Oxidation of thiols by oxygen catalysed by copper(II)ions or vitamin ...

Oxidation of thiols by oxygen catalysed by copper(II)ionsor vitamin B12Kuijpers, F.P.J.

DOI:10.6100/IR114082

Published: 01/01/1974

Document VersionPublisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers)

Please check the document version of this publication:

• A submitted manuscript is the author's version of the article upon submission and before peer-review. There can be important differencesbetween the submitted version and the official published version of record. People interested in the research are advised to contact theauthor for the final version of the publication, or visit the DOI to the publisher's website.• The final author version and the galley proof are versions of the publication after peer review.• The final published version features the final layout of the paper including the volume, issue and page numbers.

Link to publication

Citation for published version (APA):Kuijpers, F. P. J. (1974). Oxidation of thiols by oxygen catalysed by copper(II)ions or vitamin B12 Eindhoven:Technische Hogeschool Eindhoven DOI: 10.6100/IR114082

General rightsCopyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright ownersand it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal ?

Take down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Download date: 15. Mar. 2018

OXIDATION OF THIOLS BY OXYGEN

CAT AL YSED BY

COPPER (11) IONS OR VITAMIN B12

F.P.J. KUIJPERS

OXIDATION OF THIOLS BY OXYGEN

CAT AL YSED BY

COPPER CII) IONS OR VITAMIN B 12

PROEFSCHRIFT

TER VERKRIJGING VAN DE GRAAD VAN DOCTOR IN DE

TECHNISCHE WETENSCHAPPEN AAN DE TECHNISCHE

HOGESCHOOL EINDHOVEN,OP GEZAG VAN DE RECTOR

MAGNIFICUS,PROF.DR.IR. G. VOSSERS,VOOR EEN

COMMISSIE AANGEWEZEN DOOR HET COLLEGE VAN

DEKANEN IN HET OPENBAAR TE VERDEDIGEN OP

DINSDAG 9 APRIL TE 16.00 UUR

door

Franciscus Petrus Jacobus Kuijpers

geboren te Roermond

© 1974 by F.P.J. Kuijpers

DRUK ,» •. a"w VOORSCHOTEN

Dit proefschrift is goedgekeurd door de promotoren:

prof.dr. G.C.A. Schuit en

prof.dr. w. Drenth

To my parents,

to my wife.

PREFACE

This thesis deals with the mechanism of the homo­

geneously catalysed oxidation of thiols by molecular oxygen

in strongly alkaline solutions at room temperature. As

catalysts copper(II)ions as wellas vitamin B12 were used.

Thiols investigated are L(+)-cysteine, cysteamine monohy­

drochloride, thioglycollic acid, -butanethiol and

n-butanethiol.

According to the pertinent literature the main prob­

lem is whether or not the oxidation product disulfide is

formed via free thiyl radicals. After a survey of literature

and a general introduetion to this problem in chapter 1,

an answer is given in chapter 2 for the capper catalysed

and in chapter 5 for the vitamin B12 catalysed reaction.

The structure of a transient Cu(II)-thiol complex, observ­

ed during the ESR-experiments as described in chapter 2, is

elucidated in chap.ter 3. The mechanism of the oxidation of

thiols by molecular oxygen, catalysed by copper(II)ions is

presented in chapter 4. Por vitamin B12

catalysis a mechan­

ism is given in chapter 5.

An extension of the detection of thiyl radioals from

acidic to alkaline media by the ESR-rapid-mixing metbod is

described in Appendix I.

In Appendix II is shown that radioals are not

involved in the metal-ion catalysed oxidation of thiols by

molecular oxygen in alkaline media (by means of a rapid­

freezing-technique in combination with ESR-measurements

at -l70°C).

CONTENTS

PREFACE

CONTE1i!TS

CHAPTER 1 GENERAL INTRODUeTION

1.1

1.2

Survey of the literature

Radical mechanism and the rate­determining reaction

References

CHAPTER 2 THE ROLE OF THIYL RADICALS IN THE OXIDATION OF THIOLS BY MOLECULAR OXYGEN CATALYSED BY COPPER(II) IONS IN ALKALINE MEDIUM

2.1

2 .1.1

2 .1. 2

2 .1. 3

2 .1. 4

2.2

2.2.1

2.2 .2

2. 2. 3

Introduetion

Concentratien and mean life time of thiyl radicals in the catalysed homogeneaus system

Introduetion to the rapid­mixing ESR-measurements

Introduetion to the spin-trap­ESR-measurements

Check on the applicability of the spin-trap

Experimental part

Materials

Kinetic measurements and ESR­liquid-recirculation measure­ments

ESR-measurements

2.2.3.1 General

2.2.3.2 Rapid-mixing-ESR-measurements

2.2.3.3 Spin-trap-ESR-measurements

2.2.3.4 Experiments on photo-dissociation

2.3

2. 3.1

2. 3.2

2.3.3

Results

Kinetic measurements

ESR-rapid-mixing measurements

ESR-spin-trap-measurements

V

VII

1

1

2

4

7

7

7

8

9

10

10

10

10

12

12

12

14

14

15

15

18

20

CHAPTER 3

2.3.3.1 Experiments on photo­dissociation

2.3.3.2 Experiments on the catalytic system

2.4 Discussion and conclusions

2.4.1 Rapid-mixing-system

2.4.2 Spin-trap measurements

2.4.2.1 Irradiated systems: interpretation of the adduct spectra

2.4.2.2 The rate of production of thiyl radicals

2.4.2.3 Catalytic systems

Conclusions

Summary

References

TRANSIENT COPPER(II)-THIOLATE COMPLEXES

THE STRUCTURE OF THE COPPER(II)-DICYSTEINATE COMPLEX

3.1 Introduetion and survey of the literature

3.2 Experimental part

3.2.1 Materials

3.2.2 ESR-measurements

3.2.2.1 General

3.2.2.2 Rapid-mixing-experiments

3.2.2.3 Liquid-recirculation measurements

3.2.2.4 Measurements at -170°C

3.2.2.5 Absorption measurements in visible light

3.3 Results

3. 3.1 ESR-rapid-mixing-measurements

3.3.2 ESR-liquid-recirculation-measurements

3.3.3 ESR-measurements at -l70°C

3.3.4 Visible light-absorption measurements

3.4 Discussion and conclusions

3.4.1 Transient ESR-spectra

20

23

24

24

29

29

33

34

35

35

36

37

37

39

39

39

39

39

40

41

42

42

42

45

45

47

47

47

3.4.2

3.4.3

Low-temperature ESR-spectra

VIS-absorption measurements

Conclusion

Appendix

Summary

References

CHAPTER 4 MECHANISM OF OXIDATION OF CYSTEINE BY MOLECULAR OXYGEN IN 0.25 MOL/L NaOH CATALYSED BY COPPER(II)IONS

4.1 Introduetion

4.2

4.2.1

4.2.2

4.2.3

4.2.4

4.2.5

4.3

4. 3.1

4.3.2

4.3.3

4.3.4

4.4

4. 4.1

4.4.2

Experimental part

Reagents

Kinetic measurements

Quantitative Cu(I) analysis

Relative ESR intensity measurements

Quantitative product analysis

Results

Kinetic measurements

Quantitative Cu(I) analysis

Relative ESR intensity measurements

Qualitative product analysis

Discussion and conclusions

First process

Second process

Appendix

Summary

References

CHAPTER 5 OXIDATION OF THIOLS BY MOLECULAR OXYGEN IN ALKALINE MEDIUM CATALYSED BY VITAMIN B12 (CO(III))

s .1

5.2

S.3

5.4

Introduetion

Experimental

Results

Discussion and conclusions

References

49

S1

ss S6

60

61

63

63

66

66

66

67

68

68

69

69

75

77

77

81

84

97

98

101

102

104

104

105

106

107

109

APPENDIX I THE GENERATION OF PREE THIYL RADICALS WITH CE(IV) AS THE OXIDISING AGENT IN ALKALINE SOLUTIONS

References

APPENDIX II ATTEMPTS AT QUENCHING OF o; RADICALS IN CATALYTIC REACTION MIXTURES BY THE BRAY-RAPID­FREEZING-TECHNIQUE

References

SUMMARY

SAMENVATTING

DANKWOORD

LEVENSBERICHT

110

116

117

122

123

125

127

128

CHAPTER 1

GENERAL INTRODUCTION

1.1 Survey of the Ziterature

The homogeneaus oxidation of thiols by molecular oxygen

in alkaline media as catalyzed by transition metal ions and

transition metal complexes has been studied extensively

(ref. 1,2,3,4,5,6 and references cited therein).

In 1964 Wallace et al. published a mechanism based on

kinetic measurements,in which the formation of free thiyl

radicals is postulated {S).However Swan and Trimm (Sc) and

Cullis and Trimm {7) proceeding from the results of kinetic

measurements in the absence and the presence of streng

complexing ligands proposed an alternative mechanism based

on electron transfer in the coordination sphere of the metal

ion.

A decision as to which mechanism is mainly operative

obviously depends on the ability to detect the thiyl radicals

and to measure their concentration.

The formation of thiyl radicals in aqueous solutions

where they were produced by non-catalytic processes,such as

chemical oxidation of thiols with Ti(III)-H2o2

or Ce(IV) was

reported by Armstrong and Humphreys (B),Wolf et al. (9,10),

and Nicelau and Dertinger (ll).The methad used was that of

rapid-mixing-ESR.In all cases the detection only succeeded

in strongly acidic media and so far no reports are available

as totheir detection in alkaline media (12,13).Neta and

Fessenden used the reaction of hydrated electrans or hydroxyl

radicals with thiols to investigate the formation of thiyl

radicals as a function of the pH,but they only detected

carbon radicals of thiols (14).Pulse radiolysis has been used

to produce thiyl radicals (15,16,17,1B).These experiments

have led to the detection of the disulfide radical anion

(RSSR) in alkaline media.Hoffman,Hayon and Simic applied

pulse radiolysis and kinetic absorption spectroscopy to

fellow the concentratien of the thiyl radicals RS",the disul­

fide radical anion (RSSR)-,and its protonated form RSS(H)R as

a function of reaction time (13,19,20).

Parenthetically it may be remarked that these investigations

were principally concerned with an explanation ot the radio­

protective action of some thiolsin tissues (21,22,23,24).

The presence of thiyl radicals in tissues also appears to be

connected with ageing processes (25).Their relation with

biological reactions catalyzed by metallo-enzymes has been

discuss~d in a number of papers (26,27,28).

Our investigation attempted to decide between the

'radical' and 'internal' electron transfer mechanism by com­

bining kinetic and ESR-measurements by rapid-mixing and

spin-trap methods.

As a check on the applicability of the spin-trap method

to determine thiyl radicals,if actually present,we used

photo-dissociation of thiols by UV-light.

1.2 Radiaal meahanism and the rate-determining reaation

To discuss the implications of the radical mechanism the

reaction sequence according to Wallace et al. {3) was accep­

ted

- ____.. RS H20 1. RSH + OH --- +

- 2+ kl RS" Cu 1+

2. RS + Cu - +

3. RS" + RS" k2

RSSR -4. 2Cu1 ++ 02 2Cu2++ 2- a) - 02

2

4RSH + 0 2 - 2RSSR + b)

Reaction 2,i.e. the formation of thiyl, radicals is rate­

determining.

a) The presence of o; in the reaction sequence is ruled

out a priori,because these radicals were not detected

by ESR when applying the Bray-rapid-freezing-technique

(29) to our oxidation system (30),while the mean life­

time of the o; radical is known to be some hundred

ms (31,32).

b) Hydralysis of the disulfide and formation of sulfene,

sulfinic and sulfonic acids is known to occur in

alkaline solutions,particularly in solutions in aprotic,

dipolar solvents (33).In very streng alkaline solutions

direct formation of these acids from the thiyl anion

seems to be possible (34).

3

Referenaes ahapter 1 and ahapter 2

1. D.S. Tarbell in: 'Organic Sulfur Compounds',Vol. I,

N. kharash,Ed.,Pergamon Press,New York,1961,p. 97

2. B.S. Me Cormick and G.Gorin,

Inorg. Chem. !,691,(1962)

3. T.J. Wallace,A.Schriesheim,H, Hurwitz and M.B. Glaser,

Ind. Eng. Chem. Process Des. Develop. l• 237 (1964)

4. J.E. Taylor,J.F, Yan and Jin-Liang Wang,

J.Amer.Chem,Soc. ~,1663 (1966)

Sa. C.F. Cullis,J,D. Hopton and D.L. T~imm,

J. Appl. Chem. ~,330 (1968)

b. C.F. Cullis,J,D, Hopton,c.s. Swan and D.L. Trimm,

J ., Appl. Chem. ~,335 (1968)

c. c.s. Swan and D.L. Trimm,

J. Appl. Chem. 18,340 (1968)

6. P.c. Ellgen and e.D. Gregory,

Inorg. Chem. !Q,980 (1971)

7. Discuss. Faraday Soc. 144-149; 184-189 (1'968)

8. W.A. Armstrong and W.G. Humphreys,

Can. J, Chem. ~,2589 (1967)

9. w. Wolf,J.C. Kertesz and W.C. Landgraf,

J. Magn. Resonance ! 1 618 (1969)

10, J.c. Kertesz,W. Wolf and H.Hayase,

J, Magn. Resonance 22 (1973)

11. C. Nicolau and H. Dertinger,

Radiat. Res. ~,62 (1970)

12. W.A. Waters in: 'Free Radical Reactions',Organic

Chemistry Serier One,vol. 10, (1973),p. 282

13. M.Z. Hoffman and E. Hayon,

J, Phys. Chem. 990 (1973)

14. P. Neta and R.W. Fessenden

J. Phys. Chem. 2277 (1971)

15. G.E. Adams,G.S. Me Naughton and B.D. Michael in :

'Chemistry of Ioniaation and Excitation',G.R.A. Johnson

and G, Scholes.Eds.,Taylor and Francis,London 1967,p,281

16. G.E. Adams,G.S, Me Naughton and B.D. Michael,

Trans< Faraday Soc. 64,902 (1968)

4

17. W. Karmann,A. Granzow,G. Meissner and A. Henglein,

Int. J. Radiat. Phys. Chem. ! 1 395 (1969)

18. G, Caspari and A. Granzow,

J. Phys. Chem. 74,836 (1970)

19. M.Z. Hoffman and E. Hayon,

J. Amer. Chem. Soc. 2i,7950, (1972)

20. M. Simic and M.Z. Hoffman,

J. Amer. Chem. Soc. 6096 (1970)

2la.D.G. Doherty,W.T. Burnett,Jr. and R. Shapira,

Radiat. Res. 2.,13 (1957)

b.R. Shapira,D.G. Doherty and W.T. Burnett,Jr.,

Radiat. Res. 2.,22 (1957)

c.D.G. Doherty and R. Shapira,

Radiat. Res. 107 (1958)

22a.E.S. Copeland.E.C. Richardson and H.M. Swartz,

Radiat. Res. i2,542 (1971)

b.E.S. Capeland and M.M. Grenan,

Radiat. Res. !2,387 (1971)

c.E.S. Capeland and W,L. Earl,

Int. J. Radiat. Biol. !2,401 (1971)

23. W.O. Foye,Annual Reports of Medicinal Chemistry,

1669 (1970)

24. G. Nucifora,B. Smaller,R.Remko and E.C. Avery,

Radiat. Res. i2 1 96 (1972)

25. W.A. Pryor,Scientific American 223,70 (1970)

26. G. Agnes,H.A.O. Hill,J.M. Pratt,s.c. Ridsdale,

F.S. Kennedy and R.J.P. Williams,

Biochim. Biophys. Acta 207 (1971)

27a.G.N. Schrauzer,J.A. Seck,R.J. Holland,T.M. Beckham,

E.M. Rubin and s.w. Sibert,

Bioinorganic Chemistry ~,93,(1972)

b.G.N. Bchrauzer and R.J. Windgassen,

J. Amer. Chem. Soc. ~,3607 (1967)

c.R.H. Prince and D.A. Stotter,

J. Inorg. Nucl. Chem. ~,321 (1973)

28. P.C. Jocelyn: 'Biochemistry of the SH group',

Academie Press,London,1972

5

29. R,Co Brayin :'Rapid mixing and Sampling Techniques in

Biochemistry',B. Chance,R. Eisenhardt,Q.H. Gibson and

K.K. Lonberg-Holm,Eds.,Academic,New York,1964,p. 195

30. To be publisbed by F.P.J. Kuijpers and A.M. Edelbroek;

see appendix II in this thesis

31. P.F. Knowles,J.F. Gibson,F,Mo Pick,R.C. Bray

Biochem. J. !.!!,53 (1969)

32. R, Nilsson,F.M. Pick,R.C. Bray,Mo Fielden,

Acta Chem. Scand. ~,2554 (1969)

33a. T.J. Wallace and A. Schriesheim,

Tetrahedron Lett. !1,1131 (1963)

b. T.J. Wallace and A. Schriesheim,

J. Org. Chem. ~,1514 (1962)

c. T.J. Wallace and A. Schriesheim,

Te~rahedron Lett. 1!,2271 (1965)

34. H. Berger, Reel. Trav. Chim. Pays-Bas ~,773 (1963)

6

CHAPTER 2

THE ROLE OF THIYL RADICALS IN THE OXIDATION OF THIOLS

BY MOLECULAR OXYGEN CATALYZED BY COPPER(II) !ONS IN

ALKALINE MEDIUM*

.1 Int~oduction

Concentration and mean life time of thiyl ~adicals

in the cata zed homogeneaus sytem

The concentratien and the mean life time of

radicals are the parameters which determine whether they can

be detected by ESR.The concentratien has to be higher than

Sxlo- 8 mol/1 under optimal conditions and the mean life time

has to be longer than the correlation time of the applied

microwave frequency.

We define the mean life time of radicals as

T concentratien of radicals rate of disappearance of the radicals

In a steady-state it follows

[RS"J

(d[RS"] )

dt formation

Since the formation of thiyl radicals is assumed to be

rate-determining

T [RS.]

(- d~:2]) measured

*a summary is given at the end of the chapter

7

If we substitute a mean value for

- d[02] -5 -1 -1 ~ = 10 moll s (see fig. 2-3 and tabZe 2-1)

we find the following correlation between the concentratien

and the mean life time of the thiyl radicals in the homo­

geneous catalytic system:

T x1o-5 mol/1 [RS"]

9 A microwave frequency of 9x10 Hz was used.Hence the corre--10 lation time is nearly 10 s.So the only parameter relevant

for the detection of the radicals by ESR will be their

concentration.

2. 1. 2 Introduetion to the rapid-mixing-ESR-measurements

We have studied the reaction between copper(II) and

thiol in a rapid-mixing-ESR-cell.A scheme of the apparatus

is given in figure 2-2.The concentratien of the thiyl

radicals in the flow cell was calculated by assuming a

steady-state.Under this condition

~d[RS "]) \ dt formation

[d[RS • f\ \ dt Îdisappearance

( 1)

Substitution gives

(2)

k1

[Cu (II)] [RS ]0

(3)

because in the rapid-mixing cell always [RS-] >> [Cu(II)]

and therefore[RS-]is practically a constant.

8

it follows that

Combining equations (2), (3) and (5) gives

We may write equation (6) as

1 l t)} /2

0

(5)

(6)

(7)

[RS'] is a function of the reaction time and therefore of

the position in the cell.

2. 1. 3 Introduetion to the spin-trap-ESR-measurements

ESR detection of short living radicals can be impro­

ved by converting them to langer living adduct radicals.By

using spin-traps we expect a reaction between the thiyl

radical and the spin-trap forming a stable adduct radical

that hence becomes observable by ESR (35).The g0

-value and

the nitrogen hyperfine coupling constant of the spin-

adduet enables the identification of the trapped free

radical.Most spin-traps are nitroso or nitrone compounds

( 36).

In our measurements we have used nitromethane.

Nitromethane is partly ionized in basic media and radicals

can be trapped by the aci-ion (3?).

The ESR-spectrurn of the spin-adduet is a 'triplet'

of 1:2:1 triplets,caused by the interaction of the electron

9

and the nuclear spins of nitrogen and the two equivalent

hydragen atoms.

2.]. 4 Check on the applicability of the spin-trap

Befere applying nitromethane in the homogeneaus cata­

lytic system,it is necessary to check the spin-trap on its

efficiency to detect thiyl radicals by using a standard.The

efficiency is high enough when the spin-adduet in the stan­

dard salution is detectable at a lower rate of formation of

the thiyl radicals than in the catalytic system.

As standard we have chosen the system thiol/nitro­

methane/NaOH irradiated with UV light.

2.2 Experimental part

Thiols investigated were L(+)-cysteine,cysteamine­

monohydrochloride,thioglycollic acid and t-butanethiol.

2. 2. 1 '4.aterial.s

All chemieals were obtained pro anaZisi from Merck

and used as such unless otherwise stated.

As capper salt cuso4

.sH2

o was used.

As complexing agent for copper(II) ions in the rapid-mixing­

experiments L(-)-histidine was used to avoid precipitation

of capper hydroxide.

