Reactivity of tris(trimethylsilyl)silane toward diarylaminyl radicals

Article abstract of DOI:10.1021/jo010408u

Absolute rate constants for the reaction of tri-tert-butylphenoxyl radical (ArO^·) with (TMS)₃SiH were measured spectrophotometrically in the temperature range 321-383 K. Rate constants for the hydrogen abstraction from (TMS)₃SiH by diarylaminyl radicals of type (4-X-C₆H₄)₂N^· were determined by using a method in which the corresponding amines catalyze the reaction of ArO^· with (TMS)₃SiH. At 364.2 K, rate constants·are in the range of 2-50 M^-1 s^-1 for X = H, CH₃, CH₃O, and Br, whereas the corresponding value for ArO^· is 3 orders of magnitude lower. A common feature of these reactions is the low preexponential factor [log(A/M^-1 s^-1) of 4.4 and 5.2 for ArO^· and Ph₂N^·, respectively], which reflects high steric demand in the transition state. A semiempirical approach based on intersecting parabolas suggests that the observed reactivity is mainly related to the enthalpy of the reaction and allowed to estimate activation energies for the reaction of (4-X-C₆H₄)₂N^· and ArO^· radicals with a variety of silicon hydrides.

Full text of DOI:10.1021/jo010408u

J . Org. Chem. 2001, 66, 6317-6322
6317
Rea ctivity of Tr is(tr im eth ylsilyl)sila n e tow a r d Dia r yla m in yl
Ra d ica ls
†
†
,‡
Vladimir T. Varlamov, Evguenii T. Denisov, and Chryssostomos Chatgilialoglu*
Institute of Problems of Chemical Physics, Russian Academy of Sciences, Chernogolovka,
Moscow oblast, 142432 Russia, and I.Co.C.E.A., Consiglio Nazionale delle Ricerche,
Via P. Gobetti 101, 40129 Bologna, Italy
chrys@area.bo.cnr.it
Received April 20, 2001
•
Absolute rate constants for the reaction of tri-tert-butylphenoxyl radical (ArO ) with (TMS)
3
SiH
were measured spectrophotometrically in the temperature range 321-383 K. Rate constants for
•
the hydrogen abstraction from (TMS)
3
SiH by diarylaminyl radicals of type (4-X-C
6 4 2
H ) N were
•
determined by using a method in which the corresponding amines catalyze the reaction of ArO
-
1
-1
with (TMS)
CH O, and Br, whereas the corresponding value for ArO is 3 orders of magnitude lower. A common
feature of these reactions is the low preexponential factor [log(A/M
3
SiH. At 364.2 K, rate constants are in the range of 2-50 M
s
3
for X ) H, CH ,
•
3
-
1
-1
•
s
) of 4.4 and 5.2 for ArO
•
and Ph
2
N , respectively], which reflects high steric demand in the transition state. A semiempirical
approach based on intersecting parabolas suggests that the observed reactivity is mainly related
to the enthalpy of the reaction and allowed to estimate activation energies for the reaction of
•
•
(
6 4 2
4-X-C H ) N and ArO radicals with a variety of silicon hydrides.
In tr od u ction
mechanism. The ability of poly(hydrosilane)s to stabilize
polypropylene during multiple extrusion is believed to
be due to the synergism of hydrogen donation to carbon-
centered radicals5 and its capability to scavenge the
The reaction of atoms or radicals with silicon hydrides
is the main way of generating silyl radicals (eq 1), which
play a strategic role in diverse areas of science, i.e.,
6
traces of oxygen present during the extrusion process.
