PRECURSORS (IF THE METAL-COMPLEXED HYDROPEROXYL RADICAL
2207
The Precursors of the Metal-Complexed Hydroperoxyl Radical] by A. Samuni Radiation Research Laborator&s and Center for Special Studies, Mellon Institute of Science, Carnegie-MeZZon Uniwrsity, Pittsburgh, Pennsylvania 16828 (Received February 14, 1078) Publication costs assisted by Carnegie-Mellon University and the U.8. Atomic Energy Commission
The precursors of the metal-complexed paramagnetic intermediates generated in the course of redox reactions of hydrogen peroxide with metal ions have been studied by esr spectroscopy using in situ radiolysis. Thus the necessity of adding H202 and the interference of the resulting peroxy complexes were avoided. No complexed species were formed from the free metal ion in the absence of HOz radicals. However, when peroxymetal complexes were present in the system both .OH and HO2 gave rise to the same secondary species. The results indicate that the complexed radicals are formed through direct addition of H02 radical to the metal ion as well as via one-electron oxidation of peroxy-metal complexes by .OH and HOz. These results support the formulation of the species produced in all cases as M-0-0.. It was found also that ZrIV and H f I v in each case react with HOz to yield two different paramagnetic transients similar to those obtained from TiIv.
Introduction Since the assignment of the two esr singlets observed2 on mixing TiIII ions with H2Oz as the spectrum of free radicals coordinated with TiIVions,aa number of experiments have been performed to study similar metal-stabilized free radicals. 4-11 These studies were mainly carried out by generating HOt radicals in a flow system via redox reactions of HzO2 with various metal ions and following the reaction by esr spectroscopy. Although many of the transition-metal ions of the Sc, Ti, V, and Cr groups in their highest stable oxidation states were searched for the purpose of identifying the corresponding complexed radicals, in some of the systems investigated the transients have escaped detection. A better understanding of the formation and decay mechanisms of the complexed transients, as well as IinowIedge of thcir precursors, are prerequisite to the choice of the optimal experimental Conditions under which these transients can be detected by esr spectroscopy. All previous studies were made in the presence of excess Hz02 so that it was not possible to predict whether ' O H or HOz (or both) were involved in the reaction. Moreover, most of the metal ions studied form, in the presence of HzOa, various peroxy-metal complexes, and it would be helpful to know whether it is the metal ion or its peroxy complex that reacts with the primary radicals to yield the secondary ones. It was recently shownl2> l3 that inadequate understanding of the metal ion HzOz redox system has led to many erroneous interpretations. Therefore, in view of the wide use of these redox reactions as a source of free radicals and the frequent occurrence of the complexed transients throughout the reactions, a further study of these systems seemed warranted. The present work was undertaken to investigate the origin of transitionmetal complexed radicals which might be formcd via several different mechanisms. An attempt is also made
+
