Third Report of the Committee on Photochemistry: Reprint and Circular Series of the National Research Council (1938)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

SECONDARY PROCESSES IN THE PHOTOCHLORINATION OF CARBON MONOXIDE AND HYDROGEN HUGH S. TAYLOR Department of Chemistry, Princeton University, Princeton, New Jersey Received May 25, 1988 Since the publication of the Second Report of the Committee on Photo- chemistry considerable progress has been made towards a final and quanti- tative formulation of the secondary processes both in the hydrogen-chlo- r~ne combination and in the reactions, thermal and photochemical, of phosgene synthesis and decomposition. These several reactions have been the principal objective of the researches of Bodenstein and his school. At the present time Bodenstein (6) is occupied with the publication of the definitive conclusions of this long series of investigations and is attempting to incorporate within the framework of those conclusions, or reject for reasons ascertained, the auxiliary data that have accumulated from the investigations of other workers, notably Rollefson (15, 19, 20, 21), Ritchie (17, 18), Norrish (16), Allmand (1), and others. A discussion of the earlier work (3, 7, 8, 9, 10' 15, 22, 23, 24, 26) was included in the Second Report. I. THE PHOSGENE REACTIONS For the photoreaction at room temperatures and pressures over 100 mm. the kinetic expression is dECOCIs] = KLINE [C12~[CO]~/ This equation is derivable from the Bodenstein reaction scheme: Cl2 + E = 2Cl (1) Cl + CO + M = CO Cl + M COCl + M = CO + Cl + M COCl + Cl2 = COCl2 + Cl COCl + Cl = CO + C12 Cl + wall = 1/2 C12 (2) (3) (4) (5) (5') ~ Contribution No. 9 to the Third Report of the Committee on Photochemistry, National Research Council. 91

92 HUGH S. TAYLOR Equations 2 and 3 lead, on the Bodenstein interpretation, to an equilibrium expressed by the equation: CROCI= tCl]tCOl/[COCll The first five reactions, with the assumed equilibrium, yield the kinetic expression for Ki, which is obeyed by experimental results, under the given conditions, except in the beginning of the reaction and towards the end when reaction 5' becomes important with low concentration of COC1. The same chain-ending process (5') becomes important also at higher temperatures and lower concentrations (< 90 mm.~. At high tempera- tures the kinetic expression then becomes d [CO Cl21 /aft = x2labs. [C121 tCO] The reaction constants ~c' and K2 are related by the following equations to the several constants of the individual reaction steps: k4 K1 = - ; 5 KCOC1 k4 K2 = k5,Kcoc1 By reason of the additional investigations of Bodenstein, Brenschede, and Schumacher (4, 5), Bodenstein (6) has rejected the interpretation by Rol- lefson (19, 20), which makes use of C13 as an intermediate, and maintains his contention that the COC1 equilibrium exists in spite of reaction 4 in which this intermediate is steadily consumed. The thermal formation and decomposition between 350° and 450°C. yield the kinetic expression d[COCl2~/dt = Ah [C12~3/2tCO] — KthtCl211/2[COCl~] The reaction scheme pertaining to this is C12 ~ 2C1 (equilibrium; KC12 = [Cll2/[Cl21) (1) C1 + CO ~ COC1 (equilibrium; KCOCI2 = [Cl][CO]/tCOCll) (2, 3) COC1 + C12 = COC12 + C1 (formation) COC12 + C1 = COC1 + C12 (decomposition) The reaction constants Ah and Kth are then given by the equations k4 KC12 Kcoc1 and Kth = k4,~Kcl2 The numerical data for the equilibrium and reaction constants (4) (4') Bodenstein, Brenschede, and Schumacher (6) have recently completed a calculation of the numerical data for the several individual reactions and

PHOTOCHLORINATION OF CARBON MONOXIDE AND HYDROGEN 93 Their results are summarized in the fol- for the equilibria involved. lowing. For the equilibrium between chlorine molecules And atoms the accurate data of Giauque and Overstreet (13) are available 1 K 57~56 + 3 820 For the equilibrium COC1 ~ CO + Cl both the heat of reaction and the constant must be so chosen that log KCOC1 = log k3—log k2 Further, log k3 must be sufficiently greater than log k4 so that the assump- tion of practical equilibrium in spite of reaction 4 can be maintained. By trial, the equation obtained was log KCOCI = _ 5676 + 1.770 For reactions 4 and 5 the data are given in the form of equations E4.571T + log Zen + 1/2 log T—log f where E is the activation energy, Z.. the collision yield for T = 1°K., end f is the steric factor. In these equations E and f are both adjustable. log k4 = - 4267iT + I/2 log T + 10.101—3.871 log k5 = _ 4197 OT + 1/2 log T + 10.106 - 0.976 The equilibrium constant KCOC! yields the value 5676 cat. for the heat of formation of COC1. The heat of formation of phosgene from CO + C12 is 26,100 cat. With these two data and the value of 2612 cat. from log k4 the expression for k4, becomes log k4, = _4235O736 + 1/2 log T + 10.110 - 0.171 With these several equations the calculated values for the overall reactions, photochemical and thermal, from room temperatures to 450°C. agree excellently with the measured values. Bodenstein sees in this concordance the best and most convincing support for the reaction schemes assumed and for the equilibrium, CO + Cl = COC1

