739

Oxidation Communications 31, No 4, 739–757 (2008)

* For correspondence.

Hybrid AntioxidAnts

E. B. Burlakova*, E. M. Molochkina, G. a. nikiForov

N. M. Emanuel Institute of Biochemical Physics, Russian Academy of Sciences, 4 Kosygin Street, 119 334 Moscow, Russia E-mail: seren@sky.chph.ras.ru

aBstract

the article is a review of the data on synthesis and physiological activity of hybrid antioxidants. the introduction offers an explanation to the fact why in some cases it is necessary to add drug mol ecules with fragments responsible for various properties and aimed at various targets. a large group of hybrid antioxidants comprise stable nitroxyl radicals that behave as antioxidants in free-radical reactions of oxidation. the compounds of this type were synthesised extensively to form a group of antitumor agents. As a rule, the specific (antitumor) activity retained or even increased as com-pared with the initial compounds (without nitroxyl radicals); the toxicity decreased 5 to 10 times, which made it possible to apply the drug in considerably higher concen-trations. there are reported data on nitroxyl derivatives of anthracycline antibiotics, antimetabolites, alkylating agents, and the recent results on platinum complexes with nitroxyl fragments. Much attention is given to hindered phenols with ‘buoyancy’ properties, particularly, to biochemical effects making them promising agents to treat the alzheimer’s disease.

Keywords: antioxidants, synthesis, physiological activity, hybrid, the alzheimer’s disease, ‘buoyancy’ properties, oxidation.

introDuction

in 1954, the book by Professor B. n. tarusov ‘Principles of Biological Effects of radioactive Emissions’ was published1; the book made a great impression on E. B. Burlakova. the author, an outstanding soviet biophysicist, head of the Biophysics Department at the Faculty of Biology of the Moscow state university, put forward a hypothesis that the development of radiation-induced disease is associated with the induction of ramified chain reaction of oxidation of fats of cellular shells (membranes), the oxidation products are very toxic for the cell.

E. B. Burlakova had dreamed of studying the mechanism of radiation-induced disease, but nobody at the Faculty of chemistry dealt with these subjects at that time.

740

Now two areas of her scientific interests have converged – chain reactions that had always been the main subject of investigation at the chemical kinetics Department of the Faculty of chemistry (headed by n. n. semenov, nobel Prize laureate) at the Moscow state university, where E. B. Burlakova wrote her diploma work, and radiation-induced disease, that was of interest at the Biophysics Department. E. B. Burlakova was permitted to write her diploma work on biology under the guidance of academician n. M. Emanuel and Professors B. G. Dzantiev and G. B. sergeev. the aim of the work was to find out what biologically toxic products are formed upon ir-radiation of lipids. a model substrate chosen for radiolytic oxidation of fat was natural cod-liver oil. the oil was exposed to radiation, its composition was investigated, and then the oil was oxidised. the investigation results showed that even after high-dose exposures no detectable quantities of specifically new products are formed and the toxicity of the exposed oil depended only on a degree of oxidation. the higher the oxidation degree, the more the deep-oxidation products (aldehydes, ketones, peracids, etc.) and the higher the oil toxicity. at equal oxidation degrees, the toxicity of exposed and unexposed oil was equal2–4. the main irradiation effect was reduced to decom-position of natural antioxidants in oil. Generally speaking, this result (a decrease in the quantity of antioxidants upon exposure) was not unexpected: the effect had been determined previously in vitro experiments with irradiation of various fats.

However, this result interested us in view of its significance with respect to radia-tion-involved reactions. a simultaneous investigation of toxic effects of irradiation on plant and animal fats made it possible to conclude that toxicity of irradiated fat is not associated with the formation of new products of oxidation, but rather with acceleration of oxidation of exposed fat because of decomposition of natural antioxi-dants in fat. hence, an unambiguous practical conclusion was drawn: if animals could be administered with antioxidants before exposure, we could slow down processes associated with acceleration of oxidation of lipids and formation of oxidised toxic products. Therefore, we should introduce compounds that could fulfill functions of antioxidants decomposed upon exposure. We started with introduction of natural antioxidant – tocopherol and found out that we can increase the average life-span of irradiated animals. then, by analogy with works on preventing fats from oxidative decay, we introduced nontoxic synthetic antioxidants used in the food industry4. this decision was important not only because we were the first who introduced antioxi-dants to animals to protect them from irradiation (although that was a new word in radiobiology), but also because synthetic antioxidants were introduced to animals. Previously, we were sure that irrespective of a particular structure of an antioxidant (synthetic or natural), its main characteristic was its ability to react with free radicals. therefore, we believed that this ability that we had shown in model experiments can persist and manifest itself after introduction of an oxidant into lipids of animal organs. These experiments confirmed, to some extent, the Tarusov’s concept about the great role of chain (free-radical) reactions in lipids of exposed animals in the development of radiation-induced damages.

741

in 1960, n. M. Emanuel put forward the hypothesis that not only radicals of lipids, but also radicals from other biochemical components of cell (Dna, protein, polysaccharides, etc.) that are alien to normal cell and are formed upon exposure may cause multiple damages and death of the cell5.

it was shown that radicals formed upon exposures of Dna and proteins may, like lipid radicals, enter into exchange reactions with antioxidants; as a result, the free valence passes from biopolymer radical to the antioxidant molecule and forms inactive radical from the antioxidant [InH] (Refs 6–8).

r•.biol. + hin → rh + in.

in 1961, a supposition was made about the great role of free radicals alien to normal cell in the development of some other diseases and about the feasibility of inhibiting the free-radical reactions by applying synthetic inhibitors to achieve a cur-ing effect9–11. This supposition could be made only by physicochemists, first of all, by specialists in kinetics of the Semenov–Emanuel school, who understood the importance of not only (and not so much of) a change in the composition of the reaction compo-nents, but also of their physicochemical properties, i.e. when the same results may be obtained with different (in composition) components, but with the one common physicochemical property – in this case, ability to react with free radicals. Therefore, synthetic compounds of the structure other than that of natural antioxidants may be used instead of (substitute) the latter ones in reactions with free radicals.

in the 1960’s the Emanuel institute of Biochemical Physics, russian academy of Sciences, initiated studies in the new field – chemistry and biology of antioxidants. The scientists of the Institute had to solve an important task – to find out whether the biological activity of antioxidants as inhibitors of radical reactions depends on their properties. For this purpose, nontoxic different-structure antioxidants were synthesised: derivatives of hindered phenols and heterocyclic hydrocarbon hydroxy compounds12,13. the existence of homologous arrays of antioxidant derivatives made it possible to determine the structure–activity dependence and select the most efficient and least toxic compounds.

In vivo experiments, the correlation between the radioprotecting activity and antiradical properties of synthetic antioxidants was determined14,15. kinetic studies on natural antioxidants – vitamins were carried out; their constants as inhibitors of radical processes were determined16,17. in the works by khrapova et al., the chemi-luminescence method adapted to studies on bioantioxidants in lipids was used; with this method, the problems of synergism and anthagonism of synthetic and natural antioxidants were studied and the antioxidant system in membrane lipids was char-acterised as a whole18.

the above works were concerned mainly with investigating antioxidants in lipid components of cells. however, specially developed photochemiluminescence models were used to assess the antiradical activity of water-soluble natural and synthetic inhibitors; exchange reactions of this type of antioxidants with uv-induced peptide

742

radicals were studied19,20. Great attention was given to extending research concern-ing the role of bioantioxidants in the development of some or other diseases and the feasibility of using antioxidants for prophylactic or curing purposes.

It is necessary to dwell on an important property of antioxidants – the dose–effect dependence of introduced antioxidant. after introduction of antioxidant, the antioxi-dant activity (aoa) increases, then it returns to normal, and then, after a short-time aoa increasing, it decreases drastically below the normal. therefore, antioxidants may produce a curing effect by decreasing (at low doses) or increasing (at high doses) the rate of free-radical reactions.

the changes in the antioxidant activity of organs and tissues lipids in the process of carcinogenesis were studied21. the staged mechanism of changes in the antioxidant activity in the process of carcinogenesis caused by different cancinogens: benzopyr-rene, orthoaminoazotoluene, and γ-irradiation was established. at the initial stage of the carcinogen toxic effect and unduction of tumor cells, aoa decreases, then it increases, reaches the normal, and then increases above the normal at the stage of transition from diffusional to localised hyperplasia. The efficiency of synthetic antioxidants depends on their concentration and time of introduction22. At the first stage of carcinogenesis, the protective effect is caused by doses that increase aoa; at the late stages, these doses may accelerate the development of carcinogenesis and increase the number of tumors induced. at the late stages, it is necessary to introduce larger antioxidants quantities, which can cause an opposite effect – to decrease AOA (refs 23 and 24) and inhibit the process of carcinogenesis.

N. M. Emanuel and O. S. Frankfurt were the first who discovered the anticancer effect of the antioxidant dibutyloxytoluene25.

a great number of works were devoted to studying free-radical processes associ-ated with the tumor growth and the antitumor effects of antioxidants26–28. it was found out that tumor growth is associated, as a rule, with an increased level of antioxidants and only high doses of antioxidants produce the antitumor effect. in this case, anti-oxidants do not increase the aoa of organs and tissues but, on the contrary, decrease it and act as prooxidants. it should be noted that there is a general trend not only for antioxidants, but also for various antitumor agents: their efficiency is the higher, the stronger they decrease aoa (ref. 29).

Many specialists at iBcP ras studied antioxidants with respect to radiation-in-duced disease. the radioprotective effect of the compounds was in conformity with their aoa (ref. 15). similar data were obtained not only in experiments in animals, but also in model experiments with exposed solutions of Dna, proteins, and lipids30–32.

A novel field of science was commenced in the 1970’s – applications of anti-oxidants in gerontology. kinetic studies on model reactions of ageing, investigation of age-related changes in antioxidants, theoretical concepts of ageing, particular experimental studies of antioxidants as geroprotectors showed that this science field is both of theoretical and practical importance33–37. it is very strange that the idea of using antioxidants in gerontology is now put forward as a new one and the author-

743

ship is ascribed to people other than those who have been working in this field for 40 years already.

along with studies on using antioxidants for some particular diseases, extensive studies on the role of antioxidants in normal physiological processes were commenced. Palmina et al. studied the role of antioxidants in cell proliferation and showed that the factors that increase the antioxidant activity accelerate the proliferation; those that decrease it, inhibit38–40.

alesenko et al. studied the effect of antioxidants on the genetic apparatus ac-tivity41–43. the authors showed that bioantioxidants are able to affect the cell lipids composition and change the activity of lipid-dependent enzymes of synthesis and reparation of Dna and affect the activity of chromatin.

the end of 1970’s was marked by extensive studies on the role of antioxidants in the normal metabolism of cell. there was drawn the conclusion that ao are universal modifiers of composition, structure, and functional activity of membranes and that many of their effects on cell metabolism may be interpreted from these positions44,45. there was discovered the physicochemical system of regulation of cell metabolism by membranes based on interrelation between membranes lipid peroxidation (lPo), on the one hand, and changes in the composition of membrane lipids and their oxidis-ability, on the other46,47.

Proceeding from the parameters of this system, it is possible to use antioxidant to convert a cell, organ, and organism from one metabolic state to another.

in 1970’s, antioxidants found wide use in cardiology, oncology, and treatment of neurodegenerative and other classes of diseases48–50. Extensive studies on antioxidants were commenced in the field of plant growing and farming as plant growth stimulators and for preventive and curing treatment of cattle and poultry51–55.

the main conclusions made in the works by russian scientists in 1970’s are as follows:

(i) Non-toxic inhibitors of radical processes – antioxidants exhibit a wide gamut of biological activity;

(ii) the biological effectiveness of antioxidants correlates with their antioxidis-ing properties;

(iii) Depending on dose, antioxidants may either increase (at low doses) the antioxidising activity or decrease it;

(iv) the activity of antioxidants for curing any particular disease depends on the time of introduction in the course of medical treatment because the development of the disease may be accompanied by stages of changing the antioxidising activity. The compound may be efficient only if it is introduced in a low dose at the stage of reduced aoa or in a high dose at the stage of aoa elevation.

it is natural, and it could not be otherwise, that the pioneering works on antioxi-dants and free-radical reactions occurring in living systems were attacked furiously by opponents – scientists of various profiles: biologists, physicians, and even some of chemists and physicists. in spite of all arguments, vitalistic tendencies were strong:

744

nobody could even dare to think that synthetic antioxidants may substitute natural ones. ‘neither of free radical reactions can develop in a living organism’, the op-ponents claimed, ‘because these reactions are not controlled, but strict regulation is essential for a living organism’. in addition, according to the concepts prevailing at that time, the membrane structure was such that it excluded the valence transfer from one lipid molecule to another since a lipoprotein model of membrane showed that lipids were separated by protein molecules. the opponents considered the absence of specific enzymes governing these reactions a strong argument against inhibitors of radical reactions and radical processes as such in organisms. the wide-spread opinion was that the antioxidant function, even that of tocopherol, was a side effect of its activity and important only for in vitro processes but did not play any role in bioobjects life. This opinion was supported by the fact that the deficiency of tocopherol (E-avitaminosis) can not be cured completely by applying synthetic antioxidants56,57. Finally, it was considered doubtful that works, in which peroxides were detected in lipids isolated from organs and tissues, dealt with true amounts of products of free radical processes in vivo, but not with the amounts of products formed during the process of isolation.

all the objections and skepticism have been rejected in due time. antioxidant enzymes were discovered, the model of membrane was revised. the development of biochemistry and biophysics held the course in the direction of verification of this concept58,59.

these several pages of history should be concluded by the acknowledgment that the works carried out by Soviet scientists in the field of free-radical biology were pioneering and many if not all data obtained at that time remain valuable until the present time.

BioantioxiDants

in spite of the fact that antioxidants are much spoken about because of their extensive application for various purposes, we should return to the definition of bioantioxi-dants.

Bioantioxidants are substances that exhibit the properties of inhibitors in model free-radical reactions of oxidation and retain these properties when introduced into a living organism (cell culture, etc.). the absence of even one of these qualities does not permit calling a substance a bioantioxidant (Bao). although the antioxidising activity of lipids can be increased by applying substances that are synergists to natu-ral antioxidants or those transformed into antioxidants in the course of metabolism, bioantioxidants, by our definition, should necessarily possess the ability to inhibit an oxidising free-radical process in model reactions. this property of Bao makes it possible to predict the spectrum of their biological effects and perform a targeted synthesis of compounds.

745

at present, the following pathways of antioxidants effects on the cell metabolism are considered:

(i) interaction of Bao with free radicals of different nature;(ii) incorporation of Bao into the membrane structure and changes in the

membrane functional activity due to changes in the membrane viscous properties (fluidity);

(iii) effect of BAO on the activity of membrane proteins – enzymes and recep-tors;

(iv) effect of Bao on the cell genetic apparatus including gene expression;(v) effect of Bao on the regulatory systems of cell and, indirectly, on its me-

tabolism as a whole.note that the reaction rate constants of the same inhibitors with different radi-

cals differ considerably (by several orders of magnitude)60,61 (see table 1). there are cases when antioxidants that are active for some radicals can not compete in inter-acting with others, and we can not protect cell components from the effects of these radicals because the affinity of radicals to them will be higher than that of inhibitors introduced.

table 1. rate constants of interaction of inhibitors-antioxidants with biologically active radicals

inhibitors

radicals(rate constants, l/mol s)

oh • r•

proteinsro2

•

lipids

o2–•

superoxide anion radical

2-Ethyl-6-methyl-3-hydroxypyridine, hydrochloriden-3,5-di-tert-butyl-4-hydroxyphenyl propionic acid (phenozan)5,7,8-Trimethyltocopherol (α-tocopherol)

3.3 × 1010

4.4 × 1010

8 ×1010

1.9 × 106

1.2 ×106

9.0 × 105

2.2 ×104

3.4 × 106

26–4

10–2

47 × 104

(soluble form)

another obstacle to effective using of antioxidants is their extreme concentra-tion–effect dependence. As noted above, antioxidants applied in high concentrations produce an opposite effect and do not inhibit but accelerate free radical reactions. the phenomenon may be attributed either to a high activity of radicals accumulated from inhibitors or to the prevailing consumption of natural antioxidants as compared with synthetic ones introduced. Many of these effects depend on the initial characteristics of free radical processes and the initial level of antioxidants.

thus, because of the versatility of antioxidants properties and feasibility of af-fecting various normal and pathological states, we are obliged to know exactly the nature of radicals responsible for pathological changes, the time of ao introduction,

746

concentration, and elementary constants of inhibitors. a negligence in or an erroneous approach to the antioxidant therapy may lead to negative results.

a study on the mechanism of Bao effects showed that there is a whole system of relations between separate indices of cell metabolism that vary under the action of antioxidants.

investigation of the physicochemical regulatory system maintaining the level of free-radical reactions in lipids, on the one hand, and regulating the metabolism of membrane lipids and the rate of consumption of antioxidants in lipids, on the other, has been further developed46,47.

the components of this system are antioxidants, free radicals, peroxidation products, composition and oxidisability of lipids, and the rate of consumption of antioxidants.

it was shown that enhancement of the antioxidant level is associated with reduction of the lipid peroxidation (lPo) rate, reduction of oxidation products concentration, reduction of the rate of lipids exit from membranes and enrichment of membranes with unsaturated lipids, and enhancement of lipids oxidisability. the latter effect re-sults, in turn, in an increase in the rate of consumption of the antioxidant activity and, correspondingly, in returning of the aoa and lPo rate to the normal. an opposite situation is observed with decreasing in the aoa concentration, increasing the lPo rate, etc. similar systems of regulation were discovered almost for all intracellular and cellular membranes of animals, plants and microorganisms. note that changes in the composition and oxidisability degree of lipids are associated with changes in the viscosity of various layers of membranes. the above parameters affect the activity and kinetic characteristics of membrane-bound proteins – enzymes and receptors; changes in the lPo rate may result in changing not only of the structure, but also of the functional activity of membranes. in normal membranes, we observe identical relationships between the parameters; the difference concerns the system relaxation time (from several minutes to several days).

Exposure of organism to any damaging factor is associated with changes in this system of regulation. long-term changes may result from: (i) the action of a chronic factor that does not cause breaking bonds in the system of regulation; the system can return to the normal after cessation of the action; (ii) there may occur situations when exposure to a damaging factor results in transition to a new level of regulation; and (iii) there may occur breaking bonds in the system of regulation, which prevents the system to return to the normal. In the latter case, antioxidants may be efficient as a component of complex therapy. such conclusions were drawn both in experimental studies and in clinical tests.

at present, scientists-pharmacologists must answer the question: whether all pharmacologically active compounds should be multitargeted drugs, i.e. should be aimed at several targets but not at one specific target. The answer to this question is not unambiguous, although the scientists who pose it cite the data that are evidence for the fact that most diseases are associated with changes and defects of various

747

pathways of metabolism and drugs should not be aimed at one critical target, but at all affected by the disease.

Meanwhile, an alternative to this approach is complex therapy that uses several compounds aimed act at its specific target; the compounds may be applied at various concentrations (not only at those specified by the compound structure), at various times of application, in various solvents, etc.

in our opinion, both views have the right to exist and different approaches should be taken in each specific cases for the benefit of patients.

as was noted above, breaking bonds in the system of regulation point to impos-sibility of returning the system to the initial state by applying ao.

investigation of the regulation system as a whole, but not only of separate changes in the system makes it possible to decide when the AO monotherapy is sufficient and when a complex therapy is needed, in which, apart from ao, other biologically active substances aimed at other targets are needed. to some extent, this may be ac-complished by synthesis of hybrid molecules.

the term of hybrid antioxidants implies molecules whose structure contains parts responsible for antioxidant properties and fragments of molecules responsible for other specific functions.

in most cases, synthesis of hybrid molecules does not yield a new polyfunctional structure, but cross-linked or integrated molecules that produce a high ao effect and are aimed at other targets specific of a certain disease.

Very often, when designing a molecule, it is necessary to retain the specific ac-tivity of one of the hybrid components and, at the same time, to reduce side-effects, e.g. toxicity of the compound. one of the promising ways is incorporating nitroxyl radicals in the molecule structure.

nitroxyl derivatives of biologically active substances are the most numerous and earliest-synthesised hybrid antioxidants. Previously, it was shown that the nitroxyl radical exhibits the ao properties in model reactions pf oxidation and in vitro and in vivo experiments. one of the pioneers in organic chemistry who synthesised nitroxyl derivatives of BAS was a Soviet scientist – A.B. Shapiro62. since then, nitroxylation of Bas has been put into practice in pharmacology. Most extensively, antitumor compounds are nitroxylated. konovalova63–68 who had gained a great experience in the field of nitroxyl antitumor compounds made up a list of synthesised and well-studied antitumor agents referred to the class of nitroxyl-containing antibiotics, antimetabo-lites, and alkylating agents (table 2).

748

table 2. nitroxyl derivatives of antitumor agents

initial components lD50(mg/kg) nitroxyl derivatives lD50

(mg/kg)1 2 3 4

thio tEF

18 187

tEF

15 150

n n

nn n

n

tEM

1.5ono

n n

nn

n

15

o o

o oh

oh o

oh

coch3

ch3

o

hohcl nh2

ch3

rubomycin

5.6

o

no o

o oh

oh o

oh

ch3

o

hohcl nh2

ch3

c n nch3

ruboxyl

50

n

n

o

o

Fh

h

5-fluorouracil

100 n

n

o

o

F

h

n

o

ch3

c

ch3

h3c

h3c

o

magnicyl

510

n

n

o

o

Fh

o

fluoroaphur

750

to be continued

749

1 2 3 4

nhc

o

n

no

ch2ch2cl

cyclohexyl – nitrosourea

47

o

n

no

ch2ch2clno

o n

nhc

ch2ch2cl

no

n

o

ch3

n c

60

150

nitroxyl derivatives are 5 to 10 times less toxic than the initial compounds.at present, derivatives of platinum compounds have found wide use in chemo-

therapy of malignant tumors. it should be noted that, along with their high antitumor effect, these compounds exhibit a high toxicity. supposedly, their cytotoxic effect and other side-effects (nephro- and ototoxicity, nausea, etc.) are associated with intensi-fication of free-radical processes and formation of active oxygen species [O2

•–, oh•] induced in cell by cisplatinum. in fact, the antitumor effect is associated, apart from the interaction with the Dna molecule, with enhancement of free-radical reactions. as a rule, reduction of toxicity results in decreasing the antitumor effect. Develop-ment of hybrid molecules based on platinum derivatives and nitroxyl radical makes it possible to reduce the toxicity and retain the activity of the agent.

at present, extensive studies are being carried out on the synthesis, structure, and biological activity of mixed-ligand complexes of platinum (ii) with aminonitroxyl radicals69.

the largest group of antitumor compounds includes nitroxyl derivatives of anthra-cycline antibiotics. one of the major achievements of the chemistry and biochemistry of hybride antioxidants of this class is the development of emoxypin (ruboxyl) – a nitroxyl derivative of the anthracycline antibiotic – rubomycin.

this compounds have a great advantage of hybride molecules, namely, in the background of high antitumor effect (higher than that of the parental compound – rubomycin), its toxicity decreased by 40% and cardiotoxicity and depilative prop-erties vanished almost completely. the compound acquired some novel properties that are not characteristic of rubomycin and nitroxyl radical. it is of interest that there appeared no cross-resistance with rubomycin. the development of this compound is the greatest practical achievement in this field of investigation and we have passed the second stage of clinical tests.

other promising hybrid compounds are antitumor agents prepared on the basis of mixed-ligand platinum ii and platinum iv complexes containing antioxidant frag-ments of the array of aminonitroxyl radicals. the feasibility of preparing such mixed complexes relies on the fact that cisplatinum ([nh3]2Ptcl2) transforms readilty into

continuation of table 2

750

a na[(nh3)Ptcl3] complex, which interacts with primary amines on the background of sodium iodide and transforms into a (rnh2) (nh3)Ptcl complex. having been treated with silver nitrate the latter complex exchanges readily halogen anions for the mobile no3

– anion; as a result, they can be substituted for other anions – Cl- and anions of dicarboxylic acids.

the above complexes were characterised in various tumor models; it was shown that the Dna binding rate for these new complexes is comparable with that of cisplati-num, but their oxidising effect is opposite: the initial cisplatinum accelerates oxidation in model radical reaction; the platinum–nitroxyl complexes, on the contrary, inhibit oxidation. From this point of view, it is easy to explain the reduction of the toxicity of the synthesised complexes.

in the recent time, a great interest was aroused for platinum iv compounds. the scientists-organochemists of the institute of theoretical Problems of chemical Physics synthesised complexes of platinum iv with aminonitroxyl radicals. these compounds were tested on the experimental model of P-388 leukemia. In the experiments, the complexes exhibited a strong effect on mice leukemia-carriers: sometimes to complete recovery; the toxicity reduced two- to four-fold70.

the most impressive results were obtained for low doses of cisplatinum and complexes of platinum iv with amino nitroxyl radicals applied in combination for treatment of P-388 leukemia. The survival rate for leukemia mice was 100%.

as was noted above, nitroxylation of antitumor compounds was successful. there are some other examples of application of hybrid compounds containing nitroxyl radicals in their molecules.

at the institute of Problems of chemical Physics, novel biologically active com-pounds were synthesised – nitroxyl derivatives of azidothymidine of the common formula as follows:

N

N

O

O

CH3R2

O

N3

R1O

where r1 is the radical containing a nitroxyl group > n − o and r2 = r1 or h.these compounds produce the antiviral effect against rna-containing viruses

(human immunodeficiency virus and vesicular stomatitis virus) and the DNA-contain-ing virus – cytomegalovirus71.

it should be noted that hybrid molecules containing azidothymidine and antioxi-dant fragments in their structure inhibit the reproduction of cytomegalovirus; other

751

azidothymidine derivatives do not possess this property. considering that death of immunodeficiency patients is caused by cytomegalovirus-provoked diseases, this property of antioxidant- and azidothymidine-based hybrid molecules should be em-phasised particularly.

one of the most promising and important practical areas of the modern chemistry related to phenol bioantioxidants is the synthesis of hybrid compounds that combine the antioxidant activity and the capacity for structural interactions with a biosystem. this type of compounds includes the so-called ‘buoy’ compounds synthesised from quaternised derivatives of dialkylaminoalkyl-substituted 2,4- or 2,6-di-tert-butylphe-nols and dialkylaminoalkyl ethers of phenozan acid. For these compounds, a wide spectrum of biological activity was discovered – the antimicrobial, antiviral, analgetic, etc. activities. also, it was discovered that gradual (step-by-step) redox and solvolytic conversions result in the formation of a cascade of intermediates with different inherent activities – antioxidising, chelating, capacity for incorporating into the charge transfer chain, etc. in the cascade mechanism of the effect of 2,4 (2,6)-di-tert-butylphenol derivatives, the tendency to the formation of heterocyclic compounds, which make a contribution to the total biological activity, plays a great role.

another important currently developing research area is the synthesis of hybrids of functional di- and tert-butylphenols and biocompatible macromolecules. in this area, it is possible to achieve the highest values of antioxidising activity of hybrid compounds of a wide variety of hydrophobic–hydrophilic relationships and particular structural properties in solutions72.

the presence of the positively-charged nitrogen atom in the hybrid molecule provides for the antioxidant adherence to the surface of a cell membrane and its fixation in a certain place by means of the lipophilic long-chained alkyl fragment R1. such a structure ensures the targeted use of the antioxidant and favours the inhibi-tion of pathological processes in cell, e.g. intensification of LPO and disorders of cell membranes functions.

Hence, a hybrid molecule – an analog of acetylcholine [CH3cooch2ch2n+(ch3)3oh–] was constructed. in this molecule, instead of acetic acid, the ester bond is formed by carboxylic acids that contain a 2,6-di-tert-butyl-4-hydrophenyl fragment. We called these hybrid antioxidants ichfans.

it might be expected that the above structure should give rise to a bioantioxidant with an anticholineesterase activity. indeed, anticholineesterase compounds are the most efficient up-to-date therapeutic agents applied for the Alzheimer’s disease (AD); they make it possible to maintain the level of acetylcholine (responsible for memory and cognitive functions) in the disease-affected sections of brain. on the other hand, the oxidative stress, i.e. promotion of lPo in cell membranes of brain and in cells of peripheral systems and organs, plays an important role in the development of aD (ref. 73).

752

the above-described properties and interactions of the aos with biological systems, e.g. cell membranes, made it possible to suggest using these compounds as drugs for the therapy of the alzheimer’s disease74–80.

in studies performed with the use of various oxidation models, antioxidising properties of ichfans were determined and assessed quantitatively. it was found out that new hybrid AOs – ichfans – possesses the antioxidant (revealed in the model of oxidation of homogenate lipids) and anticholineesterase activity that exceeds the cor-responding indices of the initial substances. the addition of the molecule with alkyl substituents with different lengths of the aliphatic chain on the nitrogen atom promotes the aoa and inhibiting properties. With increasing the length of the carbon chain in the alkyl substituent, the capacity of the compounds to inhibit achE increases. the type of inhibition depends on an alkyl substituent. although, the inhibiting power of the substances under study for a membrane-bound enzyme is by an order of magni-tude lower than that for soluble achE, the same relationships between the structure and inhibiting activity of the compounds were detected for the membrane-bound and soluble enzyme81–84.

a strict direct correlation between the anticholineesterase and antioxidative properties of ichfans was detected; the correlation is also of the same character for membrane-bound and soluble achE.

on the basis of results of in vitro experiments, by the criteria of the anticholineeste-rase and antioxidative properties, the optimum compound was chosen for further in vivo studies. that was a hybrid with the radical r1 = c10h21; hereinafter, it will be referred to as ichfan. an additional lengthening of the tail resulted in undesirable perturbat-ing effect on membranes; the effect manifested itself by the erythrocytes increasing sensitivity to hemolysis. in accordance with the data published, the compound with r1 = c10h21 is sufficiently hydrophobic to permeate through the hematoencephalic barrier.

With regard to the achE and antioxidising effects and feasibility of permeation through the hEB, ichfan is of considerable interest in view of using it as a drug for treatment of aD.

it should be noted that the oxidative stress as an important factor of aD may be not only a source of free radicals damaging cell structures and macromolecules, but also a symptom of a disorder in the operation of the system of homeostasis of lipid peroxidation (lPo) in biological membranes. this system plays an important role in the regulation of cell metabolism; it controls the structure and structure-related func-tions of various cellular membranes.

an analysis of the data85–97 on changes in the lipid metabolism, composition and structure of the membrane lipid phase, which plays a great role in transmission, processing, and storing the information in cell, showed that the development of aD is associated, along with enhancement of lPo, with enrichment of lipids with unsaturated compounds and increasing the fluidity (decreasing the viscosity) of the lipid phase. In other words, on the background of enhancement of LPO, the lipid bilayer fluidity

753

increases; the effect favours the increasing of oxidation and, hence, the development of the pathological process. this property of membranes in aD makes it impossible to prevent from the oxidative stress by means of traditional phenol antioxidants, which promote the fluidity increasing and complicate the disease development.

We suggested that addition of a saturated fatty-acid tail (buoy) that incorporates in the antioxidant molecule membrane may make the membrane more rigid and thus contribute to the therapeutic effect of ichfan. as is seen from table 1, after introduc-tion of the compound to mice, the microviscosity of the membrane near-surface sites studied by the method of EPR spin probes either changes insignificantly or increases; the latter is a desirable effect. it should be emphasised particularly for membranes isolated from a coarse fraction of synaptosomes because aD is associated mostly with damages of nerve fibers.

according to many of the researchers, one of the aD risks is an elevated cho-lesterol level98–100. it should be noted that the cholesterol content in rat brain tissues decreased by 40% within 2 h after introduction of 15 mg/kg of the compound to the animals. the cholesterol content in a cytoplasmic fraction isolated from mice brain decreased almost two-fold within 2.5 h after introduction of 6 mg/kg of ichfan.

thus, in addition to the ability of ichfan to inhibit the cholineesterase activity, it can inhibit the oxidative stress (lPo) and, in contrast to traditional phenol-type antioxidant, rigidise the structure of the membrane lipid bilayer or at least prevent from increasing its fluidity. The combination of these properties may be beneficial for the therapy of aD through the correction of the aD-damaged system of regulation of lipid peroxidation homeostasis that participates in controlling of cell metabolism. a certain contribution to the therapeutic effect of ichfan may be provided by its lower-ing effect on the level of cholesterol.

rEFErEncEs1. B. n. tarusov: Mechanisms of Biological Effects of radioactive Emissions. Medgiz, Moscow,

1954. 140 p. (in russian).2. E. B. Burlakova, B. G. DzantiEv, G. B. sErGEEv, n. M. EManuEl: izv. vuzov, 2 (4),

533 (1959) (in russian).3. E. B. Burlakova, B. G. DzantiEv, G. B. sErGEEv, n. M. EManuEl: nauchn. dokl. vyssh.

shk., 1, 145 (1960) (in russian). 4. E. B. Burlakova, B. G. DzantiEv, G. B. sErGEEv, n. M. EManuEl: in: Proc. of confer-

ence ‘Biochemical and Physicochemical Principles of Biological Effects of radiation’, izd. MGu, Moscow, 1957, p. 10 (in russian).

5. n. M. EManuEl: in: Proc. of the Moscow society of nature researchers. nauka, Moscow, 1963, p. 73 (in russian).

6. i. i. saPEzhinskii, n. M. EManuEl: Dokl. an sssr, 163 (4), 845 (1965) (in Russian).7. G. P. zhizhina, k. E. kruGlyakova, n. M. EManuEl: Dokl. an sssr, 163 (4), 931 (1965)

(in russian). 8. n. M. EManuEl, E. B. Burlakova, k. E. kruGlyakova, i. i. saPEzhinskii: izv. akad.

nauk AN SSSR, Ser. biol., (2), 183 (1965) (in Russian).9. E. Burlakova: Moderna Medizina, 2 (4), 215 (1972) (in russian).

754

10. n. M. EManuEl, l. P. liPchina: Dokl. an sssr, 121, 141 (1958) (in Russian).11. n. M. EManuEl: Pathways of synthesis and investigation of antitumor Drugs. nauka, Moscow,

1962. 22 p. (in russian).12. v. v. Ershov, G. a. nikiForov, a. a. voloDkin: spatially-hindered phenols. khimiya,

Moscow, 1972. 351 p. (in russian).13. l. D. sMirnov, s. i. sholina et al.: Izv. AN SSSR, Ser. khim., (5), 890 (1963) (in Russian).14. E. B. BuRLAkOVA, A. V. АLESENkO, E. M. MOLOCHkINA, N. P. PALMINA, N. G. kHRAP-

ova: Bioantioxidants for radiation Damage and Malignant Growth. nauka, Moscow, 1975. 214 p. (in russian).

15. E. B. Burlakova, v. D. GaintsEva, l. v. slEPukhina, n. G. khraPova, n. M. EManuEl: Dokl. an sssr, 155, 1398 (1964) (in Russian).

16. E. B. Burlakova, n. G. khraPova, G. i. yaDykin: Biofizika, 16 (2), 201 (1971) (in rus-sian).

17. s. a. aristarkhova, E. B. Burlakova, n. G. khraPova: izv. akad. nauk, ser. khim., (2), 2714 (1971) (in russian).

18. E. B. Burlakova, s. a. BuroBina, n. G. khraPova, G. i. yaDykin: Biofizika, 16 (1), 39 (1971) (in russian).

19. i. i. saPEzhinskii: Biopolymers: kinetics of radiation-induced and Photochemical conversions. Nauka, Moscow, 1988. 214 p. (in Russian).

20. E. l. lozovskaya, i. i. saPEzhinskii: radiation Biology. Ecology, 33 (4), 601 (1993) (in russian).

21. E. B. BuRLAkOVA, E. M. МоLOCHkINA: Biofizika, 13, 443 (1968) (in Russian).22. E. B. Burlakova, s. a. BuroBina, n. G. khraPova: Dokl. akad. nauk sssr, 200, 461

(1971) (in russian).23. E. B. BuRLAkOVA, E. M. МоLOCHkINA: vopr. onkol., 20, 62 (1974) (in russian).24. E. B. Burlakova, n. P. PalMina: Vesti AMN SSSR, (3), 74 (1982) (in Russian).25. o. s. FrankFurt, l. P. liPchina, t. v. Bunto, n. M. EManuEl: Bull. Experim. Biol. Med.,

8, 86 (1967) (in Russian).26. N. M. EMANuEL, N. P. kоNOVALOVA, L. M. DRONOVA: Dokl. an sssr, 143 (3), 737 (1962)

(in russian).27. n. M. EManuEl, l. M. Dronova, n. P. konovalova: Dokl. an sssr, 152 (2), 481 (1963)

(in russian).28. E. B. Burlakova, n. P. PalMina: vopr. onkol., (10), 1155 (1990) (in russian).29. n. P. PalMina, E. B. Burlakova: Vesti AMN SSSR, (1), 91 (1985) (in Russian).30. n. M. EManuEl, r. E. kruGlyakova, n. v. nikolaEva, n. a. zakharova: Dokl. akad.

nauk sssr, 157 (4), 935 (1964) (in russian).31. i. i. saPEzhinskii, yu. v. silaEv, n. M. EManuEl: Dokl. an sssr, 159 (6), 1378 (1964)

(in russian).32. v. a. sharPatyi: Radiation Chemistry of Biopolymers. Energoizdat, Moscow, 1981. 168 p. (in

russian).33. l. k. oBukhova: uspekhi khimii, 44 (10), 1914 (1975) (in russian).34. n. M. EManuEl, l. k. oBukhova: Exp. Gerontol., 13, 25 (1979).35. l. k. oBukhova, n. sh. nakaiDzE, a. M. sErEBrjany, l. D. sMirnov, a. P. akiFiEv:

Exp. Gerontol., 14, 335 (1979).36. i. P. saDovnikova, l. k. oBukhova, t. v. Bunto, l. D. sMirnov: izv. an sssr, ser.

biol., (4), 543 (1984) (in Russian).37. n. M. EManuEl, l. k. oBukhova, t. v. Bunto, v. v. Dyakova: izv. an sssr, ser. biol.,

(1), 32 (1997) (in russian).38. E. B. Burlakova: Biofizika, 12 (1), 82 (1967) (in Russian).39. E. B. Burlakova, n. P. PalMina, n. l. ruzhinskaya: izv. an sssr, (1), 134 (1971) (in

russian).

755

40. a. v. alEsEnko, i. s. sokolova, G. v. kukushkina, E. B. Burlakova, l. B. Gor-BachEva: Dokl. an sssr, 254 (6), 1472 (1989) (in Russian).

41. a. v. alEsEnko, E. B. Burlakova: Dokl. an sssr, 207, 1471 (1972) (in russian).42. a. v. alEsEnko, E. B. Burlakova, a. a. vainson: Dokl. an sss, 202, 208 (1972) (in

russian).43. a. v. alEssEnko, E. B. Burlakova: Bioelectrochemistry, 58, 13 (2002).44. E. B. Burlakova, a. n. GoloshchaPov: Biofizika, 10 (5), 816 (1975) (in Russian).45. E. B. Burlakova, M. i. DzhalyaBova, E. M. Molochkina: in: structure, Biosynthesis

and conversions of lipids in animal and human organisms. izd. Fan, Frunze, 1975, p. 70 (in russian).

46. s. a. aristarkhova, G. v. arkhiPova, E. B. Burlakova et al.: Dokl. an sssr, 228, 215 (1976) (in russian).

47. E. B. Burlakova, M. i. DzhalyaBova, E. M. Molochkina, a. P. khokhlov: Bio-physical and Biochemical Information Transfer in Recognition and Aging. Moscow, 1979, p.1583.

48. E. B. Burlakova: kardiologiya, 20, 48 (1980) (in Russian).49. G. v. arkhiPova, E. B. Burlakova, a. F. sEMiokhina, i. B. FEDotova, l. v. krushin-

skii: Dokl. an sssr, 256 (3), 746 (1981) (in Russian).50. v. v. DisvEtova, E. i. GEniEva et al.: klinich. med., (3), 126 (1968) (in Russian).51. n. n. zoz, i. B. lEManova, s. a. akhMEDov, D. s. sulEiManov, a. M. sErEBryanyi,

i. s. Morozova, o. D. sultanova: S.-khoz. biol., (4), 71 (1985) (in Russian).52. D. D. liPsits, k. E. kruGlyakova, M. s. Postnikova: Dokl. an sssr, 145, 212 (1962)

(in russian).53. a. s. saDykov, k. E. kruGlyakova et al.: chemistry of Phyto substances. Fan, tashkent,

3, 86 (1968) (in Russian).54. E. B. Burlakova, E. P. larichEva: vniizh, 15 (1973) (in russian).55. l. D. sMirnov, yu. v. kuznEtsov, l. M. aPashEva, k. D. Poltorak, a. l. GrinchEnko,

k. M. DyuMaEv, n. M. EManuEl: Author’s Certificate 1098934, Feb. 23, 1983 (in Russian).56. s. a. krashakov, E. B. Burlakova, n. G. khraPova: Biol. Membr., 12 (2), 173 (1998)

(in russian).57. E. B. Burlakova, s. a. krashakov, n. G. khraPova: kinetic characteristics of toco-

pherols as Antioxidants. Nauka, Moscow, 1988. 247 p. (in Russian).58. E. B. Burlakova: uspekhi chimii, 44 (10), 1871 (1975) (in Russian).59. E. B. Burlakova, n. G. khraPova: uspekhi khimii, 54, 1540 (1985) (in Russian).60. D. G. PoBEDiMskii, E. B. Burlakova: in: Mechanism of antioxidant action in living organ-

isms, in atmospheric oxidation and antioxidants (Ed. j. scott), 3 (9), 223 (1993).61. E. B. Burlakova, n. G. khraPova, v. n. shtolko, n. n. EManuEl: Dokl. an sssr,

169 (3), 688 (1966) (in Russian).62. a. B. shaPiro, a. a. kroPachEva, v. i. suskina et al.: Mass-spectroscopic study of Para-

magnetic Derivatives of Ethylenephosphoramides, 4 (1), 864 (1971) (in Russian).63. n. P. konovalova: Nitroxyl Radicals as Modifiers of Biological Effects of Antitumor Compounds.

khimfizika, 10 (6), 861 (1991) (in Russian).64. n. M. EManuEl, n. P. konovalova, l. s. Povarov et al.: Author’s certificate 977462

SSSR, BI, 1980. 65. n. M. EManuEl, n. P. konovalova, r. F. Diatchkovkaja: Potential anticancer

Agents – Nitroxyl Derivatives of Rubomycin. Neoplasma, 32, 285 (1985). 66. n. M. EManuEl, r. i. zhDanov, n. P. konovalova et al.: Paramagnetic Diethyleneimides of

urethan Phosphoric acids as antitumor substances. voprosy oncol., 36, 54 (1980) (in Russian).67. P. sEMinara, F. Franchi, n. konovalova et al.: activity of a nitroxylated analog of

Daunirubomocin, ruboxyl in B-lymphoproliferative Disorders. acta haematol., 105, 77 (2001).68. n. M. EManuEl, a. v. rozEnBErG, v. a. GoluBEv et al.: urgent Problems of Experimental

chemotherapy of tumors, 5, (1987) (in Russian).

756

69. v. D. sEn’, n. a. rukina, v. v. tkachEv et al.: synthesis, structure and Biological activity of Platinum(ii) complexes, izv. an sssr, 9, 1624 (2000) (in russian).

70. v. D. sEn’, v. v. tkachEv, l. M. volkova et al.: synthesis, structure and antitumor Proper-ties of Platinum(iv) complexes with aminonitroxide. izv. an sssr, ser. khim., (2), 403 (2003) (in russian).

71. v. D. sEn’, v. a. GoluBEv, n. a. rukina, t. G. orlova, E. B. Burlakova: nitroxyl De-rivatives of 3′-azido-2′3′-dideoxythymidine with Antiviral Activity. Patent Ru 2103274 C1 1998.

72. c. BEhl: alzheimer Disease and oxidative stress: implications for novel therapeutic approaches. j. Biol. chem., 271, 26482 (1996).

