indirect contact evaporative crystallization with a heat pump - the second suggested system for energy, is shown in Fig. 6. The superheated steam with temperature (T2 = 150oC) is fed to the jacket to be cooled and to give the evaporator crystallizer the required heat to realize the crystallization state. The saturated steam out from jacket is at temperate T=25oC. The vapor resulting from the crystallization process out of crystallizer is at temperature almost 70oC. The outlet steam from the jacket is drawn to the evaporator of the heat pump. The condenser of the heat pump acts as a preheating medium for the inlet of the jacket.

The recovery of indirect contact evaporative crystal-

Fig. 5. The process of indirect contact crystallization with a heat pump when the feed to heat pump evaporator is Vo.

Fig. 6. The process of indirect contact crystallization with a heat pump when the feed to the heat pump evaporator is S.

Fig. 7. The process of indirect contact crystallization with a heat pump when the feed to the heat pump evaporator consists of S and Vo.


282

lization with a heat pump - the third suggested system for energy is shown in Fig. 7. The superheated steam with temperature (T2 = 150oC) is fed to the jacket to be cooled and to give the evaporator crystallizer the required heat to realize the crystallization state. The saturated steam out from the jacket is at temperate T=25oC. The vapor resulting by the crystallization process out of the crys-tallizer is at temperature almost 70oC. The two outlets from jacket and crystallizer are mixed and drawn to the evaporator of the heat pump. The condenser of the heat pump acts as preheating medium for the inlet of the jacket.

Material Balance for the Indirect Contact Evapora-tive Crystallization Process by Adding a Heat Pump

Fig. 7 shows a flow diagram of the indirect contact evaporative crystallization process by using superheated steam without contact phase with solution (through jack-et). The process is carried out in an evaporative column crystallizer. Steam is heated first by the condenser of the heat pump and then heated by an electrical heater or by contacting with other heating agents to reach to a desired temperature (T2). A superheated steam with a flow rate (S) is fed to the jacket that heats the solution (F) with a temperature and concentration TF, XF, respectively. In an evaporative column some of the liquid is evaporated and the steam cooled (the superheated steam temperature will decrease): the outlet steam from the jacket is mixed with the steam out from the crystallizer to give steam with a flow rate (V) and a temperature (T4), this steam is fed to the evaporator of the heat pump. The temperature will decrease from T4 at the inlet of the evaporator to T5 at the outlet. This decrease in temperature releases a definite amount of heat QL. At given COP and known amount of QL, we can estimate the amount of heat QH used for preheating the steam inlet to the jacket by the condenser of the heat pump after which a complete heating by the electrical heater or by contacting with other heating agents to reach to T2 can give the operating conditions to realize the crystallization state.

The overall material balance is:

oV++= MKF (20)


oVF k M voFX kX MX X= + + (21) where XF, XK, XM, XVo are the salt concentrations in feed, crystal, mother liquor and vapor phase, respectively. Taking into account that evaporated water contains zero amount of solute (XVo = 0), equation (21) becomes:

F k MoFX kX MX= + (22)The total material balance is:

MKD += (23)

The component material balance is:

F K MDX KX MX= + (24)

Then eq. (20) becomesoVDF += (25)


OVF D vFX DX X= + (26)An expression for the amount of crystal phase can

be obtained by solving Eq. (25) and Eq. (26) together and as we know XVo = 0 then:

D

FO

XX

FV

−= 1 (27)

The total steam fed to the evaporator of heat pump is:

SVV o += (28) Heat Balance for Indirect Contact Evaporative

Crystallization with a Heat PumpBy applying energy balance on the steam that is fed

to the jacket

2 1 o voSh Sh V H− = (29)

and then S:

32 hhHVS voo

−= (30)

The amount of heat generated from the evaporator can be calculated as:

1 - if the inlet to the evaporator is steam Vo:

Zaid Ahmed Al-Anber

283

)( 54 hhVQQ ovL −== (31)

2 - if the inlet to the evaporator is steam S:

4 5( )L soQ Q S h h= = − (32)

3 - if the inlet to the evaporator is steam V ( S +Vo):

sovL QQQ += (33)The measure of performance of a heat pump is ex-

pressed in terms of coefficient of performance COPHP, defined as:

HHP

in

QCOPW

= (34)

By applying energy balance on the heat pump

in H LW Q Q= − (35)and

H in LQ W Q= + (36)

An expression for the amount of work can be ob-tained by solving Eq. (34) and Eq. (36) :

( 1)L

inHP

QWCOP

=− (37)

The expression for the amount of heat that required increasing temperature T1 to T2 using a heater (without a heat pump):

)( 32 hhSQS −= (38)

The makeup amount of heat (using a heater) which must be added after the heat pump, QJ can be calculated by:

HsJ QQQ −= (39)

Economic Analysis for an Indirect Contact Evapora-tive Crystallization Process with a Heat Pump

To evaluate the feasibility of adding a heat pump compressor on the outlet steam line (from crystallizer) in evaporative crystallization, it is required to estimate the cost of heating for an ordinary system and compare it with a system that includes a heat pump. Depending on the operating conditions, i.e: the source of heating medium used in the heater to heat up the inlet steam temperature (fed to the jacket), the total operating cost

($/hr) can be estimated as:

heatersS SDQ Φ+×=Φ (40)

where DS is the price of the heating source ($/kJ) in case of using steam, Фheater: ($/h) is the cost of heating agent used in the heater. The cost of the process with a heat pump consists only from the cost of the electrical power supplied to the compressor which is estimated as:

a inP W η= × (41)

where Pa is the actual electrical power required for a compressor and η is the overall efficiency of the com-pressor which ranges from 65 to 75 % [8]. The operat-ing cost ($/hr) of the electrical power supplied to the compressor is estimated as:

EH aP DΦ = × (42)

The operating cost ($/hr) of the heat that is needed

after preheating from heat pump (heating make up) is estimated as:

heaterJJ SDQ Φ+×=Φ (43)

the total operating cost ($/hr) of the process with a heat pump on the outlet vapor line (from crystallizer) is sum-mation of the cost of the electrical power supplied to the compressor and the cost of heating make up steam, which can be estimated as:

HJT Φ+Φ=Φ (44)The resulting saving factor by indirect contact

evaporative crystallization is defined by:

(1 ) 100%E

T

s

S Φ= − ×

Φ (45)

where ФT the total operating cost of the added com-pressor on the outlet steam line (from crystallizer) in ($/hr) and Фs the total operating cost ($/hr) without a heat pump.


Case one: Direct Contact Evaporative Crystalli-zation with a Heat Pump

In our study we use a heat pump system to make a relation between the coefficient of performance (COP)


284

and the initial temperature of the hot air fed to the crystallizer at constant humidity. Then by solving of this relation we can see the effect of introducing a heat

pump on the operational energy saving in the process of direct contact evaporative crystallization. The followings assumptions are proposed:

Steady state, steady flow operation. Potential and kinetic energy effects are negligible. No chemical reaction. Adiabatic compression efficiency is 0.7. The price of electricity and low pressure steam for

heating (LPS) are estimated to be 0.05 ($/kW h) and 3.17x10-6 ($/kJ) respectively, and the cost of heating agent used in the heater 0.33 $/kJ [7].

The used high-efficiency heat pump has COP of (2.3-5) in the heating mode [6].

The used basis for calculation is listed in Table 1.The calculations of the heat accepted by the evapora-

tor QL, work input into heat pump win, heat generated by condenser QH, inlet air temperature T1, mass flow rate of water W, mass flow of slurry concentrated solution D, slurry concentration XD, and air moisture content at

Fig. 8. Relation between inlet air temperature to crystal-lizer and slurry concentration.

Table 2A. Material and energy results of direct contact evaporative crystallization with heat pump. COP QL

(KJ/hr)

Win

(KJ/hr)

QH

(KJ/hr)

T1

(K)

W

(Kg/hr)

D

(Kg/hr)

XD αA 1

2.3 9270 7130.8 16400.8 324.7 1.8 998.2 0.360649 0.019

2.5 9270 6180 15450 323.2 3 997 0.361083 0.017

3 9270 4635 13905 320.7 3.6 996.4 0.361301 0.016

3.5 9270 3708 12978 319.2 3.6 996.4 0.361301 0.016

4 9270 3090 12360 318.2 4.2 995.8 0.361518 0.015

4.5 9270 2648.6 11918.6 317.4 5.4 994.6 0.361955 0.013

5 9270 2317.5 11587.5 316.9 6.6 993.4 0.362392 0.011

F 1000 kg/h Cp air 1.03 kJ/kg K Xf 0.36 αA2 @ T2 0.022 kg water/kg dry air T2 25oC αR1 20 % T3 10oC αR2 100 % L 600 kg/h

Table 1. Basis for calculations [1, 4].

Zaid Ahmed Al-Anber

285

T1 are presented in Table 2A.The calculations of total operating cost without heat

pump (heating by steam) ES, the total operating cost without a heat pump (heating by electricity)EE, the total operating cost with heat pump EH, saving factor based on electrical heating SEE , and saving factor based on steam heating EES are presented in Table 2B.

The relationship between the inlet air temperature and the slurry concentration is shown in Fig. 8. It can be seen that as the inlet air temperature decreases, the concentration of slurry increases. This can be understood

by the fact that at constant flow, the humidity of air flow decreases by lowering temperature and then maximize the ability of air to carry out water from solution.

Figs. 9 and 10 show that increasing of the value of COP results in increasing of the concentration of the produced slurry. It can be explained by the fact that the relationship between COP and the temperature of inlet air is inversely proportional.

The calculations indicate that the saving factor increases in both cases by increasing the COP. A noticeable saving factor is ranged between 81 % to

Fig. 9. The coefficient of performance (COP) effect on inlet air temperature to crystallizer (T1).

Fig. 10. Effect coefficient of performance (COP) on produced slurry concentration.

COP Es ($/ hr)

EE ($/ hr)

EH ($/ hr)

SEE

SEs

2.3 0.38199 0.227788 0.069327 69.5652 81.85114

2.5 0.378977 0.214583 0.060083 72.0000 84.14589

3 0.374079 0.193125 0.045063 76.6667 87.95374

3.5 0.37114 0.18025 0.03605 80.0000 90.28669

4 0.369181 0.171667 0.030042 82.5000 91.86262

4.5 0.367782 0.165536 0.02575 84.4444 92.99857

5 0.366732 0.160938 0.022531 86.0000 93.85621

Table 2B. The cost and saving factor results of direct contact evaporative crystallization with heat pump.


286

93 % and 69 % to 86 % based on electrical and steam heating, respectively. The results are shown in Figs. 11 and 12.

Case Two: Adding a heat pump on the outlet vapor line from the crystallizer

The aim of this construction is to utilize and recover heat content of the system by adding a heat pump in three cases:

If the inlet to the evaporator (of heat pump) is outlet vapor from the crystallizer.

If the inlet to the evaporator is outlet steam from the jacket.

If the inlet to the evaporator is a mixture of the outlet vapor from crystallizer and steam outlet from the jacket

The followings assumptions are proposed: Steady state, steady flow operation.Potential and kinetic energy effects are negligibleNo chemical reaction.Adiabatic compression efficiency is 0.7.The price of electricity and low pressure steam for

heating (LPS) are estimated to be 0.05 ($/kW h) and 3.17x10-6 ($/kJ) respectively, and the cost of the heating agent used in the heater 0.33 $/kJ [9].

The used basis for calculation is listed in Table 3 for QL. There three value can be calculate, based on what is the inlet to the evaporator of heat pump:

If the inlet to the evaporator is steam Vo at T4=25oC

and out at T5=12oC;If the inlet to the evaporator is steam S at T4=25oC

and out at T5=12oC;IF the inlet to the evaporator is steam V (Vo & S)

at T4=25oC and out at T5=12oC. (See Figs. 5, 6 and 7).

The used high-efficiency heat pump manufactured has a COP of (2.3-5) in the heating mode [8].

The basis of calculation is presented in Table 3. The calculations of flow rate of outlet steam from

indirect crystallizer Vo, flow rate of inlet steam fed to the

Fig. 11. Dependency of saving factor (SEE, EES) on coef-ficient of performance (COP).

Fig. 12. Cost of heating inlet air with: a - Heat pump b - Electrical heater and c - LPS steam in heater.

Table 3. Basis for calculations [1, 4].

Fig. 13. Dependency of saving factor on different heat pump coefficients of performance.

F 1000 kg/h XD 0.6 enthalpy of superheated steam at (75 kPa, 150oC)

2778.2 kJ/kg

enthalpy of saturated steam at 25oC 2547.2 kJ/kg enthalpy of saturated steam at 70oC 2626.8 kJ/kg enthalpy of saturated steam at 12oC 2523.4 kJ/kg latent heat of vaporization at 70oC 2405.9 kJ/kg

Zaid Ahmed Al-Anber

287

jacket S, Heat accepted by evaporator QL, work input of heat pump Win, heat generated by condenser QH, heat generated by heater without heat pump QS, and the amount of the heat that needed after preheating form heat pump QJ are presented in Table 4A. All calculations are at constant feed concentration and the feed to the heat pump evaporator consists of S and Vo.

Table 4B shows the calculations of total operating cost without heat pump Фs; The total operating cost of heat pump ФH.; The operating cost of the heat heat-ing make up (after preheating by heat pump) ФJ ; the total operating cost of the process with a heat pump and heater ФT and saving factor SE. All calculations are at constant feed concentration and the feed to heat pump

COP XF Vo

(Kg/hr)

S

(Kg/hr)

QL

(KJ/hr)

Win

(KJ/hr)

QH

(KJ/hr)

Qs

(KJ/hr)

QJ

(KJ/hr)

2.3 0.22 633.3333 6596.263 834427.2 641867.094 1476294.316 1523737 47442.35

2.5 0.22 633.3333 6596.263 834427.2 556284.8148 1390712.037 1523737 133024.63

3 0.22 633.3333 6596.263 834427.2 417213.6111 1251640.833 1523737 272095.83

3.5 0.22 633.3333 6596.263 834427.2 333770.8889 1168198.111 1523737 355538.56

4 0.22 633.3333 6596.263 834427.2 278142.4074 1112569.63 1523737 411167.04

4.5 0.22 633.3333 6596.263 834427.2 238407.7778 1072835 1523737 450901.67

5 0.22 633.3333 6596.263 834427.2 208606.8056 1043034.028 1523737 480702.64

Table 4A. Material and energy balances in the case of adding heat pump on outlet vapor line from crystallizer.

COP Фs

($/hr)

ФH

($/hr)

ФJ

($/hr)

ФT

($/hr)

SE

($/hr)

2.3 5.160245 6.240374525 0.480392251 6.720767 -30.2412

2.5 5.160245 5.408324588 0.751688076 6.160013 -19.3744

3 5.160245 4.056243441 1.192543792 5.248787 -1.71585

3.5 5.160245 3.244994753 1.457057221 4.702052 8.879292

4 5.160245 2.704162294 1.633399507 4.337562 15.94272

4.5 5.160245 2.317853395 1.759358283 4.077212 20.98802

5 5.160245 2.028121721 1.853827365 3.881949 24.772

Table 4B. The Cost and saving factor in the case of adding heat pump on outlet vapor line from crystallizer.

Fig. 14. Variation of saving factor with different feed concentrations at constant COP.


288

evaporator consists - S and Vo.Fig. 13 shows the three cases at value of COP lower

than 3.5. The addition of a heat pump is useless, because there is no saving (there is economical efficiency). But for values of COP higher than 3.5, the addition of heat pump gives a noticeable saving and the saving factor is increased by increasing COP value.

Fig. 14 also shows that the highest saving factor is achieved by addition of a heat pump and mixing of steam from jacket and compressed vapor from crystallizer. It ranges from 8 to 25 %.

CONCLUSIONS

The heat saving in an evaporative crystallization processes in the two types (direct and indirect evaporative crystallizers) is studied:

In case of a direct evaporative crystallizer, a notice-able saving factor ranges between 69 to 86 % and 81 to 93 % based on electrical, steam heating, respectively, was achieved by adding a heat pump with variable COP. The process of direct contact evaporative crystallization has an advantage of producing an additional amount of distilled water as a byproduct of the crystallization.

In the case of indirect evaporative crystallizer, no-ticeable saving factor ranges between 8 to 26 % based on steam heating were achieved by adding a heat pump on the outlet vapor line with variable COP. The process of direct contact evaporative crystallization has an advan-tage of producing an additional amount of desalinated water, when the main purposes of separation is extraction

of minerals, comparing to traditional open solar pond evaporative systems.

REFERENCES

1. C.J. Geankoplis, Transport processes and unit Opera-tions, 3rd Ed, United States of America, 1993.

2. J. M. Coulson, J. F. Richardson, J. R. Backhurst, J. H. Harker, Chemical Engineering, Volume Two, 3rd Ed, 1978.

3. Mullin J.W., Crystallization, 3rd Ed., MPG Book, London, 2000.

4. Perry, H. Robert, Green .W. Don, Perry’s chemi-cal engineer’s handbook, MC Graw Hill, New York,1999.

5. Alan S. Foust, Leonard A.Wenzel, Curtis W. Clump, Louis maus, L. Bryce Andersen, Principles of Unit Operations, Bethlehem, Pennsylvania, 1960.

6. Adnan M. Al-Harahsheh, Theoretical analyses of en-ergy saving in a direct contact evaporative crystalliza-tion through the installation of heat pump, Desalination 251, 1-3, 2010, 47-52 .