Nitromethane was obtained from Fluka A.G. (boiling trajectory

98-101°C).

2. 2. 2 Kinetia measurements and ESR-Ziquid-reairauZation

measurements

The oxygen consumption in time was measured with a

Warburg-type apparatus,with the possibility for simultaneous

ESR-liquid recirculation measurements (see fig. 2-l).For

10

liquid-recirculation-measurements a Fluorocarbon Saturn pump

SPM-100 was used.Its chamber and plunger consisted of teflon.

The ESR-flow cell applied is an accessory of Varian

Associates (number E-248) .The effective volume of this cell

is 15x10- 5 l.

It was checked that the oxygen-diffusion into the solution

was so fast that it did not change any further by increasing

the speed of the stirrer.The oxidation was always performed

in 0.25 mol/1 NaOH (pH=13.4) at 24.5 °c.our thiols have pK­

values in the range of 10.4 to 10.7.The concentratien of

thiol in the reaction liquid was in the range of 7.5x1o- 3 to

5x10- 2 mol/l.The initial Cu(II) concentratien was varied

between 10-5 and 5x1o- 3 mol/l.The concentratien of nitro­

methane was 5x1o-2 mol/l.The spin-trap was added immediately

after the addition of the thiol to the system.The shortest

reaction time was 10 minutes i.e. long enough to detect

ESR-signals.

----------------------, vacuum pump

a, thiol salution

b' E.S.R floweelt

fig. 2-1, Apparatus for simultaneous measurements of

oxygen uptake and ESR absorption

(for reasans of convenience the whole apparatus was built

on a vehicle)

11

2.2.3 ESR-measurements

2.2.3.1 Gene:r>at

The ESR-measurements were performed on a Varian

E-15, X-band spectrometer with 100 kHz magnet field modula­

tion. The determination of the g-value occurred by using

an A.E.G.-nuc1ear-resonance magnet field meter in combinat­ion with a Hewlett Packard 2590 B microwave frequency con­

verter, a Hewlett Packard 5253 B frequency converter 50 te

500 MHz and a Hewlett Packard 5245 L electronic counter.

The microwave radiation had a power up to 200 mw. Saturation

did not uccur.

2.2.3.2 Rapid-mixing-ESR-measu:r>ements

The rapid-mixing-ESR-cell is a number E-249 of varian Associates.The dead volume of this cell is 2.5x10-6 1:

the cell volume is 15 x 10-5 1.

The flow rate of the mixing stream could be varied

between 0.1 and 4.0 ml/sec. A scheme of the apparatus is

given in figure 2-2.

The Cu(II) concentratien in the rapid-mixing

experiments varied between 10-4 mo1/1 and 5 x 10-2 mo1/1.

L(-)-histidine was used as complexingagent at [Cu(II}]

~ 5 x 10-4 mol/1. The concentratien of this amino acid

was always twice the Cu(II) concentratien in order to form

the Cu(II)-dihistidine complex (38).

The concentratien of thiol in the rapid-mixing­

experiments varied between 7.5 x 10-3 mol/1 and 10-1 mol/1.

We can calculate the mean concentratien of thiyl

radicals in the mixing cell by the formula

Jt2 [RS • J dt t1

[Rs']

r2 (8) 1 in which

dt

t1

12

t 1 and t2

are the times at which the mixed streams enter

and leave the cell respectively.

dead volume of the rapid-mixing cell flow rate of the mixing stream

dead volume + cell volume of the mixing cell flow rate of the mixing stream

If we substitute equation (7) into (8) we find

that the mean concentratien of the thiyl radicals in the

mixing cell will be given by:

Integration of equation (9) gives

[RS .]

1/2 2 {[Cu(II)]

0}

1 2. {2 k

1k

2[RS-]

0}

(9)

( 10)

We calculate the rate constant k 1 from the rate of oxygen

consumption measured.

13

The value of 2 k2

was taken form Behar and Fessenden who 9 -1 -1 -

found 1.9 x 10 mol/1 s for the combination of ·so3

ra-

dicals (39). This 2 k2

value is the highest value measured

for the combination of sulfur radicals and also comparable

to the cellision number.

N2~acuum pump

02

a --~-,

b i : L ___ _

,l-~1 c

' d L_ __ _

--1 : ' '

-t t-!Qm_'!)_

ij--~. rimm A-A l ~

I I A A *

11

l,•cc•••••~<(=~ __j J

a: reservoirs (ll) b: magnet ie stirrers c: flowmeters d: valves e: E SR- rapid -mixing-cel f: liquid recirculation pump

fig. 2-2, Saheme of the ESR-~apid-mixing-system

(fo~ ~easons of aonvenienae the whoZe appa~atus was built

on a vehiaZe)

2.2.3.3 Spin-t~ap-ESR-measu~ements

The concentratien of nitromethane in the catalytic

and irradiated systems was 5 x 10-2 mol/1. The spin-trap­

measurements in the catalytic system were performed in the

ESR-rapid-mixing chamber or in the ESR-recirculation

system, because of the expected high instability of the

thiyl nitroxyde radicals (40). In the ESR liquid recircul­

ation system kinetic measurements were performed in

combination with ESR measurements. The time of pas-

sage from the reactor to the liquid cell was 3 s.

2.2.3.4 Expe~iments on photo-dissoaiation

The standard solutions were irradiated by light

from a 450 W XBO-Xenon lamp in a LX 501 Xenon source,

14

with a blue filter, obtained from Carl-Zeiss.

In this set up the frequency range for irradiation

was 400-280 nm. The distance between the ESR cavity and

the UV souree was 50 cm. The concentratien of thiol in

these experiments was 10-1 mol/1. All thiol solutions

were flushed with nitrogen. The standard solutions were

irradiated in situx.

2. 3 ResuUs

2.3.1 Kinetia measurements

The consumption of oxygen versus time is shown in

fig.2-3 for several Cu(II)-thiol systems. The relevant

data are given in table 2-1.

~1SCH 2 cooH] 0 , 50x10-3 molll

fig. 2-3, Course of oxygen uptake versus time for several

thiols, based upon the overall reaation

4RSH + 02

+ 2RSSR +

To check the formation of thiyl radicals in these

solutions, frozen standard solutions were UV irradiat­

ed in a similar manner. The 3-g thiyl radical signal

then appeared immediately.

15

TabZe 2-l,Rate of oxygen uptake for the aopper aataZysed oxidation of several thioZs

'

RSH [RSH]0

mo1/1 [Cu(II) ]0

mo1/1 (d[02]/dt)steady state mol/ls

Thioglycollic acid 5.0 x 10- 3 3.2 x 10-5 10-5

Cysteamine mono- 4.6 x 10-3 3.2 x 10-5 6.0 x 10-6

chloride

L(+)-cysteine 7.3 x 10-3 3.2 x 10-5 3.0 x 10-6

L(+)-cysteine 7.5 x 10-3 10-4 1.06 x 10-5

(More kinetic data are given in chapter 4)

From the steady-state value of the rate of oxygen con­

suroption the rate constant k 1 can be calculated.

This k1

value enables us to calculate the concentratien of

thiyl radicals in the rapid mixing cell by formula (10).

We give an example of such a calculation for the

minimum radical concentratien to be expected, viz. for the

slowest oxidation reaction (i.e. the oxidation of cysteine),

the lewest Cu(II) concentratien and the lewest cysteine

concentratien in the rapid mixing experiments and the ex­

tremely high value estimated for the recombination rate

constant k 2 • The values for the parameters are:

[Cu(II)]0

= 10- 4 mol/1

= 7.57 x 10- 3 mol/1 1.08 x 10-5 mol/1

9 -1 -1 1.9 x 10 1 mol s

flow rate ~= 1.5 ml/s

2.5 x 10-G 1

15 x 10-5 1

After calculating k 1 and substituting the given para­

meters in formular (10) we find:

[RS"] 7.5 x 10-8 mol/1

This concentratien is just above the lewest detectable con­

centration (= 5 x 10-8 mol/1). So we expect the direct de­

tection of thiyl radicals in the rapid mixing system, be-

17

cause line broadening caused by a limited lifetime of the

radicals wil! not occur.

2.3.2 ESR-rapid-mixing measurements

No signa! that could be ascribed to a thiyl radical

was ever detected in an alkaline Cu(II)-thiol-rapid-mixing

system.

The original Cu(II)-signal of the copper salt or

copper-dihistidine complex vanished by mixing with thiol.

In systems with cysteine, cysteamine and thioglycollic acid

it was replaced by transient Cu(II)-signals (see figure 2

4a, 4b, 4a, 4d, 4e). These signals vanished immediately

when stopping the flow.

·Similar signa! patterns were observed when the ex­

periments were performed in a nitrogren atmosphere.

18

50 G.

fig. 2-4a, ESR-speatrum of Cuso 4 .5H20 in soZution; [Cu] =

2 x 10-4 moZ/Z, pH= 13.4

H0 = 3300 Gauss I

fig. ESR-speatrum of Cu(II)-dihistidine in solution;

[Cu] 5 x 10-3 molll~ [histidine] = 1 motfl~ pH= 13.4

H0 :3300 Gauss I

100 G.

fig. 2-4a3 ESR-speatrum

mi~ing; [Cu(II)] = 5 x -2 0

5 x 10 mollt~ ~ = 1.5

L

of Cu(II) + aysteine during rapid­-3 10 mol/l 3 [HOOCCHNH2cH2SH] =

ml/s, pH 13.4

19

H0:3300G

100G I

fig. 2-4d. ESR-speatPum

Papid-mixing; [Cu(II)] . 0 -2 5 x 10 moZ/Z, ~ = 0.5

100G

fig. 2-4e~ ESR-speatrum

[Cu(II)]0

= 10-2 moZ/Z,

3 ml/s, pH= 13.4

of Cu(II) + aysteamine duPing -3 = 5 x 10 moZ/Z, [NH 2cn2cn2SH] =

mZ/s. pH = 13.4

H0 : 3300 G

I

of Cu(II) + thioglyaollia aaid;

[HOOCCH2

SH] = 10-l mol/l, ~ =

2.3.3 ESR spin-trap-measurements

2.3.3.1 Experiments on photo-dissoaiation

Spectra obtained by irradiating the standard solut­

ions are given in figure 2-(Sa, 5b, 5a, 5d).

The adductx -signals marked by an asterisk vanished

in a few seconds by cutting the irradiation (see figure 2-8).

20

x 0

P,4Q mW

1: , 30 sec

lreq ,9493.6044 kHz

fig. 2-5a, ESR-speatrum of a soZution of t-butanethiol

+ nitromethane during UV-irradiation, pH= 15.4

asterisk_ signa i

. 2-5b, ESR-spectrum of a solution of cysteamine +

nitromethane during UV-irradiation, pH= 15.4

21

22

P:100 mW

t dO sec áh 60min

H0 :3380 G

áH: 100 G

Hm:0.1 G

treq: 9491.9835 kHz

fig. 2-5a, ESR-speatrum of a solution of aysteine +

nitromethane during UV-irradiation, pH = 13,4

Ho

I

···~lil 11=\.

1

11

a~:24.l0

I

I. I I, aH.P a.N

,I lil

,I

P:40mW

t:3sec ál 8min

G:5xt04

Hm:025 G

H0 : 3380G àH:100 G

freq: 9492.9420 kHz

1l1 asterisk-sigoal

I CHi.îoi

I

fig. 2-5d, ESR-speatrum of a solution of thioglyaollia acid + nitromethane during UV-irradiation, pH = 13,4

) I

se ar> t1 me , 30 sec t 1 sec

G 5x10 4

Hm: 01 G

P 40mW

SG

fig. 2-6, Deaay of asterisk signa~ after cutting the UV­

irradiation

In additional experiments with (RSH] = 10-2 mol/1 and

[CH3

N02

]0

10-2 mo1/1 the adductx -s~gnal intensity was

followed as a function of time during continuous uv­irradiation. The asterisk signals disappeared about 30 mins.

after starting the UV-irradiation.

2.3.3.2 Experiments on the aatalytia system

Using the aci-ion of nitromethane as a spin trap,

a thiyl radical adduct was neither observed in the rapid

mixing experiments nor in the liquid recirculation

measurements. In the rapid mixing experiments the transient

Cu(II) spectra were observed {see fig.2-4) but in the li­

quid recirculation experiments only the transient Cu(II)

spectra for cysteine and cysteamine were observable. In

both sets of experiments the original Cu(II)-signal of

23

the copper salt or of the copper-dihistidine complex

vanished by adding thiol. In the liquid recirculation ex­

periments at the end of the reaction a fifteen line radi­

cal signal of a Cu(II)-CH2No; complex was observedx, a

signal never detected during the oxidation process.

UV-irradiation of the ESR-cavity during ESR-rapid-mixing

experiments and ESR-liquid-recirculation measurements on

the catalytic systems did not change the results obtained

without UV-irradiation.

2. 4 Discussion and conclusions

2. 4. 1 Rapid-mixing system

In the rapid-mixing system we did not detect an

EqR-signal of a thiyl radical but rather transient Cu(II)

signals. With cysteine and cysteamine the Cu(II) signals

showed quintet hyperfine splitting in the two high field

peaks. In the next paper we will show in more detail that

these splittings are due to the interaction of t~o equi­

valent 14 N nuclei (I = 1) of a thiolate ion with the un­

paired electron of Cu(II) (46). So the Cu(II) ion is sur­

rounded by at least two thiol anions. Therefore, we sug­

gest a transient complex as Cu(II) (RS-) , in which x

x ~ 2 and probably ~ 4.

Because the original Cu(II) signal of the copper

salt or of the copper dihistidine complex vanished in all

cases we have to conclude that in the cases of cysteine

and cysteamine and probably also for thioglycollic acid

and t-butanethiol the first step in the reaction sequence

is a practically complete complex formation of the Cu(II)

ion with thiolate ions.

A possible reaction mechanism via thiyl radical in­

termediates would be:

x In similar experiments without nitromethane the original

Cu(II) signal returned but weaker.

24

+ RSH + OH +- RS + H2 0

Cuii + k

xRS +0 Cu(II) (RS - )x

I

k Cu(II) (RS - )x

+1 Cu(I) + RS" (x-1)RS +

RS • + RS. k2 + RSSR

2 Cu(I) + 02 + 2 Cu(II) + 2-

02

2-02 + H

20 + 2 OH + 1/2 02

Once more we calculate the mean concentration of thiyl

radicals in the rapid-mixing cell for this adjusted me­

chanism by assuming steady states in the thiyl radicals

and in the Cu(II) (RS )x complex. Then we may write the

following equations:

ki [Cu (II) (RS ) x]

and

Substitution of (12) into (11) gives:

[RS "]

Fr om

[RS-]0

>> [Cu(II)]0

we derive:

[Cu (II)] [Cu(II)] 0 -k e o

(11)

(12)

( 13)

and

(14)

25

Substitution of (14) into (13) gives rise to:

[RS.] (15)

Integration of the right hand side of equation (15) between

the limits

leads to the average thiyl radical concentration in the

cell:

1/2 2([Cu(II)]

0) ••

1/2 (2 k2k0 [RS-]~) vc

x

Equation (16) may be approximated to:

(2 k ) 1/2 2

- x We deduce the value of k0

[RS ]0

from

with Ät the average flow time

Within the flow time -d[Cu(II)]= [Cu(II)]0

,

- x !iS hence k0 [RS ] 0 ~ Vc

If we substitute the lowest value of k0 (RS-]~ in

equation (17), this formula reduces to:

26

( 1 7)

(18)

(19)

Bv substitution the same values of the parameters as befare

(see heading Results 2.3.1) the lowest concentration to

be expected in the rapid-rnixing cell is:

'] 5.3 x 10-7 rnol/1

This value of the concentratien is suitable for ESR detect­

ion. Since no thiyl radicals were observed we have to

conclude that there exists another process of eliminatien

of the thiyl radicals. This cannot be the process proposed

by Wolf and Kertesz (9~41) in which the rate of decay is

first order in the concentratien of thiyl radicals and the -1

corresponding rate constant equals 9.2 s , because the

actual eliminatien rate has to be faster than the rate of

recornbination of thiyl radicals.

Another possibility is given by the reaction of RS' with

to forrn the radical anion (RSSR) , as detected by pulse

radiolysis (15~16~17~18).

We consider the case of cysteine. The following data

are given by Sirnic and Hoffman (20).

RS' + RS k -x

(RSSR)

k -x 2.5 x 105

Both types of radicals will react with oxygen:

k RS' + 02 +y RSOÎ

ky = 8 x 10 9 1 mol- 1 s -1 (42)

-1 s

27

{RSSR) + 02 RSSR + o; (43)

4.3 x 108 1/mol s (42)

If we assume steady statas in the thiyl radioals and the

anion disulfide radioals we obtain equation (20) and (21) :

and

kx[RS.)[RS-] + ky[RS.][0 2]

(20)

(21)

Proceeding as before in the case of cysteine and inserting -3 [02 ] = 10 mol/1, we calculate for steady state conditions

4 x 10-11 mol/1, and

(ss = steady state)

Both concentrations are below the dateetion limit for ESR.

In analogy with the calculation of an average concentratien

in the cell (see sections 2.2.3.2 and 2.3.1) in this case

we can expect an average concentratien for RS. and (RSSR)

of the order of magnitude of 10-11 and 10-9 mol/1, res­

pectively. This infers that introduetion of the reaction

Rs· + RS + (RSSR) might give an explanation for the ob­

servations that thiyl radicals could not be detected.

However, it is neoessary to remark that we could not

detect the o;-radical by ESR using the Bray-rapid-freezing

technique (29,30). The long life of this species should

have allowed us to detect it (31,32). Moreover, measure­

ments by Caspari and Granzow indicate that the formation

of (RSSR) at pH = 13.5 is small compared to the optimal

pH-value of 8 (18). Finally we calculated the lowest pos-

28

sible concentrations of respectively Rs· and (RSSR) in our

experiments. The actual concentrations might be detectable

by ESR.

We now go over to the discussion of the spin-trap­

measurements.

2.4.2 Spin-trap measurements

2.4.2.1 Irradiated systems:

Interpretation of the adduct spectra

In all spectra of the irradiated standard solutions

we see two adduct signals. One signal consists of three

1:3:3:1 groups with aN 25.96 G, aH= 12.06 G and g0

=

2.00495. The corresponding lines are marked by a point.

This point-spectrum is due to CH 3No; which is formed by

trapping a· or e (3?). The spectrum of CH3No; is given in

fig.2-?. The other signal can be identified as belonging to

the trapped thiyl radical. We first give the characteristic

values of these signals marked by an asterisk (see table 2-2).

In the table are also given all radicals, presumably

generated by UV irradiation of thiols. We shall prove now

that the ESR values of the asterisk signals belang to the

thiyl radical trapped by the aci-ion of nitromethane.

In the table also the ESR values of the ·s--adduct H . -

are given; the spectrum of S-C-NO is shown in fig.2-8. H 2

The ESR values were given by Norman and Storey who generated

·s radicals by rapid-mixing of Na 2s, cn3No2//Ti(III), EDTA//

H2 o2 at pH 9 in a three way mixing chamber (44) and by

Beharand Fessenden who found the s·-adduct spectrum by

photolysing a flow of s 2o;- in the presence of nitrometane

at pH > 9.5 (45) as well as by electron irradiation of

Na2s at pH > 9.3 (3?). We found exactly the same -s·-adduct

spectrum as the authors cited by UV irradiation of Na 2s at

pH> 9.5 in the way mentioned in the experimental section.

29

Table 2-2,g0-valuea and aoupling aonatanta of the asterisk aignala

Substrate at Possible radicals after UV irradiation go ~ ~.!! ~.y ~.a pH = 13.5

: : <j=H3 I

TH3 I 1H3 I - I ·s TH3 . c - CH3 : s - c - CH3 - c - CH3

H" 2.00539 24.70 8.63 - -- I I I s - c - CH 3 CH3

I I CH3 I I and I CH3

I I and I I I CH3

I .s - I I

fig. 2-5a I I I

I I I I I I

H H H I H I H H I H I I H H I I I I I I 1 I I - I I - I S-C-C-NH .c - c - NH21 .C-NH2 1 .c-e-s I "S-C-C-NH 2.00590 24.00 6.40 - 0.75 Il h 2 I I I I I I I I

I I 2 H H H I H H H H

I I and I and I and I - I I I ·s I I

I H I .NH2 H' I - I I 1 I s-e. I I I I I

fig. 2-5b I H I I I I I

.. w -

H H I I

s-e-e-eoo I I H NH

2

fig. 2-5a

H - I s-c-coo

I H

fig. 2-5d

fig. 8

H H I I

.e-e-eoo I I H NH 2

and

H

I I

-I I I I I I I I r I I I I I I

-.l

H I

.c-coo I NH

2

and

H I -.e-s I

H

I I

- I I I I I

H H I I

.c-e-s I -eoo

1 and I I ~ .NH 2 I I I I I

~- ___ I ___ _

I I

I -I I 1 a' I ·s-e-c - eoo

I I H NH 2

H I

.c-coo I

H

and

I I H

I .e-s

I 11

and

I I I I I I

: : H

I I I I I I

I ·s-e-eoo

I

I 1--1 I I

H

- I I I I I I I I I I I I .coo I

I I

g-values are accurate to + 0.00006

: H

I I I I

I .e -

i I NH 2 H

and

.eoo

.i ---------

hyperfine constants in gauss, accurate to :t 0.05 G

2.00570 24.00 6.00

2. 00577 24.10 6.00

2.00590 24.38 6.66 -

0.625 -0.75

0.75

Ho I

-------~----l

1: '3 sec.

llto8 min

Go 5x10 4

Hmo0.5 G

H0 o 3380 G p, 40 mW

liH o100 G freq, 9493.4557 kHz

fig. 2-7, ESR-speatPum of CH 3No;, pH= 13.4

G, 2x10 4 H0 o 3380 G P, 40 mW

Hmo0063 G liH, 100 G freq o9492.3521 kHz

I

-- !