73. r. nitsh, j. k. Blusztajns, a. G. Pittas et al.: Evidence for a Membrane Defect in alzheimer Disease Brain. Proc. natl. acad. sci. usa, 89, 1671 (1992).

74. P. zaFrilla, j. MulEro, j. M. xanDri et al.: oxidative stress in alzheimer Patients in Dif-ferent stages of the Disease. curr. Med. chem., 13, 1075 (2006).

75. B. Frank, s. GuPta: a review of antioxidants and alzheimer’s Disease. ann. clin. Psychiatry, 17, 269 (2005).

76. P. i. MorEira, k. honDa, Q. liu et al.: oxidative stress: tthe old Enemy in alzheimer’s Disease Pathophysiology. curr. alzheimer res., 2, 403 (2005).

77. E. Mariani, M. c. PoliDori, a. chEruBini et al.: oxidative stress in Brain aging, neuro-degenerative and vascular Diseases: an overview. j. chromatogr. B. analyt. technol. Biomed. life sci., 827, 65 (2005).

78. P. i. MorEira, M. a. sMith, x. zhu et al.: oxidative Damage and alzheimer’s Disease: are antioxidant therapies useful? Drug news Perspect., 18, 13 (2005).

79. M. c. irizarry: Biomarkers of alzheimer Disease in Plasma. neuro rx., 1, 226 (2004).80. M. G. PErEvozkina, n. M. storozhok, G. a. nikiForov: correlation between chemical

structure and inhibiting Properties of sterically-hindered Phenols of the ichfan Group. Biomed. khim., 51, 413 (2005) (in russian).

81. F. i. BraGinskaya, o. M. zorina, E. M. Molochkina et al.: synthetic Bioantioxidants as inhibitors of the achE activity. izv. an sssr, (5), 690 (1992) (in russian).

82. F. i. BraGinskaya, E. M. Molochkina, o. M. zorina et al.: new synthetic Bioantioxi-dants – Acetylcholinesterase (AChE) Inhibitors. In: Alzheimer Disease: from Molecular Biology to Therapy (Eds R. Becker, E. Giacobini). Birkhauser-Boston, 1996, 337–342.

83. i. B. ozErova, E. M. Molochkina, E. B. Burlakova: Ichfan – New Potential Drug for treatment of alzheimer’s Disease. advances in Gerontology, 6, 30 (2001).

84. E. M. Molotchkina, i. B. ozErova, E. B. Burlakova: Ichfan – New Antioxidant Drug for the treatment of alzheimer’s Disease. Free rad. Biol. Med., 33, 229 (2002).

85. r. BuchEt, s. Piku: alzheimer’s Disease: its origin at the Membrane, Evidence and Questions. acta Biochimica Polonica, 47, 725 (2000).

86. M. r. PrasaD, M. a. lovEll, M. yatin, h. Dhillon, W. r. MarkEsBEry: regional Membrane Phospholipid alterations in alzheimer’s Disease. neurochem. res., 23, 81 (1998).

87. k. WElls, a. a. Farookui, l. liss, l. a. horrocks: neural Membrane Phospholipids in alzheimer Disease. neurochem. res., 20, 1329 (1995).

88. M. soDErBErG, c. EDlunD, k. kristEnsson, G. DallnEr: Fatty acid composition of Brain Phospholipids in aging and in alzheimer’s Disease. lipids, 26, 412 (1991).

89. j. W. PEttEGrEW, k. PanchalinGaM, r. l. haMilton et al.: Brain Membrane Phospholipid alterations in alzheimer’s Disease. neurochem. res., 7, 771 (2001).

90. M. soDErBErG, c. EDlunD, i. alaFuzoFF et al.: lipid composition in Different regions of the Brain in alzheimer’s Disease/senile Dementia of alzheimer’s type. j. neurochem., 5, 1646 (1992).

91. r. M. nitsch, j. k. Blusztajn, a. G. Pittas et al.: Evidence for a Membrane Defect in alzheimer Disease Brain. Proc. natl. acad. sci. usa, 89, 1671 (1992).

757

92. k. WElls, a. a. Farookui, l. liss, l. a. horrocks: neural Membrane Phospholipids in alzheimer Disease. neurochem. res., 20, 1329 (1995).

93. W. r. MarkEsBEry, P. k. lEunG, D. a. ButtErFiElD: spin label and Biochemical studies of Erythrocyte Membranes in alzheimer’s Disease. j. neurol. sci., 45, 323 (1980).

94. G. s. zuBEnko, u. koPP, t. sEto et al.: Platelet Membrane Fluidity individuals at risk for alzheimer’s Disease: a comparison of results from Fluorescence spectroscopy and Electron spin resonance spectroscopy. Psychopharmacology (Berlin), 145, 175 (1999).

95. W. kozuBski, M. sWiDErEk, i. kloszEWska et al.: Platelet Membrane Fluidity and recep-tor Exposition in Patients with alzheimer’s Disease. neurol. neurochir. Pol., 33, 1275 (1999) (in Polish).

96. W. E. MullEr, c. kirsch, G. P. EckErt: Membrane-disordering Effects of Beta-amyloid Peptides. Biochem. soc. trans., 29, 617 (2001).

97. F. i. BraGinskaya, o. M. zorina, n. P. Pal’Mina et al.: some Blood Biochemistry Param-eters during the cholinergic treatment of alzheimer’s Disease. neurosci. Behav. Physiol., 31, 457 (2001).

98. M. siMons, P. kEllEr, B. De strooPEr et al.: cholesterol Depletion inhibits the Generation of Beta-amyloid in hippocampal neurons. Proc. natl. acad. sci. usa, 95, 6460 (1998).

99. G. P. EckErt, c. kirsch, W. E. MullEr: Brain-membrane cholesterol in alzheimer’s Disease. j. nutr. health aging, 7, 18 (2003).

100. r. l. raFFai, k. h. WEisGraBEr: cholesterol: from heart attacks to alzheimer’s Disease. j. lipid res., 44, 1423 (2003).

Received 5 December 2007 Revised 14 January 2008

758

Oxidation Communications 31, No 4, 758–775 (2008)

ApplicAtions of diAlysis

k. a. stanchEva

Department of Analytical Chemistry, Prof. Dr. Assen Zlatarov University, 8010 Bоurgas, Bulgaria E-mail: krasimiraangelova@abv.bg

aBstract

this article provides a comprehensive overview of dialysis covering the fundamentals and recent developments of the process and its applications. theories, principles and combinations of dialysis with methods of separation, pre-concentration, and final determination are discussed.

Keywords: dialysis, separation, pre-concentration, recovery, enrichment and removal of ions, flow-through dialysis, micro-dialysis.

aiMs anD BackGrounD

nowadays, the general trends toward the instrumentation and automation of chemical analysis as well as the fashion to determine a possibly broad spectrum of analytes have inspired the increasing interest of analytical chemists to search for new meth-odological and instrumental approaches based on membrane methods. recent analyti-cal literature has revealed that membrane separation techniques have experienced a rapid growth1–3. Dialysis membrane separation processes have received an escalating scientific interest. Their implementations in the chemical analysis have attracted at-tention of many researchers. the increasing demand for faster, more cost-effective and environmentally friendly analytical methods is a major reason to improve the classical procedures used for sample treatment in environmental analysis. on the other hand, the need for determination of trace elements requires a continued search for new analytical procedures. it is now clear that the dissolved free ionic metal spe-cies is far more bio-available than most complexed species. Dialysis has been used to differentiate among the various chemical forms of soluble trace ions present in the waters. in this article, membrane methods exploiting dialysis and their practical applications are considered.

Dialysis is a spontaneous process of separation of components in a mixture by passing them through a semi-permeable membrane under the influence of a concentra-tion gradient. since its introduction, dialysis came to be used as a simple laboratory separation method. Further, in the mid of the last century, medical and engineering

759

researchers made intensive effort for using dialysis to remove unwanted, relatively small molecules from blood. at present time, the process is generally used for treat-ment of patients by hemodialysis and peritoneal dialysis. Dialysis is effective in the separation of the colloidal compounds from low-molecular weight compounds in pharmaceutical4–6 and biochemical industries7,8, especially in the field of analysis of biological samples9–13. in the Donnan dialysis, also known as diffusion dialysis, the driving force is simply a difference in chemical potential. the process, involving transfer of species through ion-selective membranes, has implemented for desalina-tion and purification of different kind of waters14–30. the Donnan dialysis has become a useful separation technique for the enrichment, recovery of metal ions31–49 and for sample pre-concentration50–59. Diffusion dialysis has shown to be a very efficient way of recovering mineral acids from acid solutions containing dissolved metals60–65. Be-cause of its passive nature, as a result of which are used non external energy sources, operating costs are low. an attention has been paid to the exploitation of dialysis as an on-line membrane-based separation and pre-concentration process for automatic sampling and sample preparation in flow-injection analysis66–97.

as already noted, some of the more common applications of dialysis and its cur-rent development are presented.

thEorEtical PrinciPlEs

Dialysis. When a semi-permeable membrane is used to stabilise the concentration gradient between a solution on the one side and pure solvent on the other, the minimum requirements for simple dialysis have been met. the kinetic movement of the solute molecules will tend to drive them through the membrane in the direction of the lower concentration. on the other hand, as a result of the osmotic-pressure difference, the net movement of the solvent molecules will be in the opposite direction. however, depending on their size, shape, and other properties, the solute molecules may not be able to enter the membrane and will be completely excluded.

This concept of dialysis was firstly proposed by Graham in 1861, as a way of separating relatively small molecules from large ones98. the test solutes he used were sucrose and gum arabic in aqueous solution. Graham called the diffusible solutes crystalloids, and those that would not pass the membrane colloids.

The phenomenological theory of dialysis is based on the Fick’s first law of dif-fusion. ideally, for a given solution, membrane, temperature and physical setup, the overall rate of dialysis is proportional to the concentration gradient, according to the following equation: Ji = −Di (dCidх) (1)

where Ji is the diffusion flux; Di – the diffusion constant, Ci – the concentration of i-th component, and x – the distance.

760

The Donnan dialysis. if two electrolyte solutions of different composition are placed on each side of a selective membrane, only ions of the same charge can move through the membrane. the cation exchange membrane permits positive ions to migrate through it. the anion exchange membrane permits only passage of negatively charged ions. since the co-ions composition remains constant, there is a resulting loss of electroneutrality in the more concentrated solution, and charge potential builds across the membrane, which is opposite in direction to the concentration gradient. this charge potential provides the driving force to transfer stoichiometrically equal amount of counter-ions from the sample (feed, donor) across the membrane into the receiver (stripping, ac-ceptor). the process of ion exchange continues until the system comes to the Donnan equilibrium99, when the electrochemical potentials in aqueous and membrane phase are equal for each ion. it can be described by the generalised equation: (C r /C f )1/z

= K (2)

where Cr and C f are the activity of the given ion in the receiver and in the feed, re-spectively; z stands for the ion charge, and K – constant for all counter-ions presenting in the system.

if an electrolyte solution containing an univalent cation n+ and a noncomlexing anion is placed on the side of the membrane, while a different concentration of a second salt having the same anion, but a different cation c z+, is placed on the other side, the Donnan equilibrium can be written as: (C r

z+ / Cf

z+) = (N r+

/Nf +)z. (3)

thus, all cations tend to concentrate in the solutions containing the higher anion concentration; cations of higher charge tend to concentrate preferentially over those of lower charge. a completely analogous treatment of the behaviour of anions in a system containing an anion permeable membrane is also valid.

This process was firstly called the Donnan dialysis by Wallace in 1967 (Ref. 100), who extended the study to the development of dialysis for the continuous concentra-tion and separation of ions in solutions. these techniques were used to concentrate uranyl ions from dilute solutions, remove excess acid from a typical feed solution, concentrate trivalent ions such a lanthanum, remove strontium from sodium nitrate solutions, and separate silver and copper ions with a process based on differences in their ionic charge, and with processes based on differences on the stability of their complexes with different ligands.

Diffusion dialysis. consider separation of acid from its metal contaminants via an acid concentration gradient between two solution compartments (contaminated with metals acid and deionised water) that are divided by an anion exchange membrane. the metals in the solution are unable to pass from the concentrate to the deionised water. the anions in the concentrate (chlorides, sulphates, nitrates, phosphates, etc.) are permitted passage. although positively charged, the hydrogen ions diffuse along with the disassociated acid anions. the passage of the hydrogen ions, which is success

761

of this process, is due to the small size of the hydrogen molecules and their mobility. The result of this process is an 80–95% recovery of the initial acid solution (somewhat diluted with deionised water) and 60–95% rejection of the metals.

Mass transfer characterisation. in literature there are many and various researches in the pursuit to exploit new applications of dialysis, mainly to obtain a selective and controllable mass transfer. several mathematical models describing the transport of ions across the membranes are presented in the journals. according to theories and experimental data, the rate of movement of any species through a membrane depends on many factors among which are: the membrane properties (kind of material, thick-ness and porosity), the dialyser configuration, the concentration, the volume and the chemical composition both of donor and recipient solution.

a number of authors31,101–103 have dealt with the description of the process taking place in the dialysis. several mathematical models concerning the phenomenon have been suggested. one of the basic relationships in the development of dialysis theory is the Nernst–Planck equation describing the ionic transport: Ji = –Di(dCi/dx – ziCiDi (F/RT) dϕi/dx) (4)

where Ji and Ci are the flux and the concentration in the membrane; Di – the diffu-sion coefficient; zi – the ionic charge; ϕi – the electrical potential; x – the distance of the membrane; F, R and T – the Faraday, gas constant and the temperature, respec-tively.

the Donnan dialysis is a continuous process and the time necessary to reach the equilibrium is too long. The fact that the Nernst–Planck equation only describes mutually independent fluxes through the membrane restricts its application. For that reason the fluxes are evaluated experimentally by linear regression techniques from the time profile of the receiver concentration of target ions and/or from the decrease of ion concentration in the sample: J = − (V/A) (dC/dt), (5)

where V means the volume of the sample solution; A – the effective membrane area, and dC/dt – the concentration gradient. The negative sign indicates that the direction of the ionic flux is towards a decreasing concentration.

The dialysis efficiency is reported by means of enrichment factor (EF). It is defined as follows: EF = (Ct /C0) ×100 (%), where Ct is the analyte concentration at time t, and C0 is initial concentration in the sample.

the recovery factor (rF) is calculated by the equation: rF = (1 – Ct /C0) ×100 (%).

Membranes. Classification of the membranes can be presented by means of their structure and chemical composition. according to the generally accepted distinction, there are two types of membranes: porous and non-porous. the porous membranes are the most utilised in the laboratories and industry. concerning their surfaces they act as inert or reactive. separation with inert membranes is based on the molecular size dif-

762

ferences. reactive ion-exchange membranes are of great interest here. heterogeneous ion-exchange membranes are prepared through the calendaring of small ion-exchange particles into an inert plastic film, mainly polyethylene. The heterogeneous functional groups provide good mechanical strength, whereas the homogeneous functional groups supply excellent electrochemical properties104–106. in homogeneous membranes the functional groups are chemically bonded with the membrane polymer chains. cation exchange membranes are often obtained by sulphonation, whereas anion-exchange membranes are obtained through chloromethylation and quaternary amination.

aPPlications

Removal of harmful ions. Today, dialysis processes are used as efficient tools in water desalination and wastewater purification. The concentration of the substances is an important parameter for water quality since in most cases the assimilation of ions by biological systems ensues from the free ion state rather than directly from complexes. Bioavailability is an important concept in metal risk assessment. Furthermore, the inorganic ligands in natural waters such as F−, cl−, so4

2−, co32−, hco3

−, may affect metal transport, and thus modify biological transport. Examples of applications where the Donnan dialysis has been studied include removal of fluoride from dilute solu-tions14–18 and troublesome ions19–30. the process is a function of concentration, ph, conduct time, membrane structure and effect of accompanying ions.

the transport rate of the ions through a perm-selective membrane greatly depends on the membrane material. For the nitrate removal from sulphate-containing water to make it drinkable, the neosepta acs appears to be the best choice20. the same anion exchange membrane proved to be a good selection for perchlorate, but was not suitable for arsenate, for which the ionac Ma-3475 allowed more than a three-order magnitude increase in flux21. For each of four examined membranes (selemion aMv and DMv; Neosepta AFN and AMX) high efficiency of anions removal was obtained: 87–98% removal of nitrates, 94–100% removal of sulphates, whereas bicarbonates were re-moved with the efficiency of 77–99% (Ref. 22). In this research the best transport properties for the examined anions has shown the neosepta aFn membrane. the same authors reported that the flux of anions depends on the ion size23. taking into account the transfer rate, ions can be put in order: no3

−>so42−>hco3

−, and the highest removal efficiency in the Donnan dialysis can be observed for sulphates, and the lowest for hco3

− ions. the transfer of nitrate ions in the presence of chloride was enhanced using immersion-modified membrane by fixation of polyethyleneimine on the strip side of the membrane, which improves cl− transference24. the anion-exchanger behaviour of two conducting polymer films (N-[3-(dimethylpyridyl-2-yl) aminopropyl] polypyrrole and n-(3-aminopropyl) polypyrrole membranes) was investigated by French scientists under the Donnan dialysis conditions25. The experiments confirmed the transport of the monovalent anion (cl−, no3

–) through the polymer film, however the transport of the divalent anion (so4

2–) is not observed. the order for recovery of boron for membranes

763

was found to be as neosepta aFn>aMh>aha (ref.26). the highest boron removal was obtained when the pH of the feed phase was 9.5, and the reverse flow transition was accelerated by counter ions; the hco3

– ion was found to be more effective than cl– and so4

2– ions. Membrane transport studies showed that the Donnan dialysis is suitable for obtaining water with concentrations of perchlorate and nitrate below the recommended levels. Portuguese scientists have reported that the ion-exchange membrane bioreactor (iEMB) process allows for the most selective removal of the target pollutant simultaneously avoiding microbial and secondary contamination of the target water stream27. a novel treatment method for the removal of arsenic (as) and mercury (hg) is this bioreactor process, which incorporates pollutant transport through an ion-exchange membrane by the Donnan dialysis, with biological removal of the pollutant28. this combination is suitable for simultaneous removal of perchlorate and nitrate from drinking water29. the Donnan dialysis with cation-exchange membrane for hardness removal of water before electrodialytic desalination is reported30. 87% of the calcium and magnesium ions (with the use of selemion cMv membrane) can be efficiently exchanged for neutral sodium ions. Schematic illustration of the Donnan dialysis removal of ions is presented in Fig. 1.

Anion-exchange membrane

Cl–

SO42–

HCO3–

Na +

Ca 2+

the feed the rece iver

Cation-exchange membrane

the feed the rece iver

Na +

Ca 2+

Mg 2+

Cl–

SO42–

BA

fig. 1. schematic representation of the Donnan dialysis removal of ions with aEM (a) and cEM (b) (refs 22, 23 and 30)

Separation, pre-concentration and recovery of ions. the ultimate purpose in any ana-lytical chemical procedure is to obtain greatest sensitivity and selectivity. to obtain correct analytical results the interferences must be overcome. in many application areas sample preparation is still the bottleneck of the analysis. classic environmental samples usually require particular pre-treatment step. the pre-treatment involves the transfer (isolation) of analytes from the primary matrix (sample) to the secondary matrix with the simultaneous removal of interferences and increase of concentration of the analytes in the receiving matrix to the value above the detection limit of the instrument (enrichment). the Donnan dialysis offers excellent pre-concentration and separation of various metal ions31–42, as well as enrichment of noble metals such as gold, copper and silver43–48. the pre-concentration is an obvious outcome when the volume of the receiver is significantly less than that of the sample. The aqueous solutions of

764

electrolytes obtained in the receiver chamber are compatible with electrochemical, spectroscopic and ion-chromatographic methods of analysis.

the chemical composition both of the sample and receiver solutions in terms of concentration and ionic compounds has a significant influence on the mass transfer efficiency31. The fluxes and recovery factor value of Al(III), Fe(III), Ti(IV) and Na(I) ions were enhanced with increasing concentration both in the feed phase and h+ ion concentration (hcl) in stripping phase32. The large fluxes of ions can be obtained using the monovalent driving ions in the Donnan dialysis, and the fluxes of bi- or higher-valent driving ions are generally smaller32–34. the recovery factor values of a trivalent chromium ion, in the presence of metals of different valences, decreased with the increasing of the metal valence and the transport was influenced with H+ ion concentration in the receiver phase34. The rate of transfer for zinc was about 25% higher than that of cobalt under the same experimental conditions (0.5 M hcl as a receiver solution, 0.1 M feed solution, and 5 h dialysis time)35, and the transfer rate was found to be greatly affected by the h+ concentration in the receiver solution and metal concentration in the feed solution. it is, therefore, possible to select the best condition for a large flux by adjusting the pH of the bulk solution36.

The chelating agents have also significant influence on the recovery of metals. For example, the transport of ti(iv) was increased, while the transport of Fe(iii) was decreased, when EDta was included37. The removal and fluxes of Al(III), Fe(III), ti(iv) and na(i) were enhanced with citrate, ethylenediaminetetraacetic acid and ni-triloacetic acid chelating agents in contrast were decreased with phosphate complexing agent38. the Donnan dialysis with complexing agents is a feasible method not only for the enrichment of metal ions, but also for their separation. to achieve a separation in multi-component systems complexing agents can be added to the feed solution as well as to the receiver. the order of effective complex agents added to the feed solu-tion containing cu(ii), ni(ii) and Fe(iii) is malonic acid > citric acid > oxalic acid > ethylenediaminetetraacetic acid, and the flux of Fe(III) was always lower than that for cu(ii) and ni(ii), because of the higher values of the stability constant of ferric ion complexes39. in this research the optimal results of Jni:Jcu:JFe can be attained about 30:10:1. in the presence of chelating agents in the feed phase the transport of Fe(iii) is decreased and the transport of ti(iv) is increased37. The fluxes of Cu(II) and Ni(II) were enhanced using EDta as the receiving agent in contrast with polyethylenimine (PEi) (ref. 40). For a cu2+/ni2+/co2+ ternary ion system the order of effective com-plexing agents is citric acid> malonic acid > oxalic acid41. the EF value of ni2+ as a single as well as mixture state was higher than that of Fe3+ (ref. 42), and the effect of accompanying anion on the transport of nickel was found to be in the following order: cl−> no3

−> so42–. the stoichiometric ratio of complexing agent to metal ions

is the primary factor on the selective transport behaviour of metal ions. For example, as the feed concentration of Ti(IV) was increased, the flux also increased proportion-ally37. the transport rate of chlorocomplexes of noble metals is considerably reduced, because of their high affinity for the tetraalkylammonium groups43,44. the tendency

765

for retention in the membrane phase of these analytes is in the order: rh (iii) ~ ir (iii) < Pd (ii) < ir (iv) < Pt (iv) < au (iii) (ref. 43).

the osmosis phenomenon becomes an important inconvenience for the classical Donnan dialysis, but the method can be improved by the combination of ion-exchange membranes with cation-exchange textiles. this hybrid process has yielded better selectivity, increased the enrichment factors and it has given rise to a very efficient separation–concentration of copper and silver49.

Recovery of acids. Because of its low energy consumption dialysis has a wide applica-tion as a simple separation technique giving possibilities for repeated utilisation of the substances. the membrane charge is one of the most important factors that affect the permeation of ions. cation-exchange membranes are used to reclaim metals with the process generally being termed the Donnan dialysis. anion-exchange membranes are utilised to recover acids with the process commonly referred to as diffusion dialysis. the distinction is arbitrary since both are basically diffusion processes and both are subjects of the same Donnan criteria of co-ion rejection and preservation of electrical neutrality (Fig. 2). Diffusion dialysis techniques are generally used to remove metals contamination from concentrated solutions.the process units are an alternative for the conventional industrial wastewater treatment and discharge of strongly acidic and basic solutions used in metal preparation processes. Diffusion dialysis is utilised for a recovery and a concentration of zn2+ and cu2+ from electroplating rinse solutions containing mixtures of different metals60, recovery of mixed acid (hF+hno3) from the titanium spent leaching solutions61, separation of h2so4+cuso4 (ref. 62) and h2so4+znso4 (ref. 63), recovery of h2so4 from waste sulphuric acid solution con-taining Fe and ni ions produced at the diamond manufacturing process64. the process has been used for recovering various inorganic acids from waste streams, because of operational simplicity, compatibility and its particularly economic advantages in terms of energy saving64,65.

Anion- exchange membrane Cation- exchange membrane

H+

An–

Mm +

contamina ted ac id rec la imed ac id

DI waterMeta l rich produc t

B

Enriched e ffluen t Deple ted e ffluen t

Mm +

An–

H+

An–

ac id s o lu tion dilu ted s o lu tion

A

fig. 2. schematic representation of recovery of acids by Diffusion dialysis (a) and the Donnan dialysis pre-concentration of metal ions (b)

Application to the methods of analysis. the Donnan dialysis has been shown to be a useful quantitative technique for matrix normalisation and ion pre-concentration in the determination of ionic species50–59. The quantification is based on linear calibration

766

curves that are generated with a fixed-time kinetic model. One of the most important parameters offering the process is the concentration enrichment factor (EF). EFs of 20-fold are obtained for 1-hour pre-concentration of cu, ni, co, cd and Mn prior to separation by ion chromatography51. For chloride, phosphate, nitrate and sulphate the enrichment factors were 16, 15, 14 and 14, respectively, when 30-min dialysis into 5-ml receiver was used53. the value of enrichment factor can be controlled by addition of complexing ligand. the transport rate of chlorocomplexes of noble metals across anion-exchange membrane can be reduced in the presence of 50-fold excess of thiocyanate as a sample matrix, and 85% of the Au(III) can be reduced by a fac-tor 7 after 2-hour Donnan dialysis43. For 10-min dialysis enrichment factor above 3 was achieved and detection limit of 35 ng ml–1 was obtained for atomic absorption spectrophotometric determination of gold44. the Donnan dialysis has been applied as a sample interface in an optical sensor for cr(vi) monitoring. in this work experi-ments with uv-vis. detection have shown that the membrane Raipore R1030–DPC (1,5-diphenylcarbazide) solution as stripping phase system exhibits features suitable for cr(vi) optical sensing54. in fractionation studies the Donnan dialysis has been recently applied for treatment of soil solutions. Determining of trace metal speciation in aqueous environmental matrices such as free zn2+, cd2+, cu2+, and Pb2+ activities in pore waters from agricultural and long-term contaminated soils was reported55 and a determination of labile cu in soils and isotopic exchangeability of colloidal cu complexes56.

the static receiver approach has disadvantages of yielding low pre-concentration. With tubular membranes and receiver recirculation high enrichments in short times have been obtained by means of decreasing the receiver volume-to-surface ratio while using volumes of milliliter range50,52,57–59. Furthermore, the tubular membranes are readily interfaced to various detectors in on-line fashion. the compatibility of dialysis with different methods of analysis is given in table 1. however, to date separations with reactive membranes have relatively seldom been used in chemical analysis. Flow-through dialysis systems are presented in Figs 3 and 4.

table 1. typical application of the Donnan dialysis pre-concentrationanalyte Matrix Detection technique references

Gold complexes aqueous solutions containing platinum group metals

Faas 43

au noble metals mixtures Faas 44cu, ni, co, cd, Mn aqueous samples ion chromatography 51cu aqueous samples ion-selective potentiom-

etry, Faas52

cl–, Po43–, no3

–, so42– aqueous samples ion chromatography 53

Pb drinking water Faas 58sr, Mg water samples icPaEs 59Pb sweeteners Faas 84

767

receiverpump

dialysateion-exchangemembrane

sample

fig. 3. Concentric type dialyser having reactive ion-exchange membrane (Refs 31, 44, 50, 59 and 84)

pump

samplereceiver

membrane

fig. 4. Typical dialysis flow system with circulated receiver and circulated sample (Refs 31, 52 and 53)

Dialysis membranes are suitable to be incorporated as а part of manifold systems for continuous flow, flow injection, sequential injection and even process analysis obviating traditional manual operations which are time-consuming, tedious, expensive and operator-intensive when are used in routine laboratories. the extensive studies have shown that the on-line sample pre-treatment offers a number of benefits in comparison with manual operations. These advantages are high efficiency in terms of throughput or sampling frequency, reduced sample and reagent consumption and waste production, improved precision (rsD), low limit of detections (loDs), minimal risk of sample contamination, improved selectivity, automation, etc.66 using dialysis techniques it is possible to analyse coloured samples, such as red wines and fruit juice67–71. Besides, sample processing in an inert closed system prevents technicians coming into contact with hazardous materials1,66. in this point of view Fia has taken a prominent position in the contemporary analysis.

The first implementation of dialysis separation in flow injection analysis was performed in 1976 by hansen and ruzicka72. the authors reported for a determi-nation of phosphate and chloride in blood serum based on continuous spectropho-tometry adapted to Fia. the basic theoretical principles of mass transport through semi-permeable membranes were studied successfully by the groups of johanssen in sweden73,74. Mathematical models describing the mass transfer in parallel-plate flow-through dialysers were offered in sequence of articles by kolev and van der linden in the netherlands75–78. the groups of van staden in south africa concentrated more on the practical aspects79,80. The dialysis flow injection systems offer a decrease of dilution factors and/or increase of dialysis yield. the rate of movement of diffusible species depends on the dialyser configuration, geometriсal dimensions of the cell and path length, carrier flow rate, flow direction, dialysis time and sample volume79–82,

768

the membrane properties70, the chemical composition both of donor and acceptor solutions82.

The flow injection Donnan dialysis (FIDD) pre-concentration of lead in drink-ing water demonstrated that 5-min dialysis provided 100-fold enrichment and loD improvement factors than direct aspiration with flame atomic absorption spectropho-tometry (FAAS) (Ref. 58). The on-line combination of FIDD with inductively-coupled plasma atomic emission spectrometry (icP-aEs) is shown to provide enrichment factors of over 200 for cations with 8-min dialysis time59. A flow injection-based on-line iminodiacetic acid–ethylcellulose (IDAEC) membrane pre-concentration and separation technique was reported in Ref. 83. The technique was successfully applied to pond and seawater analysis in which all elements studied are in ng ml−1 level. the flow-injection Donnan dialysis is demonstrated for the extraction of lead in sweeteners, such as sucrose, corn syrup, and honey, using flame atomic absorption spectroscopy (FAAS) (Ref. 84). For a 15-min dialysis procedure the LOD is 350 ng/g. an improved dialysis efficiency of the analyte ion is attained by exploitation of either the Donnan effect of passive dialysis or the fast migration of ions concomitantly present with the target species82. thus, the addition of cationic species with high transport index, such as oxonium ion, to the donor stream, or multicharged ions such as Аl3+ to the recipient stream, enhances the dialysis yield more than 62% (Ref. 82). Illustration of continuous flow set-up assembling a parallel-plate dialysis unit with a secondary flow injection system is presented in Fig. 5.

SV

W

W

sample

water

matrix modifieracceptor stream

carrier

reagent

PP

secondary flow system

sandwich-type dialyser

W

D WRC

MC

PP

membrane

IV

fig. 5. Schematic illustration of the flowing stream manifold with a parallel-plate dialyser SV – switching valve; IV – injection valve; PP – peristaltic pump; MC – mixing coil; RC – reaction coil; D – detector; W – waste (Ref. 82)

over the last few years, new methodologies have appeared to exploit the util-ity of the second generation of FIA – sequential injection analysis (SIA) for on-line sample pre-treatment and handling. the advantage of sia is that the volume of the sample used is much smaller and the consumption of reagent is much lower than that of conventional Fia. thus the analysis is more cost-effective. iron(iii) was separated from a sample matrix by dialysis in sequential injection system85. the dialysed ion was complexed with iron and then monitored spectrophotometrically. a combined

769

dilution–dialysis method, utilising two SIA manifolds (donor and acceptor stream), was suggested for more complicated reagent handling86. a dialyser unit, equipped with a passive neutral membrane, was incorporated into the conduits of a sequential injection (si) system for the on-line removal of suspended solids and simultaneous dilution of the analyte before reaction and detection of the analyte. the system was applied to the determination of zinc(ii) in fertilisers. For chloride determination on-line dialysis through a cellulose membrane was used to enable sample dilution and matrix separation87. Further development of sia is the third generation of Fia – lab-on-valve (LOV) scheme. The LOV is an ideal facility for performing on-line sample pre-treatments.

the current trend in chemical analysis is to simplify, accelerate and/or miniaturise the pre-treatment procedures. Micro-dialysis is widely used in the analysis of natural and biological samples88,89. Flow-through micro-dialysis has been traditionally used to monitor the pharmacokinetics and distribution of drugs90, as well as for dynamic monitoring of chemical events in living tissues91. recently the micro-dialysis has been employed as a sampling and sample clean-up technique in the biotechnology field for on-line monitoring of bioprocesses92,93 , for a quality control of food stuffs94, in the environmental analysis for automatic micro-sampling of transition metals (cd, cr, cu, ni and Pb) (ref. 95), for determination of chloride ions in soil samples96 and for probing low-molecular weight organic anions, such as oxalate and citrate in solid samples97.

conclusions

In conclusion, a brief review of the fields of applications of dialysis methods is pro-posed. as one of the most simple, economical and energy-saving separation processes, dialysis has considerable great range of applications in environmental and clinical analysis. Due to the possibility for full automation and interfacing with different detection methods, dialysis has found widespread use. For that reason the further development of dialysis techniques is expected to increase in the near future.

acknoWlEDGEMEnts

summarising the review of up-to-date publications on the analytical applications of dialysis, it must be emphasised that this article can not cover all of the papers pub-lished, but the author is grateful to the several scientists cited in the reference below for their help!

rEFErEncEs1. M. Miro, W. FrEnzEl: automated Membrane-based sampling and sample Preparation Exploiting

Flow-injection analysis. trends in analytical chemistry, 23, 624 (2004).

770

2. l. n. Moskvin, t. G nikitina: Membrane Methods of substance separation in analytical chemistry. j. of analytical chemistry, 59, 6 (2004).

3. n. jakuBoWska, z. PolkoWska, j. naMiEsnik: analytical applications of Membrane Extraction for Biomedical and Environmental liquid sample Preparation. critical review of ana-lytical chemistry, 35, 217 (2005).

4. k. johansEn, M. kroGh, k. E. rasMussEn: automated on-line Dialysis, trace Enrichment and high-performance liquid chromatography inhibition of interaction with the Dialysis Mem-brane and Disruption of Protein Binding in the Determination of clozapine in human Plasma. j. of chromatography, B: Biomedical sciences and applications, 690, 223 (1997).

5. t. zuPancic, B. Pihlar: Preconcentration of Quinolones by Dialysis on-line coupled to high-performance liquid chromatography. j. of chromatography, a, 840, 11 (1999).

6. t. zuPancic, B. Pihlar: optimization of a Dialytic set-up for liquid chromatography: au-tomated Separation and Preconcentration of Ciprofloxacin. J. of Chromatography, A, 975, 199 (2002).

7. W. c. DuanE, D. P. GilBoE: Measurement of Bile salt aggregation Equilibria using kinetic Dialysis and spreadsheet Modeling. analytical Biochemistry, 229, 15 (1995).

8. A. ZHELEZNOV, D. WINDMOLLER, S. kORNER, k. W. BODDEkER: Dialytic Transport of carboxylic acids through an anion Exchange Membrane. j. of Membrane science, 139, 137 (1998).

9. j. h. alDstaDt, D. F. kinG, h. D. DEWalD: Flow injection Potentiometric and voltammetric stripping analysis using a Dialysis Membrane covered Mercury Film Electrode. analyst, 119, 1813 (1994).

10. M. jurkiEWicz, s. solE, j. alMirall, M. Garcia, s. alEGrEt, E. MartinEz-FaBrE-Gas: validation of an automatic urea analyser used in the continuous Monitoring of hemodialysis Parameters. analyst, 121, 959 (1996).

11. s. zaMPoni, B. lo cicEro, M. Mascini, l. DElla ciana, s. sacco: urea solid-state Biosensor suitable for continuous Dialysis control. talanta, 43, 1373 (1996).

12. j. r. vEraart, M. c. E. Groot, c. GooijEr, h. linGEMan, n. h. vElthorst, u. a. t. BRINkMAN: On-line Dialysis–SPE–CE of Acidic Drugs in Biological Samples. The Analyst, 124, 115 (1999).

13. P. hEss, D. E. WElls: Evaluation of Dialysis as a technique for the removal of lipids Prior to the Gc Determination of ortho- and non-ortho-chlorobiphenyls, using 14c-labelled congeners. the analyst, 126, 829 (2001).

14. M. hichour, F. PErsin, j. MolEnat, j. sansEaux, c. Gavach: Fluoride removal from Diluted solutions by Donnan Dialysis with anion-exchange Membranes. Desalination, 122, 53 (1999).

15. M. hichour, F. PErsin, j. MolEnat, j. sansEaux, c. Gavach: Fluoride removal from Waters by Donnan Dialysis. Separation and Purification Technology, 18, 1 (2000).

16. H. GARMES, F. PERSIN, J. SANDEAuX, G. POuRCELLY, M. MOuNTADAR: Defluoridation of Groundwater by a hybrid Process combining adsorption and Donnan Dialysis. Desalination, 145, 287 (2002).

17. t. ruiz, F. PErsin, M. hichour, j. sanDEaux: Modelisation of Fluoride removal in Donnan Dialysis. j. of Membrane science, 212, 113 (2003).

18. F. DuRMAZ, H. kARA, Y. CENGELOGLu, M. ERSOZ: Fluoride Removal by Donnan Dialysis with anion Exchange Membranes. Desalination, 177, 51 (2005).

19. k. salEM, j. sanDEaux, j. MolEnat, r. sanDEaux, c. Gavach: Elimination of nitrate from Drinking Water by Electrochemical Membrane Processes. Desalination, 101, 123 (1995).

20. P. schaEtzEl, D. aManG, n. nWal, t. QuanG: Batch ion-exchange Dialysis to Extract Nitrate from Drinking Water: A Simplified Ion Transport Model for the Best Membrane Selection. Desalination, 164, 261 (2004).

771

21. s. vElizarov, c. Matos, M. a. rEis, j. G. crEsPo: removal of inorganic charged Micropol-lutants in an ion-exchange Membrane Bioreactor. Desalination, 178, 203 (2005).

22. j. WisniEvski, a. rozanska, t. Winnicki: removal of troublesome anions from Water by Means of Donnan Dialysis. Desalination, 182, 339 (2005).

23. j. WisniEvski, a. rozanska: Donnan Dialysis with anion-exchange Membranes as a Pretreat-ment step before Electrodialytic Desalination. Desalination, 191, 210 (2006).

24. M. aMara, h. kErDjouDj. Electro-adsorption of Polyethileneimine on the anion Exchange Membrane: application to the nitrate removal from loaded solutions. analytica chimica acta, 547, 50 (2005).

25. a. naji, M. crEtin, M. PErsin, j. sarrazin: Electrical characterization of the ionic inter-actions in n-[3-dimethylpyridil-2-il) aminopropil] Polypirrole and n-(3-aminopropyl) Polypirrole Membranes. j. of Membrane science, 212, 1 (2003).

26. h. F. ayyilDiz, h. kara: Boron removal by ion Exchange Membranes. Desalination, 180, 99 (2005).

27. s. vElizarov, M. a. rEis, j. G. crEsPo: integrated transport and reaction in an ion Exchange Membrane Bioreactor. Desalination, 149, 205 (2002).

28. A. OEHMEN, s. vElizarov, M. a. M. rEis, j. G. crEsPo: removal of heavy Metals from Drinking Water supplies through the ion Exchange Membrane Bioreactor. Desalination, 199, 405 (2006).

29. c. t. Matos, s. vElizarov, j. G. crEsPo, M. a. M. rEis: simultaneous removal of Per-chlorate and nitrate from Drinking Water using the ion Exchange Membrane Bioreactor concept. Water research, 40, 231 (2006).

30. j. WisniEWski, a. rozanska: Donnan Dialysis for hardness removal from Water before Electrodialytic Desalination. Desalination, 212, 251 (2007).

31. k. Pirzinska: Pre-concentration and recovery of Metal ions by Donnan Dialysis. Microchimica acta, 153, 117 (2006).

32. y. cEnGEloGlu, E. kir, M. Ersoz, t. BuyukErkEk, s. GEzGin: recovery and concentra-tion of Metals from red Mud by Donnan Dialysis. colloids and surfaces, a: Physical Engineering aspects, 223, 95 (2003).

33. H. MIYOSHI: Donnan Dialysis with Ion-exchange Membranes. III. Diffusion Coeficients using ions of Different valence. separation science and technology, 34, 231 (1999).

34. a. tor, y. cEnGEloGlu, M. Ersoz, G. arsalan: transport of chromium through cation-exchange Membranes by Donnan Dialysis in the Presence of some Metals of Different valences. Desalination, 170, 151 (2004).

35. i. alExanDrova, G. iorDanov: transport of cadmium and iron through a carboxylic Mem-brane Based on a Poly(vinyl chloride)/poly(methyl methacrylate-co-divinyl benzene) system. j. of applied Polymer science, 95, 705 (2005).

36. r. takaGi, M. nakaGaki: ionic Dialysis through amphoteric Membranes. separation and Purification Technology, 32, 65 (2003).

37. E. kir, y. cEnGEloGlu, M. Ersoz: the Effect of chelating agents on the separation of Fe(iii) and ti(iv) from Binary Mixture solution by cation-exchange Membrane. j. of colloid and interface science, 292, 498 (2005).

38. E. kIR, Y. CENGELOGLu, M. ERSOZ: Influence of Chelating Agents on the Recovery of Al(III), Fe(iii), ti(iv) and na(i) from red Mud by cation Exchange Membrane. separation science and technology, 41, 961 (2006).

39. j.-k. WanG: Preferential transport Behaviors of ternary system Ferric-cupric-nickel ions through cation ion Exchange Membrane with a complex agent by Dialysis. Desalination, 161, 277 (2004).

40. r.-s. juanG, G.-s. yan: Enhanced Flux and selectivity of Metals through a Dialysis Membrane by addition of complexing agents to receiving Phase. j. of Membrane science, 186, 53 (2001).

772

41. j.-k. WanG, c.-P. chanG: selective transport of a cupric-nickel-cobalt ion ternary system through cation ion-exchange Membranes with a complexing agent by Dialysis. Desalination, 164, 269 (2004).

42. E. kir, a. tor, y. cEnGEloGlu, M. Ersoz: the Effect of accompanying anion and the competitive transport of ni(ii) and Fe(iii) through Polysulfone Membranes. separation science and technology, 38, 2503 (2003).

43. k. Pyrzinska: Membrane Method for Preconcentrating and separating Gold complexes from aqueous solutions containing other Platinum Group Metals. analytica chimica acta, 255, 169 (1991).

44. k. Pyrzinska: atomic absorption spectrophotometric Determination of Gold with Preconcentra-tion by Donnan Dialysis. talanta, 41, 381 (1994).

45. a. t. chEriF, c. Gavach, j. MolEnat, a. ElMiDaoui: transport and separation of ag+ and zn2+ by Donnan Dialysis through a Monovalent cation selective Membrane. talanta, 46, 1605 (1998).

46. D.-E. acrEtchE, h. kErDjouDj: Donnan Dialysis of copper, Gold and silver cyanides with various anion Exchange Membranes. talanta, 51, 281 (2000).

47. y. yanG, P. n. Pintauro: Multicomponent space-charge transport Model for ion-exchange Membranes with variable Pore Properties. industrial and Engineering chemical research, 43, 2957 (2004).

48. R. WODZkI, P. szczEPanski: simultaneous recovery and separation of zn2+ and cu2+ in hybrid Membrane Systems. Separation and Purification Technology, 41, 289 (2005).

49. BEDrous, D. E. akrEtchE: recovery of Metals by Donnan Dialysis with ion Exchange tex-tiles. Desalination, 144, 213 (2002).

50. j. a. cox, z. tWarDoWski: tubular Flow Donnan Dialysis. analytical chemistry, 52, 1503 (1980).

51. j. E. Dinunzio, M. juBara: Donnan Dialysis Preconcentration for ion chromatography. ana-lytical chemistry, 55, 1013 (1983).

52. j. a. cox, G. r. litWinski: high sample convection Donnan Dialysis. analytical chemistry, 55, 1640 (1983).