1
organic synthesis, polymers, and materials. All available
A large number of substituted phenols and aromatic
amines have extensively been explored as autoxidation
kinetic data for reaction 1 as well as for the analogous
reaction of germanium and tin hydrides have recently
been collected and critically discussed.2
7
inhibitors. All these compounds are known to be excel-
lent traps for peroxyl radicals and to give, by hydrogen
abstraction from OH or NH groups, persistent phenoxyl
and aminyl radicals. Sulfur and phosphorus compounds
have been used in commercial practice in conjunction
Polysilanes, whose backbones consist entirely of silicon
atoms, have been shown to possess a number of interest-
ing chemical and physical properties, which have enor-
8
with hindered phenolic antioxidants as synergists. The
reactivity of alkylperoxyl radicals toward poly(hydro-
silane)s is a few orders of magnitude lower than the
3
mous potential for technological applications. A charac-
NH.9 We thought that the
reaction with ArOH or Ar
combined action of ArOH or Ar
2
teristic of poly(hydrosilane)s is the presence of Si-H
moieties in each silicon atom of the backbone. The
importance of poly(hydrosilane)s in radical chemistry is
manifested in their applications as processing stabilizers
for organic polymeric materials subject to oxidative
degradation.4 The degradation of polyolefins during
processing takes place by a widely accepted free radical
2
NH with poly(hydro-
silane)s could be of great importance provided that the
hydrogen donor ability of SiH moieties toward the
corresponding phenoxyl or aminyl radicals is reasonable.
2
Since rate constants for these reactions are lacking, we
(4) (a) For the use of poly(hydrosilane)s as process stabilizers, see:
Carrozza, P.; Borzatta, V.; Chatgilialoglu, C. PCT Int. Appl. WO97/
02322, 1997; U. S. patent 6,005,036, 1999. (b) For the use of polysilane
stabilizers containing sterically hindered amine groups, see: Carrozza,
P.; Borzatta, V.; Chatgilialoglu, C.; Da Roit, G. PCT Int. Appl. WO00/
64965, 2000.
(5) Chatgilialoglu, C.; Ferreri, C.; Vecchi, D.; Lucarini, M.; Pedulli,
G. F. J . Organomet. Chem. 1997, 545/ 546, 475-481.
(6) Chatgilialoglu, C.; Guerrini, A.; Lucarini, M.; Pedulli, G. F.;
Carrozza, P.; Da Roit, G.; Borzatta, V.; Lucchini, V. Organometallics
1998, 17, 2169-2176.
(7) Denisov, E. T.; Denisova, T. G. Handbook of Antioxidants. Bond
Dissociation Energies, Rate Constants, Activation Energies and En-
thalpies of Reactions, 2nd ed.; CRC Press: Boca Raton, 1999.
(8) Allen, N.S.; Edge, M. Fundamentals of Polymer Degradation and
Stabilization; Elsevier: London, 1992.
*
To whom correspondence should be addressed. Phone: 39-051-
6
398309. Fax: 39-051-6398349.
†
Russian Academy of Sciences.
Consiglio Nazionale delle Ricerche.
‡
(
1) Chatgilialoglu, C. Chem. Rev. 1995, 95, 1229-1251. Chatgilia-
loglu, C.; Schiesser, C. H. In The Chemistry of Organic Silicon
Compounds: Supplement; Rappoport, S., Apeloig, Y., Eds.; Wiley:
London, 2001; in press.
(
2) Chatgilialoglu, C.; Newcomb, M. Adv. Organomet. Chem. 1999,
4, 67-112.
3) For example, see: (a) West, R. In Comprehensive Organometallic
4
(
Chemistry II; Abel, E. W., Stone, F. G. A., Wilkinson, G., Eds.;
Pergamon: Oxford, U.K., 1995; Vol. 2; Chapter 3. (b) Silicon-Based
Polymer Science: A Comprehensive Resource; Ziegler, J . M., Fearon,
F. W. G., Eds.; Advances in Chemistry Series 224; American Chemical
Society, Washington, DC, 1990. (c) West, R. In The Chemistry of
Organic Silicon Compounds; Patai, S., Rappoport, Z., Eds.; Wiley:
Chichester, 1989; Chapter 19. (d) Miller, R. D.; Michl, J . Chem. Rev.
