J. Phys. Chem. 1985,89, 1339-1341 the observed increase in the decay time is compatible with the room temperature results. In conclusion, we find that, under collision-free conditions, fast intersystem crossing sets in only at energies appreciably above the origin of SI.The fact ISC rate constants obtained in condensed phases and high-pressure vapor^^^*'^ ( - 5 X lo8 s-l) are probably (16) G. M. Breuer and E. K. C. Lee, J . Phys. Chem., 75, 989 (1970). (17) A. M. Halpern and B. R. Ware, J . Chem. Phys., 54, 1271 (1971).
1339
due to pseudo-first-order collision-induced processes.
Acknowledgment. This work was partly supported by the Sheinbrom Foundation. We thank Prof. Uzi Even and Mr. Nahum Lavie of the Department of Chemistry of Tel Aviv University for helpful discussions. We are indebted to Prof. Hanazaki for communicating- his results to us prior to publication. Registry No. CH3COCH3,67-64-1; CD3COCD,, 666-52-4.
Temporary Bistabiilty and Unusual Oscillatory Behavior in a Closed Beiousov-Zhabotinsky Reaction System' Peter Ruoff*+ and Richard M. Noyes* Department of Chemistry, University of Oregon, Eugene, Oregon 97403 (Received: January 2, 1 98S}
When the organic substrate methylmalonic acid is present in stoichiometric excess, the ferroin-catalyzedoxidation by bromate in a closed system exhibits a particularly rich behavior. Thus, we observed more than 1 h during which large-amplitude oxidative excursions separated from each other by several minutes were accompanied by occasional sequences of damped or amplified regular small-amplitude oscillations with periods of about 20 s. These very different types of oscillations were centered on different potentials and could not be considered to be associated with the same steady state. After oscillations had been damped out, the system persisted for over 1 h in a condition such that small perturbations with silver or with bromide ions could initiate transitions between two different steady states each of which was stable to perturbations below a critical threshold magnitude. We believe this is the first time that an extended period of bistability has been observed in a closed system.
Introduction The Belousovz-Zhabotinsky3 reaction is probably the most investigated and best understood of all chemical oscillators! Many of the previous studies have used the organic substrate malonic acid, CH2(C02H)2, which can undergo many complicated chemical p r ~ e s s e s . ~ One of us6 has recently explored the cerium-catalyzed behavior with methylmaonic acid, CH3CH(C0zH)2. Only one hydrogen of this substrate is susceptible to bromination, and at least under some conditions7 the stoichiometrically significant change in the overall system can evidently be described well to within about 3% by the unique
+
+
-+
3Br0,5CH3CH(C02H)z 3H+ 2CH3C02H 3CH3CBr(C02H),
+
4 c 0 2 + 5 H z 0 (T)
W e now find that ferrion catalysis of reaction T can generate an unusually rich behavior when the organic reducing agent is in stoichiometric excess. Phenomena which can be observed during a single run include (a) sustained large-amplitude oscillations accompanied by occasional bursts of small-amplitude oscillations taking place around very different average electrode potentials and (b) a long period during which the system can exist in either of two steady states each of which is potentially excitable to the other but is stable to very small perturbations. We believe that these behaviors are unprecedented for closed homogeneous systems.
Materials and Method All experiments were performed in a stirred thermostat4 glass beaker a t 25 O C . The system was monitored potentiometrically with a bright platinum electrode measured against a double junction Ag/AgCl reference electrode (Orion, Model 90-02). The outer chamber of the reference electrode was filled with a 10% K N 0 3 solution. The potential was followed with a conventional x-t recorder (Leeds and Northrup Speedomax). 'Permanent address: Department of Chemistry, Rogaland Regional College, Ullandhaug, 4001 Stavanger, Norway.