53. j. a. cox, n. tanaka: Donnan Dialysis Preconcentration for the ion chromatography. analytical chemistry, 57, 2370 (1985).

54. E. castillo, M. GranaDos, j. l. cortina: chemically Facilitated chromium(vi) transport throughout an anion-exchange Membrane. analytica chimica acta, 464, 15 (2002).

55. a. l. nolan, M. j. MclauGhlin, s. D. Mason: chemical speciation of zn, cd, cu, and Pb in Pore Waters of agricultural and contaminated soils using Donnan Dialysis. Environmental science and technology, 37, 90 (2003).

56. y. B. Ma, E. loMBi, a. l. nolan, M. j. MclauGhlin: Determination of labile cu in soils and isotopic Exchangeability of colloidal cu complexes. European j. of soil science, 57, 147 (2006).

57. j. a. koroPchak, E. DaBEk-zlotorzynska: tubular Donnan Dialysis-inductively coupled Plasma atomic Emission spectrometry. analytical chemistry, 60, 328 (1988).

58. J. A. kOROPCHAk, L. ALLEN: Flow Injection Donnan Dialysis Preconcentration of Cations for Flame atomic absorption spectrophotometry. analytical chemistry, 61, 1410 (1989).

59. n. kasthurkrishnan, j. a. koroPchak: Flow injection Donnan Dialysis Preconcentration of trace Metal cations for inductively coupled Plasma atomic Emission spectrometry. analytical chemistry, 65, 857 (1993).

60. r. WoDzki, P. szczEPanski: treatment of Electroplating rinse solution by continuous Mem-brane Extraction and Diffusion Dialysis. Polish j. of Environmental studies, 10, 101 (2001).

61. t. xu, W. yanG: industrial recovery of Mixed acid (hF + hno3) from the titanium spent leaching solutions by Diffusion Dialysis with a new series of anion Exchange Membranes. j. of Membrane science, 220, 89 (2003).

773

62. z. z Palaty, a. zakova: separation of h2so4 + cuso4 Mixture by Diffusion Dialysis. j. of hazardous Materials, 114, 69 (2004).

63. z. z Palaty, a. zakova: separation of h2so4 + znso4 Mixture by Diffusion Dialysis. Desalina-tion, 169, 277 (2004).

64. j. jEonG, M.-s. kiM, B.-s. kiM, s.-k. kiM, W.-B. kiM, j.-c. lEE: recovery of h2so4 from Waste acid solution by Diffusion Dialysis Method. j. of hazardous Materials, В124, 230 (2005).

65. E. G. akGEMci, M. Ersoz, t. atalay: transport of Formic acid through anion Exchange Membranes by Diffusion Dialysis and Electro-electro Dialysis. separation science and technology, 39, 165 (2004).

66. l. WanG, E. h. hansEn: on-line sample-pre-treatment schemes for trace-level Determinations of Metals by coupling Flow injection or sequential injection with icP-Ms. trends in analytical chemistry, 22 (11), 837 (2003).

67. s. Morais, M. i. alcaina-MiranDa, F. lazaro, M. Planta, a. MaQuiEira, r. PuchaDEs: Evaluation of the Dialyzing yield of Membranes with Different composition. application to the analysis of chloride in Fruit juices by Flow injection. analytica chimica acta, 353, 245 (1997).

68. C. LINGET, C. NETTER, D. HEEMS, E. vErEttE: on-line Dialysis with hPlc for the automated Preparation and analysis of amino acids, sugars and organic acids in Grape juice and Wines. analysis, 26, 35 (1998).

69. a. n. araujo, j. l. F. c. liMa, a. o. s. s. ranGEl, M. a. sEGunDo: sequential injection system for the spectrophotometric Determination of reducing sugars in Wines. talanta, 52, 59 (2000).

70. E. MATAIX, M. D. LuQuE de CASTRO: Determination of L-(−)-malic Acid and L-(+)-lactic Acid in Wine by a Flow injection-dialysis-enzymic Derivatisation approach. analytica chimica acta, 428, 7 (2001).

71. h. a. D. F. o. silva, l. M. B. c. alvarEz-riBEiro: optimization of a Flow injection analysis system for tartaric acid Determination in Wines. talanta, 58, 1311 (2002).

72. E. h. hansEn, j. ruzicka: Flow injection analysis Part vi. the Determination of Phosphate and chloride in Blood serum by Dialysis and sample Dilution. analytica chimica acta, 87, 353 (1976).

73. E. Martins, G. johansEn: solute transfer in on-line analytical Flow-through Dialyzers. analytica chimica acta, 167, 111 (1985).

74. l. risinGEr, G. johansson, t. thornEMan: optimization of Flow-through cells for Di-alysis and other Membrane separations in Flow-injection analysis using a laminar Flow Model. analytica chimica acta, 224, 13 (1989).

75. s. D. kolEv, W. E. van der linDEn: laminar Dispersion in Parallel Plate sections of Flow sys-tems used in analytical chemistry and chemical Engineering. analytica chimica acta, 247, 51 (1991).

76. s. D. kolEv, W. E. van der linDEn: limits of Mass transfer in Parallel-plate Dialysers. analytica chimica acta, 256, 301 (1992).

77. S. D. kOLEV, W. E. van der LINDEN: Influence of the Main Parameters of a Parallel-plate Dialyser under laminar Flow conditions. analytica chimica acta, 257, 317 (1992).

78. S. D. kOLEV, W. E. van der LINDEN: Analysis of Transient Laminar Mass Transfer in Parallel-plate Dialyser. analytica chimica acta, 257, 331 (1992).

79. j. F. van staDEn, c. j. hattinG: study of Mass transfer in on-line analytical Parallel-plate Dialysers in series in Flow injection systems. analytica chimica acta, 308, 222 (1995).

80. J. F. van STADEN: Membrane Separation in Flow Injection Systems. Fresenius j. of analytical chemistry, 352, 1432 (1995).

81. M. MIRO, W. FRENZEL: A Novel Flow-through Microdialysis Separation unit with Integrated Differential Potentiometric Detection for the Determination of chloride in soil samples. analyst, 128, 1291 (2003).

774

82. M. MIRO, W. FRENZEL: Investigation of Chemical Effects on the Performance of Flow-through Di-alysis applied to the Determination of ionic species. analytica chimica acta, 512, 311 (2004).

83. x. WanG, z. zuanG, c. yanG, F. zhuy: the application of Flow injection iminodiacetic acid-ethylcellulose Membrane Preconcentration and separation technique to atomic spectrometry. spectrochimica acta, b456, 1437 (1998).

84. A. ANTONIA, L. B. ALLEN: Extraction and Analysis of Lead in Sweeteners by Flow-injection Donnan Dialysis with Flame atomic absorption spectroscopy. j. of agriculture Food chemistry, 49, 4615 (2001).

85. J. F. van STADEN, H. du PLESSIS, R. E. TALJAARD: Determination of Iron (III) in Pharmaceutical samples using Dialysis in a sequential injection analysis system. analytica chimica acta, 357, 141 (1997).

86. J. F. van STADEN, S. I. TLOWANA: On-line Separation, Simultaneous Dilution and Spectrophoto-metric Determination of zinc in Fertilisers with a sequential injection system and xylenol orange as complexing agent. talanta, 58, 1115 (2002).

87. J. E. da SILVA, M. F. PIMENTEL, V. L. da SILVA, M. C. B. S. M. MONTENEGRO, A. N. ARAuJO: simultaneous Determination of ph, chloride and nickel in Electroplating Baths using sequential injection analysis. analytica chimica acta, 506, 197 (2004).

88. R. kONSkI, A. RADOMSkA, S. GLAB: Bioanalytical Flow-injection System for Control of hemodialysis adequacy. analytica chimica acta, 418, 213 (2000).

89. J. WANG, M. Lu, F. YANG, X. ZHANG, W. R. G. BAEYENS, A. M. G. CAMPANA: Microdialy-sis with on-line chemiluminescence Detection for the study of nitric oxide release in rat Brain Following traumatic injury. analytica chimica acta, 428, 173 (2001).

90. D. j. WEiss, c. E. luntE, s. M. luntE: In vivo Microdialysis as a tool for Monitoring Phar-macokinetics. trends in analytical chemistry, 19, 606 (2000).

91. M. c. Parkin, s. E. hoPWooD, M. G. BoutEllE, a. j. stronG: resolving Dynamic changes in Brain Metabolism using Biosensors and on-line Microdialysis. trends in analytical chemistry, 22, 487 (2003).

92. n. torto, l. Gorton, t. laurEll, G. Marko-varGa: technical issues of in vitro Micro-dialysis sampling in Bioprocess Monitoring. trends in analytical chemistry, 18, 252 (1999).

93. t. ButtlEr, h. liDEn, j. Å. jonsson, l. Gorton, G. Marko-varGa, h. jEPPsson: Evaluation of Detection and sample clean-up techniques for on- and off-line Fermentation Moni-toring systems. analytica chimica acta, 324, 103 (1996).

94. n. c. van de MErBEl, h. linGEMan, u. a. t. BrinkMan: sampling and analytical strate-gies in on-line Bioprocess Monitoring and control. j. of cromatography, A725, 13 (1996).

95. n. torto, j. MWatsEtEza, G. saWula: a study of Microdialysis sampling of Metal ions. analytica chimica acta, 456, 253 (2002).

96. M. Miro, W. FrEnzEl: implantable Flow-through capillary type Microdialyzers for con-tinuous in situ Monitoring of Environmental relevant Parameters. analytical chemistry, 76, 5974 (2004).

97. M. sulyok, M. Miro, G. stinGEDEr, G. kollEnsPErGEl: the Potential of Flow-through Microdialysis for Probing low-molecular Weight organic anions in rhizosphere soil solution. analytica chimica acta, 546, 1 (2005).

98. T. GRAHAM: Liquid Diffusion Applied to Analysis. In: Proc. of the Royal Society of London, 11, 243 (1861).

99. F. G. Donnan: the theory of Membrane Equilibrium. chemical reviews, 1, 73 (1924).100. r. M. WallacE: concentration and separation of ions by Donnan Membrane Equilibrium.

industrial and Engineering chemical Process Design and Development, 6, 423 (1967).101. s. t. hWanG, k. kaMMErMEyEr: Membranes in separations. Wiley, new york, 1975.102. s. Mokrani, l. DaMMak, G. BulvEstrE, c. larchEt, B. auclair: Experimental

and theoretical studies of the crossed ionic Fluxes through a cation-exchange Membrane. j. of Membrane science, 199, 147 (2002).

775

103. s. nouri, l. DaMMak, G. BulvEstrE, B. auclaiar: studies of the crossed ionic Fluxes through a cation-exchange Membrane in the case of Donnan Dialysis. Desalination, 148, 383 (2002).

104. k. hu, t. W. xu, W. h. yanG, y. Fu: Preparation of novel heterogeneous cation-permeable Membranes from Blends of Sulfonated Poly (phenylene sulfide) and Poly (ether sulfone). J. of Ap-plied Polymer science, 91, 167 (2004).

105. k. hu, t. W. xu, y. Fu, W. yanG: Preparation of novel semihomogeneous cation-permeable Membranes from Blends of Sulfonated Poly (phenylene sulfide) and Sulfonated Phenolphthalein Poly (ether ether cetone). j. of applied Polymer science, 92, 1478 (2004).

106. k. hu, t. W. xu, W. h. yanG, y. Fu: Effects of heat treatment on the Properties of heterogene-ous cation Permeable Membranes from Blends of Poly (ether sulfone)/sulfonated Poly (phenylene sulfide) and Phenolphthalein Poly (ether cetone)/Sulfonated Poly (phenilene sulfide). J. of Applied Polymer science, 98, 494 (2005).

Received 13 October 2007 Revised 22 November 2007

776

Oxidation Communications 31, No 4, 776–785 (2008)

* For correspondence.

prediction of blood-brAin bArrier permeAtion Using topologicAl descriptors

j. shrivastava, j. sinGh, B. shaik, v. k. aGraWal*

QSAR and Computer Chemical Laboratories, A. P. S. University, 486 003 Rewa, India E-mail: vijay-agraval@lycos.com; jyoti_singh07@rediffmail.com

aBstract

Blood-brain barrier activity in terms of lgBB has been modelled using topological descriptors. it was revealed that branching plays an important role in estimating lg BB whereas the Balaban and szeged indices are not favourable. the equalised electron-egativity plays a negative role in exhibition of lgBB.

Keywords: blood-brain barrier, topological index, equalised electronegativity, the randic, Balaban and szeged indices, regression analysis.

aiMs anD BackGrounD

the blood-brain (BB) barrier is a physical barrier responsible to prevent passage of undesired substances to the brain. it is situated between the blood vessels in the central nervous system, and the central nervous system itself. the blood-brain barrier (BBB) is a very specialised system of capillary endothelial cells which protects the brain from harmful substances in the blood stream, while supplying the brain with the required nutrients for proper function.

it is an open fact that this barrier protects the brain from the chemical messenger systems flowing around the body. In fact various functions of body are controlled by hormones which are detected by receptors on interested cells throughout the body. the hormones are released on cue from the brain, so if they acted on the brain it could cause problems.

Due to the ability to prevent harmful substances to enter into brain, the blood-brain barrier protects the brain even from common infections, that is why the brain infection is rarely observed.

Prediction of the blood-brain barrier (BBB) permeability is a very important factor. For a drug designer it is essential to understand the endothelial cells which separate the brain and the cerebrospinal flood from the systematic blood-circulation because it restricts the substances targeting the central nervous system (cns).

777

BioloGical activity in tErMs oF lg BB

the ratio of the steady-state balance of the concentrations of a substance in the brain and in the blood is commonly expressed as lg (Cbrain/Cblood). it is very similar to the expression for water/n-octanol partition coefficient (lg P) (Ref. 1). It is interesting to observe that the value of lgP falls between –4 to +8 whereas lgBB covers only nar-row range between –2.00 to +1.00. A compound with lgBB > 0.3 will cross the BBB readily, while values < –1.0 show a low concentration in the brain2,3. the blood-brain barrier penetration is a very important factor in any pharmaco-kinetic study and this property is expressed in term of lg BB.

Early models for the prediction of lg BB were based on quantitative4–12 struc-ture-activity relationship (Qsar) using a limited number of variables, e.g. lgP and molecular size descriptors like electro-topological state whereas other models used lgP and polar-surface area (Psa)8,9.

One of the pioneers in this field Abraham and coworkers13,14 introduced hydrogen bond donor/acceptor property of the solute and polarisability as co-relating parameters for the modelling of lgBB.

3D molecular descriptors(volsurf program) for proposing a BBB permeation model has also been used by crivori et al. and ooms et al.7,15

We have chosen 45 compounds16 whose lgBB activities are known and tried to model the lgBB activity using very simple topological descriptors which are com-monly used in modelling various types of drug activities.

coMPutations

in the present study we have used topological descriptors, viz. the Wiener index17, Balaban index18–21, szeged index22–25, equalised electronegativity26,27 and branching index28–30 to model lgBB for a set of 45 compounds known for their lgBB activity. the topological indices were calculated using DraGon software. also we have carried out regression analysis for obtaining statistically significant models using maximum R2 method.

thE WiEnEr inDEx (W) Wiener17 in his studies on physical parameters of acyclic hydrocarbons, introduced a path number W which he called Wiener number. The number W is defined as the number of bonds between all parts of atoms in an acyclic molecule in view of the pioneering contribution of Wiener in recognising the significance of the numbers of paths in a molecular skeleton seems appropriate to continue to call the number of distances in all (acyclic and cyclic) structure, the Wiener number W(G). however, one should be aware that one-to-one correspondence between the pairs of neighbours certain number of bonds away and the number of paths of the same length holds only for acyclic systems. hence, in polycyclic structures, the Wiener number (W) is asso-

778

ciated with distance only and not with the number of paths, but in acyclic structures the two are the same.

the Wiener number17 (W) is defined as followslet the vertices of the molecular graph G be labelled by 1,2,…n. let u and v be

the two vertices of graph G and let d(u,v) be the distance between them. then the Wiener number is equal to the sum of the distances between all pairs of the respec-tive graph, i.e. W(G) = W = ∑ d(u,v). (1) u<v

thE BalaBan inDEx (j)

Balaban18–21 proposed a topological index, numbered the Balaban index which rep-resents the extended connectivity. this index, denoted by j(G) = J, is defined as follows: j(G) = j = M/(µ+1) ∑(di, dj)–0.5 (2) edges

where M is the number of edges in a molecular graph G, and di (i = 1, 2,…, N; N is the number of vertices in G) is the distance sum. the distance sum for a vertex rep-resented the sum of all entries in the corresponding row (or column) of the distance matrix D:

N di = ∑ (D)i,j. . (3)

i=1

the cyclometric number µ of a polycyclic graph is equal to the minimum number of edges necessary to be erased from G in order to transform it to the related acyclic graph: µ = M – N + 1. (4)

(i) the diagonal elements: (D)ij = 1 – (Zc/Zi) (5)

where Zc = 6 and Zj is the atomic number of the given elements.(ii) the off-diagonal elements:

(D)ij = ∑ Kr (6)r

where the summation is over all bonds.the bond parameter kr is given by the following expression:

kr = (1/br) (Zc2/Zi.Zj) (7)

where br is the bond weight values and is 1 for single bond, 2 – for double bond, 3 – for triple bond and 1.5 – for aromatic bond. The values of (D)ij for various heteroatoms and kr for various types of heterobond are available in the literature.

779

thE szEGED inDEx (sz)

the szeged index sz(G) = sz is recently introduced topological index and very little is known on its use in QSPR as well as QSAR study. This index is a modification of the Wiener index for cyclic compound and is introduced by Gutman and coworkers22–25. the szeged index sz(G) = sz of a molecular graph G is defined as follows: sz(G) = Sz = ∑ nu nv (8)

edges

where nu is the number of edges lying closer to vertex u than v. the meaning of nv is analogous. the number of edges lying equidistant from u and v are not counted for the calculation of sz.

EQualisED ElEctronEGativity (χeq)

charge conservation equation leads to general expression for equalised electronega-tivity26,27 (χeq) as given below: χeq = N/Σ (V/χ) (9)

where N = Σ V = total number of atoms in the species, V – the number of atoms of a particular element in the species, and χ – the electronegativity of that element.

Now, the group negativity is defined as: χG = N0 /(V/χ) (10)

where N0 is the number of atoms in the group formula.note that groups are fundamentally different from atoms in their ability to donate

or withdraw charge. the important difference between atom and a group is that the groups have the ability to dissipate the charge over several atoms increasingly with increasing N0. a group can be treated as ‘pseudo atoms’ in electronegativity discus-sion because a polyatomic group can be considered as a reservoir of enhance charge capacity potentially able to withdraw considerable amount of charge with only small variation in electronegativity.

First order connectivity index (1χ). The connectivity index χ = χ(G) of a graph G is defined by Randic28 as follows: 1χ = 1χ(G) = Σ [δi δj] –0.5 (11)

ij

where δi and δj are the valence of a vertex i and j, equal to the number of bonds con-nected to the atoms i and j, in G.

Indicator parameter. the indicator parameter iP1 accounts for the presence of halogen in the structure. if the halogen is present it has been assigned a value of 1, otherwise 0.

780

rEsults anD Discussion

the compounds which are known drugs along with their biological activity value lg BB are reported in table 1. the value of indicator parameter (iP1) has also been reported in table 1. the star (*) denotes the compounds which are serious outliers. these outliers (compounds nos 11, 12, 17, 19, 23 and 43) were deleted while carry-ing out the regression analysis.

table 1. name of compounds, their lg BB values and values of indicator paramaterno name of compound lg BB iP1 no name of compound lgBB iP11 Difloxacin –0.14 1 24 levodopa –1.33 02 Rufloxacin –0.27 1 25 Dopamine –0.87 03 Ciprofloxacin –0.26 1 26 nordazepam 0.31 14 Pefloxacin –0.32 1 27 Pirenzepine –0.43 05 Sparfloxacin –0.16 1 28 loratadine 0.95 16 Iomefloxacin –0.15 1 29 Desa-loratadine 1.29 17 Fleroxacin –0.24 1 30 clobazam 0.09 18 Enoxacin –0.34 1 31 Perphenazine 1.01 19 Norfloxacin –0.28 1 32 Cis-flupentixol 0.62 1

10 Ofloxacin –0.28 1 33 Progesterone 0.57 011 Mefloquine* 0.35 1 34 testosterone 0.15 012 cetirizine* 0.28 1 35 Estradiol 0.06 013 ioperamide 0.22 1 36 corticosterone –0.40 014 terfenadine –0.47 0 37 cortisol –0.94 015 Fexofenadine –1.20 0 38 aldosterone –0.65 016 Diphenylhydramine 0.57 0 39 Piroxicam –0.92 017 Doxylamine* 0.77 0 40 tenoxicam –0.77 018 carebastine –0.63 0 41 Meloxicam –0.52 019 Ebastine* 0.02 0 42 isoxicam –0.77 020 carmoxirol –0.66 0 43 Flurosemide* –1.21 121 astemizol –0.19 1 44 rivastigmine 0.31 022 roxindole –0.62 0 45 Delavirdine –1.30 023 carbidopa* –1.58 0

* Denotes deleted case; iP1 = 1, if halogen is present in the compound, otherwise – 0.

the calculated parameters (W, sz, j, lg BB, 1χ ) are reported in Table 2.to understand the inter-correlation among various parameters used, we have

calculated correlation matrix which is reported in table 3. a close look of this table shows that the Wiener index is highly correlated with 1χ and Sz. Similarly, 1χ is highly correlated with sz. the Balaban index (j) shows a moderate correlation with 1χ and Sz, but the coefficient of correlation is negative.

781

table 2. value of various topological indices calculated for the compounds used in the present studycompd.

no W1χ j sz χeq

compd.no W

1χ j sz χeq

1 2102 13.8462 1.5344 3840 2.323 24 321 6.5029 2.3261 451 2.4232 1352 11.9692 1.3594 2970 2.405 25 160 5.2363 2.2753 248 2.3543 1234 11.5586 1.5315 2037 2.403 26 229 6.2540 1.5815 438 2.3954 1277 11.4179 1.7291 2330 2.387 27 1540 12.6310 1.3129 2617 2.3735 1512 11.9347 1.6465 2712 2.436 28 1599 12.7751 1.2728 2784 2.3446 1212 11.4516 1.8263 2211 2.432 29 918 10.8265 1.3890 1732 2.3237 1360 11.8700 1.8312 2434 2.447 30 837 10.0753 1.4889 1425 2.4018 1116 11.0241 1.7498 2048 2.446 31 2000 13.2415 1.2979 3418 2.1959 1116 11.0241 1.7498 2048 2.401 32 2618 14.4528 1.3442 4387 2.359

10 1484 12.3799 1.3838 3251 2.406 33 1052 10.8598 1.6030 2117 2.28211 1491 12.0873 1.8601 2536 2.472 34 802 9.9491 1.6231 1683 2.28212 2068 13.1310 1.4930 3070 2.471 35 724 9.5931 1.5650 1536 2.26813 3458 16.3488 1.4389 5135 2.319 36 1328 11.4525 1.6759 2347 2.30014 4424 16.7809 1.2109 6506 2.298 37 1598 12.2739 1.7460 2776 2.32215 5214 17.7242 1.2007 7602 2.313 38 1597 12.4012 1.7567 2790 2.32116 552 8.8601 1.7379 956 2.296 39 1116 10.9255 1.7356 1902 2.44317 805 9.6268 1.9958 1111 2.325 40 988 10.4255 1.7256 1513 2.40918 5490 17.8413 1.1356 7878 2.283 41 1115 10.8194 1.7405 1779 2.44319 4676 16.8980 1.1396 6758 2.305 42 1115 10.8194 1.7405 1779 2.46220 2540 13.7035 1.1098 3717 2.310 43 939 9.7867 1.9594 1294 2.48721 4090 16.6741 1.0668 6085 2.335 44 746 8.9516 2.5202 932 2.30322 442 7.3868 2.5345 600 2.404 45 3189 15.1924 1.3202 4760 2.36823 2076 12.7928 1.0974 3081 2.299

table 3. correlation matrix showing inter-correlation among various parameters used in the present study

lg BB W 1χ j sz χeq iP1

lg BB 1.0000W –0.1903 1.00001χ –0.0560 0.9271 1.0000j –0.1999 –0.6628 –0.7621 1.0000sz –0.1463 0.9888 0.9548 –0.7139 1.0000χeq –0.3848 –0.3461 –0.3018 0.3163 –0.3460 1.0000iP1 0.5047 –0.0300 0.1251 –0.2627 0.0434 0.2383 1.0000

a perusal of table 3 clearly indicates that no mono-parametric correlation is feasible which is statistically significant.

782

tabl

e 4.

reg

ress

ion

para

met

ers a

nd q

ualit

y of

cor

rela

tions

Mod

. n

oPa

ram

eter

s us

edA i

whe

re i

=1,2

,..5

con

st. (

B)

sER2

RF-

ratio

Q=R

/sE

Prob

.

1χ eq iP

1

A 1 = –

5.25

04 (±

1.1

595)

A 2 = 0

.775

0 (±

0.1

449)

11.8

001

0.43

820.

5252

0.72

4719

.908

1.65

381.

5060

2 j χ eq iP

1

A 1 = 0

.306

8 (±

0.2

268)

A 2 = –

5.93

59 (±

1.2

533)

A 3 = 0

.851

4 (±

0.1

540)

12.8

878

0.43

330.

5488

0.74

0814

.188

1.70

963.

2690

3 1 χ χ eq iP

1

A 1 = –

0.07

19 (±

0.0

233)

A 2 = –

6.42

91 (±

1.1

106)

A 3 = 0

.862

5 (±

0.1

334)

15.3

875

0.39

410.

6266

0.79

1619

.578

2.00

861.

2630

4 W χ eq iP

1

A 1 = –

1.93

78 (±

4.9

404×

10–5

)A 2 =

–6.

6826

(± 1

.045

9)A 3 =

0.8

027

(± 0

.122

7)

15.4

920

0.37

040.

6701

0.81

8623

.703

2.21

001.

4870

5 sz χ eq iP

1

A 1 = –

1.38

85 (±

3.4

661)

A 2 = –

6.79

06 (±

1.0

468)

A 3 = 0

.842

9 (±

0.1

229)

15.7

862

0.36

800.

6744

0.82

1224

.169

2.23

151.

1870

6 W 1 χ χ eq iP

1

A 1 = –

3.19

40 (±

1.3

464)

A 2 = 0

.059

9 (±

0.0

597)

A 3 = –

6.62

91 (±

1.0

472)

A 4 = 0

.747

7 (±

0.1

344)

14.8

967

0.37

040.

6796

0.82

4418

.032

2.22

574.

9540

7 j sz χ eq iP

1

A 1 = –

0.41

06 (±

0.2

579×

10–5

)A 2 =

–1.

8924

(± 4

.639

1)A 3 =

–6.

4321

(± 1

.049

0)

A 4 = 0

.765

3 (±

0.1

298)

15.7

769

0.36

020.

6971

0.83

4919

.566

2.31

781.

9610

8 1 χ j sz χ eq iP

1

A 1 = 0

.091

8 (±

0.7

34)

A 2 = 0

.295

6 (±

0.2

718)

A 3 = –

3.14

22 (±

1.0

999×

10–4

)A 4 =

–6.

5709

(± 1

.046

3)A 5 =

0.7

434

(± 0

.129

9)

15.1

935

0.35

730.

7108

0.84

3116

.218

2.35

964.

505×

10–4

783

We have used regress-1 software for carrying out the regression analysis. step-wise regression analysis gave many correlation equations. some of the statistically significant models are reported in Table 4.

We have obtained mono-parametric, bi-parametric, tri-parametric, tetra-parametric and penta-parametric correlations which are reported in table 3.

the highest value for R obtained in bi-parametric correlation (R = 0.7247) sug-gests the following model: lg BB = –5.2504 (± 1.1595) χeq + 0.7750 (± 0.1449) IP1 + 11.8001 (12)

n = 39, SE = 0.4382, R = 0.7247, F-ratio = 19.908, Q = 1.6538

to improve our result, we have tried for tri-parametric models by adding one by one other molecular descriptors to the above-mentioned bi-parametric correlation.

slight improvement in R value has been observed when sz index is added to the above model. the best model found is as below: lg BB = –1.3885 (± 3.4661×10–5) Sz – 6.7906 (± 1.0468) χeq + 0.8429 (± 0.1229) IP1 + 15.7862 (13)

n = 39, SE = 0.3680, R = 0.8212, F-ratio = 24.169, Q = 2.2315

the quality of model follows the following sequence:Sz, χeq, iP1 > W, χeq, iP1 > 1χ, χeq, iP1 > J, χeq, iP1.

in table 4, we have reported some four-parametric correlations. one with j, sz, χeq, iP1 is the best (R = 0.8349) model and is shown below: lg BB = –0.4106 (± 0.2579) J – 1.8924 (± 4.6391×10–5) sz – 6.4321 (± 1.0490) χeq + 0.7653 (± 0.1298) IP1 + 15.7769 (14)

n = 39, sE = 0.3602, R = 0.8349, F-ratio = 19.566, Q = 2.3178

We have obtained one penta-parametric correlation which is statistically signifi-cant. this model contains 1χ, J, Sz, χeq, and iP1 as correlating parameters. in this penta-parametric correlations the R-value is equal to 0.8431. This model is as follows: lg BB = 0.0918 (± 0.0734)1χ – 0.2956 (± 0.2718) J – 3.1422 (±1.0999×10–4)sz – 6.5709 (± 1.0463)χeq + 0.7434 (± 0.1299) IP1 + 15.1935 (15)

n = 39, sE = 0.3573, R = 0.8431, F-ratio = 16.218, Q = 2.3596.

We have used the Pogliani’s quality factor29,30 Q (R/sE) for selection of the most appropriate model. the higher the value of Q the better will be the model. on the basis ot Q-value which comes out to be 2.3596 (equation (15)) as compared to 2.3178 (equation (14)), we conclude that the model expressed by equation (15) is the most appropriate model for prediction and modelling of lg BB of the present set of compounds.

We have also calculated observed and estimated lgBB and residual values (dif-ference between observed and estimated lgBB). a comparison of observed versus

784

estimated lg BB is shown in Fig. 1. the predictive R2 value comes out to be 0.7107. this shows that the proposed model can be used for prediction of lg BB of present set of compounds.

y = 0.7107x – 0.0665R2 = 0.7107

-1.5

-1

-0.5

0

0.5

1

1.5

-1.5 -1 -0.5 0 0.5 1 1.5

observed lg BB

estim

ated

lg B

B

fig. 1. comparison between observed and estimated lgBB using the best model

conclusions

The positive coefficient of 1χ suggests that the branching index plays a dominant role in exhibition of lgBB activity.

The coefficient of IP1 in proposed models is positive. recall that iP1 accounts for the presence of a halogen group in the compound suggesting that the lgBB activity will be favoured in the presence of halogens.

The equalised electronegativity plays a negative role as its coefficient is negative in the model.

The coefficients of Sz and J indices are found to be negative, suggesting that cyclisation is not favourable for exhibition of lgBB activity.

acknoWlEDGEMEnt

authors are thankful to Prof. istvan lukovits for providing regress-1 program. they are also thankful to Prof. P.v. khadikar for his expert guidance and fruitful sugges-tions.

rEFErEncEs 1. c. hansch, j. P. Bjorkroth, a. lEo: j. Pharm. sci., 76, 663 (1987). 2. k. van BEllE, s. sarrE, G. EBinGEr, y. MichottE: j. Pharmacol. Exp. ther., 272,

1217(1995). 3. j. kElDEr, P. D. j. GrootEnhuis, D. M. BayaDaa, l. P. c. DElBrEssinE, j. P. PloE-

MEn: Pharm. res., 16, 1514 (1999). 4. j. h. lin: Pharmacol. rev., 49, 403 (1997). 5. u. norinDEr, P. sjöBErG, t. östErBErG: j. Pharm. sci., 87, 952 (1998). 6. j. M. luco: j. chem. inf. comput. sci., 39, 396 (1999). 7. P. crivori, G. cruciani, P. a. carruPt, B. tEsta: j. Med. chem., 43, 2204 (2000).

785

8. M. FEHER, E. SOuRIAL, J. M. SCHMIDT: int. j. Pharm. sci., 201, 239 (2000). 9. r. liu, h. sun, s. s. so: j. chem. inf. comput. sci., 41, 1623 (2001).10. j. a. Platts, M. h. aBrahaM, y. h. zhao, a. hErsEy, l. ijaz, D. Butina: Eur. j. Med.

chem., 36, 719 (2001). 11. t. hou, x. xu: j. Mol. Model., 8, 337 (2002). 12. k. rosE, l. h. hall, l. B. kiEr: j. chem. inf. comput. sci., 42, 651 (2002).13. M. h. aBrahaM, h. s. chanDha, r. c. MitchEll: Drug Des. Discov., 13, 123 (1995). 14. M. h. aBrahaM, h. s. chanDha, r. c. MitchEll: j. Pharm. sci., 83, 1257 (1994). 15. F. ooMs, P. WEBEr, P. a. carruPt, B. tEsta: Biochem. Biophys. acta Mol. Basis Dis., 118,

1587 (2002). 16. c. h. MichaEl: j. comput. aided Mol. Des., 17, 415 (2003).17. h. WiEnEr: j. am. chem. soc., 69, 17 (1947).18. A. T. BALABAN: chem. Phys. lett., 89, 399 (1982).19. P. v. khaDikar, n. v. DEshPanDEy, P. P. kalE, a. DoBrynin, i. GutMan, G. DoMo-

tor: j. chem. inf. comput. sci., 35, 547 (1995).20. r. toDEschini, v. consonni: handbook of Molecular Descriptors. Wiley-vch, Weinheim

(Ger.), 2000.21. M. v. DiuDEa (Ed.): QsPr/Qsar studies by Molecular Descriptors. Babes-Bolyai university,

cluj, romania, 2000.22. v. k. aGraWal, j. sinGh, M. GuPta, P. v. khaDikar, c. t. suPuran: Eur. j. Med. chem.,

41, 360 (2006).23. v. k. aGraWal, s. Bano, P. v. khaDikar: Bioorg. Med. chem., 13, 4039 (2003)24. P. v. khaDikar, a. srivastava, v. k. aGraWal, s. shrivastava: Bioorg. Med. chem.

lett., 13, 3009 (2003).25. P. v. khaDikar, s. karMarkar, v. k. aGraWal, j. sinGh, a. shrivastava, i. lu-

kovits, M. v. DiuDEa: letters in Drug Design & Discovery, 2, 604 (2005).26. l. PaulinG: the nature of the chemical Bond. cornell univ. Press, itacha, new york, 1969. 27. n. GorDi: Phys. rev., 69, 604 (1946); P. PolitzEr, h. WEinstEin: j. chem. Phys., 71, 4218

(1979); r. t. sanDErson: science, 121, 207 (1955).28. M. RANDIC: chem. Phys. lett., 211, 478 (1993).29. l. PoGliani: chem. rev., 100, 3827 (2000).30. l. PoGliani: j. Phys. chem., 98, 1494 (1994).

Received 29 December 2007 Revised 16 January 2008

786

Oxidation Communications 31, No 4, 786–797 (2008)

* For correspondence.

dissociAtive AttAcHment of low-energy electrons (below ionisAtion or electronic excitAtion tHresHolds) in frozen AqUeoUs pHospHAte solUtions

o. s. nEDElina*, o. n. BrzhEvskaya, E. n. DEGtyarEv, a. v. zuBkov

Emanuel Institute of Biochemical Physics, Russian Academy of Sciences, 4 Kosygin Street, 117 977 Moscow, Russia E-mail: olga.nedelina@gmail.com

aBstract

We investigated the ESR spectra of the products of interaction of Н2Ро4– with

low-energy electrons emitted by irradiated fluorophores (Flu). Our selection of this particular source of electrons enabled an efficient electron-injection process upon near uV light excitation (4.0–4.5 eV). Our experiments were conducted in fro-zen aqueous dilute solutions of Н2Ро4

– following the scheme: Flu+ hν →(Flu)*→Flu + еaq, еaq+ Н2Ро4

– → H• + НРо42–, (Flu-ferrocyanide ions, acetate, tryptophan

(λ>240 nm), NADH, phenothiazine, 1,3,6,8-pyrenetetrasulphonic acid (λ>340 nm). the photoinduced Esr spectra of solvated electrons (g=1.998, ∆Hpp 0.15 Gs), of the hydrogen atom (g=2.0043, ∆Hpp ~508 Gs), and of secondary acceptors were the basic indicators provided evidence for the electron attachment to Н2Ро4

– and for the subsequent interaction of an electron-adduct [Н2Ро4

–] with secondary acceptors (specifically with vanadate). In our experiment we observed: (i) the reverse relation between the Esr intensity of the hydrogen atom and the free electron with acetate in SDS-micelle as fluorophore; (ii) an interdependent ESR signal relation of the hydrogen atom and the donor-acceptor system, and (iii) disappearance of the hydrogen-atom spectra after addition of the electron scavenger kno3. We used the Esr method to visualise the discharge channel of photoejected electron (or some of the form of its relaxation to eaq) to the dissociative attachment е– + Н2Ро4

– → H• + НРо42–. our

experimental results suggest that the possible range of the functional activity of phosphates can be extended to direct involvement in redox-reactions.

Keywords: low-energy electrons, fluorophores, frozen aqueous phosphate solution, Esr.

aiMs anD BackGrounD

Electron transfer is an elementary and very important step in many chemical and biochemical reactions, where certain molecules act as electron-donors and others

787

as electron-acceptors. Molecular anions formed in the electron–target interaction are considered to be the driving force for the respective enzymatic reaction. the chemical behaviour of enzymes can largely be controlled by electron transfer from the substrate. the cation and anion produced in this way are better oxidants and re-ductants, respectively, than either of the neutral ground state molecule. Determining redox intermediates is necessary to establish the mechanism of elementary chemical events of enzymatic reactions.

in recent years, increasing importance has been associated with electron-induced chemical processes in biological environments. identifying the underlying mechanisms involves several research methodologies, including studies of the electron interaction with the bio-molecules such as Dna, its component subunits, and amino acids1,2. us-ing ionising radiation as a method for electron extraction and electron attachment is very powerful for preparing and studying the electron-gain and electron-loss species; the utility and simplicity of the method is stressed in ref. 3. Electrons are removed or added without the need of oxidising or reducing agents whose presence often leads to complications.

in studying reaction mechanisms, model photochemical systems with photon energies up to 4.5 eV (λ>250 nm) are preferable to radiolytic systems, because they are conducive to direct stabilisation of the functional reactive intermediates, while avoiding concurrent generation of highly active products of water radiolysis, which initiate the formation of nonspecific half-products. This region is not absorbed by water, and some solutes of photochemical reactions are only possible in the presence of some recyclable absorbing sensitisers or fluorophores. Photochemical methods (namely fluorophores photoionisation associated with monochromatised electron release) are advantageous for studying redox process intermediates and, in particular, are used frequently to determine potential biological electron-donors or -acceptors. Many of these products may be intermediates of dark chemical reactions4.

Photoinduced electron detachment from fluorophores is a rapid, efficient charge-transfer reaction. in this reaction, charge separation between a photoexcited sensitiser and an electron-carrier is one of the most important steps in production of long-lived photoinduced charge separation for energy conversion and storage. to harvest the light energy put in the system, the oxidising and reducing power of the photoinduced species must be used before the electrons are transferred back. in the presence of ap-propriate acceptors, the net yield of charge separation grows, and the back electron transfer slows down significantly, compared with the values of the same parameters in their absence5.

it is advantageous to detect the products of electron impact in frozen solutions, in which the products are matrix isolated and chemical transformations of primary photoproducts are stopped. the advantage of using low-temperature rigid matrices is that highly reactive species are rendered impotent by immobility, while the lifespan of the unimolecularly unstable species gets extended by the low temperatures and sometimes by the inhibitory effects of the matrix on their tendency to fragment. the

788

results are very similar in aqueous and frozen solutions; this strongly suggests that both phases generate the same radiation-produced intermediates, which, in turn, react in similar ways6.

In competitive reactions, electron capture can be remarkably specific in the solid state, suggesting that, at least in certain media, electrons migrate over large distances prior to being captured. the reactions are strongly favoured by the environment because of the charge, which reflects the role of the solvent. Thus, if protic solvents are used, solvation is rapid and strong for anions, including electrons. as a result, the ejected electrons (e–

D) are very rapidly solvated (e-s) and thereby stabilised. in contrast, elec-trons in rigid media may not be solvated at all, even in such media as glassy alcohols, or may be solvated very slowly, and are far more likely than in fluid solutions to be captured by reactive solutes prior to being deeply trapped. Photoejected electrons e–

D are likely to be more reactive than e–

s , the margin being close to the solvation energy. this is in a good agreement with the functional shape of the observed electron decays and with the dependence of the decay lifetimes on the scavenger concentrations and on the initial electron yield7.

Photochemical methods combined with low-temperature Esr spectroscopy make it possible to distinguish between: (1) direct electron capture by molecule aBc to give thermodynamically stable anion [aBc]•– , and (2) dissociative electron capture to give a– + Bc• (ref. 3). the same is true for resonant capture of free electrons at subexcitation energies by molecule M e + M→(M–(*)), which forms either unstable species (M–(*)) or a transient molecular anion, referred to as ‘temporary negative ion’ (TNI) (Refs 8 and 9). TNI typically involves multi-electron resonance, where extra-electrons are bound temporarily to electronically-excited molecules. a tni can decay either via electron autodetachment or via dissociative electron attachment (DEa) M–(*)→M•r+x–. it has been shown that, at energies well below the ionisation potential of M, DEA is the only mechanism that efficiently controls molecular fragmentation. Furthermore, recent studies on DEa in low-energy electron attachment to gas phase molecules in the biology context have shown that hydrogen loss is the predominant reaction channel10.

In an ESR study of biologically significant frozen aqueous matrices modelling the medium for uV-light-induced ATP synthesis (ADP + Pi), we were the first to demonstrate11 the presence of atomic hydrogen in the multicomponent mixture of free radicals. in such a system, it is possible to observe both stable anions and products of disintegration by the DEa mechanism. The doublet with the 508 Gs separation has been assigned to trapped hydrogen atoms produced from the reaction of the mono-phosphate ions present in the matrix with electrons at subexcitation energies (below 4.5 eV) resulting from photoionisation of the adenine base with λ >260 nm.

The assumption of electron attachment to phosphate was confirmed by our experi-ments with nanosecond laser photolysis, in which we showed that hydrated electrons are quenched by phosphate12. Pulse photoexcitation of an aqueous solution of naDh (nicotineamide adenine dinucleotide reduced) (0.2 mM) or pyrene with 337-nm light

789

of n2-laser produced two intermediate products in time less than the resolution time of the recording system (10 ns). these products were: (1) hydrated electrons, with a typical absorption spectrum with the maximum at 715 nm and lifespan about 120 ns (quenched by o2 and naDh), and (2) cation-radical naDh+, with absorption maximum at approximately 550 nm. introducing nah2Po4 into the solution did not influence the kinetics of degradation of the NADH+ radical, but decreased the lifespan of eaq and its release. the rate constant of the dynamic quenching was about 1×107

l/mol s.the process of electron interaction with phosphate does not produce stable

phosphate anion. in fact, in our experiments we did not obtain Esr spectra with ex-tremely large P31 hyperfine interaction, which would correspond to stable phosphate anion. indeed, it is known that attempts to add electrons to monophosphate anions or their salts in various solvents have failed even in radiolysis, as determined by Esr spectroscopy13.

hydrogen atoms found in these systems seem to indicate another important chan-nel of low-energy photoejected electron consumption, namely electron attachment to phosphate еaq + h2Po4

– → [H2Po4–] →H• + hPo4

2–, which leads to the formation of an important intermediate of photochemical conversions.