---- ·---~----·- -------'

fig. 2-8, ESR-speatrum of -SCH2No;

of a salution of Na 2s, pH= 13,4

during UV-irradiation

By irradiating the thiols we expect the ·s--adduct

when the s-e bond ruptures. We never saw the -s·-adduct spec­

trum, so we must rule out a breaking of the e-s bond under

generation of the corresponding radicals.

In the adduct-spectrum of t-butanethiol there is no

further splitting of the three 1:2:1 groups. This implies

that the 'eH 3 radical is not trapped and therefore not gene­

rated. eonsequently, its 'partner-radical' se· eH3

eH3

is

not formed. Hence the radical trapped must be the thiyl

radical.

32

In the adduct-spectra of cysteamine, cysteine and

thioglycollic acid we abserve another 1:2:1 triplet splitting

of each line of the three 1:2:1 groups with a coupling con­

stant of 0.75 gauss. This indicates that two equivalent hy­

drogens of the radical trapped are coupling with the unpaired

electron of the N·o; group. For the case of cysteamine the

radical trapped should be either 'CH2 1 its 'partner-

radical' 'CH2

or the thiyl radical, because "NH2 is not

and therefore its 'partner-radical' 'CH2

CH2

S- is

not formed.

For the case of cysteine a similar reasoning yields

that the radical trapped should be either 'CH2 or the

thiyl radical. Por the case of thioglycollic acid the ra­

dical trapped should be either ·cH2 S or the thiyl radical.

By camparing the coupling constants of the spectra

of cysteamine, cysteine and thioglycollic acid significant

differences in the values of the corresponding parameters

can be distinghuished. Hence, the radical trapped cannot

be the 'CH2 S radical of the three thiols. This implies

for the cases of cysteine and thioglycollic acid that the

thiyl radical has been trapped. Por the case of cysteamine

we are now permittod to rule out the trapping of 'CH 2NH2

because its 'partner-radical' 'CH2 S- is not formed.

Moreover, we do not see the nitrogen hyperfine interaction

of the NH 2 group of 'CH2 NH2

(an a~ can be observed in the

adduct spectra of CH 2 =NO; (39).

So, we have to conclude that also for the case of

cysteamine it is the thiyl radical that has been trapped.

2.4.2.2 The rate of production of thiyl radicaZs

We are now able to calculate the rate of production

of the thiyl adduct-radicals from the data in gure 2-6

where the rate of disappearance of the thiyl adduct­

radicals is given after cutting out the UV irradiation.

We assume a steady state in thiyl adduct-radical so the

rate of disappearance is:

33

d{RS •] dt 1013 - 1014 spins/4 s

10-6 - 10-5 mol 1-1s-1

(Veff.liquid cel! 0.15 ml),

and hence the rate of production is between 10-S and 10-6 -1 -1 mol 1 s • Therefore the rate of production of thiyl ra-

dicals is in the range of 10-5 to 10-6 mol/1.

Another way to caLculate the mean maximal rate of

formation of the thiyl radicals in the irradiated systerns is following the adductx-signal intensity as a function of

time during continuous UV irradiation (see 2.3.3.1).

d [RS ·] (RSH]0

or [CH 3N0 2]0

dt period in which adduct signal exists

10-2 or 10-2 molll 5 x 10-6 mol -1 -1 L s 30 x 60 s

The agreement between the results of the two calculations

is satisfactory.

2.4.2.3 CataZytia systems

The rate of formation of thiyl radicals following

the mechanism of Wallace et al in the catalytic system is

For this case, we therefore expect trapping of the thiyl

radicals by the spintrap in the catalysed system. However,

we never saw such a spin adduct signa!.

34

CONCLUSIONS

1. The homogeneaus oxidation of thiols by molecular oxygen

in NaOH (pH= 13.4) catalysed by copper(II)-ions does not

praeeed via free thiyl radicals.

2. The electron transfer and the formation of product con­

sequently will occur in the coordination sphere of a

transient Cu(II)-complex.

SUMMARY

The homogeneaus oxidation of thiols by oxygen with

copper(II) ions as catalyst was studied at pR = 13.4

by measuring the oxygen consumption and the con­

centration of thiyl radicals, the latter by ESR­

rapid-mixing and ESR-spin-trap measurements.

Similar ESR experiments were performed for the

detection of thiyl radicals as obtained by uv­irradiation of thiols.

Thiyl radicals obtained by UV-irradiation were

detected in alkaline media but not in the catalytic

system. From the data thus obtained it is con­

cluded that the catalytic oxidation does not praeeed

via a mechanism invalving free thiyl radicals. The

observation of an ESR signal ascribable to a Cu(II)­

ion coordinated to two or more thiol ligands lends

support to the assumption that the oxidation occurs

via an internal electron shift in the roetalion complex.

35

x Beferences chapter 2

35. E.G. Janzen, Accounts Chem. Res. i• 31 (1971)

36. c. Lagercrantz, J. Phys. Chem. ~, 3466 (1971)

37. D. Behar and R.W. Fessenden, J. Phys. Chem. 76,

1710 (1972)

38. H. Sigel and D.B. McCormick, J. Amer. Chem. Soc.

93, 2041 (1971)

39. D. Beharand R.W. Fessenden, J. Phys. Chem. 76,

1906 (1972)

40. I.H. Leaver, G.C. Ramsay and E. Suzuki, Aust. J.

Chem. 22, 1891 (1969)

41. J.C. Kertesz, Ph. D. Thesis, University of Southern

California (1970)

42. J.P. Barton and J.E. Packer, Int. J. Radiat. Phys.

Chem. ~. 159 (1970)

43. J.E. Packer and R.V. Winchester, Can. J. Chem. 48,

417 (1970)

44. R.O.C. Norman and P.M. Storey, J. Chem. Soc •. (B),

1009 (1971)

45. D. Behar and R.W. Fessenden, J. Phys. Chem. ~,

2752 (1971)

46. F.P.J. Kuijpers and Th.L. Welzen, to be published,

see chapter 3 in this thesis

47. F.P.J. Kuijpers, Th.L. Welzen, A.H. Schoonbeek,

J.F. Timmers, to be published, see chapter 4 in this

thesis

x References 1 to 34 are given at the end of chapter 1.

36

CHAPTER 3

TRANSIENT COPPER(II)-THIOLATE COMPLEXES

THE STRUCTURE OF THE COPPER(II)-DICYSTEINATE COMPLEX *

J.l Intpoduetion and suPvey of the Ziterature

In our investigation of the catalyzed

oxidation of thiols by molecular oxygen in aqueous alkaline

solutions and in the presence of copper(II) ions we

encounterd transient Cu(II)-thiolate complexes (l).These

observations arose out of attempts to establish the mechanism

of the homogeneously catalyzed oxidation of thiols and in

particular of the role of free thiyl radioals in the oxida­

tion process.Rapid-mixing-ESR-experiments and ESR-liquid­

recirculation measurements were consequently performed.During

those experiments we observed transient Cu(II)-signals"As

will be shown these signals can be to Cu(II)-(RS-} x complexes (2~x,4) "Special attention is to the system

where RS is cysteinate,bacause we are particularly interes-

ted in the mechanism of the copper oxidation of

cysteine"As will be described in a paper the

existence of a Cu(II)-dicysteinate complex is

important in this connectiono

We have studied the structure of this complex by per­

forming ESR-measurements at -170°C and visible light absorp­

tion measurements during the oxidation processo

The study of complexes of cysteine with transition

metal ions is nat deeply explored in literatureoMC Cormick

and Gorin studied the reaction of Co(II) with cysteine in

the presence of oxygen as a function of pH (2J.They observed

bis- and triscysteinecobaltate complexes, on the

* a summary is given at the end of the ehapter

37

pH.Models for Co(II)-cysteine complexes in enzymes have been given by Garbett et al. (J).

Srivastava et al. determined the thermodynamics for the for­mation of bis-cysteino-Ni(II)-chelate (4).

Perrin and Sayce reported about a zn3 (cysteinate) 4 complex ( 5).

Recently complexes of Mo(V) with cysteine were discussed by

Kroneck and Spenee in their model studies for molybdenum containing enzymes (6,7).

Kolthoff and Stricks showed the existence of a copper(I)­

cysteinate complex by means of a polarographical metbod (8)"

Perkins reported cu2L and cu2L2 complexes,in which L is

cysteinate (9).Noteworthy is the publication by Klotz et al.

who reported about mixed Cu(I)-Cu(II)-thiol complexes for

thiomaric and thioglycollic acid (10).They did not detect

such a mixed complex in the case of cysteine (10).Lohmann et

al. have studled the behaviour of the Cu(II)-ESR-signal when

adding cysteine toa Cu(II)-solution (11).At a concentratien

ratio of Cu(II) : cysteine = 3 : 5 no further Cu.(II) signal was observed.We shall return to these measurements in a

forthcoming paper.By visible light absorption measurements

Cavallini et al. showed the presence of a Cu(II)-dicysteinate

complex during the copper catalyzed oxidation of cysteine

(12).They also presentedan ESR-spectrum of a frezen Cu(II) + cysteine reaction system,but for the interpretation

of this spectrum they refer to a private communication with

Blumberg and Peisach.Blumberg and Peisach reported about the interpretation of ESR-spectra,recorded at 1.4 K of complexes

of Cu(II) in which two ligands of the type N-N-C-S are

coordinated to copper each by one nitrogen and one sulfur

atom (15J.However they based their interpretation on the a

priori assumption that in the Cu(II)+cysteine complex and in

the Cu(II)+penicillamine complex Cu(II) was coordinated by

two N and two S atoms.Hardly any experimental data are given

for this assumption. To our knowledge no more contributions are known on Cu(II) +

cysteine complexes.Interesting in this field is also the

review by vänngárd about copper proteins,studied by ESR (14).

38

3.2 Experimental part

3. 2.1 Material-a

The thiols examined were L(+)-cysteine,cysteamine­

monohydrochloride and thioglycollic acid.

As copper(II) salt cuso4 .sH2o was used.

The copper(II) ions were complexed with L(-)-histidine when

precipitation of the hydroxyde could be expected (for capper

concentrations greater than 5?10- 4 mol/1) •

All chemieals were obtained pro analiai from Merck and used

as such.

Doubly distilled water was always used.

3.2.2 ESR-meaaurements

3.2.2.1 General

The ESR-measurements were performed on a Varian

E-15,X-band spectrometer with 100 kHz magnet field modula­

tion.The determination of the g-value occurred by using an

A.E.G.-nuclear-resonance magnet field meter in combination

with a Hewlett Packard 2590 B microwave frequency converter,

a Hewlett Packard 5253 B frequency converter 50 to 500 MHz

and a Hewlett Packard 5245 L electronic counter.The micro­

wave radiation had a power up to 200 mW.Saturation did not

occur.

x 3.2.2.2 Rapid-mixing-experiments

The rapid-mixing-ESR-cell is a Varian Associates

accessory number E-249.The dead volume of the cell is

2.5xlo-6 l;the cell volume is 15Xl0-5 l.The flow rate of

both feed streams could be varied between 0.1 and 4.0 ml/s.

The initial Cu(II)-concentration in the experiments

reported here varied between 10- 4 and 5xlo-3 mol/l.Above

ft Theee measurements are also described in ehapter 2

39

sxto-4 mol/1 L(-) - histidine was added to avoid precipi­

tation of the capper hydroxyde by formation of the Cu(II)­

àihistidine complex (15). The concentratien of histidine

was twice the capper concentration, that of thiol 0.1 mol/1.

The measurements were in 0.25 mol/1 NaOH

(pH= 13.4) at 23°C. The liquià pump used was a Fluorocarbon

Saturn pump SPM-100. Its chamber and plunger consisted of

teflon.

A scheme of the apparatus is given in figure 3-1.

-~acuum pump 'N2

02

a y,

b: --~-:

i '

,l-~1 L----

d

d

a reservoirs (11 l b magnetic stirrers c: tlowmeters d: valves e: ES R rap<d-m<xmg-c"l f !iqu<d recircu!ation pump

Fig. 3-1, Saheme of the system.

(for reasans of aonvenienae the whole apparatus was

built on a vehiale).

x 3.2.2.3 Liquid-reciraulation-measurements

The flow cell is an accessory of Varian Associates

(number E-248). The effective volume is lSxlo-51. A line

diagram of the apparatus is given in 3-2. The time

x These measurements are also described in ahapter 2

40

of passaqe between flow cell and reactor is 3 seconds.

The liquid-chamber and the plunger of the liquid purr,p were

constructed of teflon (Fluorocarbon Saturn pump SPM-100).

The initial Cu(II)-concentration in these experiments varied

between 10-5 and 5xl0-3 mol/1. Hhen necessary L(-) - histi­

dine \.;as added.

The thiol concentration varied between 5xl0- 3 mol/1 and

10-l mol/1.

All raeasurements were performed in an oxygen atmosphere ancl

in a solution of 0.25 mol/1 NaOH at 23°c. The oxygen vlas

continuously pumped through the reaction liquid.

3-2~ AppaFatus foF ESR OF VIS liquid-Feeireulation­

measurements

Feasons of aonvenienae this apparatus was built on a

vehicZeJ

3.2.2.4 Ueasurements at -1

Liquid samples from the recirculation measurements

were rapidly transmitted to ESR-sarnple tubes by means of a

syringe with teflon plunger fitted to a Pt/Rh needle

(I!amilton 1005 syrinqe with Kel-F Hub and KF 727 Pt needle),

41

and frozen immediately bere after in liquid nitrogen.The

frozen samples were measured at - 170°C by means of the

Varian low temperature accessory.

3.2.2,5 Absorption measurements in visibte light (VIS)

VIS-spectra were recorded with a Unicam SP. 800

spetrometer,while the reaction liquid was recirculated

between sample holder and reactor.The same apparatus as in

the ESR-recirculation measurements was used,in which the

flow cell was replaced by a VIS-sample holder with variable

optica! path (see figure 3-2).

The raferenee solution was identical to the reaction

liquid .in the absence of copper.The optical path of the

sample and raferenee solutions could be varied between 2.0

and 0.01 cm by use of variable liquid cells.

The concentratien of copper(II) and of thiol were similar to

these in the recirculation-ESR-measurements.

J.J

J.J.l

Results

x ESR-rapid-mixing-measurements

The original copper(II)-signal of the copper salt

or of the Cu(II)-dihistidine complex vanished always after

mixing of the copper stream and the thiol stream (see

figure 3-3).

Transient ESR-spectra obtained by the rapid mixing are given

in figure J-4.Similar signals were detected when operating

in a nitrogen atmosphere instead of under oxygen.The tran­

sient signals vanished immediately after stopping the flow.

x These resuZts are atso given in ahapter 2

42

~ 50 G.

Ho

3-3a~ ESR-speatrum of Cuso 4 .sn2o in so [Cu] ~ -4 2 x 10 mol/l, pH = 13.4

100 G.

H0 , 3300 I

3-Jb, ESR-speatrum of Cu(II)-dihistidine in soZution;

[Cu] = 5 x 10-3 moZ/t, [histidine] = 10- 2 mol/t, pH 13.4

43

44

100 G. -

fig. 3-4a, ESR-speetrum

mixing; [Cu(II)] = 5 x -2 0

5 x 10 moZ/Z, , = 1.5

H0 ,3300 Gauss I

of Cu(II) + eysteine during rapid--3 .

10 moZ/Z, [HOOCCHNH 2CH2SH] = mZ/s, pH= 13.4

H0 ,3300G

100 G I

fig. 3-4b,ESR-speetrum of Cu(II) + cysteamine during

rapid-mixing; [Cu(II)] = 5 x 10- 3 moZ/Z, [NH 2CHéH2SH] = -2 0

5 x 10 moZ/Z, , = 0.5 mZ/s, pH= 13.4

H0 - 3300 G

100G ~~ ~ r ~~ ~~~

/\;\) )1(! \ I I \ V I

11 i

u I

fig. 3-4c, ESR-speotrum of Cu(II) + thioglycollic acid;

[Cu(II}]0

= 1 moZ/Z, [HOOCCH 2SH] = 1 1 mol/l, !2i

3 ml/s, pH= 13.4

3,3,2 x

ESR-Ziquid-recirculation-measurements

With the ESR-liquid recirculation metbod we are able

to detect similar ESR-spectra in the cases of cysteine and

cysteamine, but only when oxygen was admitted to the

reaction system and pumped through the liquid (see figure

3-2).

3,3,3 ESR-measurements at -170°C

The frezen samples, taken at different times from the

copper-cysteine system, all showed an identical signal; this

is given in figure 3-5.

At the end of the oxidation reaction of cysteine (no further

oxygen consumption being measured) the original Cu(II)­

signal reappears though weaker. The ESR-signals of cuso 4 and Cu(II)-dihistidine are given in figure 3-6.

* These results are aZso given in chapter 2

45

46

tOOG. -

H0 :3200G. I

fig. 3-5# ESR-spectPum of Cu(II) + cysteine at -170° C;

[Cu(II)J0

= 10-3 moZ/Z, [HOOCCHNH2CH 2SH] 10-1 moZ/Z

Ho:3200G. I

tOOG g~2.02

fig. 3-6a, ESR-spectrum of Cuso4 at -170° C; [Cu]0

= 2 x 10-4 mol/Z

1\:1:32006. I

fig. 3-6b, ESR-spectrum of Cu(Il)-dihistidine at -170° C;

[Cu(II)) = 10-3 moZ/Z, [histidine] = 2 x 10-3 mol/Z 0

3.3.4 VisibZe Zight-absorption measurements

The VIS-spectra of the Cu+cysteine reaction liquid as

a function of oxidation time are shown in figure 3-7. -1 -1 -1 Absorptions appear at 16700 cm , 23000 cm and 28200 cm o

221 - 2.0

c 0

a: 5 °/o conversion 1.8 ·~ b' 18 '/, c

c' 105'/,

d' 145'/,

16 ~

e, 195'/,

b I I

, ___ ... /a [cu~0 ol85x10- 4 rnalil

[cys 8]0

, 9 o x 10-2 mol !l

1.4

1.2

1.0

0.8

0.6

0.4

L-~3o~o~o~o----------~~--------~2~o~o~oo~--------~15~o~o~o----~o -1 25000

~- wavenumber <cm l

fig. 3-7, VIS-spectra of Cu(II) + cysteine during the

capper cataZysed oxidation of cysteine

3,4 Discussion and concZusions

3,4,1 Transient ESR-spectra

The spectra of the copper salt CuS04 o3H2o, the

Cu(II)-dihistidine complex and the transient spectra of the

Cu(II)/thiol systems are all characterized by four lines

with the intensity ratio 1:1:1:1.

The signals are due to the unpaired electron of Cu(II) (d9

system) ,the four-line splitting being caused by interaction

of the copper-nucleus (I=1o5) with the unpaired electrono

47

The line width of each of the four lines increases with the

magnitude of the magnet field.This relaxation-phenomenon is

well known for Cu(II) and can be explained by slow tumbling

of the complex (16).

The presence of cysteine or cysteamine leads to an additio­

nal hyperfine interaction of five linesoThis superfine

structure is only detectable in the two high field peaks

(eee figure 3-B),presumably because the copper lines are

much more smaller at higher H0

•

·········------------

lOG.

fig. 3-8, The two high field peake of the ESR-speotrum

of Cu(II) + oyeteine in eolution; [Cu(II)-dihistidine] = -2 5 x 1 mol/l, [HOOCCHNR2CR2SH]

0 = 5 x 10 mol/l,

~ = 1,5 ml;s. pH= 13.4

This interaction is most probably caused by the

interaction of two equivalent 14N-nuclei with the unpaired

Cu(II)-electron.The intensity ratio of the five lines should

48

be 1:2:3:2:1 but is difficult to determine quantitatively in

the spectra of Cu(II) - cysteine and Cu(II) - cysteamine

because of the poor resolution.

So,we may conclude that for ligands such as cysteine and

cysteamine Cu(II) is coordinated by at least two thiol

ligands.Hence,this will presumably also be the case for the

Cu(II) - thioglycollic acid complex,because the ESR-spectra

due to the Cu(II) ion are very similar for the three thiols

examined.