0022-3654/85/2089-1339$01.50/0
All chemicals except the methylmalonic acid, MeMA (Fluka, >99%), were of analytical grade. The reaction volume was 50 mL, and oscillations were started by mixing reagents in the order H2S04,MeMA, NaBr03, KBr. After the yellow color of initially formed bromine had disappeared, ferroin solution (BDH, 1,lophenanthroline-ferrous sulfate complex solution, 0.025 M) was added. Perturbations were performed by adding the perturbant dropwise with a conventional Pasteur pipet. As in an earlier study: the average drop size was taken to be 31 pL.
Results Figure 1 illustrates approximately the first 4 h of a run with the indicated initial composition. After all of the components had been added, the solution first alternated between extended periods of 30-40 min each in reduced and oxidized steady states. Particularly when in the reduced state, the potential exhibited noisy (1) No. 61 in the series "Chemical Oscillations and Instabilities"; No. 60 is Noyes, R. M. In "Non-Equilibrium Dynamics in Chemical Systems" Vidal, C., Pacault, A., Ed.; Springer-Verlag: Berlin, 1984; pp 60-64. (2) Belousov, B. P. Ref.Radiats Med., Moscow 1959, 145-147. (3) Zhabotinsky, A. M. Dokl. Akad. Nauk SSSR 1964, 157, 392-395. (4) Reviews and references can be found in "Oscillations and Traveling Waves in Chemical Systems", Field, R. J., Burger, M., Ed,Wiley: New York, in press. ( 5 ) (a) Jwo, J. J.; Noyes, R. M.J . Am. Chem. SOC.1975,97, 5422-5431. (b) Ganapathisubramanian, N.; Noyes, R. M. J . Phys. Chem. 1982, 86, 5 158-5 162. (6) Ruoff, P.; Schwitters, B. 2.Phys. Chem. (Frankfur?am Main) 1983, 135, 171-184. (7) Hansen, E. W.; Gran, H. C.; Ruoff, P. J . Phys. Chem. 1985, 89, 682-84. (8) Ruoff, P. J . Phys. Chem. 1984, 88, 2851-2857. (9) Epstein, I. R.; Dateo, C. E.; De Kepper, P.; Kustin, K.; Orbin, M. In 'Nonlinear Phenomena in Chemical Dynamics", Vidal, C., Pacault, A,, Ed.; Springer-Verlag: Berlin, 1981; p 188. (10) Field, R. J.; Karos, E.; Noyes, R. M. J. Am. Chem. SOC.1972, 94, 8649-8664. (11) Field, R. J.; Noyes, R. M. J . Chem. Phys. 1974, 60, 1877-1884. (12) (a) Showalter, K.;Noyes, R. M.; Bar-Eli, K. J . Chem. Phys. 1978, 69, 2514-2524. (b) Ganapathisubramanian, N.; Noyes, R. M. J. Chem. Phys. 1982, 76, 1770-1774.
0 1985 American Chemical Society
1340 The Journal of Physical Chemistry, Vol. 89, No. 8, 1985
Letters
1 2
3
+ I
1 5
c
v-\
1 Min
A 15Min F 7 5 s e c 4 1-1
9
Jiu B
8
Figure 1. Oscillatory behavior of a ferroin-catalyzed methylmalonic acid BZ system. Initial concentrations: [H2S04],= 0.5 M; [MeMA], = 0.3 M; [NaBrOplo= 0.1 M; [KBr], = 0.1 M. Parts A and B represent a continuation of the same run with an overlap of one ‘spike”. (A) 1: addition of KBr; 2: after bromine color has disappeared, the ferroin is added; 3: after staying for a while in the reduced state, the system goes spontaneously into the oxidized state; The color of the solution changes from red to blue; 4: at the beginning of the oscillating period, oscillations
t
1
t
2
Figure 2. Bistability at the end of the oscillating region in the ferroin-
of high frequency and increasing amplitude are observed in the oxidized state; 5 : train of small-amplitude oscillations in the oxidized state; 6: oxidizing excursions or “spikes”. Note also the considerable “noise” which is observed in the reduced state. (B) 6 and 8: oxidizing excursions; 7: train of small-amplitude oscillations in the oxidized state. Period length in the expanded region is 20 s; 9: at the end of the oscillating region, small-amplitude damped oscillations in the oxidized state are observed (period length in the expanded region is 19 s).
