.A
H
Rate Constants for the Reactions of Deuterium Atoms with Silanes bi, H, S. Sandhu, H. E. Gunning, and 0 . P. Strausz" Departmsnl of Chemistry, University of Alberta, Edmonton, Alberta, Canada
(Received June 26, 1972)
Publication sosts assisted by the National Research Council of Canada
The absolute rate constants for the abstraction reactions of deuterium atoms, produced by the photolysis of C&4 with Si&, SizHs, CH3SiH3, (CH3)2SiH2, and (CH3)3SiH, have been determined E t room temperatuiv in competition with addition to ethylene. Silanes are more reactive than t counterparts and the reactivities have the same order of magnitude as ethylene. The reactivity per Si-H bond i!j in the order Si2He > (CH3)3SiH > (CH3)2SiH2 > CH3SiH3 > SiH4. Potential energies of activiLion of these reactions have been computed using a modified BEBO method.
Introduction Quar tatative kinetic data for the gas-phase hydrogen atom transfer rcactions of alkyl radicals with a variety of silicon compounds have become available during the past 5 years.1-18 In general, these reactions proceed ab much faster rate3 than those with carbon analogs and the increased reactrvily has been shown to be due to lower activation energy requirements, the A factors being essentially identical. Quantitative data involving H atoms, however, are still sparse and are restricted to a study of the H plus (Glrl3)3SiHlhand H plus SiH4I7 systems. It has been shown recently from this laboratoryle that the hydrogen atoms produced in the mercury photosensitization of disilane undergo a displacement type reaction with the substrate ro give SiH4 and SiH3, parallel to the more comnioii abstraction reaction producing H2 and Siz. Hg. The rates of thei,e two processes are comparable at room temperature. p\ o such displacement reaction has been observed with methylated siianes.20,21 In previous publica1 ions5-8 from this laboratory Arrheni ~ paramelers s rind kinetic isotope effects have been reported for the reactlions of alkyl radicals with monosilane, &silane, and met hyl- and phenyl-substituted silanes which were compared to theoretical values derived on the basis of transition state theory. The present article reports the rate constants of the reactions of D atoms with monosilane, disrlant?,and methyl-substituted silanes. Experimental Section A conventional cylindrical quartz reaction cell, 35 mm x I40 mm and 168 cc volume, was attached to a greasefree high-vacuum systim. The quartz iodine lamp,22 filled
with ca. 10 Torr of Ne and excess iodine crystals, was dis-
(1) J. A. Kerr, D. H. Slater, and J. 6 . Young, J. Chem. Soc. A , 104 (1966). (2) J A. Kerr, D. H. Slater, and J. C. Yoting. J , Chem. Soc A , '134 (1967). (3) T. N . Beli and 8. B. Johnson, Aust. J. Chem., 20, 1545 (1967), (4) W. J. Cheng and M. Szwarc, J. Phys. Chem., 52,494 ((968). ( 5 ) 0. P. Strausz, E. Jakubowski, H. S. Sandhu, and H . E. Gunning, J. Chem. Phys., 51,552 (1969). (6) E. Jakubowski, H . S. Sandhu, H. E. Gunning, and 0 . P. Strausz, J . Chem. Phys., 52,4242 (1970). (7) R. E. Berkley, Ph.D. Thesis, University of Alberta. Ednonton, Canada, 1970; R. E. Berkley, I . Safarik, ? E. i. Gunniilg, ana 0 . F.Strausz. to be submitted for publication. (8) I . Safarik, R. E. Berkley, and 0. P. Strausz, J. Chem. Phys., 