94 HUGH S. TAYLOR which Rollefson (19) especially has questioned. The heat of formation of COCl, 5676 car., is materially lower than the value of 10 kg-car. originally estimated, the higher figure justifying the objection of Rollefson. With the newer numerical data, Bodenstein is of the opinion that the several reactions are described very satisfactorily and, he believes, in final and definitive form. II. THE HYDROGEN—CHLORINE PHOTOREACTION The reaction sequence in the hydrogen-chlorine combination is, it is quite generally agreed, the Nernst chain mechanism, with the chains normally terminated by interaction of atomic hydrogen with oxygen im- purities. The reaction scheme thus becomes C12 + E = 2C1 (1) C1 + H2 = HC1 + H—800 cat. (2) H + HC1 = H2 + C1 + 800 cal. H + Cl2 = HCl + C1 H + O: + M = HO2 + M (2~) (3) (4) The numerical data for the several reactions are still subject to final revision but, according to Bodenstein, the most reliable data now available are obtained from the following equations for velocities, the units being in moles per liter per second. log k2 = 5750 1/2 log T + 10.47 - 0.92 = 6.40 at 288°K. logk2' = 4 CHIT + 1/2 log T + 10.48 - 1.39 = 6.55 at 288°K. log k3 = - 4 5575~0T + 1/2 log T + 10.50 - 0.78 = 9.01 at 288°K. log k4 = - 4 571T + 1/2 log T + 10.42 - 1.60—log f In the last expression the datum—1.60 represents the logarithm of the number of three-body collisions (moles per liter). The value of log f varies with M according to the best evidence. Bodenstein assigns the following values: log f = - 1.37 for M = C12, H2, O2; log f = - 0.77 for M = HCl; and log f ~—1.07 for M = H2 + C12 mixture. The value E2 = 5750 cal. is obtained from Hertel's value for the tem- perature coefficient for 10° = 1.37. Hence, since E2—E2, = 800 car.,

PHOTOCHLORINATION OF CARBON MONOXIDE AND HYDROGEN 95 E2, becomes 4950 cal. From a comparison of H + HCl = H2 + Cl with H + Cl2 = HC1 + Cl (in pare-hydrogen) and the variation of k2,/ks with temperature, En, E3 = 2400 cal. and so E3 = 2550 cat. From Hertel's data (14) it follows also that E3 would be 2060 cal. if H + 02 + M were temperature-inde- pendent. To reconcile the two data for Ea we therefore can set Ed equal to 500 cat. The absolute value of log k2 = 6.40 at 288°K. comes from a comparison, by Brenschede and Schumacher (4, 5), of CO + Cl + Cl2 = COCl2 + Cl with Cl + H2 = HCl + H Hence log f = - 0.92. The absolute value of log k4 = 8.60, in the mean, was obtained by Bodenstein from analysis of data by Frankenburger and Klinckhardt (12) and by Bates (2) on peroxide formation from atomic hydrogen. From data of Ritchie (18), with experiments in which both water and hydrogen chloride were formed, log k3—log k4 = 0.41 and hence log k3 = 9.01. The steric factor corresponding is then log fa = - 0.78. From Hertel's data already discussed log k3—log k2, = 2.46. Hence log k2 at 288°K. is 6.55 and log f2' = - 1.39. The uncertainty in the values for log k is estimated by Bodenstein to be not greater than 0.3. REFERENCES (1) ALLMAND: J. Chem. Soc. 1937, 1878, and earlier papers. (2) BATES: J. Chem. Phys. 1, 457 (1933~. (3) BODENSTEIN: Z. physik. Chem. 130, 422 (1927~. (4) BODENSTEIN, BRENSCHEDE, AND SCHUMACHER: Z. physik. Chem. B28, 81 (1935~. (5) BODENSTEIN, BRENSCHEDE, AND SCHUMACHER: Z. physik. Chem. Bee, 382 (1937~. (6) BODENSTEIN, BRENSCHEDE, AND SCHUMACHER: Z. physik. Chem., in press (1938~. (7) BODENSTEIN, LENHER, AND WAGNER: Z. physik. Cheryl. Be, 459 (1929~. (8) BODENSTEIN AND ONODA: Z. physik. Chem. 131, 153 (1927~. (9) BODENSTEIN AND PLAIJT: Z. physik. Chem. 110, 399 (1924~. (10) BODENSTEIN AND SCHENK: Z. physik. Chem. Bee, 435 (1933~. (11) BODENSTEIN AND WINTER: Sitzber. preuss. Akad. Wiss., Physik.-math. Klasse 1936, 2-18. (12) PRANCKENBURGER AND KLINCKHARDT: Z. physik. Chem. B15, 421 (1932~. (13) GIAIJ~IJE AND OVERSTREET: J. Am. Chem. Soc. 54, 1731 (1932~. (14) HERTEI,: Z. physik. Chem. B15, 325 (1932~.

96 HUGH S. TAYLOR (15) LECHER AND ROLLEFSON: J. Am. Chem. SOC. 62, 500 (1930~. (16) NORRISH AND RITCHIE: Proc. ROY. SOC. (LOndOn) A140, 713 (1933~. (17) RITCHIE AND NORRISE: Proc. ROY. SOC. (London) A140, 99, 112 (19331. (18) RITCHIE: J. Chem. SOC. 1937, 857. (19) ROLLEFSON: Trans. Faraday Soc. 27, 465 (1931~. (20) ROLLEFSON: J. Am. Chem. SOC. 66, 579 (1934~. (21) ROLLEFSON: Z. PhYSik. Chem. B27, 472 (1937~. (22) SCHULTZE: Z. PhYSik. Chem. B6, 368 (1929~. (23) SCHUMACHER: Z. PhYSik. Chem. 129, 253 (1927~. (24) SCHIrMACHER: J. Am. Chem. SOC. 62, 3132 (1930~. (25) SCHIrMACHER: Z. angeW. Chem. 40, 613 (1936~. (26) SCHUMACHER AND STIEGER: Z. PhYSik. Chem. B13, 157, 169 (1931~.