catalyzed methylmalonic acid BZ reaction. The composition corresponds to that at the end of the time in Figure 1. (A) The system starts in the oxidized state. 1: one drop (31 pL) of a 0.004 M KBr solution (2.5 X lod M in 50 mL) shows that the steady state is stable. 2: two drops of a 0.004 M KBr solution drive the system to the reduced state. The color of the solution changes from blue to red. (B) The system starts in the reduced state. 1: five drops of 0.004 M AgNOpsolution (1.3 X M) shows that the steady state is stable. 2: eight drops of 0.004 M AgNOSsolution (2.0 X M) drives the system to the oxidized state.
fluctuations which seemed to be far too large to explain as instrumental effects. Then followed a little more than 1 h during which the solution spent most of the time in a reduced state exmpt for seven oxidizing excursions of about 200 mV each. Many of these excursions were sharp “spikes” of oxidation, but during those marked 5 and 7 in Figure 1 the oxidative excursion included several minutes of regular small-amplitude oscillations extending 40-1 00 mV with period about 20 s each. These small-amplitude oscillations generated a regular envelope as they first damped and then amplified until the system passed a threshold and switched back to a reduced steady state. After the system had exhibited regular small-amplitude oscillations, the time in the reduced state was longer before the next oxidative excursion, and when the time between excursion spikes had been reduced to about 5 min, the next oxidative excursion exhibited small-amplitude oscillations. At the excursion marked 9 in Figure 1, the small-amplitude oscillations damped instead of being amplified as in excursions 5 and 7. The system then entered an oxidized steady state which persisted for 1 h or more if the system was not perturbed. Such an oxidized steady state was stable to perturbation by small amounts of bromide ion as shown by point 1 in Figure 2A, but a larger amount of bromide could convert it to a reduced steady state as shown by point 2. That reduced steady state was also stable to perturbation by a small amount of silver ion but could be converted back to the oxidized steady state by a larger perturbation as shown in Figure 2B. Therefore, the system generated at the end of Figure 1 exhibits a true bistabiiity in that there are two steady states each of which persists for an extended time in the absence of perturbation but each of which can be converted to the other by a sufficient but still very small perturbation. Of course the bistability cannot persist to the final equilibrium state, but we believe this is the first time that such behavior has been demonstrated for an extended duration in a closed system. In fact, bistability has been claimed
to be a phenomenon necessarily restricted to open system^.^ The features in Figures 1 and 2 were reproducible, although absolute times of transitions varied somewhat from run to run. Because the organic substrate was in stoichiometric excess, a system like that in Figure 1 eventually went to a reduced steady state which persisted for several hours before exhibiting chaotic fluctuations of the order of 100 mV with periods of the order of tens of minutes. Of course these fluctuations eventually died out before the ultimate approach to equilibrium.
Discussion The observations reported here add still another chapter to the remarkable variety of Belousov-Zhabotinsky systems! The two locally stable steady states of Figure 2 are both presumably undergoing net reaction by the stoichiometry of equation T. The oxidized state has a blue color and a very low concentration of bromide ion. The reduced state has a red color and a concentration of bromide ion in excess of the critical valuelo of about lo-’ M. Past experience indicates that the rate of overall reaction is much greater in the oxidized than in the reduced state. The small-amplitude oscillations are centered on a very different potential than are the largeamplitude excursions. Therefore, each set of oscillations is associated with a different steady state. We believe that dynamic behavior of the system must be influenced by small amounts of other intermediate or product organic species which are present to different extents during different stages of the reaction and whose presence is not recognized in simple process T. Presumably, the relative concentrations of these species determine whether an oxidative excursion will or will not lead to small-amplitude oscillations and whether those oscillations will be amplified until the system goes to a reduced state or will be damped until an oxidized steady state is attained. Investigation of this remarkably rich behavior is continuing. It is clear that such variety cannot be accomodated either to the original Oregonator model” or to an extension used to model large-
J. Phys. Chem. 1985,89, 1341-1344
1341
and small-amplitude oscillations associated with a single stationary state in a CSTR.I2
(NTNF) to P.R. and by a grant from the National Science Foundation to the University of Oregon.