Thus, our photochemical experiment combined with ESR spectroscopy confirmed the early results reported in refs 14 and 15, which indicate that photochemically and ra-diolytically solvated electrons can be converted into hydrogen atoms via interaction with protons or the Brønsted acids: eaq+ HX→H + X–, where hx is any proton-containing acid. it is reported in ref. 16 that this role may be implemented by proton-donors (oxyac-ids). lately, electrochemical experiments have demonstrated direct cathode release of hydrogen from phosphate and from some acids, rather than from water17,18.

this postulate of electron attachment is based primarily on the suppression of the yields of trapped hydrogen atoms by added electron scavengers. the yields of hydrogen atoms are decreased by added electron scavengers such as nitrate ion in ices containing mononegative ions; the addition was not effective in polynegative oxyanion solutes. the relative rates of decrease by nitrate and scavengers forming stable anion indicate that electron is a mobile precursor of the trapped hydrogen atoms16.

in case of pulse radiolysis and photolysis of aqueous solutions under exposure to light with λ< 200 nm, when the medium is filled with extra-electrons of various origins, there are several channels through which the hydrogen atom is formed and utilised, including the products both of water radiolysis and of photodissociation of dissolved molecules. hence, the proportion of h atoms formed as a result of electron attachment to monophosphate ions is small (0.36% of its concentration)15.

in view of this, it was reasonable to restrict the scope of our experiment, limit-ing it to a collision study of only two components: extra low-energy electron and phosphate. in this study we attempted to directly detect the products of interaction between orthophosphate and monochromatised low-energy electrons in a medium containing predominantly these components.

790

In this work, we investigated the interaction of Н2Ро4– with low-energy pho-

toejected electrons, whose sources were photoexcited fluorophores Flu*; this particular choice allowed for producing an efficient process of monochromatised electron injec-tion upon near-uV irradiation of 340 or 260 nm with quantum energy 4.0–4.5 еV. The investigation was conducted in frozen aqueous dilute solutions following the scheme Flu+ hν→ Flu*→ Flu+ + е–

,, е–+ h2Po4–→ [H2Po4

–] → H• + hPo42– (Flu-ferrocyanide

ions, acetate, tryptophan (λ>240 nm), phenothiazine, 1,3,6,8-pyrenetetrasulphonic acid (λ>340 nm)). Modern experimental approaches, such as ESR, fluorescence, and nanosecond laser photolysis, permit both detection of all intermediates in these main processes and selective determination of specific properties of the high-energy products generated in these processes (such as excited states, free radicals including atomic hydrogen, or solvated electrons).

ExPEriMEntal

A DRSh-1000 high-pressure mercury lamp, equipped with light filters uFS-1 (240 nm <λ< 300 nm) and BS-6 (λ> 320 nm), was used as an illumination source. The excitation light was passed through an uFS-1 light filter (240 nm <λ< 300 nm) for tryptophan and acetate, and through a BS-6 filter (λ> 320 nm) for NADH and 1,3,6,8-pyrenetetrasulphonic acid. The irradiation time was 8 min. The quantum yields of these fluorophores are different, the lowest (10%) being characteristic of NADH. In this study, the following reagents were used: nah2Po4 (extra pure), tryptophan and sodium acetate of reagent grade, and naDh from acros organics. the EPr absorp-tion of frozen samples irradiated at 77 k was measured using a Bruker EMX-8 EPR spectrometer (frequency 9.6 Ghz) at 77 k.

In continuous photolysis of frozen solutions of fluorophores in the presence of phosphate, Esr spectra, including lines from counter radical and hydrogen atoms (doublets with splitting of ~508 Gs), were found in broad scanning (650 Gs) of EPR spectra (77 k) in all systems containing different photosensitisers (ferrocyanide ions, acetate, tryptophan – λ>240 nm, NADH, phenothiazine, 1,3,6,8-pyrenetetrasulphonic acid λ>340 nm).

rEsults anD Discussion

Among photochemical events in fluorophores, photoionisation is paramount in produc-ing both electrons and fluorophores cation Flu*→ Flu.+ + e•. Electrons in the excited state can either revert to the ground state or may be stabilised either by physical trap-ping or by electron capture with the electron-acceptors present in the matrix. in the absence of these events, electrons may return to their cations. these experiments did not detect directly the Esr line of photo-ejected electron. instead, we used the effect of phosphate reactants to demonstrate its presence as recorded in the EPr spectra of the fluorophore cation radical and of free radicals of acceptors Ac ascribed to stable

791

radical ac•– formed in the fast bimolecular reaction e– +ac > ac•– or products of DEa е– + h2Po4

– → [H2Po4–] → H• + hPo4

2–.Figure 1 shows photoelectron capture by h2Po4

– or D2Po4–, as monitored by Esr

spectra of hydrogen or deuterium atoms in the tryptophan (5×10–4 M solutions in the presence of 0.5 M nah2Po4 (pH 4.9) irradiated with λ>240 nm at 77 k.

3000 3100 3200 3300 3400 3500 3600 3700 3800-10000

0

10000

20000

30000

40000

50000

60000D

H

edc

b

a

magnetic field (Gs)

inte

nsity

(a.u

.)

fig. 1. x-band Esr spectra of 0.5 M solutions of nah2Po4 (ph= 4.5) in h2o or D2O after 8 min uV radiation at 77 k: a – in the presence of Trp (5×10–4 M) in h2o; b – the same in D2o; c, d – in H2o; e – in D2o. Microwave power 2 µW, modulation amplitude 3 Gs. conditions of uv irradiation (1 kW mercury lamp) as follows: a, b, c – λ>240 nm; d, e – without filter)

We established that the signal from atomic hydrogen was recorded only in the presence both of fluorophore and monophosphate anion H2Po4

– in a weak acid medium, the signal intensity being dependent on the phosphate concentration. our investigation of the ph dependence of hydrogen atom showed that the intensity of the signal from hydrogen was maximum in the range of maximum concentration of the monoanion (Figs 2 and 3).

792

1 2 3 4 5 6 7 8 9

0.0

0.2

0.4

0.6

0.8

1.0

I Hpp /

I Hpp(

max

)

pH

fig. 2. Dependence of the normalised intensities of the H-atom ESR spectrum high-field component on ph for various photosensitisers after uv irradiation at 77 k in 0.5 M solution of nah2Po4 in water: ●, ○ – Ade (5×10–4 M); ◊ – Trp (5×10–4 M); ∆ – NADH (5×10–4 M); □ – k4[Fe(cn)6] (5×10–3 M). solid and dotted lines show the mole fraction f of h2Po4

– and hPo42– ions in total phosphate concentration

(ƒ(h2Po4–)+ƒ(hPo4

2–) = 1). Microwave power 2 µW, modulation amplitude 1 Gs. conditions of uv irradiation (1 kW mercury lamp) as follows: Ade, Trp – λ>240 nm; NADH – λ>320 nm; k4(cn)6 – without filter)

3000 3100 3200 3300 3400 3500 3600 3700 3800

0.0 0.2 0.4 0.6 0.8 1.0

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

B

A

magnetic field (Gs)

[P

pH

i] (mol dm–3)

fig. 3. x-band Esr spectra of 0.3 M phosphate buffer solutions in the presence of trp (5×10–4 M) after 8 min uV radiation at 77 k for different pH: A – pH = 4.4; B – pH = 9.2. The inset shows the depend-ence of the double integral of trp radical and h-atom Esr spectrum components on the phosphate buffer concentration: ■, □ – pH = 4.4; ●, ○ – pH = 9.2, accordingly. The conditions of irradiation and recording are the same as in Fig. 1

We observed increases in the total yield of hydrogen and cation radical with in-creasing concentration of the phosphate acceptor. one can interpret the low hydrogen yield and the low cation radical yield in alkaline media, as opposed to the respective yields in weak acid media, as evidence of absence of another competing process,

793

caused by the presence of an electron-acceptor, namely of monoanion phosphate. the absence of monophosphate as electron scavenger is evidently responsible for the rapid decrease in yields of both hydrogen and cation radical spectra (Fig. 3).

note that, under these conditions, the structure of the signal from atomic hydrogen was similar in all the systems that we studied. this also indicates that hydrogen atoms had the same fixation locus, and that the environment of this locus was homogene-ous. By using partially deuterated naD2Po4, with deuterium connected to all oxygen atoms, it is possible to show that all these resonances originate from abstraction of hydrogen from oxygen sites but not from water (Fig. 1).

the acceptor properties of phosphate with respect to a photo-ejected electron were additionally corroborated by the results obtained by the method of competing electron-acceptors. these results demonstrated competitive relations between well-known electron scavenger kno3 forming stable anion and h2Po4

–, revealed on signal intensity of atomic hydrogen.

the ratio k(e– + no3–)/k(e– + Po4

–) = 340±20 obtained in our experiments agrees well with the value reported by kevan16 for γ-irradiated phosphates (Fig. 4) and characterises different degrees of affinity of kNO3 and phosphate for the electron.

3000 3100 3200 3300 3400 3500 3600 3700 3800-500000

-400000

-300000

-200000

-100000

0

100000

200000

300000

400000

C

B

A

magnetic field (Gs)

inte

nsity

(a.u

.)

0.00 0.01 0.02 0.03 0.04 0.05 0.06

-2

0

2

4

6

8

10

12

14

16

18

20

22

24

(I H 0-I H)/

I H

[NO3-]/[H2PO4

-]

fig. 4. Esr spectra of 0.5 M solutions of nah2Po4 (ph= 4.4) in h2o, containing 5×10–3 M of k4[Fe(cn)6] and different amounts of kno3 after 8 min uV radiation at 77 k: A – 0 M, B – 0.005 M, C – 0.03 M kno3, accordingly. the insert shows kinetic plot for competitions of electrons between h2Po4

– and no3

–. From this slope the ratio k(e– + no3–)/k(e– + Po4

–) = 340±20 was determined

the dependence of the pattern of the Esr spectrum of eaq on the presence of orthophosphate or the electron scavenger kno3 was also studied in a medium con-taining the anionic detergent sodium dodecyl sulphate (sDs). the reason is that a more reliable line corresponding to eaq is detected in a micellar structure rather than in an aqueous solution. Since photoionisation in micelles is a sufficiently effective way of generation of eaq, this is a good method for a spectroscopic study of chemical

794

reactions, in an aqueous solution, of organic compounds triggered by electron trap-ping. Photo-ejected electrons are released into the ambient aqueous medium, and their return into anionic micelles is hampered by the repulsing electrostatic potential. thus, in microheterogeneous micellar structures, photoproducts are separated via hydrophilic–hydrophobic interactions19. Photo-ejected electrons may be scavenged by dissolved acceptors located on the periphery of micelles. in this case, the line cor-responding to eaq disappears, and the spectrum of the products of interaction of the electron with the acceptor appears instead.

Figure 5 shows the EPr spectra obtained as a result of photolysis of frozen samples both in the absence and in presence of phosphate. in our experiments, we recorded the eaq spectrum (g = 1.9987, Hpp= 0.15 Gs) in a phosphate-free medium. the g-factor value obtained in our experiments was somewhat smaller than the g-factor value characteristic of a free electron (2.0023); this can be explained by the fact that, similarly to F-centers, electrons in frozen systems are trapped in the field of molecules and ions. Depending on the structure of the system, traps have different characteristics. using EPr spectroscopy it was shown that the g-factor of the electron in the F-center may be 1.995 (ref. 20). this fact indicates that bound electrons in ions contribute markedly to the magnetic moment of the electron contained in the F-center. therefore, a considerable part of the lifespan of the electron in the F-center passes in the vicinity of ions surrounding the site at which it was trapped.

3000 3100 3200 3300 3400 3500 3600 3700 3800

-10000

-5000

0

5000

10000

B

A

magnetic field (Gs)

inte

nsity

(a.u

.)

fig. 5. Esr spectra of sodium acetate (5×10–2 M) and sDs (2×10–2 M) solutions (ph= 4.9) in water after 8 min uV radiation at 77 k in the absence (A) and presence (B) of 0.5 M NaH2Po4

the signal of eaq (g = 1.9987, Hpp= 0.15 Gs) was quenched when both phosphate and kno3 were added. the disappearance of the line corresponding to eaq was ac-companied by the appearance of an EPr spectrum of the product of interaction of

795

eaq with the aforementioned acceptor – either the hydrogen signal or a characteristic triplet, respectively.

in a micellar solution we observed the same dependencies as in our previous study12, when investigating the intensity of signal from atomic hydrogen in depend-ence on ph, orthophosphate concentration, and the effect of deuteration led us to assume that the signal from the hydrogen atom characterises the acceptor interaction of h2Po4

– with a photo-ejected electron. in this study, in contrast to radiolytic conditions (mobile electron energy up to

20 eV), hydrogen atoms were stabilised under relatively mild conditions (~ 4–4.5 ev) when there is no energy transfer from the excited states of a sensitiser to the phosphate molecule. in this case, direct photo-dissociation of phosphate h2Po4

–→h + hPo4

2– is excluded (we never saw the Esr signal of hPo4 .– , Fig. 1), and their

precursors seem to be a mobile electron trapped by phosphate е– + h2Po4

– → [H2Po4–].

→H• + hPo42– in a dissociative electron attachment DEa.

note that redox processes involving phosphate may be initiated not only by light or radiation, but also by electron transmission in dark processes. For example, dark one-electron reduction of vanadate by ascorbate and naDh and related free-radical generation in a phosphate buffer were investigated by Esr and Esr spin trapping21,22. the vanadate reduction was stimulated by phosphate, the vanadium(iv) yield increas-ing with increasing phosphate concentration. addition of formate to the incubation mixture generated the carboxylate radical (•coo–), which indicated the formation of reactive species in the vanadium reduction mechanism. We posit that a phosphate electron adduct revealed by hydrogen loss may be mediating this process. in this system of vanadate reduction by photoejected electrons we observed a decrease in appearance of atomic hydrogen and vanadium in the presence of phosphate (Fig. 6).

2600 2800 3000 3200 3400 3600 3800 4000 4200 4400 4600

-20000

0

20000

40000

60000

80000

100000

120000

140000

x 0.2D

C

B

A

inte

nsity

(a.u

.)

magnetic field (Gs)

fig. 6. x-band Esr spectra of 5×10–4 M solutions of Trp (pH= 4.4) in water, containing: A – 0.5 M nah2Po4 ; B – 5×10–4 M vo3

– +0.5 M nah2Po4; C – 5×10–4 M vo3– after 8 min uV radiation at 77 k;

6D shows Esr spectrum of the paramagnetic 5×10–4 M solution of vanadyl vo2+ in 0.5 M solution of nah2Po4 in water at 77 k

796

the interrelations of photo-induced Esr spectra of photo-ejected electron, the hydrogen atom, and the fluorophore cation radical were suggested to be the basic indicators of the process of electron attachment to Н2Ро4

– and, in some cases, of subsequent interaction of an electron-adduct [Н2Ро4

–]• with secondary electron-ac-ceptors (vanadate).

This confims the role of phosphate monoanion and its electron adduct as ac-ceptor-donor intermediates in models of photochemical and dark-electron transport. the Esr method is assumed to visualise the discharge channel of the photoejected electron (or some form of its relaxation to eaq) to the dissociative electron attachment е– + h2Po4

– → [H2Po4–]• → H. + hPo4

2–. in the study by atkins23, it was assumed that electrons apparently attack phosphate at the electrophilic hydrogen atom, which then leads to its cleavage, the acceptor level being the localised σ* O–H bond. Acceptor interaction of phosphate with electrons in the physiological ph range is of interest for studying the mechanism of many biochemical reactions involving orthophosphate, including the synthesis of atP, because the highly active intermediates obtained in the interaction are included in the reaction. according to these studies, the possible range of the functional activity of phosphates can be extended to direct involvement in redox reactions.

this process may be important for studying the radiolytic and photolytic chemistry of biological systems, because solvated electrons are the main reagent in these proc-esses. in the presence of acceptors converting them into hydrogen atoms, reactions mediated by the latter may be decisive for the observed results. this aspect may also be of interest in studying the effect of low-energy electrons on the Dna damage that results in free-radical dissociation of the C–O– P bond. Our results coincide with the outcomes of the computational work24, which assumes that the most direct mechanism of single strand breaks occurring in Dna at subexcitation energies (<4 ev) is due to resonance electron capture DEa directly to the phosphate group.

rEFErEncEs 1. j. BErDys, i. anusiEWicz, P. skurski, j. siMons: j. am. chem. soc., 128, 6445 (2004). 2. y. v. vasil’Ev, B. j. FiGarD, v. G. voinov, D. F. BaroFsky, M. l. DEinzEr: j. am. chem.

soc., 128, 5506 (2004). 3. M. c. r. syMons: Pure and appl. chem., 53, 223 (1981). 4. l. i. GrossWEinEr, j. F. BauGhEr: j. Phys. chem., 81, 93(1983). 5. j. r. Bolton: solar Energy, 57, 37 (1996). 6. l. j. kEvan: Phys. chem., 70, 2529 (1966). 7. s. P. Mishra, M. c. r. syMons: Faraday Discuss. chem. soc., 63, 175 (1977). 8. o. inGolFsson, F. WEik, E. illEnBErGEr: intern. j. of Mass spectrometry and ion Processes,

155, 1(1996). 9. y. zhEnG, P. cloutiEr, D. j. huntinG, l. sanchE, j. r. WaGnEr: j. am. chem. soc.,

127, 16592 (2003).10. B. liu, P. hvElPlunD, s. B. niElsEn, s. toMita: j. chem. Phys., 121, 4175 (2004).

797

11. s. n. DoBryakov, o. n. BrzhEvskaya, i. s. solov’Ev, E. M. shEkshEEv, o. s. nEDElina: Dokl. an sssr, 384, 119 (2002). in: Dokl. Biochem. Biophys. (Engl. transl.), 384, 136 (2002).

12. o. n. BrzhEvskaya, E. n. DEGtyarEv, P. P. lEvin, t. a. lozinova, o. s. nEDElina: Dokl. Biochem. Biophys. (Engl. transl.), 405, 395 (2005). in: Dokl. an sssr, 405, 259 (2005).

13 s. P. Mishra, D. j. nElson, M. c. r. syMons: int. j. radiat. Phys. chem., 7, 581 (1975).14. j. jortnEr, M. ottolEnGhi, j. raBani, G. stEin: j. chem. Phys., 37, 2488 (1962).15. l. kEvan, P. n. Moorthy, j. j. WEiss: j. am. chem. soc., 86, 771 (1964).16. l. kEvan, c. FinE: j. am. chem. soc., 88, 869 (1966).17. v. Marinovich, a. DEsPic: Electrochimiya, 33, 1044 (1997). in: russ. j. Electrochem. (Engl.

trans.), 33, 965 (1997).18. v. Marinovich, a. DEsPic: Electrochimica acta, 44, 4073 (1999).19. M. GratzEl, j. k. thoMas: j. Phys. chem., 78, 2248 (1983).20. E. j. hart, M. anBar: the hydrated Electron. Wiley-interscience, new york, 1970.21. M. DinG, P. M. GanEtt, y. rojanasakul, k. lui, x. shi: j. inorg. Biochem., 55, 101

(1994).22. s. yoshino, G. sullivan, a. stErn: arch. Biochem. Biophys., 272, 76 (1989).23. P. W. atkins, n. kEEn, M. c. r. syMons, h. W. WarDalE: j. chem. soc., 5594 (1963).24. c. köniG, j. koPyra, i. BalD, E. illEnBErGEr: Phys. rev. lett., 97, (2006).

Received 18 October 2007 Revised 20 November 2007

798

Oxidation Communications 31, No 4, 798–803 (2008)

* For correspondence.

HydroxycinnAmides of some Amino Acids And tHeir AntioxidAnt Activity

i. stankova*, k. chuchkov

Department of Chemistry, South-West University ‘Neofit Rilski’, 2700 Blagoevgrad, Bulgaria E-mail: ivastankova@abv.bg

aBstract

several hydroxycinnamides were prepared from the sinapic, p-coumaric and ferulic acids and thiazole containing trifluoroacetic acid (TFA).valine-4-carboxylic acid ethyl ester, and their antioxidant activities that were evaluated using 1,1-diphenyl-2-picryl-hydrazyl (DPPh) test.

hydroxycinnamides (sinapoyl-2-valyl-thiazole-4-carboxylic acid ethyl ester (5b), p-coumaroyl-2-valyl-thiazole-4-carboxylic acid ethyl ester (5c) obtained from amino acid (thiazole containing valine ethyl ester) exhibited insignificant antioxidative activity, and only feruloyl-2-valyl-thiazole-4-carboxylic acid ethyl ester (5a) showed a borderline activity.

Keywords: hydroxycinnamides, amino acid, thiazole, antioxidant activity, DPPh test.

aiMs anD BackGrounD

the biochemical properties of polyphenolic secondary plant metabolites such as esters of cinnamic acids (caffeic, ferulic, p- and o-coumaric, sinapic) attract much attention in biology and medicine. these compounds show antioxidant, antiviral, antibacte-rial, vasoactive and other properties1–7. cinnamic acid conjugates are also commonly isolated from plant sources as the corresponding n-substituted amides. While esters of cinnamic acids occur widely in higher plants, amides of cinnamic acids seem to be rare. the growing interest during the last years to natural and synthetic amides of phenyl propenoic acids is due to the higher metabolical stability of the amide group in comparison to the ester group.

in the course of exploring new hydroxycinnamides as an antioxidative agent, we synthesised hydroxycinnamic acid amides with thiazole containing tFa.valine-4-carboxylic acid ethyl ester. We have considered the role of heterocyclic amino

799

acid and the effect of these structure modification on antioxidant activity. Herein, we describe the preparation of hydroxycinnamides from the corresponding acids (sinapic, p-coumaric and ferulic acids) and tFa.valyl-thiazole-4-carboxylic acid ethyl ester. We also have evaluated the effectiveness of these compounds in DPPh radical-scavenging activity.

recently it has been shown that 3,4-dihydroxycinnamic acid (l-alanine methyl ester) amide and 4-hydroxycinnamic acid (l-alanine methyl ester) amide possess useful biological activity as an anti-atherosclerotic agent with inhibition of cellular cholesterol storage3,5. it is presumed that induction of the accumulation of hydroxycinnamic acid amides is a part of the defense system of the plants. their production is activated in response to various environmental stimuli (wounding, fungal infection, heavy metal ions, etc.)8–10. these compounds are postulated to contribute to the formation of a phenolic barrier, which makes cell walls more resistant to enzymatic hydrolysis11.

lee et al. studied a variety of hydroxycinnamic acid amides12. they prepared several hydroxylated cinnamic acid derivatives from the corresponding acids and amino acid residues, and their hypocholesterolemic activities evaluated in high cho-lesterol-fed mice. 3,4-Dihydroxycinnamides obtained from amino acid derivatives containing a hydrophobic side chain such as alanine, valine, phenylalanine, and isoleucine exhibited potent cholesterol-lowering activities.

Fifteen amides of cinnamic, ferulic and sinapic acids with natural and unnatural c-protected amino acids have been synthesised and their antioxidant activity in bulk phase lipid autoxidation has been studied13–15. the highest antioxidant activity has been found for the compounds n-feruloyl-l-phenylalanine t-butyl ester and n-sinapoyl-phenylalanine t-butyl ester, containing the same phenylalanine moiety.

the variety of natural products containing thiazole, oxazole and imidazole rings have encouraged numerous synthetic efforts16. During the last decade a particularly broad spectrum of 5-ring heterocycles containing natural products, has been isolated from marine organisms. in particular, thiazole, oxazole and imidazole amino acids that may play a key role in biological activities of unusual peptides are also important intermediates for natural product synthesis.

We decided to synthesise cinnamoyl-, feruloyl and sinapoyl-thiazole-amino acid amides in order to define their antioxidant activity.

ExPEriMEntal

the cinnamoyl-, feruloyl- and sinapoyl-amides were synthesised according to the procedure presented in Fig. 1. amide 1 was obtained following the Pozdnev’s method17 from Boc –Val-OH. The amide was converted into thioamide 2 by the lawesson’s reagent (2,4-bis(4-methoxyphenyl)-1,3-dithia-2,4-diphosphetane-2,4-disulphide)18. cyclocondensation of thioamide with 3-bromo-2-ethyl-propionic acid19 leads to va-lyl-thiazole-4-carboxylic acid ethyl ester 3 in 82% yield. Deprotection was affected with trifluoroacetic acid/CH2ch2.

800

(i) (Boc)2o, nh3hco3; (ii) lawesson's reagent 2,4-bis(4-methoxyphenyl)-1,3-dithia-2,4-diphosphetane-2,4-disul-phide; (iii) 3-bromo-2-ethyl-propionic acid; (iv) tFa/ch2ch2; (v) hydroxycinnamoyl rest/EDc/DMaP/Et3n

fig. 1. synthesis of hydroxycinnamic acid amides with tFa.2-valyl-thiazole-4-carboxylic acid ethyl ester

a solution of sinapic, p-coumaric and ferulic acids in dimethylformamide (DMF) was treated with triethylamine and tFa.2-valyl-thiazole-4-carboxylic acid ethyl ester using the coupling agent 1-[3-(dimethylamino)propyl]-3-ethyl carbodiimide hydro-chloride (EDc) and 4-(dimethylamino)-pyridine (DMaP) as a catalyst, to produce the amide derivates (5 a–c).

rEsultsDPPh raDical scavEnGinG activity

the radical scavenging activity of the new compounds was evaluated by the method of Pekkarien et al.20 For each compound and concentration tested (0.9, 1.8 and 3.6 mM), the reduction of DPPh radical was followed by monitoring the decrease in the absorbance at 516 nm (table 1). the absorption was monitored at the start and at 10 and 20 min. The results are expressed as % RSA = [(Abs516 nm (t=0) – Abs516nm (t=t’)) × 100/abs516 nm (t=0)], as proposed by Pekkarien et al.

801

table 1. antioxidative activity of hydroxycinnamides by DPPh•

compounds

% RSA (radical scavenging activity)0.9 mM 1.8 mM 3.6 mM

reaction time (min)10´ 20´ 10´ 20´ 10´ 20´

tFa.val-thz-oEt 1.3 1.6 1.4 1.9 2.1 2.4sinapic acid (sa) 16.1 17.2 26.5 31.9 69.0 69.6sa-val-thz-oEt (5a) 6.5 7.4 9.9 11.1 17.9 19.4Ferulic acid (Fa) 12.0 13.8 21.0 25.1 36.7 44.3Fa-val-thz-oEt (5b) 10.7 12.9 16.6 20.1 28.9 33.2p-cumaric acid (ca) 2.1 2.9 3.7 4.7 4.5 6.1ca-val-thz-oEt (5c) 1.2 1.6 1.2 1.6 3.4 3.9

the tested hydroxycinnamoyl amides demonstrated higher radical scavenging activity than tFa.val-thz-oEt. as seen in table 1 the amides of sinapic and p-cou-maric acids showed lower antioxidative effect than the free hydroxicinnamic acids against DPPh test. the amide of ferulic acids exhibited a borderline activity.

conclusions

three novel hydroxycinnamic acid derivatives have been synthesised. Firstly, thiazole containing amino acid derived from Boc-valine-4-carboxylic acid was synthesised.

Secondary, the cinnamoyl-, feruloyl and sinapoyl–thiazole-amino acid amides synthesised by the method in peptide chemistry (EDc) were pure (after the reaction the residue was purified on kieselger 60 F254 (Merck) using the solvent system hex-ane:Etoac = 4:5).

thirdly, the hydroxycinnamoyl thiazole conjugates have been studied for their antioxidant activity by the DPPh test. amide of ferulic acids exhibited a borderline activity. the rest of compounds have a lower antioxidative effect than the free hy-droxycinnamic acids.

These results demonstrate that hydroxycinamic acid modified with thiazole, containing amino acid does not possess antioxidative effect in comparison with those modified with natural amino acids. Further systematic research is needed for investigation the role of other amino acids containing oxazole, imidazole and thia-zolyl-thiazole rings.

aBBrEviationsBOC – tert-butyloxycarbonylDMAP – 4-(dimethylamino)pyridineDMF – dimethylformamideDPPH – 1,1-diphenyl-2-picryl-hydrazyl radicalOEt – ethyl ester

802

RSA – radical scavenging activityTFA – trifluoroacetic acidTHZ – thiazoleVAL – valine.

rEFErEncEs 1. F. suDina, k. MirzoEva, a. PushkarEva, a. korshunova, v. suButya, D. var-

FoloMEEv: caffeic acid Phenethyl Ester as a lipoxygenase inhibitor with antioxidant Properties. FEBs lett., 329 (1–2), 21 (1993).

2. M. naMiki: antioxidant antimutagens in Food. crit. rev. Food sci. nutr., 29, 273 (1990). 3. s. lEE, han jonG-Min, h. kiM, E. kiM, t. s. jEon, W. s. lEE, k. h. cho: synthesis of

cinnamic acid Derivatives and their inhibitory Effects on lDl-oxidation, acyl-coa: cholesterol acyltransferase-1 and -2 activity, and Decrease of hDl-Particle size. Bioorg. Med. chem. lett., 14, 4677 (2004).

4. j. h. Moon, j. j. tErao: antioxidant activity of caffeic acid and Dihydrocaffeic acid in lard and human low-density lipoprotein. j. agric. Food chem., 46, 5062 (1998).

5. s. lEE, c. h. lEE, j. oh, E. kiM, y. y. choi, W. lEE, s. h. Book, t. s. ikon: anti-atherogenic Effects of 3,4-dihydroxyhydrocinnamides. Bioorg. Med. chem. lett., 13, 2681 (2003).

6. PErEz-alvarEs, v. BoBaDilla, P. MuriEl: structure-hepatoprotective activity relationship of 3,4-dihydroxycinnamic acid (caffeic acid) Derivatives. j. appl. toxicol., 21, 527 (2001).

7. r. BurkE, r. FEsEn, a. MazuMDEr, j. WanG, M. carothErs, D. GrunBErGEr, j. Driscoll, k. kohn, j. PoMMiEr: hydroxylated aromatic inhibitors of hiv-1 integrase. j. Med. chem., 38, 4171 (1995).

8. J. NEGREL, F. JAVELLE, M. PAYNOT: Wound-induced Tyramine Hydroxycinnamoyl Transferase in Potato (Solanum tuberosum) tuber Discs. j. Plant Physiol., 142, 518 (1993).

9. h. PEiPP, W. MaiEr, j. schMiDt, v. Wray, D. strack: arbuscular Mycorrhizal Fungus-induced changes in the accumulation of secondary compounds in Barley roots. Phytochemistry, 44, 581 (1997).

10. W. Fink, M. liEFlanD, k. MEnDGEn: comparison of various stress responses in oat in compatible and non host resistant interactions with rust Fungi. Physiol. Mol. Plant Pathology, 37, 309 (1990).

11. j. nEGrEl, s. loFty, F. javEllE: Modulation of the activity of two transferases in Wound healing Potato tuber Discs in response to Pectinase or abscisic acid. j. Plant Physiol., 46, 318 (1995).

12. s. lEE, c.-h. lEE, E. kiM, s.-h. junG, h. k. lEE: hydroxylated hydrocinnamides as hypo- cholesterolemic agents. Bull. korean chem. soc., 28 (10), 1787 (2007).

13. M. sPasova, v. kortEnska-kanchEva, i. totsEva, G. ivanova, l. GEorGiEv, ts. Milkova: synthesis of cinnamoyl and hydroxycinamoyl-amino acid conjugates and Evalu-ated of their antioxidant activity. j. Peptide sci., 12, 369 (2006).

14. M. sPasova, G. ivanova, l. GEorGiEv, ts. Milkova: in: Peptides (Ed. M. Flegel, M. Fridkin, G. Gilon, j. slavonova).2004, 950–951.

15. v. kortEnska-kanchEva, v. D. Bankova, v. s. PoPova: antioxidant capacity of new chalcones from Propolis of El salvador during Methyl linoleate oxidation in Micellar solutions. oxid. commun., 28 (3), 525 (2005).

16. G. viDEnov, D. kaisEr, c. kEMPtEr, G. junG: synthesis of naturally occurring conforma-tionally Resticted Oxazole and Thiazole Containing Di – and Tripeptide Mimetics. Angew. Chem.,

803

108, 1604 (1966); G. VIDENOV, D. kAISER, C. kEMPTER, G. JuNG: angew. chem. int. Ed. Engl., 35, 1503 (1996).

17. v. PozDnEv: activation of carboxylic acids by Pyrocarbonates. application of Di-tert-butyl Pyrocarbonate as condensing reagent in the synthesis of amides of Protected amino acid and Peptides. tetrahedron lett., 36, 7115 (1995).

18. S. SCHEIBYE, B. S. PEDERSEN, s. o. laWEsson: studies on organophosphorus compounds. The Dimer of p-methoxyphenylthionophophine Sulfide as a Thiation Reagent. A New Route to thiocarboxamides. Bull. soc. chim. Belg., 87 (3), 229 (1978).

18. D. BOGER, S. MIYAZAkI, S. H. kIM, J. H. Wu, O. LOISELEuR, S. L. CASTLE: Diastereoselec-tive total synthesis of the vancomycin aglycon with ordered atropisomer Equilibrations. j. am. chem. soc., 121, 3226 (1999)

20. s. PEkkarinEn, k. schWarz, M. hEinonEn, a. hoPia: antioxidant activity and Partition-ing of Phenolic Acids in Bulk and Emulsified Methyl Linoleate. J. Agric. Food Chem., 47, 3036 (1999).

Received 1 February 2008 Revised 15 March 2008

804

Oxidation Communications 31, No 4, 804–811 (2008)

* For correspondence.

AntioxidAnt systems And free proline content in different genotypes of soybeAn

Dj. MalEncic*, M. PoPovic, D. stajnEr, D. Prvulovic

Faculty of Agriculture, University of Novi Sad, 8 Trg D. Obradovica Square, 21 000 Novi Sad, Serbia E-mail: malencic@polj.ns.ac.yu

aBstract

superoxide dismutase and guaiacol peroxidase activities, lipid peroxidation intensity, free proline, protein and glutathione contents were measured in 20 soybean cultivars and hybrids. in addition, the total non-enzymatic antioxidant activity of plant extracts was evaluated by the DPPh-radical scavenging activity assay. several different but compatible antioxidant systems were established in examined soybean genotypes. on the basis of the results obtained more tolerant genotypes with higher free proline and glutathione contents, higher enzymes and DPPh-radical scavenging activities were chosen. The selected genotypes could be used in field production as well as in breeding.

Keywords: DPPh-radical scavenging activity, glutathione, guaiacol peroxidase, lipid peroxidation, proline, superoxide dismutase.

aiMs anD BackGrounD

oxidative metabolism of normal cells and different stress situations generate highly reactive oxygen species (ros). the ros, such as superoxide radical (o2

•), hydrogen peroxide (h2o2), hydroxyl radical (•oh), and singlet oxygen (1o2), have been impli-cated in a number of physiological disorders in plants1. Environmental factors such as uv-light and other forms of radiation, photooxidation, air polution, drought, herbi-cides, pathogen invasion, certain injuries, hyperoxia, ozone, temperature fluctuations and other stresses are known to induce free radical formation in most aerobic organ-isms2. ros inhibit chloroplast development, decrease seed viability and root growth, stimulate leaf abscission and desiccation, cause peroxidation of essential membrane lipids in the plasmalemma and intracellular organelles (i.e. lipid peroxidation). the main cellular components susceptible to damage by free radicals are polyunsaturated fatty acids in membranes, proteins, carbohydrates, nucleic acids and pigments such as

805

chloroplasts or carotenoids3. antioxidant defence systems have co-evolved with aero-bic metabolism to counteract oxidative damage from ros. this includes antioxidant enzymes such as superoxide dismutase (soD), catalase (cat), glutathione reductase (Gr), glutathione-s-transferase (Gst) and different kinds of plant peroxidases (Px), as well as some non-enzymatic plant antioxidant systems such as glutathione, toco-pherols, carotenoids, ascorbate, flavonoids and other phenolic compounds4–6.

Growth of legume plants is depressed under drought conditions. cellular water deficit can result in a concentrations of solutes, changes in cell volume and membrane shape, disruption of water potential gradients, loss of turgor, disruption of membrane integrity, and denaturation of protein7,8. in addition, leaves close stomata under water stress and the imbalance between photosynthetic electron transport and co2 fixation rates may result in the overreduction of the electron transport chain components and facilitate the transfer of electrons to o2 and generation of ros (ref. 9).

in response to osmotic stress, many organisms accumulate compatible osmolytes to adjust their intracellular osmotic potential and to protect subcellular structures against stress damage. one of the most widespread accumulated osmolytes is the amino acid proline10. according to some authors11–13, accumulated proline acts as a cytosolute, an osmoprotectant and a protective agent for cytosolic enzymes and vari-ous cellular structures. Many other functions are assigned to proline as well, such as a mediator of osmotic adjustment, a free radical scavenger, a redox potential buffer and as an important component of cell wall proteins.

in order to achieve an improvement of plant adaptions to different types of stresses, the aim of our study was to select genotypes of soybean which are able to accumulate a larger quantities of free proline, proteins and Gsh, and at the same time have a good resistance to toxic action of ros.

ExPEriMEntal

twenty domestic and introduced genotypes (cultivars and their F1 hybrids) of soybean [Glycine max (L.) M e r r.] were grown on experimental fields at the Institute of Field and vegetable crops at the location of rimski sancevi, near novi sad, in 2006. the trial was set in a complete randomised block design in four replications. leaves for the biochemical assays were collected in the stage of full blossoming.

Oxygen metabolism enzymes. 1 g of fresh leaves was homogenised with 5 cm3 0.1 mol/dm3 k2hPo4 at ph 7.0. after centrifugation at 15 000 rpm for 10 min at 4o c, aliquots of the supernatant were used for enzyme activity measurements. superoxide dismutase (soD; Ec 1.15.1.1) activity was assayed by measuring its ability to inhibit the photochemical reduction of nitroblue-tetrazolium (nBt) (ref. 14). For the guaiacol peroxidase (GPx; Ec 1.11.1.7) activity, aliquots of the supernatant were evaluated using guaiacol as substrate15.

806

Lipid peroxidation (lP) was measured at 37 oc as malonyldialdehyde (MDa) pro-duction, spectrophotometrically at 532 nm with the thiobarbituric acid (tBa). the total amount of tBa-positive substance is given as nmol MDa/g fresh material, as described in refs 16 and 17.

Soluble proteins were determined using Bsa as a standard, according to the method of Bradford18.

Glutathione, in reduced form (Gsh), was determined with the Ellman reagent19.

Free proline (Pro) was extracted from 0.5 g of fresh leaf tissue into 10 cm3 of 3% sulphosalicylic acid. Proline content was determined spectrophotometrically at 520 nm following the ninhydrin method20, using pure proline (Merck, Darmstadt, Ger-many) as a standard.

DPPH assay. 0.2 g of plant material were excised and homogenised at 4 oc in 2.0 cm3 of absolute ethanol with mortar and pestle as described in refs 5 and 21. a 0.5 cm3 aliquot of extract was mixed with a 0.5 mmol/dm3 DPPh ethanol solution (0.25 cm3) and 100 mmol/dm3 acetate buffer (ph 5.5; 0.5 cm3). after standing for 30 min, the absorbance of the mixture was measured at 517 nm against a blank containing 0.5 cm3 of absolute ethanol instead of aliquot. DPPh-radical scavenging activity is expressed as % of control.

the obtained results on the antioxidant enzymes activities, as well as for the pro-teins, Gsh, MDa and free proline contents and DPPh-radical scavenging activities, were done in triplicate and analysed using analysis of variance. The least significant difference (lsD) values at P = 0.05 were calculated.

rEsults anD Discussion

the results for the enzyme activities showed high activity in all examined genotypes (Table 1). The total leaf SOD activity ranged from 1028.42 to 1827.11 u/g, while GPx activities showed bigger differences among examined genotypes, ranging from 257.25 to 986.12 u/g. Plants were grown in experimental field without irrigation, under natural agroecological conditions. soD and GPx are inducible enzymes and our results showed that the examined genotypes were partially susceptibile to the action of toxic ros. increased enzyme activities are probably due to high outside temperatures and water deficit in the mid-July when the plant material was collected. as our previous study have shown, the formation of ros is favoured under such conditions and one can expect some adaptive changes in the activity of antioxidant enzymes to mitigate the situation22. according to some authors23, soD activity is clearly increased depending upon the degree of water stress indicating that the leaf has the enhanced capacity to scavenge o2

.– during stress conditions. in our study, genotypes Venera, BL-8, Morava and F1 hybrid (7×8) featured significantly higher soD activities compared to other genotypes.

807

table 1. superoxide dismutase (soD) and guaiacol peroxidase (GPx) activities and soluble proteins content in soybean genotypes

Genotypes soD (u/g) GPx (u/g) sol. proteins (mg/g) 1. ln92-7369 1288.03 343.00 14.19 2. 1581/99 1613.81 703.15 15.92 3. 1511 1574.97 420.17 16.33 4. 1499/99 1564.19 411.60 15.92 5. lori 1439.05 600.25 16.66 6. linda 1391.84 437.33 18.21 7. Balkan 1348.73 866.08 15.06 8. BL-8 1745.96 574.52 16.12 9. alisa 1314.86 634.55 13.5410. tara 1469.50 986.12 14.4411. Meli 1453.66 883.22 14.3812. sava 1289.12 565.95 13.8213. venera 1827.11 806.05 16.0014. Morava 1691.56 385.88 17.3515. F1 (1×2) 1461.96 814.63 16.0616. F1 (4×2) 1134.37 968.98 17.1617. F1 (4×3) 1028.42 831.78 14.1118. F1 (5×1) 1237.84 754.60 17.8519. F1 (6×1) 1596.81 257.25 17.8820. F1 (7×8) 1651.63 737.45 18.84lsD0.05 177.80 158.50 0.88

the highest GPx activities were recorded in the leaves of tara cultivar and F1 hybrid (4×2), 985.7 and 968.7 u/g, respectively, where the activity of SOD was lower (table 1). this suggests that in these genotypes, but also in some other with increased GPx activity (Balkan, F1 (4×3)), peroxyl radicals were accumulated. results obtained for GPx activity in F1 hybrids are in agreement with the results for lP (table 2). in these genotypes MDa content was low and in positive correlation with the GPx activ-ity, although no correlation was recorded in investigated cultivars. it has been shown by the earlier investigations24,25 that the plant species with higher antioxidant enzyme activities remove ros more effectively, thus preventing membrane damage.

the lP is one of the most widely-used indicators of oxidative stress. increased contents of MDa, one of the main end-products of lP, were observed in genotypes F1 (1×2), (7×8), BL-8, 1581/99 and Balkan, but their values were not above average and they are in agreement with our previous findings, ranging from 166.7 to 263.0 nmol/g fresh material (table 2). in spite of the enhanced activities of soD and GPx in some of these genotypes, increased formation of MDa occurred. this means that scavenging system against ros is less effective in some genotypes, and the activity of antioxidant enzymes is not enough to prevent oxidative damage, which is in agree-ment with findings of some other authors23.

808

table 2. Glutathione (Gsh), malonyldialdehyde (MDa), and free proline (Pro) contents, and DPPh-radical scavenging activity in soybean genotypes

Genotypes Gsh (µmol/g) MDa (nmol/g) Pro (µmol/g) DPPH (%) 1. ln92-7369 10.54 197.19 0.522 48.17 2. 1581/99 9.51 231.25 0.286 25.55 3. 1511 9.82 209.56 0.306 45.98 4. 1499/99 11.81 191.12 0.375 43.79 5. lori 12.36 166.82 0.468 46.71 6. linda 16.96 222.79 0.442 35.04 7. Balkan 11.58 230.60 0.266 31.39 8. BL-8 11.28 233.64 0.404 36.98 9. alisa 11.24 211.51 0.482 29.6810. tara 11.09 205.65 0.300 27.7311. Meli 10.44 187.00 0.491 29.9212. sava 14.48 201.53 0.271 22.8713. venera 11.39 204.35 2.286 34.5514. Morava 11.36 205.87 0.369 43.8015. F1 (1×2) 10.80 263.36 0.313 40.1516. F1 (4×2) 12.96 190.68 0.322 45.2517. F1 (4×3) 9.52 196.11 0.316 45.2518. F1 (5×1) 13.61 187.00 0.276 47.2019. F1 (6×1) 13.63 177.45 0.531 43.0620. F1 (7×8) 18.31 243.61 0.571 24.33lsD0.05 0.53 18.13 0.090 6.96

The tripeptide glutathione (γ-Glu-Cys-Gly) is involved in many aspects of metabolism, such as removal of hydroperoxides, protection from ionising radiation, maintenance of the sulphhydryl status of proteins, etc. Gsh contents showed great differences among examined soybean genotypes, ranging from 9.51 (cultivar 1581/99) to 18.31 µmol/g fresh material (F1 hybrid (7×8)) (Table 2). Since this hybrid also showed elevated enzyme activities as well, it seems that several antioxidant systems were active in this genotype. F1 hybrid (7×8) may be the typical example of the com-plementary action of two antioxidant protective mechanisms in plants – enzymatic and non-enzymatic.