7. Adnan M. Al-Harahsheh, A heat pump in a coun-tercurrent crystallization process Applied Thermal Engineering, 2005, 545-555.

8. Yunus A. Cengel Michael A. Boles, Thermodynamics An Engineering Approach 5th Ed., MC Graw Hill, 2006.

9. L. Ying, P.C. Flynn, Deregulated power price com-parison of diurnal patterns, Energy Policy 32, 2004, 657-672.

Mohammad Javad Ghaderi, Mahdi Shafiee Afarani, Ghodratollah Roudini

289


SYNTHESIS OF ALUMINA POROUS SUPPORTS VIA DIFFERENT COMPACTION ROUTES: VIBRATION AND PRESSING


Department of Materials Engineering, Faculty of Engineering, University of Sistan & Baluchestan,Zahedan, IranE-mail: [email protected]

ABSTRACT

Porous alumina supports were fabricated using vibration and pressing compaction methods, and their physical and mechanical properties were investigated. After packing of alumina powder with and without silica additive, the samples were sintered at 1325°C - 1625°C for 2 hours. The effects of sintering temperature and silica addition (about 5 mass %) on porosity, thermal conductivity, compressive and flexural strength were studied. Results showed that compaction routes strongly affect sintering and consequently, properties of the samples. Also, samples with high porosity and desirable me-chanical properties could be obtained with combination of silica addition and vibration forming. A transition from first stage sintering to intermediate stage has been observed in the temperature domain of 1400-1475°C.

Keywords: porous alumina, compaction, vibration, sintering, mechanical properties, thermal conductivity.

Received 09 July 2012Accepted 15 April 2013

INTRODUCTION

Porous ceramics have many desirable properties such as their light weight, high chemical stability, and low thermal conductivity. These properties are quite attractive for environmental, energy, biotechnology, and other applications. The porous ceramic materials are made by sintering and the physical properties and mechanical strength can be controlled by the micro-structure of the sintered powders. To achieve optimal sintered porous ceramics, control of powders parameters, dimensional tolerances, dense and homogeneous pack-ing of powders, as well as of the additives is required [1-3]. Porosity is the main cause for reduction in me-chanical properties of ceramic and brittle solids. These classes of ceramics are essential for many industries where high permeability, high surface area, and insulat-ing characteristics are required. The search for porous ceramics with good mechanical strength has stimulated the development of several technologies. The properties of porous ceramics greatly depend on pore morphology,

size and distribution [3].The mechanical properties of porous alumina bodies

were correlated with relative density and total porosity [4, 5]. Hashimoto et al. [1] investigated mechanical properties of porous alumina supports fabricated by anisotropic alumina particles with uniaxial pressing at 1 and 3 MPa and heated at 1400°C for 1h. The relative densities of the resulting alumina supports were 25.0 % and 35.5 %, respectively. The compressive strength of the compacts that were uniaxially pressed at 1 and 3 MPa were 0.8 and 4.3 MPa, respectively.

The additives affect properties of sintered alumina. Silica addition decreases shrinkage rate, hardness and toughness of specimens and formation of liquid phase at high temperatures which leads to abnormal grain growth [6-8].

Roudini et al. [9] produced alumina by vibrating the pure powder with narrow size distribution cold isostatic pressing (CIP) and sintering at 1283-1530°C for 2h for synthesis of alumina particle reinforced aluminum com-posites. They measured only the contiguity, green and


290

final density of these performs. Their results indicated that the green density of vibrated performs was 0.52 g/cm3 and the final density was 0.52 - 0.60 g/cm3; the contiguity was in 0 - 0.119 range. Despite the existence of a huge number of open-literature investigations on the processing, sintering and properties of porous and dense supports, the synthesis of alumina porous supports via vibration method is more than scarcely discussed.

The forming method, silica addition and sintering temperature dramatically change relative densities and connectivity of alumina particles and consequently, the bending strength, compression strength and thermal conductivity of final samples. It can be concluded that significant improvements in the mechanical properties and thermal conductivity of porous ceramics can be obtained by control of processing (e.g. forming method), of the sintering aids and temperature.

The aim of the present work was the synthesis of porous supports and the study of their physical and me-chanical properties. The effects of compaction methods, including vibration and uniaxial pressing, sintering temperatures and silica addition have been investigated.

EXPERIMENTAL

High purity α-Al2O3 powder (WDR4, Indal chemi-cal, India) was used as the starting material. The physi-cal properties and chemical composition of this powder declared by the manufacturer are given in Table 1. Two types of granules were prepared for manufacturing of the samples. One was pure alumina powder with proper-ties given in Table 1 and the other mixture of alumina

powder and silica (SiO2) containing 95 mass % Al2O3 and 5 mass % SiO2 (SYLOID AL-1 FP Pharmaceuti-cal Excipient, d50 = 6.8 – 8.1 μm, purity 99.4 %) as an additive. For granulation, 4 mass % PVA (0.2 mass % aqueous solution) was added to the powders, gently blended and aged for 48 h at room temperature. Then, sieved (< 355 µm) granules were used.

Green compacts were formed using the uniaxial pressing and vibration method. For pressed specimens, powders were uniaxially pressed in the pressure range of 2 - 5 MPa in a designed mold, and for vibrated samples, powders were poured in graphite molds and vibrated with a 50 Hz shaker for 5 minutes. Cylindrical green compacts with 18 mm diameter and 45 mm height were produced for the compressive and thermal conductivity tests. Rectangular green compacts with 50.5×12×7 mm3 in size were fabricated for flexural test.

Sintering was performed at 1325, 1400, 1475, 1550 and 1625°C with 2 h soaking time. Heating rates from room temperature to 1000°C and from 1000°C to the sintering temperature were 10°C min-1 and 5°C min-1, respectively. Relative densities of the specimens were determined using mass and dimensions measurements. Porosity was determined with 2 methods: a) by mass and dimensions measurements and b) according to the ASTM C373-88 standard. Three point flexural strength measurements were performed on 58.4×4.70×3.90 mm3 rectangular blocks with a mechanical testing machine (Instron model 4208). Surface area measurements and adsorption isotherms determination were done using a micromeritic apparatus. Structure and microstruc-ture analysis of samples were performed using XRD (Unisantis XMD400) and FE-SEM (HITACHI, S4160) techniques, respectively. Thermal conductivity (W/m K) of specimens were measured with a thermal conduction apparatus (Armfield HT10XC) and calculated by Fourier’s law equation:

dTq Kdx

= − (1)

where, q is the heat transferred through the sample sur-face (W m-2), dxis the length (m), and dT is temperature difference between two heads of samples (°C).

All experiments (except heat conductivity) were performed at least 3 times. Results are presented as the mean± SD.

Median particle size (d50)

1.0 μm

BET surface area 6 m2/g Specific gravity 3.90 g/cm3

Al2O3 purity 99.5 wt.% Na2O 0.4 wt.% SiO2 0.03 wt.% Fe2O3 0.03 wt.% TiO2 0.006

wt.%

Table 1. Characterization of alumina powder (as indicated by the manufacturer).


291


A continuous trend of relative density increase and porosity decrease as a function of sintering temperature, both for pressed and vibrated samples, is shown in Fig. 1a and b. Since samples sintered at 1325°C were weak, porosity values of this temperature were not measured. For pressed specimens, porosity has been lower than for vibrated ones due to higher initial compaction of the pressed green compacts. It seems that the difference be-tween porosity values in samples with and without silica, is due to larger mean size of silica and consequently different compactions of green samples.

It can be observed that for vibrated samples, po-

rosity values remains higher than 30 % for sintering temperatures up to 1550°C. For pressed samples, this temperature is 1475°C. A considerable porosity decrease (or relative density increase) due to transition from initial to intermediate stage of sintering is observed both for pressed and vibrated samples within the temperature domain of 1400-1475°C. Within the transition zone, the densification process undergoes acceleration. Similar behavior may also be observed in the shrinking versus sintering temperature curves (Fig. 2).

Surface area and nitrogen adsorption isotherms de-termined of BET technique versus sintering temperature for vibrated alumina containing SiO2, are shown in Fig. 3 and Fig. 4, respectively. As seen, a considerable surface

Fig. 1b. Porosity versus sintering temperature for pressed and vibrated samples.

Fig. 3. Surface area versus sintering temperature for vi-brated alumina with SiO2 samples.

Fig. 2. Shrinkage versus sintering temperature for pressed and vibrated samples.

Fig. 1a. Relative density versus sintering temperature for pressed and vibrated samples.


292

area and nitrogen adsorption decrease with sintering temperature increase due to open porosity decline is observed.

Funnel shaped corners between large particles can be considered as mesopores (2 - 50 nm) which are good for condense vapor. Mean mesopore size evaluation us-ing nitrogen adsorption isotherms could help the better understanding of the sintering process. Considering a Maxwellian pore volume distribution, the total pore volume )(PV calculated by the following equation:

−−−+−+−= )]exp()[exp(2)]exp()1(1[

0

max

0000)()( r

rrr

rt

rr

rrVV SP

−−−+−+−= )]exp()[exp(2)]exp()1(1[

0

max

0000)()( r

rrr

rt

rr

rrVV SP

(2)

where SV , is the total pore volume (cm3g-1), 0r - a con-stant (nm), r - the pore radius (nm), maxr - the thresh-old pore radius (nm) and t - the statistical thickness of the adsorbed layer (nm). Using nitrogen adsorption isotherms at sintering temperature and equation (2), the mesopores size distribution could be evaluated at each sintering temperature [10]. The change in the mean mesopores radius as a function of the sintering tempera-ture is shown in Fig. 5. For the last sintering temperature (1625°C) it has not been evaluated due to scattering data. The mean mesopore radius first decreases in the 1400-1475°C range, then increases in the 1475-1550°C tem-perature range. For this phenomenon, two mechanisms have been suggested. They are shown schematically in Fig. 6. First, sample shrinkage causing reduction of the funnel tip diameter (Fig. 6b) and second - tip filling by different mass transfer mechanisms causing to enlarge-ment of the mesoperes (Fig. 6c) [10].

The fracture surface micrographs of specimens are shown in Fig. 7. At 1375°C, particles are partially bonded together with little connectivity and wide po-rosity for different samples (vibrated/pressed and with/without silica addition) as expected for the initial sin-tering step of porous ceramics (Fig. 7a &c). In samples without silica, solid state sintering is dominant, as shown in Fig. 7b, but in samples with silica liquid state sintering occurred (Fig. 7d). As sintering temperature increased, porosity decrease and grain growth could be observed (Fig. 7b and d). Obviously, for liquid phase sintered samples, these phenomena were more significant than for the solid state sintered ones.

Fig. 4. Nitrogen adsorption versus relative pressure for vibrated alumina with SiO2 samples sintered at different temperatures.

Fig. 5. Mesopore radius versus sintering temperature for vibrated alumina with SiO2 samples.

Fig. 6. The changes of mesopore during sintering progres-sion: (a) Initial diameter before transition zone (b) Diameter increase due to tip filling by different mass transfer mecha-nisms (c) Diameter reduction due to specimen shrinkage.


293

The XRD pattern for a sample of alumina with silica additive is shown in Fig. 8. Alumina (Card No: 01- 071- 1126) is the dominant phase and mullite (Card No: 00 – 001- 613) is a minor phase formed due to reac-tion between alumina and silica.

Like relative density and shrinkage curves, a con-tinuous trend of compressive and bending strength increase with sintering temperature increase, both for

pressed and vibrated samples is shown in Figs. 9 and 10, respectively. Two main parameters affect strengths of samples: porosity and binding nature (solid or liquid phase sintering) of particles. Considering vibrated and pressed samples sintered at 1625°C shows more porosity has led to less strength. Furthermore, in samples with same porosity and different (solid/liquid) state sinter-ing, (for example, pressed samples sintered at 1625°C)

Fig. 7. Microstructure of alumina samples sintered at different temperatures: (a) Pressed sample without silica sintered at 1325°C, (b) Pressed sample without silica sintered at 1625°C (c) Vibrated sample with silica sintered at 1325°C and (d) Vibrated sample with silica sintered at 1625°C.

Fig. 8. XRD pattern of pressed alumina with silica sample sintered at 1625°C.

Fig. 9. Compressive strength versus sintering temperature for pressed and vibrated samples.


294

particles’ bonding strength is higher when solid state sintering is dominant.

The variation in thermal conductivity of uniaxilly pressed and vibrated samples as a function of sintering temperature is shown in Fig. 11. It gives evidence that the sintered alumina at higher temperatures has more thermal conductivity paths between connected parti-cles. Several factors including composition, compac-tion, grain size and boundaries, rigidity and strength of samples have influence on the thermal conductivity of

polycrystalline ceramics. In porous ceramics, porosity is a very important and dominant parameter. With increase of porosity, a dramatic thermal conductivity decrease occurs, as shown in Fig. 11 for pressed and vibrated samples sintered at 1625°C. Moreover, in samples with the same porosity and different (solid/liquid) sintering state, (for example, pressed samples sintered at 1625°C) solid state sintering has led to higher strength of parti-cles’ connections and rigidity and therefore - to increase of thermal conductivity.

CONCLUSIONS

For both pressed/vibrated alumina samples with/with-out silica addition, a transition from initial to intermediate stage of sintering is observed within the temperature domain of 1475–1550°C. In this region, physical and mechanical properties like surface area, strength and thermal conductivity may strongly change, mainly due to densification progress and porosity decrease.

Vibrated samples without silica sintered at 1550°C cann give high porosity (50 %) and appropriate strength. For pressed samples with silica addition, the optimum sintering temperature for giving both proper porosity and strength is 1475°C.

REFERENCES

1. S. Hashimoto, S.Horita, Y. Itoa, H. Hirano, S. Honda, Y. Iwamoto, Synthesis and mechanical properties of porous alumina from anisotropic alumina particles, Journal of the European Ceramic Society, 30, 2010, 635-639.

2. S. Bhattacharjee, L. Besra, B.P. Singh, Effect of addi-tives on the microstructure of porous alumina Journal of the European Ceramic Society, 27, 2007, 47-52.

3. T.D. Senguttuvan, H.S. Kalsi, S. K. Sharda, B.K Das, Sintering behavior of alumina rich cordierite porous ceramics, Materials Chemistry and Physics, 67, 2001, 146-150.

4. J.S. Magdeski, The Porosity dependence of me-chanical properties of sintered alumina, Journal of the University of Chemical Technology and Metallurgy, 45, 2, 2010, 143-148.

5. V.H. Hammond, D.M. Elzey, Elevated temperature me-chanical properties of partially sintered alumina, Com-posites Science and Technology, 64, 2004 ,1551-1563.

Fig. 10. Flexural strength versus sintering temperature for pressed and vibrated samples.

Fig. 11. Thermal conductivity versus sintering temperature for pressed and vibrated samples.


295

6. Z. Misirli, A. Uguz, T. Baykara, Effect of additives on the microstructure and mechanical properties of commercial alumina ceramics, Materials characteri-zation,1994, 33, 329-341.

7. N. Louet, H. Reveron, G. Fantozzi, Sintering be-haviour and microstructural evolution of ultrapure α-alumina containing low amounts of SiO2, Journal of the European Ceramic Society, 2008, 28, 205-215.

8. N. Louet, M. Gonon, G. Fantozzi, Influence of the amount of Na2O and SiO2 on the sintering behavior and on the microstructural evolution of a Bayer

alumina powder, Ceramics International, 2005, 31, 981-987.

9. G. Roudini, R. Tavangar, L. Weber, A. Mortensen, Influence of reinforcement contiguity on the thermal expansion of alumina particle reinforced aluminium composites, International Journal of Materials Re-search, 101, 9, 2010, 1113-1120.

10. C. Falamaki, M. Shafiee Afarani, A Aghaie, Initial sintering stage pore growth mechanism applied to the manufacture of ceramic membrane supports, Journal of the European Ceramic Society, 2004, 24, 2285-2292.


296


OPTIMIZATION OF THE BASIC PARAMETERS OF CATHODIC DEPOSITION OF Ce-CONVERSION COATINGS ON D16 AM CLAD ALLOY

S. Kozhukharov1, J.A.P. Ayuso2, D.S. Rodríguez2, O.F. Acuña2,

M. Machkova1, V. Kozhukharov1

1 University of Chemical Technology and Metallurgy 8 Kl. Ohridski, 1756 Sofia, Bulgaria E-mail:[email protected] University of Vigo, Lagoas, Marcosende 36310 (Spain)

ABSTRACT

The present research work is investigation on the probabilities for application of a new cerium compound, for cathodic electrodeposition of cerium based conversion coatings (CeCC) for protection of D16 AM alloy against corrosion. For the purpose of the present study, diammonium pentanitrocerate (NH4)2Ce(NO3)5 was used, where the cerium is represented in the anionic moiety, instead of the electrolytes used up to nowadays. The barrier ability and durability against corro-sion of all coatings were evaluated by electrochemical methods – Linear Sweep Voltammetry (LSV) and Electrochemical Impedance Spectroscopy (EIS). Additionally, selected specimens underwent morphological characterization by means of canning Electronic Microscopy (SEM) combined with Energy Dispersive X-ray spectroscopy (EDX). As a result, various parameters and conditions of deposition, such as the preliminary treatment, concentration of the basic substance and ad-ditives, density of the applied electric current and duration of deposition were elucidated.

Keywords: corrosion protection, aluminium alloy CeCC, EIS, LSV, SEM and EDS.