to account for the contradictory experimental reports and interpretations regarding the appearance of a twolined spectrum in the case of the elements of group I V (TiIV, ZrIV, HfIV). The complexed radical was produced through a steady-state, in situ radiolysis of the acid aqueous solutions of the metal ions. Using this technique it was possible to avoid the necessity of adding hydrogen peroxide to the system. Thus the interference of peroxy-metal complexes in the reactions was prevented. Moreover, by properly choosing experimental conditions, the formation of either .OH or HO2 radicals in the irradiated solution was controlled thus enabling us to distinguish among the various formation mechanisms of the radicals. The results obtained in the present study indicate that in the absence of H202 no detectable reaction of .OH with any of the transition metal ions studied takes (1) Supported in part by the U. S. Atomic Energy Commission. (2) (a) W. T. Dixon and R. 0. C. Norman. Nature (London). 196, 891 tl962); (b) W. T. Dixon and R. 0. C. Norman, J . Chem. Soc., 3119 (1963). (3) Y . S. Chiang, J. Craddock, D. Mickewich, and J. Turkevich, J . Phys. Chem., 70, 3509 (1969). (4) iM.9 . Bains, J. C. Arthur, Jr., and 0. Hinojosa, Inorg. Chem., 9, 1570 (1970). ( 5 ) M. 8. Bains, J. C. Arthur, Jr., and 0. Hinojosa, J . Amer. Chem. Soc., 91, 4673 (1969). (6). (a) M. C. R. Symons, J . Chem. SOC.A, 1889 (1970); (b) Y. Shimizu, T. Shiga, and K. Kuwata, J . Phys. Chem., 74, 2929 (1970). (7) R. W. Brandon and C. S. Elliott, Tetrahedron L e t t , 44, 4375 (1967). (8) T. Oeawa, Y. Kirino, M. Setaka, and T. Kwan, h'ippon Kagaku Zasshi, 95, 304 (1971). (9) V. F. Shuvalov, N. M . Bazhin, V. M. Berdnikov, A. P. Merkulov, and V. K. Fedorov, Zh. Strukt. Khim., 10,548 (1969). (10) M . Setaka, Y. Kirino, T. Oeawa, and T.Kwan, J . CataZ., 15, 209 (1969). (11) A. Samuni and G. Ceapski, J . Phys. Chem., 74, 4592 (1970). (12) G. Ceapslti, ibid., 75, 2957 (1971). (13) G. Ceapski, A. Samuni, and D. Meisel, ibid., 75, 3271 (1971). The Jozirnal of Physical Chemistry, Vol. 76, N o . 16, 1978
A. SAMUNI
2208 place to yield complexed radicals. On the other hand, some metal ions such as ThIV, UV1,TiIV, ZrIV, HfIV, and CeIII, react with HOZ to form the complexed hydroperoxyl radical. I n the presence of hydrogen peroxide the complexed radicals are generated probably via one-electron oxidation of the peroxy-metal complexes by both . OH and Hi), radicals. Experimental Section Reagents. Reagent grades were used when available (Fisher Scientific Co. Certified thorium nitrate; Fisher Scientific Go. Purified ammonium vanadate, uranyl sulfate, zirconium sulfate, and titanium sulfate; Research Inorganic Chemicals lanthanum nitrate, zirconium nitrate, sodium metavanadate, hafnium nitrate, and niobium pentoxide; G. Frederick Smith Chemical Co. thorium perchlorate, zirconyl perchlorate, lanthanum perchlorate, uranyl perchlorate, and cerous perchlorate; and Baker Analyzed Reagents hydrogen peroxide 30y0 and perchloric acid). All the chemicals were used without further purification. The niobium pentoxide was dissolved in sulfuric acid as described f0rmer1y.I~ Acidities were adjusted (to pH 1, unless otherwise stated) using perchloric acid. Solutions were prepared with doubly distilled water. Regularly the perchlorate salts of the metal ions studied were used, but no differences in the obtained results were found when the perchlorate was replaced by the sulfate anion. When peroxy-metal complexes were studied, the metal ion solutions were mixed with hydrogen peroxide prior to irradiation. Introduction of oxygen into the solution was carried out by bubbling a nitrogen-oxygen mixture rather than pure oxygen since the oxygenation of the solution resulted in a considerable broadening of the esr lines. On measuring the esr spectra in deaerated and aerated solutions, the line widths were found to be practically the same. Deoxygenation of the solutions was done by bubbling with Airco prepurified nitrogen. The steady-state, in situ radiolysis esr technique and the experimental setup have been described elsewhere. 