5 4 , 1919 (1971). (9) E. R . Morrisand J. C. J. Thynne, J. Phys. Chern., 73,3294 (1969). (10) E. R , Morris and J. C. J. Thynne, Trans. Faraday Soc., 65, 183 (1970). (11) A. U. Chaudhry and B. G. Gowenlock. J . Organometai. Chem., 1 6 , 221 (1969). (12) J. A. Kerr, A. Stephens, and J . C. Young, Int. J . Chem. Kinet., 1, 339, 371 (1969). (13) T , N. Bell and U. F. Zucker, J. Phys. Chem., 4 4 , 979 (1970); Can. J. Chem., 48, 1209 (1970). (14) T. N. Beli and A. E. Platt, lnt. J. Chern. Kinet.. 2, 299 (1970); J . Phys. Chem., 75, 603 (1971). (15) J . A. Kerrand D. M. Timlin, Int. J. Chem. Kinet., 3, 1, 69 (1971). (16) M. A. Contineanu, D. Mihelcic, R. N. Schindler, and P. Potzinger, Ber. Bunsenges. Phys. Chem., 75,428 ('1971). (17) G. K. Moortgat, Diss. Abstr. B, 31, 1879 (1970). (18) H. E. O'Neal, S. Pavlou, T. Lubin, M. A. Ring, and L. Batt, J Phys. Chem., 75,3945 (1971). (19) T. L. Pollock, H. S.Sandhu, A. Jodhan, and 0 , P. Strausz, J. Amer. Chem. Soc., in press. (20) M . A. Nay, G. N. C. Woodall, 0. P. Strausz, end ti. E. Gunning, J . /?mer. Chem. SOC.,87,179 (1965). (21) A. G , Alexander, Ph.5. Thesis, University of Alberta, Edmonton, Canada, 1972; A . G. Alexander, R. W. Fair, and 0 . P. Strausz, to be submitted for publication; A. G. Alexander and 0 . P. Strausz, to be submit!ed lor publication.
as1 1
K. Obi, H. S. Sandhu, H. E. Gunning, and 0.P. Strausz
39112
terms of steps 1-5. In addition, small amounts of I-butene, n-hexene, and 1-pentane were also detected. These minor products, which could be completely suppressed in the presence of nitric oxide, probably arise from addition reactions of ethyl radicals to the olefins and/or secondary radical reactions of products. They have been neglected in the following kinetic treatments. The ratio of the primary quantum yields is given by the relation
TABLE I: Product Yields in the Photolysis of Ethylene-d4 at 25'* _ s _ _ _ l l l l -
Products, bmol P(CZD~)~ Torr
Material balance
n-CdD10 ~-
50
2.94
50.4 50.2
286 2 89
4.05 348
3.91
50.4b
2 84
3.20
50.7c 10.4c
'.23 0.68
50.0e9d
0.66
1.7: 0 77 0 72
a Exposure time 5 min.
0.21
1.37
CZD4.1
0.21 0.20 0.18 0.09
1.26 1.37
cZD4.2
1.10
C2D4.4
0.59 0.29 0.34
C2D4.2
0.04 0.04
660 Torr of Xe added.
c204.3
c2D4.2 c2D4.0
Screen filter used.
Exposure time 2 min.
charged by a 2450 MHz/sec microwave generator. Constant iodine pressure (40 p ) was maintained by an icewater bath. Of the nine emission lines between 2062 and 1618 the 2062-A line was the strongest and the shorter wavelength lines were effectively eliminated by keeping the lamp 1 cm away from the cell. Fresh air was flushed through continuously in order to cool the lamp and to remove ozone produced during the irradiation. When f%& was present, a screen filter (28% transparency) was used to reduce the light intensity. Ethylene-& (Merck) was purified by gas chromatography on medium activity silica gel column. Using mass spectrometry it was found that 4.5% ethylene-& was present as an impurity. Monosilane, disilane, methylsilane, and dimethylsilane (Peninsular) were purified by lowtemperaturie distillations. Trimethylsilane (Procurement) was first purified by preparative gas chromatography on a medium activity silica gel column and then subjected t o distillation.