INTRODUCTION

Regardless the employment of entire generations of new materials as carbon fibres and various composites in the transport and especially in the aircraft industry, the aluminium alloys still remain the basic constructive materials. Especially, AA2024 and AA7075 alloys are objects of special attention, due to their remarkable mechanical strength [1], being the basic constructional material for commercial [2] and military [3 - 5] aircraft. In the former case, the aluminium fuselages render “visibility” for the aircraft and airport radars, whereas in the military air-transport, the Al-frames shield the on-board navigation and communication equipment, against exterior electromagnetic influence. Recently, the importance of these alloys increased, due to their capabilities to be employed in the automobile industry [6 - 8]. In addition, the aluminium alloys encountered

application in the marine industry for production of sport boats and even ships [9 - 11]. The main advantage of the aluminium alloys, compared to the steels is that the former are much lighter (about 2.723 tonnes/m3) than the latter (about 7.840 tonnes/ m3) [12]. Finally, the alu-minium alloys are valuable for military naval building, rendering to the ships lower radar detection visibility [13], compared to the steel constructions.

Nevertheless, the corrosion protection by coatings is indispensable for all kinds of aluminium details and equipment. The conventional coatings are always composed by multilayer systems, where each layer has its own function [14-17]. On the other hand, the environmental restrictions regarding the employment of chromium and other heavy metals in EC [18, 19] and USA [20, 21] impose demants for alaboration of environmentally acceptable coatings. In that means, the Ce(III) compounds have proven to be excellent

S. Kozhukharov, J.A.P. Ayuso, D.S. Rodríguez, O.F. Acuña, M. Machkova, V. Kozhukharov

297

inhibitors for localised, particularly pitting corrosion, whereas the Ce(IV) species act rather as corrosion ac-tivators [22-24]. The importance of the anionic part of the Ce-compounds has also been described, elsewhere [25]. All the facts mentioned above predetermine the nowadays elevated interest to the application of Ce ox-ides/hydroxides in form of Cerium containing primary coating layers on aluminium alloys [3, 4, 26-34] and stainless steel [37 - 40].

Arenas et al. [31] define the conversion coatings as products of chemical or electrochemical process, consisted on formation of a metallic oxide, with differ-ent properties, being substitute of the native superficial oxide layer of the respective substrate. Consequently, CeCC could be deposited by various methods, such as spontaneous deposition [3, 4, 26, 27, 30], spray coating [34] electrodeposition [35], spin coating, [41], etc.

EXPERIMENTAL

Composition and of the metallic substrates All substrates were composed of aluminium alloy,

D16-AM, described in GOST 172342 – 99. According to [42], this alloy is typical presenter of the AA2024 class. The alloy sheet for the present work is produced in Ukraine, and delivered by Klöckner Metalsnab AD, (Bulgaria) [43]. Because the standards allow some deviations from the basic nominal composition, the Al-sheet was submitted to an additional analysis, in order to determine its exact chemical composition. The element determination was performed by Inductively Coupled Plasma Optical Emission Spectroscopy (ICP-OES) in the Central Scientific Research Laboratory “Geokhimia”- Sofia. The exact composition of the alloy is summarized in Table 1.

The alloy sheet was cut to metallic coupons with size dimensions, shown in Fig. 1. Afterwards, they were sub-sequently polished up to 1000 grits emery paper, etched in NaOH solution (50 g/l) at 55°C for 5 minutes, and activated in diluted HNO3 (1:1) for 5 minutes, at ambi-ent temperature. Before each stage of the procedure, the

samples passed cleaning by vigorous tap-water, followed by distilled water cleaning.

Coating depositionThe CeCCs were deposited via electrodeposition at

either -2, or -5 mA/cm2, for 5 min from water solutions of (NH4)2Ce(NO3)5.4.H2O – (“Alfa Aesar”, Karlsruhe, Germany) with concentrations, varying from 0.03 to 0.10 M. Prior the respective electrodeposition, 30 % H2O2 (“Valerus”, Sofia, Bulgaria) was added to the solutions, in 1 to 10 ml. additions to 100 ml. of CeCC deposition solution. This approach enabled to determine the optimal composition of the coating solution.

Measurements and characterizationsThe electrochemical CeCC depositions were per-

formed in a Galvanostatic regime, allowing recording of the potential during coating deposition and the fol-lowing of its kinetics. Afterwards, the CeCC corrosion protective properties were characterized by linear sweep voltammetry and electrochemical impedance spectros-copy. Both the depositions and characterizations were performed in three electrode “flat cell”, according to ISO 16773-2:2007 (E) Standard [44]. In both cells, Ag/AgCl-3M KCl, commercial electrode, model 6.0733.100, product of Metrohm (Netherlands) served as reference. The original “horseshoe” – shaped counter electrode was used for the needs of electrodeposition, whereas cylinder-shaped platinum net with highly developed superficial area was used for counter electrode in the test procedures. Other differences between the deposi-tion and test cell, were the areas of the samples exposed to the liquid medium (shown in Fig. 1(c)), as well as its composition. The corrosion protective properties of CeCC, were examined after 24 h of exposition to natu-rally aerated 3.5% NaCl model corrosive medium. This exposure extension was selected, since in previous works [45, 46] was established that the Open Circuit Potential (OCP) stabilizes its value after at least 8 – 9 hours of exposition. The specimens were not submitted to larger exposition extensions, because the anodic polarization

Element Al Cu Fe Mg Mn Ni Si

Content (%wt.) Residual 3.716 0.404 1.259 0.537 0.055 < 0.01

Table 1. Composition of AA2024 alloy according to ICP – OES analysis.


298

curves cause irreversible damage on the electrodes. In addition, the durability of the coatings was not examined, because they are subjected to further investigations after their sealing by phosphate layer, as is proposed in [4, 29, 47, 48].

All the electrochemical procedures were performed by PG-stat Autolab 30/2, product of Ecochemie (the Netherlands). The respective impedance spectra were acquired at the following conditions: frequency range from 104 to 10-2 Hz, distributed in 7 frequencies per decade, with signal amplitude of 10 mV according to OCP. Afterwards, individual cathodic and anodic polari-zation curves were recorded in a larger potential interval (from OCP ± 30mV to OCP ± 500 mV, respectively), at 1 mV/s potential sweep rate. The latter curves were recorded after quenching of the polarization, caused by recording of the cathodic curves.

Superficial morphology and coating compositionThe SEM observations were performed by

TESCAN, SEM/FIB LYRA I XMU, supported by detector BRUKER-Quantax 200 for the EDS char-acterisations.


Influence of concentration of the basic substance of the conversion bath on the features of the CeCC

Five metallic substrates were submitted to elec-trodeposition at -2 mA/cm2, for 5 min. The concen-tration of the basic ingredient of the coating solution was selected to be: 0.03, 0.04, 0.05, 0.07, 0.10 M of (NH4)2Ce(NO3)54H2O, and 10 ml. of 30 % H2O2 per liter of coating solution for all of them.

The barrier abilities of the respective coated speci-mens were examined by electrochemical measurements after 24 hours of exposition to the corrosive medium. The EIS-spectra of the samples are shown in Fig. 2.

The Bode plots of the impedance spectra look almost

Fig. 1. Photographs of the deposition cell (a), test cell (b) and illustration of the coated and tested metallic sample (c).

Fig. 2. EIS-spectra of samples prepared by CeCC deposi-tions from solutions with different NH4)2Ce(NO3)5 contents: 1 - 0.03M; 2 - 0.04M; 3 - 0.05M; 4 - 0.07M; 5 - 0.1M.


299

equal, whereas the Nyquist diagrams reveal remark-able differences. The largest semi-circle belongs to the sample coated in 0.05 M (NH4)2Ce(NO3)5 solution. Furthermore, its log|Z| - log(f) curve at 10 mHz also has the highest value, revealing that this coating possesses the best protection ability. The phase shift / log(f) curves reveal that there are two overlapped maxima, revealing presence of two, almost indistinguishable superficial layers on the substrates.

The anodic curves also partially confirm the results, obtained by EIS spectra. Fig. 3 depicts anodic curves, acquired after 24 hours of exposition to model corrosive medium of samples, coated in solutions with different (NH4)2Ce(NO3)5 contents.

Although the fact that the curve of the sample, pre-pared by 0.05 M (NH4)2Ce(NO3)5 deposition solution shows the best barrier ability, it also reveals clear features of pitting corrosion. The potential oscillations in -375 to -250 mV resemble the dynamic pitting nucleation/repas-sivation state, and the further swift rise reveals a stable pitting growth [49]. Nevertheless, the anodic current densities of this sample, (together with those, prepared by 0.1 M of the Ce-salt) stay at relatively two orders of magnitude lower than all of the rest.

Influence of oxidant addition on the features of the CeCC

The hydrogen peroxide is described as accelerator of the deposition process [50]. In order to investigate the impact of the oxidant addition, entire group of samples was submitted to CeCC deposition at equal conditions, as

follows: The depositions were performed for 5 minutes at -2 mA/cm2. The basic substance was 0.05 M in the conversion bath solution.

During the deposition at galvanostatic regime at -2 mA/cm2, the equipment was continuously measuring the potentials versus the Ag/AgCl reference electrode and their evolution within the deposition process. The obtained potential/time diagrams are shown in Fig. 4. There, after its initial immediate fall down to almost -1.650 V, the potential reverses its values reaching about -1.300 to -1.350 V. The initial potential drop is related to the current spent for hydrogen evolution on the metallic surface. This process appears due to both of the reduc-ing role of the cathodic current, and the generally acidic character of the deposition solution.

The subsequent reversion of the potential relays to removal of the H2 bubbles from the metallic surface. Probably, the reason for this removal and the reversion of the potential is the so called “cathodic dissolution of the aluminium” [51, 52]. The most extended continua-tion is observable for the worst samples – with 100 and 50 ml/l H2O2 (175 s for curve 1, and 125 s for curve 2). Curves 3 and 4 of the samples with more uniform, dense and homogeneous Ce-coatings achieve maxima after 10 - 15 s.

After reaching maxima, the potentials start to drop gradually again for all curves. This phenomenon is related either to a gradual growth of uniform coatings (curves 3 and 4), or to occupation of the metallic surface by Ce-containing agglomerates. For the best coating, (10 ml/l H2O2), this gradual drop of potential continued

Fig. 3. Anodic polarization curves of AA2024 substrate prepared by CeCC depositions from solutions with different NH4)2Ce(NO3)5 contents: 1 - 0.03M; 2 - 0.04M; 3 - 0.05M; 4 - 0.07M; 5 - 0.1M.

Fig. 4. Chronopotentiometric curves obtained during the galvanostatic deposition of CeCC coatings with different additions of H2O2 to the coating solution: 1 – 100ml/l H2O2; 2 - 50 – 10ml/l H2O2; 3 – 25ml/l H2O2; 4 - 10ml/l H2O2.


300

from the 50th (point (a)) to 200th (point (b)) second after the beginning of the deposition. During this time, the potential drops with about 200 mV. Taking into account that the deposition is performed in galvanostatic regime (-2 mA/cm2) and applying the Ohm law, it was found that the resistance of the deposited film increases with 100 Ω cm2 for a second. In other words, for the entire period of film growth (between points (a) and (b)) the total resistance of the best film (curve 4) raises up to 750 kΩ cm2. Consequently, the increment of the H2O2 as OH- - provider in the deposition solutions favors the dissolution of the underlying aluminum, hindering the formation of uniform and homogeneous CeCC.

Fig. 5 represents cathodic (a) and anodic (b) polari-zation curves recorded after 24 hours of exposition to the model corrosive medium of four specimens coated at different H2O2 additions.

The linear voltammogams, resemble distinguish-able features of the respective specimens. All the four samples could be conditionally divided into “better” and “worse”. The former have lower content of the oxidant (10 and 25 ml/l of 30% H2O2), whereas the latter were prepared with 50 and 100 ml/l of 30% H2O2 content, respectively. The anodic curve of the sample prepared by 50 ml/l. H2O2 shows a lack of whatever passive region. Both the cathodic and anodic curves of the sample with 100 ml/l H2O2 content stay at lower current densities, than those of the sample coated at 50 ml/l H2O2 addition. Nevertheless, the respective anodic curve (sample with 100 ml H2O2) has shorter region of passivity (from -750 to -550 mV vs. Ag/AgCl) than the curves of the samples

with lower H2O2 additions. According Bethencourt et al. [53], the shorter passivity regions indicate lower strength against pitting nucleation.

Between the “better” samples, the cathodic curve of the sample prepared at 25 ml/l H2O2 stays at lower current densities, than this of the sample prepared at 10 ml/l content of the oxidant. However, the respective anodic curves stay at the same current densities, reveal-ing very similar barrier ability. Furthermore, the curve of the sample with 10 ml/l oxidant addition has relatively larger passivity region (Fig. 5b, curve 4). From these relatively equivocal features of the polarization curves, it can be concluded that the optimal addition of peroxide should be in the range of 10 to 25 ml of 30 % H2O2 for a liter of conversion bath.

Influence of deposition current on the features of the CeCC

For investigation of the influence of the current ap-plied, electrodepositions at -5 mA/cm2 were performed on other set of specimens. The rest conditions were as described in the previous section (alkaline etching of the substrates, 0.05 M (NH4)2Ce(NO3)5 and 10, 25, 50 and 100 ml/l H2O2). In that manner, the coatings obtained by these conditions could be compared with the previous ones (e.g. prepared at - 2 mA/cm2).

As could be seen from Fig. 6, the curves obtained during the electrodeposition at -5 mA/cm2 do not possess the indicative slope for deposition of uniform, homoge-neous layer described in the previous section.

The barrier abilities of the respective samples were additionally evaluated by electrochemical measurements

Fig. 5. Cathodic (a) and anodic (b) polarization curves recorded after 24 hours of exposition to the model corrosive medium of four specimens coated at different H2O2 additions: 1 – 100ml/l H2O2; 2 – 50 ml/l H2O2; 3 – 25ml/l H2O2; 4 - 10ml/l H2O2.


301

after 24 hours of exposition to the model corrosive me-dium, as well. Fig. 7 shows EIS – spectra of the samples with films deposited at - 5 mA/cm2.

Any significant differences among the shapes of the spectra were not registered in the respective Bode plots. However, when the Nyquist plots are compared, clear differences among the radia of the respective spectra is observable. The coating, prepared at 25 ml/l of H2O2 obviously excels the rest two, being almost identical to those of the sample, prepared at lower oxidant content and current density. The similarity between the spectra of the specimens, prepared at -2 mA/cm2 and 10 ml/l. of

H2O2 (Fig. 2b, curve 3) and at -5 mA/cm2 and 25 ml/l. H2O2 (Fig. 7b, curve 2) reveal compensation between the reductive effect of the cathodic current applied during the deposition, and the addition of H2O2 as an oxidant. In other words, at 0.05 M concentration of the cerium compound, in both cases, identical coatings could be obtained. The former approach of lower current and oxidant addition is preferable by economic point of view. Indeed, the visual inspection of the coatings ob-tained at the higher deposition current proved formation of rather less uniform deposits. The samples revealed rough and grain-formed agglomerates, when the higher current was used. The unsatisfying homogeneity of the coatings favors the access of corrosive species to the metallic surface, resulting in appearance of Warburg impedance tails [54] (Fig. 7b, curves 2 and 3). Curve 1 in the same figure does not possess such a tail and its radius is slightly larger than this of curve 3. Both these features of the former curve are indications of hampered corrosion products inside the defects of the coating, caus-ing obstruction for the access of corrosive species to the substrate surface. The relation between the presence of corrosion products and the change of the shape of the EIS spectra is described elsewhere [55, 56].

Fig. 8 reveals that there are lower differences among the polarization curves of the samples prepared with different H2O2 additions, compared to those in Fig. 5.

This phenomenon could be easily explained, because the deposition current plays a role of reducer. As a result,

Fig. 6. Chronopotentiometric curves obtained during gal-vanostatic deposition of CeCC at - 5 mA/cm2 and different additions of H2O2 to the coatins solution: 1 – 100 ml/l H2O2; 2 - 50 ml/l H2O2; 3 – 25 ml/l H2O2; 4 - 10 ml/l H2O2.

Fig. 7. Electrochemical impedance spectra, acquired after 24 hours of exposition to the model corrosive medium of four specimens coated at -5 mA/cm2 at different H2O2 addition: a – Bode plot; b-Nyquist plot 1 - 50 ml/l H2O2; 2 – 25ml/l H2O2;

3 - 10ml/l H2O2


302

at higher current densities (- 5 mA/cm2), the influence of the H2O2 as oxidant is less notable. The supply of electrons towards the specimen (by the electric cur-rent) leads to deactivation of the peroxide by obtaining of OH- ions [3] and acceleration of oxygen reduction. Both these processes result in supplemental alkalisation of the medium near the substrate surface and accelera-tion of Ce-precipitation [3, 57]. However, the obtained Ce(OH)3/Ce(OH)4 precipitates do not form any coating layer, but rather conjunction of clusters. The acceleration of the deposition process at higher current (e.g. - 5 mA/cm2) disturbs the gradual growth of dense crystalline

structures, promoting formation of agglomerates, similar to the described in previous works [58].