15,16 A conventional X-band spectrometer was used and the radiation source a 2.8-MeV Van de Graaff accelerator. The flow rate of the sample solution was about 1-03 cc/sec. All experiments were carried out at -15". Results and Discussion There is some controversy in the literature concerning the formation, reactions, and chemical nature of the complexed radicals, For instance, the eightlined spectrum which appears during the oxidation of V 0 2 + by HzOz has been accounted for in differentways
+ HOz +VOz+-HO2 vo(o2)++ .OH --+ vo(.02)2+ + OHVO(02)f + HOz +V O ( * O Z ) ~++ HOzV02+
The Journal of Physical Chemistry, Vol. 76, N o . 16,1978
(1)*$10
(2p7
(3)''
+
while in the case of titanium HzOz the number of the reaction schemes suggested was even higher. Denoting schematically the metal ion (VOz+), its peroxy complex (VO(O*)+), and the complexed radical formed (VOz+-HO,) as R4, M-HzOz, and P, respectively, the various mechanisms can be written generally for all metals as follows 9
+ *OH+ *P 14 + HOz + . P
(5)
M-HzOz -b *OH--ic .P
(6)
+ HO:,
(7)
31
lI-HzOz
JI-HzOz l!t-HzOz
---j
-P
+ *OH* OH- + . P + HOz *HO2- + . P
(4)
(8) (9)
where reactions 4-7 are addition reactions and reactions 8 and 9 are one-electron redox processes. (The schematic notations 14 and M-HzO2 are adopted for the metal ion and its peroxy complex(es), irrespective of their hydrolyzed forms, both in the highest stable oxidation state of the metal. ' P represents the complexed paramagnetic species, and no attempt has been made here to draw any conclusions concerning the various present in ionic species, either of the metal ion or .P, the reaction mixtures.) Although there is some evidence that most of the complexed radicals detected hitherto might be produced in the absence of OH radicals, it was not yet made clear whether reactions 4, 6, and 8 should be excluded also when .OH radicals do occur in the s y ~ t e m . ~ * ~ To* -check ~ ~ the existence of reactions 4-9 the occurrence of the possible reacting species (14, M-HzOz, .OH, HOz) was controlled by choosing proper experimental conditions and the solutions of the metal ions studied were irradiated in the following manner: (a) in the absence of both 0 2 and H202 where neither HOz nor R4-Hz0z were present; (b) in the presence of Oz where both HOz and .OH do occur in the solution; (c) in the presence of both 0 2 and HzOz (excess) (As a result the dominant species were HOz and M-H202.); (d) in the presence of a low IH2O,] and in the absence of oxygen (in such a case neither M nor HOzwere present in the solution due to binding of the HzOzmolecules by the metal ions). (14)N. Adler and C. F. Hiskey, J . Amer. Chem. Soc., 79, 1827, 1831, 1834 (1957). (15) R. W. Fessenden and R. H. Schuler, J . Chem. Phys., 39, 2147 (1963). (16) K. Eiben and R. W. Fessenden, J . Phys. Chem., 75, 1186 (1971). (17) H. B.Brooks and F. Sicilio, Inorg. Chem., 10, 2530 (1971). (18) K. Takakura and B. Ranby, J. Phys. Chem., 72, 164 (1968). (19) F. Sicilio, R. E. Florin, and L. A . Wall, ibid., 70, 47 (1966). (20) J. Stauff and H. J. Huster, 2.Phys. Chem., (PTanlcfurt a m M a i n ) , 55, 39 (1967). (21) N . C. Verma, K. P. Mishra, and B. B. Singh, I n d i a n J . Chem., 9, 882 (1971). (22) V. H. Fischer, Ber. Bunsenges. Phys. Chem., 71, 685 (1967).