a
The value of 1.84 f 0.03, calculated from the data given in Table I, is identical with that determined by Akimoto and Tanaka2?for CzH4 at 1931A. If a silicon compound containing abstractable hydrogen is introduced into the system, the abstraction, displacement, and exchange reactions will compete with the addi-
-
+ silane HD + si!yl radicai D + silane silane-dl -2- radical D + silane H + silana-dl D
-
(6)
(71 (8)
tion step 3. Steady-state treatment of reactions 1-3 and 6-8 leads to the follo,wing rate expression
Thus, a plot of the left-hand side of this equation against ethylene-dh/silane concentration should give a straight line, from the slope of which k3/k6 values can be determined. The importance of steps 7 and 8 relative to step 6 can be estimated from the intercept. Monosilane. The isotopic composition of hydrogen formed in the presence of SiH4 is shown in Table 11. The esults values have been corrected for the presence of ethylene-& in ethylene-de. The deuterium atoms were generated by the photolysis Monosilane is transparent to wavelengths longer than of CzD4. Effective light intensities were calculated from 1850 A;28 consequently its photolysis is slow and the yield the published relative emission intensitiesz3 and the exof HZ is less than 5% of the total hydrogen yield from tinction coefficient of oxygen24 assuming the transparency equimolar concentrztions of C2D4 and SiH4. A plot of eq I of the reaction cell windows to be independent of waveusing the data presented in Table 11 for SiH4 is shown in length. About 80% of the absorbed radiation was in the Figure 1. From 4(1)/4(2) and the slope, the ratio of the 2062-1830-a region. rate constants for addition reaction 3 to that of abstraction The primary and secondary processes occurring in the 9 -~~ photolysis of CZW* are reasonably well u n d e r s t o ~ d . ~ ~ reaction Analogous reactions of C2D4 are D + SiH4 HD f SiH3 (9) C2D.4 hv C2D2 D2 (1) k3/k9 = 1.88 f 0.07. From the intercept, an upper limit of less than 0.3 can be obtained for klo/h9 C2C4 hv CzD2 + 2D (2)
+ +
-
D f C2D4
---f
C2D5
C Z D+~ C2D5 ---* n-C1Dlo e2D5 I- C2D5
-
-
+
+
C2D6 f C2D4
(3) (4) (5)
'The importance of primary process 2, relative to 1, increases with increasing photonic energy.25 The enthalpy change of reaction 2 is -1143 kcal/mol and consequently little excess translational energy can be associated with the D atoms. In addition a concurrent study21 has shown that excess translaticna? energy up to 35 kcal/mol has no discernible effect on the reaction of H or D atoms with silanes. The major products of the reaction, D2, CZDZ,C2D6, and n-CdD~o,and the experimental conditions employed are given in Table I. Their formation can be explained in The Journal of Physical Chemistry. Vol. 76, No. 26, 7972
D
+ SiH4
-
H + SiH39
(10)
and we believe that step 10 does not occur at a11.21 Disilane. Disilane absorbs strongly up to 2100 A and can be readily decomposed in this region29 whereas significant absorption by ethylene occurs ,only below 1850 A.30 In P. Harteck, R. R. Reeves, and 5 . A. Thompson, Z. Naturforsh. A, 19, 2 (1964). K. Obi, Ph.D. Thesis, Tokyo Institute of Technology, 1966. K. Watanabe, E. C. Y. inn, and M. Zelikoff. J. Chern. Phys., 21, 1026 (1953). J. R. McNesbyand H.Okabe, Advan. Photachem., 3,228 (1964). R. A. Back and 0. W. L. Griffiths, J. Chem. Phys.. 416, 4839 (1967). H. Akimoto and I. Tanaka, Z. Electrochem., 72,134 (1968). V. R. Scharz and F. Heinrich, Z. Anorg. Chem., 221, 277 (1935). T. L. Pollock, H. S, Sandhu, and 0. P. Strausz. to be submitted for publication. R. McDiarmid and E. Charney, J. Chem. Phys., 49, 1517 (1967)
20
16
12
Y
n L
\ (v
n
a
4
1
0
I
4
1
8
P(CzDd)/P(SILANE) ure 1. Plot a i D;I/HD as a function of P(CzDd)/P(silane)
P(C2D4), Torr
P(SiHe), Torr
50.5
Eio.:>
50.9
1 7.0
50.0
18.2
51.2
6.9 5.2
_I__.-
50.4 a
-__
k 3 / k l l = 0.20 reactions
Hydrogen, pmol
E2
3.i4 3.20 3.16 3.05 2,9'i
HD
H2
1.13
0.49
0.14 0.10
0.32 0.22 0.16
0.07 0.05 0.03
Expasuie time B min
TABLE Ill: Isotopic Composition of Hydrogen Formed in the photolysis of c2D-4 in the Presence of Si2H6 at 25"a,b ~~
Hydrogen, pmol P(c21)4)1
PISIZHb),
Toir
Torr
D2
nu
50.9 50.7
50.7
5.2 2.6 -!.7
t .9 0.97
0.06 0.07 0.06 0.07
1.34 1.26 1.13
50.1 50.6
0.16 0.35 0.45 0.57 0.63
0.06
0.94
a Exposure?iirne
--
D + SizHe
TABLE II: Isotopic ComFiosition af Hydrogen Formed in the Photolysis of c204 in the Presence of SiH4 at 250a
H2
1.10
5 mii?. Scrsen filter used.