We shall now specify the Cu(II) - cysteine complex in

more detail from the ESR-measurements at -170°C and the VIS-

absorption-measurements.

3. 4. 2 Low-temperature ESR-speotra

The ESR-spectra of the frozen samples of the Cu(II)

+ cysteine system do not change with reaction time,except

when the oxidation of cysteine is completed and the original

Cu(II)-spectrum has reappeared.

A complexing ligand like L(-) - histidine does not influence

the ESR-spectrum of the Cu(II)-cysteine system. Conseguently

the ESR-spectra belang to a Cu(II)-cysteine complex.

It appears possible to describe the ESR-spectrum of the

Cu(II)-cysteine complex with help of an axially symmetrie

spin Hamiltonian:

H= g s H 11 z

s z + g (H S + H S )+ A S I + A (S I + S I ) 1 XX yy llzz 1 XX yy

The first two terms describe the Zeeman-interaction vli th

anisotropy in the g-value. The hyper fine splittin0 is

described by the last two terms.

On the basis of the Hamiltonian we expect two ESR-absorp­

tions, centered around g 1 and g1 1

• Each line will be split

into four equidistant lines, caused by the interaction of

the nuclear spin of the Cu(II) ion.

This pattern is clearly observed for the g1 1

-absorption,

three of the four components are visible. The fourth is

overlapped by the g1-peak, which is not split, because Al.

49

is in genPral much smaller than A11

for Cu(II)-complexes

( 1?).

The spectrum of the frozen Cu(II)-cysteine system does not 14

show the N-hyper fine splitting. This disappearance may

be caused by exchange broadening.

He do see in this spectrum an extra absorption at g = 1.95;

(we also see an extra absorption in the spectrum of the

copper salt at g = 2.02). Theoretica! considerations on the

intensity function and lineshape of ESR-signals in polv­

cristalline or glassy materials have shown that for definite

values of g, A and MI extra absorptions will appear (18,19).

This phenomenon is observed in particular for a number of

Cu(II)-complexes (17,19).

By using computer analysis it is shown in the appendix in

accoràánce to the method of Neiman and Kivelson (18) that

the absorption at g = 1.95 for a frozen Cu(II)-cysteine

system is an extra one.

In tabZe 3-1 the values of the ESR-parameters of the

Cu(II)-cysteine complex are given.

50

Parameter Value Reference

go 2.076

gil 2.14

gl 2.04 *

A0

(Cu) 86.5

A (Cu) 190,5

A (Cu) 34.5 **

A0

(N) 10.5

*calculated according to g0

**calculated according to A0

2.13

2.03

202

TabZe 3-1,VaZues of ESR-parameters of the frozen

Cu(II) + aysteine sytem

(13)

(13)

( 13)

In iabZe 3-1 are also given the values reported by Blumberg

and Peisach (13). The agreement is good.

An important feature of the signals described so far is the

absence of absorptions that might be ascribed to Cu Cu

interactions. We are obviously dealing with monoméric Cu(II)

species. This is not entirely surprising in view of the

system's high pH (3}.

i'le now praeeed to the interpretation of the results of

the VIS-absorption measurements in order to further specify

the syrnmetry of the Cu(II)-cysteine complex.

3.4.3 VIS-absorption-measurements

We observed three absorption bands at 16700

(weak band), 23000 cm-1 (shoulder} and 28200 cm- 1

band).

-1 cm

(strong

Because of the Jahn-Teller effect Cu(II) prefers a square

planar coordination or more generally a tetragonallv dis­

torted octahedron coordination (19,20). The appearance of

three absorption bands is in agreement with such a coor­

dination. (20).

Cavallini and co-vJOrkers reported (12) a braad band at

330 nrn ( 30300 crn- 1 ) and a shoulder at 400 nm ( 25000 cJ11- 1 )

for a yellow reaction liquid. Under our experimental condi­

tions the reaction liquid is brown as lon0 as oxygen is ad­

mitted. The pattern observed by Cavallini at al. is siJ11ilar

to our spectrum, but our spectrum is shifted nearly 2000 cm-l

to higher frequencies, compared by that of Cavallini et al.

From the ESR-liquid spectra we know that there are tvlO

equivalent Cu (II) -N bonds. We shall novr identify the other

two bonds frorn ESR-evidence in the literature for

square-planar Cu(II)-L4

cornplexes (L = N,O,S).

In tabZe 3-2 several square planar Cu(II)-cornplexes with

their corresponding ESR-parameters are given. Vle have only

considered Cu(II)-complexes containing two bidentate

ligands, because the bidentate character of cysteine is

well-known (2,3,4,5,6,?,9,12)

51

Table 3-2,Values of EER-parameters of aomplexes~coordinated by 2N and 20,28 and 20,and 48

Ligand Coordination go gl gil Al All ref. (aaus"'-\

o, / N salicylaldoxime Cu 2,11 2.06 2.22 5.0 58.2 17

N/ "'-o

diphenylcarbazide 0'- /N

2.12 2.08 2.20 3.9 50 17 Cu N/ '-o

o, / N salicylamide Cu 2.10 2.08 2.14 3.9 55.7 17

N/ "'-o

0" /N 2.10 2.08 2.14 17.1 180 21 salicyldimine Cu

N/ ""-..o 2.10 2.045 2.20 22.5 198 22

(S)-S-(2-pyridyl- N' /0 ethyl)-L-cysteine

oj'N 2.124 2.055 2.262 30 173 23

I Ligand Coordination gl gil Al All ref. (gauss) (gauss)

maleonitrile- s"'- /s dithiolate C•_i 2.04 2.02 2.08 51 176 24

s/ ""-s 11

s"'- /s -1

diethyldithio- Cu 2.048 2.027 2.089 36.4 172 25

carbonate s/ "'-s 2.051 2.023 2.108 24 163 26

piperidinedithio-s""- /s

2.028 Cu 2.047 2.085 35.8 172 25

carbonate s/ "'-s

i

N-thiobenzyl-N- s""- /0 phenylhydroxyl- Cu 2.076 2.041 2.148 42.8 203 27

amine o/ "'-s

? /N '" L(+)-cysteine Cu 2.076 2.04 2.14 34.5 190.5 this paper

N/ "-..?

Comparing tbe ligand 90ordination of Cu(II)-complexes and

tbeir corresponding gJ-values some tendencies are notewortby.

Copper, square planar coordinated by N and 0 bas g0-values

in tbe range 2.10 - 2.12. Copper, coordinated by four 8 atoms

bas g0-values in tbe range 2.04 - 2.05. Copper, coordinated

by two 8 atoms and two 0 atoms bas tbe same g0-value as in

the Cu(II)-dicysteinate complex.

We may exclude the coordination of Cu(II) by 2N and 20

in the Cu(II)-dicysteinate complex because the g0

-value of

Cu(II)-dicysteinate certainly does not fit in the range 2.10-

2.12 which is found for double N and double 0 coordination.

When Cu(II) is coordinated by four 8 atoms a notable shift

in g0-value from 2.10

0'\ /~ for Cu ' to 2. 045 occurs.

N/ "-o

When Cu(II) is coordinated by two S and two 0 atoms the

g0-value is tbe mean of the former two values. ~ecause tbe

g0-value of Cu(II) is significantly lower tban 2.11 we have

to conclude that copper is coordinated by 8 atoms in a Cu(II)

dicysteinate complex. Moreover, since the g0

-value of Cu(II)­

dicysteinate is relevantly higher than 2.045 we have to con­

clude that copper is surrounded by less than four S atoms

in this complex.

Because from the ESR-Liquid measurements was concluded that

two copper coordination honds are Cu-N bonds, tbe other two

bonds are most prohably ascribable to cu-s honds. The exis­

tence of two Cu-S honds is in agreement with the g0

-value

of the di-thiobenzoyl-N-phenylhydroxylamine-Cu(II) complex

invalving 2S and 20 ligands (27), (see tabZe J-2).

54

Conclusion

During the oxidation reaction of cysteine with molecular

oxygen, catalysed by copper(II) ions, there exists a Cu(II)­

dicysteinate complex in which copper(II) ion is coordinated

to two bidentate cysteinate ions, via N and S-, in a square

planar canfiguration.

Finally we will cansider the cavalency of the capper

cysteine bond. For ionic bands in a square planar Cu(II)

complex the g tensor values are given by Abragam and

Pryce (28).

2(1 + 4À) L\1

and g1

2(1 +

where À is the spin orbit coupling constant (= 828 cm-1 for

free Cu(II)) and 6 1 and 62

are the energy level

separations between d ~ d 2 2 and d ~ d , d . xy x -y xy xz yz 2

For partially covalent bands a correction factor a

can be introduced, leading to the formulae (29,30):

and g 1

2 2(1 + ~,

/1,2 ,

where a 2 expresses the ionic character: a 2 = 1 stands for

complete ionic bonding and a 2 = 0.5 far complete covalent

bonding.

The value of is related to the capper-hyperfine

splitting in the ESR spectrum and is given by the formula

( 28):

- 2) + 3/7(gl - 2) + 0.04,

where P is the coefficient for hyperfine splitting (=0.036 -1 2 cm for free Cu(II)). Calculating a we find the value of

2 -1 0.681. Inserting this value of a and 6 1 = 28200 cm ,

/1.2

= 23000 cm- 1 in the given formulae for respectively

55

g 11 and gJ. we calculated:

2.05.

The agreement with the measured g-values is satisfactory.

It is an interesting and probably more than coincidental

feature that high covalency, i.e. tendency to locate ligand

electrans on the cations, runs parallel with catalytical

activity, i.e. tendency to reduction of this cation.

APPENDrx CHAPTER 3

The appropriate spin-Hamiltonian for a square planar

Cu(II) complex is (30):

ii s z z <ii s x x

<s t + s t > x x y y (1)

The energy corresponding to the Hamiltonian in eq. (1)

can be written as (31,32,33):

(2)

MI is the nuclear spin quanturn number of the copper atom

(MI = ± 1/2, ± 3/2).

So, the field H at which resonance occurs at a frequency

u0

is dependent upon the angle ~. In polycristal line or

amorphous substances the molecules are randomly oriented

and the spectrum is the sum of the resonances of molecules

in all orientations.

56

The number of molecules (dN) whose symmetry axis forms

an angle, with respect to the applied magnotie field,

between ~ and ~ + d~ is given by (31).

dN N /2 sin ~di; 0

where N0

is the total number of molecules

So, dN dH

dN dE;

(3)

(4)

For the case of absence of a nuclear spin eg. (2) yields:

hu (g~ 2 . 2~::) -1/2 H 0

rç + g .l sln , (5)

G hu hu ) o~H

0 E; 2!.---+ H

0 --"';

gli30 gll~o 2

Eliminatien of the angle E; yields a relation between and

H:

dN dH

where H0

11 2) [ ( H /H) 2_ 2], -1/2

gl go o g.l J

The derivative spectrum is proportional to

(6)

This second derivative has a singularity at both extreme

values of H corresponding to E; 0 and ~ = ~· Therefore one

would expect to see a weak 'derivative line' at H=hv0

/g 11 ïT

(~ o) and a streng one at H hv0

/g1S0

(~ = 2) · For the case for which =f.o, H is given by the relation:

(7)

where K ~)1/2.

57

(8)

In contrast to the expression in eq. (6) the angle ~ can

not be explicitly eliminated from this expression for ~:. One must therefore obtain both H and ~: as functions of ~. This has been done for the frozen Cu(II) dicysteine. The

values of the parameters are:

Au Al=

gil

gl

so

go

190.5

34.5

2.14

2.04

4.669

2.076

-1 0.01781 cm

0.003225 cm-1

-5 -1 x 10 cm /gauss

2 4.58 2 2 gil -g~Aro. oo145

g/ 4.16 glAl=0.000043

g H = 0 0

2.076 "o

By putting in H0

= 3400 gauss that is to say g0

H0

7058 we

found the extra absorption to occur for MI = -3/2. The

calculated value was 3577.3 gauss (see table 3-J} which is

in good agreement with the measured g value of 1.95 for the

extra absorption peak.

So the absorption at g = 1.95 is showed to be an extra

one. The corresponding calculations are given in table 3-J

(the computer programme is available on request).

58

Tabte 3-3,Cateutated vaZues of magnet­

ia field and the eorresponding inten-

sity simulation of the ESR-speetrum

of frozen Cu(II)-dieysteinate

Value Nuclear Spin -1.5

Angle Magnetic Field Intensity

0 3565.0 +.1074'-1

.04 3565.1 +.1077'-1

.08 3565.3 +.1087'-1

. 12 3565.7 +.1103'-1

. 16 3566.2 +.1127'-1

.20 3566.8 +.1158'-1

.24 3567.5 +.1200'-l

.28 3568.4 +.1253'-1

. 32 3569.3 +.1320'-l

.36 3570.3 +.1405'-1

.40 3571.3 +.1514'-1

.44 3572.3 +.1657'-1

.48 3573.3 +.1847'-1

.52 3574.2 +. 2110 '-1

.56 3575.1 +.2493'-1

.60 3575.9 +.3093'-1

.64 3576.5 +.4160'-1

.68 3577.0 +.6554'-l

.72 3577.3 +.1665'+0

.76 3577.3 -.2706'+0

.80 3577.0 -.7238'-1

.84 3576.4 -.4112 1 -1

.88 3575.5 -.2844'-1

.92 3574.2 .2160'-1

.96 3572.5 -.1734'-1

1. 00 3570.4 -.1446'-1

1. 04 3567.8 .1239'-1

59

SUMMARY

Rapid mixing and liquid recirculation ESR measurements

of Cu(II) thiol systems in 0.25 mol/1 NaOH showed the

existence of transient Cu(II) thiolate complexes of the

formula Cu(II)-(RS-) with 2~ x~ 4. x For cysteine the coordination of the transient complex

was moreover investigated by visible light absorption

measurements and ESR measurements at -170°C. The

Cu(II) dicysteinate complex appeared to be square

planar, capper being coordinated to two bidentate

cysteinate ions via N and S-.

Referencee chapter 3

1. F.P.J. Kuijpers, to be published, see chapter 2 in this

thesis.

2. B.J. McCormick and G. Gorin, Inorg. Chem. 691 (1962}

3. K. Garbett, G.W. Partridge and R.J.P. Williams, Bio­

inorganic Chemistry l• 309 (1972)

4. S.K. Srivastava, E.V. Ragu and H.B. Mathur, J. Inorg.

Nucl. Chem. 253 (1973)

5. D.D. Perrin and I.G. Sayce, J. Chem. Soc. A 53 (1968)

6. P. Kroneck and J.T. Spence, Inorg. Nucl. Chem. Lett. 9

177 (1973)

7. J.T. Spenee and P. Kroneck, International Conference of

the Chemistry and Uses of Molybdenum, September 17-21,

1973, Reading, Great Britain.

8. I.M. Kolthoff and w. Stricks, J. Amer. Chem. Soc. 73,

1728 (1951)

9. D.J. Perkins, Biochem. J. 55, 649 {1953)

10. I.M. Klotz, G.H. Czerlinski and H.A. Fiess, J. Amer.

Chem. Soc. 2920 (1958)

11. W. Lohmann, M. Momeni and P. Nette, Strahlentherapie

134, 590 (1967)

12. D. Cavallini, C de Marco, S. Duprè and G. Rotilio,

Arch. Biochem. Biophys. 130, 354 (1969)

13. W.E. Blumberg and J. Peisach, J. Chem. Phys.

(1968)

1993

14. T. Vänngard in: 'Biological Applications of Electron

Spin Resonance', Ed. H.M. Swartz, J.R. Bolton and

o.C. Borg, John Wiley and Sons, 1972, p. 411.

15. H. Sigel and O.B. McCormick, J. Amer. Chem. Soc. 22, 2041 (1971)

16. H. McConnell, J. Chem. Phys. ~, 709 (1956)

17. K. Wiersema and J. Windle, J. Phys. Chem. 68, 2316

(1964)

18.

19.

R. Neiman and D. Kivelson, J. Chem. Phys.

(1961)

H.R. Gersmann and J.D. Swalen, J. Chem. Phys.

(1962)

156

3221

61

20. C.J. Ballhausen, 'Introduction to ligand field theory',

Me Graw-Hill, {1962) p.268

21. F.A. Cotton and G. Wilkinson, Advanced Inorganic Chem.

third ed., Interscience Publishers, John Wiley and Sens

(1972) p.556

22. A.H. Maki and B.R. Me Garvey, J. Chem. Phys. 29, 35

(1958)

23. R.H. Fish, J.J. Windle, W. Gaffield and J.R. Scherer,

Inorg. Chem. !l• 855 {1973)

24. E. Billig, R. Williams, I. Bernal, J.H. Waters and

H.E. Gray, Inorg. Chem. ~, 663 (1964)

25. O.M. Petrukhin, I.N. Marov, V.V. Zhukov, Yu.N. Dubrov

and A.N. Ermakov, Russian J. Inorg. Chem. !2• 973

{1972)

26. T. Ramasubba Reddy and R. Srinivasan, J. Chem. Phys.

!l· 1404 {1965)

27. D. Rerorek, R. Kirmse und Ph. Thomas, z. Anorg. Allg.

Chem. 395, 103 (1973)

28. A. Abragam and M.H.L. Pryce, Proc. Roy. Soc. (Londen)

A 206, 164 (1951)

29. A.H. Maki and B.R. Me Garvey, J. Chem. Phys. ~, 31

(1958)

30. R. Neiman and D. Kivelson, J. Chem. Phys. ~, 149

{1961)

31. R.H. Sands, Phys. Rev. ~, 1222 (1955)

32. B. Bleaney, Phil. Mag. 42, 441 (1951)

33. A. Abragam and M.H.L. Pryce, Prov. Roy. Soc. (Londen)

A 205, 135 (1951)

62

CHAPTER 4

MECHANISM OF OXIDATION OF CYSTEINE BY MOLECULAR OXYGEN IN

0.25 MOL/L NaOH CATALYSED BY COPPER (II) IONS *

4.1 Introduetion

In a preceeding paper we proved that the oxidation of

thiols by molecular oxygen in 0. 25 mol/1 NaOH catalysed by

copper (II) ions does not proceed via free thiyl radioals

but that the electron transfer and the formation of the

product could be supposed to occur in the copper coordination

sphere (1). In this paper a detailed mechanism is proposed

that fits the kinetic,analytical and speetral data and that

is based on the previously discussed Cu(II) dicysteinate

complex (2).

To obtain more information about the actual nature of the

oxidation process we have performed further kinetic measure­

ments,quantitative Cu(I) analysis, relative ESR intensity

measurements and product analysis by absorption measurements

in the ultraviolet region.

The investigations on the oxidation of cysteine in alka­

line medium as catalysed by copper(II) ions were started by

Kolthoff and Stricks using polarographical measurements (3).

They found formation of copper(I) cysteinate and cystine. A

review covering the literature till l9GO is given by Tarbell forse'ieral tra;'lsition roetal ions including Cu(II) (4). The

general features appeared to be the independency of the rate

of oxygen uptake of the thiol concentration, the rate of

oxygen uptake being proportional to the oxygen concentratien

in the reaction mixture. The pH-dependency of the observed

rates proved complex. The rates of oxygen consumption were

found to be accelerated by several roetal ions, e.g. Fe(III),

* a summary is given at the end of the chapter

63

Cu(II). Noteworthy is the kinetic study by Taylor et al.

who found a two third order with respect to the iron (III)

concentratien and a zero order with respect to cysteine and

oxygen (5). They observed a marked pH effect with a pro­

nounced maximum in rate at pH = 8.01; the existence of a

iron-cysteinate complex was postulated long before the

latter investigation. The pH-effect was discovered by

Mathews and Walker (6), also for the non metal ion cata­

lysed autoxidation reaction of cysteine.

Very recently Bridgart et al. reported about the oxi­

dation of cysteine by hexacyanoferrate (III)-ions in acidic

media catalysed by small amounts of copper ions (?). From

kinetic measurements they concluded a mechanism involving

thiol complexes of Cu(I-III) and an intermediate believed

to be the radical species (RSSR) or its protonated form

RSS(H)R. A similar investigation was performed fora

reaction in the presence of EDTA by Bridgart and Wilson (8).

Copper catalysis which is dominant in the absence of EDTA

persists but with changes in behaviour which appear to re­

flect the change in the reduction potential of Cu(II) and

the differential lability of the complexes of the II and I

oxidation states.

Cu(I) analysis was performed by Cavallini et al. (9).

In a set of experiments they showed Cu(I) not to be

complexed by neocuproin in an aqueous alkaline solution

when cysteine was present. Therefore they concluded that

during the oxidation process of cysteine all copper was

present in the Cu(II) form. Only at the end of the oxi­

dation reaction Cu(I) could be trapped by neocuproin. We

will comment on their measurements in this paper.

ESR intensity measurements on the system Cu(II) cysteine

have been performed by Lohmann et al. (10). The copper (II)

signal decreased in intensity by addition of thiol. It

vanished at a copper:thiol concentratien rate of 3:5. We

confirmed their results as will be described in this paper.