The obtained results on free proline content have shown significant differences among examined genotypes (Table 2). They varied from 0.266 to 2.286 µmol/g fresh material. the highest content of the aminoacid was recorded in genotypes venera, Meli, alisa and ln92-7369, and F1 hybrids (7×8) and (6×1). Except chinese cultivar ln92-7369, all other genotypes are domestic, and it seems that those genotypes are better adapted to our agroecological conditions (semiarid climates). it is known that soybean contain proline-rich proteins isolated from the cell walls. the proteins are similar in amino acid content, containing 20% proline, 20% hydroxyproline, 20% lysine, 16% valine, 10% tyrosine, and 10% glutamate. The characterisation of a

809

developmentally regulated proline-rich cell wall protein (sbPrP1) gene of soybean has been reported26; two other closely related members of this family of proline-rich protein (PrP) genes (sbPrP2 and sbPrP3), have been characterised as well. sbPrP3 is the major form of PrP gene expressed in aerial parts; it is highly expressed in leaves, although no expression is detected in the roots. it seems that some of our plant mate-rial, especially venera cultivar, possess PrP gene sbPrP3, but further investigation in this field is neccessary.

DPPh-radical scavenging activity is a measure of non-enzymatic antioxidant activity. higher levels of DPPh activity have been correlated with tolerance to differ-ent stress conditions5. in this assay system, antioxidants can react with the stable free DPPh radical and produce 1,1-diphenyl-2-picrylhydrazine. the change of absorb-ance produced in this reaction is assessed to evaluate the antioxidant potential of test samples. the preliminary distribution pattern of antioxidant activity in plant extracts may be categorised as: active (>80% inhibition), moderately active (50–80% inhibi-tion) and inactive (<50% inhibition)27. Our samples showed less than 50% inhibition which categorises them as inactive. the highest non-enzymatic antioxidant activity was recorded in leaves of genotypes ln92-7369 and F1 (5×1), 48.17 and 47.20%, respectively, where enzymatic antioxidant system were less active. it seems that plant phenolics and other non-enzymatic antioxidants were active and sufficient to prevent increased MDa formation in these genotypes (table 2).

conclusions

results showed that plants growing in arid and semiarid climates often face some degree of drought stress. in combination with some other environmental factors such as high outside temperatures and uv-radiation during summer months, oxidative stress may occur in plants. this affects many physiological and biochemical proc-esses, including increases in free proline, MDa and Gsh contents, and soD and GPx activity. these results demonstrate that there are a large number of parallel changes in plant responses to stress and complementary protective mechanisms. together, these antioxidant systems enhance the capability of plants to survive, adapt and grow dur-ing unfavourable periods. soybean genotypes with the best antioxidant ability with elevated free Pro and protein contents, such as F1 hybrid (7×8), are of special interest for field production, as well as for breeding.

acknoWlEDGEMEnt

this study was carried out within a project of the Ministry of science and Environ-mental Protection of the Republic of Serbia, Grant No TR-6852.

rEFErEncEs1. j. G. scanDalios: oxygen stress and superoxide Dismutases. Plant Physiol., 101, 7 (1993).

810

2. o. Blokhina, E. virolainEn, k. v. FaGErstEDt: antioxidants, oxidative Damage and oxygen Deprivation stress: a review ann. Bot., 91, 179 (2003).

3. s. PhilosoPh-haDas, s. MEir, B. akiri, j. kannEr: oxidative Defense systems in leaves of three Edible herb species in relation to their senescence rates. j. agric. Food chem., 42, 2376 (1994).

4. a. nizzaMuDDin: naDPh-dependent and o2.–-dependent lipid Peroxidation. Biochem. Edu.,

15, 58 (1987). 5. h. M. kanG, M. saltvEit: reduced chilling tolerance in Elongating cucumber seedling

radicles is related to their reduced antioxidant Enzyme and DPPh-radical scavenging activity. Physiol. Plant., 115, 244 (2002).

6. s. hErBEttE, c. lEnnE, n. lEBlanc, j. l. juliEn, j. r. DrEvEt, P. roEckEl-DrEvEt: two GPx-like Proteins from Lycopersicon esculentum and Helianthus annuus are antioxidant Enzymes with Phospholipid hydroperoxide Glutathione Peroxidase and thioredoxin Peroxidase activities. Eur. j. Biochem., 269, 2414 (2002).

7. j. G. strEEtEr: Effects of Drought on nitrogen Fixation in soybean root nodules. Plant cell Environ., 26, 1199 (2003).

8. h. hErzoG, k. P. Götz: Influence of Water Deficit on uptake and Distribution of Nitrogen in soybeans Monitored by soil injected 15n. j. agr. crop sci., 190, 161 (2004).

9. l. c. PurcEll, c. a. kinG: Drought and nitrogen source Effects on nitrogen nutrition, seed Growth, and yield in soybean. j. Plant nutr., 19, 969 (1996).

10. P. arMEnGauD, l. thiEry, n. Buhot, G. GrEniEr-DE March, a. savourE: tran-scriptional regulation of Proline Biosynthesis in Medicago truncatula reveals Developmental and Environmental Specific Features. Physiol. Plant., 120, 442 (2004).

11. F. lahrEr, l. lEPort, M. PEtrivalsky, M. chaPPart: Effectors for the osmo-induced Proline response in higher Plants. Plant Physiol. Biochem., 31, 911 (1993).

12. D. rEntsch, B. hirnEr, E. schMElzEr, W. B. FroMMEr: salt stress-induced Proline Transporters and Salt Stress-repressed Broad Specificity Amino Acid Permeases Identified by Sup-pression of a yeast amino acid Permease-targeting Mutant. Plant cell, 8, 1437 (1996).

13. c. B. taylor: Proline and Water Deficit: ups, Downs, Ins, and Outs. Plant cell, 8, 1221 (1996).14. r. s. DhinDsa, P. PluMP-DhinDsa, t. a. thorPE: leaf senescence: correlated with increased

levels of Membrane Permeability and lipid Peroxidation, and Decreased levels of superoxide Dismutase and catalase. j. Exp. Bot., 32, 93 (1981).

15. l. M. siMon, z. Fatrai, D. E. jonos, B. Matkovics: study of Metabolism Enzymes during the Development of Phaseolus vulgaris. Biochem. Physiol. Pflanzen, 166, 387 (1974.)

16. z. a. PlacEr, l. l. cushMan, B. c. johnson: Estimation of Product of lipid Peroxidation Malonyldialdehyde in Biochemical systems. anal. Biochem., 16, 359 (1966).

17. h. GiDrol, h. sErGihini, B. noBhani, B. MocQuot, P. Mazilak: Biochemical changes Induced by Accelerated Agent of Sunflower Seeds. Lipid Peroxidation and Membrane Damage. Physiol. Plant., 76, 591 (1987).

18. M. M. BRADFORD: a rapid and sensitive Method for the Quantitation of Microgram Quantities of Protein utilizing the Principle of Protein-dye Binding. anal. Biochem., 72, 248 (1976).

19. i. sEDlak, r. h. linDsay: Estimation of total Protein-bound and non Protein sulfhydryl-groups in tissue with Elmans reagent. anal. Biochem., 25, 192 (1968).

20. l. s. BatEs, l. P. WalDrEn, j. D. tEarE: rapid Determination of Free Proline for Water stress studies. Plant soil, 39, 205 (1973).

21. n. aBE, t. Murata, a. hirota: novel 1,1-Diphenyl-2-picrylhydrazyl-radical scavengers, Bisorbicillin and Demethyltrichodimerol, from a Fungus. Biosci. Biotechol. Biochem., 62, 661 (1998).

22. Dj. MalEncic, M. PoPovic, D. stajnEr, n. MiMica-Dukic, P. Boza, i. MathE: screen-ing for antioxidant Properties of Salvia nemorosa l. and Salvia glutinosa l. oxid. commun., 25, 613 (2002).

811

23. r. Baisak, D. rana, P. B. B. acharya, M. kar: alterations in the activities of active oxy-gen scavenging Enzymes of Wheat leaves subjected to Water stress. Plant cell Physiol., 35, 489 (1994).

24. h. Masaki, s. sakaki, t. atsuMi, h. sakurai: active-oxygen scavenging activity of Plant Extracts. Biol. Pharm. Bull., 18, 162 (1995).

25. Dj. MalEncic, D. vasic, M. PoPovic, D. DEvic: Antioxidant Systems in Sunflower as Af-fected by oxalic acid. Biol. Plant., 48, 243 (2004).

26. j. c. honG, r. t. naGao, j. l. kEy: Developmentally regulated Expression of soybean Proline-rich cell Wall Protein Genes. Plant cell, 1, 937 (1989).

27. s. k. lEE, z. h. MBWaMBo, h. chunG, l. luyEnGi, E. j. c. GaMEz, r. G. MEhta, a. D. kinGhorn, j. M. PEzzuto: Evaluation of the antioxidant Potential of natural Products. combinatorial chemistry & high throughput screening, 1, 35 (1998).

Received 13 December 2007 Revised 9 January 2008

812

Oxidation Communications 31, No 4, 812–818 (2008)

ozonAtion of pHenols in wAter solUtions intermediAtes And mecHAnism of destrUction. A review

a. DEMirEv

Technical University, Branch of Plovdiv, 4000 Plovdiv, Bulgaria E-mail: a.demirev@mail.bg

aBstract

Hereby has been made a scientific literary review of the mechanisms of destruction of phenols under ozonation in water medium and formed intermediates and final products of the oxidation. organic compound, constantly altering in the course of the process, predefines for each single moment the parameters of the treatment waters and in view of toxicity and recalcitrant nature of phenols the influence of intermediates on the resistance of ozonated substances to the biological process has been measured. in this respect the changes in chemical oxygen demand (coD), bidogical oxygen demand (BoD), biodegradability (BoD/coD ratio) and the toxicity of the treatment waters have been followed.

Keywords: phenols, ozonation, degradation, mechanism, intermediates.

aiMs anD BackGrounD

Phenols are present in industrial wastewaters of a solid number of manufacturers and this usually hampers the purification of waters due to the refractory and toxic character of this pollutants. some opinions claim that the standard methods can not be effective against the disturbing recalcitrant compounds1. toxicity of pollutants is a limitation for the use of biological purification of wastewaters with suspended and biofilm culture2. some of the destructive methods, i.e. wet catalytic oxidation of phenols, even increase their toxicity3. in many cases advanced oxidation (aoPs) processes prove flaws – presence of salts deactivates the hydrogen peroxide4 and the Fenton’s reagents5. the optically active phenols shield the dissolved molecular ozone6 and eliminate the synergism in the action of the gas and the uv-rays.

at the same time, ozone, as a powerful oxidant, is capable of oxidising a great number of toxic compounds, this resulting in organic acids of lower toxicity, sus-ceptible to biological treatment7. the technology of ozonation has a complex effect – pollutants are oxidised, and waters are deodorised, disinfected and enriched with oxygen alongside. ozone forms no additional sludge and removes colour8.

813

in view of the toxicity of phenols and their resistance to biological destruction, of great importance are the intermediates, composed in the course of ozonation, and final products of the oxidation process, and the mechanisms, respectively, leading to their formation. usually, the original organic compound is highly transformed and, in effect, the nature of the intermediates defines the parameters (toxicity, COD, BOD, BoD/coD ratio) of ozonated wastewaters.

Discussion

Degradation of organic matter under the influence of ozone runs at high speed, es-pecially in basic range. under those conditions the process is dominated by highly reactive hydroxyl (ho•) radicals – products of decomposition of ozone molecules in water. under acidic, neutral and low alkaline medium conditions ozone attacks the pollutant directly9.

According to the conditions, selective reaction molecular ozone – pollutant af-fects substitutes, should such exist, or the aromatic rings of refractory compounds. Formed ozonides are unstable and easily hydrolysed, emitting oxygen. the aromatic structures are decomposed to short-chain acyclic products – saturated and unsaturated carbonyl compounds, which later take part in the oxidation process.

hydroxybenzenes, dissolved in water, react easily with molecular ozone, in most cases for the oxidation of a molecule phenol are necessary 2 to 4 ozone molecules. The OH group is an electron–donor group and phenols show high electronic densi-ties on carbons located in the ortho- and para-positions, and so are highly reactive to ozone at these positions. The chemical reaction is multistage and at first stage a reactive phenoxyl radical is formed, which rapidly reacts with oxygen to quinone. Further oxidation of the initial products leads to cleavage of aromatic structure and formation of derivatives of unsaturated aldehydes and carboxylic acids with lower toxicity. the chemical interaction can lead to the end products of the degradation – CO2 and h2o, at high initial phenol concentration. if the initial concentrations are low enough the products of degradation, i.e. maleic acid, remain in the solution due to their low reactivity.

A much more active attack of the hydroxyl radicals over the effluent leads to the formation of organic peroxyl radicals at first. usually this reaction is irreversible and the peroxyl radicals via chain reaction results in the formation of organic oxyl radicals, or molecular products. as initial non-molecular products of phenols destruction are observed phenoxyl and semiquinone radicals10,11.

the nature of the pollutant establishes the way to transformation into the respec-tive molecular products. in the case of hydroxybenzene, o- and p-carbon atoms have increased electron density due to the presence of electron-donating hydroxyl group. As it is expected, in the first moments of the reaction, the electrophilic ozone attacks those particular positions. as electrophilic agents hydroxyl radicals attack primarily

814

o- (48%) and p- (36%) positions12. the development of oxidation process includes an opening of aromatic ring. in the mechanism of oxidation of hydroxybenzene by molecular ozone, proposed by Bailey13, catechol has been pointed out as the primary product of the destruction. Muconic acid and o-benzoquinone has also been registered. along with them Mokrini et al. also point out hydroquinone14. as intermediate products of oxidation process in the pH interval 2–12 Poznyak and Araiza15 also prove the pres-ence of catechol and muconic acid, along with fumaric and maleic acids. according to the authors, end product of degradation is oxalic acid. shang and yu observe an increasing toxicity and colouration of the ozonated at ph 7 phenol composite16 which they explain with the low dose of ozone.

Many sources in literature prove divalent representatives to be the initial mo-lecular products of phenol ozonation of phenol – catechol, resorcinol and hydroqui-none17–19. the same are decomposed to the corresponding quinones and acetals after destruction of aromatic ring under the influence of ozone20, also to the muconic acid, already mentioned. its further oxidation, by direct mechanism of ozonation as well as by radical mechanism, results in composure of maleic and fumaric acids. The final product of the oxidation process is oxalic acid21, 22.

Direct electrophilic attack of molecular ozone on catechol under acidic condi-tions – pH = 2 causes discomposure of the aromatic ring and formation of unsaturated carboxylic (c6) acids. They immediately transform into the identified maleic, fumaric, oxalic (with highest concentration registered), formic and glioxalic acids, also glyoxal, acetaldehyde and formaldehyde23. as a whole the substrate has no registered mutagenic action and has a biodegradability a few times higher than that of refractory catechol. according to sotelo et al.24, the reaction ozone–resorcinol also leads to the formation of unsaturated and saturated carboxylic acids.

among cresol isomers, the highest reactivity to ozone is manifested by m-cresol and the lowest – by o-cresol25. The identified intermediates during its oxidation are ac-ids – salicylic, glyoxylic, propionic, acetic and formic acids26. After approximately 50% reduction of coD there has been established a good anaerobic biodegradability.

a product of the initial dechlorination of o-chlorophenol during its oxidation by ozone is catechol27. Dechlorination of mono- and di-substituted representatives in the initial phase of ozonation (pH 2–12) is confirmed by Poznyak et al.28 on the other hand, according to shang et al.29, in neutral medium conditions 2-, 3- and 4-chloroph-enols react with ozone to chlorocatechol, chloromuconic acid and chlorinated dimers. these chlorinated intermediates could be the reason for the established increase of toxicity of the ozonised mixture. Formation of chlorinated intermediates (along with catechol) in the initial phase of ozonation of 2-chlorophenol is observed30. short-chain carboxylic acids – tartaric, oxalic, maleic and glycolic acids, were registered during the direct mechanism of ozonatio (ph 3).

Pentachlorophenol reacts directly with molecular ozone, while the initial nucleo-philic attack is directed against p-oriented substitute. the initial molecular products

815

of ozonolysis – chlorinated p-benzoquinone and p-hydroquinone are decomposed, now with the participation of hydroxyl radicals, to oxalic and (probably) to glyoxalic and formic acids31. the authors observe a much lower toxicity and an increasing with the time of ozonation biodegradability of the oxidising mixture, which is proved by the high value of the coefficient BOD5 / COD of the treated solution – 0.6 at the end of the process.

chlorinated cresol and 1,4-chlororesorcinol are decomposed to the respective quinones32. the presence of chlorides has been observed which indicates partial mineralisation of organic matter.

ozonation decreases strongly the toxicity of substituted chloro-, nitro- and aminophenols33. there is a highly increased biodegradability shown by the ozonated chloro- and nitrophenols, while with the aminophenols the result in this respect is negative. complete detoxication (and a considerable degree of mineralisation) of the ozonated nitrophenols is also established by trapido et al.34 Formation of nitrate ions was observed following the decomposition of 2- nitrophenol by ozone35.

intermediates of catalytic (tio2) and photocatalytic ozonation of p-nitrophenol and p-chlorophenol, with some exceptions are analogical to those, composed in other oxidation systems36. They could be classified in three different groups: polyphenols (catechol, resorcinol), unsaturated (maleic and fumaric) and saturated (formic and oxalic) acids.

Maleic, glyoxalic, oxalic and formic acids, along with hydroquinone and methyl-amine, are accepted as products of ozonation of water solutions of some aminophenols. Methylamine is the main analysed product in which nitrogen of metal is converted37. the toxicity of ozonated mixture is the lowest about the 20 min of the process after which it increases along with the increase in the concentration of methylamine.

conclusions

Phenols dissolved in water react easily with ozone. the initial nonmolecular products of reaction are ozonides, also phenoxyl and quinine radicals and their substituted representatives. the ozonides are easily hydrolysed, emitting oxygen, to short-chain acyclic compounds. Peroxyl radicals immediately attach to oxygen, forming the initial aromatic products – quinones, polyvalent and substituted phenols, aromatic acids.

Further development of oxidation process leads to destruction of aromatic struc-tures and composure of: aldehydes – acetaldehyde, formaldehyde, glyoxal; hydroxi-acids – glycolic, glyoxalic, tartaric acids; unsaturated and saturated carboxylic acids – fumaric, maleic and muconic, propionic, acetic, formic acids.

As a final product of destruction of phenols during their reaction with ozone many authors point out oxalic acid. its building up in the reactionary mixture could be the reason for the decrease in the speed of the macroprocess in its end.

816

indication for partial mineralisation of organic matter are the registered oxides of carbon, chlorides and nitrates – the last of them being observed during oxidation of chlorophenols and respectively of amino- and nitrophenols.

Biodegradability of ozonated organic matter increases, but according to alvares et al., over 90% parent compound transformation was required to give best results in the subsequent biological process8. For most phenols the phenol biodegradability ratio BoD5 / COD reaches the value ~ 0.4 after 50–70% reduction of COD. This shows a susceptibility to biological destruction of ozonated mixtures, if we accept for the index of biodegradability values of coefficient BOD5 / COD ~ 0.15 (Ref. 38).

ozonation strongly decreases the phenols toxicity33,34. and yet, there are data which show an unwanted alteration of this parameter during different phases of the process29,37. Obviously, the degree up to which the toxicity of the effluent can be overcome, is to be established manually for each particular system.

rEFErEncEs 1. M. j. cuGat, F. Borrull, M. calull: comparative study of capillary zone Electrophoresis

and Micellar Electrokinetic chromatography for the separation of twelve aromatic sulphonate compounds. chromatographia, 46 (3), (1997).

2. n. nikolova, v. nEnov: azo Dye schwarz Grs Bioconversion under various conditions. Water, air and soil Pollution, Focus 4 (2004).

3. a. santos, P. yustos, a. Quintanilla, F. Garcia-ochoa, j. a. casas, j. j. roD-riGuEz: Evolution of toxicity upon Wet catalytic oxidation of Phenol. Environmental science and technology, 38 (1), (2004).

4. h. GriGoroPoulou, c. PhiliPPoPoulos: homogeneous oxidation of Phenols in aqueous solution with hydrogen Peroxide and Ferric ions. Water science and technology, 36 (2), (1997).

5. k. a. FlahErty, c. P. huanG: in: chemical oxidation. vol. 2. continuous Flow applications of Fentons reagent for treatment of refractory Wastewaters (1st ed. technomic inc.). lancaster, 1992, 58–77.

6. M. traPiDo, j. kallas: Advanced Oxidation Processes for the Degradation and Detoxification of 4-nitrophenol. Environmental technology, 21 (2000).

7. h. stockinGEr, E. hEinzlE, o. M. kut: removal of chloro and nitro aromatic Waste-water Pollutants by ozonation and Biotreatment. Environmental science and technology, 29 (8), (1996).

8. A. B. C. ALVARES, C. DIAPER, S. A. PARSONS: Partial oxidation by ozone to remove recal-citrance from Wastewaters. a review. Environmental technology, 22 (4), (2001).

9. j. hoiGnE, h. BaDEr: the role of hydroxyl radical reactions in ozonation Processes in aque-ous solutions. Water research, 10, (1976).

10. h. utsuMi, s. k. han, k. ichikaWa: Enhacement of hydroxyl radical Generation by Phe-nols and their reaction intermediates during ozonation. Water science and technology, 38 (6), (1998).

11. s. k. han, k. ichikaWa, h. utsuMi: Generation of p-semiquinone radicals from chlorophe-nols in Water during ozonation. Water research, 32 (6), (1998).

12. E. Mvula, M. n. schuchMann, c. von sonntaG: Reactions of Phenol–OH–adduct radicals. Phenoxyl radical Formation by Water Elimination vs. oxidation by Dioxygen. j. of the chemical. society, Perkin transaction ii, 2 (2001).

817

13. P. s. BailEy: in: ozone in Water and Wastewater treatment. chap. 3. reactivity of ozone with Various Organic Functional Groups Important to Water Purification (Ed. F. L. Evans). 1972.

14. a. Mokrini, D. oussi, s. EsPluGas: oxidation of aromatic compounds with uv radiation/ozone/hydrogen Peroxide. Water science and technology, 35 (9), (1997).

15. t. Poznyak, G. B. araiza: ozonation of non-biodegradable Mixtures of Phenol and naphthalene Derivatives in tanning Wastewaters. ozone: science & Engineering, 27 (5), (2005).

16. n. c. shanG, y. h. yu: the Biotoxicity and color Formation results from ozonation of Wastewa-ters containing Phenol and aniline. j. of Environmental science and health, Part a: toxic/hazardous substances & Environmental Engineering, 36 (3), (2001).

17. P. c. sinGEr, M. D. Gurol: Dynamics of the ozonation of Phenols. i. Experimental observa-tions. Water science and technology, 17, (1983).

18. E. GILBERT: ozonation of aromatic compounds ph-dependence. Water science and technology, 14 (8), (1982).

19. x. F. zhu, x. h. xu: the Mechanism of Fe (iii)-catalyzed ozonation of Phenol. j. of zhejiang university science, 5 (12), (2004).

20. s. rakovsky, D. chErnEva, M. DEnEva, v. Ershov: kinetics and Mechanism of the reaction of ozone with Pyrocatechol and its Derivatives. oxid. commun., 20 (2), (1997).

21. y. c. hsu, h. c. yanG, j. W. liu: the ozonatoin of catechol and hydroquinone solutions using Gas-inducing reactor. j. of the chinese institute of Environmental Engineering, 13 (3), (2003).

22. z. ParishEva, a. DEMirEv: ozonization of aqueous solutions of resorcinol and catechol. Environment Protection Engineering, 27 (2), (2001).

23. c. zaror, v. carrasco, l. PErEz, G. soto, M. a. MonDaca, h. Mansilla: kinetics and toxicity of Direct reaction between ozone and 1, 2-dihydrobenzene in Dilute aqueous solu-tion. Water science and technology, 43 (2), (2001).

24. j. l. sotElo, F. j. BEltran, M. GonzalEz: ozonation of aqueous solutions of resorcinol and Phloroglucinol. 1. stoichiometry and absorption kinetic regime. industrial & Engineering chemistry research, 29 (1990).

25. y. zhEnG, c. h. kuo: Destruction of cresols by chemical oxidation. j. of hazardous Materials, 34 (2), (1993).

26. y. t. WanG: Effects of Preozonation on anaerobic Biodegradability of o-cresol. j. of Environmental Engineering, 115 (2), (1989).

27. M. a. Boncz, h. BrunninG, W. h. rulkEns, j. r. suDholtEr, G. h. harMsEn, j. W. BijstErBosch: kinetic and Mechanistic aspects of the oxidation of chlorophenols by ozone. Water science and technology, 35 (4), (1997).

28. T. POZNYAk, R. TAPIA, J. VIVERO, I. CHAIREZ: Effect of ph to the Decomposition of aqueous Phenols Mixture by ozone. j. of the Mexican chemical society, 50, 1 (2006).

29. n. c. shanG, y. h. yu, h. W. Ma: variation of toxicity during the ozonation of Monochloroph-enolic solutions. j. of Environmental science and health. Part a. toxic/hazardous substances & Environmental Engineering, 37 (2), (2002).

30. o. j. junG: Destruction of 2-chlorophenol from Wastewater and investigation of By-products by ozonation. Bulletin of the korean chemical society, 22 (8), (2001).

31. P. k. anDrEW honG, y. zEnG: Degradation of Pentachlorophenol by ozonation and Biodegrad-ability of intermediates. Water research, 36 (17), (2002).

32. M. traPiDo, y. vErEssinina, j. k. hEntunEn, a. hirvonEn: ozonation of chlorophenols: kinetics, By-products and toxicity. Environmental technology, 18 (3), (1997).

33. c. D. aDaMs, r. a. cozzEns, B. j. kiM: Effects of ozonation on the Biodegradability of sub-stituted Phenols. Water research, 31 (10), (1997).

818

34. M. traPiDo, y. vErEssinina, j. kallas: Degradation of aqueous nitrophenols by ozone combined with uv-radiation and hydrogen Peroxide. ozone: science & Engineering, 23 (4), (2001).

35. y. ku, j. j. hunG, W. y. WanG: Decomposition of 2-nitrophenol in aqueous solution by ozone and uv/ozone Processes. Water Environment research, 78 (9), (2006).

36. F. j. BEltrán, F. j. rivas, o. GiMEno: comparison between Photocatalytic ozonation and other oxidation Processes for the removal of Phenols from Water. j. of chemical technology & Biotechnology, 80 (9), (2005).

37. r. anDrEozzi, M. s. lo casalE, r. Marotta, G. Pinto, a. Pollio: n-methyl-p-ami-nophenol (metol) ozonation in aqueous solution: kinetics, Mechanism and toxicological charac-terization of ozonized samples. Water research, 34 (18), (2000).

38. Z. SHIYuN, Z. XuESONG, L. DAOTANG: ozonation of naphthalene sulfonic acids in aqueous solutions. Part i. Elimination of coD, toc and increase of their Biodegradability. Water research, 36 (5), (2002).

Received 14 November 2007 Revised 5 December 2007

819

Oxidation Communications 31, No 4, 819–825 (2008)

* For correspondence.

effect of rAdiAtion on tHe gUineA pigs’ Kidney tissUe protein oxidAtion, lipid peroxidAtion, glUtAtHione peroxidAse, And tHe effect of n-Acetylcysteine on tHis system

a. aricioGlua*, G. yilDiriMoGlua, M. akMansub, z. turkozEra

aDepartment of Biochemistry, Medical Faculty, Gazi Unıversity, Besevler, Ankara, Turkey bDepartment of Radiation Oncology, Medical Faculty, Gazi University, Besevler, Ankara, Turkey E-mail: aysela@gazi.edu.tr

aBstract

The most frequently used method for determining the oxidative modifications of proteins is the determining of carbonyl groups that are oxidation products of proteins. although the mechanisms of damage in biological systems by ionising radiation have not been completely clarified yet, it is believed that one of these mechanisms runs through free radicals. in our study, we investigated carbonyl group levels, malondialde-hyde (MDa) amounts, and glutathione peroxidase (GPx) activities in kidney tissue of guinea pigs which were administered 612 cGy radiation. Furthermore, we studied the effect of n-acetylcysteine (nac) on changes caused by radiation. the level of protein carbonyl groups, MDa and GPx activities were estimated spectrophotometrically.

While statistically significant changes were seen in carbonyl group levels and MDa levels in renal tissue related to radiation as compared to the control group (p<0.05), no significant changes were seen in GPx activities as compared to the control group.

reduction in protein carbonyl groups and MDa levels were seen in the group nac administered before radiation compared to the group to which only radiation was applied. however, this reduction in protein carbonyl groups was not found to be statistically significant (p>0.05). in GPx enzyme levels, however, statistically significant increases were seen in the group of which NAC was administered prior to radiation compared to the group with radiation alone (p<0.05). When only nac administered group was compared to the control group, no significant increases were found in renal tissue carbonyl group, MDa levels, and GPx activity.

With the direction of the data obtained from this study, it has been concluded that radiation increases the protein oxidation and lipid peroxidation of the renal tissue,

820

and application of nac before the radiation had protective effect against oxidative damage by reducing protein oxidation and lipid peroxidation.

Keywords: kidney irradiation, carbonyl groups, lipid peroxidation, glutathione per-oxidase, n-acetylcysteine.

aiMs anD BackGrounD

the kidneys are probably the most radiosensitive of the abdominal organs. renal irra-diation is associated with a chronic and persistent oxidative stress1. oxidative stress is caused by exposure to reactive oxygen species (ros), such as superoxide anion (o2

–), hydrogen peroxide ( h2o2) and hydroxyl radical (oh•), which can damage proteins, nucleic acids, and lipids2.the oxidation of proteins is caused by interaction of proteins with ros, which can be generated by ionising radiation, metal ion catalysed reac-tions, photochemical processes and enzyme catalysed redox reactions3. oxidatively modified proteins accumulate in different pathological conditions, including inflam-matory diseases, atherosclerosis, neurological disorders, ischemia reperfusion injury, carcinogenesis and renal disorders. Determination of carbonyl content in proteins can be used as a measure of oxidative protein damage4,5.

oxidative stress contributes to the pathophysiology of kidney injury. the admin-istration of various natural or synthetic antioxidants has been shown to be of benefit in prevention and attenuation of radical scavenging in numerous animal models of kidney diseases. these include vitamins, n-acetylcysteine (nac), alpha lipoic acid, melatonin and many others6.

nac, a synthetic precursor of reduced glutathione (Gsh), is a thiol containing compound, which stimulates the intracellular synthesis of Gsh, enhances glutath-ione–S-transferase activity, and acts solely as a reactive species scavenger7. Glutath-ione peroxidase (GPx) has been implicated in mediating the radioprotective effects of GSH. A mechanism has been proposed by which GSH may influence radiosensitivity via its role as the specific thiol substrate for GPx. Radiation-induced hydroperoxides are detoxified by enzimatically-mediated reduction. Depletion of both GSH and GPx together does not radiosensitise the kidney any more than was achieved by Gsh depletion alone8. chronic renal failure (crF) and hemodialysis (hD) patients have increased plasma protein oxidation manifested by oxidation of thiol groups and for-mation of carbonyl groups9.

in this study, the effect of nac on decreasing radiation-induced kidney dam-age was sought by measuring carbonyl groups, lipid peroxidation product MDa and GPx levels.

ExPEriMEntal

Materials and methods. 24 guinea pigs (having 400–500 g body weight) were divided into four groups: control group (group i) included non-irradiated guinea pigs, they

821

were only given intraperitoneal 0.9% NaCl for 5 days at a dose of 0.1 cm3/10 g; group II – the guinea pigs received 0.9% NaCl intraperitoneally at a dose of 0.1 cm3/10g followed by radiotherapy; group III – this group received N-acetylcysteine 300 mg/kg (intramuscular) for 5 days prior to irradiation; group iv- this group received n-ace-tylcysteine 300 mg/kg (intramuscular) for 5 days only.

all guinea pigs except the control group were irradiated with co-60 machine, total dose being 612 cGy. 24 hours following irradiation phenobarbital anaesthesia was given to all the guinea pigs. their kidneys were removed and rinsed in cold saline. kidney tissues were immediately put into liquid nitrogen to be stored at –70oc for further biochemical investigation.

Determination of protein carbonyl content in kidney was realised by 2,4-dini-trophenylhydrazine (DnPh) method as described by levine et al.10 the content of carbonyl groups was calculated using the molar extinction coefficient of 22 000 M–¹ cm–¹. the results were expressed as nmol carbonyl per mg protein.

Determination of lipid peroxidation was estimated by measuring the formation of thiobarbituric acid reactive substances (tBars) according to the method of uchiyama and Mihara11. 1,1,3,3-tetraethoxypropane was used as a standard, and the results were expressed as MDa/g tissue.

the glutathione peroxidase (GPx) activity was evaluated according to the Mishra et al.12 modification of the method described by Paglia and Valentine13. GPx activity was expressed as nmol oxidised naDPh/min/mg protein.

the protein concentration was determined using the method of lowry et al.14

The results were expressed as mean ± SD for six animals from each group (n=6). Statistical analysis was done using the Mann–Whitney u tests. Statistical significance was set up at p≤ 0.05.

rEsults

levels of carbonyl groups and MDa, and GPx activities in kidney tissues are shown in Table 1 and Figs 1–3.

table 1. levels of carbonyl groups and MDa and GPx activities in kidney tissues

Groups carbonyl groups(nmol/mg protein)

MDa(nmol/g tissue)

GPx(nmol oxide naDPh/

min/ mg protein)control (group i) 1.06±0.18 19.16±5.48 131.66±9.06radiation (group ii) 2.19±0.82a 105.33±86.93a 152.16±20.53b

radiation+nac (group iii) 1.56±0.31b, c 26.5±12.88b, c 220.33±38.44b, d

nac (group iv) 1.03±0.36a, d, e 19.83±7.86b, c, f 134.0±9.43a, c, e

Results were expressed as mean ± standard deviation. a p<0.05, compared to the corresponding value of control group; b p>0.05, compared to the corresponding value of control group; c p<0.05, compared to the corresponding value of radiation group; d p>0.05, compared to the corresponding value of radia-tion group; e p<0.05, compared to the corresponding value of nac group; f p>0.05, compared to the corresponding value of nac group.

822

0.00

0.50

1.00

1.50

2.00

2.50

3.00

carbonyl groups

cont

ent (

nmol

/mg

prot

ein)

controlradiationNAC + radiationNAC

fig. 1. levels of carbonyl groups in kidney tissue

0

50

100

150

200

250

MDA

cont

ent(

nmol

/gtis

sue)

controlradiationNAC + radiationNAC

fig. 2. levels of MDa in kidney tissue

0

50

100

150

200

250

300

GPx

GPx

activ

ity(n

mol

oxid

eN

AD

PH/m

in/m

gpr

otei

n)

controlradiationNAC + radiationNAC

fig. 3. levels of GPx activity in kidney tissue

Discussion

nucleic acids, proteins and membrane lipids are major targets of ros. oxidative stress occurs in several human diseases15,16. among these diseases are those in which high levels of protein carbonyl (co) groups have been observed, including alzheimer’s disease (aD), rheumatoid arthritis, diabetes, sepsis, chronic renal failure and respira-tory distress syndrome. Oxidative modification of proteins in vivo may affect a variety of cellular functions involving proteins, receptors, signal transduction mechanism,

823

transport systems and enzymes. oxidative protein damage may cause a decline in enzyme activities of the antioxidant system. carbonyl groups are commonly accepted as an oxidative modification marker. Accordingly, increases in the protein carbonyl content have been demonstrated in tissues after a pro-oxidant stress17.

In the present study, significantly elevated protein carbonyl content was observed in irradiated kidney tissues as compared to the control groups. Our findings are in agreement with the animal and human studies from other investigators. in recent years the oxygen effect in molecular as well as in cellular radiation biology has attracted more and more attention, because in many cases oxygen is still the best radiosensi-tiser18. Fliss and Menard19 showed for the first time that whole body exposure to high dose of ionising radiation resulted in a significant increase in protein oxidation in rat tissues including kidney. Exposure of the lungs to high doses of ionising radiation can initiate an injurious acute inflammatory response, consisting of a rapid neutrophil influx and oxidant production. Oxidative stress can cause significant oxidation of cellular proteins. the antioxidant dithiothreitol (Dtt), a potent oxidant scavenger and radioprotective agent, can protect against the oxidative damage to proteins. Fliss and coworkers shown that methionine is highly reactive with neutrophil oxidants and that the level of oxidised methionine in proteins can serve as a good indicator for oxidative stress20,21. oxidised methionine in the proteins of the lungs, heart, liver, kidney and jejunum increased significantly in 2 h after irradiation19. Protein carbonyls were studied in ageing and exercise22–25. Proteins of rat kidneys exhibited significant age-related increase in the amount of carbonyl while those of brain and liver did not22. radiation nephropaty is a characterised by a chronic progressive reduction in renal hemodynamics associated with progressive glomerular and tubular alterations resulting in glomerulosclerosis, tubulointerstial fibrosis and ultimate renal failure26,27. At the present time, the specific biochemical pathways by which patients with renal failure are subjected to increased oxidative stress are unclear. Patients with chronic renal failure (crF) develop increased levels of plasma protein oxidation, likely as a consequence of oxidant stress9.

the mechanism for the radioprotective effect of antioxidant enzyme GPx and antioxidant compound nac is still unclear. Depletion of both Gsh and GPx together did not radiosensitise the kidney any more than was achieved by Gsh depletion alone7,28,29. Our data show that no significant changes were seen in GPx activity in renal tissue related to radiation as compared to the control group. When nac was given before radiation, GPx enzyme activities statistically increased compared to the radiation group. a high dose of ionising irradiation causes excessive oxidative stress in kidney1. oxidative protein damage may cause the decline in enzyme activities of the antioxidant system25.

in our study, MDa content increased in irradiated kidney tissues, but decrease in the group to which nac was given before radiation when compared to the group to which only radiation was administered. our results are in agreement with other authors. NAC protects patients with moderate chronic renal insuffiency from contrast-

824

induced deterioration in renal function after coroner angiography procedures, with minimal adverse effects6,30.

in accordance with the direction of the data obtained from this study, it has been concluded that radiation increases the protein oxidation and lipid peroxidation of the renal tissue, and application of n-acetylcysteine before the irradiation had protective effect against oxidative damage by reducing protein and lipid peroxidation.

rEFErEncEs 1. M. E. c. roBBins, W. zhao, c. s. Davis, s. toyokuni, s. M. BonsiB: radiation-induced

kidney injury: a role for chronic oxidative stress? Micron, 33, 133 (2002). 2. G. storz, j. a. iMlay: oxidative stress. current opinion in Microbiology, 2, 188 (1999). 3. j. M. FaGan, B. G. slEczka, i. sohar: Quantitation of oxidative Damage to tissue Proteins.

int. j. Biochem. cell Biol., 31, 751 (1999). 4. E. r. staDMan: role of oxidized amino acids in Protein Breakdown and stability. Methods

Enzymol., 258, 379 (1995). 5. E. r. staDtMan, c. n. olivEr: Metal-catalyzed oxidation of Proteins. j. Biol. chem., 266,

2005 (1991). 6. l. tylicki, B. rutkoWski, W. h. hörl: antioxidants: a Possible role in kidney Protection.

kidney Blood Pres. res., 26, 303 (2003). 7. o. aruoMa, B. halliWEll, B. hocy, j. ButlEr: the antioxidant action of n-acetylcysteine:

its reaction with hydrogen Peroxide, hydroxyl radical, superoxide, and hypochlorous acid. Free radic. Biol. Med., 6, 593 (1989).

8. G. n. stEvEns, M. c. joinEr, B. joinEr, h. johns, M. r. l. stratForD: role of Glu-tathione Peroxidase in the radiation response of Mouse kidney. int. j. radiation oncology Biol., 16, 1213 (1989).

9. j. hiMMElFarB, E. McMonaGlE, E. McMEnaMin: Plasma Protein thiol oxidation and carbonyl Formation in chronic renal Failure. kidney intern., 58, 2571 (2000).

10. r. l. lEvinE, D. GarlanD, c. n. olivEr, a. aMici, i. cliMEnt, a. G. lEnz, B. W. ahn, s. shaltiEl, E. r. staDtMan: Determination of Carboyl Content in Oxidatively Modified Proteins. Methods in Enzymology, 186, 464 (1990).

11. M. uchiyaMa, M. Mihara: Determination of Malondialdehyde Precursor in tissues by thio-barbituric acid test. analytical Biochem., 86, 271 (1978).

12. o. P. Mishra, D. M. PaPaDoPulos, l. c. WaGErlE: anti-oxidant Enzymes in the Brain of newborn Piglets during ischemia Followed by reperfusion. neuroscience, 35, 211 (1990).

13. D. E. PaGlia, W. n. valEntinE: studies on the Quantitative and Qualitative characterization of Erythrocyte Glutathione Peroxidase. j. Biol. chem., 249, 7130 (1974).

14. o. h. loWry, n. j. rosEBrouGh, a. l. Farr, r. j. ranDal: Protein Measurement with the Folin Phenol reagent. j. Biol. chem., 193, 265 (1951).

15. i. DallE-DonnE, r. rossi, D. Giustarini, a. Milzani, r. coloMBo: Protein carbonyl Groups as Biomarkers of oxidative stress. clinica chimica acta, 329, 23 (2003).

16. i. DallE-DonnE, r. rossi, r. coloMBo, D. Giustarini, a. Milzani: Biomarkers of oxidative Damage in human Disease. clin. chem., 52, 601 (2006).

17. r. l. lEvinE: Carbonyl Modified Proteins in Cellular Regulation, Aging, and Disease. Free Radical Biology and Medicine, 32, 790 (2002).

18. h. schuEslEr, k. schillinG: oxygen Effect in the radiolysis of Proteins. Part 2. Bovine serum albumin. int. j. radiat. Biol., 45, 267 (1984).

19. h. Fliss, M. MEnarD: rapid neutrophil accumulation and Protein oxidation in irradiated rat lungs. j. appl. Physiol., 77, 2727 (1994).

825

20. h. Fliss: oxidation of Proteins in rat heart and lungs by Polymorphonuclear leukocyte oxidants. Mol. cell. Biochem., 84, 177 (1988).

21. h. Fliss, h. WEissBach, n. Brot: oxidation of Methionine residues in Proteins of activated human neutrophils. Proc. natl. acad. sci. usa, 80, 7160 (1983).

22. s. Goto, a. nakaMura, z. raDak, h. nakaMoto, r. takahashi, k. yasuDa, y. sakurai, n. ishii: carbonylated Proteins in aging and Exercise: immunoblot approaches. Mechanisms of aging and Development, 107, 245 (1999).

23. E. r. staDtMan: Protein oxidation in aging and age-related Diseases. annals of the new york academy of sciences, 928, 22 (2001).

24. E. r. staDtMan: Protein oxidation and aging. science, 257, 1220 (1992).25. l. tjan, Q. cai, h. WEj: alterations of antioxidant Enzymes and oxidative Damage to Macro-

molecules in Different organs of rats during aging. Free radical Biology and Medicine, 24, 1477 (1998).

26. M. E. c. roBBins, s. M. BonsiB: radiation nephropathy: a review. scanning Microscopy, 9, 535 (1995).

27. M. E. c. roBBins, y. o’MallEy, W. zhao, c. s. Davis, s. M. BonsiB: the role of the tubulointerstitium in radiation induced renal Fibrosis. radiation reserch, 155, 481 (2001).

28. j. E. BiaGloW, M. E. varnEs, E. P. clark, E. r. EPP: the role of thiols in cellular response to radiation and Drugs. radiat. res., 95, 437 (1985).