Acknowledgment. This work was supported by a grant from the Norwegian Research Council for Science and Technology
Registry No. NaBr03, 7789-38-0; methylmalonic acid, 5 16-05-2; ferroin, 14708-99-7.
Micellar Effects on Belousov-Zhabotlnsky Oscillations with Trls(2,2’-blpyridyl)ruthenium( I I ) as a Catalyst Melina Maritato, Jacqueline Nikles, Laurence S. Romsted,* and Monica Tramontin Department of Chemistry, Rutgers, The State University of New Jersey, New Brunswick, New Jersey 08903 (Received: January 24, 1985)
Micellar solutions of cetyltrimethylammonium sulfate dramatically reduce the amplitude and lengthen the induction period of the Ru(bpy)?+-catalyzed Belousov-Zhabotinsky oscillations compared to reaction in aqueous solution, whereas micellar solutions of sodium lauryl sulfate shift oscillations to higher absorbances with a longer oscillation period but no change in the induction period. The ability of micelles to selectively compartmentalize ions and molecules may account for their effects on the amplitude and period of the BZ reactions.
Introduction The Belousov-Zhabotinsky (BZ) reaction, the metal ion catalyzed bromination of organic compounds by strongly acidic aqueous bromate solutions, is the most thoroughly studied oscillating reaction in homogeneous solution.’-7 The reaction is catalyzed by a number of different metal ionss including Ru(bpy)j2+.g11 Malonic acid is the most often used organic reactant, but a variety of organic corn pound^'^-'^ and mixtures of compounds produce oscillation^.'^ The fundamental processes of the BZ reaction are currently best described by the Field-Kbrbs-Noyes (FKN) mechanism,’ and mathematical models were developed which simulate, semiquantitatively, the characteristics of BZ Re-
(1) Field, R. J.; K M s , E.; Noyes, R. M. J . Am. Chem. SOC.1972, 94, 8649. (2) Noszticzius, Z.; Bodis, J. Ber. Bunsenges. Phys. Chem. 1980,84,366. (3) Vital, C.; Roux, J. C.; Rossi, A. J. Am. Chem. SOC.1980, 102, 1241. (4) Geisler, W.;Bar-Eli, K. J . Phys. Chem. 1981, 85, 908. (5) Ganapathisubramanian,N.; Noyes, R. M. J . Phys. Chem. 1982,86, 2317, 5155, 5158. ( 6 ) Noszticzius, Z.; Noszticzius, E.; Schelly, Z . A. J . Phys. Chem. 1983, 87, 510. (7) Bar-Eli, K.; Haddad, S.J. Phys. Chem. 1979,83, 2944. (8) Bolletta, F.; Balzani, V. J . Am. Chem. SOC.1982, 104, 4250 and
references therein. (9) K M s , E.; Burger, M.; Friedrich, V.; Ladanji, L.; Nagy, Zs.; Orban, M.Faraday Symp. Chem. SOC.1974, no. 9, 28. (IO) Demas, J. N.; Diemente, D. J . Chem. Educ. 1973, 50, 357. (11) Mitteilung, K. Z . Phys. Chem. ( h i p r i g ) 1983, 264, S593. (12) Kasperek, G. J.; Bruice, T. C. Inorg. Chem. 1971, 20, 382. (13) Rastogi, R. P.; Rastogi, P.Indian J. Chem., Sect. A 1980, 29A, 1. (14) Salter, L. F.; Siieppard, J. G. Inr. J . Chem. Kiner. 1982, 14, 815. (15) Rastogi, R. P.; Verma, M. K. Indian J . Chem. Sect. A 1983, 2 2 4 917. (16) Field, R. J.; Noyes, R. M. J . Chem. Phys. 1974, 60, 1877. (17) Field, R. J. J. Chem. Phys. 1975, 63, 2289. (18) Edelson, D.; Field, R. J.; Noyes, R. M. Int. J. Chem. Kinet. 1975, 7,
417. (19) Edelson, D.; Noyes, R.M.; Field, R. J. Inr. J. Chem. Kiner. 1979, 21,
155.