order to reduce absovption by disilane, low disilane concentmations were used. and the light intensity was attenuated by a screen filter. The results of isotopic analyses are shown in Table 111. It can be seen that disilane is the major absorbing speclies even at 2% concentrations, since Hz constitutes 68% of total hydrogen yield. From the plot of Dz/HD ab a function of P(CzD4)/P(SizHG), shown in Figure 2, the relative rate constant for abstraction reaction I1
&
HD
+ Si&,
(11)
0.04. For the displacement and exchange
D + SizHs
SiH3D
D + SizH6
+
SiH3
H + SizHsD
612) (13)
the upper limit of (klz k13)/kll is estimated to be 0.4. Since slep 13 is unlikely to occur21 this leads to an upper limit of k12/kll = 0.4. From independent studies19 the measured value of k l z / k l l for the attack by H atoms i s 0.52. Methylsilane. The isotopic composkion of hydrogen ob. tained in the presence of CH3SiH3 is given in Table IV. For a fourfold excess of methylsilane, HZ is 15% of the total hydrogen yield. Thus, although methylsilane is known to decompose a t 1783 A31 its absorption at longer wavelengths is much weaker than that of ethylene-&. As methane was not observed among the products, the displacement reaction
D + CH3SiH3
-
CHs + S H 3 D
does not occur. It should be mentioned here that exchange reaction 8 does not occur with the methylated silaries.16JO A plot of D2/HD ratios as a function of P(@2D4)/P(C&SiH3) is displayed in Figure 3. For the abstrac*' cion reaction
D -t- CH3SiH3
-
HD f CHsSiH2
(14)
= 1.72 rf: 0.02. Dimethylsilane. The isotopic composition of hydrogen formed in the presence of (CH3)zSiNz is showr, in Table V. Since the HZ yield is only 5% of total hydrogen yield kS/k14
(31) A. G . Alexander, 0 . P. Strausz, R. Pottier and G. P. Semeluk, Chem. Phys. Left., 1 3 , 608 (1972), The Journal of Physical Chemistry, Vol. 76. No. 26, 7972
K. Obi, H . S.Sandhu, H. E. Gunning, and 0.P. Strausz
3914
TABLE V: Isotopic Composition of Hyarogen Formed in the Photolysis of CZD4 in the Presence of (CH3)2SIH2 at 250a
P(C2D4) Torr
Hydrogen, pmol _ ^ _ I _ _ _ _ . ~
P((CH3)2SiHz) Torr
I
I
HD
D2
Hz I
50.1
100.5 50.1 25.3 17.0 10.3
50.1 49.4
50.0 49.0 a Exposure time
2.40 2.64 2.65 2.62 2.78
1.31
0.20
0.99 0.63
0.12 0.06 0.05 0.03
0.45 0.32
5 min.