(9) The rate constants of cumylperoxyl radical with (TMS)
3
SiH and
poly(phenylsilane) are 66.3 and 24.7 (referring to each PhSiH moiety)
-1
-1
M
s
at 345.5 K, respectively; see: Chatgilialoglu, C.; Timokhin, V.
I.; Zaborovskiy, A. B.; Lutsyk, D. S.; Prystansky, R. E. J . Chem. Soc.,
Perkin Trans. 2 2000, 577-581.
1
989, 89, 1359-1410.
1
0.1021/jo010408u CCC: $20.00 © 2001 American Chemical Society
Published on Web 08/21/2001

6
318 J . Org. Chem., Vol. 66, No. 19, 2001
SiH,10 as a reliable
Varlamov et al.
chose tris(trimethylsilyl)silane, (TMS)
model for this purpose. However, it was recently reported
that the reaction of phenothiazinyl radical (Ar
3
•
2
N ) with
(
TMS) SiH does not occur, being endothermic of 20 kJ /
3
1
1
mol. Herein we report a kinetic study of the reaction of
tri-tert-butylphenoxyl radical (ArO ) with (TMS)
which is 11 kJ /mol endothermic as well as a number of
•
3
SiH
1
2
•
diarylaminyl radicals (Ar
-15 kJ /mol exothermic.15 We also applied a semi-
empirical model of intersecting parabolas for estimating
2 3
N ) with (TMS) SiH which are
0
•
activation energies for the reaction of Ar
2
N with a variety
of silicon hydrides.7
,18
Resu lts a n d Discu ssion
Rea ction of 2,4,6-Tr i-ter t-bu tylp h en oxyl Ra d ica l
w ith (TMS)
radical (1) with (TMS)
3
SiH. The reaction of tri-tert-butylphenoxyl
SiH in chlorobenzene as the
3
solvent under Ar was followed spectrophotometrically
F igu r e 1. (a) Plot of ln[1] vs time: at 321.5 K with 74.4 mM
either at 400 or 625 nm, where the radical 1 absorbs with
of (TMS)
3
SiH (line 1); at 343.7 K with 72.9 mM of (TMS)
3
SiH
3
2
-1
-1
extinction coefficients of 2 × 10 and 4 × 10 M cm
,
(line 2); at 383.2 K with 49.1 mM of (TMS)
initial time refers to the beginning of the clean slow decay (see
text and ref 21). (b) Dependence of the observed first-order
3
SiH (line 3). The
respectively.1
9,20a
In all experiments, the kinetic curves
of radical 1 disappearance showed a short and fast decay
at the initial phase followed by a long and slow process.
The later time profile was used for the kinetic investiga-
tion.21
3
rate constant, kobs, from the concentration of (TMS) SiH. Initial
-4
concentration of radical 1 in the range of 1.03 × 10 - 7.12
-4
-4
×
10 M (λmon ) 400 nm for [1] e 3 × 10 M and λmon ) 625
-4
nm for [1] > 3 × 10 M).
When a large excess of tris(trimethylsilyl)silane is
used, i.e., [(TMS)
radical 1 disappearance followed clean first-order kinetics
Figure 1a). The pseudo-first-order rate constants, kobs
3
SiH] g 100 × [1], the time profile of
3
were proportional to [(TMS) SiH] and independent of [1]
(Figure 1b). These results suggest that the hydrogen
abstraction from silane by radical 1 is the rate determin-
ing step (eq 2). In analogy with the well-known reaction
of aryloxyl radicals with hydroperoxides and hydrocar-
(
,
(
10) (a) Chatgilialoglu, C. Acc. Chem. Res. 1992, 25, 188-194. (b)
Chatgilialoglu, C.; Ferreri, C.; Gimisis, T. In The Chemistry of Organic
Silicon Compounds, Vol. 2; Rappoport, S., Apeloig, Y., Eds.; Wiley:
London, 1998; pp 1539-1579.