PRECURSORS OF THE METAL-COMPLEXED HYDROPEROXYL RADICAL
2209
a. Deaerated Solutions. The esr signals of the radicals present during the steady-state irradiation of the acid (0.1 M HClOI) aqueous solutions containing 4 mM metal ion were recorded for the following transition metal ions: TiIV, ZrIV, ThIV, Ce'I', Vv, Kbv, Uvl, and HfIV. The irradiations were carried out first in acid solutionF in the absence of oxygen where H and *OHradicals are the only species which react with the solute. Under these experimental conditions no esr signals due to metal-complexed radicals were observed in any of the irradiated solutions. Considering the high production ratel6 of the primary radicals and the slow decay of the secondary ones searched, the absence of any detectable signal leads to the exclusion of the possibility formerly suggestedz0 that P is directly formed via reaction 4. This result is also in accord with previous conclusions based on the similarity between the esr signals of .Pformed from vanadiumGaand titaniumGband those of peroxy species (XI-0-0.) rather than monoxy (31-0.) species.23 Therefore, it is possible to conclude that no detectable addition reaction of .OH radical to any of the metal ions studied takes place. b. Aerated Solutions. Acid aqueous solutions of the metal ions have been irradiated in the presence of oxygen, the esr spectra were recorded, and the results are summarized in Table I. Under those experimental conditions all . H atoms are converted into HOz radicals
-
"2
"$
+
5,
b
.H
+
0 2
+HO,
(10)
leaving [ * O H ]practically unaffected. Hence, the formation of the observed complexed radicals undoubtedly involves a direct addition of the HOz radical to the metal ion (i.e., reaction 5), a conclusion which agrees with former r e ~ u l t s . ~ , gIt* ~is~seen in Table I that ThIV, Uvl, TiIV, G I v , HfIV, and CeIrl react directly with the HOz radical. In the case of ZrxV and Ce"' the present experimental results differ considerably from those previously r e p ~ r t e d . ~ , ~Furthermore, ~" the complexed species thought to bes-lo formed from HOz and vanadium(V) ions (VOz+) was not detected even when OZ was introduced into the solution, a result which indicates that neither 'OH nor HOz react with VOz+ to yield the complexed radical. Thew discrepancies could stem from the fact that premixing of the metal ion with HzOz is essential in some cases (as for Vv) but unnecessary in others. To check this possibility, the experiments were repeated when hydrogen peroxide was added to the solutions. c. Aerated Solutions Containing HzOz. In the presence of an excess of HzOzthe .OH radicals also are converted into HOzradicals through reaction 11 *OH
+ Hz02 +HzO + HOz
(11)
(23) P. It. Edwards, S. Subramanian, and M. C. R. Symons, Chem. Commun., 799 (1968). The Journal of Physical Chemistry, Vol. 76, N o . 16, 1972
2210
A. SAMUNI kl4
TirV
+ HzOz k
TirV-H2O2
(14)
14
as well 8s reaction 11, taking into account, of course, the self-recombination and cross-recombination reactions of .OH and H&. Now, considering the corresponding rate constants (klz = 6 X lo2 M - l s e ~ - l , ~k13 g Table I1 : Peroxy Complexesa of the Metal Ions at 0.1 M HC104 Metal ion
TiIV VV
NbV
UV’ ZrIV
Composition of complex
Log
Kw
Monoperoxy Monoperoxy Diperoxy Monoperoxy Diperoxy Monoperoxy Diperoxy Mono- and diperoxy
4.27 4.54 0.11 12.7 32 60
Reference
24 25 25 26, 14 14 27 27 28
Remarks
HzSO440%
a Except in the case of TiIV, additional complexes with different ratio of peroxy group per metal ion do exist, but at the low metal ion concentration and pH used, the mono- and diperoxy Complexes are dominant. See ref 28 for detailed description.
Figure 1. The esr spectra (second derivative) obtained on radiolysis of aerated 0.1 M HCIOd solutions containing 4 mM of (a) TiIJ’, (b) ZrIV, and ( c ) H P . Field increases from left to right. The relative positions of the lines in the different radicals are properly indicated (see Table I for g values).