TABLE VI: Isotopic Composition of Hydrogen and Methane Formed in the Photolysis of C2D4 in the Presence of (CH3)3SiH at 250a __
Torr
50.6
99.8
1.88
0.8%
50.4
49.7 24.9 17.3 10.4
2.26 2.46 2.69 2.92
0.68 0.46
50.1
50.2
"
50.4
0
4
la
as a function of P(C2D4)/P(silane)
TABLE I V : isotopic Composition of Hydrogen Formed in the Photolysis of C2D4 in the,Presence of CH3SiH3 at 250a Hydrogen, pmoi P(CH3SirlZ), Torr
50.8
203.0 100.3
51.1 49.9
5Q.8 39.5
50.2 50.2 49.8 100.0
21.2 16.3
lG.5
D2
HD
Hz
2.38
1.70 1.62 1.11 0.96 0.62 0.34 0.22
0.72 0.47 0.28 0.23 0.07 0.06 0.06
2.78 2.83 2.81 3.02 2.98 3.62
HD
Methane, pmol
0.09 0.06
0.22 0.04
0.37
0.03
0.01
0.26
0.02 0.01
(CH&SiH2. Trimethylsilane. The results on this sytern are given in Table VI. Small amounts of H2 and methane indicate that absorption by trimethylsilane is much weaker than that by ethylene-d4 in the region above 1783 A. At the highest trimethylsilane concentration used, ea. 92% of methane product was CH4 and the remaining was CH3D. The only possible source of CH4 formation i s the direct photolysis of (CH3)3SiH. Both molecular and radical abstraction steps for methane formation have been shown to occur in the photoiysis of methyl~ilane3~ and dimethyisilane.21 The small amount of CH3D could arise from the disproportionation reactions of CH3, formed in the direct photolysis, with CzDs CH3 + C2D5
a Exposure time 5 niin
W2
I I _
DO
a Exposure t!me 5 min.
P (C2D4)/P(SILANE) Fjgure 3. Plot of D2/WD
Hydrogen, pmol
P((CH3)3SiH), Torr
P(CZD4).
-+
CH3D -i-CzD4
A plot of D2/HD ratios as a function of P ( C Z D ~ ) / P ( ( C H ~ ) ~ for a twofold excess of dimethylsilane the amount of abSiH) is shown in Figure 3. The slope of this plot yields sorption by dimethylsilane must be minimal. Only small amounts of methane (0.004 pmol at the highest dimethD (CH3)3SiH HD (CH3)3Si (16) ylsilane concentration) were present in the products. A plot of D2/HD as a function of P ( C ~ D ~ ) / P ( ( C H B ) ~isS ~ Hfor ~ )the rate constant ratio k3/k16 = 2.28 $: 0.07. Bond Energy Bond Order (BEBO) Calculations. Acshown in Figure 1. From the slope the ratio of the rate cording to this meth0d,~3,34the potential energy of the constant of addition reaction 3 to that of the abstraction system, V , along the line of constant bond order, m = 1 reaction n, is given by D fCH3)2SiH2+HD (CH3)zSiH (15) V = E1,(1 - n p ) - Ezsm4 + Vt,. (11) 1s k 3 / k i b = 1.75 A: 0.04. where A possible source of methane formation is the displaceV,, = 0.5&,l[ex~(-b'r3)1[~ + 0.5 exp(-flr3)1\ ment reaction (111) r3 = R1 R2 D i- !CH3)&iH2 CH3 CH3SiHD Rl = R1,- 0.26 In n followed by H atom abstraction by methyl. However, this R2 = Rzs - 0.26 In m mechanism can be discounted since the Hg 6(3P1) plus dimethylsilane reaction, which leads to exclusive Si-H bond (32) K . Obi, A . Clement, H. E. Gunning, and 0. P. Strausz, J. Amer. Chem. Soc., 91,1622 (1969). cleavage, does not give rise to the production of CH4.20 (33) H. S. Johnston and C. Parr, J. Amer. Chem. Soc., 85,2544 (1963). Consequently the small amount of CHI formed in the (34) H. S. Johnston, "Gas Phase Reaction Rate Theory," Ronaia Press, New York, N. Y . , 1966, p339. present system must arise from the direct photolysis of
+
+-
-
+
-
+
The Journai o f Ph;/sica/ Chemistry, Voi. 76, No. 26, 1972
+
+
a The bone energy irdices used are H-H = 1.041 and C-H = 1.087 lnciudes zero point energy. (ref 34); ana Si-k! =: 1.004 (ref 5 ) . CWeighted average cf Ss-H stretching frequencies taken from R. P. Wollantlsworth and M. A. Ring, inorg. Chem., 7, 1635 (1968). d R . E. Wiide, J. Mol. S p e c t r o x . , 8 , 424 (1962). eCalculaled from the expres. 34, p 82. g Reference sion 0 := 1.2177 X 10' o : ~ ( @ / D ~ ) " *Reference 41. Estimated l r c r the relationship D(CH3,-H)/P(C2HS-W) cx 13(Sii-i3--.H)/D(Sip115--rl). Esrimated using the relationship [D(SiH3-H) D(Mex':;iH3..,--H)I/[D(G:;3-ii) D(MexCk+2-H)] = [D(SIH3-H) D(Mel!ji-n)]/[o(Cn,-lij - D(Me3C-H)]. J Reference 42. Reference 40. 1-1