29. s. l. MarklunD, n. G. WEstMan, G. roos, j. carlsson: radiation resistance and the cuzn superoxide Dismutase, Mn superoxide Dismutase, catalase and Glutathione Peroxidase activities of seven human cell lines. radiat. res., 100, 115 (1984).

30. j. kay, W. h. choW, t. M. chan, s. k. lo, o. h. kWok, k. yi Fan, c. h. lEE, W. F. laM: acetylcysteine for Prevention of acute Deterion of renal Function Following Elective coronary angiography and intervention: a randomized controlled trial. jaMa, 289, 553 (2003).

Received 31 January 2008 Revised 19 March 2008

826

Oxidation Communications 31, No 4, 826–840 (2008)

* For correspondence.

epoxidAtion of metHAllyl cHloride witH Hydrogen peroxide over ti-mcm-48 cAtAlyst

a. WroBlEWska*, j. WajzBErG, E. MilchErt

Institute of Organic Chemical Technology, Szczecin University of Technology, 10 Pulaskiego Street, PL 70 322 Szczecin, Poland E-mail: agnieszka.wroblewska@ps.pl

aBstract

The epoxidation of methallyl chloride (MAC) with a 30% hydrogen peroxide has been presented. in the process there were used methanol as a solvent and mesoporous titanium silicalite Ti-MCM-48 as a catalyst. The process has been carried out by two methods: under autogenic pressure and under atmospheric pressure. the effect of the following parameters has been studied: temperature, molar ratio of Mac/h2o2, solvent concentration, catalyst concentration and reaction time. the process was described by the following functions: the selectivity of transformation to 2-methylepichlorohydrin (MEch) in relation to Mac consumed, the selectivity of transformation to organic compounds in relation to h2o2 consumed, the conversions of Mac and hydrogen peroxide. 2-Methylepichlorohydrin (MEch) is widely applied, mainly for produc-tion of epoxy resins.

Keywords: Ti-MCM-48 catalyst, zeolites, methallyl chloride, epoxidation, 2-me- thylepichlorohydrin.

aiMs anD BackGrounD

in recent years much attention has been paid to development of new catalysts of epoxidation by hydrogen peroxide, taking into regard the improvement of their se-lectivity (also shape-selectivity of the catalysts), easy regeneration and recovery and the structure stability during the process1. the zeolite titanium silicalite catalysts have been studied since the 1980’s. The first catalyst from this group is TS-1 (Refs 2–5). It is effective in epoxidation processes of linear olefins with 30% hydrogen perox-ide, however its effectiveness in epoxidation of unsaturated compounds of branched structure and large molecules is much lower. the molecules of such compounds can not reach the active sites of the catalyst localised in the narrow pores of 5.6–5.3 Å in diameter. in view of the above, one direction of the new catalysts development was to obtain titanium silicalite materials with pores of larger size. the catalysts of this

827

group include: Ti-MWW (Ref. 6), Ti-Beta (Ref. 7), Ti-MCM-41 (Ref. 8), Ti-MCM-48 (Refs 9–11). Among them Ti-MCM-48 belongs to mesoporous materials with three-dimensional structure of pores and their diameter reaching 100 nm (ref. 10).

Epoxidation of methallyl chloride (MAC) with a 30% hydrogen peroxide over Ti-MCM-48 as a catalyst permits getting 2-methylepichlorohydrin (MECH), used for production of epoxy resins and further glues for adhesion of metals, glass, ceramics and wood, varnishes and epoxy paints used for protection against corrosion12. 2-Me- thylepichlorohydrin has been also used for modification of the thermoplastic proper-ties of polycarbonates. Because of their ability to fill up cavities and hollows these compounds are used for production of small elements13. Polyeterols and polyeterol esters obtained on the basis of 2-metyloepichlorohydrin are employed for production of polyurethane. They permit production of polyurethane fibres with hydrophilous, antistatic and antistain properties14. Pure 2-methylepichlorohydrin has also been used for production of pesticides15.

The paper presents the results of investigations the influence of different techno-logical parameters on Mac epoxidation under the atmospheric or autogenic pressure. the obtained results of experiments allowed calculations of the values of the following functions: the selectivity of transformation to MEch in relation to Mac consumed (SMEch/Mac), the selectivity transformation to organic compounds in relation to h2o2 (S org/h2o2

) consumed, the conversion of Mac (CMac) and the conversion of h2o2 (Ch2o2).

these functions were calculated from the following equations:

SMECH/MAC =amount of MECH obtainedamount of MAC consumed × 100%

,

Sorg/H2O2=

amount of organic comp. obtainedamount of H2O2 consumed

× 100% ,

CMAC =amount of MAC consumed

initial amount of MAC × 100% ,

CH2O2=

amount of H2O2 consumedinitial amount of H2O2

× 100% .

on the basis of analysis of these functions the best technological parameters of the Mac epoxidation process have been chosen.

ExPEriMEntal

Raw materials. in the epoxidation process there were used: methallyl chloride (Mac) (98 wt.%, Fluka), hydrogen peroxide (30 wt.%, POCh Gliwice, Poland), methanol (analytical grade, POCh Gliwice, Poland) and Ti-MCM-48 catalyst (obtained in In-stitute of organic chemical technology, szczecin university of technology).

828

Preparation of Ti-MCM-48 catalyst. Ti-MCM-48 catalyst was synthesised by the method proposed by schumacher et al.16 By xrF measurement the content of titanium in this catalyst expressed in tio2 was 0.41 wt. %. The IR spectrum of the catalyst, taken on a jasco Ft/ir instrument with the concentration of the catalysts in kBr pellets 1.4 wt.%, revealed an intense absorption band at 960 cm–1. the uv-vis. spec-trum recorded on a sPEcorD M40 showed a band at 220 nm and no band at 300 nm, which means that ti4+ ions had been incorporated into the crystalline structure of silica and no tio2 in the form of anatase had been detected. the crystalline structure of the Ti-MCM-48 catalyst was confirmed by the XRD method using an XPERT PRO diffractometer with a cobalt lamp as the source of irradiation at 0.179 nm (Fig. 1).

2 Θ (degree)

inte

nsity

(a.u

.)

0

2000

4000

6000

8000

10000

12000

14000

16000

2 3 4 5 6 7 8 9 10 11

fig. 1. XRD pattern of the Ti-MCM-48 catalyst obtained

Epoxidation procedure, apparatus and analytical methods. in the pressure method, a stainless steel autoclave equipped with a teflon insert of 7 cm3 in capacity was used. into the autoclave the reagents were introduced in the following sequence: the catalyst, Mac, hydrogen peroxide and methanol (as a solvent). in the method of synthesis under the atmospheric pressure a glass reactor was used, equipped with a reflux, thermometer, magnetic stirrer and a dropping funnel. The reagents placed subsequently in the glass reactor were: the catalyst, Mac, methanol and the contents were heated to the temperature of the process. having reached a desired temperature 30% hydrogen peroxide was dropped on intense stirring.

The influence of the following technological parameters of the process were tested: temperature in the range 20–120oC (under autogenic pressure), 20–60oc (under atmos-pheric pressure), molar ratio of Mac/h2o2 in the range 1:1–5:1, solvent concentration (methanol) in the range 5–90 wt.%, Ti-MCM-48 catalyst concentration in the range 0.1–5.0 wt.%, and the reaction time in the range 0.5–5.0 h. The initial parameters for both methods were the same: molar ratio of Mac/h2o2 of 1:1, methanol concentration 40 wt.%, Ti-MCM-48 catalyst concentration 1 wt.%, and reaction time 3 h.

829

Quantitative analysis of the products was made by gas chromatography on an instrument CHROM 5 with a flame-ionisation detector (FID). The parameters of chromatographic separation were as follows: nitrogen flow 40 cm3/min, air flow 40 cm3/min, hydrogen flow 30 cm3/min, detector temperature 250oc, and sample chamber temperature 250oc. the oven temperature was changed according to the following programme: isothermally at 170oc for 14 min, temperature increase at the rate 15oc/min, isothermally at 200oc for 10 min, temperature increase at the rate 20oc/min, iso-thermal at 220oC for 8 min, and cooling by 2 min to 170oc. the unreacted hydrogen peroxide was iodometrically determined17. after the mass balance in each synthesis, the above-defined functions characterising the process were calculated.

rEsults

The epoxidation of MAC over Ti-MCM-48 catalyst with 30% hydrogen peroxide and in methanol as a solvent leads to formation of 2-methylepichlorohydrin:

CH2 C CH2 Cl

CH3H2O2,Ti-MCM-48

-H2OH2C C CH2 Cl

CH3

O

methallyl chloride 2-methylepichlorohydrin

2-Methylepichlorohydrin is susceptible to the attack of nucleophilous particles present in the reaction medium, which leads to the formation of 1-chloro-2-methyl-propane-2,3-diol,3-chloro-2-methoxy-2-methylpropane-1-ol, and 1-chloro-3-meth-oxy-2-methylpropane-2-ol.

H2C C CH2 Cl

CH3

O

H2OH2C C CH2 Cl

CH3

OHOH

2-methylepichlorohydrin 1-chloro-2-methylpropane-2,3-diol

H2C C CH2 Cl

CH3

O

CH3OHH2C C CH2 Cl

CH3

OCH 3OH

or H2C C CH2 Cl

CH3

OHOCH 3

2-methylepichlorohydrin 3-chloro-2-methoxy- 1-chloro-3-methoxy-2- 2-methylpropane-1-ol methylpropane-2-ol

in much smaller amounts the formation of ethers, polymers and copolymers of 2-methylepichlorohydrine and methallyl chloride takes place18.

Effect of temperature. the effect of temperature on the course of Mac epoxidation by the autogenic pressure method was studied in the range 20–120oc.

830

as results from Fig. 2, the selectivity of transformation to MEch in relation to Mac consumed increases with increasing temperature in the range from 20 to 60oc, whereas for temperatures the range 60–120oc this selectivity is constant. the low selectivity in low temperatures is related to a slow rate of epoxidation. With tempera-ture increasing to 60oc, the rate of the reaction increases, more MEch is formed at increasing MAC conversion. The MAC conversion initially very low of 0.2 mol % at 20oC, with increasing temperature grows and reaches about 11 mol. % at 60oc, with a further increase in temperature it remains constant to 120oc. With temperature in-creasing from 20 to 60oc the conversion of h2o2 increases from 10 to 50 mol. %, and with a further temperature increase to 100oc it increases by a few percent. the rapid increase in the conversion of h2o2 above 100oc is caused by a fast decomposition of hydrogen peroxide. as results from these temperature dependencies the optimum temperature of the process is 60oc. the effect of temperature on the Mac epoxida-tion under the atmospheric pressure was analysed in the range 20–60oc, at the other parameters unchanged.

0

10

20

30

40

50

60

70

80

90

100

20 40 60 80 100 120

temperature (°C)

S MEC

H/M

AC, S

org/

H2O

2, CM

AC, C

H2O

2 (mol

.%)

fig. 2. Influence of temperature on MAC epoxidation under autogenic pressure: selectivity of transfor-mation to MECH in relation to MAC consumed (♦), selectivity of transformation to organic compounds in relation to h2o2 consumed (■), conversion of MAC (▲), conversion of H2o2 (×)

as results from Fig. 3, with temperature increasing from 20 to 40oc, the selectiv-ity of transformation to MEch in relation to Mac consumed increases from 0 to 10 mol. %, and it does not change in higher temperatures. The selectivity of transformation to organic compounds in relation to h2o2 consumed slightly decreases above 40oc. With temperature increasing in the range 20–60oc, the conversion of h2o2 increases. in the range of temperatures 40oc the consumption of h2o2 is mainly related to the proc-ess of epoxidation. With increasing temperature the rate of h2o2 decomposition into oxygen and water increases, so the selectivity of transformation to MEch in relation to h2o2 consumed decreases. the conversion of Mac undergoes the smallest changes in this temperature range. as indicated by the course of the function: SMEch/Mac, Sorg/h2o2

,

831

CMac, Ch2o2 , the optimum temperature of Mac epoxidation under the atmospheric

pressure is 40oC. This conclusion follows first of all from the course of the function of selectivity of transformation to MEch in relation to Mac consumed and selectivity of transformation to organic compounds in relation to h2o2 consumed.

0

10

20

30

40

50

60

70

80

90

100

20 30 40 50 60temperature (°C)

S MEC

H/M

AC, S

org/

H2O

2, CM

AC, C

H2O

2 (mol

.%)

fig. 3. Influence of temperature on MAC epoxidation under the atmospheric pressure: selectivity of transformation to MECH in relation to MAC consumed (♦), selectivity of transformation to organic compounds in relation to h2o2 consumed (■), conversion of MAC (▲), conversion of H2o2 (×)

Influence of molar ratio of MAC/H2O2. The influence of the molar ratio of MAC/H2o2 on the syntheses performed by both autogenic and atmospheric pressure methods was studied in the range 1:1–5:1. In the autogenic pressure method the reaction was conducted at 60oc, while the other parameters were left unchanged (Fig. 4).

0

10

20

30

40

50

60

70

80

90

100

1 2 3 4 5

molar ratio (mol/mol)

S MEC

H/M

AC, S

org/

H2O

2, CM

AC, C

H2O

2 (mol

.%)

fig. 4. Influence of the molar ratio of MAC/H2o2 on Mac epoxidation under autogenic pressure: selectiv-ity of transformation to MECH in relation to MAC consumed (♦), selectivity of transformation to organic compounds in relation to h2o2 consumed (■), conversion of MAC (▲), conversion of H2o2 (×)

832

in the autogenic pressure method the selectivity of transformation to MEch in relation to MAC consumed decreases from 44 mol.% at the molar ratio 1:1 to 0 mol.% at the molar ratio 5:1. With the molar ratio increasing in the above range, the selectiv-ity of transformation to organic compounds in relation to h2o2 consumed increases from 11 to 17 mol.%, but a greater increase in this selectivity occurs at the molar ratio of Mac/h2o2 above 3:1. With increasing molar ratio the conversion of Mac decreases from 5 to 0.7 mol.% and the conversion of H2o2 also decreases. it results from a considerable dilution of the reaction medium and points to a low contribution of the side reactions involving Mac. For further analysis the molar ratio of 1:1 was chosen because of a high selectivity of transformation to MEch in relation relative to the Mac consumed.

The influence of the molar ratio MAC/H2o2 under the atmospheric pressure was studied at 40oc, while the other parameters were the same as in the autogenic pressure method. With increasing the molar ratio of Mac/h2o2 the selectivity of transforma-tion to MECH in relation to MAC consumed decreases from 10 to 0 mol.% for the ratio above 3:1 (Fig. 5). simultaneously, the selectivity of transformation to organic compounds in relation to h2o2 consumed increases from 19 mol.% at MAC/H2o2=1:1 to 100 mol.% above MAC/H2o2=3:1. this observation indicates a substantial increase in the effective use of h2o2 in the synthesis of MEch and its derivatives. the conver-sions of h2o2 and Mac in the molar ratio range studied are practically unchanged.

0

10

20

30

40

50

60

70

80

90

100

1 2 3 4 5molar ratio (mol/mol)

S MEC

H/M

AC, S

org/

H2O

2, CM

AC, C

H2O

2 (mol

.%)

fig. 5. Influence of the molar ratio of MAC/H2o2 on Mac epoxidation under the atmospheric pressure: selectivity of transformation to MECH in relation to MAC consumed (♦), selectivity of transformation to organic compounds in relation to h2o2 consumed (■), conversion of MAC (▲), conversion of H2o2 (×)

the optimum Mac/h2o2 molar ratio for the two methods is 1:1. in each method the increasing molar ratio causes a decrease in the selectivity of transformation to MEch in relation to Mac consumed. under the atmospheric pressure this function decreases to 0% at a molar ratio of 3:1, whereas under the autogenic pressure – above a molar ratio of 5:1. in the optimum conditions and under the autogenic pressure the

833

values of SMEch/Mac are by about 30 mol.% higher than under the atmospheric pres-sure.

Iinfluence of solvent concentration. the influence of methanol concentration on the synthesis performed by the autogenic pressure method was studied in the range 5–90 wt.%, at 60oc and at a molar ratio of Mac/h2o2 of 1:1. the other parameters were unchanged.

0

10

20

30

40

50

60

70

80

90

100

5 15 25 35 45 55 65 75 85

methanol concentration (wt.%)

S MEC

H/M

AC, S

org/

H2O

2, CM

AC, C

H2O

2 (mol

.%)

fig. 6. Influence of methanol concentration on MAC epoxidation under autogenic pressure: selectivity of transformation to MECH in relation to MAC consumed (♦), selectivity of transformation to organic compounds in relation to h2o2 consumed (■), conversion of MAC (▲), conversion of H2o2 (×)

as shown in Fig. 6, with an increase in the methanol concentration from 5 to 40 wt.%, the function SMEch/Mac rapidly increases from 0.8 (5 wt.% of methanol) to 44 mol.% (40–50 wt.% of methanol). Further increase in the methanol concentra-tion to 90 wt.% results in a decrease in this function to 0 mol. %. The reason for this behaviour is methanolysis of the epoxy ring. the selectivity of transformation to organic compounds Sorg/h2o2

and conversion of Mac increase on increasing methanol concentration. A high concentration of methanol increases the efficiency of the use of h2o2 for obtaining organic compounds. At the methanol concentration of 5 wt.% the conversion of MAC is 0 mol.%, while at the methanol concentration of 90 wt.% it increases to 13 mol.%. The selectivity of transformation to organic compounds in relation to h2o2 consumed (Sorg/h2o2

) increases from 0 at 5 wt.% of methanol to 37 mol.% at 90 wt.% of methanol. The high concentration of methanol facilitates for-

834

mation of a 5-member active compound whose structure – assuming the mechanism proposed by a. corma19 – can be illustrated as:

OTi

O

SiO OSiCH3OH

HO

CH 2

C(CH 3)

CH 2Cl

H3C

in this mechanism important is the role of the molecules of h2o2 and ch3oh. The influence of methanol concentration on the course of the epoxidation under the

atmospheric pressure was studied at 40oc and at the Mac/h2o2 molar ratio of 1:1.as results from Fig. 7, the selectivity of transformation to MEch increases from

2 to 10 mol.% on increasing methanol concentration from 5 to 40 wt.%. Further in-crease in the methanol concentration leads to a decrease in this function to 0 mol.% at the methanol concentration of 80 wt.%. With the methanol concentration increas-ing above 40 wt.% the selectivity of transformation to organic compounds Sorg/h2o2

significantly increases. At the methanol concentration of at least 80 wt.% H2o2 is totally reacted to organic compounds (2-methylepichlorohydrin and its derivatives), which is unfortunately accompanied by a decrease in the conversion of h2o2, so the product contains an increasing contribution of unreacted h2o2.

0

10

20

30

40

50

60

70

80

90

100

5 15 25 35 45 55 65 75 85

methanol concentration (wt.%)

S MEC

H/M

AC, S

org/

H2O

2, CM

AC, C

H2O

2 (mol

.%)

fig. 7. Influence of methanol concentration on MAC epoxidation under the atmospheric pressure: selectiv-ity of transformation to MECH in relation to MAC consumed (♦), selectivity of transformation to organic compounds in relation to h2o2 consumed (■), conversion of MAC (▲), conversion of H2o2 (×)

The influence of methanol concentration on the epoxidation performed by the two methods is similar. the processes are the most effective at the methanol concen-

835

tration of 40 wt.%. Increase in the methanol concentration to 60 wt.%, does not lead to significant changes in the functions SMEch/Mac and CMac. applying the autogenic pressure method higher values of the functions describing the process are obtained. When applying the autogenic pressure method, the function SMEch/Mac at the optimum concentration of methanol of 40 wt.%, is by about 30 mol.% higher than when car-rying out the epoxidation under the atmospheric pressure method. in both methods, the selectivity of transformation to organic compounds in relation to h2o2 consumed (Sorg/h2o2

) increases with increasing methanol conversion. in the autogenic pressure method, when the methanol concentration gets over 80 wt.%, all the H2o2 is used in the process of epoxidation. in the atmospheric pressure method about half of the h2o2 introduced into the reactor is used in the process.

Effect of catalyst concentration. The influence of the catalyst concentration in the autogenic pressure method was studied at 60oc, at a molar ratio of Mac/h2o2 1:1, at the methanol concentration of 40 wt. % and after 3 h of the reaction (Fig. 8). In the autogenic pressure method the selectivity of transformation to MEch in relation to MAC consumed increases from 42 to 55 mol.% with the catalyst concentration increasing from 0.1 to 2 wt.%. Further increase in the concentration of Ti-MCM-48 causes a decrease in the function SMEch/Mac to 40 mol.% at 5 wt.% of Ti-MCM-48. the selectivity of transformation to organic compounds in relation to h2o2 consumed reaches the highest value of 34 mol.% at the catalyst concentration of 5.0 wt.%. In the analysed range of the catalyst concentration the conversion of Mac increases from 0.4 to 9 mol.%. The conversion of H2o2 reaches the highest value of about 41 mol.% at the catalyst concentration of 2 wt.%. Hence, the optimum concentration of the catalyst is 2 wt.%. At its concentrations above 3 wt.%, the conversion of MAC is higher, but the contribution of the side reactions increases, which decreases the selectivity of transformation to MEch in relation to Mac consumed.

0

10

20

30

40

50

60

70

80

90

100

0 0 5 1 1 5 2 2 5 3 3 5 4 4 5 5

Ti-MCM-48 concentration (wt.%)

S MEC

H/M

AC, S

org/

H2O

2, CM

AC, C

H2O

2 (mol

.%)

fig. 8. Influence of Ti-MCM-48 concentration on MAC epoxidation under autogenic pressure: selectivity of transformation to MECH in relation to MAC consumed (♦), selectivity of transformation to organic compounds in relation to h2o2 consumed (■), conversion of MAC(▲), conversion of H2o2 (×)

836

in the syntheses taking place under the atmospheric pressure (Fig. 9) the se-lectivity of transformation to MECH increases from 2.5 mol.% at the Ti-MCM-48 concentration of 0.1 to 18 mol.% at the catalyst concentration of 5 wt.%. The selectiv-ity of transformation to organic compounds in relation to h2o2 consumed decreases from 37 to 16 mol.% with the catalyst concentration increasing in the whole range analysed. the conversion of Mac increases with the catalyst concentration increas-ing in the whole range studied, reaching 15 mol.% at the catalysts concentration of 5 wt.%. The conversion of H2o2 is the highest at the catalyst concentration of 5 wt.% and reaches 94 mol.%. Therefore, the optimum concentration of the catalyst in this method is 5 wt.%.

0

10

20

30

40

50

60

70

80

90

100

0 0 5 1 1 5 2 2 5 3 3 5 4 4 5 5Ti-MCM-48 concentration (wt.%)

S MEC

H/M

AC, S

org/

H2O

2, CM

AC, C

H2O

2 (mol

.%)

fig. 9. Influence of Ti-MCM-48 concentration on MAC epoxidation under the atmospheric pressure: selectivity of transformation to MECH in relation to MAC consumed (♦), selectivity of transformation to or-ganic compounds in relation to h2o2 consumed (■), conversion of MAC (▲), conversion of H2o2 (×)

The studies of the influence of changes in the catalyst Ti-MCM-48 concentra-tion showed considerable differences in the course of the function SMEch/Mac in the two methods investigated.. under the autogenic pressure there was achieved the higher selectivity of transformation to MEch than under the atmospheric pressure. simultaneously, the higher selectivities transformation to MEk concern low catalyst concentrations. in the reaction under the atmospheric pressure the highest selectivity is reached at the highest catalyst concentration. in the process under the autogenic pressure, with increasing concentration of the catalyst, the selectivity of transforma-tion to organic compounds Sorg/h2o2

increases and reaches 31 mol.% at 5 wt. % of Ti-MCM-48 catalyst. under the atmospheric pressure this value of the selectivity of transformation to organic compounds in relation to hydrogen peroxide is reached at the catalyst concentration of 0.1 wt.%. So, the influence of the catalyst concentration on this function is the opposite in the two methods. a similar course was found for the conversion of Mac in both methods.

837

Effect of reaction time. the effect of the reaction time was studied for the parameters recognised as the optimum in the above studies. in the pressure process the selectivity of transformation to MEch in relation to Mac consumed increases from 25 to 59 mol.% with the reaction time increasing from 0.5 to 4 h (Fig. 10), and remains con-stant with the further increase in the duration of the reaction. as the main by-product is formed 1-chloro-2-methyl-2,3-propandiol accompanied by smaller amounts of 1-chloro-2-methoxy-2-methyl-2-propanol and 3-chloro-2-methoxy-2-methyl-1-pro-panol. simultaneously, with increasing function SMEch/Mac the conversion of Mac increases. the selectivity of transformation to organic compounds in relation to h2o2 consumed in the reaction time studied is at a level of 20–25 mol.%. The conversion of h2o2 increases with increasing reaction time to 43 mol.% after 5 h.

0

10

20

30

40

50

60

70

80

90

100

0 5 1 1 5 2 2 5 3 3 5 4 4 5 5

time (h)

S MEC

H/M

AC, S

org/

H2O

2, CM

AC, C

H2O

2 (mol

.%)

fig. 10. Effect of the reaction time on Mac epoxidation under the autogenic pressure: selectivity of transformation to MECH in relation to MAC consumed (♦), selectivity of transformation to organic compounds in relation to h2o2 consumed (■), conversion of MAC (▲), conversion of H2o2 (×)

With increasing duration of the reaction under the atmospheric pressure the se-lectivity of transformation to MEch in relation to Mac consumed decreases from 100 mol.% after 0.5–2 h to about 16 mol.% after 3–5 h. This decrease is related to the consumption of MEch in the reaction of hydrogenation to 1-chloro-2-methyl-2,3-propandiol. the selectivity of transformation to organic compounds in relation to h2o2 consumed increases from 0.9 to 17 mol.% with the reaction duration increasing in the range studied. this implies an increase in the amount of the organic compounds related to the consumption of hydrogen peroxide. the conversion of Mac reaches 93 mol.% after 2 h and then remains constant. As follows from the above, the optimum reaction time under the atmospheric pressure is 2 h, while under the autogenic pres-sure the optimum reaction time is 4 h.

838

0

10

20

30

40

50

60

70

80

90

100

0 5 1 1 5 2 2 5 3 3 5 4 4 5 5time (h)

S MEC

H/M

AC, S

org/

H2O

2, CM

AC, C

H2O

2 (mol

.%)

fig. 11. Effect of the reaction time on Mac epoxidation under the atmospheric pressure: the selectivity of transformation to MECH in relation to MAC consumed (♦), selectivity of transformation to organic compounds in relation to h2o2 consumed (■), conversion of MAC (▲), conversion of H2o2 (×)

conclusions

The studies of the MAC epoxidation by a 30% H2o2 in the presence of ti-McM-48 catalyst carried out by the autogenic pressure method and atmospheric pressure method showed that the optimum parameters of the process determined for the two methods are similar. table 1 presents the optimum parameters and the corresponding functions describing the process.

table 1. optimum parameters and functions of Mac epoxidation Method under autogenic

pressure under atmospheric

pressure optimum parameters

temperature molar ratio of Mac/h2o2 methanol concentration (wt.%) Ti-MCM-48 concentration (wt.%) reaction time (h)

601:140

2 4

401:140

5 2

S MEch/Mac (mol. %) 60 100

CMac (mol. %) 11 2

Ch2o2 (mol. %) 43 93

Sorg./ h2o2 (mol. %) 26 2

in the two methods (under the autogenic and atmospheric pressure) the optimum molar ratio of Mac/h2o2 is 1:1 and the optimum methanol (solvent) concentration is

839

40 wt.%. In the atmospheric pressure method the optimum temperature of the process is 40oc, while in the autogenic pressure method the optimum temperature is 60oc. the pressure method needs the lower catalyst concentration 2 wt.%, and the longer reaction time 4 h. in the pressure method the selectivity of transformation to MEch in relation to Mac consumed is lower than that obtained under the atmospheric pressure. in the absolute values the yield of MEch, calculated as W= SMEch/Mac × CMac is higher in the autogenic pressure method. similarly, the yield of the organic compounds in relation to the amount of h2o2 introduced into the reaction is also higher. the lower conver-sion of h2o2 implies the necessity of its decomposition before isolation of MEch by distillation, for the safety reasons. High efficiency of the decomposition is achieved in the presence of metallic silver, or the compounds of Mn2+, cu+ deposited on active carbon or on ceramic adsorbents. the lower values of the functions conversion and selectivity are a consequence of the presence of the cl atom in the Mac molecule in the vicinity of the double bond. the 5-member active compounds forming at the ti4+ centres are less stable than in case when cl atom is replaced for example by the –OH group, as the active hydrogen is able to form a hydrogen bond with the oxygen atoms from hydrogen peroxide.

rEFErEncEs 1. G. GriGoroPoulou, j. h. clark, j. a. ElinGs: recent Developments on the Epoxidation

of alkenes using hydrogen Peroxide as an oxidant. Green chemist., 5, 1 (2003). 2. M. taraMasso, G. PErEGo, B. notari: Preparation of Porous crystalline synthetic Material

Comprised of Silicon and Titanium Oxides. uS Patent 4410501 (1983). 3. a. tuEl, y. BEn taarit: synthesis of titanium silicalite-1 using hexapropyl-1,6-hexanediam-

monium ions as templating agent. zeolites, 14, 594 (1994). 4. c. B. khouW, M. E. Davis: catalytic activity of titanium silicates synthesized in the Presence

of alkali Metal and alkaline Earth ions. j. org., 151, 77 (1995). 5. t. tatsuMi, M. yako, M. nakaMura, y. yuhara, h. toMinaGa: Effect of alkene struc-

ture on selectivity in the oxidation of unsaturated alcohols with titanium silicalite-1 catalyst. j. Mol. catal., 78, l41 (1993).

6. Wu. PERG, T. TAkASHI: A Novel Titanosilicate with MWW Structure III. Highly Efficient and selective Production of Glycidol through Epoxidation of allyl alcohol with h2o2. j. catal., 214, 317 (2003).

7. M. A. CAMBLOR, A. MIFSuD, J. PEREZ-PARIENTE: Influence of the Synthesis Conditions on the crystallization of zeolite Beta. zeolites, 11, 792 (1991).

8. k. A. kOYANO, T. TATSuMI: Mechanochemical Collapse of M41S Mesoporous Molecular Sieves through hydrolysis of siloxane Bonds. chem. lett., 12, 659 (1997).

9. k. W. GALLIS, C. C. LANDRY: Synthesis of MCM-48 by a Phase Transformation Process. Chem. Mater., 9, 2035 (1997).

10. u. ciEsla, F. schÜth: review ordered Mesoporous Materials. Microp. Mesop. Mater., 27, 131 (1999).

11. G. oyEa, j. sjoBloMB, M. stockErc: synthesis characterization and Potential applications of new Materials in the Mesoporous range. adv. coll. inter. sci., 89–90, 439 (2001).

12. Z. FLORIANCZYk, S. PECZEk: Chemia olimerów. Oficyna Wydawnictwa Politechniki Warsza-wskiej, (1997)(in Polish).

840

13. L. E. CHuMBLEY: Polycarbonates Incorporating 2-methylepichlorohydrin. uSA Patent 4168368 (1979).

14. k. koErtE: Polyetheresters, their Production and use. usa Patent 5112940 (1992).15. k. M. PATEL, T. M. STEVENSON: Herbicidal Oxazine Ethers. uSA Patent 5510318 (1996).16. k. SCHuMACHER, M. GRÜN, k. k. uNGER: Novel Synthesis of Spherical MCM-48. Microp.

Mesop. Mater., 27, 201 (1999).17. W. F. BRILL: The Origin of Epoxides in the Liquid Phase Oxidation of Olefins with Molecular

oxygen. j. am. chem. soc., 85, 141 (1963).18. M. RZEPkOWSkA, A. WROBLEWSkA, E. MILCHERT: Epoksydacja alkoholu metallilowego

i chlorku metallilu nadtlenkiem wodoru na katalizatorach tytanowo-silikalitowych. Przem. chem., 83, 197 (2004) (in Polish).

19. W. aDaM, a. corMa, t. j. rEDDy, M. rEnz: Diastereoselective catalytic Epoxidation of chiral allylic alcohols by ts-1 and ti-BEta: Evidence for a hydrogen Bonded, Peroxy-type loaded complex as oxidizing species. j. org. chem., 62, 3631 (1997).

Received 11 January 2008 Revised 18 February 2008

841

Oxidation Communications 31, No 4, 841–852 (2008)

* For correspondence.

Kinetics of oxidAtion of Heterocyclic compoUnds by qUinoliniUm dicHromAte

h. suantE, G. s. chauBEy, M. k. Mahanti*

Department of Chemistry, North-Eastern Hill University, 793 022 Shillong, India E-mail: mkmahanti@yahoo.com

aBstract

Quinolinium dichromate in sulphuric acid oxidises heterocyclic aldehydes (to the corresponding acids) and heterocyclic carboxylic acids (to the corresponding hy-droxy-substituted acids) in 50% (v/v) acetic acid–water medium. The kinetic results supported a mechanistic pathway proceeding via a rate-determining decomposition of the chromate ester.

Keywords: kinetics, oxidation, heterocyclic compounds, quinolinium dichromate.

aiMs anD BackGrounD

in the oxidation of heteroaldehydes and heteroacids, there exists the distinct possibility of the reaction occurring either at the heteroatom or at the carbonyl function. With a view to establish the site of reaction in the oxidation of heterocyclic compounds, we have focused attention on the oxidation of heteroaldehydes (2-furaldehyde, 2-pyr-rolecarbaldehyde, 2-thiophenecarbaldehyde, pyridine-2-aldehyde and pyridine-3-al-dehyde) and heteroacids (pyridine-2-carboxylic acid, pyridine-3-carboxylic acid and pyridine-4-carboxylic acid) by quinolinium dichromate (QDc, (c9h7nh+)2cr2o7

2–), in acid medium, in 50% acetic acid−water (v/v), under nitrogen. This investigation forms part of our sustained efforts to use QDc for a variety of organic transformations1.

ExPEriMEntal

Materials, methods and stoichiometry. 2-Furaldehyde, 2-thiophenecarbaldehyde, pyridine-2-aldehyde and pyridine-3-aldehyde (Aldrich) were purified by distillation under reduced pressure. Pyridine-2-carboxylic acid, pyridine-3-carboxylic acid, py-ridine-4-carboxylic acid (spectrochem) and 2-pyrrolecarbaldehyde (aldrich) were recrystallised before use. Quinolinium dichromate (QDc, (c9h7n+h)2cr2o7

2–), was prepared by the reported method2; the infrared spectrum (kBr) exhibited bands at 930, 875, 765 and 730 cm–1, characteristic of the dichromate ion. acetic acid (ar grade,

842

S.d. fine-chem. Ltd.) was distilled before use, and the fraction distilling at 116oc was used. sulphuric acid (E. Merck) was used after checking its physical constants. ir spectra were recorded on a FT-IR (DA-8, Bomem) spectrophotometer, and NMR spectra – on a FT-NMR (300 MHz, Bruker) spectrometer.

Pseudo-first order conditions were used for all the kinetic runs ([substrate] >> [QDC]). All the reactions were performed at constant temperature (± 0.1 k), under nitrogen, and followed by monitoring the absorption band at 440 nm spectrophoto-metrically (Systronics, Model-108), as described earlier3. rate constants were evalu-ated from the linear ( r > 0.996) plots of lg [QDc] against time. the values reported were the mean of two or more runs (reproducibility ± 3%). The reactions were carried out in aqueous medium, and water–acetic acid mixtures were used for studying the effect of dielectric constants on the rates of the reactions.

Stoichiometric ratios, ∆[QDC]/∆ [substrate], in the range of 0.66–0.69 (for het-eroaldehydes) and 1.02–1.05 (for heteroacids) were obtained1, which conformed to the following overall equations: 3c5h4o2 + 2cr(vi) + 3h2o → 3c5h4o3 + 2cr(iii) + 6 h+ (1)

(2-furaldehyde)

c6h5no2 + cr(vi) + h2o → c6h5no3 + cr(iv) + 2h+ (2) (pyridinecarboxylic acid)

Product analysis. Doubly distilled water (30 ml) was taken, cooled in ice, and con-centrated h2so4 (7.9 g, 0.08 mol) was added slowly with constant cooling. After the acid solution had cooled to room temperature, QDc (9.52 g, 0.02 mol) was added and the mixture was warmed to 323 k for complete dissolution of QDc. to this mixture, 0.015 mol of the substrate (1.45 g of 2-furaldehyde, 1.43 g of 2-pyrrolecarbaldehyde, 1.69 g of 2-thiophenecarbaldehyde, 1.61 g each of pyridine-2-aldehyde and pyridine-3-aldehyde, and 0.18 g each of the pyridinecarboxylic acids), taken in 25 ml of 50% acetic acid-water solution, was added. the reaction mixture was stirred at 323 k for 48 h under nitrogen. The organic layer was extracted three times with diethyl ether (25 ml each time), and the combined organic extracts were washed with water and dried over anhydrous na2so4. the oxidised products were obtained after complete removal of ether (yields ≈ 85–90%), subjected to IR (kBr) and 1h nMr (cDcl3 /1 drop DMso-d6) analyses, and characterised as follows:

(i) 2-furancarboxylic acid (from 2-furaldehyde): ν = 3000, 2860 (br, s, –OH), 2583, 1690 (s, C=O), 1470, 1305, 1245, 1020, 930, 760 cm–1; δ 10.5 (s, 1H, COOH), 7.9 (d, 1h, 5-h), 7.3 (d, 1h, 3-h), 6.7 (t,1h, 4-h);

(ii) 2-pyrrolecarboxylic acid (from 2-pyrrolecarbaldehyde): ν = 3000, 2850 (br, s, –OH), 1692 (s, C=O), 1545, 1355, 1105, 925, 750 cm–1; δ 10.2 (s, 1H, COOH ), 6.9 (d, 1h, 5-h), 6.7 (d, 1h, 3-h), 6.2 (t, 1h, 4-h);

(iii) 2-thiophenecarboxylic acid (from 2-thiophenecarbaldehyde): ν = 3090, 2850 (br, s, –OH), 2621, 1690 (s, C=O), 1530, 1350, 1100, 910, 750 cm–1; δ 11.4 (s, 1H, COOH), 7.8 (d, 1H, 5-H), 7.5 (d, 1H, 3-H), 7.1 (t, 1H, 4-H);

843

(iv) pyridine-2-carboxylic acid (from pyridine-2-aldehyde): ν = 3030 (br, s, –OH), 2500, 1720 (s, C=O), 1590, 1305, 1255, 1040, 830, 680 cm–1; δ 8.9 (s, 1H, COOH), 8.3 (d, 1H, 6-H), 8.1 (d, 1H, 3-H), 7.7 (m, 2H, 4,5-H);

(v) pyridine-3-carboxylic acid (from pyridine-3-aldehyde): ν = 3030 (br, s, –OH), 2445, 1720 (s, C=O), 1590, 1300, 1195, 830, 680 cm–1; δ 9.1 (s, 1H, COOH ), 8.8 (s, 1H, 2-H), 8.2 (d, 1H, 6-H), 7.5 (m, 2H, 4,5-H);

(vi) 3-hydroxypyridine-2-carboxylic acid (from pyridine-2-carboxylic acid): ν = 3080 (b, s, –OH), 2500 (O–H, str.), 1710 (s, C=O), 1610, 1410, 1389 (O–H bend.), 1250 (C–O, str.), 950 (O–H, str.), 720 cm–1;

(vii) 2-hydroxypyridine-3-carboxylic acid (from pyridine-3-carboxylic acid): ν = 3060 (b, s, –OH), 2510 (O–H, str.), 1690 (s, C=O), 1610, 1420, 1370 (O–H bend.), 1248 (C–O, str.), 946 (O–H, str.), 745 cm–1;

(viii) 3-hydroxypyridine-4-carboxylic acid (from pyridine-4-carboxylic acid): ν = 3075 (b, s, –OH), 2505 (O–H str.), 1700 (s, C=O), 1610, 1410, 1385 (O–H bend.), 1245 (C–O, str.), 950 (O–H, str.), 740 cm–1.

rEsults anD Discussion

Kinetic results. The observed pseudo-first order rate constants for the oxidation of heterocyclic compounds did not alter appreciably with changing QDc concentrations (10-fold range), at constant substrate concentration (large excess); this indicated a first order dependence on QDc (tables 1 and 2). in the range of substrate concentrations (10-fold), at constant acidity, the order of the reaction with respect to [substrate] was unity (Tables 1 and 2). In the range of acid concentrations used (0.5–1.5 mol dm–3 for the oxidation of heteroaldehydes, and 3.0 – 5.0 mol dm–3 for the oxidation of heteroac-ids), the observed pseudo-first order rate constants showed a first order dependence on [acid]. it could be suggested that in the range of acid concentrations used, the oxidant QDc was converted to the protonated cr(vi) species, wherein the dichromate ion (or its protonated form) was the predominant species1. Earlier reports have established the involvement of a protonated cr(vi) species in chromic acid oxidation reactions4.

the data in table 2 showed that an increase in [h+] favoured a higher rate of oxida-tion. such an effect would not support a decarboxylation process. this was similar to the reported behaviour of other aromatic acids, where the unionised carboxyl group underwent decarboxylation with great difficulty or not at all5–7.

in the experimental range of temperatures used, the dissociation constants of the heteroacids were reported to be as follows8(a): pyridine-2-carboxylic acid = 3 × 10–6 (pk = 5.52); pyridine-3-carboxylic acid = 1.4 × 10–5 (pk = 4.85); pyridine-4-carboxylic acid = 1.1 × 10–5 (pk = 4.96). these values were quite low, indicating that these substrates remained undissociated in the presence of the high concentrations of mineral acid used.

844

table 1. rate data for oxidation of heteroaldehydes at 313 k[substrate]

×102 (mol dm–3)

[QDc] × 103

(mol dm–3)[h2so4]

(mol dm–3)

k1 × 104 ( s–1)

2-furaldehyde 2-pyrrolecarbal-dehyde

2-thiophenecar-baldehyde

1.0 1.0 0.5 1.25 1.18 1.142.5 1.0 0.5 3.12 2.91 2.835.0 1.0 0.5 6.21 5.90 5.677.5 1.0 0.5 9.32 8.71 8.50

10.0 1.0 0.5 12.5 11.8 11.21.0 0.75 0.5 1.22 1.15 1.141.0 0.50 0.5 1.25 1.18 1.131.0 0.25 0.5 1.24 1.19 1.151.0 0.10 0.5 1.27 1.14 1.121.0 1.0 0.75 1.88 1.75 1.641.0 1.0 1.0 2.50 2.36 2.201.0 1.0 1.25 3.20 2.90 2.701.0 1.0 1.50 3.80 3.55 3.40

table 2. rate data for oxidation of heteroacids at 323 k [substrate]

× 102 ( mol dm–3)

[QDc] × 103 (mol dm–3)

[h2so4](mol dm–3)

k1 × 104 ( s–1)

1 2 31.0 1.0 3.0 1.72 1.48 1.645.0 1.0 3.0 8.68 7.47 8.20

10.0 1.0 3.0 17.50 15.00 16.501.0 0.75 3.0 1.76 1.49 1.601.0 0.50 3.0 1.67 1.46 1.631.0 0.10 3.0 1.65 1.47 1.621.0 1.0 3.5 2.01 1.71 1.931.0 1.0 4.0 2.34 1.96 2.20

1 – Pyridine-2-carboxylic acid; 2 – pyridine-3-carboxylic acid; 3 – pyridine-4-carboxylic acid.

the reactions were studied in solutions containing varying proportions of water and acetic acid. the dielectric constants (εr) of water–acetic acid mixtures were cal-culated from the εr of the pure solvents8(b). A decrease in εr resulted in an increase in the rate (tables 3 and 4). the magnitude of this effect established that for the equi-librium 2hcro4

– cr2o7 2– + h2o, the equilibrium was shifted to the right due to the reduced water activity, thus favouring the dichromate form over the chromate form. the absence of any salt effects indicated that these reactions were not of the ion–ion type. If the reactions were between two neutral molecules, then the plot of lg kex versus (εr – 1)/(2εr + 1) would have been linear; this was not found to be so. the plots of lg kex versus 1/εr were linear, with positive slopes, which indicated that the reactions were of the ion–dipole type9.