22
bons, the silyl radical should react with another radical
1 (eq 3).23 For [(TMS)
SiH] . [1], the kinetic expression
describing the consumption of radical 1 is shown in eq
, where kobs ) 2k [(TMS) SiH].
3
(11) Lucarini, M.; Pedrielli, P.; Pedulli, G. F.; Valgimigli, L.; Gigmes,
D.; Tordo, P. J . Am. Chem. Soc. 1999, 121, 11546-11553.
4
2
3
(
12) Based on the following bond dissociation energies (in kJ /mol):
1
,2,13
14
D° [(TMS)
t-Bu-C
3
SiH] ) 351,
and D° [ArOH] ) 339.9, where Ar ) 2,4,6-
6
H
2
.
(
13) Laaehoven, L. J . J .; Mulder, P.; Wayner, D. D. M. Acc. Chem.
Res. 1999, 32, 342-349.
14) Mahoney, L. R.; Ferris, F. C.; DaRooge, M. A. J . Am. Chem.
(
Soc. 1969, 91, 3883-3889. Mahoney, L. R.; Mendenhall G. D.; Ingold,
K. U. J . Am. Chem. Soc. 1973, 95, 8610-8614.
(
15) The D° [Ar NH] values were determined by an equilibration
2
16
method using UV-vis detection (see Table 4) or by combination of
the equilibrium acidities and the oxidation potentials of their conjugate
anions.17 The two sets of data are similar, the later being slightly
2
higher (1-4 kJ /mol). Values of 359 and 364 kJ /mol for D° [Ph NH]
1
1,13
are also recently reported.
16) Varlamov, V. T.; Denisov, E. T. Bull. Acad. Sci. USSR, Div.
(
Chem. Sci. 1990, 657-661 (Translated from Izv. Akad. Nauk SSSR,
Ser. Khim. 1990, 743-749).
(
17) Bordwell, F. G.; Zhang, X.-M.; Cheng, J .-P. J . Org. Chem. 1993,
5
8, 6410-6416.
18) Denisov, E. T. Mendeleev Commun. 1992, 2, 1-2. Denisov, E.
T. Kinet. Catal. 1994, 35, 617-635 (Translated from Kinet. Katal. 1994,
5, 671-691). Denisov, E. T. Russ. Chem. Rev. 1997, 66, 859-876
(
3
(
Translated from Usp. Khim. 1997, 66, 953-971). Denisov, E. T. In
General Aspects of the Chemistry of Radicals; Alfassi, Z. B, Ed.;
Wiley: Chichester, 1999; pp 79-137.
(
19) Cook, C. D.; Norcross, B. E. J . Am. Chem. Soc. 1959, 81, 1176-
The rate constants obtained from the slope of the linear
plot of kobs vs [(TMS) SiH] are reported in Table 1. Linear
regression analysis of log(k s ) vs 1/T plot yields
1
180.
(
20) (a) Varlamov, V. T. Kinet. Catal. 1992, 33, 25-32 (Translated
3
0
-
1
-1
from Kinet. Katal. 1992, 33, 27-35). (b) Varlamov, V. T.; Denisov, E.
2
/M
T. Kinet. Catal. 1989, 30, 943-946 (Translated from Kinet. Katal. 1989,
the Arrhenius parameters given in Table 2.
3
0, 1079-1083).
(21) The following observations suggest that the short and fast decay
at the initial phase is due to the reaction of 1 with “impurities” present
in the stock solution of silane: (i) the consumption of radical 1 at the
initial phase is not temperature dependent but increases by increasing
silane concentration, (ii) addition of parent phenol in concentration
from 10 to 50 times higher than 1 does not influence the reaction decay
at the initial phase, and (iii) faster decay of 1 at the initial phase is
(22) Neuman, M. B.; Mamedova, Yu. G.; Blenke, P.; Buchachenko,
A. L. Dokl. Nauk SSSR 1962, 144, 392-394. Roginskii, V. A. Phenolic
Antioxidants; Nauka: Moscow, 1988; pp 53-128.