Product analysis has been performed by many authors and

all designated cystine as the main oxidation product.

However, further oxidation to sulfinic and sulfonic acids

is reported (4,11). The formation of the sulfonic acid is

accelerated in aprotic dipolar solvents (12). For non

aqueous alkaline media Berger showed that disulfide is

formed from unionised thiol whereas the acids originate in

the thiolate ion (13).

In order to campare our results with the mechanism of

Swan and Trimm (14) who used ethane thiol as a model com­

pound for studying the homogeneously catalysed oxidation of

thiols by roetal ions we have performed additional measure­

ments of oxygen uptake and UV absorption for the system

capper + n-butane thiol. According to Cullis et al. (15)

n-butane thiol shows the same kinetic behaviour as ethane­

thiol but is more easily handled because its lower vapour

pressure.

Swan and Trimm gave the follovling mechanism:

* k * * 1. Cu{I) ((SR) 2 ] + o2 - Cu(II) [(SR) 2 (02 )]

* * * **

* ** * 3. Cu(I) [(SR) 2 (RSSR)] ~ Cu(I) [(SR) 2 ] + RSSR

4.

(coordinated atoms are marked with an asterisk).

Swan and Trimm based their mechanism on experimental

rates of oxygen uptake in the presence and in the absence of

strong complexing ligands in the reaction systems as well as

on the observed exclusive formation of disulfide. Because

of the latter feature they excluded an outer sphere mecha­

nism. The overall reaction is:

4RSH + o2 ~ 2RSSR +

65

This mechanism with step 1 being rate determining leads to

a first order dependency in the catalyst concentration. It

is not clear whether they allow Cu(I) to be formed when no

0 2 is present: the mechanism given does not encompass a

possibility herefore.

4.2 ExperimentaZ part

4.2.1 Reagents

Thiols examined were L(+)-cysteine and n-butanethiol.

As souree of capper (II) ions cuso4 ·5H 2o was used. L(-)­

histidine was used at [Cu(II)]0

~ 5.10-4 mol/1 to avoid

precipitation of capper hydroxides by forming the Cu(II)-di­

histidine complex (16). For Cu(I) analysis neocuproin

(2,9-dimethyl phenanthroline) was used.

All chemieals were obtained pro anaZisi from Merck anrl

used as such. Doubly distilled water was always used.

4.2.2 Kinetia measurements

Measurements of oxygen uptake were performed in a

Warburg type apparatus in which the oxygen was pumped

through the reaction liquid. A line diagram of the appara­

tus is given in figure 4-1. It was checked that the rate

of oxygen consumption was not limited by the speed of the

stirrer. Measurements were performed in solutions of 0.25

mol/1 NaOH (pH = 13.4) at 24.5°C, 31°C and 37°C for the

case of cysteine. The initial capper (II) concentrations

varied between 1.0 x 10-5 and 18.5 x 10-5 mol/1. The -3 initial concentratien of cysteine had a value 7.57 x 10 or

4.35 x 10-2 mol/1.

In the Cu(II) + n-butane thiol system the initial Cu(II)

concentratien amounted to 5 x 10-4 mol/1 and the initial

thiol concentratien to 9.3 x 10-2 mol/1. For this system

66

measurements were also performed at a pH of 11.5. The pH­

value was kept constant by adding 8 mol/1 NaOH with an auto­

matically driven burette (see figure 4-1).

f:"1

~recorder ~ j, Ç ~b-urette f NaOH E'j PH-meter ' iaN t1trator

burett

fig. 4-1, Apparatus measurements of oxygen uptake

4.2.3 Quantitative Cu(I) analysis

The determination of the amount of Cu(I) in the

reaction liquid at different times during the oxidation

reaction of cysteine was performed by complexing Cu(I) with

2.9-dimethyl phenanthroline (1?). The samples were taken

from the reaction liquid by using a teflon syringe with a

Pt/Rh needle (Hamilton syringe 1005 with KEL-F hub and KF

727 Pt needle) . The procedure of the analysis was as

follows: transfer a sample with 20-200 ~g copper to a

separatory funnel. Add immediately 10 ml 0.1% neocuproin

solution in absolute ethanol. The mixture colours to

orange. Extract for 30 with 10 ml chloroform and there­

after with 5 ml. Transfer the chloroform layer to a 25 ml

receiver and fill up with a absolute ethanol to the mark. -1

Measure the extinction of the salution at 457 nm (21880 cm )

against a chloroform ethanol reference (ratio 3:2). CuCl

was used as standard. The cells had an optical path length

67

of 10 ~m. The extinction coefficient was 62.9 x 10-S ml/

~g ~m. The measurements were perforrned on a Zeiss M.M. 12

37085 spectrophotometer.

4.2.4 Relative ESR intensity measurements

The ESR measurements were performed on a Varian E-15,

X-band-spectrometer with 100 kHz magnet field modulation.

The microwave radiation had a power up to 200 mW.

Saturation did not occur.

To a Cu(II)-dihistidine salution cysteine was added in

s.mall portions under anaerobic conditions. The decrease

of the relative spin concentratien was followed as a

function of cysteine added unti~ the Cu(II)-dihistidine

signal had vanished.

The measurements were performed in an ESR liquid re­

circulation system as described in a previous paper (1).

The used liquid cell was an accessory of Varian Associates

(E-248). A Fluorocarbon Saturn pump SPM-100 was used. Its

chamber and consisted of teflon.

4.2.5 Qualitative produat analysis

During the oxidation process of respectively

cysteine and of n-butanethiol samples were taken out of the

reaction liquids and brought into 1 mm quartz cells. Rafe­

renee solutions were respectively 0.25 mol/1 NaOH and 0.5

mol/1 NaOH. The measurements were perforrned on a Unicam

SP 800 spectrophotometer.

The reaction liquids were allowed to stand for some days

in an oxygen atmosphere and thereafter UV absorption measu­

rements were performed again.

In order to campare the results additional UV measure­

ments were perforrned in a similar way for the copper cata­

lysed systems of cysteamine monohydrochloride, thioglycollic

acid and S-mercapto propionic acid.

68

4.3 Results

4.3.1 Kinetia measurements

The results of the measurements of oxygen uptake at

24.5°C, 31°C and 37°C are given in s 4-(2a.1,2a.2,

2b,2e). The oxygen uptake is referred toa value of 100%

for the overall reaction

4RSH + 0 2 ~ 2RSSR +

time(minl-

fig. 4-2a.l, Curves of oxygen uptake versus time for the

copper catalysed oxidation of cysteine

time(mml--

fig. 4-2a.2, Curves oxygen versus time for the

capper cata oxidation of cysteine

69

i 150 .. -"'

"' 'g-100

a: ~un] 0 :7.92,1ó 5 mol/l c b: .. :5.27·10-5 ..

c: :2.64·10-5

d d: :106·10- 5

[cySH]0

:7.57,10-3mol/l a

b

4 8 12 time(minl-

fig. 4-2b, Curves of o~ygen uptake versus time for the

aopper aataZysed o~idation of aysteine

fig. 4-2a, Curves of o~ygen uptake versus time for the

aopper aataZysed oxidation of aysteine

From figure 4-(2a.l, 2a.2, 2b, 2a) the following features

appear:

a. The ultimate conversion always exceeds 100%

b. At a certain conversion level (range 110-150 %, dependent

on [Cu]0

) a marked break in the oxygen uptake appears.

Evidently a change in reaction occurs at this moment.

Around this point also a sudden change in the colour of

the solution from red brown to light green is observed;

see also the visible light absorption spectra in the

foregoing paper (2).

c. The secondary reaction seems also to be catalysed by

copper. In an experiment with [Cu] = 1.32 x 10-4 mol/1 -3 0

and [cysteine]0

= 7.57 x 10 mol/1 the oxygen uptake

70

was followed until it stopped. This occurred at a value

of 600% corresponding to a reaction

2CySH + 20 2 -

The reaction after the first uptake around 130% is how­

ever very slow.

The oxygen uptake for the case of n-butanethiol shows

a similar conversion pattern (see fig. 4-6). The secend

reaction appears to be strongly dependent on pH. At

pH= 11.5 it is hardly observable for n-butanethiol (see

fig. 4-6). However for acid the oxygen

consumption increases almost linearly to at least 350% (1)

d. The initial part of the curves in fig. 4-(2a.1, 2a.2, 2b,

2c) and fig. 4-6 is nearly linear in time. Values of

rates of oxygen uptake calculated for this part are given

in tabte 4-1.

In figure 4-3 the rates of oxygen uptake at 24.5°C are

plotted as a function of the copper concentration.

I 'û w !I>

0

E 1.oL

·~~~·I' ~-o L_____, o~

"'~ 0 , 2 4

A [CySH]0

,4 35x10-2mol/l

o [CySH] 0 :7.57x10-3

L . .l ' 6 8 10 20 105x[Cull]

0< mol /I)-

fig. 4-3, Rate of oxygen uptake versus the totat copper

concentratien for the oxidation of cysteine

The course of the curve in fig. 4-3 suggests an order in

[Cu]0

higher than one. The rate is zero order in the cy-

steine concentration, presumably because 0

>> [Cu]0

71

Table 4-l,Rates of o~ygen uptake foP the aopper aatalysed o~idation of aysteine at 24.5,31.0 and 37.0°C

[Cu (II)] 0

x 10-5 mol/1 (d[o 2]/dtl 55 x 10-5 mol/ls

[CySH]0

7.57 mol/1 [CySH]0

4.35 10- 2 mo1/1 24.5°C 31. n I

.0°C = x 1 = x

1. 06 0.07 0.095 0.12

2.64 0.22 0.27 0.29

5.27 0.50 0.55 0.73

7.92 0.80 0.92 1.13

13.20 1. 43 - -----------------------------------------------------------1----------- --------- ----------

0 0.008

l. 06 0.07

2.60 0.22

6.60 0.65 10.5 1.17

13.2 l. 37 15.8 l. 87 18.5 2.10

By plotting -d[02]/dt as a function of the initial [Cu]0

at double logarithmic scales linear relations are found (aee

figure 4-4).

5 4

o.o. x [CySHJ0

, 7.57x10-3rnolll

6 [cyS~0 ,4 35,10-2mol/l

:r. ·~· 1 2 4 6 8 10 102

10\[cu~ l mol/tl-o

fig. 4-4, Plots of rate of oxygen uptake ~ersus the total

aopper aonaentration for the oxidation of aysteine at

double Zogarithmia saales

At all teroperatures the order in copper as derived froro

the slope of the straight lines is equal to 1.2. So we roay

write the 'power rate law':

-d{02]/dt k(Cu]l. 2 0

In figure 4-5 the relation between -d[o2]/dt and 1/T is

drawn for several [Cu]0

• The activatien energy is calcula­

ted to be approximately 5.6 kcal/mol (the activatien energy

of the non roetal ion catalysed autoxidation of ethane thiol

has a value of 16.5 kcal/mol).

73

l 102

~b a :[cun]

0:7.92x10-S mol/I

u b " 5.27x 10-5 .. ..

"' :2.64x1o-5 :::: C: " 0 sor- d: .. : 1.06x 10-5 E ~ ...

~ ~12 "' td02J ~

~' : k Cu

~r dt ss .:::...........

k:k0 xex{~;akt] ~ c :?

10 ~d In~ In k [cun] _LIEakt dt ss 0 o RT

5 101ogld0~ : C x_!_ dt ss T

a: LIEakt :5.5 kcal/mol

b :5.5

c: 4.9

d: 6.5

LIEakt: 5.6 kcal/mot 32 3.3 3.4 1/T-

fig. 4-5, Rate of oxygen uptake versus the reciprooat of temperature; oatoutation of the aotivation energy for the

oopper oataZysed oxidation of oysteine

Finally in figure 4-8 the rate of oxygen uptake for n­butane thiol is given at 23°C and pH = 13.5 and pH= 11.5.

74

1150

.. ..: .. ~ 100

c .. ~ 0

50

x CuS04 .sH20 pH:ll.S

A Cuthistl2 pH :11.5

o CuS04 5H20 pH:136

• Cu(hist l2 pH:135

A auto<~dat<on pH:135

time\mînl---

fig. 4-8, Curves of oxygen uptake versus time for the

aopper cataZysed oxidation of n-butanethiot

4.5.2 Quantitative Cu(IJ anaZysis

During the steady state in the first part of the

oxidation of cysteine our Cu(I) analysis gave a value of

60% of the initial copper concentration. An examp1e of the

extinction measurements for [cysteine]0

and [Cu]0

10-4 mo1/l is given in tab

-3 7.3 x 10 mol/1

4-2. As can be

seen in the table at the end of the first straight line

period of the oxidation reaction (at approximately 130% 0 2 uptake) there is a marked decrease in the Cu(I) concentra­

tion.

TabZe 4-2~Peraentage of [Cu(I)] the aapper

oxidation of aysteine

* Time (min) Oxygen uptake {%) Extinction [Cu {I)] (%)

I

3.5 25 0.352 60.7

10.0 66 0.337 58.3

15.5 100 0.355 61.2

21.0 119 0.347 60.0 32.0 137 0.247 42.7 38.0 145 0.202 34.9

* Based upon the initial capper concentratien [Cu(II)]0 -4 3 10 mol/1; [cysteine]

0 7.3 x 10- mol/1

I

•

From the data in tabZe 4-2 we conclude to a steady state

in Cu(I) of 60% of the initial capper concentratien during

the earlier oxidation reaction (oxygen uptake 110-150%, some­

what dependent on [Cuj0

). This observation disagrees with the

conclusion of Cavallini et al. that capper is preponderantly

75

present as Cu(II) during the oxidation process (9). They

based their conclusion on the following set of experiments:

a. Copper (I) chloride was allowed first to react under an­

aerobic conditions with neocuproine in basic solution and

the typical VIS band of the red complex at 450 ~m could

be recorded.

b. After addition of cysteine still under anaerobic

conditions the peak at 450 ~m disappeared.

c. After admission of oxygen the characteristic band of a

Cu(II)-dicysteinate complex was observed at 330 nm.

d. When in the presence of o2 all cysteine was oxidised, the

spectrum of the red complex of Cu(I)-neocuproine re­

appeared again.

'The rate of the oxidation reaction of cysteine did not

change in the presence of neocuproine and the VIS-spectrum

of the Cu(II)-dicysteine complex was observed just as in the

absence of neocuproine.

Their experimental conditions were: [CuCl] = 10- 4 mol/1;

[cysteine] = 10-3 mol/1; [neocuproine] = 3x10-4 mol/1;

[NaOH] = 0.1 mol/1.

Because they showed that neocuproine is substituted by

cysteine in a Cu(I)-neocuproine complex and that neocuproine

is not coordinatively active as long as cysteine is present

we have to conclude that neocuproine is a weak complexing

ligand for Cu(I) compared to cysteine. Therefore we do not

agree with their conclusion that copper is completely

present as Cu(II) during the oxidation reaction.

In their experiments Cavalline et al. added neocuproine

to the aqueous reaction liquid. In our measurements an

ethanolic solution of neocuproine is added to the reaction

liquid and the immediately formed orange Cu(I)-neocuproine

complex is extracted with chloroform. Thus in alcoholic

medium neocuproine appears to be a very attractive comple­

xing ligand in the presence of cysteine. As is known metal

ion neocuproine complexes are relatively much more stable

in organic than in aqueous solutions (17).

76

4. 3.;, lative ESR intensity measurements

The copper (II) ESR signal decreased in intensity by

addition of thiol.

The course of the relative spin concentratien as a

function of the ratio [Cu]0

/[cysteine]added is given in

figur>e 4-7.

• 10

• 0

0

j__ ' ' 1 _j _j

01 0 05 sp!n concen:raLon (relative un1tsl-

fig. 4-7, Cour>se of the relative spin aonaentration

ver>sus [Cu(II)]0

[ ]added at double

scales

This figure clearly demonstratos that no Cu(II}-dihistidi-

ne signal could be

at a ratio [Cu]0

: [

detected anymore in a ni trog en atmosphere

added = 3:5. These measurements

confirm the results of Lohmann et al. (10).

4.3.4 Qualitative analysis

UV spectra during copper catalysed oxidation of re­

spectively cysteine, n-butane thiol, cysteamine monohydro-

77

chloride, thioglycollic acid and e-mercaptopropionic acid

are given in figure 4-(Ba,Bb,Ba,Bd,Be).

78

/

/

a: immediately alter "' 130 '/o conversion

b alter 20 hours

c : alter 40 hours

[RS~ 0 , 5x10-s M

[cu10

:32x10-5

M

T: 23°C

Ll.--~--~····· _j_----

50000 45000 - wavenumber km-1 I

40000 35000

2.6

I 1.4 c 0

12 ö c:

1.0 >< 01

0.8

0.6

0.4

0.2

0

fig. 4-Ba. UV-speatra of oxidation produats of ayeteine

c 0

a,b,c :during steady state

d,e on !he "nd of the oxidation reaction

1.6 •t; c

u;;

[RSfi0

:9.30x 10-2M

[cu~ 0 : Sx10"4 M

T: 23°C

1.2

10

0.8

0.6

\~--.;;:====d0.4

----------j 0.2

LS~QQ~Q~Q--------- c"~-------4~0~QQ~Q-~~-----3····s··LQ~Q~Q---~o

- wavenumber

"'

fig. 4-Bb. UV-speatra of oxidation produate of n-butane­thiol

c 0 1.4 ·;:; u c

' ' \

a' immediately af ter 100 '/, conversion

b' aft er 20 hours 12 ·~

' \ \ \

a \b \

\ \

\ '

[RSH]0

: 5 x 10- 5 M

[cu~0 , 3 20 x 10-5 M

T, 23 °C

~--

10

0.8

0.6

0.4

0.2

~5~oo~o~o----------~4~5~oo~o~---------4~o~o~oo~--------~3~5o~o~o------~o - wavenumber (cm-1)

fig. 4-Bc, UV-spectra of oxidation products of cysteamine

monohydrachloride

I~ -----~---

50000 !,5000 - --- wavenumber lcm-1 l

a:1mmediately after complete boafter 72 hours

convers1on

[RsH]0

, o 28 M

~u[):r~o' 3.2 x10- 5M

L 23°C

l16 t 114 ~ I -

-,1 2-.::;

~10 ~ ~0 8

ks 1o4

::::---------__ 02

0 40000 35000

fig. 4-Bd, UV-spectra of oxidation products of thio­

glycollic acid

7'2

50000 ..,...._. wavenumber

a: during staady state [RSf-)0,0 3 M. diluted 100x

b:after 20 hours [RSf-) 0 , 5xHf4M

h(Q]Jo = 3 2 x w·5M T 23"C

. L -----40000 35000

oa 06

04

fig. 4-Be, UV-spectra of oxidation products of S-mercapto­

propionic aaid

The UV spectra of respectively di-n-butane disulfide and

cystamindihydrochloride are given in figures 4-9a and 4-Bb. -1

The disulfide has a UV absorption band around 40 000 cm

(18) (see aZso fig. 4-9a,9b). The UV absorption band of -1

sulfonic acid is around 46 500 cm (19,20).

c·)mparing the change in the uv spectra of several

thiols with reaction time indicates that the formation of

sulfonic acid occurs after the conversion of thiol to di­

sulfide is practically completed.

80

------~16 t 1.4

c 0

1.2;:;; u c:

1 0 -; "'

08

6

02

s~o~o~oo~--------~,s~o~o7o~--------~4~o=oo~o-----------~~======:Jo - wavenumber<cm·1)

fig. 4-Ba, VV-speatrum of aystamindihydroahZoride

RSSR 41000 cm1

[RSSR], 8 xl0-4 M

fig. 4-9b, UV-speet~um of di-n-butyZdisulfide

4.4 Discussion and coneZusions

1.6 t 1.4

c 0

1.2 •t; c

10 ~

0.8

0.6

04

We now praeeed to the construction of a mechanism for

the capper catalysed oxidation of cysteine encompassing all

results observed.

Kinetic, analytical and speetral data suggest that two

processes occur, i.e. the relatively fast oxidation of thiol

to disulfide (further called the first process) and the re­

latively slow oxidation of disulfide to sulfonic acid

(further called the second process). The observed UV spectra

show that sulfonic acid is formed by oxidation of the di­

sulfide, so the two processes are sequential.

The curves of oxygen uptake versus time after the break

arenotparallel (see fig. 4-(2a.1, 2a.2, 2b,2c) which

suggests capper to be catalytically active also in the

second process. We have checked this suggestion by platting

the time reduced curves of fig. 4-2(a1,a2,b,e) with respect

to 100% oxygen uptake. The parts of the reduced curves

after the break did nat overlap eachother. So we conclude

the overall rate of the second process has an order with

respect to capper higher than that of the first reaction.