845

table 3. solvent effect for oxidation of heteroaldehydes at 313 k[substrate] = 1.0 × 10–2 mol dm–3; [QDc] = 1.0 × 10–3 mol dm–3; [h2so4] = 0.5 mol dm–3

h2o: acoh(%, v/v)

Dielectric constant

εr

k1 × 104 ( s–1)

2-furaldehyde 2-pyrrole- carbaldehyde

2-thiophene- carbaldehyde

50:50 39.79 1.25 1.18 1.1445:55 36.44 1.82 1.58 1.4140:60 33.09 2.51 2.09 1.7835:65 29.74 3.80 2.95 2.5130:70 26.39 6.30 4.27 3.55

table 4. solvent effect for oxidation of heteroacids at 323 k[substrate] = 0.01 mol dm–3 ; [QDc] = 0.001 mol dm–3; [h2so4] = 3.0 mol dm–3

h2o : acoh(%, v/v)

Dielectric constant

εr

k × 104 ( s–1)pyridine-2-car-

boxylic acidpyridine-3-car-

boxylic acidpyridine-4-car-

boxylic acid50:50 38.1 1.72 1.48 1.6445:55 34.9 2.56 2.23 2.4040:60 31.8 3.82 3.58 3.7535:65 28.6 5.56 5.30 5.4530:70 25.4 10.6 9.93 10.1

the effect of a change in the solvent (from h2o to D2o) was studied in order to ascertain the extent of the solvent isotope effect. the rates of oxidation were increased in D2o medium (tables 5 and 6), in agreement with earlier reported observations10. since D3o+ is about three times stronger than h3o+ (refs 10 and 11), the solvent isotope effect being greater than unity suggested a proton-catalysed reaction. this supported the protonation of the oxidant (QDC), an observation reflected in the acid dependence on the rates of the reactions (tables 1 and 2).

table 5. solvent isotope effect for oxidation of heteroaldehydes at 313 k[substrate] = 0.01 mol dm–3; [QDc] = 0.001 mol dm–3; [h2so4] = 0.5 mol dm–3

substratekh2o kD2o kD2o/kh2o

kex × 104 (s–1)2-Furaldehyde 1.25 2.03 1.622-Pyrrolecarbaldehyde 1.18 1.86 1.582-thiophenecarbaldehyde 1.14 1.75 1.54

846

table 6. solvent isotope effect for oxidation of heteroacids at 323 k[substrate] = 0.01 mol dm–3; [QDc] = 0.001 mol dm–3; [h2so4] = 3.0 mol dm–3

substratekh2o kD2o kD2o/kh2o

kex × 104 (s–1)Pyridine-2-carbocylic acid 1.72 2.80 1.63Pyridine-3-carboxylic acid 1.48 2.31 1.56Pyridine-4-carboxylic acid 1.64 2.61 1.59

The oxidation reactions were studied in the temperature range 303–333 k (Ta-bles 7 and 8). The negative values of ∆S* suggested the formation of a transition state containing both the oxidant and the reducing species.

table 7. temperature and activation parameters for oxidation of heteroaldehydes[substrate] = 1.0 × 10–2 mol dm–3; [QDc] = 1.0 × 10–3 mol dm–3; [h2so4] = 0.5 mol dm–3

Temp. (±0.1 k)k1 × 104 ( s–1)

2-furaldehyde 2-pyrrolecarbaldehyde 2-thiophenecarbal-dehyde

303 0.62 0.59 0.56308 0.94 0.89 0.84313 1.25 1.18 1.14318 1.78 1.74 1.71323 2.54 2.41 2.35

∆H* (kj mol–1) 51 ± 2.2 54 ± 2.3 56 ± 2.1∆S* (j k–1 mol–1) –152 ± 3.5 –145 ± 3.7 –142 ± 4.1

table 8. temperature and activation parameters for oxidation of heteroacids[substrate] = 1.0 × 10–2 mol dm–3; [QDc] = 1.0 × 10–3 mol dm–3; [h2so4] = 0.5 mol dm–3

Temp. (±0.1 k)k1 × 104 (s–1)

pyridine-2-carboxylic acid

pyridine-3-carboxylic acid

pyridine-4-carboxylic acid

313 0.87 0.75 0.82318 1.30 1.12 1.20323 1.72 1.48 1.64328 2.62 2.25 2.41333 3.50 2.95 3.26

∆H* (kj mol –1) 54 ± 2.2 55.7 ± 2.4 54.8 ± 2.3∆S* (j k–1 mol–1) –145 ± 4.1 –135 ± 3.6 –141 ± 3.5

the induced polymerisation of acrylonitrile or the reduction of mercury(ii) chloride12 was not observed, showing that an one-electron oxidation was unlikely. control experiments, performed in the absence of the substrate, did not show any appreciable change in [QDc].

847

MEchanisM

(i) Oxidation of heteroaldehydes. it has been shown that aldehyde oxidation reactions proceeded via the hydrated form13–18. table 9 lists the experimental rate constants (kex) for the oxidation of the aldehydes by QDc, and the aldehyde hydrate dehydration constants (Kdeh) for the following reaction: rch(oh)2 rcho + h2o. (3)

From kex and Kdeh , it was possible to compute two sets of rate constants for the oxidation of the aldehyde :

(a) the values of khy were obtained by assuming that only the hydrate form ap-pears in the following rate law: ν = khy[QDc][rch(oh)2]. (4)

and(b) the values of ka were calculated using the concentration of free aldehyde ac-

cording to the rate law as follows: ν = ka[QDc][rcho][h+]. (5)

table 9. QDc oxidation of heteroaldehydes at 313 k

aldehydes Kdehkex × 104

(s–1)khy

(M–2 s–1)ka

(M–2 s–1)2-Furaldehyde 1.29 1.25 2.38 ± 0.22 57.7 ± 1.322-Pyrrolecarbaldehyde 0.92 1.18 1.88 ± 0.18 54.5 ± 1.252-thiophenecarbaldehyde 0.84 1.14 1.59 ± 0.16 52.1 ± 1.20

the values of khy and ka are given in Table 9. using the σ+ values derived from a consideration of the electrophilic substitution for the five-member hetero systems19, a plot of lg khy against σ+ was linear (r = 0.990), with a slope of ρ = +2.0. On the other hand, the correlation of σ+ with ka gave a value of ρ = +0.64 (r = 0.992). the positive value of ρ could be interpreted as being due to a superimposed effect of the ring substituents on the hydration equilibrium, wherein it had been shown that all the alde-hydes oxidised by chromic acid were completely hydrated in aqueous solution20. the chromic acid oxidation of benzaldehyde had yielded a value of ρ = +1.06, which was in consonance with a pathway proceeding by way of an intermediate chromic acid ester of hydrated benzaldehyde21. in the range of acid concentrations used (present study), the hydrated forms of the substrates would remain as undissociated molecules (since [h+] would be much greater than the dehydration constants of the substrates). in the present investigation, the correlation with khy supported the mechanistic pathway for the oxidation reactions as proceeding via the hydrated form of the aldehydes. hence, a mechanism involving a direct hydrogen-transfer reaction between a free aldehyde and QDc was very unlikely.

848

in the case of heteroaldehydes, the heteroatoms are strong resonance donors in these five-member ring systems, an effect which completely overrides their induc-tive withdrawal. By treating these rings as transmitting systems, one could look for a correlation between the structure and the reactivity of these heterocyclic aldehydes. the observed order of reactivity was: 2-furaldehyde > 2-pyrrolecarbaldehyde > 2-thiophenecarbaldehyde (table 1), which was in conformity with the decreasing electronegativities of o, n and s atoms (electronegativities were: o = 3.50; n = 3.07; s = 2.44) (ref. 22). the inference was that electronegative substituents increased the oxidation rates by increasing the equilibrium concentrations of the intermediate chromate ester of the aldehyde hydrate. thus, the rate-accelerating effect of the elec-tronegative substituents could be interpreted in terms of greatly increased hydration (table 1).

the close resemblance in the structures of aldehyde hydrates and alcohols would favour similar pathways in their oxidation processes. the oxidation of alcohols had demonstrated the rate-determining decomposition of the protonated acid chromate ester23. in a similar manner, the oxidation of aldehydes could be visualised as proceed-ing via the formation of a similar intermediate (an ester of the aldehyde hydrate). if the chromium was coordinated through the –OH group (of the aldehyde hydrate) in the cyclic transition state24,25, this would facilitate the formation of the chromate ester and enhance the ease of its oxidation to the corresponding carboxylic acid. such a transition state envisaged the transfer of electrons towards the chromium, occurring by the formation of the carbon–hydrogen–oxygen bonds, as well as by the carbon–oxy-gen–chromium bonds.

the slow step of the reaction involved the participation of the aldehyde hydrate, protonated QDc, and two electrons in a cyclic system. removal of the hydrogen (on the carbon) was part of this step, as evidenced by the kinetic isotope effect for the oxidation of 2-furaldehyde-d1 (kh/kD = 5.8). Since the five-member heterocyclic ring system was a planar pentagon with sp2 hybridised carbon atoms, and possessed con-siderable aromatic character arising from delocalisation of the two-paired electrons, it would undergo a reaction via an electrocyclic mechanism involving six electrons; being a hückel-type system (4n + 2), this was an allowed process26.

the sequence of reactions for the oxidation of heteroaldehydes by QDc, in acid medium (scheme 1) showed that QDc was converted to the protonated dimetallic cr(vi) species (pq) (in the acid range used for the present investigation, the protonated QDc would have the cr(vi) existing mainly as cr2o7

2– (ref. 27)). the substrate (A) was converted to the hydrated form (Hy), which reacted with pq, giving the monochromate ester intermediate (e) and a cr(vi) monomer. the monochromate ester (e) underwent decomposition, in the rate-determining step, to give the product (the corresponding acid), along with the cr(iv) species. the rate law has been derived as follows: –d [QDC]/dt = k3 [E] = k3 [hy][PQ], (6)

where [PQ] = K1 [QDc][h+], and [hy] = K2 [a] [h2o].

849

s c h e m e 1

+product+Cr(V) substrate

K

K

O

O

OO

O

slow3k

fast

fast

Cr(III)

+ 2Cr(V)Cr(VI)Cr(IV)

Cr(IV)+OC

OH

(monochromate ester)(E)

Cr(VI)monomer

Cr +O

O

O

OQ

H

-+ + H

HQ

O

O

O

O+

CrH

OH

C

O

O

Cr

+O O

O

O

O

Cr Q

Q

2 H

H+

(PQ)

+-+HC

OH

OH

(Hy)

-

(Hy)(A)

+

(PQ)

OH

OH

C H

2

1

2OH+

(QDC)

O

HC

+H

H2

Q

QCr

O

O

O

OO+

Cr+2H+Q +

substituting the values of [PQ] and [hy] in equation (6) (taking the activity of water to be unity), we obtain –d [QDC]/dt = K1K2 k3 [a][QDc][h+] (7)

This rate expression (equation (7)) indicated that the reaction exhibited first order dependence on the concentrations of each reactant (substrate, oxidant and acid).

(ii) Oxidation of heteroacids. the oxidation of heteroacids by QDc was performed in solvent mixtures containing water−acetic acid (50:50, %, v/v). The data in Table 4 showed that with increasing proportions of acetic acid (decrease in εr of the medium), there was an increase in the value of the pseudo-first order rate constant, suggesting the absence of decarboxylation. This experimental observation was significant in view of the fact that a reduction in the pseudo-first order rate constant with decreasing εr had earlier been used as evidence to support the process of decarboxylation28. hence,

850

a change in εr of the solvent medium (table 4) affected the rate by decreasing the concentration of available protons required for decarboxylation.

in the case of heteroacids, the observed order of reactivity was as follows: pyridine-2-carboxylic acid > pyridine-4-carboxylic acid > pyridine-3-carboxylic acid (table 2).

The presence of the carboxylic group in the 2-position would influence the distri-bution of charge as a result of the steric configuration. This steric arrangement would not be possible either with the carboxylic group in the 3-position or the 4-position. steric considerations would favour the mechanistic pathway for the reaction as pro-ceeding via the formation of a cyclic intermediate involving an attack by the oxygen (of the oxidant) at the position adjacent to the carboxylic group.

s c h e m e 2

2

1slowk

(E)(S)

OQHOCrO

OQH OQH

OCrO OQH

OQH

OH

H

O

Cr(IV) + Cr(VI) 2Cr(V)

Cr(III)

OHCr

O

OQHHO+

Cr(IV)

+C

O

OH

2H O

+OQHCr

O

C

O

O

CrO

O+

OH+

+OQH

OCr

C

O

O+ Cr

+

O

O 2+

(PQ)

+-

HO

O

C

K-

+

(PQ)(QDC)

+2O

O

+

Cr+H++

Q

N N

N

N

pyridine-2-carboxylic acid

Cr(V) substrate product+ +

3-hydroxypyridine-2-carboxylic acid

Cr(VI) monomer

fast

fast

scheme 2 shows the sequence of reactions for the oxidation of heteroacids by QDC. The first step was the reaction between the protonated QDC (pq) and the substrate (s). one could envisage a mechanism involving the formation of the cyclic transition state, in which the carbon (of the substrate) was bonded to the oxygen (of the oxidant), to form the chromate ester (e). this would enable the transfer of electrons

851

towards the chromium, facilitating the formation of the chromium–oxygen–carbon bond. such an electrocyclic mechanism involved six electrons; being a hückel-type system (4n+2), this was an allowed process26. the rate of oxidation of the substrates did not increase rapidly by a decrease in the water concentration in water–acetic acid mixtures (table 4), suggesting that a molecule of water was not involved in a kineti-cally important stage of the oxidation reaction. this intermediate underwent reaction with a nucleophile (e.g. water), in the reaction mixture, to form the product, along with the formation of the cr(iv) species. the conversion of cr(iv) to cr(iii) was via the reaction cr(iv) + Cr(VI) → 2Cr(V). The standard potential for the Cr(VI)–Cr(V) couple was favourable (Eo = 0.62 v), and this reaction proceeded rapidly29. the Cr(V)–Cr(III) couple has a potential of 1.75 V, enabling the rapid conversion of Cr(V) to cr(iii), after reaction with the substrate29,30. the driving force for this reaction was the electrophilic nature of the oxygen (of the oxidant) and the electron-donor character of the charged oxygen atom of the substrate. the rate law was given by:

–d [QDC]/dt = k1 [E] = k1 [S] [PQ] (8)

where [PQ] = K [QDc] [h+]. hence,

–d [QDC]/dt = K k1 [s] [QDc] [h+]. (9)

This rate expression (equation (9)) showed that the reaction exhibited first order dependence on the concentrations of each reactant (substrate, oxidant and acid).

the cyclic mechanism proposed for the oxidation of heterocyclic acids indicated that the hydroxyl group was located ortho to the carboxyl group (in 2-hydroxypyri-dine-3-carboxylic acid). this ortho isomer was the only one isolated as a pure com-pound from the oxidation of pyridine-3-carboxylic acid. this was in contrast to the kolbe–Schmitt reaction31, wherein reaction conditions favoured entry of the carboxyl group at the para position.

conclusions

the experimental protocol in the present investigation has established that in the QDc oxidation of heteroaldehydes, there was an attack of the oxidant on the aldehydic function, leaving the heteroatom site intact. the QDc oxidation of heteroacids offers an unique advantage for the production of hydroxylated heterocarboxylic acids.

acknoWlEDGEMEnt

Financial support from the Department of science and technology, new Delhi, under the Fund for infrastructure in science and technology (Fist) Programme, is grate-fully acknowledged.

rEFErEncEs 1. G. s. chauBEy, s. Das, M. k. Mahanti: croatica chem. acta, 76, 287 (2003). 2. k. BalasuBraManian, v. PrathiBa: indian j. chem., 25b, 326 (1986). 3. B. kuotsu, E. tiEWsoh, a. DEBroy, M. k. Mahanti: j. org. chem., 61, 8875 (1996);

852

4. k. B. WiBErG: oxidation in organic chemistry. Part a.academic Press, new york, 1965, p. 69; k. k. BanErji: indian j. chem., 17A, 300 (1979).

5. k. R. LYNN, A. N. BOuRNS: Chem. Ind. London, 782 (1963). 6. j. l. lonGriDGE, F. a. lonG: j. amer. chem. soc., 90, 3092 (1968). 7. G. E. Dunn, s. k. Dayal: canadian j. chem., 48, 3349 (1970). 8. R. C. WEAST: Handbook of Chemistry and Physics. CRC Press, Ohio, 1978. 9. E. s. aMis: solvent Effects on reaction rates and Mechanisms. academic Press, new york, 1967,

p. 42.10. h. Maskill: the Physical Basis of organic chemistry. oxford university Press, oxford, 1993,

398–401. 11. s. n. isaacs: Physical Organic Chemistry. Longman, Harlow (uk), 1995, 389–390.12. j. s. littlEr, W. a. WatErs: j. chem. soc., 1299 (1959).13. l. c. GruEn, P. t. MctiGuE: j. chem. soc., 5217 (1963).14. r. P. BEll: adv. Phys. org. chem., 4, 1 (1964).15. s. kanDlikar, B. sEthuraM, t. n. rao: indian j. chem., 17A, 264 (1979).16. A. L. JAIN, k. k. BANERJI: J. Chem. Res. (M), 678 (1983).17. k. k. BanErji: tetrahedron, 43, 5949 (1987).18. V. k. SHARMA, k. SHARMA, N. MISHRA: Oxid. Commun., 16, 33 (1993).19. S. CLEMENTI, P. LINDA, G. MARINO: Tetrahedron Lett., (17), 1389 (1970). 20. j. rocEk: tetrahedron lett., 5, 1 (1959).21. G. t. E. GrahaM, F. h. WEsthEiMEr: j. am. chem. soc., 80, 3030 (1958).22. F. a. cotton, G. Wilkinson: Advanced Inorganic Chemistry. Wiley Eastern, New Delhi, 1985,

p. 115.23. j. rocEk, F. h. WEsthEiMEr, a. EschEnMosEr, l. MolDovanyi, j. schrEiBEr: helv.

chim. acta, 45, 2554 (1962).24. u. klanninG: acta chem. scand., 11, 1313 (1957); ibid., 12, 576 (1958).25. c. G. sWain, r. F. W. BaDEr, r. M. EstEnE, r. n. GriFFin: j. am. chem. soc., 83, 1951

(1961).26. j. s. littlEr: tetrahedron, 27, 81 (1971).27. M. crEslak-Golonka: coord. chem. rev., 109, 223 (1991).28. W. W. kAEDING: j. org. chem., 26, 3144 (1961); ibid., 29, 2556 (1964).29. F. h. WEsthEiMEr: chem. rev., 49, 419 (1949).30. j. F. PErEz-BEnito, c. arias, D. laMrhari: chem. commun., 472 (1992).31. a. s. linDsay, h. jEskEy: chem. rev., 57, 583 (1957).

Received 29 November 2007 Revised 10 December 2007

853

Oxidation Communications 31, No 4, 853–859 (2008)

* For correspondence.

pd(ii) cAtAlysis in oxidAtion of d-glUcose by cHlorAmine-t in Acidic mediUm. A Kinetic stUdy

sh. srivastava*, P. sinGh

Chemical Laboratories, Feroze Gandhi College, 229 001 Raebareli, (U.P.) India E-mail: she_ila72@yahoo.com

aBstract

kinetic investigations on Pd(ii)-catalysed oxidation of D-glucose by acidic solution of chloramine-t in the presence of mercuric acetate, as a scavenger have been made in the temperature range of 30–45°C. The rate shows first order kinetics in case of chloramine-t and order of reaction is zero and one with respect to substrate and Pd(II), respectively. Increase in [Cl‾] showed positive effect, while [H+] showed zero effect.

negligible effect of mercuric acetate and ionic strength of the medium was ob-served. a transient complex formed between Pdcl2 and chloramine-t. Pdcl2, being the reactive species of palladium(ii) chloride, disproportionates in a slow and rate- determining step. various activation parameters have been calculated. a suitable mechanism in agreement with observed kinetics has been proposed.

Keywords: Pd(ii) chloride, chloramine-t, D-glucose.

aiMs anD BackGrounD

the kinetics and mechanism of Pd(ii)-catalysed oxidation of some compounds by chloramine-t in perchloric acid medium has been reported1–3..

chloramine-t has been used as an oxidant in oxidation of some compounds such as amino alcohols4 and α-amino acids5, etc. the catalytic as well as inhibition action of Pd(ii) in various redox reactions has been reported over the past decade, there has been a considerable interest on the speciation of aqueous Pdcl2 solutions and complexes6–8 of Pd(ii) with cl– ions. the use of Pd(ii) chloride as a non-toxic and homogeneous catalyst has been reported by several workers9–12. scant work has been done for Pd (ii)-catalysed oxidation by chloramine-t which prompted us to un-dertake the kinetic study of Pd(ii)-catalysed oxidation of D-glucose by chloramine-t in acidic medium.

854

MathErials anD MEthoDs

Materials and methods. aqueous solution of D-glucose (E. Merck), chloramine-t (cDh grade), and mercuric acetate (E. Merck) were prepared by dissolving the weighed amount of sample in triply distilled water. Perchloric acid (60%) of E. Merck grade was used as a source of hydrogen ions. Palladous(ii) chloride (johnson Matthey) was prepared by dissolving the sample in hydrochloric acid of known strength. all other reagents of analytical grade were available. sodium perchlorate (E. Merck) was used to maintain the ionic strength of the medium. the reaction stills were blackened from outside to prevent photochemical effects.

Kinetics. a thermostated water bath was used to maintain the desired temperature within ±0.1°C. Requisite volume of all reagents including substrate, were taken in reaction vessel and thermostated at 35°C for thermal equilibrium. A measured volume of chloramine-t solution, which was also maintained separately at the same tempera-ture, was rapidly poured into the reaction vessel.

the kinetics was followed by examining aliquot portion of reaction mixture for chloramine-t iodometrically using starch as an indicator, after suitable time inter-vals.

rEsults anD Discussion

reaction mixtures containing excess of chloramine-t over D-glucose in different ratios were allowed to equilibrate at 35°C for about 24 h. The estimation of unconsumed oxidant showed that 1 mol of oxidant was consumed per mol of D-glucose, according to the following stoichiometric equation:

rcho + h2O + RNHCl → RCOOH + RNH2 + hcl

where r= c5h11o5. Identification of the end product formed in the above reaction, i.e the correspond-

ing acid (gluconic acid), was carried out as follows: 5 ml of acid were neutralised with excess of ammonia in a boiling test tube. then the solution boiled to remove the excess of ammonia, cooled and few drops of neutral Fecl3 solution were added. A reddish brown-coloured precipitate was obtained, which confirms the presence of carboxylic group.

the kinetic results were collected at several initial concentrations of reactants (Table 1). First order rate constants, i.e. (–dc/dt), were calculated from the plots of unconsumed chloramine-T versus time. It was observed that the values of (–dc/dt) were doubled when the concentration of chloramine-t was made two times, showing thus first order dependence on chloramine-T. The kinetic results recorded at various [Pd(ii)], ionic strengths of the medium along with kinetic effects on successive ad-dition of mercuric acetate, potassium chloride and sodium perchlorate are given in table 2. First order dependence on [Pd(ii)] is evident from the close resemblance

855

between the slope values (2.77 × 10–2), of (-dc/dt) versus [Pd(ii)] (Fig. 1) and average of experimental k1 (-dc/dt/[Pd(ii)]) values (2.57 × 10–2 at 35oc, respectively). this can be also justified by least square method (Fig. 2).

table 1. Effect of variation of oxidant, substrate, catalyst and perchloric acid at 35oc[substrate] = 2.00 × 10–2 mol dm–3; [hclo4] = 1.00 ×10–3 mol dm–3; [Pd(ii)] = 2.25 × 10–6 mol dm–3; temperature 35oc

oxidant ×103

(mol dm–3)[sub] × 102

(mol dm–3)[Pd(ii)] × 106

(mol dm–3) [hclo4] ×103

(mol dm–3)(–dc/dt) ×107

(mol dm–3 s–1)0.801.001.251.672.505.001.001.001.001.001.001.001.001.001.001.001.001.001.001.001.001.001.001.00

2.002.002.002.002.002.000.330.440.550.661.002.002.002.002.002.002.002.002.002.002.002.002.002.00

2.252.252.252.252.252.252.252.252.252.252.252.251.122.253.374.505.726.742.252.252.252.252.252.25

1.001.001.001.001.001.001.001.001.001.001.001.001.001.001.001.001.001.000.801.001.251.672.505.00

1.782.663.003.334.007.882.752.402.752.002.602.661.672.664.165.336.828.042.672.662.572.332.852.00

0

1

2

3

4

5

6

7

8

9

0 2 4 6 8

[Pd (II)] × 106 (mol dm–3)

(–dc

/dt)

× 10

7(m

ol d

m–3

s–1)

G

fig. 1. Plot between [Pd(ii)] and (-dc/dt) for oxidation of D-glucose

856

table 2. Effect of variation of [kcl], mercury(ii) acetate and sodium perchlorate at 35oc[hg(oac)2] = 1.25 × 10–3 mol dm–3; [kcl] = 1.00 ×10–3 mol dm–3; temperature 35oc

[kcl] × 103

(mol dm–3)[naclo4]×103

(mol dm–3)[hg(oac)2] ×103

(mol dm–3)(-dc/dt) ×107

(mol dm–3 s–1)0.801.001.251.672.505.001.001.001.001.001.001.001.001.001.001.001.001.00

––––––

0.801.001.251.672.505.00

––––––

1.251.251.251.251.251.251.251.251.251.251.251.250.801.001.251.672.505.00

2.252.662.963.283.603.942.332.502.542.662.362.122.502.672.662.002.802.33

0

1

2

3

4

5

6

7

8

9

0 2 4 6 8

(a +

bx)

G

[Pd (II)] × 106 (mol dm–3)

fig. 2. Plot between [Pd(ii)] and (a + bx) for oxidation of D-glucose

negligible effect of variation of ionic strength of the medium, addition of mer-curic acetate and positive effect of chloride ions on reaction rate are obvious from the kinetic data in table 2. change in ionic strength has a negligible effect. kinetic results obtained on varying concentrations of hydrogen ions indicate negligible effect of hydrogen ion variation, which means that the rate of the reaction is not affected by increase or decrease of [h+] concentrations.

857

The rate measurements were taken at 30–45°C and specific rate constants were used to draw a plot of lg k versus 1/T, which was linear (Fig. 3). the value of energy of activation (ΔE*), the arrhenius factor (A), entropy of activation (ΔS*) and free energy of activation (ΔG*) were calculated from rate measurement at 30, 35, 40 and 45º , and these values are recorded in table 3.

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.031 31.5 32 32.5 33 33.5

4 +

lg k

1/T 104 (K–1)

fig. 3. Plot between lg k and 1/T for oxidation of D-glucose[chloramine-t]= 1.00 × 10–3 mol dm–3; [D-Glucose] = 2.00 × 10–2 mol dm–3; [hg(oac)2] = 1.25 × 10–3 mol dm–3; [kcl] = 1.00 × 10–3 mol dm–3; [hclo4] = 1.00 × 10–3 mol dm–3; [Pd(ii)] = 2.25 × 10–6 moldm–3.

table 3. rate constants and activation parameters for D-glucose[hclo4] = 1.00 × 10–3 mol dm–3; [chloramine-t] = 1.00 ×10–3 mol dm–3; [kcl] = 1.00 × 10–3 mol dm–3; [Pd(ii)] = 2.25 ×10–6 mol dm–3

temperature(oc)

(-dc/dt) ×107

(mol dm–3 s–1)30354045

arrhenius parameters at 35ocΔE* (kj mol–1)lg AΔS* (j k–1 mol–1)ΔG* (kj mol–1)ΔH* (kj mol–1)

1.862.663.745.35

53.04 9.42

–16.9074.1869.14

negligible effect of mercuric acetate excludes the possibility of its involvement either as a catalyst or as an oxidant because it does not help the reaction proceeding without chloramine-t. hence, the function of mercuric acetate is to act as a scaven-ger11 for any cl– ion formed in the reaction. it helps to eliminate the parallel oxidation

858

by cl2 which would have been formed as a result of the interaction between cl– and rnhcl ion.

Pd(ii) chloride has been reported to give a number of possible chloro species depending on ph of the solution. under the experimental ph range in the present investigation, [Pdcl2] has been proposed and confirmed as the reactive species13 dominant in the ph range 1.00 to 3.00.

in acidic solution of chloramine-t quick formation of rnhcl has been reported. First-order dependence on [chloramine-t] suggests that rnhcl is itself involved in slow step as an oxidant.

the kinetic results reported in tables 1, 2 and 3 and the above statements lead us to suggest the following reaction scheme which gives the details of various steps in the title reaction.

Rate law. cl– exists as the following equilibrium in acidic solution of palladium(ii) chloride:

k1

Pdcl+ + cl– Pdcl2 (1) [c1] k–1 [c2]

Positive effect with respect to cl– in the present investigation suggests that the equilibrium would shift to the right. therefore, [Pdcl2] is the active species of palladium(ii) chloride in acidic media.

k2

[Pdcl2] + RNHCl ––––––→ [RNHCl-----PdCl2] (2) [c2 ] slow step [c3]

[rnhcl…Pdcl2] + rcho + h2O → PdCl2 + rcooh + rnh2 +hcl (3) [c3]

now considering the above steps and applying the steady-state treatment with reasonable approximation, the rate law may be written in term as: rate = – d[RNHCl]/dt = k2 [c2] [rnhcl] (4)

[Pd(ii)]t = [c1] + [c2] (5)

– d[C1]/dt = k–1[c2] – k1 [c1] [cl–] (6)

[c1] = k–1[c2]/ k1[cl–]. (7)

By putting the value of [c1] in equation (5) we obtain the following:[Pd(ii)]t = k–1[c2]/k1[cl–] + [c2] (K1 = k1/k–1)

[Pd(ii)]t = [c2]/ K1[cl–] + [c2] (8)

[Pd(ii)]t =

[c2] + K1[cl–] [c2] (9)

K1 [cl–]

859

[c2] = [Pd(ii)]t K1 [cl–]

1+ K1[cl–]

By putting the value of [c2] in equation (4) we obtain the following:

rate = k2 K1 [Pd(ii)]t [cl–] [rnhcl] .

1 + K1 [cl–]

conclusions

the experimental results as shown reveal that the reaction rate doubles when the concentration of catalyst is doubled. the rate law equation is in conformity with all kinetic observations and proposed mechanistic steps are supported by the negligible effect of ionic strength. the high positive values of free energy of activation (∆G*) indicates highly solvated transition state, while fairly high negative values of entropy of activation (∆S*) suggest the formation of an activation complex with reduction in the degree of freedom. From the present investigation, it is concluded that rnhcl is the reactive species of chloramine-t in acidic medium and the reactive species of Pd(ii) in acidic medium is [Pdcl2] under the experimental ph range.

rEFErEncEs 1. W. r. BErGrEn, hon G. MG. et al.: science, 176, 683 (1972). 2. c. k Mythily, k. s. ranGPPa: int. j. chem. kinet., 23 (1–2), 127 (2004). 3. PuttasWaMy, r. v. jaGaDEEsh: int. j. chem. kinet., 37 (1–4), 201 (2005). 4. a. v. Galavin, M. a. FaDotov: Mach. cat. study, Abs. 1, 172 (2002). 5. a. Gauri, h. Prakash et. al.: j. Phys. org. chem., 11, 31 (1998). 6. a. shukla, s. GuPta,.s. k. uPaDhyaya: int. j. chem. kinet., 23 (1–4), 279 (2004). 7. n. GrovEr, n. kaMBo et al.: int. j. chem., 41A, 2482 (2002). 8. o. haMED, P. M hEnry, c. thoMPson: j. org. chem., 66, 7745 (1999). 9. El-Qisairi, P. M hEnry: j. org. Met. chem., 603, 50 (2000).10. o. haMED, P. M hEnry: j. org. chem., 66, 180 (2001).11. MallaMMa, a. ranGasWaMy: inorg. Met. org. chem., 33, 1555 (2003).12. sh. a. chiMataDar, sanGaPPa et al.: tr. Mat. chem., 26, 662 (2001).13. M. h. konDarasaiah, s. ananDa, PuttasWaMy et al.: inorg. M. org & nano M. chem.,

33, 1145 (2003).

Received 16 October 2007 Revised 17 November 2007

.

860

Oxidation Communications 31, No 4, 860–866 (2008)

* For correspondence.

Kinetics And mecHAnism of tHe oxidAtion of dl-metHionine by morpHoliniUm cHlorocHromAte

n. sonia, P. P. raob, v. sharMaa*aDepartment of Chemistry, J. N.V. University, 342 005 Jodhpur, India bDepartment of Chemistry, PBS College, Vijaywada, A. P., India E-mail: anantsharma11@gmail.com

aBstract

the oxidation of methionine (Met) by morpholinium chlorochromate (Mcc) in dimethylsulphoxide (DMso) leads to the formation of the corresponding sulphox-ide. The reaction is of first order with respect to each MCC and Met. The reaction is catalysed by hydrogen ions. the hydrogen-ion dependence has the form: kobs = a + b[h+]. the oxidation of methionine was studied in 19 different organic solvents. the solvent effect was analysed by the kamlet’s and swain’s multiparametric equations. solvent effect indicated the importance of the cation-solvating power of the solvent. a suitable mechanism has also been postulated.

Keywords: halochromate, kinetics, mechanism, methionine, oxidation.

aiMs anD BackGrounD

Pyridinium and quinolinium halochromates have been used as mild and selective oxidising reagents in synthetic organic chemistry1–4. Morpholinium chlorochromate5 (Mcc) is also such a compound. We have been interested in the kinetics of oxidations by cr(vi) species and have already reported on the oxidation of methionine by other halochromates6–8, too. there seems to be no report on the oxidation of methionine by Mcc. Methionine (Met), a sulphur-containing essential amino acid, is reported to behave differently from other amino acids towards many oxidants9,10, due to elec-tron-rich sulphur center which is easily oxidisable. therefore, in continuation of our earlier work on the oxidation studies by halochromates, we report here the kinetics of oxidation of Dl-methionine by Mcc in dimethylsulphoxide (DMso) as a solvent. a suitable mechanism has also been proposed.

861

ExPEriMEntal

Materials. Mcc was prepared by the reported method5 and its purity checked by an iodometric method. Methionime (Merck) was used as supplied. Due to non-aqueous nature of the solvent, toluene-p-sulphonic acid (tsoh) was used as a source of hy-drogen ions. Other solvents were purified by the usual methods11.

Product analysis. Product analysis was carried out under kinetic conditions. the oxidation of Met by Mcc resulted in the formation of the corresponding sulphoxide, which was determined by the reported method12. the yield of sulphoxide was 95±4%. the oxidation state of chromium in completely reduced reaction mixtures i.e. after > 10 half-lives, as determined iodometrically, was +4.

Kinetic measurements. The pseudo-first order conditions were attained by maintaining a large excess (× 15 or more) of the Met over Mcc. the solvent was DMso, unless specified otherwise. The reactions were followed, at constant temperatures (±0.1 k), by monitoring the decrease in [Mcc] spectrophotometrically at 365 nm. no other reactant or product has any significant absorption at this wavelength. The pseudo-first order rate constant, kobs, was evaluated from the linear (r = 0.990–0.999) plots of lg [MCC] against time for up to 80% reaction. Duplicate kinetic runs showed that the rate constants were reproducible to within ±3%. All experiments, other than those for studying the effect of hydrogen ions, were carried out in the absence of tsoh.

rEsults

Stoichiometry. oxidation of Met by Mcc resulted in the formation of the correspond-ing sulphoxides. the overall reaction may, therefore, be represented as follows:

Me – S – R + CrO2clo−Mh+ → Me – S – R + CrOClO−Mh+ (1) o

Mcc undergoes two-electron change. this is in accordance with the earlier observations with halochromates6–8. it has already been shown that both PFc (ref. 13) and Pcc (ref. 14) act as two-electron oxidants and are reduced to chromium(iv) species by determining the oxidation state of chromium by magnetic susceptibility, Esr and ir studies.

Rate laws. The reactions were found to be first order with respect to MCC. In individual kinetic runs, plots of lg [Mcc] versus time were linear (r2 > 0.995). Further, it was found that the observed rate constant, kobs, does not depend on the initial concentration of Mcc. the reaction rate increases linearly with an increase in the concentration of methionine (table 1).

862

table 1. Rate constants for the oxidation of methionine by MCC at 298 k

[Mcc]*×103

(mol dm–3)[Met]

(mol dm–3)[tsoh]

(mol dm–3) kobs×104 (s–1)

1.00 0.10 0.00 1.171.00 0.20 0.00 3.301.00 0.40 0.00 6.571.00 0.60 0.00 9.811.00 0.80 0.00 13.31.00 1.00 0.00 16.22.00 0.20 0.00 3.854.00 0.20 0.00 3.546.00 0.20 0.00 2.808.00 0.20 0.00 3.381.00 1.00 0.10 18.11.00 1.00 0.20 21.01.00 1.00 0.40 26.21.00 1.00 0.60 31.31.00 1.00 0.80 37.11.00 1.00 1.00 42.51.00 0.40 0.00 6.93*

* contained 0.001 mol dm–3 acrylonitrile.

table 2. rate constants and activation parameters for the oxidation of Met by Mcc

k2×104 (dm3 mol−1 s−1) ∆H*

(kj mol−1)∆S*

(j k–1 mol−1)∆G*

(kj mol−1)288 k 298 k 308 k 318 k7.38 16.2 36.0 77.4 57.2.±0.7 −107±2 88.8±0.6

Induced polymerisation of acrylonitrile. the oxidation of Met, in an atmosphere of nitrogen, failed to induce the polymerisation of acrylonitrile. Further, the addition of acrylonitrile had no effect on the rate of oxidation (table 1). therefore, an one-electron oxidation, giving rise to free radicals, is unlikely.

Effect of acidity. the reaction was studied at different acidities by adding varying amount of tsoh to the reaction mixtures. the reaction is catalysed by hydrogen ions (table 1). the hydrogen-ion dependence has the form kobs = a + b[h+]. the values of a and b are 1.54±0.01×10−3 s–1 and 2.69±0.02×10−3 mol–1 dm3 s–1, respectively (r2 = 0.9997).

Discussion

the observed hydrogen-ion dependence suggests that the reaction follows two mecha-nistic pathways, one acid-independent and another acid-dependent. the acid-catalysis may well be attributed to a protonation of Mcc as (equation (2)) to yield a protonated cr(vi) species which is a stronger oxidant and electrophile.

863

Mhocro2cl + h+ M+hocr(oh)ocl (2)

Formation of a protonated cr(vi) species has earlier been postulated in the reac-tions of structurally similar BPcc (ref. 15) and QBc (ref. 16).

Solvent effect. the oxidation of Met was studied in 19 organic solvents. the choice of solvents was limited due to the solubility of Mcc and its reaction with primary and secondary alcohols. there was no reaction with the chosen solvents. the kinetics are similar in all the solvents. the values of k2 are recorded in table 3.

table 3. Effect of solvents on the oxidation of methionine by MCC at 298 ksolvents k2×105 (s−1) solvents k2 ×105 (s−1)chloroform 63.1 acetic acid 30.21,2-Dichloroethane 66.1 cyclohexane 1.95Dichloromethane 53.7 toluene 14.8DMso 162 acetophenone 58.9acetone 51.3 thF 25.7n,n-Dimethylformamide 79.4 t-butyl alcohol 27.5Butanone 38.9 1,4-dioxane 23.4nitrobenzene 69.1 1,2-dimethoxyethane 16.2Benzene 18.6 carbon disulphide 6.92Ethyl acetate 19.0

the rate constants for oxidation, k2, in 18 solvents (CS2 was not considered, as the complete range of the solvent parameters was not available) were correlated in terms of the linear solvation energy relationship of kamlet et al.17 as follows: lg k2 = A0 + pπ* + bβ + aα (3)

where π* is the solvent polarity, β – hydrogen bond acceptor basicities, α – the hydro-gen bond donor acidity, and A0 is the intercept term. it may be mentioned here that out of the 18 solvents, 12 have a value of zero for α. the results of correlation analyses in terms of equation (3), a biparametric equation involving π* and β, and separately with π* and β are given below. lg k2 = − 4.42 + 1.58 (±0.19) π* + 0.14 (±0.16) β + 0.12 (±0.15) α (4)

R2 = 0.8518; sd = 0.18; n = 18; ψ = 0.42

lg k2 = − 4.45 + 1.53 (±0.18) π* + 0.18 (±0.15) β (5)R2 = 0.8451; sd = 0.17; n = 18; ψ = 0.42

lg k2 = − 4.49 + 1.58 (±0.18) π* (6)r2 = 0.8290; sd = 0.18; n = 18; ψ = 0.43

lg k2 = − 2.66 + 0.46 (±0.34) β (7)r2 = 0.1040; sd = 0.41; n = 18; ψ = 0.97

where n is the number of data points and ψ – the Exner’s statistical parameter18.

864

the kamlet’s17 triparametric equation explains ca. 85% of the effect of solvent on the oxidation. however, by the Exner’s18 criterion the correlation is not even satisfac-tory (equation (7)). the major contribution is of solvent polarity. it alone accounted for ca. 84% of the data. Both β and α play relatively minor roles.

the data on the solvent effect were analysed in terms of the swain’s equation19 of cation- and anion-solvating concept of the solvents as well: lg k2 = aA + bB + C (8)

where A is the anion-solvating power of the solvent, B – the cation-solvating power and C – the intercept term. (A + B) is postulated to represent the solvent polarity. the rates in different solvents were analysed in terms of equation (8), separately with A and B and with (A + B). lg k2 = 1.18 (±0.04) A + 1.51 (±0.03) B − 3.82 (9)

R2 = 0.9951; sd = 0.03; n = 19; ψ = 0.07

lg k2 = 1.97 (±0.50) A − 2.79 (10)r2 = 0.1836; sd = 0.40; n = 19; ψ = 0.93

lg k2 = 1.42 (±0.21) B − 3.44 (11)r2 = 0.7248; sd = 0.23; n = 19; ψ = 0.54

lg k2 = 1.40 ± 0.05 (A + B) − 3.81 (12)r2 = 0.9810; sd = 0.06; n = 19; ψ = 0.14

the rates of oxidation of methionine in different solvents show an excellent cor-relation with the swain’s equation with both the cation- and anion-solvating powers playing significant roles, though the contribution of the cation-solvation is slightly more than that of the anion-solvation. the solvent polarity, represented by (A + B), also accounted for ca. 98% of the data. However, the correlations individually with A and B were poor. in view of the fact that solvent polarity is able to account for ca. 98% of the data, an attempt was made to correlate the rate with the relative permittivity of the solvent. however, a plot of lg k2 against the inverse of the relative permittivity is not linear (r2 = 0.4879; sd = 0.32; ψ = 0.74).

the observed solvent effect points to a transition state more polar than the reac-tant state. Further, the formation of a dipolar transition state, similar to those of sn2 reactions, is indicated by the major role of both anion- and cation-solvating powers. however, the solvent effect may also be explained assuming that the oxidant and the intermediate complex exist as ion-pair in non-polar solvent like cyclohexane and be considerably dissociated in more polar solvents.