•
•
(23) The removal of (TMS)
than the backward reaction 2, i.e., k
can be estimated to be in the range of 0.09-0.7 M
temperature range 283 - 322 K from the relation ∆H ≈ -2.3RT logK,
3
Si by ArO (eq 3) is expected to be faster
•
3
[ArO ] . k-2[ArOH]. Indeed, k-2
-1 -1
s
at the
observed for a prolonged storage of the (TMS)
3
SiH solution in chlo-
8
-1 -1
robenzene. To minimize the disappearance of 1 with impurities, we
where K ) k
2
/k-2. Assuming k
3
≈ 10
M
s , it can be calculated
adjusted the concentration of (TMS)
3
SiH so that less than 20% of the
that the backward reaction 2 is unimportant under the present
experimental conditions.
starting 1 is consumed by this way.

Reactivity of Tris(trimethylsilyl)silane
J . Org. Chem., Vol. 66, No. 19, 2001 6319
Ta ble 1. Ra te Con sta n ts for th e Rea ction of Ra d ica l 1
w ith (TMS)3SiH a t Va r iou s Tem p er a tu r esa
The amine is regenerated during the reaction 5, and
then it returns into the catalytic cycle (eqs 6/-6). The silyl
radical formed in reaction 5 should decay by recombina-
tion with radical 1 as happens in the absence of the
catalyst (eq 3), since the concentration of radical 1
exceeds by several orders of magnitude the overall
concentration of all the other radicals. We predicted that
1
T, K
k2, M- s-1
-
-
-
-
3
3
2
2
3
21.5
43.7
64.2
83.2
(1.78 ( 0.01) × 10
(4.97 ( 0.05) × 10
(1.19 ( 0.02) × 10
(2.49 ( 0.05) × 10
3
3
3
a
the decay of radical 1 in the presence of (TMS)
3
SiH
Chlorobenzene as the solvent.
should be accelerated by addition of Ar NH. In the
absence of side reactions, amines behave as the catalyst
because they are not consumed and do not affect the
2
Rea ction s of Dia r yla m in yl Ra d ica ls w ith (TMS)
3
-
SiH. We thought that the rate constants for the reaction
•
2 3
of a variety of diarylaminyl radicals (Ar N ) with (TMS) -
composition of the products. To achieve this scenario, the
SiH (eq 5) could be determined by using a method in
which an amine catalyzes the reaction of radical 1 with
26
[Ar
2
NH] and [2] should be used as higher as possible
whereas the ratio [Ar
2
NH]/[2] should be kept at mini-
mum. The [1] and [(TMS) SiH] should also be kept
(
TMS)
3
SiH (eq 2). This approach is based on the known
27
3
•
2
methodology used for determining the reactivity of Ar N
relatively low and high, respectively.
Under such experimental conditions, the time profile
of radical 1 disappearance in the presence of five amines
of type (4-X-C H ) NH (cf. structures 3-7 for the cor-
6 4 2
responding aminyl radicals) followed clean first-order
kinetics. Two examples are shown in Figure 2a and 2b
toward hydroperoxides24 and hydrocarbons. We briefly
20
describe the method and the necessary modifications. The
reactions of radical 1 with Ar
2
NH are relatively fast and
reversible (eqs 6/-6). For example, k
6
and k-6 are in the
2
4
6
7
-1 -1
range of 10 -10 and 10 -10 M
s
at 294 K, respec-
2
5
tively, depending on the nature of the aryl moieties and,
therefore, several orders of magnitude higher than k
2
.
It is possible to reach a quasi-equilibrium regime (eqs
/-6), if 2 and Ar NH are added to the solution of radical
and (TMS) SiH. The time (τ) to reach the quasi-
6
1
2
for Ph
2
NH and (4-MeO-C
6 4
H )PhNH in the absence (line
3
1
) and in the presence (lines 2 and 3) of amine. In the
equilibrium conditions can be estimated from the relation
presence of amine, the overall rate of radical 1 dis-
appearance is equal to the sum of “noncatalyzed” and
-
1
-3
-2
-3
τ ≈ (k-6[2]
0 0
) . For [2] ) 10 -10 M, a τ value of 10 -
-
5
1
0
s is calculated, which means that a quasi-equilib-
“catalyzed” reaction rates and, consequently, kobs is given
rium is reached immediately after mixing the reactants.