l o 9 M-‘ S ~ C - ’ , 1c14 ~ ~ = 1.2 X lo2 M-l sec-1,25 Icll 1.2 X lo’ M-l se~-l,~O log K14 = 425)it seems evident that reactions 11-14 take place concurrently. Consequently, provided [HzOz] > [Ti11x], titanous ions, titanic ions, peroxy-titanic ions, as well as .OH and thus increasing the HOz chemical yield by a factor of HOz radicals are all present in the system simulta1.8 compared with the oxygenated H20z-freesolutions neously. Thus the formation of the peroxy-metal (primary yields of - H , .OH, and eaq- are about 0.6, complexes, before the primary reactioa is over, makes it 2.8, and 2.8, respectively). Furthermore, peroxyalmost impossible to ascertain which are the precursors metal ion complexes were present in the irradiated solution due to the presence of the hydrogen p e r o ~ i d e ~ ~ of - ~the ~ long-lived secondary radicals. The comparison of the results obtained in the pres(see Table 11). As a result the complexed-transients ence and in the absence of HzOl (Table 1)indicates that, involving Vv and KbV have become detectable, indiin the case of group V metals, reaction 5 probably does cating that peroxy complexes are, in this case, the prenot occur (to any detectable extent)-a conclusion cursors of . P rather than the metal ions themselves, which supports previous similar suggestions. l1 It is Le., reactions 7 and/or 9. also seen in Table I that the esr spectra of the radicals The results showp in Table I and the interpretation formed from Uvr, ThIV, CelI1, TirV,arid HfrVwere not given above can be compared with previous ones8p8i11 changed on the addition of HzOz into the solution. It attributing the complexed species formed to the metal might suggest that the same complexed paramagnetic ions and/or their peroxy complexes. The contradictory explanatious stem from the role of hydrogen peroxide in the metal ion-HzOz redox systems. TO (24) E. Gastinger, 2.Anorg. Allg. Chem., 275, 331 (1954). (25) M. Orhanovic and R. Wilkins, J . Arne?”. Chem. SOC.,89, 278 demonstrate the complications which arise from (1967). studying the complexed radical using the (widely used) (26) L. G. Sillen, “Stability Constants of Metal Ion Complexes,” Chem. SOC.S p e c . Puhl., No. 17, 206 (1964). rapid-mixing technique, let us consider the titanium(27) A. I. Moskvin, Radwkhimiya, 10, 13 (1968). (111) H202redox system. The reactions which take (28) J. A. Connor and B. A. V. Ebsworth, Advan. Inorg. Chem. place at 0.1 M HC104 are Radiochem., 6 , 279 (1964)
= 3 X =
+
I
Tilr1
+ H,Oz 5TiIV + OH- + .OH
Ti111 + .OH L$TiIV
+ OH-
The Journal of Physical Chemistry, Vol. 76, N o . 16, 1979
(12) (13)
(29) G . Csapski, A. Samuni, and D. Meisel, submitted for publication. (30) M. Anbar and P. Neta, Int. J . A p p l . Radiat. Isotopes, 18, 493 (1967).
PRECURSORS OF THE METAL-COMPLEXED HYDROPEROXYL RADICAL species are formed in the presence and in the absence of HzOz. It is noteworthy that HtOz formed by the ionizing radiation was present in the system throughout all of the experiments. Kevertheless, no esr signals due to the complexed radicals were observed unless either oxygen or hydrogen peroxide was added to the solution prior to irradiation (see Table I). Indeed, the calculation of the [H202]produced through the irradiation and the [M-HzO2] consequently formed shows that [.PI u-as too lorn- to be detected (for the case of A4 = TiIV). d . Deaerated Solutions, [ H z O z ]< [ M I . The complexed radicals have already been obtained in systems where .OH radicals were absent as by C P Ht024,8,11 reaction or via electrolytic oxidation of peroxy titan i ~ m . ~Since l .OH is a more efficient oxidant than HOz, such a mechanism implies that .P radicals formed through mere oxidation of peroxy metals would be formed from .OH radical also. This possibility was studied by irradiating deaerated solutions where [HzOz] < [MI. Under such experimental conditions no HOz radicals are formed through reactions 10 or 11 either, provided the peroxy complex formed between the metal ion and H20zis very stable. As a result, the reacting species in the system turn out to be R!I-H202 and .OH rather than 14 and H02. Considering Table I1 it is seen