Morphological observationsTo follow how the surfaces of the metallic speci-

mens change after CeCC electrodeposition, and what is the level of coverage by the coatings, several SEM observations combined by EDX elemental analyses were performed. The SEM photographs (Fig. 9, positions a, c,) reveal that the entire surface is covered by the CeCC. In addition, larger oval hills are clearly distinguishable on the surface. Probably these oval hills are formed as

Fig. 8. Cathodic (a) and anodic (b) polarization curves recorded after 24 hours of exposition to the model corrosive medium of four specimens coated at different H2O2 additions: 1 - 50 ml/l H2O2; 2 – 25ml/l H2O2; 3 - 10ml/l H2O2.

Fig. 9. SEM images (a, c) and EDX point analyses (b, d) of bright and dark zones of CeCC deposited from 0.05 M (NH4)2Ce(NO3)5 with 10ml/l H2O2 at -2 mA/cm2


303

consequence of preferential deposition on the locations of the intermetallics, as was observed in [22].

In order to clarify the real composition of the samples and the distribution of the elements on their surfaces, EDX – point analyses were executed during the SEM observations. The elemental distributions of the most important chemical elements: Al, Ce, Cu, and oxygen were monitored. These elements were selected because the aluminium could be presented not only from the metallic substrate, but also to compose corro-sion products, in form of Al(OH)3, etc. The cerium was selected because this element together with the oxygen are the basic components of the coating. The copper was also selected to be monitored, because it should reveal whether the S-phase intermetalics are preferable loca-tions of deposition, and is there a copper re-deposition

as evidence of corrosion during the deposition. In order to clarify the compositions of the bright and dark zones, quantitative point analyses were performed on them. Fig. 9 shows two point analyses on a bright (position c) and dark (position d) areas of a CeCC coating, elec-trodeposited from 0.05 M (NH4)2Ce(NO3)5 with 10ml/l. H2O2 at -2 mA/cm2.

After the execution of these analyses it was conclud-ed that the bright areas are almost completely composed by Ce-oxides/hydroxides, whereas in the darker zones the Ce-content is much lower. This fact reveals that ei-ther the coating in these zones is covered or mixed with Al-containing corrosion products, or it is thin enough to allow detection of the substrate material. The latter case is possible because the depth of focus is higher than the CeCC thickness [59]. Table 2 summarizes the content of the monitored elements in the bright (Fig. 9a) and dark (Fig. 9b) zones of the specimen.

As could be seen in Table 2, that the atomic con-centration Ce is almost three times higher than this of the aluminium. Surprisingly, there is unexpectedly low content of copper. This fact reveals that the superficial intermetallics are completely covered by rather thick Ce-layer. In the dark zone, more than 62 atomic units belong to Al, and only traces of Ce are represented, as can be seen in Table 2.

Comparing the tabular data for the dark and bright zones, it could be concluded that the coatings are not enough homogeneous, but they are rather represented by mixture of Ce-oxides/hydroxides, surrounded by Al-corrosion products deposited on the coating, or involved

Table 2. Elemental composition determined by the point analysis of the bright and dark zone points of the coating

Fig. 10. EDX map data of CeCC deposited at 0.05 M (NH4)2Ce(NO3)5 with 10ml/l H2O2 at -2 mA/cm2.


304

inside the CeCC-layer. Indeed, the map analysis (Fig. 10) reveals that the difference of the compositions, registered by the point analysis (Fig. 9) evinces that the Al-presence is a result of involution of corrosion products in the coating structure.

In addition, clear features of uniform copper re-deposition, as a result of the CeCC deposition procedures and/or preliminary treatment are observable in Fig. 10. This phenomenon is well described in the literature [57, 60, 61].

CONCLUSIONS

The CeCC depositions and the subsequent complex elucidation of the data obtained by the electrochemi-cal and morphological studies provided the following conclusions:

For the investigated concentration range, from 0.03 to 0.1 M (NH4)2Ce(NO3)5, the optimal concentration is about 0.05 M. The optimal content of peroxide was determined to be in the range of 10 and 25 ml of 30 % H2O2 for liter of deposition solution.

It is established that the application of the lower current e.g. - 2 mA/cm2 is preferable for deposition of dense and uniform coatings. The higher current density promoted formation of friable agglomerates.

Compensation was registered between the detrimen-tal effects of the higher H2O2 content and the stronger current applied. The coating, deposited at -5mA/cm2 at 25 ml/l of H2O2 was almost identical to those, deposited at - 2 mA/cm2 and 10ml/l of H2O2 content.

By the chronopotentiometrical measurements during the deposition it was concluded that the efficient dura-tion of the deposition of the optimal coating is about 200 seconds, any further continuation should not affect the coating characteristics, being just time wasting.

The complete coverage by the coating observed by SEM evinced that the application of cathodic currents results in the uniformity of the CeCC obtained.

Akcnowledgements

The authors gratefully acknowledge the financial support of project BG 051PO001-3.3.06-0038. Dr. G. Pelaez Lourido is acknowledged for the opportunity for the international collaboration activities.

REFERENCES

1. E. A. Starke, Jr., and J. T. Staley, Application of Modern Aluminum. Alloys to Aircraft, Prog. Aerospace Sci., 32, 1996, 131-172.

2. AIRBUS A380, Airplane characteristics, Issue 30.03.05 (2005), www.content.airbusworld.com/SITES/Techni-cal.../AC_A380.pdf

3. M. J. O’Keefe, S. Geng, S. Joshi, Cerium-Based Conver-sion Coatings as Alternatives to Hex Chrome; Rare-earth compounds as providers resistance against corrosion for aluminum alloys in military applications, Metalfinish-ing, 2007, 25 – 28.

4. S. Geng, S. Joshi, W. Pinc, W.G. Fahrenholtz,(1)M.J. O’Keefe, T.J. O’Keefe, P. Yu, Influence of processing parameters on cerium based conversion coatings, Pro-ceed “TRI – Service – 2007” - corrosion conference, accessible via: https://www.corrdefense.org/Techni-cal%20Papers/Influence%20of%20Processing%20Parameters%20on%20Cerium%20Based%20Conver-sion%20Coatings.pdf

5. B. Hindin, Corrosion protection of aluminium alloys by corrosion prevention compounds as measured by electrochemical techniques, accessible via: https://www.corrdefense.org/Academia%20Government%20and%20Industry/XVIII%20-%20HINDIN%20-%20Corrosion%20Protection%20of%20Aluminum%20Alloys%20by.pdf

6. P. Ostash, I. M. Andreiko, Yu. V. Holovatyuk, O. I. Se-menets, Effect of corrosive media on the fatigue life of degraded D16- and V95-type aluminum alloys, Mater. Sci., 44, 2008, 672 – 682.

7. J. A. Rodríguez-Martínez, A. Rusinek, A. Arias, Thermo-Viscoplastic behaviour of AA2024 aluminium sheets subjected to low velocity perforation at different tem-peratures, accessible via:

http://e-archivo.uc3m.es/bitstream/10016/11941/1/rodri-guez_thermoTWS2011ps.pdf

8. Y. Komatsu, New Pretreatment and Painting Technology for All Aluminum Automotive Body, paper no. 910787, SAE Technical Paper Series-International Congress/AIRP Meeting (Warrendale, PA: SAE International, 1991).

9. http://www.sbfships.com.au/10. http://www.aluminiumleader.com/en/around/transport/

ship


305

11. http://www.advancedrobotic.com/blog/cnc-routers/cost-savings-in-the-manufacture-of-aluminium-ships

12. D. Eyres, “Ship construction”, 6th ed. 2007, Linacre House, pp. 50–52, ISBN 10:0-75-068070-9

13. http://nation.time.com/2011/07/05/u-s-navys-brand-new-aluminum-ship-foiled-by-seawater/

14. M. L. Zheludkevich, I.M. Salvado M.G. Ferreira, Sol-gel coatings for corrosion protection of metals, J. Mater. Chem., 15, 2005, 5099 - 5111.

15. G. Tsaneva, V. Kozhukharov, S. Kozhukharova, M. Ivanova, J. Gerwann, M. Schem, T. Schmidt, Func-tional nanocomposite coatings for corrosion protection of aluminium alloy and steel, J. Univ. Chem. Technol. Met. (Sofia), 43, 2, 2008, 231 – 238.

16. S. A. Kulinich, A. S. Akhtar, On Conversion Coating Treatments to Replace Chromating for Al Alloys: Re-cent Developments and Possible Future Directions, Russian Jour. Non-Ferrous Met., 53, 2012, 176-203.

17. D. Balgude, A. Sabnis, Sol–gel derived hybrid coatings as an environment friendly surface treatment for cor-rosion protection of metals and their alloys, J. Sol-Gel Sci. Technol. DOI 10.1007/s10971-012-2838-z

18. Directive 2004/107/EC of the European Parliament and of the Council of 15 December 2004 relating to arsenic, cadmium, mercury, nickel and polycyclic aromatic hydrocarbons in ambient air. Official Jour-nal of the European Communities L 23, 26.1.2005, p. 3–16, Special edition in Bulgarian: Chapter 15 Volume 13 P. 124-137.

19. EU Directive 2002/95/EC “Restriction of Hazardous Substances in Electrical and Electronic Equipment” (RoHS directive 2002), www.broadcom.com/docs/ и www.chem.agilent.com/

20. U.S. Department of Health and Human Services, Public Health Service, Agency of Toxic Substances and Disease Registry (2008) Toxicological profile for Chromium, www.atsdr.cdc.gov/toxprofiles/tp7.pdf.

21. U.S. Environmental Protection Agency Washington, DC, August (1998) Toxicological review of hexavalent chromium. http://www.epa.gov/iris/toxreviews/0144tr.pdf

22. J. E. Pernas, S. Kozhukharov, E. Matter, A.A. Salve, M. Machkova, Influence of the oxidation state of Ce-ions on the inhibition of AA2024 alloy corrosion in a model corrosive medium, J. Univ. Chem. Technol. Met. (Sofia), 47, 3, 2012, 311 - 318.

23. E.A. Matter, S. Kozhukharov, M. Machkova, V. Kozhu-

kharov, Electrochemical studies on the corrosion inhibition of AA2024 aluminium alloy by rare earth ammonium nitrates in 3.5 % NaCl solutions, Mater. Corros., 63, 2012, 1-7; DOI: 10.1002/maco.201106349

24. E. A. Matter, S. Kozhukharov, M. Machkova, V. Kozhukharov, Comparison between the inhibition efficiencies of Ce(III) and Ce(IV) ammonium nitrates against corrosion of AA2024 aluminum alloy in solu-tions of low chloride concentration, Corr. Sci., 62, 2012, 22 - 33.

25. M. Machkova, E. A. Matter, S. Kozhukharov, V. Kozhukharov, Effect of the anionic part of various Ce(III) salts on the corrosion inhibition efficiency of AA2024 aluminium alloy, Corr. Sci., 69, 2013, 396-405.

26. A. de Frutos, M. A. Arenas, Y. Liu, P. Skeldon, G.E. Thompson, J. de Damborenea, A. Conde, Influence of pre-treatments in cerium conversion treatment of AA2024-T3 and 7075-T6 alloys, Surf. Coat. Technol. 202, 2008, 3797-3807.

27. A. Pardo, S. Feliu, M. C. Merino, R. Arrabal, E. Matykina, The effect of cerium and lanthanum sur-face treatments on early stages of oxidation of A361 aluminium alloy at high temperature, Appl. Surf. Sci., 254, 2, 2007, 586-595.

28. E. Stoyanova, R. Andreeva, D. Guergova, D. Stoychev, Formation of cerium-containing conversion coatings on aluminium and its alloys as alternative to chromate (Cr6+) conversion coatings, accessible via:

http://mech-ing.com/journal/Archive/2011/8/95_Dimi-tar%20Stoichev.pdf

29. D. K. Heller, W. G. Fahrenholtz, M. J. O’Keefe, The effect of post-treatment time and temperature on cerium-based conversion coatings on Al 2024-T3, Corros. Sci., 52, 2010, 360-368.

30. B. F. Rivera, B. Y. Johnson, M.O’Keefe, W.G. Faren-holz, Deposition and characterization of cerium oxide conversion coatings on aluminum alloy 7075-T6, Surf. Coat. Technol., 176, 2004, 349-356.

31. M. A. Arenas, J. J. de Damborenea, Generación de capas de conversión con elementos de tierras raras sobre acero galvanizado Rev. Metal, Extr.-Vol., 2005, 433-437.

32. L.E.M. Palomino, I.V. Aoki, H.G. de Melo, Micro-structural and electrochemical characterization of Ce conversion layers formed on Al alloy 2024-T3 covered with Cu-rich smut, Electrochim. Acta, 51, 2006, 5943


306

– 5953.33. A. Conde, M. A. Arenas, A. de Frutos, J. de Dambore-

nea, Effective corrosion protection of 8090 alloy by cerium conversion coatings, Electrochim. Acta, 53, 2008, 7760-7768.

34. W. Pinc, S . Geng, M. O’Keefe, W. Fahrenholtz, T. O’Keefe, Effects of acid and alkaline based surface preparations on spray deposited cerium based conver-sion coatings on Al 2024-T3, Appl. Surf. Sci, 255, 2009, 4061-4065.

35. K. Brunelli, F. Bisaglia, J. Kovac, M. Magrini, M. Da-bala, Effects of cathodic electrodeposition parameters of cerium oxide film on the corrosion resistance of the 2024 Al alloy, Mater. Corros., 60, 7, 2009, 514 - 520.

36. B. Jegdic, L. Živcovic, J. Popic, J. Bajat, V. MiŠkovic-Stankovic, Electrochemical methods for corrosion testing of Ce-based coating prepared on AA6060 alloy by dip immersion method, J. Serb. Chem. Soc., 78, 2013, 1-21.

37. D. Guergova, E. Stoyanova, D. Stoychev, I. Avra-mova, G. Atanasova, P. Stefanov, Corrosion stability of stainless steel, modified electrochemically with Ce2O3–CeO2 films, in 3.5% NaCl media, Bulgarian Chem. Commun. 43, 1, 2011, 150-157.

38. D. Nickolova, E. Stoyanova, D. Stoychev, I. Avra-mova, P. Stefanov, Protective effect of alumina and ceria oxide layers electrodeposited on stainless steel in sulfuric acid media, Surf. Coat. Technol., 202, 2008, 1876-1888.

39. E. Stoyanova, D. Guergova, D. Stoychev, I. Avramova, P. Stefanov, Passivity of OC 404 steel modified elec-trochemically with CeO2-Ce2O3 oxide layers in sulfuric acid media, Electrochim. Acta 55, 2010, 1725 - 1732.

40. D. Guergova, E. Stoyanova, D. Stoychev, G. Atanasova, I. Avramova, P. Stefanov. Influence of calcination of SSOC 4004 with alumina or ceria layers on their pas-sive state in different acid media. Bulgarian Chem. Commun., 40, 2008, 227-232.

41. H. Hasannejad, M. Aliofkhazraei, A. Shanaghi, T. Shah-rabi, A. R. Sabour, Nanostructural and electrochemical characteristics of cerium oxide thin films deposited on AA5083-H321 aluminum alloy substrates by dip im-mersion and sol-gel methods, Thin Solid Films, 517, 17, 2009, 4792-4799.

42. accessible via: http://www.splav.kharkov.com/en/e_mat_start.php?name_id=1438

43. accessible via: http://www.metalsnab.com/en/news/

iso_cert.php44. J. Vogelsang, The new standards for EIS measurements

on coatings. Short course on EIS, proceeds. 6–7 No-vember, 2008, Lisbon (Portugal).

45. M. Machkova, E. A. Matter, S. Kozhukharov, V. Kozhukharov, Effect of the anionic part of various Ce(III) salts on the corrosion inhibition efficiency of AA2024 aluminium alloy, Corr. Sci. 69, 2013, 396 – 405.

46. E. A. Matter, S. V. Kozhukharov, M. S. Machkova, Effect of preliminary treatment on the superficial morphology and the corrosion behaviour of AA2024 aluminum alloy, Bulgarian Chem. Comm., 43, 1, 2011, 23-30.

47. D. K. Heller, W. G. Fahrenholtz, M. J. O’Keefe Effect of Phosphate Source on Post-Treatment of Cerium-Based Conversion Coatings on Al 2024-T3, J. Electrochem. Soc., 156, 11, 2009, C400-C406.

48. H. Zhang, Y. Zuo, The improvement of corrosion resistance of Ce conversion films on aluminum alloy by phosphate post-treatment, Appl. Surf. Sci., 254, 2008, 4930-4935.

49. K. Srinivasa Rao, K. Prasad Rao, Pitting Corrosion of heat-treatable aluminium alloys and welds: a review, Trans Indian Inst. Met., 57, 2004, 593-610.

50. F.H. Scholes, C. Soste, A. E. Huges S. G. Hardin, P. R. Curtis, The role of hydrogen peroxide in the deposition of cerium-based conversion coatings, Appl. Surf. Sci., 253, 2006, 1770-1780.

51. M. Mokaddem, P. Volovitch, F. Rechou, R. Oltra, K. Ogle, The anodic and cathodic dissolution of Al and Al-Cu-Mg alloy, Electrochim. Acta, 55, 2010, 3779 -3786.

52. K. Ogle, M. Serdechnova, M. Mokaddem, P. Volovitch, The cathodic dissolution of Al, Al2Cu, and Al alloys, Electrochim. Acta, 56, 2011, 1711 – 1718.

53. M. Bethencourt, F.J. Botana, M.J. Cano, M. Marcos, J.M. Sánchez-Amaya, L. González-Rovira, Behaviour of the alloy AA2017 in aqueous solutions of NaCl. Part I: Corrosion mechanisms, Corros. Sci. 51, 2009, 518-524.