Because Cu(II) coordinates much better to cysteine than to

81

cysteine (2,9) we strongly suggest that only the capper

catalysed oxidation of thiol to disulfide will occur as

long as cysteine is present in the reaction solution. We

seperately measured the rate of the non roetal ion catalysed

autoxidation of disulfide to sulfonic acid. This conversion

is negligible in the considered periods of time. Thus the

initial straight line part of the curves in fig. 4-2(al,a2.

b,a) can be safely considered to represent the capper cata­

lysed oxidation of cysteine to cysteine. As may be concluded

from kinetic, analytical and speetral data in the visible

region (2) the oxygen uptake for the first process continues

to around 130%.

To be certain we also considered the less acceptable

assumption that the two processes occur parallel. We shall

prove that the order of 1.2 in capper for the oxidation of

cysteine to cystine is not influenced by this procedure.

To separate the contributions of the two processes the

following procedure, admittedly approximative is proposed.

Replace the curve, representing the volume of o2 adsorbed as

a function of time by two straight lines. One is the tangent

to the rate of oxygen uptake in the period after the break

point (see figuPe 4-10). This line is extrapolated to zero

time and gives an intercept (A) with the conversion axis.

A line parallel to the abscissa is drawn from A that inter-

sects the perpendicular in C The straight line OC

is now considered to be a measure of the rate of the first

process, AB a measure of the rate of the secend process.

82

1 300

;;- 250

"' -"'

"' a. ::> 200 ---c "' 0>

A >-x 0 150

60 80 100 120 140 160 t1me ( m!n)______.,

fig. 4-10, ExampZe of a correction for the rate of oxygen

uptake for the first process, assuming that both proces­

ses occur simuZtaneousZy; oxidation of cysteine

Correcting the oxygen uptake in the way indicated the

oxygen consumption for the first process still continues

above 100% at a constant rate. The rates of the first pro­

cess are plotted t:ersks copper concentration at doubly loga­

rithmic scales. For all temperatures the slopes do not

significantly differ from the value of 1.2. This value was

also derived when the amounts of oxygen uptake were not

corrected fora parallel reaction (see figure 4-4). So,

whether the sequential processes 1 and 2 occur simultaneously

or not it is permitted to conclude that the rate of the

oxygen uptake for the copper catalysed reaction of cysteine

to cystine has an order of 1.2 with respect to the initial

copper concentration.

We will now present a mechanism of the copper catalysed

oxidation of cysteine to cystine. Thereafter we will take

into account that the oxygen uptake continues to above 100%

without changing its rate.

83

4,4.1 Fi~st p~oaess

The data to account for when constructing a mechanism

for the first process are:

a. The order of -d[02]/dt with respect to [Cu]0

is 1.2.

b. The reaction does nat praeeed via free thiyl radicals (1).

c. A transient Cu(II)-dicysteinate complex exists during the

oxidation process (2.9) in which two cysteinate ligands

are coordinated bidentately via N and S- (2).

d. The Cu(II)-dicysteine complex decays rapidly in a nitrogen

atmosphere (1).

e. By addition of cysteine in small portions to a Cu(II)

salution in a nitrogen atmosphere the Cu(II)-ESR signal

vanished at a ratio [Cu(II)] : [cysteine] = 3:5.

f. There exists a Cu(I}-cysteine complex as appears from

the cited experiments by Cavallini et al. (9). Also

Kolthoff and Stricks give evidence for such a complex (3).

g. There is a steady state in Cu(I) during the oxidation

reaction. The steady state concentratien is three fifth

of the initial Cu(II) concentration.

h, o; is not involved in the reaction sequence (1).

i. The oxygen uptake continues to around 130% at a constant

rate.

j. The formation of product occurs via Cu(II) even befare

oxygen takes part in the reaction (1).

We distinguish two alternative mechanisms. One in which

the disulfide radical anion (RSSR) participates and another

one in which it does nat. We name the types respectively

type A and type B.

An example of a type A mechanism is:

k 2. Cu(II) + 2RS-~ Cu(II) (RS-)

2 k_l

84

k 3. Cu(II) (RS-)

2 ~ Cu(I) (RSSR)-

4. Cu(I) (RSSR) Cu(II) + RSSR + o;-

Assuming steady statesin Cu(II) (RS-)2

and Cu(I) (RSSR)

this mechanism leads to:

-d[o 2 ]/dt = K[Cu(II)] [L]2

, where

K

This mechanism would always have strictly first order be­

haviour in the copper concentration and is therefore excluded.

A variant of this mechanism with stap 3 being

Cu (II) (RS ) 2

k2 ~ Cu(I) (RSSR) also shows first order

k3

with respect to capper because in that case

-d[0 2]/dt = K[Cu(II)] [L]2

, where

A combination of type A and type B mechanism would be:

A + B: l. RSH + OH ::;;;::::!: RS + H20

Cu(II) 2RS - kl Cu (II) (RS - ) 2 2. + ~ -k_l

3. Cu(II) (RS -

) 2 k2

Cu(II) (RSSR) ~ -k -2

85

4. Cu (I) (RSSR) + Cu (II) k3 ~ 2Cu(I) + RSSR

5. 2Cu(I) + 02 k4 - 2Cu(II) + 2-

02

6. 2- + H20 ---20H - + 1/2 02 02

Steady statesin Cu(II) (RS-) 2 , Cu(I) (RSSR) and Cu(I) were

assumed in the derivationof this rate equation. The order

with respect to copper appears to be between 1 and 2. When

this mechanism would apply a plot of [Cu]0/(d[02]/dt)

versus 1/[Cu]0

should give a straight line. Experimentally

a straight line was not observed. Therefore this mechanism

was excluded.

After a discussion of five variations of the type B

mechanism, which is completely given in the appendix, we

found the following mechanism in agreement with all observ­

ations.

1. RSH + OH

II - k1 II 2. Cu + 2RS + Cu (RS-)2 k -1

3. Cu11 (RS-)2

+ Cu11 (RS-)2

~2 2 Cu1 (RS-) + RSSR

2 OH + 1/2 02

86

We will check this mechanism with respect to the following

data:

a. 1.2

k[Cu]0

where [Cu]0

stands for the total copper concentration.

b. steady state in Cu(I) with [Cu(I)] [Cu]

0

3 5

Step (2) cannot be entirely rate-determining since

this would lead to a first order dependency on the total

cu-concentration. Neither could one of the steps (3) and

(4) be taken as rate-determining since this would lead to

a second order dependency on Cu. Because the actual order

in Cu is nearly, but not entirely first order this let us

expect that reaction (2) is the slowest, although commen­

surable with (3) and (4). Because reaction (2) is relativ­

ely slow it cannot be at equilibrium either. Finally,

step (4) must be somewhat slower than (3) in view of the

experimental [Cu(I)]/[Cu(II)J ratio, being 3:2.

We shall now analyse the situation more quantitativ­

ely, assuming steady-states for Cuii(RS-)2

and Cui(RS-).

This leads to

and (1)

d[CUIIL2

] Assume dt to be zero.

87

This yields:

[CuiiL2 ] is a root of equation (2), so we write

- k_l ± {(k-1)2 + 8 klk2[Cuii) [L)2}

Substitution of eq. {3)

d [02] -(--)

dt meas.

4 k2

in (1) and (2) yields:

Equation (4) may be rewritten as:

d [02] -(--)

dt meas.

1/2

(k ) 2 As a first guess we neglect 8 ;~ with respect to

d[02] (--) and

dt meas.

Hence,

88

0 (2)

{3)

(5)

1/2

(k1

[Cuii] [L] 2 ) (6)

II II I [Cu]

0 = [Cu ] + [Cu L2 ] + [Cu ] (7)

Because we have assurned steady-states in CuiiL2

and

Cui we are permitted to write, in accordance to eq. (7):

II [Cu ] = a[Cu]0

; a being a constant (8)

1/2 Substituting eq. (8) in eq. (6), and dividing by [Cu]

0

yields:

1/2 [Cu]

0

1/2

~ k 1 [L]2

a[Cu]0

x

1/2 2

(a k1

[L] )

Platting the left hand side of eq. (9) versus

1/2 [Cu]

0 yields a straight line (see fig.4-11).

(9)

89

rs ., 16

u

"""' ::::_tn -.. 14 0

E

oa, oot corrected s1mulatioo

• b, after iteration

slope curve a, 0.1323 sec-1 -4 112

intereepi y_axis,-3.1x10 !mol/ll5

ec-1

slope curve b, 01374 seé1 112

1otercept y_axis,-3.25x10 41mol/ll5

eë1

s a 10 V g:un]

0 x 103 ! molll

fig. 4-11, Graphs of fa[o 2y ) \ dt ss

[Cu]! [Cu]~ befare

and after iteration; oxidation of aysteine

The value of the slope for the -1

best fit is 0.1323 cm 1 2 2 k

1a[L]

The value of the intercept on the y-axis is

1/2 - 3.1 x 10-4 (mol/Z) s-1 This va1ue is approximative

2 II 2 because we neglected (k_1) /8 k 2 with respect to k 1 [cu ] [L]

Such a negleetien is certainly not permitted for low

capper concentrations. Therefore, we cannot distinguish

between the plus and the minus sign in eq. (9) according

to the derivation given. Hence, we have first to discuss

90

the ± sign in eg. (9) befare correcting this eguation by an

iteration procedure.

We will therefore consider

(d[

(d [

0.2 [Cu]

0 so the value of

increases with the total capper

concentration.

Combining eg. {4) and (8) we can derive:

Hence,

In the right hand

only appears as a

the minus sign in

that the value of

(k )2 8 a[L] 2

--.=.1--=- + 2 --=--'=---- }

[Cu] 0

[Cu]0

(k ) 2 _21 k1 J. [L]2 + -1 + x

8 k 2 [Cu]0

side of eq.(11) [Cu] or [Cu] 2 0

x

( 10)

1/2

(11)

denorninator. Therefore, we have to choose

eq. (11) in order to fulfil the condition

has to increase with the total

capper concentration.

91

We are now able to calculate an approximated value of

from the intercept on the y-axis in figure 4-11.

k_1 2 1/2 ---'"-----:1~/r-:-2 ( ct k 1 [ L ] )

2(2 k2)

2 This yields after substitutionof the value of a k 1 [L] :

2(2 k ) 1/ 2 2

-4 1/2 6.02 x 10 (mol/Z) , hence

This value is small but not entirely negligible. We now cor­

rect eq.(9) by an iteration procedure, inserting the neglect­

ed values of (k-1) 2 in the original equation (4).

~

After iteration the slope of the straight line beoomes for the best fit (see fig.4-ll) 0.1374 s-1 •

-4 1/2 1 The intercept then has a value of 3.25 x 10 (mol/Z) s- . This yields:

-7 3.96X10 mol/Z

1.2 -d[021 So, a mathematica! simulation of dt k[Cu]

0 may be

presented as:

92

d[02] - (--) dt meas.

-7 -4 0.1374[Cu}0

+ 3.96 x 10 - 6.2 x 10

-7 1/2 (0.2748[Cu]

0 + 3.96 x 10 )

mol z-1 s-1

d[o 2 ] The measured and calculated values of-(~)

are shown graphically in figur•e 4-12.

:;: 21.

"' 0 20 E

161

'~1!1 12~

• m<>asur<>d

o calculat<>d

steady state

fig. 4-12, Graph of the measured and eaZauZated rate of

oxygen uptake versus the totaZ eopper concentration;

oxidation of eysteine

The agreement is very good.

Using the datum (Cu(I)] =~[Cu] in combination with eq. (1), 0 2

(2) 1 (3) and (7) we are able to calculate a, k 1 [L] 1 k_1 ,

k 2 and k 3 [o2]. We found the values (for [Cu]

0 = 10-4 mol/l):

a 2.026 x 10-5

k1[L)2 1. 354 x 10-5 s-1

k_1 14.6 x 10-2 s-1

k2 6.705 x 10 3 l mol -1 s-1

k3[o2] 3.03 10 3 z -1 s-1 x mol

93

As an internal check we now can use eq. (3)

(k ) 2 2 1/2

[CuiiL2

] - k1 8 klk2 a[L] [Cu]

0 4k+ (__::.!._ +

2 k 2 2

16 k2 16 2

Substitution of the calculated parameters yields

II -5 -4 [Cu L2 ] = 4.09 x 10 mol/Z (for [Cu]

0 = 10 mol/Z)

II Because the value of [Cu ] = a[Cu]

0 is negligible compared

II to the value of [Cu L2 ] 1 the value of [Cu(I)] will be -5 5.91 x 10 mol/Z. This value is in very good agreement

with the measured value of [Cu(I)] 1 being 60% of the total

copper concentration.

There is still the question why the oxygen uptake

proceeds to values above 100% without changing its rate.

The apparent possibility of the oxygen uptake to increase

above 100% at a constant rate indicates that the presence

of thiyl anions has to be accepted even while the overall

oxygen consumption indicates that they should be completed

converted to RSSR. In other words there must be a reaction

in which thiyl anions are formed from the initial product

RSSR. We have chosen for this reaction the hydrolysis of

the disulfide 1 because the cleavage of the sulfur-sulfur

bond by a nucleophilic attack of the hydroxyl ion is wellknown

(20) 1 especially for the case of cystine (20).

So 1 the complete mechanism of the oxidation of cysteine to

cystine can be represented by:

l. RSH + OH- -+ -<- RS

2. Cuii + 2 RS- t Cuii(RS-)2

3. Cuii(RS-)2

+ Cuii(RS-)2

-+ 2 Cui(RS-) + RSSR

4. 2 Cui(RS-) + 02

-+ 2 Cuii + 02

2- + 2 RS

2- 1 5. 0 2 + H2 0 -+ 2 OH + 2 0

2

- -+ 6. RSSR + OH -<- RSOH + RS

94

The unstable sulfenic acid is catalytically oxidized to sul­

fonic acid as will be discussed later on.

As has been shown before, reaction 3 and 4 are about

equally rate-determining, while the equilibrium in reaction 2

is not attained. Because of the high value of the pH of 13.4

the thiol is completely present in the ionized farm (21).

Reaction 5 is assumed to be fast.

The actual picture of the electron transfer and the formation

of product now may be represented by

fig. 4-13, Model of the electron transfer and formation

of product in the colZieion complex; oxidation of eysteine

The mechanism described can account for all the observed

phenomena. The ratio [Cu(II)]0/[cysteine]added = 3:5 for the

point where no further copper-ESR-signal is detectable in a

uitragen atmosphere can be explained by the hydralysis of the

disulfide as given in reaction 6.

The proposed mechanism differs essentially from the

mechanism proposed by Swan and Trimm (14) in that their me­

chanism postulates that 02

is necessary for the reaction

to occur at all. It has been shown that this is not the

case since reduction of Cu(II) also occurs in the absence

of oxygen. Obviously, this would involve the interaction

of two Cu-complexes which would lead to an order in Cu

higher than one. Swan and Trimm suggest a first order de­

pendency but our data definitely show a higher order.

We agree with Swan and Trimm that the reoxidation of the

roetal ion is rate-determining but we wish to remark that the

formation of product is a concurrent in determining the rate

95

of the overall reaction. It must be remarked that the up­

take of oxygen exceeding 100% was not observed by Cullis

et al (15), and Swan and Trimm (14). Their experimental

conditions were [NaOH] = 2 mol/Z; [RSH] = 0.5 mol/Z. We do

·not have an explanation for this difference in results, be­

cause we have found the limited uptake of oxygen being 100%

for the case of n-butanethiol at pH= 11.5. The pK value of

n-butanethiol is 10.7. An uptake of oxygen till 140 to 150%

is also reported by xan et al (11), butforsmaller ratios

of [RS-] /[OH-] than in Swan's and Trimm's and in our exper-o iments. Finally, we want to remark that hydralysis of the

disulfide, which leads to an oxygen consumption exceeding

100%, is wellknown in alkaline media (20).

96

4.4.2 Seaond prooess

We now try to construct a tentative mechanism for

the second process.

Around 130% of oxygen uptake a break in the curve of

oxygen uptake versus time is observed. After this break the

Cu(II)-dicysteinate complex is no langer present (2) and

the Cu(I) concentratien decays. At the break point a sudden

change in colour from brown (belonging to cu11 (RS-)2

) to

green is observed. These features indicate that the concen­

tratien of thiyl anion in salution is small with respect

to the capper concentratien and the second process may start.

As indicated befare the second process is capper catal-

yzed with an order in capper f~r higher than one. Be-

cause the unstable sulfenic acid now can coordinate with

capper (green colour) sulfonic acid can be formed catal­

Ytically. A strongly tentative mechanism then would be:

1. RSOH - + -+ OH + RSO + H2 0

2. (RS0-)2

+ Cuii(RS0-)2

~

- -+ 3. RS0 2SR + OH + RS0 2 + RSOH

-4. 2 RS0 2 + 02 + 2 RS0 3

5. 2 cu1 {RSO-) + 02 -+ 2 + 2 RSO + 02 2-

6. 2- + H2 0 2 OH - + .!. 02 02 ~

2

To our knowledge the capper catalyzed of cysteine

developing into a capper catalyzed oxidation of cystine

to the corresponding sulfonic acid is not previously re­

ported.

97

Appe.ndi:x;

In the appendix some variations of the type

B-mechanism are discussed. For each variation we have

considered the posibility that the equilibrium in the first

step either is completely established or is not reached.

Furthermore, we always have assumed steady-states in Cu ("II) (RS-)

2 and Cu (I).

B-a

1. Cuii + 2 RS kl ... +

k_l

2. CUII(RS-)2

+ Cuii ~2 2 Cui + RSSR

fast equilibrium: second order with respect to [CuJ0

A plot of (Cu]0

/ versus l/[Cu]0

did not yield a straight line.

Therefore this mechanism was excluded.

B-b

1. Cuii + 2 RS

98

fa st

no equilibrium: -

second order with respect to [Cu)0

2 2 (kl)2 k2[Cuii [L]

(k ) 2 -1

Mechanism B-b was excluded because it leads to a second

order in Cu.

B-e

2. (RS ) 2

3. 2 Cu1 (RS-) + 02

~ 3 2 cu11 (RS-) + O~­

4. 2 ) + 2 k4 ~ 2 Cu11 (RS-)

2 -4

fast equilibrium first step: second order with respect to [Cu]

0

d [ 02] no equilibrium first step: - ~

k1

[Cu (II)] [L]2

= k2 ( k ) -1

2

Mechanism B-e was excluded because it leads to a second

order in Cu.

B-d

l. + 2 RS kl

Cuii (RS-) 2

+ +-k_l

2. + 02 k2 + Cu

11 + RSSR + 2-02

99

2 Cui + RSSR

for fast and no equilibrium:

I

where k 2

A plot of

[Cu(II)){k21

kl + k 3k1 [Cu(II)][L] 2} I

- k 2 - k 3 [Cu(II)) }

versus [Cu]0

did not yield a straight line. Therefore, mechanism B-d

was excluded.

100

SUMMARY

Measurements of the rate of oxygen uptake in 0.25

mol/1 NaOH at different temperatures yielded the

following data:

a. The value of the order with respect to capper

is 1.20.

b. More oxygen was consumed than could be described

by the stoechiometry 4 RSH + o 2 ~2RSSR + 2H 2o. c. A secoud oxidation reaction exists, also catalysed

by Cu(II), in which cysteine is transformed into

higher oxidation products.

Product analysis by UV spectroscopy showed that

cysteine was first oxidised to cystine which was

further on transformed in the corresponding sulfonic

acid. Quantitative Cu(I) analysis during the oxidation

process showed a steady state in Cu(I), the Cu(I) con­

centration being 60% of the original capper concen-

tration, To campare our results with literature

similar oxygen uptake and UV measurements were per­

formed in the case of n-butanethiol, which yielded

a same pattern of results as for the case of cysteine.

Based on the data given a mechanism could be

established for the oxidation of cysteine to cystine

in which the electron transfer and the formation of

oxidation product praeeed via the combination of two

Cu(II) dicysteinate cornplexes. Ligand coordination

to Cu(II) (first order in Cu) and the electron trans­

fer (second order in Cu) are commeasurable in speed:

the order of the rate of oxygen uptake with respect

to Cu is hence 1.2 .

A tentative mechanism for the oxidation of cystine

to the corresponding sulfonic acid is proposed.

101

Referenaes ahapter 4

1. F.P.J.

of this thesis

2. F.P.J.

see chapter 3

3. I.M. Kolthoff

1728 (1951)

, to be published, see chapter 1 and 2

, am: T.L. Welzen, to be published,

of this thesis

and w. Stricks, J. Amer. Chem. Soc. 73,

4. D.S. Tarbell in: 'Organic Sulfur Compounds', vol. 1,

N. Kharash, ed. Pergamon Press, New York 1961, p. 97

5. J.E. Taylor and J.F. Yan Jin-Liang Wang, J. Amer. Chem.

Soc. 88. 1663 (1966)

6. A.P. Mathews and S. Walker, J. Biol. Chem. ~, 21, 29,

289, 299 (1909)

7. G.J. Bridgart, M.W. Fuller and I.R. Wilson, J. Chem.

Soc., Dalton Trans. 1274 (1973)

8. G.J. Bridgart and I.R. Wilson, J. Chem. soc., Dalton

Trans. 1281 (1973)

9a D. Cavellini, C. de Marco, S. Duprè and G. Rotilio,

Arch. Biochem. Biophys. 130, 354 (1969)

b D. Cavallini, C. de Marco and s. Duprè, Arch. Biochem.