cently, however, the fundamental role of Br- as the control intermediate in the reaction was ~ h a l l e n g e d ~and l - ~ ~an alternative mechanism proposed.24 The issue is far from ~ e t t l e d ? ~and J ~ new evidence supporting the FKN mechanism continues to a p ~ e a r . ~ ’ . ~ ~ One reason for the current interest in oscillating reactions is the numerous periodic phenomena occurring in biological syst e m ~including ,~~ enzyme-catalyzed reaction^^^*^' and reactions across membra ne^.^^-^^ Recently sodium lauryl sulfate (NaLS) micelles were used to promote oscillations in the fluorescence intensity from irradiated solutions,35and a cationic surfactant was used in generating spontaneous sustained oscillations in the electrical potential across a liquid membrane.36*37We wondered whether micelles of ionic surfactants, which are often used as models for enzymes and membrane interface^,^^ might influence
(20) Showalter, K.; Noyes, R. M.; Bar-Eli, K. J. Chem. Phys. 1978, 59, 2514. (21) Noszticzius, Z. Acta Chim. Acad. Sci. Hung. 1981, 106, 347. (22) Nosticzius, Z.; Noszticzius, E.; Schelly, Z. A. J. Am. Chem. SOC. 1982, 104,6194. (23) Noszticzius, Z.; Noszticzius, E.; Schelly, Z. A. J. Phys. Chem. 1983, 87, 510. (24) Noszticzius, Z.; Farakas, H.; Schelly, Z . A. J. Chem. Phys. 1984,80, 6062. (25) Noyes, R. M. J. Chem. Phys. 1984, 80, 6071. (26) Tyson, J. J. J. Chem. Phys. 1984,80,6079. (27) Koros, E.; Varga, M.; Gyorgyi, L. J . Phys. Chem. 1984, 88, 4116. (28) Ruoff, P. J . Phys. Chem. 1984,88, 2851. (29) Hess, B.; Boiteux, A. Berg. Bumenges. Phys. Chem. 1980, 84, 346. (30) Hervagault, J. F.; Friboulet, A.; Kernevez, J. P.; Thomas, D. Ber. Bunsenges. Phys. Chem. 1980, 84, 358. (31) Goldbeter, A.; Caplan, S. R. Annu. Rev. Biophys. Bioeng. 1976, 5 , 449. (32) Muller, P. Ber. Busenges. Phys. Chem. 1980, 84, 341. (33) Botre, C.; Lucarini, C.; Memoli, A. Bioelectrochem. Bioenerg. 1979, 6, 451. (34) Chay, T. R.; Keizer, J. Biophys. J . 1983, 42, 181. (35) Tsuchiya, S.; Kanai, H.; Seno, M. J. Am. Chem. SOC.1981, 203, 7370. (36) Yoshikawa, K.; Matsubara, Y. J. A m . Chem. Soc. 1983, 205, 5967. (37) Yoshikawa, K.; Matsubara, Y. J . Am. Chem. SOC.1984,206,4423. (38) Fendler, J. H. “Membrane Mimetic Chemistry”; Wiley: New York, 1982.
0022-3654/85/2089-1341$01.50/00 1985 American Chemical Society