54. M. Garcia-Heraz, A. Jimenez-Moralez, B. Casal, J. Galvan, S. Radzki, A.Villegas, Preparation and electro-chemical study of cerium-silica sol-gel films, J. Alloys Compd 380, 2004, 219 - 224.

55. A. A. Salve, S. Kozhukharov, J.E. Pernas, E. Matter, M. Machkova, A comaparive research on hybrid and hybrid nano-composite protective primary coatings for


307

AA2024 aircraft alloy, J. Univ. Chem. Technol. Met. (Sofia), 47, 3, 2012, 319-326.

56. S. Kozhukharov, V. Kozhukharov, M. Schem, M. Aslan, M. Wittmar, A. Wittmar, M. Veith, Protective ability of hybrid nano-composite coatings with cerium sulphate as inhibitor against corrosion of AA2024 aluminium alloy, Prog. Org. Coat., 73, 2012, 95-103.

57. K. A. Yasakau, M. L. Zheludkevich, S. V. Lamaka and M. G. S, Ferreira, Mechanism of Corrosion Inhibition of AA2024 by Rare-Earth Compounds, J. Phys. Chem. B, 110, 2006, 5515-5528.

58. E. Matter, S. Kozhukharov, M. Machkova, SEM & EDS determination of the impact of inhibitor containing cor-rosive media over the AA2024 superficial morphology,

Ann. Proceeds. “Angel Kanchev” University of Rousse (Bulgaria), 50, 9.1, 2011, 60-64.

59. P. E. López, J-B. Carda Castelló, E. C. Cordoncillo, Esmaltes y Pigmentos Ceramicos, Faenza Edirtice Iberica, S. L. Catellon (Spain) 2001, pp. 287-300.

60. M. Bethencourt, F.J. Botana, J.J. Calvino, M. Marcos, M.A. Rodriguez-Chacón, Lanthanide compounds as environmentally-friendly corrosion inhibitors of alumin-ium alloys: a review, Corros. Sci., 40, 1998, 1803-1819.

61. Y. Liu, M.A. Arenas, S.J. Garcia-Vergara, T. Hashi-moto, P. Skeldon, G.E.Thompson, H. Habazaki, P. Bailey, T.C.Q. Noakes, Behaviour of copper during alkaline corrosion of Al-Cu alloys, Corros. Sci., 50, 2008, 1475-1480.


308


STATIONARY AND PULSE ELECTRODEPOSITION OF CoNi AND CoNiCu COATINGS

K. Ignatova1, Y. Marcheva2

1University of Chemical Technology and Metallurgy 8 Kl. Ohridski, 1756 Sofia, Bulgaria2Technical University – Sofia, Sofia 1000, Bulgaria

ABSTRACT

The kinetics of stationary and pulse potentiostatic electrodeposition of Co, Ni and Cu as well as the possibility for their co-deposition in alloys CoNi and CoNiCu from slightly acidic citrate electrolyte was investigated. The morphologies and the elemental composition of coatings were studied using SEM and EDSA analysis respectively. It was established that at pulse deposition of CoNi the coatings were with smooth surface, crystalline structure with rounded crystals and average size about 200-300 nm at frequency 500 Hz. In the case of triple alloy CoNiCu deposition, with the increase of frequency up to 1 000 Hz a finer nanosized structure was formed, with the Ni content up to 27 %. The Co content in the triple alloy was not influenced substantially by the pulse frequencies and was about 71-76 %, while copper content decreased from 8 % to less than 5 %. The X-Ray analysis indicated that copper, cobalt, and nickel crystallize in cubic lattice (fcc) in all studied alloy coatings. Besides the cubic phases of the three metals, the presence of cobalt-containing phase with hexagonal crystal lattice (h.p.c.) was ascertained.

Keywords: alloy electrodeposition, pulse potential modes, voltammetry, morphology, elemental and phase com-position of coathings.

Received 23 November 2012Accepted 15 May 2013

INTRODUCTION

In recent years an increasing interest exists towards electrodeposition of Co and CoNi-alloyed [1-7] and multi-layer coatings [6-8]. The particular interest to-wards nano-sized Co alloys is due to their increasing application in magnetosensor technologies and mag-netoelectronics where miniaturization of items is the underlying purpose [9-14]. Due to their high hardness, wear resistance, endurance and corrosion resistance, the cobalt alloys are widely used in medicine, nuclear-power systems, chemical- and oil industry [12, 15, 16]. The hardness of nano-sized coatings increases with the reciprocal of the square root of crystals grain size ac-cording to the Hall-Petch dependency [17].

The СоNi coatings are deposited mainly from sul-phate [18] and citrate [1,6,7,19] electrolytes. However, the literature does not provide detailed data about the kinetics and the deposition conditions of these alloys. The citrate electrolyte is used in recent years because of its ability to serve as a buffer, to form complexes, and to

add coating luster, thus avoiding the need of introduction of special organic additives [7]. The difficulties in using this electrolyte come from its stability. It was found [21], that the stability of citrate electrolyte for deposition of CuNi coating can be controlled by modifying the pH. It decreases upon reaching рН levels bellow 4, which corresponds to complexes CuHCit, Cu2Cit2

-2 and NiНCit. The stable electrolyte corresponds to рН = 5-6.

The phase content of CoNi changes depending on the deposition conditions [2, 3, 25]. The alloys, depos-ited at more positive potentials contain ε-phase with hexagonal close-packed (hcp) lattice and α-phase with cubic lattice (fcс), the proportion between phases re-mains constant [25]. The CoNi alloys, deposited at more negative potentials contain pure cobalt and mixture of α- and ε-phases. Anomalous galvanostatic deposition in glucinate bath of CoNi [21] and potentiostatic deposition of CoNi and CoNiCu in citrate bath [25] were performed. It was established formation of CoNi solid solution with hexagonal close packed lattice [21] and solid solution of NiCoCu with face centered cubic lattice [25].

K. Ignatova, Y. Marcheva

309

It was established [7] that alloying the CoCu system with low amounts of Ni, can improve the properties of the thin films. The low nickel percentages in CoCu coat-ings can favor the segregations of small ferromagnetic particles and increase of the magnetoresistance; decrease the stress in the copper/ferromagnetic interface and improvement in the corrosion resistance of the depos-its [7]. The possibility of preparing CoNiCu granular magnetoresistive coatings is proposed as alternative to Cu-Co-Ni/Cu multilayers preparation [26].

The possibility of simultaneous CoNiCu was tested in sulphate-citrate bath [24] with copper incorporated in CoNi deposits in amounts 5 to 60 mass %. It would be interest-ing also to study the possibility for codeposition of ternary CoCuNi alloys from citrate solutions in pulse conditions.

The subject of the present article is to study the ability for codeposition of Cu, Ni, and Co in CoNi and CoNiCu coatings from citrate electrolyte using station-ary and pulse potentiostatic electrodeposition, as well as the morphology, elemental and phase composition of these alloys.

EXPERIMENTAL

The experimental studies were carried out in three-electrode cell with 150 dm3 total volume at room temperature (20oC±1oC) with disk-shaped cathode from pure electrolytic copper (surface area 1cm2) and Pt plate anode. The anode surface was more than 30 times larger than that of the cathode. Saturated calomel electrode (SCE) was used as reference electrode.

The study was carried out in slightly acidic cit-rate bath with content: CuSO4 from 4 to 12 g dm-3; CoSO4.7H2O from 60 to 90 g dm-3; NiSO4 from 40 to 60 g dm-3; Na citrate 50 g dm-3.

The рН=5.3-5.5 of electrolyte was measured using рН-meter and adjusted with NaOH and Н2SO4. The ki-netics of deposition was studied by means of the method of linear and cyclic voltammetry (CV) with potentioscan type Wenking (Germany), enabling varying the sweep rate from 0 to 150 mVs-1. The electrodeposition of cobalt coatings was carried out both in stationary potentiostatic and pulse mode with rectangular potentiostatic pulses, with varying the applied potential (the polarizations in pulse mode, resp.) and the pulse frequencies.

The pulse deposition of coatings was carried out through potentiostatic pulses of rectangular shape of

potential. For the purpose a pulse generator was used connected to the input of an especially designed po-tentiostat connected to the three-electrode cell. The average values of ΔĒ (calculated as difference between the potential at current and the equilibrium potential) and the average current Iav were measured respectively by means of digital voltmeter with high input voltage, and milliammeter. The amplitude values of polariza-tion ΔΕp were measured using an oscilloscope. The theoretical relation between the average (ΔĒ) and the amplitude (ΔΕp) values of polarization in poten-tiostatic mode with rectangular shape of the pulses is:

pEE ∆=∆ .θ , where zp

p

τττ

θ+

=;

τp – time of pulses, and

τz – time of pauses between pulses. The relations

avE I∆ − and p avE I∆ − were determined for all pulse frequencies

1( , , )p zf Hz TT

τ τ= = + , 5,0=θ . The morphology and the elemental composition

were investigated using SEM and Energy Dispersive Spectral Analysis (EDSA) respectively through equip-ment of Oxford Instruments, JSM-6390- Jeol.

The phases presented in the deposited coatings were identified using X-Ray analysis. For the purpose, a vertical automatic powder diffractometer Philips PW 1050 was used with secondary graphite monochromator operating with Cu Kα radiation and scintillation counter. The diffraction curves were recorded in angular interval from 10 to 100 degrees 2θ with step 0.04 degrees 2θ and exposure 1 sec.


Deposition of Cu, Co and NiIn order to get information on the nature of polari-

zation in deposition of Cu, Co and Ni, the polarization dependencies in the corresponding electrolytes were obtained with varying the concentration of the basic salts in the solution (Fig. 1, a-c) on copper electrode at potential sweep rate 30 mV s-1.

In the electrolytes both for copper and nickel depo-


310

sition (Fig.1а and Fig.1с), plateaus of the current are formed that increase with the increase of concentration of salts in the solution. In the case of cobalt deposition of (Fig.1,b), instead of plateau - a pronounced cathode peak of the current is reached that increases with the increase of concentration of cobalt and it is shifted towards more negative values of potential. The increase of current with

the concentration in the three electrolytes suggests that a part of the polarization is related to diffusion limita-tions. Evidence of this are the dependencies (Fig.2, a-c) indicating the impact of the potential sweep rate on po-larization characteristics in electrolytes for Cu, Co and Ni. In all electrolytes the increase of the potential sweep rate results in increase of the current corresponding to

0,0 0,5 1,0 1,5 2,00

10

20

30

40

50

3

2

1

i,mA

cm-2

-E(SCE),V

1 4 g dm-3CuSO4

2 8 g dm-3CuSO4

3 12 g dm-3CuSO4

0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4 1,6 1,8 2,00

10

20

30

40

50

3 2 1

3

21

i,mA

cm-2

-E(SCE),V

1 25 g dm-3 CoSO47H2

2 60 g dm-3 CoSO4H2O3 90 g dm-3 CoSO4H2O

0,0 0,5 1,0 1,5 2,00

5

10

15

20

25

30

35

40

32

1

i, m

A cm

-2

-E(SCE),V

1 - 40 g dm-3 NiSO4

2 - 50 g dm3 NiSO4

3 - 60 g dm-3 NiSO4

Fig. 1. Effect of concentration of components in citrate electrolyte for deposition of Cu (a), Co (b), and Ni (c); 50 g dm-3 Na citratе, v= 30 mVs-1.

0,0 0,5 1,0 1,5 2,00

5

10

15

20

25

30

35

40

3

2

1

i c,mA

cm-2

-E(SCE),V

v, mV s-1: 1- 30 (Cu) 2- 50 3- 80

0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4 1,6 1,8 2,00

10

20

30

40

50

60

3

2

1i c,mA

cm-2

-E(SCE),V

v, mV s-1: 1- 30 (Co) 2- 50 3- 80

0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4 1,6 1,80

2

4

6

8

10

12

14

16

18

20

22

24

3

2

1

i c, m

A cm

-2

-E(SCE),V

v, mV s-1: 1- 30 (Ni) 2- 50 3- 80

Fig. 2. Voltammerograms in citrate electrolyte at different potential sweep rates, v: (1) 30 mVs-1; (2) 50 mV s-1; (3) 80 mVs-1 in electrolytes for deposition of Cu, Co, and Ni with content: (а) 12 g dm-3 CuSO4; (b) 60 g dm-3 CoSO4.7H2O; (с) 60 g dm-3 NiSO4; (а-с) 50 g dm-3 Na citratе.


311

the cathode maximum either in the plateau in deposition of Cu (Fig. 2a) and Ni (Fig. 2c), or in the cathode peak in deposition of Со (Fig. 2b).

In electrolyte for Cu deposition at lower potential sweep rates (Fig. 2,a), two pronounced plateaus in the curves of current occur, corresponding to the two subse-quent stages of electron transfer. Between them a chemi-cal stage exists also, which is due to the reaction of copper disproportion. The described mechanism was proved by the authors in amine-nitrate electrolyte [22] and is also proved by other authors in other electrolytes [23].

There is no pronounced plateau of limiting current in the electrolyte for nickel deposition (Fig. 2c) at all potential sweep rates. After 2-3 repeats of cyclic voltam-mograms (Fig. 3), when the electrode is coated with nickel, the plateau gradually disappear and the impact of the potential sweep rate is slight. This indicates that the larger share in the polarization of nickel deposition has an activation nature. The absence of anode branch in the cyclic curves (Fig. 3) is also in favor of that claim.

From the kinetics results obtained we can conclude that the electroactive forms in the citrate electrolyte are complex compounds of copper, cobalt and nickel ions with specific mechanism of deposition, with predomi-nance of the diffusion polarization for Cu and Co and activation one for Ni. In the investigated electrolyte composition the deposition of cobalt initiates at –1.0 V, of nickel initiates at –1.1 V and of copper at –1.25 V.

Codeposition of metals in CoNi and CoNiCu alloys The possibility for codeposition of Co, Cu, and Ni

in alloy coatings can be evaluated through comparison of the polarization dependencies for their separately deposition to the dependency at the simultaneous pres-ence of the corresponding metals in the solution, as it is made in Fig. 4 (a) for CoNi, and in Fig. 4b for CoNiCu.

The curve taken at simultaneous presence of Ni and Co (Fig. 4a) is almost a sum of deposition curves of copper and cobalt. The data show that the deposition potentials of Co and Ni on Cu substrate are very close and almost coincide in the range from -1.2 V to -1.4V (SCE), just in the way as the values of their standard potentials are close to each other: Е0(Co2+/Co) = -0,277 V; Е0(Ni2+/Ni) = -0,250 V. Most likely the closeness of the deposition potentials is due to the closeness of the corresponding stability constants of the citrate com-plexes of both metals.

0,0 0,5 1,0 1,5 2,0

-5

0

5

10

15

20

25

30

v=60 mV s-1

2*

1*

1

i, m

A cm

-2

-E,V(SCE)

0,4 m Ni +50 g dm-3 Na citrat

Fig. 3 Cyclic voltammograms (v= 30 mV s-1) in electrolyte for Ni with content: 60 g dm-3 NiSO4; 50 g dm-3 Na citrate: 1, 1* - first, resp. second repeat in straight direction; 2* - second and subsequent repeats in reverse direction.

0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4 1,6 1,8 2,0 2,20

10

20

30

40

50

60

70

3

2

1

i c, m

A cm

-2

-E(SCE),V

1- Ni2- Co3- NiCo

0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4 1,6 1,8 2,0 2,20

10

20

30

40

50

60

70

4

3

2

1

i c, m

A cm

-2

-E(SCE),V

1- Cu2- Ni3- Co4- CuNiCo

Fig. 4. Comparison between the curves of individual deposition and co-deposition of alloys (v = 30 mVs-1): (a) CoNi: Ni (1), Co (2), and CoNi (3); and for: (b) CoNiCu: Cu (1), Ni (2), Co (3), and CoNiCu (4) in electrolytes with content: 4 g dm-3 CuSO4; 90 g dm-3 CoSO4 7H2O; 60 g dm-3 NiSO4; 50 g dm-3 Na citratе.


312

Fig. 4,b shows the curves of deposition of each of the metals (Cu, Ni, Co) and the curve of their codeposi-tion from electrolyte, containing the three components. The deposition potentials of metals converge in the range -1.3 V to -1.42 (SCE), which could be the range of potentials for codeposition of CоNiCu alloy coatings in stationary mode.

Deposition of CoNi and CoNiCu alloys in potentio-static pulse mode

The effect of the pulse frequency (varied in the range from 100 Hz to 10 000Hz, θ=0,5) on the dependencies

avE I∆ − (Fig. 5, curves 1-6) and p avE I∆ − (Fig. 5, curves 1* - 6*) was studied. The data for CoNi deposition are similar to those for CoNiCu and this is the reason the dependencies for the first alloy only to be presented (Fig. 5).

The ohmic drop of solution (between the Lugin capillary and the working electrode), as well as the polarization, used to charge the double electric layer (DEL), are also included in the values of the average and amplitude polarization. While the first becomes negligible with increasing the polarization, the second

is significant and is about one third of the total polariza-tion, especially at frequencies above 1 000 Hz. At such frequencies the time of pulses is of the order of the time for loading DEL ( 0,5 )p msτ << . Although the deposition at frequencies higher than 1 000 Hz is ineffective due to the mentioned reasons, CoNi coatings were deposited at frequencies 5 000 Hz and even 10 000 Hz in order to determine the effect of the pulse mode on the structure of alloys in such non-standard conditions.