Biophys. 124, 18 (1968)

c c. de Marco, S. Duprè, C. Crifè, G. Rotilio and

D. Cavallini, Arch. Biochem. Biophys. 144, 496 (1971)

10. w. Lohmann, M. Momeni and P. Nette, Strahlentherapie

134, 590 (1967)

11. J. xan, E.A. Wilson, L.D. Roberts and N.H. Norton,

J. Amer. Chem. Soc. 63, 1139 (1941)

12. T.J. Wallace and A. Schriesheim, Tetrahedron ~, 2271

( 1965)

13. H. Berger, Reel. Trav. Chim. Pays Bas ~, 773 (1963)

14. C.J. Swan and D.L. Trimm, J. Appl. Chem, ~, 340 (1968)

15. C.F. Cullis, J.D. Hopton, C.J. Swan and D.L. Trimm,

J. Appl. Chem. 335 (1968)

102

16. H. Sigel and D.B. Me Cormick, J. Amer. Chem. Soc. 22, 2041 (1971)

17. A.A. Schilt: 'Analytica! Applications of 1. 10-phenan­

throline and Related Compounds', Pergamon Press 1969

p. 72

18. G. Gorin and G. Dougherty, J. Org. Chem. 241 (1956)

19. Spectrum of benzenesulfonic acid, methylester

(c7 ) , see Sadtler U. V. tables: 4550 U.V. (1960)

20. J.P. Jocelyn in 'Biochemistry of the SH group'.

Academie Press, London 1972, p. 108, 131

21. R.E. Bene.sh and R. Benesh, J. Amer. Chem. Soc. 77,

5877 (1955)

103

CHAPTER 5

OXIDATION OF THIOLS BY MOLECULAR OXYGEN IN ALKALINE MEDIUM

CATALYSED BY VITAMIN B12 (Co(III))

5.1 Introduation

Comparative data for the activities of various transi­

tion metal ions for the oxidation of ethane thiol are given

by Cullis and Trimm (1). An interesting feature is that

activities of various complexes of one and the same cation

can vary widely. For instance, although cabalt ions are

not particularly active the complex vitamin B12 is the

most active catalyst found so far for the thiol oxidation,

Another interesting aspect concerns the stereochemistry of

the catalyst. Cabalt in vitamin B12 has at least five of

its coordination sites occupied by nitrogen ligands, be­

longing either to the corrin ring or to an imidazole ligand

(see figure 5-1), i.e. it can nat bind two thiyl ligands

simultaneously. Consequently, an intramolecular electron

transfer appears impossible and vitamin B12 therefore either

has to resort to a radical type of mechanism, or alternative­

ly any formation of disulfide is always connected with two

vitamin B12 molecules. Consequently, an investigation as

to the presence or absence of free thiyl radicalsduring the

vitamin B12 catalysed oxidation of thiols is highly relevant.

1M

fig. 5-1, Spatial aonfiguration of ayanoaobalamin

5.2 Experimental

Vitamin B12 as cyanocobalamin (c 63H88 coN14o14P) was

used. All chemieals were pro analisi obtained from Merck

and used as such. Doubly distilled water was always used.

As described in a previous paper we performed measurements

of oxygen uptake and ESR-liquid-recirculation measurements

during the oxidation process of respectively L(+)-cysteine,

thioglycollic acid, t-butanethiol and n-butanethiol using

the spin trap nitromethane.

Also ESR rapid mixing experiments combined with spin

trap measurements were performed in a similar manner

as described before (2). The initial concentrations

respectively of vitamin B12 , thiol and nitromethane were

10- 4 - 10-5 , 10-1 and 10-2 mol/1. Measurements were per-

105

formed in 0.5 mol/1 NaOH at 23°C. The apparatus used was

the same as in the investigation of the capper catalysed

reaction (2). ESR rapid mixing measurements were performed

in the absence as well as in the presence of oxygen.

5.5 Resu~ts

The amount of oxygen uptake versus time for

of n-butanethiol + vitamin B12 , where vit. (B 1 ~ 0

the case

10-5mol/l

is given in figure 5-2. The rate of oxygen uptake during

the initia! part of the process (before the break) was

93 x 10-6 mol 1-ls-1 • The rates of oxygen uptake for the

other thiols investigated are comparable to this value.

No ESR signal was ever detected, neither of a thiyl

radical spin trap adduct nor of a Co-complex during the

ESR-liquid-recirculation-spin trap and ESR-rapid-mixing

spin trap measurements. In similar experiments without

the presence of nitromethane also no ESR signa! was

observed.

A most remarkable observation was made when the ESR

cavity containing vitamin B12 and thiol during liquid re­

circulation or in rapid mixing experiments, in the presence

of nitromethane was irradiated by UV light. Only the ESR

signa! of CH 3No 2-was detected and in the normal intensity

but signals from thiyl radical adducts were entirely absent.

Returning to the capper catalysed oxidations we found its

behaviour different: the concentratien of thiyl radical

adducts during irradiation and in the presence as well as

in the absence of capper was similar.

During the kinetic experiments and ESR liquid recircu­

lation measurements a colour change from the original red

to a purple red was observed. At the end of the oxidation

reaction the original red colour reappeared. In the rapid

mixing experiments where much higher thiol concentrations

were used the colour changes from red to purple red brown.

106

x 0

20 30 40

o, v1tamine B12 pHo11.5

A' vitamine 812 pH:13.6

[vit 812} 10-S mol/I

[rL BuSH]~ 9.3x 10-2 mol/I

50 60 110 120 time (min)-

fig. 5-2, Curves of oxygen uptake versus time for the

vitamin 2 oatalysed oxidation of n-butanethiol

5.4 Discussion and aonclusions

Assurning the rate of formation of thiyl radicals, if

actually present to be rate determining we are able to

campare the rate of oxygen uptake with the rate of formation

of thiyl radicals in a non-catalysed standard system. For

this standard system UV irradiation of solutions of thiols

to which the spin trap nitromethane was added was chosen.

The standard solution was identical to the catalytic

system without the presence of vitamin B12 . As described

before, the rate of formation of thiyl radicals in the

standard salution is around 5 x 10-6 mol 1- (2).

Because the rate of oxygen uptake in the vitamin -6 -1 -1 catalysed system is at least 93 x 10 mol 1 S we

2

should expect the detection of thiyl radicals, if actually

present, as adducts of the aci-ion of nitromethane.However,

the UV irradiation experiments at the vitamin catalysed

systems give strong evidence for a very rapid reaction of

thiyl radicals with vitamin B12 . This removal of

radicals might occur by the reaction

107

Co(III) (RS- + RS'- Co (II) + RSSR.

Similarly the following reaction might be conceivable:

Co (III) (RS-) ---Co (II) + RS ••

(Co(II) can be reoxidized to Co(III) by oxygen in a

following reaction) • Therefore we have to be careful to

conclude that the vitamin a12 catalysed oxidation of thiols

does not proceed through free thiyl radicals.

We will now summarise the results obtained for Cu(II)

and Co(III) (vitamin a12 ) catalysts for the oxidation of

thiols by molecular oxygen in alkaline medium.

1. In the Cu(II) catalysed system thiyl radicals do not

play a role. The oxidation occurs via Cu(II)-thiol com­

plexes (shown to be present) in a bimolecular reaction not

invalving oxygen (reaction occurs in the absence of oxygen,

the product being Cu(I) and the kinetics being in agreement

with this supposition). Thiyl radicals, if formed by

radlation are apparently not able to interset with the

Cu(II)-thiol complex.

2. The Co(III) (vitamin B12 J catalysed system occurs

according to a similar but nat necessarily equal mechanism.

Kinetically it is completely equivalent and again it

proceeds via an interaction between two molecules of a

Co(III) thiol complex (not identified), since it also occurs

in the absence of oxygen. However, thiyl radicals produced

by UV irradiation react very rapidly with Co(III) and the

reaction between two complex molecules can therefore be

written as:

Co(III) (RS-)--- Co(II) + RS'

Co(III) (RS-) + RS'~ Co(II) + RSSR

in stead as for the Cu(II) catalysed oxidation

2Cu(II) (RS-) 2 ~2Cu(I) (RS-) + RSSR

108

In the vitamin B12 system 'hidden free thiyl radicals'

(3,4,5) may occur but this is almost certainly not the case

for the Cu{II) catalysed oxidation of thiols.

It is interesting to speculate on the reasons for the

possible difference in reaction mechanism for the Co(III)

and Cu(II) complexes. It is known that there are Cu(II)

complexesin which Cu(II) - Cu(II) bondsexist (6). Such

an interaction in the colloison complex could lead to a

delocalisation of the electron system and therefore to an

decrease of electron density between copper (II) ion and

ligands and also to a decrease of electron density between

the two ligands, i.e. a more radical like character of the

lig.ar:ds in the colloison complex. No such cation-cation

bonds are known for Co(III)-complexes and the mechanism

envisaged therefore seems improbable for Co(III) complexes.

On the other hand any electron donation of a ligand to a

single Co(III) complex is probably supported by delocalisa­

tion of the electron over the corrin ring.

Beferenoes ohapter 5

l.a C.F. Cullis and D.L. Trirnrn, Discuss. Faraday Soc., 46

144 (1968)

b C.F. Cullis, J.D. Hopton, C.J. Swan and D.L. Trimrn,

J. Appl. Chem. 335 (1968)

2. F.P.J. Kuijpers, to be published, see chapter 2 in this

thesis

3. H. Beinert in: 'Flavins and flavoproteins', E.C. Slater

ed., B.B.A. Library, vol. 8, p. 49 Amsterdam, Elsevier.

4. P. Hernrnerich, H. Beinert and T. vänngard, Angew. Chem.

Int. Ed. Engl. 422 (1966)

5. P. Hernrnerich, Proc. Roy. Soc. A 302, 335 (1968)

6. F.A. Cotton and G. Wilkinson: 'Advanced Inorganic

Chemistry', third ed. Interscience Publishers, John

Wiley and Sons, 1972,919

109

APPENDIX I

THE GENERATION OF FREE THIYL RADICALS WITH Ce(IV) AS THE

OXIDISING AGENT IN ALKALINE SOLUTIONS

The technique of generating free radicals by Ce(IV) is

well-known in acidic media. It was also used by Wolf,

Kertesz and Landgraf to study the decay of thiyl radicals

by ESR in aqueous, acidic solutions in combination with a

trigger apparatus (1,2).

Until now the Ce(IV) technique is limited to strongly

acidic solutions. The upper pH limit is 2. Above this

value Ce(IV) is not stable anymore and precipated probably

as the hydroxide. So far, no reports on the detection of

thiyl radicals in alkaline media generated by a chemical

method were available (3}. The direct study of thiyl radi­

cals in alkaline solutions is limited to irradiation tech­

niques such as pulse radiolysis in combination with kinetic

absorption spectroscopy (4,5,6,7) or x-ray radiation under

formation of hydrated electrens and hydroxyl radicals in

combination with ESR (8). This latter technique only yielded

carbon radicals of thiols (8). The transient UV spectra

obtained during pulse radiolysis are difficult to interprete

because of the lack of a reference (5,6,7). With a chemical

method of generating thiyl radicals the reference problem

does not arise because ESR spectra of thiyl radicals in acid

solutions (1,2,3) and in solids (9,10) are well investi­

gated. Therefore we have looked for a chemical method of

generating thiyl radicals in alkaline medium.

One possibility is to use an extension of the method of

Armstrong and Humphreys (11) also applied by Nicolau and

Dertinger (12) who used a Ti(III)~2o2 system to form hydro­

xyl radicals which rapidly attack the thiol under formation

of thiyl radicals. An extension to alkaline solutions is

110

possible when Ti(III) ions are complexed by EDTA. However,

in this case a three way mixing chamber is necessary because

Ti(III)-EDTA, H2o2 and thiol have to be introduced separate­

ly in rapid mixing experiments in alkaline solution. So

far such a rapid mixing cell is not commercially available.

Another possibility is to complex the Ce(IV) ion in a

way that it can be used as oxidising agent in alkaline

medium. After some trial and error experiments with

several complexing agents we found acetylacetone to be a

convenient ligand for Ce(IV) in alkaline medium.

The Ce(IV) acetylacetone complex in aqueous alkaline

salution is a deep yellow coloured colleidal salution that

can be used in ESR rapid mixing experiments to generate

radicals in alkaline solution. With a two way rapid

mixing cell using Ce(IV) acetylacetone in alkaline salution

we were able to detect the spectrum given in figure I-1 for

the case of thioglycollic acid.

P:40 mW

:3380 G

5G

llt: 2min

t:.H:100 G fr~>q: 9491.325 5 kHz

fig. I-1, ESR-speatrum of Ce(IV) + aaetylaaetonate/thio­

glyaollia acid during rapid-mixing; ~ = 1.3 ml/s, nitro-

gen atmosphere, pH ~ 10

This spectrum shows a typical 1:2:1 triplet hyperfine

splitting and is therefore due to a free radical interacting

with two equivalent neighbouring H-nuclei. The -value

111

calculated was 2.00969 and the value of the hyperfine

splitting constant 13.75 Oe. The signal immediately dis­

appeared after stopping the flow.

The ESR spectra of thiyl radicals in acidic media gene­

rated in a rapid mixing system by Ce(IV) or Ti(III)-H2o2 are given in figure I-2(a,b1>b2>a). All signals immediately

disappeared after stopping the flow.

~I g

0:2.o!!'!}

I I I

g:2 .. ~.~

P: 40 mW

G:2.5x10 3

't: 1 sec

At 2min

H0 : 3380 G

Hm:10G

AH: 100 G

treq :9492.366 kHz i

fig. I-2a~ ESR-speatrum of Ce(IV)/aysteine during rapid­

mi~ing; ~ = 2.3 mZ/s~ nitrogen atmosphere~ pH= 0.5

P:40 mW

G:12x104

t :1 sec At: 2 min

H0 :3380G

Hm:0.5G

AH :100 G

treq :9497.040 kHz

fig. I-2b1~ ESR-speatrum of Ti(III)/H202 + aysteine during

r'apid-mi~ing; ~ = 1. 7 ml-/s, n-ltrogen atmosphere> pH = 1. 0 .

112

~.} 9=2.01264 }~

P:40mW G:3.2x10 4

H0 =3380 G Hm: 0.5 G

1::1 sec

ê.H:100G

fig. I-2b2, ESR-speatrum of Ti(II1)/H 2o2 + aysteine during

rapia-mixing; ~ 1.3 m~/s, oxygen atmosphere, pH= 1.0

P: 40 mW

G 5x10 4

1: 3 sec

At: 4min

H0 :3380G

Hm:O.SG

ê.H:100G

freq :9494.786 kHz

fig. I-2a, ESR-speatrum of Ti(IIIj/H2o

2 + thiogZyaol.Zia

aaid during rapid-mixing; ~ = 2.0 mZ/s, nitrogen atmos­

phere, pli = 1.0

The spectra are identical to the spectra observed by Wolf

et al. (1,2) (generation by Ce(IV)); Armstrong and Hurnphreys

( 11) ( generation by Ti ( III) /H 2o2 ) and Nicolau and Dertinger

(12) (generation by Ti(III}/H2o 2 }. The g0-values and

coupling constants are given in tabZe I-1. These values are

in good agreement with the reported values (1,2,11,12).

113

TabZe I-1,g0

and a8 vaZues of thiyZ radiaaZs in aaid soZution

Thiol Oxidising agents Spectr. no. go aH

L(+)-cysteine Ce(IV) I-2a 2.01031 9.23

L(+)-cysteine Ti (III) /H2o2 I-2b1 2.01010 9.30 * L(+)-cysteine Ti (II) /H 2o2 I-2b2 2.01031 9.47 ** Thioglycollic Ti(III)/H2o2 I-2c 2.00960 9.11

acid

* In an oxygen atmosphere two additional signals appear

indicated in fig. I-2b.2 at g = 2.01384 and g = 2.01264.

** The observed extra doublet splitting of the central peak

may be caused by an intermolecular interaction in

accordance to: 0~ H ~c- b

(see also raferenee 13).

- s·

I A,/ OH---'

Although the g0-value of around 2.0100 is not equal to one

third of the sum of the anisotropic g-values of the sulfur

radical being 1/3 (2.003 + 2.025 + 2.053) 2.027 (10,11)

the value of 2.0106 was shown to be characteristic for the

thiyl radical (1,2,11,12). We are hence allowed to conclude that the radical

detected in the alkaline system is the thiyl radical of

thioglycollic acid, a conclusion that is confirmed by the

triplet pattarn of the radical spectrum. The value of the

hyperfine splitting constant is greater than in acidic

media because the relativa electron density on the sulfur

radical is higher for -OOC CH2S-than for HOOCCH 2s-. This is the place to draw the attention to a secend

method to evaluate the rate of decay of thiyl radicals by

measuring their steady state concentratien and their rate

114

of formation. So far, we have used the method of trapping

the thiyl radicals by a convenient spin trap during UV

irradiation of solutions of thiols in aqueous alkaline

solutions (14). However, to apply this method a search

for a convenient spin trap is necessary. We found this

spin trap to be nitromethane (14) and not t-nitrobutane (15).

The Ce(IV) acetylacetonate is strongly yellow, its re­

duction product colourless. Therefore, a measurement with

a stop flow methad should in principle give the rate of the

reaction:

Ce(IV) + RS----- Ce(III) + RS"

Lack of time prohibited a further elaboration of this

method. Evidently the method derived its applicability

from the discovery of the Ce{IV)-acetylacetonate complex.

115

Referenaes appendix I

1. w. Wolf, J.C. Kertesz and w.c. Landgraf, J. Magn.

Resonance !• 618 (1969)

2. J.C. Kertesz: Dissertation, University of Southern

California, January 1970. Dissertation Abstract no.

V313, 1192-B.

3. W.A. Waters in 'Free Radical Reactions', Organic

Chemietry Series One, vol. 10 (1973), p. 282.

4. G.E. Adams, G.S. McNaughton and B.D. Michael, Trans.

Faraday Soc. 64, 902 (1968)

5. M. Sirnic and M.z. Hoffrnan, J. Arner. Chern. Soc. 92

6096 (1970)

6. M.Z. Hoffrnan and E. Hayon, J. Arner. Chern. Soc. 94

7950 (1972)

7. M.Z. Hoffrnan and E. Hayon, J. Phys. Chern. 12, 990

(1973)

8. P. Neta and R.W. Fessenden, J. Phys. Chern. 75, 2277

(1971)

9. Y. Kurita and W. Gordy, J. Chern. Phys. 34, 282 (1961)

10. Y. Kurita and W. Gordy, J. Chern. Phys. 2!1 1285 (1961)

11. W.A. Armstrong and W.G. Hurnphreys, Can. J. Chern. 45

2589 (1967)

12. C. Nicolau and A. oertinger Radiat.Res. 42, 62

(1970)

13. J.C. Kertesz, w. Wolf and H. Hayase, J. Magn. Reso­

nance, 22 (1973)

14. F.P.J. Kuijpers, to be published, see chapter 2 in

this thesis

15. J. Zwart, graduate report, Eindhoven University of

Technology, Department of Chemistry, Labaratory for

Inorganic Chernistry and Catalysis, October 1973

116

APPENDIX II

ATTEMPTS AT QUENCHING OF o; RADICALS IN CATALYTIC REACTION

MIXTURES BY THE BRAY RAPID FREEZING TECHNIQUE

When a reduced transition metal is reoxidized by

molecular oxygen two alternative reactions paths are

possible:

1. Me(n-1)+ + o2 ---Me n+ + o;

Me(n-1)+ + ---oo-Men+ + 2-02

2. 2Me(n+l)+ + o2-2Me n+ 2-

+ 02

These paths involve transfer of one and two electrans in

To distinguish between the two one step respectively.

pathways it is necessary

the capacity to detect

free radical o; is ESR.

to develop a technique which has

An obvious methad to detect the

However, the rate of decay of o; is rapid and therefore it is not possible to detect o; in

reaction mixtures directly by ESR. Consequently, the

measurements have to be performed at optimal o; concentra­

tien (ESR rapid mixing) or by reducing the rate of decay

of o; (addition of spin trap (1), addition of Vycor glass

(2) or rapid freezing). We have chosen the rapid freezing

technique by which radicals are quenched in isopentarre at

-140°C and measured by ESR at this low temperature. This

technique has been developed by Bray (3) and modified by

Beinert and Palroer (4) and by Ballau (5).

In our investigation we have used a simply modified

Bray rapid freezing technique (3). A line diagram of the

apparatus is given in fig. II-1.