Under these conditions, some of the radical 1 will be
by eq 7 which, for practical purposes, can be written as
eq 8.
•
•
replaced by Ar
2], where K
hydrogen abstraction by Ar
higher than that for the analogous reactions of radical 1
2
N radical, i.e., [Ar
2
N ] ) K
6 2
[1][Ar NH]/
[
6
) k /k-6. Since the rate constants for
6
•
2
N radical are usually much
2
0
(
4-6 orders of magnitude), the replacement of 1 by
2
N radical, even at a low extent, suffices for performing
•
Ar
the desired measurements.
Ta ble 2. Ar r h en iu s P a r a m eter s for th e Rea ction of (TMS)3SIH w ith Som e P er sisten t Ra d ica ls

6
320 J . Org. Chem., Vol. 66, No. 19, 2001
Varlamov et al.
rameters given by eq 9, where θ ) 2.3RT kJ /mol and the
errors correspond to one standard deviation.
log(k K /M s-1) ) (5.23 ( 0.28) - (50.8 ( 1.90)/θ
-
1
5
6
(9)
5 6
Combination of the Arrhenius expression for k K
with
the known parameters30 for the equilibrium 6/-6 yields
the temperature dependence of k (Table 2). The value
5
1
-
-1
of k
5
can be calculated as 4.4 M
s
at 298 K.
The Arrhenius parameters obtained for the reaction
•
2 3
of radical 1 and Ph N with (TMS) SiH and the analogous
3
1
reaction of 2,2,6,6-tetramethylpiperidinyl radical are
reported in Table 2 for comparison. A remarkable feature
of the three reactions is the low value of logA which is
normally in the range of 8.5-9.0 for hydrogen abstrac-
tion.32 The low value can be explained in terms of the
high steric demand for a reaction in which both sub-
strates are sterically hindered.33 Indeed, the steric con-
straints in the transition state should manifest both in
the entropy of activation and the energy of activation,
the former being more pronounced due to the loss of
rotational and vibrational degrees of freedom.
F igu r e 2. Plot of ln[1] vs time in the absence and presence
of Ar NH. The initial time refers to the beginning of the clean
slow decay (see text and ref 21). (a) At 383.2 K with 49.1 mM
of (TMS) SiH. Initial concentration of compounds 2 and Ph
NH, respectively: 1.5 and 0 mM (line 1); 4 and 1 mM (line 2);
2
3
2
-
3
2
and 6 mM (line 3). (b) At 364.2 K with 50.1 mM (line 1 and
) or 64.4 mM (line 3) of (TMS) SiH. Initial concentration of
)PhNH, respectively: 10 and
mM (line 1); 6 and 0.47 mM (line 2); 3 and 0.47 mM (line 3).
Figure 4 shows the dependence of kobs/2[(TMS)
the ratio [Ar NH]/[2] at 364.2 K for other amines. The
values for these amines including Ph NH are
reported in Table 4 together with the equilibrium con-
3
SiH] on
3
compounds 2 and (4-MeO-C
0
6
H
4
2
k
5
K
6
2
1
6
stants and D°(Ar
2
NH) taken from the literature and the
values. The reactivity of (4-X-
N radicals with (TMS) SiH is increasing with the
decrease of the electron-donating properties of substitu-
calculated absolute k
5
•
C
6
H
4
)
2
3
-
1
ents X, the rate constants being in the range 2-46 M
-
1
s
at 364.2 K. A reasonably good linear correlation was
2
5 2
found when logk was plotted vs D°(Ar NH) (r ) 0.95)
indicating that the observed effects are primarily caused
by the ability of substituent to stabilize the aminyl
+
radical. A plot of logk
5
against the Σσ
p
substituent
3
4
+
constants also shows a good linear relation with F )
2
0
.73 (r ) 0.96), which is not surprising in view of the
fact that correlation was also observed when D°(Ar NH)
values, with F ) 9.02 (r ) 0.97).