With the increase of frequency of pulses from 100 Hz to 10 000 Hz, the average polarization (Fig. 5, curves 1-6) is lower than that in stationary mode (Fig. 5, curve 7), which is explained with the possibility to decrease the diffusion limitations due to relaxation of the diffusion gradient during pauses (effect, more significant at lower frequencies). In the same time the increase of frequency of pulses results in almost double increase of the am-plitude values of polarization (Fig. 5, curves 1*-6*) in the frequency range investigated. Since the polarization is a measure for saturation, it is reasonable to expect achieving more fine crystal structure of coatings when higher frequencies of pulses are applied.

Morphology and elemental content of the alloysNiCo Coatings of CoNi and CoNiCu were deposited in a

wide range of variation of pulse frequencies (from 500 Hz to 10 000 Hz for CoNi and from 500 Hz to 1 000 Hz for CoNiCu), θ=0,5. With the aim the stationary and pulse plate conditions to be comparative, an average polarization corresponded to an average current Iav (S = 1cm2), was applied (for CoNi Iav = 35 mA, for CoNiCu Iav = 45 mA)

The SEM images of Ni-Co stationary deposited coat-ings are shown in Fig. 6а, b and pulse deposited coatings (Fig. 6c-f). The elemental content (in wt.%) according to the data from EDSA analysis is given in the capture to the figure. The pulse deposited coatings have low Ni content (up to 12 mass %) compared to the stationary deposited coatings. At high frequencies presence of oxides could be found in the coating.

The application of pulse mode especially the higher pulse frequencies results in increased share of the more oval crystals compared to the needle-shaped ones, their size decreases (reaching 200-400 nm) and the surface smoothes as a whole. The increase of pulse frequency above 1 000 Hz results in deposition of alloys with high

Fig. 5. Relation between the average cathode polarization and average current p avE I∆ − (curves 1-4), and between the amplitude polarization and average current p avE I∆ − (curves 1* - 4* ) at pulse frequencies: 100 Hz (1, 1*); 500 Hz (2, 2*); 1,000 Hz (3, 3*); 2,500 Hz (4, 4*); in electrolyte for CoNi: 90 g dm-3 CоSO4 7H2O; 60 g dm-3 NiSO4; 50 g dm-3 Na citrate; curve 5 (dashed): polarization dependence in stationary mode in the same composition.

0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4 1,6 1,8 2,0 2,2 2,4 2,60

10

20

30

40

50

60

70

80

6*

5*

4*

3*

2*1*7

65

43

2

1

aver

age

curre

nt, m

A

average (1-6) and amlitude (1*-6*) polarization,V

1-6 average polarization1*-6* amplitude polariration 7--- stationary regime


313

Fig. 6. SEM and EDSA of CoNi coatings deposited in stationary mode (a, b) and in pulse mode (c-f) at pulse frequencies: 500 Hz (c); 1000 Hz (d); 5000 Hz (e); 10000 Hz (f) in composition: 4 gd m-3 CuSO4; 90 g dm-3 CоSO4.7H2O; 60 g dm-3 NiSO4; 50 g dm-3 Na citrate. The applied deposition polarizations correspond to the same average current Iav =45 mA (S=1 cm2) in the polarization curves from Fig. 5. Coating composition: (a) 76 % Co-21 % Ni (3 % O); (b) 76 % Co-21 % Ni (3 % O); (c) 88 % Co -12 % Ni ; (d) 96 % Co- 4 % Ni; (e) 89 % Co-5 %Ni (6 % O); (f) 90 % Co-5 % Ni (5 % O).

Fig. 7. SEM and EDSA of CoNiCu coatings deposited in pulse mode with pulse frequencies: 500 Hz (a); 1 000 Hz (b) in electrolyte: 4 g dm-3 CuSO4; 90 g dm-3 CоSO4.7H2O; 60 g dm-3 NiSO4; 50 g dm-3 Na-citrate. The applied deposition polarizations correspond to the same average current Iav = 35 mA (S = 1 cm2).

a) a*)

b) b*)

a) b) c)

d) e) f)


314

cobalt content (about 90 mass % Со). The blurred SEM images at higher frequencies can be explained with ob-taining nanosized structure and smoothing the surface.

CuNiCo Triple Cu-Ni-Co alloy coatings were pulse de-

posited, the data of the morphology and the elemental content are presented in Fig. 7.

It was found that the CuNiCo alloy coating deposited with pulse frequency 500 Hz have larger crystals and are more inhomogeneous compared to CoNi coatings (Fig. 6с and Fig. 7a). The increase of pulse frequency up to 1 000 Hz results in obtaining rounded, more uniform, and finer crystals of average diameter 400-500 nm (Fig.7,b). Moreover, the percentage of nickel in the alloys increases (up to 27 wt % Ni). The percentage of cobalt, however, does not change significantly and remains about 71-76 wt %, while the copper content decreases from 8 wt % to less than 5 wt % at the potentials applied.

Phase content of NiCo and CuNiCo coatings All diffractograms are characterized with relatively

wide peaks due to their fine nanosized structure. This makes difficult the interpretation of data since some-times the identification of a certain phase, and mostly

the phases of cobalt, is made upon a single peak only. The X-Ray analysis indicates that CoNi stationary

coatings are characterized with the highest and narrow-est reflections (Fig. 8 b) corresponding to their larger crystal structure in comparison to the pulse deposited coatings (Fig. 8 с-g). This result well agrees with the SEM observations of the same samples. In all alloy coatings copper, cobalt, and nickel crystallize in cubic lattice (f.c.c.). Only in the triple CoNiCu coating depos-ited at pulse frequency 1 000 Hz (Fig. 8 а), it was found presence of cobalt both with face-centered cubic crystal lattice (f.c.c.) and hexagonal crystal lattice (h.p.c.). The clearly pronounced spectrum of copper is mainly due to the fact that the layers are relatively thin (about 3 µm) and are deposited on copper.

CONCLUSIONS

It was found that at potentiostatic pulse deposition and higher frequency coatings with smooth surface and finer crystals with rounded shape were produced. The average grain size was about 200-300 nm for CoNi and 500 nm for CoNiCu. The optimal frequency of pulses for obtaining fine-crystal nanosized alloys of maximum homogeneity in both cases is 1 000 Hz. In CoNi coatings the metals crystallize in cubic lattice (f.c.c.). In CoNiCu coatings except phases of cubic copper and nickel, it was found presence of cobalt both with face centered cubic lattice (f.c.c.) and hexagonal crystal lattice (f.c.c.) at pulse frequency 1 000 Hz.

REFERENCES

1. B.Y.C.Wu, C.A.Schuh, P.J.Ferreira, Nanostructured NiCo alloys with tailorable grain size and twin den-sity, Metallurgical and Materials Transactions A, 36A, 2005, 1927-1936.

2. J. Horkans, I.C.H.Chang, P.C.Andricacos, E. J. Podlaha, Determination of Partial Current of CuNi and CuCo alloys: Electrodeposition using Rotating Ring-disk electrodes, J.Electrochem. Soc., 138, 1991, 411.

3. A. Bastos, S.Zaefferer, D.Raabe, Three-dimensional EBSD study on the relationship between triple junc-tions and columnar grains in electrodeposited Co–Ni, Advanced Materials Research, 15-17, 2006, 953-958.

4. A. Bastos, S. Zaefferer, D. Raabe, Characterization of the Microstructure and Texture of Nanostr.

Fig. 8. Х-Ray diffraction curves of CoNiCu coatings (a) and CoNi coatings (b-f) deposited in stationary (b) and pulse mode at frequencies of pulses: 500 Hz (c); 1 000 Hz (d); 5 000 Hz (e); 10 000 Hz (f) in electrolytes of content: 4 gdm-3 CuSO4; 90 g dm-3 CоSO4.7H2O; 60 g dm-3; 50 g dm-3 Na citrate. The applied deposition polarizations cor-respond to average currents Iav = 35 mA (a) and Iav =45 mA (b-f); S=1 cm2.


315

Electrodeposited NiCo by use of Electron Backscatter Difractom., J. Microscopy, 230, 2008, 487-498.

5. A. Hiroki, I.Masanobu, Investigation of solid states, Papers of Technical Meeting on Magnetic, IEE, Japan, 50-63, 2005, 49-52.

6. C.P. Singh, R.K. Shukia, S.C.Srivastava, Electrodepo-sition of NiCoCu alloy from a citrate bath, Surface Technology, 16, 1982, 219-225.

7. S. Pané, E. Gómez, E. Vallés, Magnetoresistive granu-lar CuCoNi coatings prepared by electrodeposition, J. Electroanalytical chem., 596, 2006, 87-94.

8. M. Uhlemann, A. Gebert, M. Herrich, A. Krause, A. Cziraki, L. Shulz, Electrochemical deposition and modification of Cu/Co-Cu multilayers, Electrochim. Acta, 48, 2003, 3005-3011.

9. E. Chassaing, Effect of organic Additives on the Elecrodeposition and the Magnetoresistance of CuCo Multilayers, J. Electrochem. Soc., 148, 2001, C690-694.

10. J.Garcia-Torres, L.Péter, A.Révész, L.Pogány, I.Bakonyi, GMR in electrodeposited films, Thin Solid Films, 517, 2009, 6081-6090.

11. S.K. Gupta, M.G. Gartley, XRD and VSM Analysis of nanostructured CuCo Alloys, J. CPDS International Centre for Diffraction Data, 1999, 688-697.

12. G.R.Pattanaik, D.K.Pandya, S.C. Kashyap, Preparation of CuCo Alloy Thin Films on nSi by Galvanostatic DC Electrodeposition, J.Electrochem Soc, 149, 2002, C363-369.

13. G.Binasch, P. Grünberg, F.Saurenbach, W.Zinn, Enhanced magnetoresistance in layered magnetic structures with antiferromagnetic interlayer ex-change, Phys. Rev. B, 39, 1989, 4828-4830.

14. M. Gu, Initial stages of electrocrystalization of CoCu alloys on GCE from the Co rich electrolytes, Еlectrochimica Acta, 52, 2007, 4443-4448.

15. D. Landolt, XPS and AES analysis of passiv films on Fe25Cr-X model alloys, J. Electrochem. Soc.,

149, 2002, S9. 16. I. Flis-Kabulska, Investigation of thick Co films elecro-

deposited on Au substrates, J. Appl. Electrochem., 36, 2006, 131-137.

17. N.J.Petch, The cleavage strength of polycrystals, J. Iron Steel Inst., 174, 1953, 25-28.

18. S.M.S.I.Dulal, E.A.Charles, S.Roy, Electrodeposition and composition of CoNi(Cu)/Cu, Electrochimica Acta, 49, 2004, 2041-2049.

19. A.M. El-Sherik, Syntesis of bulk nanocrystalline Ni by pulsed electrodeposition, Ph.D.Thesis, Queen’s University, Kingston, Canada, 1993.

20. Q. Huang, D. Davis, E.J. Podlaha, Simulation of Pulsed Electrodeposition for GMR of FeCoNiCu/Cu, J. Appl. Electrochem., 36, 2006, 871-882.

21. T. A.Green, A.E.Russel, S.Roy, The Development of a Stable Citrate Electrolyte for the Electrodeposition of CuNi Alloys, J.Electrochem. Soc., 145, 1998, 875-881.

22. L.Petkov, K.Ignatova, Investigation of the pro-cesses and mechanism of electrodeposition of Cu from ammonium nitrate electrolyte by the method of cyclic voltamperometry, Bulg. Chem. Comm., 41, 2009, 39-45.

23. D. Grujicic, B.Pesic, Reaction and nucleation mech-anisms of Cu electrodeposition, Electrochimica Acta, 50, 2005, 4426-4443.

24. K. Ignatova, Y. Marcheva, Еlectrodeposition and structure of Cо coatings (CoCu, NiCo and CoNiCu) in potentiostatic and pulse potential modes, SED-Sofia, Nov. 2012, Bulg. Chem. Commun, 2013, (in press).

25. G. Nabiyouni, W. Schwarzacher, Z. Rolik, I. Bakonyi, GMR and Magnetic Properties of elec-trodeposited CoNiCu/Cu multilayers, J. Magn. Mater, 253, 2002, 77-85.

26. E. Gómez, S. Pané, E. Vallés, Electrodeposition of CoNi and CoNiCu system in Sulphate Citrate Medium, Electrochimica Acta, 51, 2005, 146.


316


CRYSTALLIZATION IN THE THREE-COMPONENT SYSTEMS Rb2SeO4 - MeSeO4 - H2O (Me = Mg, Ni, Cu) at 25°C

V. Karadjova


ABSTRACT

The solubility in the three-component systems Rb2SeO4-MeSeO4-H2O (Me = Mg, Ni, Cu) is studied at 25°C by the method of isothermal decrease of supersaturation. It has been established that double compounds, Rb2Me(SeO4)2·6H2O (Me = Mg, Ni, Cu), crystallize from the ternary solutions within wide concentration ranges. The X-ray diffraction data reveal that the title compounds crystallize in the monoclinic space group P21/c (C2h

5) with lattice parameters: Rb2Mg(SeO4)2·6H2O – a = 6.353(3) Å, b = 12.674(5) Å, c = 9.412(3) Å, b = 105.22(2)º , V= 731.2(3) Å3; Rb2Ni(SeO4)2·6H2O – a = 6.349(1) 9.324(2) Å, b = 12.617(2) Å, c = 9.324(2) Å, b = 105.35(1)º, V = 720.3(1) Å3; Rb2Cu(SeO4)2·6H2O – a = 6.361(1) Å, b = 12.569(2) Å, c = 9.414(2) Å, b = 104.68(1), V = 728.1(2) Å3. The infrared spectra are discussed with respect to the normal vibrations of the selenate ions and water molecules. The unit-cell group theoretical treatment of the double salts is presented. Infrared spectroscopy experiments show that the effective spectroscopic symmetry of the selenate ions is close to C3v. Comparatively strong hydrogen bonds are formed in the rubidium Tutton selenates as deduced from both the wavenumbers of the stretching modes of the water molecules and the water librations due to the strong proton acceptor strength of the selenate ions.

Keywords: rubidium magnesium selenate hexahydrate, rubidium nickel selenate hexahydrate, rubidium cop-per selenate hexahydrate, solubility diagrams, X-ray powder diffraction, infrared spectra.


INTRODUCTION

The rubidium double selenates belong to a large number of isomorphous compounds with a general for-mula Me2

+Me2+(XO4)2∙6H2O (Me+ = K, NH4+, Rb, Cs;

Me2+ = Mg, Mn, Co, Ni, Cu, Zn; X = S, Se) known as Tutton salts. They crystallize in the monoclinic space group P21/c (C2h

5) with two formula units in the unit-cell. The crystal structures of these compounds are built up from isolated octahedra, [Me2+(H2O)6], (three crystal-lographically different water molecules are coordinated to the Me2+ ions) and tetrahedra XO4. The polyhedra are linked by hydrogen bonds. All atoms, except the divalent metal ions, which lie at centre of inversion Ci, are located at general positions C1. Recently, the crystal structures of some rubidium Tutton compounds have been reported in [1-3]. As an example the crystal

structures of Rb2Co(SeO4)2∙6H2O is presented in Fig. 1.In this paper we present the results on the study of

the solubility in the three-component systems Rb2SeO4-MeSeO4-H2O (Me = Mg, Ni, Cu) at 25 °C. There are no available data in the literature about the phase diagrams of the above systems. The solubility data for the systems Rb2SeO4-MeSeO4-H2O (Me = Co, Zn) are reported only in the literature [4, 5]. The Tutton salts are characterized by means of both the infrared spec-troscopy and the X-ray powder diffraction methods. For comparison the infrared spectra of Rb2Co(SeO4)2∙6H2O and Rb2Zn(SeO4)2∙6H2O are also presented and dis-cussed. A practical point of studying is that the Tutton compounds could be considered as proton conductors due to the existence of comparatively strong hydrogen bonds determined by the strong proton acceptor capa-bilities of the selenate ions.

V. Karadjova

317

EXPERIMENTAL

Rb2SeO4, MeSeO4·6H2O (Me = Mg, Co, Ni, Zn) and CuSeO4·5H2O were prepared by neutralization of the respective carbonate and hydroxide carbonates with dilute selenic acid solutions at 60-70°C. Then the solutions were filtered, concentrated at 50-60°C, and cooled to room temperature. The crystals were filtered, washed with alcohol and dried in air. The solubility in the three-component systems was studied by the method of isothermal decrease of supersaturation. The equilib-rium between the liquid and solid phases was reached in about 20 hours.

About 3-4 g from the liquid and wet solid phases were taken from the systems and analyzed as follows: the metal ion contents were determined complexono-metrically at pH 9.5-10 using eriochrome black as in-dicator (magnesium ions) and at pH 5.5-6 using xylenol orange as indicator (copper and nickel ions); the sum of the selenate ions was determined after precipitation with Pb(NO3)2 solutions and the concentration of the excess Pb2+ ions was determined complexonometri-cally using xylenol orange as indicator; the content of the rubidium selenate was calculated by difference [G. Schwarzenbach, H. Flaschka, Die komplexometrische Titration, Ferdinand Enke Verlag, Stuttgart, 1965]. The composition of the solid phases was identified by means

of X-ray diffraction and infrared spectroscopy methods as well. The cobalt and zinc analogues are prepared according to [4, 5]. All reagents used were of reagent grade quality (Merck).