117

O.Sml

a: thermocouples

110 m/sec gtass bar

outside:; 2mm _i.!l§ide:il1mm

cappiUary

ins i de: fJ 0,2 mm

!.i<LIIII!l'!!ler

dew_<!~

ESR _tube ~~liquid (inside:-35mml

fig. II-1, Line diagram of the quenah-apparatus

The used concentrations of respectively capper (II) and

thiol in the separate reaction streams were 10-3 mol/1 and

2 x 10-2 mol/1. The concentratien of NaOH in bath streams

was 0.5 mol/1. The separate reaction products were trans­

ported into the mixing pipe line by manually pressing of the

plungers in the syringes at a constant velocity.The plungers

were constructed out of teflon; the Y-part was made of

glass. The squirting mouth of this pipe was a conus in

order to avoid mixed streams with a diameter longer than

that of the squirting mouth which was 0.2 mm. The speed

manually obtainable was 10 m/s Bray reported an optima!

speed of 31 m/s . with respect to the size and the hardness

of the frozen particles (J) but for our purposes a speed

of 10 m/s appeared to be satisfactory. The mixed stream

was pressed into isopentane which was kept at -140°C by

cooling with nitrogen gas of -180°C. The temperature of

the isopentane could be regulated by the flow of the nitro­

gen gas. The temperature in the dewar with isopentane was

measured at three points, as indicated in fig. II-1 by

means of tnermo couples. The frozen crystals were sampled

118

in a funnel and pressed with a glass bar into an ESR tube

(~ inside = 3.5 mm) constantly operating in isopentane at

-l40°C. According to Bray the isopentane was chosen be­

cause isopentane is a liquid at -l40°C and becomes solid

at -l60°C. In such a liquid the heat transfer from frozen

particles to the neighbouring liquid does not cause evapo­

ration of the liquid which would give thus a great harrier

for heat transfer.

As a test case for the apparatus described to quench o;­radicals we have chosen the generation of o; radicals by

reaction of KI0 4 and H2o2 (6) in a solution of 0.5 mol/1

NaOH. The spectrum detected on the quenched particles of

the reaction mixture is shown in figure II-2.

g o2.123

go2.006

fig. II-2, ESR-spectrum at -180° C of a sorution of KI0 4 + H2o 2 in 0.5 mor/r NaOH, quenched by rapid-freezing

This 2 g value spectrum with g11 = 2.123 and g.l= 2.006

is undoubtedly due to o; (6,7,8). The catalytical system

119

was imitated by mixing Cu(II) and RS streams, bath oxygen

saturated and quenching in the way described. No o;-signal

was ever detected in frozen species of these catalytical

systems. The absence of this signal cannot be due to a

short lifetime of the radicals because the mean lifetime is

reported to be more than 200 ms (6,7,8). So if the concen­

tratien of o; in the frozen samples is high enough

(~5 x 10-8 mol/1) we should detect these radicals by ESR

if actually present. We can calculate the concentratien

of o; in the reaction salution by assuming a steady state

in o; and the formation of o; to be rate-determining.

[0-2 ] --=--=s~s___ (1 )

(d[o; ]/dt)disappearance (d[o; ]/dt)formation

' x (d[02]/dt)formation -3 -4 200x10 x10 mol/1

2 x 10-5 mol/1

So the lowest concentratien of [o;] in the frozen samples

will be given by

x ótquenching ( 2 )

The quench time is the sum of the mixing time and the

freezing time. The freezing time is the sum of the cooling

time and the time of crystallisation of the frozen species.

The crystallisation of the particles occurs around -30°C;

the time of this process is generally twice the cooling time

(9). The mixing time is given by the sum of the dead

volume of the Y-part, the volume of the mixing pipe line

and the volume of the apparent liquid cylinder from the

120

squirting mouth to the surface of the isopentane, divided

by the velocity of the mixed stream up to the surface of the

isopentane. So, the mixing time is in the order of

2.5 x 10-3 s. Bray reported a quench time of 10 to 20 ms,

measured by using the intensity of the colour of the

reaction ofiron(III) with rhodanide as a standard (3). By

using this method the crystallisation time is not taken into

account. Therefore the real quench time will be in the -3 * range of 50 x 10 S .

To be certain the cooling time was also estimated by

using the differential equation for non-instationary

heat transfer of a partiele with a spherical shape (9).

This yielded in our hands a freezing time of 15 x 10-3 S

(10) and hence a quench time in the range of 50 x 10-3 S.

Inserting this value in equation (2) the concentratien

of a; in the frozen samples is calculated to be 15 x 10 -6

mol/1. This concentratien of a; radioals should be detected

by ESR. Moreover, the actual value will be higher because

we have neglected the rate of formation of a; radicals

during the freezing process and only considered the upper

limit of the rate of disappearance.

So we have to conclude that o; radioals if actually

present in the catalytic systems should have been detected

by ESR after rapid freezing in the way described.

Since this was not the case we conclude that o; radicals

are not involved in the copper catalysed oxidation of thiols.

121

RefePenoes Appendi~ II

1. F.P.J. Kuijpers, to be pub1ished, see chapter 2 in

this thesis.

2. Y. Fujita and J. Turkevich, Discuss. Faraday Soc. no.

41, 407 (1966).

3. R.C. Bray in: 'Rapid Mixing and Sampling Techniques in

Biochemistry', B. Chance, R. Eisenhardt, Q. Gibson and

K. Lonberg-Holm Eds., Academie, New York, 1964.

p. 195.

4. G. Palroer and H. Beinert in: 'Rapid Mixing and Sampling

Techniques in Biochemistry', B. Chance, R. Eisenhardt,

Q. Gibson and K. Lonberg-Holm Eds. Academie,

New York 1964, p. 205.

5. D. Ballou, Ph.D. thesis

6. P.F. Knowles, J.F. Gibso·n, F.M. Piek and R.C. Bray,

Biochem. J. !!!• 53 (1969)

7. R. Nilsson, F.M. Piek, R.C. Bray and M. Fielden, Acta

Chem. Scand. 2554 (1969).

8. W. Orme-Johnson and H. Beinert, Biochem. Biophys. Res.

Commun. 36, 905 (1969)

9.a R.B. Bird, W.E. Stewart and E.N. Lightfoot/ 'Transport

Phenomena', chapter 8,9 and 11.

b H.S. Carlslaw and J.C. Jaeger: 'Conduction of heat in

solids'

10. A.M. Edelbroek, graduate report, Eindhoven,

University of Technology, Dept. of Chem. Eng.,

Laberatory of Inorganic Chemistry and Catalysis,

November 1972.

122

SUMMARY

This thesis reports on an investigation into the

mechanism of the oxidation of thiols by gaseaus oxygen in

alkaline solution at roomtemperature under the influence

of Cu(II) or vitamin B12 catalysts.

The literature on the subject mentions two types of

mechanistic proposals, one of which involves the aceurenee

of free radicals while the other envisages electron trans­

fer in coordination complexes. So far, thiyl radicals have

been observed in acid media but not in alkaline solutions.

Whether free thiyl radicals could be detected in the homo­

geneaus catalytic oxidations mentioned was investigated by

ESR-rapid mixing and ESR-spintrap methods both in combi­

nation with measurements of the rate of oxygen consumption.

These studies led to the proof that thiyl radicals are not

involved in the Cu-catalysed oxidation in alkaline media.

It was shown in a separate study that thiyl radicals could

be detected in alkaline media when generated by ultraviolet

light; the detection being accomplished by spintrap methods

under conditions where their rate of formation was of the

same order as expected for the catalytic reaction (chapter

1 and 2).

During the ESR-rapid mixing experiments transient

signals were observed that could be ascribed to Cu(II)­

(thiolate)x complexes with 2 ~ x ~ 4 (chapter 3). The struc­

ture of the Cu(II)-cysteinaat complex was further inves­

tigated by visible light absorption measurements and ESR­

studies at -170°C. The complex was found to contain two

bidentally bonded cysteinate ligands, connected to Cu(II)

by N and S in a square planar configuration (chapter 3).

The mechanism of the copper-catalysed oxidation of

cysteine was investigated in details from its kinetics.

The electron transfer and the formation of the product ap­

pear to occur in a bimolecular reaction of two Cu(II)­

dicysteinate complexes. Oxygen serves only to reoxidize the

123

so formed Cu(I)-cations. Formation of the transient complex,

interaction of two complexes and reoxidation of Cu(I) are of comparable rate. The mechanism proposed accounts for all

observations made so far (chapter 4).

In the reaction catalysed by vitamin B12 two reaction

mechanisms, one via free thiyl radicals, the other by elec­

trontransfer between two complex molecules, still remain

possible. The uncertainty sterns from the observation that

free thiyl radicals generated by UV-light could not be

observed any more in the presence of vitamin s 12 , contrary to the case of Cu(II) which presence does not influence the

radical concentratien (chapter 5) •

Appendix I gives a description of the generation and

the direct ESR-detection of thiyl radicals in alkaline

solution via the interaction of Ce(IV) + acetylacetonate

and thiol in a rapid mixing system, the metbod being an

alternative for the spintrap method.

Appendix II gives an experimental proof via the "rapid

freezing" metbod in combination with ESR-measurements that

o; radicals are not present and therefore do not participate

in the catalytic oxidation of thiols by oxygen in alkaline

media.

124

The work described in this thesis was supported in

part by the Nether~ands Foundation for Chemiaal

Research (SON) with financial aid from the Netherlands

Organization for the Advanaement of Pure Reaearah(ZWOl.

SAL"1.ENVATTING

In dit proefschrift wordt een onderzoek beschreven

naar het mechanisme van de oxydatie van thiolen door mole­

kulaire zuurstof, gekatalyseerd door koper(II)-ionen of

vitamine

ratuur. 2

, in sterk alkalisch milieu en bij kamertempe-

Er bestaat in de korresponderende literatuur een

dilemma over de vraag of de oxydatie reaktie verloopt via

vrije thiylradicalen. Het bestaan van vrije thiylradicalen

in zuur milieu is bekend, in basisch milieu zijn zij echter

tot nog toe niet waargenomen. Of vrije thiylradicalen in de

genoemde homogeen gekatalyseerde oxydaties aangetoond konden

worden, werd onderzocht met behulp van de ESR-rapid-mixing

methode en de ESR-spin-trap methode, beide in kombinatie

met metingen van de zuurstofopname-snelheid. Via deze aanpak

kon worden bewezen, dat vrije thiylradicalen geen rol spelen

in de door koperionen gekatalyseerde oxydatie van thiolen in

basisch milieu. Thiylradicalen konden, middels hun spin-trap

adduct, wel aangetoond worden in basische thioloplossingen

bestraald met U.V.-licht, waarbij de vormingssnelheid verge­

lijkbaar was met de snelheid van de snelheidsbepalende stap

in het gekatalyseerd systeem (hoofdstuk 1 en 2). Tijdens de

ESR-rapid mixing experimenten werden transient signalen waar­

genomen, welke toegeschreven konden worden aan Cu(II)­

(thiolaat)x complexen (2 ~ x s 4) (hoofdstuk 3). De struktuur

van het Cu(II)-(cysteinaat)x complex werd onderzocht met be­

hulp van metingen van de zichtbaar licht absorptie en de elec­

tronenspinresonantie metingen bij -170°C. Het complex bleek

opgebouwd te zijn uit twee cysteinaat liganden, welke biden­

taat aan Cu(II) gebonden zijn, via N en S in een vlak vier­

kant omringing (hoofdstuk 3) •

Het mechanisme van de door koper-ionen gekatalyseerde

oxydatie van cysteine is via de kinetiek van de reaktie in

detail onderzocht: de electrenenoverdracht en de vorming van

het produkt cystine blijken te verlopen in een bimolekulaire

125

reaktie van twee Cu(II)-dicysteinaat complexen. zuurstof

is alleen noodzakelijk om het gevormde Cu(I) te reoxyderen.

In het mechanisme zijn de vorming van het transient com­

plex, de produktvorming, en de reoxydatie van Cu(I) on­

geveer gelijkelijk betrokken in de bepaling van de reak­

tiesnelheid. Het voorgestelde mechanisme verklaart alle

tot nog toe bekende gegevens (hoofdstuk 4).

In de door vitamine B12 gekatalyseerde oxydatie

van thiolen blijven twee mogelijkheden bestaan: de vorming van het reaktieprodukt via de bimolekulaire koppeling van

twee vitamine B12-monothiolaat complexen en de reaktieweg

via vrije thiylradicalen (hoofdstuk 5). De onzekerheid stamt uit de waarneming dat door U.V.-licht gevormde radicalen

in aanwezigheid van vitamine B12 zo snel reageerden, dat

zij niet meer gedetekteerd konden worden, terwijl in aan­wezigheid van Cu(II) hun koncentratie niet wezenlijk ver­anderde.

In Appendix I wordt de generatie en de direkte ESR­

detektie van thiylradicalen in basisch milieu met behulp

van (Ce(IV) + acetylacetonaat) en thiol in een rapid-mixing­

system beschreven. Deze methode is een alternatief van de

ESR-spin-trap methodiek.

In Appendix II wordt met behulp van een snelle in­

vries methode in kombinatie met ESR-metingen bij -l80°C

aangetoond, dat 02 radicalen niet aanwezig zijn in de homo­

geen gekatalyseerde oxydatie van thiolen door molekulaire

zuurstof in basisch milieu.

126

Het onderzoek beeahreven in dit proefschrift werd

mogeZijk gemaakt door finanaiëZe steun van de Neder­

Zandse Organisatie voor Zuiver-Wetenschappelijk­

Onderzoek (ZWO) via de Stiahting Saheikundig Onderzoek

NederZand (SON).

DANKWOORD

De auteur wil allen bedanken die aan de tot stand­

koming van het in dit proefschrift beschreven werk hebben

bijgedragen, met name prof.dr. G.C.A. Schuit, en de af­

studeerders ir. Th.L. welzen, ir. A.M. Edelbroek,

Mej. A.H. Schoonbeek (H,B.O.), de heer J.F.Timmers (H.B.O.),

ir. J.Zwart en ir. H.J.K.M. Duijkers. Prof.dr. W. Drenth

van de Rijksuniversiteit Utrecht dank ik voor de discus­

sies en zijn kritische beschouwing van het manuscript.

Veel dank ben ik verschuldigd aan de heer W. van Herpen

voor het maken van de tekeningen. Mevr.Th.de Meijer-van

Kempen en Mej. M.den Dekker dank ik voor het typen van het

manuscript.

De glasinstrumentmakerij van de T.H. Eindhoven ben

ik zeer erkentelijk voor hun jarenlange accurate en snelle

service. De reproduktiedienst van de T.H. Eindhoven wil ik

bedanken voor het fotograferen en afdrukken van de tekenin­

gen. De T.H. Eindhoven en de Stichting S.O.N.-Z.W.O. dank

ik voor de financiële steun, welke het mij mogelijk maakte

de "Summerschool in Theoretical Chemistry" van prof.dr.C.A.

Coulson F.R.S. te Oxford te volgen, evenals het congres over

metallo-enzymes te Oxford (september 1972) .

In mijn dank wil ik tot slot mijn ouders en mijn echt­

genote betrekken voor hun waardevolle steun tijdens mijn

studie en werkzaamheden.

F.P.J.

127

LEVENSBERICHT

Frans Kuijpers werd geboren op 25 maart 1947 te

Roermond. Na het behalen van het getuigschrift HBS-B aan

het St.Maartenscollege te Maastricht begon hij in septem­

ber 1964 met de studie voor scheikundig ingenieur aan de

Technische Hogeschool Eindhoven. Het ingenieursexamen werd

afgelegd met lof in januari 1970. Vanaf 1 mei was de auteur

in dienst van z.w.o. - S.O.N., eerst als doctoraal assis­

tent en vanaf 1 januari 1972 als wetenschappelijk ambtenaar.

Naast zijn promotie-werkzaamheden verrichtte hij in samen­

werking met drs. Barry H.van Vught, drs. N.J.Koole en

prof.dr.W.Drenth van de Rijksuniversiteit Utrecht onderzoek

aan de oxydatie van cyclohexeen door molekulaire zuurstof,

gekatalyseerd door tris (trifenylfosfine) rhodium(I).

Een tiental SVIII-praktikanten en afstudeerders van

de T.H.E. en twee afstudeerders van het H.B.O.-Eindhoven

werden door de auteur gecoached. De auteur nam aktief deel

aan de organisatie in de groep anorganische chemie.

Van medio 1973 tot medio maart 1974 verrichtte de

auteur onderzoek aan de katalytische oxydatie aktiviteit en

werking van geimmobiliseerde coordinatie-complexen aan de

University of Delaware, Newark, u.s.A. Dit onderzoek werd

financiiel gesteund door de N.A.T.O. (research grant no.695).

Last, but not least, op 12 mei 1972 trouwde hij met

Rikie Pauw, en op 28 april 1973 werd hun dochter Tineke ge­

boren.

128

Stellingen behorende bij het proefschrift van

F.P.J. Kuijpers,

9 april 1974

1. De bewe ring van Fusi et al. dat a -naftol en hydrochinon

äe oxydatie van cyclohexeen door moleculaire zuurstof

in benzeen,gekatalyse erd door t ris(trifenylfosfine )chloro­

rhodium(I) , r emmen in hun functie van radicaalvange r is

o njuis t .

- A. Fusi,R. Ugo,F. F o x,A. Pasini and S. Cenini,

J. Organometal. Chem. ~,417 (1971)

- Barry H. van Vu g t,N.J. Ko ole and W. Drenth

and F.P.J. Kuijpers,

Ree l . Tr av . Chim . Pays-B as , to b e p ublished

2. De oxydatie van thiolen door moleculaire zuursto f,gekata­

lyseerd d oor kope r-ionen in sterk alkali sch milieu,ver­

loopt niet via vrije thiylradicalen.

- Hoo fd st uk 1 e n 2 in dit proefschri ft

3 . De conclusie va n Ca v a ll ini et a l. dat Cu(II) i n het

Culii)-dicyste inaat complex omringd is door 2 S en 2 N

atomen van twe e bidentaat coordinerende cys teine liganden

berus t noch o p experime ntele noch op t heor e tische

g eg e vens.

- D. Ca vallini, C . de Mar co , S . Du pr è a nd

G. Rat i 1 ia,

Ar e h. Bi o chem. Bi o ph y s . .!lQ, 354 (1969)

- Hoof ds tuk 3 in dit pr o ef schrift

4. De toewijzing van d e nobelpri j s voor de v rede l i jkt een

po l itieke z aak te worden .

5. De biochemische theorie over "hidde n free radicals" in

metaa lion-disulfide clust ers dient te worde n herzie n

voor kop erionen .

- H . Bein e r t i n "F l a v i n s a nd f l a v o protein s ", E . C.

Slat e r, ed . , B . B .A. Li br a r y , vol . B, p . 49,

Amsterdam Elsevier

- P. Hemmerich,H. Beinert and T. Vänngard,

Aogew. Chem. Int. Ed. Eog1. 2,422 (1966)

- P. Hemmerich,

Proc. Roy. So c. A 302,335 (1968)

- Hoofdstuk 4 en S in dit proef s chrift

6. De direkte detektie van thiylradicalen in basisch milieu

m.b.v. elektronenspinresonantie blijkt mogelijk.

- Appendix I in dit proefschrift

7. Het verdient aanbeveling wegg ebruike r s te bekeuren ,die

op een dergelijke manier door plassen rijden dat zij

andere verkeersdeelnemers een onwelkome douche bezorgen.

8 . In de huidige theorieën over de s tralings besche rmende

werking van bepaalde aminothiolen in weefsels wordt ten

onrechte geen r ekening gehouden met he t bestaan van he t

disulfide radicaal anion.

- G. Caspari and A. Graozow,

J. Phys. Chem. I!!_, 836 (1970)

-M.M. Gr e n a n and E.S. Cop e1and,

Radiat. Res . il,387 (197 1 )

- G. Nucifora a nd B. Sma11er,R. Remko and

E.C. Avery,

Radiat. Res . i2_,96 (1972)

- M. Z . Hoffman and E. Hayon,

J. Phys. Chem. J.]_, 990 (1973)

9. Het r efer e r e n aan " privat e communicati o n" in een publi ­

catie is niet wetenschappelijk.

10. Het hechten van een Rh(I)-complex aan een copolymeer van

styreen en d i vinylbenzeen vo lge ns Grubbs et al . verdient

de voorke ur boven de methode beschreven dóor Collman et

al.

- R.H. Grubbs,L.C. Kroll and E.M. Sweet,

J. Macromol. Sci.-Chem. A7(5),1047 (1973)

- J.P. Collman,L.S. Hegedus,M.P. Cooke,

J.R. Norton,G. Dolcetti,D.N. Marquardt,

J. Amer. Chem. Soc. 94,1789 (1972)

11. De onrechtvaardige verhouding tussen rijke en arme mensen

in Zuid-Amerika wordt gestabiliseerd door de internatio­

nale handelspolitiek tussen de rijke en de Zuid­

Amerikaanse landen.

Ontleend aan Dom Helder Camara,bisschop van

Olinda en Recife,Brazilië

12. Het vieren van Carnaval is een van de hoogste vormen

van bezinning op de relativiteit van het leven,en

verdient als zodanig alle aanbeveling.