V I I!: Potential ~ ? of Activation ~ and ~ Experimental ~ ~ Activation Energies 01 ydregen Abstraction by H Atoms
6
~
~
_ _ _ _ _ I _ _
I___~s___l_l_lii__
Substrate
V
kea1 mol-'
Ea, kcai r i a l
14.0 10.6 9.62
-'
;2.Qb 9.4c 8.2C
7.8 9.1 (6.2) 7.0 (4.8)
6.BC 2.5-3.9,d 4.1e I .4-2.8d 2,3-3.7d 2.1-3.4d 1.8-3.2,d 2.301
5.6 (3.6) 5.1 (3.3) 4.4 (2.80)
i n parentheses are tile values calculated using modified Vtr, eq !V. M . J. Laidler, "Chemical Kinetics," McGraw-Hill, New York, N. Y., !9E5, 12 129. 8 . A . Thrush, Progf. React. Kinet.. 3, 89, 95 (1965). Estimated assunning an A factor in the range 5 X 1012-5 X 1013 cc mol-' !;ec-! per S,--H b o n d @Reference17. Reference 16.
*
p and. q are the b'ond energy indices, E,,, EZs,and E3a are the sin.gle bond energies, R l s , RzS, and Rss denote the single bond distances, i?~ and Rz are the progress distances, 3 i s the Morse partmeter o f the R-R' bond in the triatomic transition state 12---H---R' (see ref 34 for a detaiied discussion and notation), and Vtr represents the
triplet interaction energy between the end atoms. This expression has been wjdely used to predict the activation energk?s for e v a r i e ~ yof H atom abstraction reactions and the calculated values are found to agree with experiment to within A 2 kcaJ m o k l . Tabie lists the bond properties needed in the computations. The final results are given in 'Table VlZI which also contains the estimated values for siianes fuicle inj5-a) and the literature values for alkane reactions. ~
~
~
~
~
$
~
In carder to calcu1ai;e absolute rate constants from reiative rate measurements, 123 must be estimated. For the analogous reaction the high-pressure raw constants reported in the litera38 are in good agreement, yielding an average value of 6.7 x llcill cc mol-3 sec--I. The rate of the H plus CzD4
Q
W. Braunand M. Lenzi, Discuss. FaradaySoc., 44,252 (1967). R, 3 . Penshorn and B. deB. Darwent, J. Chern. Phys., 65, 1508 (1971). J. ~A . Eyre, T. liikida, and t.. M . Dorfman, J. Chem Phys.. 53, 1281 (1 970). VI. J. Kuryla, N. C . Peterson, and W . Braun, J Chem. Fhys., 5 3 , 27713 (1970) A . F. Trotman-Dickenson, Advan. Free-.Rad!cs/ Chew.. '1, 7 (1965). J. A K w r , Chem. Rev., 68,465 (1966). F. E. Saafold and J . J, Svec, J. Phys. Chem.. 7 0 , 1753 (l,966). S. J. .&?id, I . ?ri. -7. Cavidson, and C. 4. Lar,:kert, J . Chem. SOC.A, 2068 ( 1 968) N. I . Parsamyan, E. A . Arakelyan, V. V. Azatyan, ana A . b. 'Nalbandyen, l z v Akad. N a u k S S S R , Ser. Khim.. 1196 (1968). IN I . Parsamyan and A. B. Nalbandyan, l i v . Akad. !dauk SSSR, Ser. Khim., 750 f: 9 6 8 )
The Journal of Physical Chemistry. Voi. 76, iV0. 26, 7972
391 6
K. Obi, H. S . Sandhu, H. E. Gunning, and 0. P. Strausz
TABLE IX: Kinetic Data of Hydrogen Transfer Reactions of Silanes Db
x lo-’’ CH3 cc mol-’ sec-’ Per Si-H Log A, Total bond cc mol-’ sec-’ kabstr
Substrate
SiHB-H C H $3H2 -- li (CH3)2SiH-H (CH3)3Si-H SiW3SiH2-H
k3/kabstra
1.88 5 0 . 0 7 1.72 rk 0.02 1.75 f 0.04 2.28 rk 0.04 0.20 f 0.04
2.82, 1.32c 3.08 3.03 2.94, l . l l d 26.5, 13.0e
0.71 1.03 1.51 2.32 4.41
CF3 Ea9
kcal mol-’
Log A , cc mol-’ sec-’
____ Ea I
kcal mol-’
1?.80,f11,82g
6.99,16.89g
11.90,h11,90’
4.93,h5.111
11.34s 11.967
7.83g 5.63f
12.26i
5.561
a Errors quoted are based on 90% confidence limit. Present work, temperature 25”. Absolute rate constants calculated with k 3 = 5.3 X 10” cc mol-’ sec-’ (see text). CLower limit for H plus SiH4 reaction, ref 17. dReference 16, log A = 12.73 (cc mol-’), Ea = 2.3 kcal mol-’. e H plus SizHe reaction, ref 19, f Reference 5. g Reference 9. Reference 6 . Reference 10.