The infrared spectra were recorded on a Bruker model

Rb2Co(SeO4)2∙6H2O

Fig. 1. Crystal structure of Rb2Co(SeO4)2·6H2O (structural data according to [1]; plane ba).

Fig. 2. Solubility diagrams of the three-component systems Rb2SeO4 - MeSeO4 - H2O (Me = Mg, Ni, Cu) at 25°C.


318

IFS 25 Fourier transform interferometer (resolution < 2 cm-1) at ambient temperature using KBr discs as ma-trices. Ion exchange or other reactions with KBr have not been observed.

The X-ray powder diffraction spectra were collected within the range from 10° to 50° 2θ with a step 0.02° 2θ and counting time 35 s/step on Bruker D8 Advance diffractometer with Cu Kα radiation and LynxEye de-tector. The lattice parameters of the double salts were calculated using the program ITO and refined with the program LSUCR.


Solubility diagrams of the three-component systems Rb2SeO4-MeSeO4-H2O (Me = Mg, Ni, Cu) at 25ºC

The solubility diagrams of the above systems are presented in Fig. 2 (the respective experimental data are summarized in Tables 1-3). It is seen from Fig. 2 that the simple salts Rb2SeO4, MgSeO4∙6H2O, NiSeO4∙6H2O and CuSeO4∙5H2O crystallize within very narrow con-centration ranges, whereas the rubidium double selenates crystallize within wide concentration ranges, thus indi-cating that strong complex formation processes occur in the ternary solutions. For example, the magnesium compound crystallizes from solutions containing 1.01 mass% magnesium selenate and 60.94 mass% rubidium selenate up to solutions containing 32.21 mass% mag-

nesium selenate and 5.28 mass % rubidium selenate; the copper compound - from solutions containing 1.43 mass % copper selenate and 61.02 mass % rubidium selenate up to solutions containing 18.80 mass % cop-per selenate and 10.56 mass % rubidium selenate; the nickel compound - from solutions containing 1.06 mass % nickel selenate and 60.74 mass % rubidium selenate up to solutions containing 27.54 mass % nickel selenate and 6.94 mass % rubidium selenate.

Liquid phase mass%

Wet solid phase mass%

Composition of the solid phases

Rb2SeO4 MgSeO4 Rb2SeO4 MgSeO4

62.02 - - Rb2SeO4 60.94 1.01 71.58 10.36 Rb2SeO4+Rb2Mg(SeO4)2·6H2O 54.96 1.47 53.66 23.87 Rb2Mg(SeO4)2·6H2O 44.81 1.66 51.42 21.96 “ − “ 37.47 2.24 50.94 23.59 “ − “ 29.06 4.47 49.06 23.84 “ − “ 21.51 8.02 49.83 25.71 “ − “ 9.49 15.23 46.17 26.74 “ − “ 6.50 25.02 47.25 28.04 “ − “ 6.28 32.21 23.44 42.10 MgSeO4+Rb2Mg(SeO4)2·6H2O

- 35.43 MgSeO4·6H2O

Table 1. Solubility in the Rb2SeO4 - MgSeO4 - H2O system at 25°C.

10 20 30 40 50

Rb2Ni(SeO4)2·6H2O

Rb2Cu(SeO4)2·6H2O

2Θ (degree)

Inten

sity

Rb2Mg(SeO4)2·6H2O

Fig. 3. X-ray powder diffraction patterns of Rb2Me(SeO4)2

·6H2O (Me = Mg, Ni, Cu).

V. Karadjova

319

X-ray powder diffraction data of the rubidium Tutton compounds

The X-ray powder diffraction patterns of the rubidium Tutton compounds are shown in Fig. 3 (d-spacing, hkl and relative intensities are given in Table 4). The double salts form monoclinic crystals (SG P21/a (C2h

5)). The calculated unit-cell parameters have values of: Rb2Mg(SeO4)2·6H2O – a = 6.353(3) Å, b = 12.674(5) Å, c = 9.412(3) Å, b = 105.22(2)º, V = 731.2(3) Å3; Rb2Ni(SeO4)2·6H2O – a = 6.349(1) 9.324(2) Å, b = 12.617(2) Å, c = 9.324(2) Å, b = 105.35(1)º, V = 720.3(1) Å3; Rb2Cu(SeO4)2·6H2O – a = 6.361(1) Å,

b = 12.569(2) Å, c = 9.414(2) Å, b = 104.68(1), V = 728.1(2) Å3. Our results coincide well with those deter-mined from single crystal X-ray diffraction data [1, 2].

Infrared spectra of rubidium Tutton compoundsThe free tetrahedral ions (XO4

n-) under perfect Td symmetry exhibit four internal vibrations: n1(A1), the symmetric X–O stretching modes, n2(E), the symmetric XO4 bending modes, n3(F2) and n4(F2), the asymmetric stretching and bending modes, respectively. The normal vibrations of the free tetrahedral ions in aqueous solu-tions are reported to appear, as follows: for the selenate

Liquid phase mass%



Rb2SeO4 NiSeO4 Rb2SeO4 NiSeO4

62.02 - Rb2SeO4 60.74 1.06 69.17 16.42 Rb2SeO4+Rb2Ni(SeO4)2·6H2O 55.83 0.92 51.45 24.08 Rb2Ni(SeO4)2·6H2O 47.65 1.58 50.19 28.85 “ − “ 39.17 1.63 48.76 27.69 “ − “ 29.31 0.84 45.58 25.27 “ − “ 22.90 2.07 45.72 28.93 “ − “ 13.36 5.64 42.07 27.45 “ − “ 8.67 13.79 42.93 29.24 “ − “ 7.21 20.30 39.62 29.78 “ − “ 6.97 27.54 28.55 44.69 NiSeO4 + Rb2Ni(SeO4)2·6H2O

26.47 NiSeO4·6H2O

Table 2. Solubility in the Rb2SeO4 - NiSeO4 - H2O system at 25°C.

Liquid phase mass%



Rb2SeO4 CuSeO4 Rb2SeO4 CuSeO4

62.02 Rb2SeO4 61.02 1.43 62.17 20.76 Rb2SeO4+Rb2Cu(SeO4)2·6H2O 58.10 0.55 54.63 21.74 Rb2Cu(SeO4)2·6H2O 52.04 0.96 52.41 20.08 “ − “ 49.79 0.98 51.78 21.93 “ − “ 41.09 2.14 49.21 19.57 “ − “ 33.92 2.24 46.89 19.86 “ − “ 28.24 4.48 46.05 21.71 “ − “ 17.07 8.58 46.76 24.68 “ − “ 11.14 14.24 43.41 25.19 “ − “ 10.56 18.80 18.73 46.57 CuSeO4+Rb2Cu(SeO4)2·6H2O 6.34 17.45 1.34 61.37 CuSeO4·5H2O

15.72 “ − “

Table 3. Solubility in the Rb2SeO4 - CuSeO4 - H2O system at 25°C.


320

ions – n1 = 833 cm-1, n2 = 335 cm-1, n3 = 875 cm-1, n4= 432 cm-1 [6]. On going into solid state, the normal modes of the XO4

n- ions are expected to shift to higher or lower frequencies.

The static field (related to the symmetry of the site on which the XO4

n- ions are situated) will cause a re-moval of both the doubly degenerate n2 modes and the triply degenerate n3 and n4 modes. Since the tetrahedral ions in the structures of Tutton compounds occupy site symmetry C1, two bands for n2 (2A) and three bands for n3 and n4 (3A), respectively, are expected to appear in the vibrational spectra as predicted from the site group

Fig. 4. Correlation diagram between Td point symmetry, C1 site symmetry and C2h factor group symmetry (SeO4

2- ions).

Rb2Mg(SeO4)2·6H2O Rb2Ni(SeO4)2·6H2O Rb2Cu(SeO4)2·6H2O

dobs, Å hkl I/I0 dobs, Å hkl I/I0 dobs, Å hkl I/I0 7.380 011 100 7.320 011 42 7.376 011 31 6.340 020 <5 6.125 100 <5 6.153 100 6 6.131 100 5 5.513 110 9 5.527 110 8 5.524 110 5 4.495 002 12 5.172 021 <5 5.305 -111 <5 4.395 120 14 4.553 002 10 4.543 002 5 4.264 111 82 4.397 120 13 4.403 -120 <5 4.193 -102 100 4.309 111 63 4.288 111 21 3.811 031 83 4.205 -102 100 4.216 -102 89 3.679 121 9 3.806 031 79 3.830 031 58 3.663 022 <5 3.687 022 7 3.696 121 <5 3.466 130 5 3.463 130 <5 3.479 -130 <5 3.411 -131 16 3.402 -131 14 3.424 -131 5 3.238 102 23 3.284 102 19 3.261 102 7 3.155 040 21 3.177 112 13 3.169 040 20 3.137 112 21 3.142 040 35 3.097 131 11 3.070 032 60 3.094 131 15 3.070 -211 20 2.9758 041 <5 3.072 -211 61 2.9673 -113 <5 2.9447 -113 12 2.9507 013 10 2.9444 013 7 2.9154 013 26 2.9105 122 29 2.9209 -202 9 2.8810 122 16 2.8292 -221 10 2.9004 122 <5 2.8292 -221 13 2.7660 -141 <5 2.8318 -221 6 2.7745 -141 <5 2.6443 -222 <5 2.7852 -141 <5 2.5659 132 <5 2.5862 042 <5 2.5326 -142 11 2.5288 -231 10 2.5272 -231 8 2.4741 -133 12 2.5204 -142 13 2.5171 -142 8 2.4603 -213 8 2.4560 -133 32 2.4636 -133 22 2.2710 004 11 2.2592 142 19 2.2777 231 7 2.2030 -240 <5 2.1968 240 10 2.2703 142 15 2.1476 -242 5 2.1401 -242 9 2.1984 240 7 2.0409 300 <5 2.1172 024 9 2.1901 -124 <5 1.9246 -313 <5 2.0410 300 16 2.1370 -242 11 1.8449 044 14 2.0356 -134 9 2.1025 -204 <5 1.8407 330 8 1.9200 -313 8 2.0948 060 <5

1.8787 -162 5 2.0499 -312 29 1.8298 044 12 2.0243 310 15 1.9467 250 <5 1.9204 -313 7 1.8750 -162 7

Table 4. X-ray powder diffraction data of Rb2Me(SeO4)2·6H2O (Me-Mg, Ni, Cu).

V. Karadjova

321

analysis (the non-degenerate n1 mode is activated). Ad-ditionally, the factor group analysis (C2h factor group symmetry) predicts a splitting of each species of A

symmetry into Ag + Au + Bg + Bu (related to interactions of identical oscillators, correlation field effects). Con-sequently, 18 infrared bands (9Au + 9Bu) and 18 Raman bands (9Ag + 9Bg) correspond to the normal vibrations of the tetrahedral ions. The correlation diagram between the Td point symmetry, C1 site symmetry of the selenate ions and C2h factor group symmetry is presented in Fig. 4. Unit-cell theoretical treatment for the translational lat-tice modes (Rb, Me2+, SeO4

2- and H2O) and librational lattice modes (SeO4

2- and H2O) yields: 69 modes of Ag, Au, Bg and Bu symmetry and 48 modes of Ag, Au, Bg and Bu symmetry for the translational and librational modes, respectively (see Table 5; the factor group analysis is made according to [7]).

Infrared spectroscopic investigations of the potas-sium Tutton sulfates and selenates are widely discussed in the literature [8-17 and Refs. therein], while those of the rubidium compounds are briefly commented [3].

Infrared spectra of Rb2Me(SeO4)2∙6H2O (Me = Mg, Co, Ni, Cu, Zn) in the region of 4000−400 cm-1 are shown in Figs. 5-6. Some structural and spectroscopic data are summarized in Table 6 (for comparison the data for the respective potassium and ammonium Tutton selenates are presented; the data are taken form [25]). It is readily seen that the band positions and the shape of the spectra are similar owing to the isostructureness of the double salts. The eight infrared bands expected according to the factor group analysis for the stretching modes of the se-lenate ions coalesce into three bands for each compound: Rb2Mg(SeO4)2·6H2O – 895 and 877 cm-1 (n3) and 835 cm-1 (n1); Rb2Co(SeO4)2·6H2O – 893 and 873 cm-1 (n3) and 831 cm-1 (n1); Rb2Ni(SeO4)2·6H2O – 895 and 881 cm-1 (n3) and 831 cm-1 (n1); Rb2Cu(SeO4)2·6H2O – 889 and 867 cm-1 (n3) and 835 cm-1 (n1).; Rb2Zn(SeO4)2·6H2O – 893 and 877 cm-1 (n3) and 831 cm-1 (n1).

1000 900 800 700 600 500 400

432

573623735

831

877893

Wevanumber (cm-1)

417

571613739

835

867889 434

623744

Inten

sity

582

831

895881 432

Zn

Cu

Ni

Co

582623732

831

873893

426597625709739835

895877

Mg

Fig. 5. Infrared spectra of Rb2Me(SeO4)2·6H2O (Me = Mg, Ni, Cu) in the region of 1100-350 cm-1 (normal vibrations of the SeO4

2- ions and water librations).

Species N ni

SeO4

ni

H2O

nT' nR nT Activity

Ag 45 9 6 18 12 0 R Au 45 9 6 18 12 0 R Bg 45 9 6 17 12 1 IR Bu 45 9 6 16 12 2 IR Ʃ 180 36 24 69 48 3

Table 5. Unit-cell theoretical analysis of Rb2Me(SeO4)2·6H2O (Me = Mg, Ni, Cu).


322

The distortion of the polyatomic ions at various lattice sites (as compared to the free ions in solution or gas) established by spectroscopic studies (infrared and Raman) is called an energetic distortion in order to distinguish it from the geometrical distortion revealed by structural data [18, 19]. Both the site group splitting of the asymmetric modes (Dnas) and the value of Dnmax (the difference between the highest and the lowest wave-numbered components of the stretching and bending modes, respectively) are an adequate measure for the degree of energetic distortion of the polyatomic ions, i.e. the strength of the electrostatic field at the lattice sites where these ions are located [19-21]. The small energetic distortions of the selenate ions (Dn3 for the magnesium

compound has value of 18 cm-1; for the cobalt compound – 20 cm-1; for the nickel compound – 14 cm-1; for the copper compound – 22 cm-1, and for the zinc compound – 16 cm-1) are due to the small geometric distortion of the selenate ions (for the magnesium compound Dr has value of 0.014 Å; for the cobalt compound – 0.015 Å; for the nickel compound – 0.012 Å; for the copper com-pound – 0.024 Å; for the zinc compound – 0.013Å; Dr is the distance between the longest and the shortest bond length in the selenate tetrahedra – calculations are made according to Refs. 1, 2). So, the infared spectroscopic experiments show that effective spectroscopic symmetry of the selenate ions is closed to C3v. The close values of Dn3 and Dnmax for potassium, ammonium and rubidium selenates reveal that the nature of the Me+ and Me2+ ions do not influence on the energetic distortion of the SeO4

2- ions since the latter are not coordinated to the divalent metal ions and form weak electrostatic bonds with Me+ ions (see Table 6). The bending modes of the selenate ions appear in the spectral region below 400 cm-1 (only one band in the region of 417–434 cm-1 is observed in the spectra, which is attributed to asymmetrical bending modes of the selenate ions).

The normal vibrations of the water molecules appear in the high frequency region of 3000−4500 cm-1. The three crystallographically different water molecules (C1 site symmetry) in the structures of the Tutton compounds are expected to display six infrared bands corresponding to the asymmetric and symmetric modes n3 and n1, respectively. However, due to the strong interactions of identical oscillators O−H the different normal modes overlap and as a result one broad band centered at about 3200 cm-1 is observed in the spectra of the double Tutton salts (with exception of the copper compound). Three bands correspond-ing to n2 of three crystallographically different water molecules are observed in the spectra of cobalt and zinc compounds (1704, 1640 and 1550 cm-1; 1703, 1624 and 1556 cm-1 for the cobalt and zinc salts, respectively). In the case of magnesium, nickel and copper compounds the three bands expected coalesce into two bands (1703 and 1546 cm-1;1703 and 1562 cm-1; 1708 and 1614 cm-1 for the magnesium, nickel and copper compounds, respectively) (see Fig. 6). The

1640

1550

32011704

3231

1703

Mg

1546

15621703

3200

Ni

3315

Co

16143063

3201

Cu

Inten

sity

Zn

Wavenumber (cm-1)

1708

1556

16241703

4000 3000 12002000

Fig. 6. Infrared spectra of Rb2Me(SeO4)2·6H2O (Me = Mg, Ni, Cu) in the region of the stretching and bending modes of the water molecules.

V. Karadjova

323

band positions of the stretching and bending modes indicate that comparatively strong hydrogen bonds are formed in the selenates and the hydrogen bond strengths do not depend on the Me2+ chemical nature (for the magnesium, cobalt, nickel and zinc salts). The appearance of a band at a lower frequency (3069 cm-1) in the spectrum of the copper compound evidences that stronger hydrogen bonds are formed in this salt as compared to other rubidium ones. This spectroscopic finding is owing to the stronger synergetic effect of the Cu2+ ions, i.e. to the strong Cu−OH2 interactions (increasing of the acidity of the water molecules). The formation of comparatively strong hydrogen bonds in the studied compounds is due to the strong proton ac-ceptor capacity of the selenate ions [22-24].