that some of the metal ion investigated forms very stable peroxy complexes with HzOz. Kevertheless, as shown in Table I, all metal ions yielded observable esr signals in deaerated solutions where [HzOzJ< [M], indicating the formation of the complexed radical also from Pf-HzOz and .OH. Again, the complexed species formed through addition of HOz to M (reaction 5) were indistinguishable, except for the case of ZrIV, from those produced from ’OH and nI-Hz02. Therefore, assuming the radicals formed through the various mechanisms are identical implies the exclusion of the existence of reactions 6 and 7. This conclusion suggests the following formation mechanisms for the metal ions investigated: (a) reaction 5 in the case of Ce’II and maybe ThIV, too
+
Ce3+ (or CeOH2+)
+ HO, +Ce(.02)2++ H+
(b) reactions 8 and 9 in the cases of VV and NbV
+ HOz +HOz- + VO(*Oz)2+ + *OH+OH- + VO(*0z)2+
VO(Oz)+ VO(O2)’
(c) reactions 5, 8, and 9 in the cases of TiIV, HfIV, Uvl, and ZrIV Ti02+ TiO(0,) TiO(02)
+ H 0 2 +TiO(.02)+ + H +
+ Hi), TiO(.Oz)+ + H o t + .OH +TiO(.Oz)+ + OH--f
e. Transients with Higher Peroxide Content. Most of the metal ions investigated are capable of forming
2211
complexes with peroxide-to-metal ratios higher than 1: 1. Nevertheless, the monoperoxy complexes are dominant in the solutions (except for Uvl) provided [HtO,] < [RI]. On introducing HzOz into the metal ion solution no change in the magnetic parameters of the obtained spectra was observed. These results suggest, as mentioned above, that complexed radicals formed in the presence and the absence of H20z are either identical or indistinguishable by esr spectroscopy. The latter possibility is supported by the observation of the invariance of the spectra, even under a hundredfold increase of the H20zconcentration. In other words, the unpaired electron being located on a peroxy group associated with the metal ion is presumably affectedvery little by coordination of an additional peroxy group to that ion. Evidently this is not the case with zirconium(1V) , since increasing [H20z]in the ZrIVion solution resulted both in the disappearance of the high-field line as well as the broadening of the lowfield line and shifting it towards a lower field. Hence, the previously d c t e ~ t e d ~ t gbroad a ~ ~ line in the ZrIV H202systems is probably attributable t o higher peroxy complex(es) of this ion. f. Two-Lined Spectra. In view of the close similarity between certain of the transition metals of group IV, particularly Ti, Zr, and Hf, the previously reported marked differences among the esr signals of their complexes with HO2 seemed obscure. The two esr lines (2.5 G apart) appearing in the Ti system are ascribed by all recent workers to two different paramagnetic intermediates involving the TiIV ion. On the other hand, the ZrIV ion was found to yield a comparatively with AH > 2 G. In the case of wide single line4*9,11 HfIV, a two-lined spectrum (2.5 G apart) has been observed4 and was attributed to a single species whose esr line has been split due to a hyperfine i n t e r a ~ t i o n . ~ Previously, Fischerz2related the very weak “satellite” lines accompanying the intense pair of lines formed from TiIV to hyperfine splitting caused by the metal ion. Since Zr and Hf as well as Ti are labeled by nonzero nuclear spin isotopes roughly at the same level (47Ti, I = b/ 2, natural abundance: 7.75%; 49Ti,I = 7/2, 5.51%; 91Zr,I = 5 . / ~11.23%; , 177HF,I = 7/z, 18.39%; 179Hf,I = ”2, 13.78%), it would be reasonable to anticipate similar behavior of these three systems. As shown in Figure 1, each of the three metal ions, Ti, Zr, and Hf, when irradiated in the presence of 0, gave rise to an esr spectrum consisting of a pair of lines separated by 2.54, 1.54, and 2.72 G, respectively. Previously it was reported4 that no change in the intensities ratio of the two-lined spectrum of HfIVHOz was observed when the experimental conditions were varied. This result led Bains, et ~ l .to, ~relate the spectrum to a single species. However, the results obtained in the course of the present study show that
+
(31) H. B. Brooks and F. Sicilio, J . Phys. Chem., 7 4 , 4565 (1970).