sec-1. This is an order of magnitude smaller than those of the alkane reactions. On the other hand, the A factors for the reactions of methyl and trifluoromethyl radicals with several silanes have been found to have values essentially identical with those of the alkanes. The cause of this discrepancy is not understood but clearly more experimental work will be required for the elucidation of the problern. If we now arbitrarily assign a value to the A factor of the H plus silane systems then it is possible to estimate the magnitude of the activation energies involved. The values derived this way by taking a range of A from 5 X 1OI2 to 5 x 10’3 cc mo1-l sec-1 per Si-H bond are listed in Table VI11 in comparison with the measured activation energies of the alkane reactions and calculated potential energies of activation for both systems. As seen from the table, the measured and calculated values are in good agreement for the alkanes; for the silanes, however, the computed values are persistently too high. The discrepancies, which were also found before in the CHs and CF3 plus silane systems,”S-S cannot be attributed to possible inaccuracies in the input parameters because the spectroscopic and bond eneirgy data for the monosilane molecule are reasonably well established, and it is highly unlikely that they could lend to a 5-6 kcal mol-1 error. One crucial. point in the BEBO method is the estimation of the triplet repulsion term, Vtr. In the original formulation of the method, Johnston and Parr33 assumed Sato’~ anti-Morse ~~ function with a coefficient of 0.5, eq 111, to represent the triplet repulsion between two H atoms. This same function, however, would not be expected to provide a satisfactory description of the repulsive
The Journal of Physical Chemistry, Vol. 76,
No. 26, 1972
term in cases where one or both end atoms are different from hydrogen. Also, M a t ~ e nhas ~ ~questioned, on quantum mechanical grounds, the hypothesis that the repulsive term arises only from spin-spin interactions. Thus, the overestimation of the activation energy in the hydrogen transfer reactions of silicon hydrides is probably related to the inadequacy of the anti-Morse function in the description of the repulsive potentials. Since, apart from the case of the Hz molecule, neither experimental nor computed repulsive potential data are available, we modified the potential energy function, V,,, on empirical grounds by introducing a coefficient (Y into the exponential factors Vtr =
0.5E3sllexp(-aprs)][l -t0.5 exp(--aflr.dlj
((IV)
The value of a was determined from the measured activation energy of the CM3 + SiH4 reaction. The potential energies of activation computed with this modified potential function, listed in Table VIII, are in reasonable agreement with the experimentally estimated values of the action energies. It should be mentioned that other modifications of Vt, have also been suggested recently in the literature.47-49
Acknowledgment. The authors thank the National Research Council of Canada for financial support. (45) S. Sato, J . Chem. Phys., 23, 2465 (1955). (46) F. A. Matsen, J. Amer. Chem. SOC.,92,3525 (1970). (47) S. W. Mayer, L. Schieler, and H. S. Johnston, Proc. int. Symp. Combust., 711h, 387 (1967) (48) S. W. Mayer, J. Phys. Chem., 73,3941 (1969), (49) C. M. Previtali and J. C. Scaiano, J. Chem. Soc. 5, 2317 (1971).