The water librations (rocking, twisting and wag-ging) appear in the region below 1000 cm-1 and a strong overlapping with vibrations of other entities in the structure is expected. Two types of water libra-tions for the Tutton sulfates are discussed briefly in the literature – rocking and wagging, the former observed

at higher frequencies [10]. Each type is characterized with two broad bands. The water molecules bonded to the Me2+ ions via shorter Me2+-OH2 bonds display water librations at higher frequencies as compared to those forming longer Me2+-OH2 bonds (equatorial water mol-ecules). The former Me2+-OH2 bonds are much more polarized due to the stronger synergetic effect of the Me2+ ions (stronger metal-water interactions). The mean wavenumbers for the rocking librations are reported to have values of 855 and 740 cm-1, and 770 and 680 cm-1 for the potassium and ammonium sulfates, respectively. The respective wagging modes have mean values of 570 and 441 cm-1 for the potassium compounds, and 544 and 425 cm-1 for the ammonium ones [10]. Thus, the broad bands in the spectral interval of 750-710 cm-1 are attributed to the rocking modes of the water molecules and the two bands in the region of 630-570 cm-1 to the wagging modes. The close wavenumbers of the water librations confirm the claim that hydrogen bonds of close strength are formed in the rubidium Tutton compounds.

Compounds V/n Å3

X─O Å

Dr(XO4) Å

n3 cm-1

n 3 cm-1

n1 cm-1

Dn3 cm-1

Dnmax cm-1

K2Mg(SeO4)2⋅6H2O 174 1.634 0.018 899, 877 888 835 22 64 Rb2Mg(SeO4)2⋅6H2O 182 1.632 0.014 895, 877 886 835 18 60 (NH4)2Mg(SeO4)2⋅6H2O 183 1.635 0.016 901, 875 888 835 26 66 K2Co(SeO4)2⋅6H2O 174 1.633 0.021 899, 877 888 831 22 68 Rb2Co(SeO4)2⋅6H2O 181 1.638 0.020 893, 873 883 831 20 62 (NH4)2Co(SeO4)2⋅6H2O 182 1.638 0.017 899, 873 886 833 26 66 K2Ni(SeO4)2⋅6H2O 172 1.633 0.015 896, 880 888 830 16 66 Rb2Ni(SeO4)2⋅6H2O 179 1.637 0.019 895, 881 888 831 14 64 (NH4)2Ni(SeO4)2⋅6H2O 180 1.637 0.021 899, 873 886 831 26 68 K2Cu(SeO4)2⋅6H2O 173 1.628 0.013 894, 881sh 888 837 13 57 Rb2Cu(SeO4)2⋅6H2O 181 1.636 0.017 889, 867 878 835 22 54 (NH4)2Cu(SeO4)2⋅6H2O 181 1.628 0.030 893, 882sh, 877 884 835 16 58 K2Zn(SeO4)2⋅6H2O 174 1.633 0.020 896, 879 888 833 17 63 Rb2Zn(SeO4)2⋅6H2O 181 1.631 0.011 893, 877 885 831 16 62 (NH4)2Zn(SeO4)2⋅6H2O 182 1.639 0.017 898, 875 887 833 23 65

Table 6. Some structural and spectroscopic characteristics for the XO42- ions in the neat Tutton salts

(V/n, unit-cell volumes divided by the numbers of the XO42- ions; X--O, mean values of the X-O bond

lengths; Dr(XO4), the difference between the longest and the shortest X-O bond lengths in the respective tetrahedra; Dn max, the difference between the highest and the lowest wavenumbered components of the stretches of the XO4

2- ions; the structural data are taken from Refs. [1, 2, 25]).


324

CONCLUSIONS

Tutton compounds, Rb2Me(SeO4)2·6H2O (Me = Mg, Ni, Cu), have been obtained from the three-component systems Rb2SeO4 - MeSeO4 - H2O (Me = Mg, Ni, Cu) at 25°C. The X-ray powder diffraction data reveal that the double salts are isostructural and have close lattice parameters. The analysis of the infrared spectra in the region of the normal vibrations of the selenate ions show that the effective spectroscopic symmetry of these ions is close to C3v. Both the wavenumbers of the stretching modes of the water molecules and the water librations indicate that comparatively strong hydrogen bonds are formed in the rubidium Tutton selenates due to the strong proton acceptor capacity of the selenate ions.

REFERENCES

1. A. Kirfel, A. Meents, B. Barbeir, H. Euler, New crys-tal structures, Z. Kristallogr., 218, 2003, 265-268.

2. U. Kolitsch, M. Fleck, New crystal structures, Z. Kristallogr., 217, 2003, 15-16.

3. Zd. Micka, L. Prokopova, I. Cisarova, D. Havlicek, Crystal structure, thermoanalytical properties and infrared spectra of double magnesium selenates, Collect. Czech. Chem. Commun., 61, 1996, 1295-1306.

4. T. Ojkova, D. Stoilova, D. Barkov, On the double salt formation in the systems Rb2SeO4-ZnSeO4-H2O and Cs2SeO4-ZnSeO4-H2O at 25 ºC, Monatsh. Chem., 131, 2000, 1019-1023.

5. T. Oykova, D. Mihov, P. Pavlova, Phase interaction in the systems Sodium selenate- cupper selenate-water and sodium selenate-zinck selenate-water at 25ºC, Cryst. Res. Technol., 26, 1991, 1071-1075.

6. K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordination Compounds, John Wiley & Sons, New York, 1986.

7. D.L. Rousseau, R. P. Bauman, S. P. Porto, Normal mode determination in Crystals, J. Raman Spectrosc., 10, 1981, 253- 290.

8. V. Ananthanarayanan, Studies on the vibrational spec-trum of the SO4 ion in crystalline M2’M’’(SO4)2∙6H2O (M’ = K or NH4, and M’’ = Mg, Zn, Ni, or Co): Observations on the symmetry of the sulfate ion in crystals, J. Chem. Phys., 48, 1968, 573-581.

9. R.G. Brown, S. D. Ross, Forbidden transitions in the infrared spectra of tetrahedral anions – VI. Tutton salts and other double sulfates and selenates, Spectrochim. Acta, 26A, 1970, 945-958.

10. J.A. Gampbell, D.P. Ryan, L.M. Simpson, Interoionic forces in crystal – II. Infrared spectra of SO4 groups and “octahedrally” coordinated water in some Alums, Tutton salts, and the double salts obtained by dehydrating them, Spectrochim. Acta, A26, 1970, 2351-2361.

11. S. Gupta, B. Singh, B. Khanna, The vibrational spectra of SO4

2- ions in the hydrated double sulfates: the Tutton salts, J. Mol. Struct., 112,1984, 41-49.

12. V. Ivanovski, Th. Mayerhöfer, J. Popp, Investigation of the peculiarities in the polarized reflectance spec-tra of some Tutton salt monoclinic single crystals using dispersion analysis, Vibr. Spectrosc., 44, 2007, 369-374.

13. D. Marinova, M. Georgiev, D. Stoilova, Vibrational behavior of matrix-isolated ions in Tutton com-pounds. I. Infrared spectroscopic study of NH4

+ and SO4

2- ions included in magnesium sulfates and selenates, J. Mol. Struct., 929, 2009, 67-72.

14. D. Marinova, M. Georgiev, D. Stoilova, Vibrational behavior of matrix- isolated ions in Tutton com-pounds. II. Infrared spectroscopic study of NH4

+ and SO4

2- ions included in copper sulfates and selenates, J. Mol. Struct., 938, 2009, 179-184.

15. M. Georgiev, D. Marinova, , D. Stoilova, Vibrational behavior of matrix- isolated ions in Tutton com-pounds. III. Infrared spectroscopic study of NH4

+ and SO4

2- ions included in cobalt sulfates and selenates, III, Vibr. Spectrosc., 53, 2010, 233-238.

16. D. Marinova, M. Georgiev, D. Stoilova, Vibrational behavior of matrix- isolated ions in Tutton com-pounds. IV. Infrared spectroscopic study of NH4

and SO4

2- ions included in nickel sulfates and selenates, Cryst. Res. Technol., 45, 2010, 637-642.

17. D. Marinova, M. Georgiev, D. Stoilova, Vibrational behavior of matrix- isolated ions in Tutton com-pounds. V. Infrared spectroscopic study of NH4

+ and SO4

2- ions included in zinc sulfates and selenates, Solid State Sci., 12, 2010, 765-769.

18. H.D. Lutz, E. Alici, T. Kellersohn, Single crystal Raman studies on Sr(CIO3)2 and BrO3

- and IO3- ions

matrix isolated in Sr(CIO3)2, J. Raman Spectrosc.,

V. Karadjova

325

21, 1990, 387-393. 19. H.D. Lutz, H. Haeuseler, Infrared and Raman spectra

of inorganic solids - State of the art, Trends Appl. Spectrosc., 2, 1998, 59-88.

20. H.D. Lutz, H. Christian, W. Eckers, Infrared and single crystal Raman studies of alkaline earth chlorate, bromated and iodate monohydrates and of halite ions matrix isolated in Ba(CIO3)2∙H2O type host lattice, Spectrochim. Acta A41, 1985, 637-648.

21. H.D. Lutz, T. Kellersohn, B. Müller, J. Henning, Infrared and Raman studies of halite ions matrix isolated in Sr(CIO3)2 and Sr(CIO3) 2∙H2O type host lattices, Spectrochim. Acta A44, 1988, 497-504.

22. H.D. Lutz, Bonding and structure of water molecules in solid hydrates Correlation of spectroscopic and structural data, Struct. Bond. (Berlin), 69, 1988,

97-123.23. H.D. Lutz, Structure and strength of hydrogen

bonds in inorganic salts, J. Mol. Struct., 646, 2003, 227-236.

24. M. Georgiev, D. Marinova, D. Stoilova, Infrared spectroscopic study of Tutton compounds. II. Vibrational behavior of NH4

+ ions included in K2Me’’(XO4)2∙6H2O (Me’’= Mg, Co, Ni, Cu, Zn; X = S, Se), J. Univ. Chem. Technol. Metal., 45, 2010, 195-200.

25. M. Georgiev, D. Marinova, D. Stoilova, Infrared spectroscopic study of Tutton compounds. I. Vibrational behavior of SO4

2- ions included in Me2

’Me’’(SeO4)2∙6H2O (Me’ = K, NH4; Me’’ = Mg, Co, Ni, Cu, Zn), J. Univ. Chem. Technol. Metal., (Sofia), 45, 2010, 75-82.

326


Journal of Chemical Technology and Metallurgy is a specialized scientific edition presenting original research results in the field of chemical technology and metallurgy, chemical engineering, biotechnology, indus-trial automation, environmental protection and natural sciences. The articles published in Journal of Chemical Technology and Metallurgy refer to:

Inorganic Chemistry; Organic Chemistry; Analytical Chemistry; Physical Chemistry;

Organic Synthesis and Fuels; Polymer Engineering; Textile and Leather; Cellulose, Paper and Polygraphy; Inorganic and Electrochemical Productions;

Chemical Engineering; Industrial Automation; In-formation Technology; Biotechnology; Economics and Management of Chemical and Metallurgical Industry; Sustainable Development and Environmental Protection;

Physical Metallurgy; Metallurgy of Iron and Casting; Metallurgy of Non-ferrous and Semiconducting Materi-als; Technology of Silicates; Nanomaterials.

Journal of Chemical Technology and Metallurgy publishes full-length research papers, critical and book reviews. Research papers are expected to be complete and authoritative accounts of work that has significance and interest, presented clearly and concisely. Critical reviews are commissioned by the Editor-in-Chief. Au-thors intending to offer critical reviews are invited first to contact the Editor-in-Chief.

Contributions will only be considered for publica-tion if they are relevant to the topics pointed above. Presentation and discussion should be at the level of the Journal status gained. The language can be a reason for rejection if below an acceptable level of clarity. De-tailed descriptions of the equipment used should only be given if the latter is new. Papers reporting experimental data without adequate interpretation are not acceptable. Contributions will be accepted for publication only on the recommendation of referees.

Submission of ManuscriptsManuscripts should be submitted to the Editor-in-

Chief at the following address:University of Chemical Technology and Metallurgy8 Kliment Ohridski blvd., Sofia 1756, Bulgaria,

(for JCTM)Manuscripts may be submitted electronically to e-

mail: [email protected]

GUIDE FOR AUTHORS

Manuscript preparationIt is important that the file be saved in the native

format of the wordprocessor used. All contributions should be typed, double-spaced, 12-pt font. All pages must be numbered in sequence. Most formatting codes will be removed and replaced on processing the article. In particular, do not use the wordprocessor’s options to justify text or to hyphenate words. However, do use bold face, italics, subscripts, superscripts, etc. Do not embed graphically desighed equations or tables, but prepare these using the wordprocessor’s facility. When preparing tables, if you are using a table grid, use only one grid for each individual table and not a grid for each row. If no grid is used, use tabs, not spaces, to align columns.

Essential title page informationTitle. It should be concise and informative. Titles

are often used in information-retrieval systems. Avoid abbreviations and formulae where possible.

Author names and affiliations. The names of the authors have to be clearly indicated. Present the authors’ affiliation addresses (where the actual work was done) below the names. Provide the full postal address of each affiliation, including the country name.

Corresponding author. Clearly indicate who will handle correspondence at all stages of refereeing and publication, also post-publication. Ensure that phone numbers (with country and area code) are provided in addition to the e-mail address and the complete postal address. Contact details must be kept up to date by the corresponding author.

Present/permanent address. If the author has moved since the work described in the article was done, or was visiting at the time, a “Present address” (or “Permanent address”) may be indicated as a footnote to that author’s name. The address at which the author actually did the work must be retained as the main, af-filiation address. Superscript Arabic numerals are used for such footnotes.

AbstractA concise and factual abstract is required. The ab-

stract should state briefly the purpose of the research, the principal results and major conclusions. An abstract is often presented separately from the article, so it must be able to stand alone. For this reason, References should be avoided, but if essential, then cite the author(s) and year(s). Also, nonstandard or uncommon abbreviations

327


should be avoided, but if essential they must be defined at their first mention in the abstract itself.

Article structureIntroductionState the objectives of the work and provide an

adequate background. Avoid a detailed literature survey or a summary of the results.

ExperimentalProvide sufficient details to allow the work to be

reproduced. Methods already published should be indi-cated by a reference: only relevant modifications should be described.

Results and discussionResults and the discussion provided should be clear

and concise.ConclusionsThe main conclusions of the study have to be pre-

sented in a short Conclusions section.AcknowledgementsAcknowledgements are given in a separate section

at the end of the article before the references.References should be cited in the text by an Arabic

numeral in square brackets, listed numerically and typed double-spaced on a separate sheet at the end of the paper. References should be arranged in the order in which they appear in the text. Journal titles should be abbreviated ac-cording to the latest Chemical Abstracts Service Source Index. Only articles that have been published or are in press should be included in the references. Unpublished results should be cited as such in the text. References in non-English language should be translated in English. The original language must be shown in brackets. In the reference list, the styling, punctuation and capitalization should be as follows:

For journals:S.K. Maji, A. Pal, T. Pal, A. Adak, Modeling and

fixed bed column adsorption of As(III) on laterite soil, Sep. Purif. Technol., 56, 2007, 284-290.

For books:A. Krueger, Carbon Materials and Nanotechnology,

Weinheim, WILEY-VCH Verlag GmbH & Co. KGaA, 2010.

For edited books:F.W. Cooke, J.E. Lemons, B.D. Ratner, in: B.D.

Ratner, A.S. Hoffman, F.J. Lemons (Eds.), Biomaterials Science, Academic Press, San Diego, 1996, p. 11.

For conference proceedings, symposia, etc.:E. Motta, D. Rajpathak, Z. Zdrahal, R. Roy, The

Epistemology of Scheduling Problems, F.V. Harmelen (ed.), Proceedings of the 15th European Conference on Artificial Intelligence, Lyon, France, 2002, 215-219.

When you are citing papers published in Journal of Chemical Technology and Metallurgy you should use the abbreviation of the title as follows:

J. Chem. Technol. Metall.

Formulae should be typewritten (Equation Editor) with ample space around them. The meanings of all symbols have to be given immediately after the equa-tion in which they are first used. Equations should be sequentially numbered (on the right side of the equation and in parentheses). Subscripts and superscripts should be clearly indicated. Greek letters and other non-Latin symbols should be typed using Times New Roman Gr or Symbols.

Units should be given in SI system. They should be in the form, e.g. g cm-1 rather than g/cm.

Figures should be numbered consecutively with Arabic numerals. All figures should be supplied on separate sheets. All illustrations must be readable when reduced to a width of 75 mm (single column figure) or 160 mm (double column figure). Photographs, charts and diagrams are all to be referred to as “Figures(s)” and should be numbered consecutively in the order to which they are referred. They should accompany the manuscript, but should not be included within the text. All illustrations should be clearly marked with the figure number. Figure captions should be typed double-spaced on a separate sheet. Figures should not include text. Graphic files should be in TIFF or JPG with resolution not less than 300 dpi.

Tables should be numbered consecutively with Arabic numerals in the order of appearance in the text. Each table should be given on separate sheet with a short descriptive title directly above it, with essential footnotes below.

One copy of the manuscript in English and one in Bulgarian (for Bulgarian authors) with a set of original illustrations and a disk must be submitted. The manu-script should be written with true-type font Times New Roman size 12 with double line spacing. Manuscripts of papers submitted to the journal will not be sent back to the authors.

journal - dl.uctm.edu · new techniques that are currently being developed include capillary...

Documents