The Journal of Phvsical Chemistry, Vol. 76, N o . 16, 1972
2212
A. SAMUNI
this is not the case. On changing the concentration of reasons: (a) the loss of sensitivity by an order of magthe metal ion, the relative intensity of the high-field nitude due to the splitting of the line (13'3La,I = 7/*, line compared with that of the lowfield line was not natural abundance = 1 0 0 ~ o; ) (b) the too low steadystate concentration of the radicals caused either by a kept constant in the cases of Zr and Hf. These results exclude the assignment of the HfIV-H02 two line specvery fast decay rate or a very losv stability of the COMtrum to a line splitting and indicate that two different plex LaTT1-H02. This explanation is in accord with the complexed radicals are formed by the H 0 2 addition to very poor signal-to-noise ratio achieved when the LaIIIthe metal ion. As to the TiIV-Hoe, several attempts H 0 2was produced by the rapid mixing technique.36 have been made to account for the appearance of its The Ce111-H02 complexed radical has been "doublet" spectrum. The two lines werc atrributed to: to be a short-lived radical too; nevertheless, it was (1) H02 and .OH radicals;32(2) HOz radical and .OH detected p r e v i o ~ s l yusing ~ ~ ~the ~ ~rapid-mixing ~~~ esr or HOa radical associated with Ti1V;33(3) .OH radical technique. On comparing the magnetic parameters and .OH radical complexed with TiIV;19(4) .OH and of the Ce111-H02 and the ThIV-HO2 signals (Table I), HOz radicals, both complexed with TiIV; 20*34 35 (3 the two spectra seemed to be identical and the possibil'OH and HO2 radicals, both complcxed with peroxyity that one of the metal ions was contaminated by the TiIV.,18 ( 6 ) TiIV-OO. and a dimeric species TiIV-0other was considered. Since both signals were observed O-TiIV-OO. j Z 2 (7) HOZ radicals coordinated with even at low concentrations of the metal ions and their Ti1V-(Hz02)ncomplexes of different types.6b Recently intensities were of the same order of magnitude, the it was even suggestedz1that these long-lived paramagpossibility that one species only is responsible for both netic species are formed from Tilll-OH-TilV or TiI'Isignals seemed unlikely. The incompatibility of the H02-TiIV complexes. present result with those previously reported4 wherc I n view of the results obtained in the present work, Ce111-H02 was not found can be explained as follow. it appears that both lines can be generated in the absence for adjusting The use of sulfuric acid by Bains, et d,* of peroxy complexes of the metal ions (Ti, Zr, and Hf), the pH prevented the detection of this species while, where HC104 was used, the CeI1I-HO2 was observ;.e., from & ions !I and NO2 radicals solely. Regarding the pair of esr lines appearing in the Ti system, it is able.37,3s To verify this explanation sulfuric acid has almost generally accepted to relate the low-field line been added to the aerated solution of CelII prior to Indeed, the to the TiIV-H02 radical (TiO( .02)+). irradiation. This resulted in a complete disappearance present results confirm that assignment. But it seems of the Ce1I1-HO2esr signal, even at [H,SOa] = 0.1 M , that the suggested complexes of TirV-( 0H),819,20,34 while ThlV-HOz signal was observed even in sulfuric Ti11T-OH-Ti1V,21 and Ti1V-(H20z)n-H026bshould be acid. (The different behavior in the various acids has ruled out on the grounds of the experimental results. already been discussed e l s e ~ h e r e . ~ ~It) should be However, the possibility of an existence of a dimeric noted that exclusion of a Ce1V-HO2 stable radical by form as previously suggested22cannot be excluded as is Bains, et al.,4 is apparently correct, since HOz reduces the assignment of the two lines to different hydrolyzed CeIV very rapidly in both sulfuric and perchloric forms of the metal ions. (The same arguments hold for ZrIVand HfIVas well.) g. Short-Lived