MEchanisM

in view of the absence of any effect of radical scavenger, acrylonitrile, on the reac-tion rate, it is unlikely that an one-electron oxidation giving rise to free radicals, is operative in this oxidation reaction. the experimental results can be accounted for

865

in terms of electrophilic oxygen transfer from Mcc to the methionine-sulphur via an intermediate complex (equation (13)), similar mechanism has been suggested for oxidation of sulphides and iodide ions by periodate ion20 and for the oxidation of sulphides by hydrogen peroxide21 and PFc (ref. 22). the electrophilic attack on the sulphide-sulphur can be viewed as an sn2 reaction. an sn2 like transition state is supported by the observed solvent effect also.

the oxidation of methionine by Mcc may involve a cyclic intermediate as has been suggested in many reactions of cr(vi) (ref. 23). the cyclic transition state will be highly strained in view of the apical position of a lone pair of electrons or an alkyl group (equation (14)). the formation of a cyclic transition state entails a more exacting specificity of orientation and should result in much larger negative entropy of activation than that observed.

it is of interest to compare here the mode of oxidation of methionine by PFc (ref. 6), pyridinium chlorochromate (Pcc) (ref. 7), pyridinium bromochromate (PBC) (Ref. 8) and MCC. The oxidation by PFC, PBC and MCC presented a similar kinetic picture, i.e. the reactions are of first order with respect to the reductants. While in the oxidation by PCC, the Michaelis–Menten type kinetics was observed with respect to the reductants. it is possible that the values of the formation constants for the reductant-PFc/PBc/Mcc complexes are very low. this resulted in the observa-tion of second-order kinetics. no explanation of the difference is available presently. solvent effects and the dependence of the hydrogen ions are of similar nature in all these reactions, for which essentially similar mechanisms have been proposed.

Me S R + CrO2ClOMHMe

RS O Cr

OMH

O Cl

CrO2ClOMH+Me S R

O

Me S R + CrO2ClOMH

S

O Cr

O

Cl

O-MH+

Me

R

••

S

O Cr

O

Cl

O-MH+

Me

••OR

R

products

(13)

(14)

866

conclusions

the oxidation reaction is proposed to proceed in terms of electrophilic oxygen transfer from Mcc to the methionine-sulphur via an intermediate complex. the cyclic transi-tion state is highly strained in view of the apical position of a lone pair of electrons or an alkyl group. the electrophilic attack on the sulphide-sulphur can be viewed as an sn2 reaction like transition state, which is supported by observed solvent effect also.

acknoWlEDGEMEnts

Thanks are due to university Grants Commission, New Delhi for financial support in the form of Major research Project no F. 32−207/2006 (sr) dated 22.02.2007.

rEFErEncEs 1. E. j. corEy, W. j. suGGs: teerahedron lett., 2647 (1975). 2. F. S. GuZIEC, F. A. LuZIO: Synthesis, 691 (1980). 3. M. n. BhattacharjEE, M. k. chouDhuri, h. s. DasGuPta, n. roy, D. t. khathinG:

Synthesis, 588 (1982). 4. k. BalasuBraManian, v. PrathiBa: indian j. chem., 25b, 326 (1986). 5. h. n. shiEkh, M. sharMa, M. hussain, B. l. kalsotra: oxid. commun., 28, 887

(2005). 6. v. sharMa, P. k. sharMa, k. k. BanErji: j. chem. res. (s), 290 (1996). 7. v. sharMa, P. k. sharMa, k. k. BanErji: j. indian chem. soc., 74, 607 (1997). 8. V. SHARMA, P. k. SHARMA, k. k. BANERJI: Indian J. Chem., 36A, 418 (1997). 9. D. s. MahaDEvaPPa, s. ananDa, n. M. M. GouDa, k. s. ranGaPPa: j. chem. soc.,

Perkin Trans. 2, 39 (1985) and references cited therein.10. s. Mittal, v. sharMa, k. k. BanErji: J. Chem. Research (S), 264 (1986).11. D. D. PERRIN, L. ARMAREGO, D. R. PERRIN: Purification of Organic Compounds. Pergamon

Press, oxford, 1966. 12. j. MitchEll: organic analysis. vol. ii. interscience, new york, 1, 375, (1953).13. M. n. BattacharjEE, M. k. chouDhary, s. PurkEsyatha: tetrahedron, 43, 5389

(1987). 14. h. c. BroWn, c. GunDu rao, s. u. kulkarni: j. org. chem., 44, 2809 (1979).15. sh. v. yas, P. k. sharMa: oxid. commun., 24, 248 (2001).16. P. k. sharMa: int. j. chem. kinet., 38, 364 (2006).17. M. j. kaMlEt, j. l. M. aBBouD, M. h. aBrahaM, r.W. taFt: j. org. chem., 48, 2877 (1983)

and references cited therein.18. O. EXNER: Collect. Czech. Chem. Commun., 38, 411 (1973).19. c. G. sWain, M. s. sWain, a. l. PoWEl, s. alunni: j. am. chem. soc., 105, 502 (1983). 20. F. ruFF, a. kucsMan: J. Chem. Soc., Perkin Trans. 2, 683 (1985).21. G. MoDEna, l. Maioli: Gazz. chim. ital., 87, 1306 (1957).22. k. k. BanErji: tetrahedron, 44, 2969 (1988).23. y. W. chanG, F. h. WEsthEiMEr: j. am. chem. soc., 82, 1401 (1960); j. rocEk, F. h.

WEsthEiMEr: j. am. chem. soc., 84, 2241 (1962).

Received 29 November 2007 Revised 18 December 2007

867

Oxidation Communications 31, No 4, 867–873 (2008)

* For correspondence.

Kinetics And mecHAnism of oxidAtion of Bis(2,2′,6′,2″-terpyridine)iron(ii) By iodate and vAnAdiUm(v) in sUlpHUric Acid mediUm

r. n. rEDDya,c*, t. satyanarayanab, n. r. aniPinDic

aSchool of Chemistry, Andhra University, 530 003 Visakhapatnam, Andhra Pradesh, India bDepartment of Chemistry, Ideal College of Postgraduate Courses, 533 003 Kakinada, Andhra Pradesh, India cRational Labs Pvt. Ltd., Plot No 177, Phase-II, IDA Mallapur, 500 076 Hyderabad, Andhra Pradesh, India E-mail: drnageswar@yahoo.com

aBstract

The kinetics of oxidation reaction of bis(2,2′,6′,2″-terpyridine)iron(II) by iodate and vanadium(v) has been studied in sulphuric acid medium by spectrophotometric method. The reaction follows first order kinetics in substrate and oxidant. The oxida-tion process is catalysed by sulphuric acid and increase in ionic strength increases the reaction. the reaction obeys the rate law:

kK[h+][oxidant]t [Fe(terpy)22+]

rate = –––––––––––––––––––––––––– .1 + K [h+]

Keywords: kinetics, mechanism, oxidation, 2,2′,6′,2″-terpyridine, bis(2,2′,6′,2″-terpyridine)iron(ii), vanadium(v), iodate.

aiMs anD BackGrounD

the kinetic and mechanistic studies of iodate oxidations of inorganic and organic substrates were reported by several workers. Manikyamba et al.1 studied the oxidation of α- and β-napthols by iodate to the corresponding 1,2-napthoquinones in aqueous acetic acid in the presence of sulphuric acid. these workers reported that the order of the reaction depends on the concentrations of the substrate and oxidant and also catalysed by acid. they further reported2 on the oxidation of benzaldehydes and substituted benzaldehydes by acid iodate in aqueous methanol and observed that the reaction varies with the concentrations of substrate and oxidant. radhakrishnamurti and coworkers3 reported the kinetics of osmium(viii) and ruthenium(iii) catalysed

868

oxidations of styrene and stilbene by iodate in aqueous acetic acid and perchloric acid media. These authors observed that the reaction is zero order in iodate and first order each in substrate and catalyst. the kinetics of oxidation of aquoiron(ii) by iodate in dilute perchloric acid medium in the presence of allyl alcohol was reported by hig-ginson and Mccarthy4. ramana and appa rao5 studied the kinetics of oxidation of oxalic acid by iodate in aqueous sulphuric acid medium and observed that the rate of the reaction is first order each in substrate and oxidant. Cyfert and coworkers6 investigated the kinetics of oxidation of tris(1,10-phenanthroline)iron(ii) by iodate and periodate ions in neutral medium. these workers observed that the iodate ion is unreactive with ferro-ion in aqueous solution and is autocatalytic in the case of io4

–. the autocatalysis disappears in the presence of excess of io3

–. the kinetic and mechanistic studies of the oxidations by vanadium(v) were re-

viewed by Waters7, Waters and littler8, stewart9, littler10 and turney11. the kinetics of the reaction between vanadium(v) and iron(ii) was investigated by Daugherty and newton12. the kinetics and mechanism of the reactions between vanadium(v) with cyanobipyridyliron(ii) and of vanadium(v) with tris(bipyridine)iron(iii) were reported by Birk and Weaver13. subba rao et al.14–17 found that oxalic acid and EDta catalyse the oxidation of vanadium(v) reactions. nandibewoor and coworkers18 studied the oxidation of antimony(iii) and thallium(i)(ref. 19) by vanadium(v).

The kinetics of oxidation of bis(2,2′,6′,2″-terpyridine)iron(II) by iodate and vanadium(v) has not been reported by the research workers so far. the present work is an attempt to study the oxidation of (2,2′,6′,2″-terpyridine)iron(II) by iodate and vanadium(v) in sulphuric acid media.

ExPEriMEntal

All the chemicals are of A.R. grade. 2,2′,6′,2″-terpyridine was obtained by G.F.S. Chemical Co., Ohio (uSA). The preparation of bis(2,2′,6′,2″-terpyridine)iron(II) solu-tion was described20 as earlier. aqueous iodate solution (0.1 mol dm–3) was prepared from potassium iodate (E. Merck). it was standardised iodometrically21,22.

the sodium vanadate solution was prepared and standardised according to the method of Gopala rao and coworkers23. the method consists in mixing a calculated amount of ammonium vanadate (riedel Gr) with a slight excess of sodium carbon-ate (BDh analr) in 500 ml of water and boiling the solution till ammonia ceases to evolve. the mixture is cooled, made up to 1 l and standardised against standard ferrous ammonium sulphate solution using n-phenyl anthranilic acid as an indicator in 6 mol dm–3 sulphuric acid medium.

the absorption spectra of the reactants and products were recorded on a shi-madzu uv-260 uv-vis. recording spectrophotometer with a six-cell holder in which the temperature can be controlled to ± 0.1oc by the Peltier effect. a control Dynam-ics ph meter was used for ph measurements. all computations were made using a compaq Presario (with intel Pentium) personal computer using Microsoft Excel. a

869

Shimadzu TB-85 Thermo Bath was used to equilibrate the reaction solutions at the experimental temperature.

Experiments were carried out in the [h+] range of 0.1 to 0.5 mol dm–3. a de-tailed study of the oxidation of bis(terpyridine)iron(ii) by iodate and vanadium(v) was monitored at 552 nm (ε = 12 500 dm3 cm–1 mol–1), where it shows maximum absorbance. These reactions were performed under pseudo-first order conditions with [oxidant] >> [Fe(terpy)2

2+], and kinetic runs were followed up to the disappearance of the terpyridine complex to colourless at 552 nm. the rate constants were evaluated from the plots of lg(abs.) versus time. the kinetic runs were performed in duplicate and are reproducible within ± 5%.

Product analysis. known amounts of bis(2,2′,6′,2″-terpyridine)iron(II), oxidant(iodate/vanadium(v)), sulphuric acid and sodium bisulphate solutions were taken in a 50-ml volumetric flask. After allowing the reaction mixtures to stand for sufficient time so that the reaction goes to completion. the product solutions show maximum absorbance at 703 nm with ε = 704 dm3 cm–1 mol–1 with respect to iron concentration. this value agrees well with the reported values24 for Fe(terpy)2

3+ thus confirming that Fe(terpy)22+

is oxidised to the corresponding iron(iii) complex.

rEsults anD Discussion

The pseudo-first order rate constants kobs values are independent of substrate and the data exhibits first order kinetics with respect to substrate. The kobs values are linearly depend on the concentration of oxidant (iodate/vanadium(v)) (tables 1 and 2) and the rate follows first order with respect to the oxidant. The order of the reaction is fractional in [h+] (tables 3 and 4) and plots of 1/kobs versus 1/[h+] are linear with positive slopes and intercepts.

table 1. Effect of iodate on the rate[Fe(terpy)2

2+] = 4.0 × 10–5 mol dm–3, [h+] = 1.0 × 10–1 mol dm–3, μ = 0.5

[iodate] × 10(mol dm–3)

kobs × 102 (min–1)25 ± 0.1oc 30 ± 0.1oc 35 ± 0.1oc

2.0 1.46 2.12 3.26 4.0 2.91 4.24 6.52 6.0 4.37 6.36 9.78 8.0 5.82 8.48 13.0410.0 7.28 10.60 16.3012.0 8.73 12.72 19.56

870

table 2. Effect of vanadium(v) on the rate[Fe(terpy)2

2+] = 4.0 × 10–5 mol dm–3, [h+] = 2.0 × 10–1 mol dm–3, µ = 0.5

[vanadium(v)] × 104

(mol dm–3)kobs × 102 (min–1)

30 ± 0.1oc 35 ± 0.1oc 40 ± 0.1oc 2.0 2.98 4.51 6.57 4.0 5.81 8.92 12.96 6.0 8.69 13.51 19.52 8.0 11.62 17.58 26.3110.0 14.42 22.55 30.3312.0 17.54 26.46 39.21

table 3. Effect of hydrogen ion on the rate[Fe(terpy)2

2+] = 4.0 × 10–5 mol dm–3, [io3–] = 4.0 × 10–3 mol dm–3, µ = 0.5

[h+](mol dm–3)

kobs × 102 (min–1)25 ± 0.1oc 30 ± 0.1oc 35 ± 0.1oc

0.1 2.83 4.16 6.250.2 4.12 5.92 8.620.3 5.02 6.96 9.870.4 5.85 8.45 11.920.5 6.67 9.87 13.93

table 4. Effect of hydrogen ion on the rate [Fe(terpy)2

2+] = 4.0 × 10–5 mol dm–3, [v(v)] = 4.0 × 10–4 mol dm–3, µ = 0.5

[h+](mol dm–3)

kobs × 102 (min–1)30 ± 0.1oc 35 ± 0.1oc 40 ± 0.1oc

0.1 3.65 5.05 6.840.2 5.92 8.89 12.680.3 8.30 11.24 17.240.4 9.98 14.55 21.150.5 13.92 19.85 27.76

Mechanism of Fe(terpy)22+–iodate reaction. in the present experimental conditions,

iodate exists as io3– and hio3. hio3 is considered as the reactive species and is pro-

posed the following mechanism:

h+ + io3– hio3 (1)

hio3 reacts with Fe(terpy)22+ in the rate-limiting step giving Fe(terpy)2

3+ and hio3

–• anion radical.

Fe(terpy)22+ + hio3 →k Fe(terpy)2

3+ + hio–•3 (2)

the hio3–• anion radical formed reacts with another mol of substrate forming

Fe(terpy)23+ and io2

–•.

K

871

fast Fe(terpy)2

2+ + hio3–•H+

–––––→ Fe(terpy)2

3+ + io2–•

+ h2o (3)

io2–• anion radical further reacts with Fe(terpy)2

2+ to form i2 and Fe(terpy)23+ in

a series of fast steps as follows: rate = k [Fe(terpy)2

2+] [hio3]e (4)

[io3–]t = [io3

–]e + [hio3]e (5)

kK[h+][io3–]t[Fe(terpy)2

2+] rate = –––––––––––––––––––––––– . (6)

1 + K[h+]

The pseudo-first order rate constant is given by the following equations: kK[h+][io3

–]t kobs = –––––––––––– . (7)

1 + K[h+]

1 1 1/kobs = –––––––––––––––– + ––––––––––– . (8)

kK[h+][io3–]t 1 + K[h+]

the values of k and K evaluated at 25, 30 and 35oc are 0.0226, 0.0316, 0.0422 dm3 mol–1 min–1 and 1.01, 1.82, 2.32 dm3 mol–1 min–1, respectively. thermodynamic parameters are Ea = 86 kJ mol–1 and ∆S* = –69 J k–1 mol–1. the reported value5 of K is 6.369 dm3 mol–1 min–1 and our experimental value of K is 2.32 dm3 mol–1 min–1, i.e. these values are of the same order and, hence, it gives to the proposed mechanism.

Test for free radicals for Fe(terpy)22+–iodate reaction. the reaction mixture is kept in

the lower limb of a thumberg tube and the acrylonitrile is kept in the upper limb of the tube. the tube is closed and the air inside the tube is removed by using a vacuum pump. then both the solutions are mixed by shaking the tube. a dense polymer was found in the reaction mixture, which indicates that the free radical formation occurs in the reaction.

Mechanism of Fe(terpy)22+ – vanadium(V) reaction. the kinetics of the reaction be-

tween vanadium(v) and iron(ii) was investigated by Daugherty and newton12. they proposed the following rate law: –d[V(V)]/dt = k´[vo2

+][Fe2+] (9)

where k´ = a[h+]–1 + b + c[h+].in the present experimental conditions, vanadium(v) exists as vo2

+ and v(oh)32+.

a suitable mechanism is proposed by considering the v(oh)32+ as the reactive species

to explain the kinetic results. K

vo2+ + h3o+ v(oh)3

2+ (10)

872

Fe(terpy)22+ + v(oh)3

2+ →k Fe(terpy)2

3+ + v(iv) (11)

rate = k [v(oh)32+]e[Fe(terpy)2

2+] (12)

[v(v)]t = [vo2+]e + [v(oh)3

2+]e (13)

K [h+] [v(v)] t [v(oh)3

2+]e = –––––––––––––––– (14) 1 + K[h+]

kK[h+][v(v)]t[Fe(terpy)22+]

rate = –––––––––––––––––––––––– . (15)1 + K [h+]

The pseudo-first order rate constant is given by the following equations:kK [h+] [v(v)]t

kobs = –––––––––––––––– (16) 1 + K[h+]

1 1 1/kobs = –––––––––––––––– + ––––––––––– . (17)

kK[h+][v(v)]t 1 + K[h+]

according to equation (7), a plot of 1/kobs versus 1/[h+] is a straight line with positive slope and intercepts. the slope is equal to 1/kK[v(v)]t and the intercept – to 1/k[v(v)]t, respectively. the values of k and K at 30, 35 and 40oC are 0.0951, 0.1308, 0.1843 dm3 mol–1 min–1 and 0.93, 1.08 and 1.52 dm3 mol–1 min–1, respectively, and the Ea and ∆S* values are 71.0 kj mol–1 and –15.0 J k–1 mol–1. subba rao and cowork-ers19 reported the value of K = 1.98 for the oxidation of oxalic acid by vanadium(V). hence, our experimental value (K = 1.52) at 40oc agrees well with the reported value (K = 1.98).

in sulphuric acid concentration employed in these investigations, vanadium(v) exists as v(oh)3

2+ formed from the vo2+ + h3o+ v(oh)3

2+. this reacts with the substrate in the rate-determining step to Fe(terpy)2

3+ and vo2+. the hydrogen ion catalysis noticed in these reactions suggests that the protonated forms of oxyanions are the reactive oxidising species. the rate of the reaction reached limiting value and this can be ascribed to complete protonation of oxyanions at higher acidities and the protonated species are the reactive oxidising species in all these systems. the higher values of specific rate constants are noticed in iodate oxidation when compared with vanadium(v). the iodate ion has more resonance forms and the resonance contributes to the greater stabilisation of the oxyanion. the activation energies for iodate and vanadium(v) oxidation reactions are low, which suggests that the electron-transfer in these reactions occurs by outer-sphere mechanism.

873

rEFErEncEs1. P. ManikyaMBa, k. vijayalaxMi, E. v. sunDaraM: j. indian chem. soc., 58, 574

(1981). 2. P. ManikyaMBa, P. raGhunatha rao, E. v. sunDaraM: j. indian chem. soc., 20A, 574

(1981). 3. P. s. raDhakrishnaMurty, k. s. triPathy: indian j. chem., 25A, 762 (1986). 4. C. E. WILLIAM HIGGINSON, D. A. McCARTHY: J. Chem. Soc., Dalton Trans., 797 (1980). 5. P. v. raMana, r. v. aPPa rao: indian j. chem., 30A, 971 (1991). 6. M. cyFErt, B. latko, M. WaWrzEnczyk: int. j. chem. kin., 28, 103 (1996). 7. W. a. WatErs: in: Mechanisms of oxidation of organic compounds (Eds h. j. Emeleus, D.W. G.

style, r. P. Bell). Methuen and co. ltd., london, 1964. 8. W. A. WATERS, J. S. LITTLER: In: Oxidation in Organic Chemistry (Ed. k. B. Wiberg). Academic

Press inc. ltd. london, 1965. 9. r. stEWart: in: oxidation Mechanisms (Eds r. Breslow, M.carplus). W. a. Benzamin inc., new

york, 1964. 10. M. t. P. international review of science (Ed. W. a. Waters). Butterworths, london, 10, 1973,

p. 245. 11. t. a. turnEt: oxidation Mechanisms. chapter iii. Butterworths, london, 1965. 12. n. a. DauGhtErty, t. W. nEWton: j. Phys. chem., 67, 1090 (1963).13. j. P. Birk, s. v. WEavEr: inorg. chem., 11, 92 (1972). 14. P. v. suBBa rao, r. v. s. Murty, B. a. Murthy, k. s. Murthy, P. s. n. Murthy: indian

j. chem., 16A , 1056 (1978). 15. P. v. suBBa rao, r. v. s. Murty, P. s. n. Murthy: indian j. chem., 18A, 228 (1980). 16. P. v. suBBa rao, r. v. s. Murty, k. v. suBBaiah: react. kin. catal. lett., 10, 287 (1979). 17. P. v. suBBa rao, r. v. s. Murty, P. s. n. Murthy: j. inorg. nucl. chem., 40, 295 (1978).18. P. L. TIMMANAGOuDAR, G. A. HIREMATH, S. T. NANDIBEWOOR: J. Chem. Soc. Dalton

trans., 3623 (1995).19. P. l. tiMManaGouDar, G. a. hirEMath, s. t. nanDiBEWoor: indian j. chem., 35A,

416 (1995). 20. a. a. schilt: in: analytical applications of 1,10-phenanthroline and related compounds. intern.

series of Monographs in analytical chemistry. vol. 32. Pergmon Press, new york, 1969.21. j. M. BoBBit: adv. carbohydrate chem., 11, 1 (1956); j. r. DyEr: Methods Biochem. analysis,

3, 11 (1956).22. l. M. Frolova, v. M. chistyanrkov, v. a. ErMakov, a. G. rykov: radiokhimya, 18,

276 (1976) (in russian).23. G. GoPala rao, v. P. rao, B. v. s. r. Murthy: z. anal.chem., 147, 161 (1955).24. M. h. ForD-sMith, n. sutin: j. am. chem. soc., 83, 1830 (1961).

Received 8 August 2007 Revised 13 September 2007

874

Oxidation Communications 31, No 4, 874–878 (2008)

* For correspondence.

Kinetics of tHe oxidAtion of sodiUm tetrAHydroborAte by cobAlt(iii)

M. DasGuPta, M. k. Mahanti*

Department of Chemistry, North-Eastern Hill University, 793 022 Shillong, India E-mail: mkmahanti@yahoo.com

aBstract

the kinetics of the oxidation of sodium tetrahydroborate by cobalt(iii) in aqueous acidic medium at constant ionic strength has been studied. the rate of the reaction was proportional to the concentrations of each – cobalt(III), sodium tetrahydroborate and acid. The formation of the cobalt(II) species as the final product indicated an one-electron transfer process.

Keywords: kinetics, oxidation, sodium tetrahydroborate, cobalt(iii).

aiMs anD BackGrounD

Mechanistic studies on the reaction of sodium tetrahydroborate with organic com-pounds have been reported, with the reaction following second-order kinetics1. in continuation of our work on redox reactions between sodium tetrahydroborate (naBh4) and transition metal ions2, we report here the kinetics of the reaction between naBh4 and the hexamine cobalt(iii) chloride complex, co(nh3)6cl3 , in aqueous acidic me-dium, at constant ionic strength.

ExPEriMEntal

Materials, methods and stoichiometry. sodium tetrahydroborate (loba chemical co.) was kept under vacuum. Its purity was checked by IR analysis (FT-IR, DA-8, Bomem spectrophotometer). two sharp peaks were obtained at 2290 and 1120 cm–1, both be-ing characterised for naBh4. these peaks have been assigned as follows3: (i) 2290 cm–1 as (B–H)asym stretching, and (ii) 1120 cm–1 as H–B–H deformation. The (B–H)asym stretching mode was further split (2380 and 2220 cm–1). it has been suggested that the splitting was a consequence of the inability of the tetrahedral anion to rotate freely in the crystal lattice4. the hexamine cobalt(iii) chloride complex was prepared by the standard method5. hydrochloric acid was an E. Merck sample. sodium perchlorate (aldrich) was used to maintain a constant ionic strength.

875

solutions of naBh4 (in water) and co(nh3)6cl3 (in hcl) were separately pre-pared, thermostated for 1 h at 30oc, and mixed in equal volumes by syringing into the spectrophotometric cell. the progress of the reaction was followed spectropho-tometrically (Du 650, Beckman) by monitoring the disappearance of cobalt(iii) at 475 nm (Ref. 6). The pseudo-first order rate constants (k1) were evaluated from linear plots (r > 0.994) of lg [co3+] against time, and were reproducible within ± 3%. All reactions were performed under nitrogen.

Product analysis. sodium tetrahydroborate (0.25 mol dm–3) in water was mixed with 0.25 mol dm–3 of hexamine cobalt(iii) in 0.1 mol dm–3 hcl, and the ionic strength adjusted to 0.1 mol dm–3 using sodium perchlorate. the mixture was allowed to stand at 30oC for 48 h under nitrogen. The mixture was filtered.

(i) Precipitate. the precipitate was washed with water and dried. a small portion of the precipitate (0.25 g) was dissolved in 5 ml of concentrated hcl, and a few crystals of ammonium thiocyanate were added. the blue coloured solution obtained was shaken with 10 ml amyl alcohol. The blue colour passed into the alcohol layer, confirming the presence of co2+ ion.

(ii) Filtrate. The filtrate was used to confirm the presence of boric acid in the final product, characterised by known chemical and spectral methods.

Stoichiometry. reaction mixtures containing an excess of co3+ were allowed to react with 0.01 mol dm–3 naBh4 at various acidities, at 30oc, and then analysed spectro-photometrically at 475 nm for the co3+ which was left. the results gave the overall reaction corresponding to: 8Co3+ + Bh4

– + 3h2O → 8Co2+ + B(oh)3 + 7h+ (1)

rEsults anD Discussion

Kinetic results. under pseudo-first order reaction conditions, the rate of disappear-ance of cobalt(III) was observed to be first order in each of the reactants (Table 1). A plot of k1 against a twenty-fold range of [Bh4

–] concentration was linear and passed through the origin. at constant [Bh4

–] concentration (large excess), the pseudo-first order rate constant (k1) did not change, with changing co(iii) ion concentration (ten-fold range), indicating a first order dependence on Co(III) ion concentration. The rate of the reaction was observed to vary as a function of ph (table 1). the logarithm of the disappearance of co(iii) divided by [Bh4

–], was plotted against ph. the plot was linear, indicating a first order dependence of the rate on the hydrogen ion concentra-tion. under the present experimental conditions, the rate law could be expressed as follows: –d[Co (III)]/dt = k [co (iii)] [Bh4

–] [h+]. (2)

876

the effect of a change in temperature was studied (table 1). From the linear plot of lg k1 against 1/T, the activation energy was calculated (E = 76 ± 3 kJ mol–1). the other activation parameters calculated were: ΔH* = 73 ± 3 kJ mol–1; A = 2 × 1011 s–1; ΔS* = –112 ± 4 J k–1 mol–1. The negative value for ΔS* indicated that electron transfer played a dominant role in this reaction.

table 1. rate data for the oxidation of naBh4 by co(iii) at 30oc (µ = 0.1 mol dm–3)

[co(iii)] × 105

(mol dm–3)[Bh4

–] × 103

(mol dm–3)[hcl]

(mol dm–3) k1 × 104

(s –1)0.51.02.55.07.5

10.010.010.010.010.010.010.010.010.010.010.010.010.0

1.01.01.01.01.01.02.03.05.0

10.020.01.01.01.01.01.01.01.0

0.50.50.50.50.50.50.50.50.50.50.51.02.03.00.50.50.50.5

8.9 8.2 8.5 8.8 8.7 8.6 16.7 24.7 40.0 79.0159.0 17.3 34.0 51.0 4.3a

6.6b

12.8c

17.4d a20oc; b25oc; c35oc; d40oC; all temp. ± 0.1oc.

Variations in the ionic strength of the medium (μ = 0.01 to 0.10 mol dm–3) did not affect the rate, since the high concentrations of the acid predominated over the ionic strength.

Mechanism. When hydroborates react with acidic species, diborane is generally formed in a molar amount equal to one-half of the number of moles of hydroborate consumed, as for example7: naBh4 + h2so4 → ½ B2h6 + nahso4 + h2. (3)

it could be postulated that Bh4 – reacted with the aqueous proton:

slow Bh4

– + h+ ––––→ Bh3 + h2 (4)

however, other rate-determining steps could be considered, such as:slow

Bh4– + h+ ––––→ Bh5 → BH3 + h2 (5)

877

slow h3o+ + Bh4

– ––––→ h3o. Bh3

+ + h – (6)

slow Bh4

– + h+ ––––→ Bh5 → BH3 + h2 (7)

reactions represented by equation (5) usually proceed faster in D2o than in h2o, because of the weaker basicity of D2o with respect to h2O (Ref. 8). We have observed that the hydrolysis of Bh4

– was faster in h2o than in D2o by a factor ≈ 10, thus ruling out the mechanism given in equation (5).

We have also observed that BD4– reacts faster than Bh4

– (≈ 8), and BH4

– hydro-lyses faster in h2o than in D2o ( ≈10). this would indicate that, in the formation of the activated complex, a B–H bond was strengthened, and an O–H bond was weak-ened. these observations would rule out the possibility of the mechanism given in equation (6).

it has been shown that when Bh4 –

was hydrolysed in D2O, about 4% of the evolved hydrogen was h2 (ref. 9), whereas when BD4

– was hydrolysed in water, about 1% of the hydrogen was D2o. it would seem likely that in a process such as that shown in equation (4), the reacting proton (which essentially plucks a hydride ion from the Bh4

– ion) would always end up in the molecular hydrogen. these arguments would exclude the mechanism shown in equation (4).

it would seem more consistent to visualise that intermediates having the com-position, Bh4D or BD4h, would be formed. these would mainly decompose to give, respectively, hD and Bh3 (or BD3). Hence, it would be justified to postulate that the mechanism shown in equation (7) would be the most favourable pathway for the reaction.

In the present study, it would be appropriate to suggest that the first step involved the reaction of h+ with Bh4

– to give the intermediate, h+Bh4–, which was supported

by the experimental observation that the rate of the reaction was dependent on the first powers of the concentrations of both, H+ and Bh4

– ions (table 1). hence, the first step would be:

slow Bh4

– + h+ ––––→ h+ Bh4

– (8)

since this redox reaction involved an one-electron transfer, the steps following equation (8) could be represented as follows: h+ Bh4

– → ½ B2h6 + h2 (9)

h2 + 2co3+ → 2Co2+ + 2h+ (10)

½ B2h6 + 3 h2O → B (OH)3 + 3 h2 (11)

the overall stoichiometric reaction would be: 8Co3+ + Bh4 – + 3h2O → 8Co2+ + B(oh)3 + 7h+ (12)

878

conclusions

the kinetics of the oxidation of sodium tetrahydroborate by cobalt(iii) was studied. the reaction pathway involved an one-electron transfer process, giving cobalt(ii), characterised by chemical and spectral methods.

acknoWlEDGEMEnts

Financial assistance from the university Grants commission, new Delhi, under the special assistance Programme, is gratefully acknowledged.

rEFErEncEs1. h. c. BroWn, E. j. MEaD, B. suBBa rao: j. amer. chem. soc., 77, 6209 (1955).2. M. DasGuPta, s. DasGuPta, M. k. Mahanti: oxid. commun., 29, 363 (2006).3. W. c. PricE, h. c. lonGuEtt-hiGGins, B. ricE, t. F. younG: j. chem. Phys., 17, 217

(1949).4. T. C. WADDINGTON: J. Chem. Soc., 4783 (1958).5. G. BrauEr (Ed.): handbook of Preparative inorganic chemistry. vol. 2, 1969, p. 1531.6. F. Basolo, r. G. PEarson: Mechanisms of inorganic reactions. Wiley, new york, 1973,

p. 655.7. h. G. WEiss, i. shaPiro : j. amer. chem. soc., 81, 6167 (1959).8. L. MELANDER: Isotope Effects on Reaction Rates. Ronald Press, New York, 1960.9. W. j. jolly, r. E. MEsMEr: j. amer. chem. soc., 83, 4470 (1961).

Received 12 December 2007 Revised 17 January 2008

879

Oxidation Communications 31, No 4, 879–884 (2008)

* For correspondence.

oxidAtion of α-Hydroxy Acids by pyrAziniUm cHlorocHromAte

k. G. sEkar*, a. PraBakaran

Department of Chemistry, National College, 620 001 Tiruchirappalli, Tamilnadu, India E-mail: drkgsekar@yahoo.co.in; drpraba@gmail.com

aBstract

The kinetics of pyrazinium chlorochromate (PzCC) oxidation of α-hydroxy acids, viz. glycolic, lactic and mandelic acids, has been studied in the presence of perchloric acid in aqueous medium. the order with respect to [Pzcc], [substrate] and [h+] is one each. no appreciable change in rate is observed by lowering the dielectric constant of the medium. Addition of sodium perchlorate and manganese sulphate has no significant effect on the rate of oxidation. absence of any effect on acrylonitrile rules out the possibility of one-electron oxidation. the proposed mechanism involves the removal of water molecule from the protonated structure in the rate-determining step. From the kinetic data obtained, the activation parameters have been calculated. the order of reactivity is mandelic acid > lactic acid > glycolic acid.

Keywords: oxidation, pyrazinium chlorochromate, α-hydroxy acids.

aiMs anD BackGrounD

the oxidation of chromium(vi) has been well studied. the pathway in the mechanistic aspect varies with the nature of chromium(vi) species and the solvent used. recently, a variety of chromium(vi) oxidants together with special reaction conditions have been developed for the chemospecific, regiospecific and stereospecific oxidative generation of functionality in highly sensitive system. Pyrazinium chlorochromate (Pzcc) (ref. 1), one of the chromium(vi) compounds, is reported to be a neutral and mild oxidant for selective substrate. the kinetics of oxidation of benzyl alcohols by pyrazinium chlorochromate has already been reported2. The oxidation of α-hydroxy acids by various chromium compounds has also been studied extensively3–9. now we report the kinetics and mechanism of oxidation of α-hydroxy acids by pyrazinium chlorochromate in aqueous medium.

880

ExPEriMEntal

Materials and method. Pyrazinium chlorochromate (Pzcc) was prepared by litera-ture method1 and the purity of the sample was determined as 99% by an iodometric procedure. α-Hydroxy acids were purified by recrystallisation to a constant melting or boiling point. all other chemicals used were of ar grade and doubly distilled water was used for entire kinetic reaction.

Kinetic measurements. The reaction was carried out under pseudo-first order condi-tions [substrate] >> [PzCC] in 100% aqueous medium containing perchloric acid. The reaction was followed colorimetrically making use of digital photoelectric colorim-eter manufactured by Equiptronics (EQ-650). the kinetic runs were followed with 470 nm filter throughout the study10. the rate constant (k1) computed from the linear plot of lg [Pzcc] versus time (r>0.999) by least square method, were reproducible within ±2%.

Stoichiometry and product analysis. reaction mixture containing excess of oxidant over substrate was kept at room temperature in the presence of perchloric acid for 24 h. Estimation of unreacted pyrazinium chlorochromate showed that 1 mol of substrate consumed 1 mol of oxidant. the stoichiometric amount of substrate and oxidant were mixed and kept at room temperature under kinetic conditions for 24 h. the product was identified by spot test11 and ir spectral data.

rEsults anD Discussion

The kinetics of oxidation of α-hydroxy acids by PzCC was investigated at several initial concentrations of the reactants. the oxidation of lactic acid by Pzcc proceeds smoothly at 313 k in 100% aqueous medium. The reaction was found to be first order with respect to Pzcc as evidenced by the linear plot of lg [Pzcc] versus time. But, the rate of the reaction decreased with increase the concentration of oxidant12–14, the plot of [Pzcc] versus 1/k1 which is also linear. The pseudo-first order constants in Pzcc (k1) calculated at different initial concentrations of the substrate are found to increase linearly with the increase in lactic acid concentration. a plot of lg k1 versus lg [lactic acid] gave a straight line with a slope of unity, the correlation coefficient being 0.998 (Table 1).

increase in ionic strength of the medium by adding sodium perchlorate has no effect on the reaction rate indicating the involvement of an ion and neutral molecule in the rate-determining step15. at constant concentration of the reactants, the reaction rate increased with the increase in the concentration of the hydrogen ion (table 2). But, the plot of lg k1 versus lg [h+] was found to be linear with a slope of unity indicating first order dependence with respect to hydrogen ion concentration.

881

table 1. Dependence of the oxidation rate on oxidant and substrate at 313 k[h+] = 1.47×10–1 M; solvent = 100% H2o

[Pzcc] × 102 (M) [lactic acid] × 101 (M) k1 × 104 (s–1)0.39 0.60 3.700.59 0.60 2.950.79 0.60 2.560.99 0.60 2.141.18 0.60 1.820.59 0.40 1.940.59 0.60 2.950.59 0.80 4.130.59 1.00 5.150.59 1.20 6.240.59 1.40 6.95

table 2. Dependence of the oxidation rate on ionic strength and [h+] at 313 k[Pzcc] = 0.59×10–2 M; [lactic acid] = 0.60×10–1 M; solvent = 100%H2o

[naclo4] × 102 (M) [hclo4] × 101 (M) k1 ×104 (s–1)– 0.73 1.66– 1.47 2.95– 2.21 4.38– 2.94 5.54– 3.68 6.82– 4.41 8.23

0.00 1.47 2.950.20 1.47 3.090.30 1.47 2.840.40 1.47 3.25

the rate of the reaction was unaffected by lowering the dielectric constant of the medium. there was no appreciable change in the rate with the addition of manganous sulphate (table 3). no polymerisation was observed with acrylonitrile, indicating the absence of free radicals during the reaction.

table 3. Effect of solvent and Mn(ii) on reaction rate at 313 k [Pzcc] =0.59×10–2 M; [lactic acid] = 0.60×10–1 M; [h+] = 1.47×10–1 M

Water: acetonitrile (%) [Mn(ii)] × 102 (M) k1 × 104 (s–1) 90:10 – 3.48 80:20 – 3.52 70:30 – 3.54 60:40 – 3.53100:00 0.00 2.95100:00 0.15 2.83100:00 0.30 2.98100:00 0.45 3.01

882

the rate constants were measured at 4 different temperatures and activation pa-rameters have been calculated from a plot of lg k2 versus 1/T using the Eyring equation (table 4). the low Ea and ∆H* values supported the proposed concerted mechanism. the linear trend between enthalpies (∆H*) and entropies (∆S*) of activation (Fig. 1) shows that the reaction is controlled by both parameters. the high negative values of the entropy of activation (∆S*) suggest assumption of highly solvated transition state due to its increased polarity. Free energy of activation (∆G*) values are nearly constant which indicates that all the α-hydroxy acids are oxidised by the same mechanism.

table 4. Thermodynamic parameters for the oxidation of α-hydroxy acids with PzCC

compoundk1 × 104 (s–1) ∆H*

(kj mol–1)

–∆S*

(j k–1 mol–1)

–∆G* (kj mol–1)at 313 k

Ea (kj mol–1)at 313 k 303 k 313 k 323 k 333 k

Mandelic acid 5.61 9.33 13.96 19.73 15.25 164.11 66.61 17.85lactic acid 1.48 2.95 4.75 8.16 20.40 151.87 67.93 23.00Glycolic acid 0.62 1.27 2.32 4.98 24.86 140.49 68.83 27.46

10

15

20

25

30

–170 –160 –150 –140 –130

∆S*

∆H*

glycolic acid

(J K–1 mol–1)

(kJ m

ol–1

)

mandelic acid

lactic acid

r = 0.999

fig. 1. isokinetic plot

the activation enthalpies, entropies are linearly related by the following equa-tion16,17:

∆H* = ∆H0 + β∆S*

where β is the isokinetic temperature, ∆H0 – the enthalpy of activation at ∆S*= 0 and usually has no physical significance. The isokinetic temperature obtained from the slope is 407 k (r=0.999).

The genuine nature of isokinetic relationship was verified by the Exner crite-rion by plotting lg kobs at 313 and 303 k (Fig. 2). this plot is found to be linear with r=0.999. This linearity of the Exner plot is suggestive of unified mechanism for the PzCC oxidation of all α-hydroxy acids studied.

883

0.5

1

1.5

2

2.5

0.5 1 1.5 25 + lg k1(303 K)

5 +

lg k

1 (31

3 K

)

mandelic acid

lactic acid

glycolic acid

fig. 2. the Exner plot

MEchanisM anD ratE laW

The reaction shows first order with respect to [PzCC], [substrate] and [H+]. the rate of the reaction was unaffected by lowering the dielectric constant of the medium. increase in ionic strength of the medium by adding sodium perchlorate has no effect on the reaction rate indicating the involvement of an ion and neutral molecule in the rate-determining step. there was no appreciable change in the rate at the addition of manganous sulphate. no polymerisation was observed with acrylonitrile, indicating the absence of free radicals during the reaction. Based on the above facts, the follow-ing mechanism (scheme) was proposed:

s c h e m e

884

ratE laW

the above mechanism leads to the following rate law: –d[PzCC]/dt = k3 [product] = k3 K2 K1 [s] [o] [h+] = kobs [s] [h+]

this rate law explains all the observed experimental facts.

structurE anD rEactivity oF α-hyDroxy aciDs

In the present study, the order of reactivity for oxidation of α-hydroxy acids by pyrazinium chlorochromate has been found as follows: mandelic acid > lactic acid > glycolic acid.

lactic acid is oxidised at a slower rate than mandelic acid as the methyl group in lactic acid substitution in place of phenyl group in lactic acid may destabilise the complex intermediate18. the larger rate of mandelic acid is due to the resonance stabi-lisation of the intermediate. however, introduction of alkyl group increases the rate of the reaction among the acids. so, lactic acid reacts at higher rate than glycolic acid.

acknoWlEDGEMEnt

the authors thanks the Directorate of collegiate Education, chennai, for granting me a fellowship.

rEFErEncEs 1. B. h. Davis, M. r. shEEts, M. j. BrannFors, W. W. PaulDEr, l. G. GarD: heterocycles,

20 (10), 2029 (1983). 2. s. D. choi, y. c. Park: j. korean ind. Eng. chem., 14 (7), 994 (2003). 3. k. k. BanErji: j.chem. res. (S), 193 (1978). 4. n. k. saran, r. c. acharya: j. indian chem. soc., lxii, 747 (1985). 5. r. asoPa, P. Bhatt, k. k. BanErji: indian j. chem., 31A, 706 (1992). 6. s. M. sunDaraM, v. sathiyEnDiran: oxid. commun., 21 (1), 71 (1998). 7. s. M. sunDaraM, v. sathiyEnDiran: oxid. commun., 21 (1), 77 (1998). 8. k. R. SANkARAN, C. ANBuSELVAN: Asian J. Chem., 10 (4), 806 (1998). 9. P. Pohani, anjana, P. k. sharMa: indian j. chemistry, 45A, 2218 (2006).10. B. BhattacharjEE, M. n. BhattacharjEE, M. BhattacharjEE: int. j. chem. kin.,

17, 629 (1985).11. F. FEIGL: Spot Tests in Organic Analysis. Elsevier, Amsterdam, 1966, p. 482.12. M. k. DorFMan, j. W. GryDEr: inorg. chem., 1, 799 (1962).13. n. vEnkatasuBraManian, v. s. srinivasan: indian j. chem., 8, 57 (1970).14. M. krishnaPillay, a. thirunavukkarasu: indian j. chem., 20b, 583 (1981).15. P. chockalinGaM, P. s. raMakrishnan, s. j. arulraj, k. naMBi: j. indian chem.

soc., 69, 247 (1992).16. j. E. lEFFlEr: j. org. chem., 20, 1202 (1955).17. r. c. PEtErson: j. org. chem., 29, 3133 (1964).18. S. D. PAuL, D. G. PRADHAN: Indian J. Chemistry, 9, 835 (1977).

Received 17 August 2007 Revised 30 September 2007