Complexed Radicals. The lifetime Conclusions of the complexed HOz radicals has been shown6-11 to It would be justifiable to refer to the various combe very long compared with the decay time of the free plexed species produced in the metal-H20z system as HO2. This fact was one of the reasons which made it hydroperoxyl radicals complexed by metal ions easier to detect these species. Hence, a complexed (31-0-0 ) . radical with a shorter lifetime might sometimes escape Transition metal ions form complexed radicals detection. Recently the esr spectra of the very unthrough direct addition of KO2 radical but not of .OH stable complexed radicals NbV-H02 and La1I1--HO2 radical. were identified36and previous failure to detect them4 When H202 is present in the system, the secondary was accounted for. The difficulty caused by the fast decay rate of the radicals was overcome in the present (32) L. H. Piette, G. Bulow, and K. Loeffler, American Chemical study by generating the radicals at the observation Society, Division of Petroleum Chemistry, C9-Cl9 (1964). point by in situ radiolysis. As can be seen in Table I (33) A. R. Metcalf and W. A. Waters, J . Chem. Sac. B , 340 (1967). the ten-lined spectrum of NbV-H02 (s3Nb, I = @/2, (34) M .S. Bains, J. C. Arthur, Jr., and 0. Hinojosa, J.Phys. Chem., 72, 2250 (1968). natural abundance = 100oJo) was observed on irradiat(35) W. A. Armstrong, Can. J . Chem., 47, 3737 (1969). ing the peroxy-niobium solution. On the other hand, (36) G. Czapski and A. Samuni, unpublished results. no esr signal due to LaIII-HOzss was observed even on (37) G. Czapski and A. Samuni, Israel J . Chem., 7, 361 (1969). the addition of high LaIII concentrations. The failure (38) G. Czapski, H. Levanon, and A. Samuni, ibid., 7, 375 (1969). (39) A. Samuni and G. Czapslri, ibid., 8, 551 (1970). to detect the LaIII-H02 could stem from the following 8
I
The Journal of Physical Chemistry, Vol. 76, N o . 16, lQ7R
RATECONSTANT FOR H
+ HzCO
2213
paramagnetic species are produced via an oxidation of the peroxy complexes of the metal ions by both HOZ and .OH radicals. The complexed radicals produced from different peroxy complexes of the same metal ion are (except for ZeIV) indistinguishable by esr spectroscopy. TiIV, ZrIV, and HfIV exhibit similarity by yielding
Measurement of the Rate Constant for H
two different complexed radicals each on addition of HOz to the metal ion, even in the absence of HzOz. This is probably due to different hydrolyzed or dimeric forms of the metal ions. NbV and Vv do not form complexed radicals through the reaction of the free metal ions but via their peroxy complexes only.
+ H,CO
---t
H,
+ HCO at 297-6520K
by A. A. Westenberg* and N. deHaas Applied Physics Laboratory, The Johns Hopkins Umiversz'ty, Silver Spring, Maryland .%?OB10 (Received February 14, 1973) Publication costs assisted by the Bureau of Naval Ordnance
+
+
Absolute measurements of the rate constant for the reaction H HzCO+ Hz HCO (ICl) over the temperature range 297-652'K are reported. A fasbflow reactor with esr detection for H atom decay was used. The Arrhenius expression obtained was IC1 = 1.35 X 10" exp(-3760/RT) cma mol-' sec-1. Comparisons with other data are made.
The rate constant of the reaction
H
+ HzCO +Hz + HCO
(1)
has been measured directly only a t room temperature,' while IC1 relative to other reactions has been determined by more complex photo1ysis2va and explosion limit4 studies at higher temperatures. We report here the direct measurement of kl over a temperature range 297-652' K. The technique was that used in several previous studies from this laboratory, which has been adequately described.6-7 It is a fast-flow system with movable stable gas (HzCO) injector and a fixed esr cavity (outside the heated reactor) to monitor H decay, the latter being furnished by a microwave discharge in a trace of Hz (