2
+
+
2
was plotted vs Σσ
p
Therefore, polarized transition states should be un-
important and the reactivity trend simply reflects the
(25) Varlamov, V. T.; Denisov, N. N.; Nadtochenko, V. A.; March-
enko, E. P.; Petrov, I. V.: Plekhanova, L. G. Kinet. Catal. 1994, 35,
773-775 (Translated from Kinet. Katal. 1994, 35, 838-841). Varlamov,
V. T. Kinet. Catal. 2000, 41, 320-331 (Translated from Kinet. Katal.
F igu r e 3. Linear regression analysis of kobs/2[(TMS)
Ph NH]/[2] at different temperatures.
3
SiH] vs
[
2
2
000, 41, 353-365).
•
(26) For a quasi-equilibrium the rates of reactions 6 and -6 should
be close to each other and much higher than those of reactions 2 and
Ta ble 3. k5K6Va lu es for th e Rea ction of P h 2N Ra d ica l
w ith (TMS)3SiH a t Va r iou s Tem p er a tu r esa
5
, respectively. Therefore, [Ar
SiH)/k-6
27) Since the cross-coupling of Ar
2
NH].k
2
[(TMS)
3 6 5 3
SiH)/k and [2].k [(TMS) -
1
.
T, K
k5K6, M- s-1
(
2
N• and 1 radicals is unimpor-
-
-
-
-
4
3
3
2
28
•
3
21.5
43.7
64.2
83.2
(9.24 ( 0.63) × 10
(3.01 ( 0.14) × 10
(9.46 ( 0.40) × 10
(1.87 ( 0.18) × 10
tant, the self-termination of Ar
t
reaction. Indeed, these rate constants (2k ) are in the range of 10 -
29
N radicals is the only possible side
2
6
3
3
3
8
-1 -1
10 M
s , depending on the nature of substituents. Therefore, the
rate of the reaction 5 should be much higher than the rate of self-
•
2 5 3 t 6 2
termination of Ar N radicals, i.e., k [(TMS) SiH)/2k K [1].[Ar NH]/
a
Chlorobenzene as the solvent.
[2].
(
28) Karpukhina, G. V.; Maizus, Z. K. Bull. Acad. Sci USSR, Div.
Chem. Sci. 1968, 915-920 (Translated from Izv. Acad. Nauk SSSR.,
Figure 3 shows the dependence of kobs/2[(TMS)
the ratio [Ph NH]/[2] according to eq 8 at different
temperatures, whose slopes provide the k values for
this amine (Table 3). Linear regression analysis of
log(k ) vs 1/T plot yields the relative Arrhenius pa-
3
SiH] on
Ser. Khim. 1968, 957-961).
(
29) Efremkina, E. A.; Khudyakov, I. V. Denisov, E. T. Khim. Fiz.
2
1
987, 6, 1289-1291.
(30) The temperature-dependent function for the equilibrium 6/-6
5
K
6
1
6
6
is as follows: logK ) 0.01-24.7/θ, where θ ) 2.3RT kJ /mol.
(
31) Lucarini, M.; Marchesi, E.; Pedulli, G. F.; Chatgilialoglu, C. J .
Org. Chem. 1998, 63, 1687-1693.
32) Benson, S. W. Thermochemical Kinetics, 2nd ed.; Wiley-Inter-
5
K
6
(
science: New York, 1976.
(
24) Varlamov, V. T. Bull. Acad. Sci. USSR, Div. Chem. Sci. 1989,
(33) Howard, J . A. Adv. Free-Radical Chem. 1971, 4, 49-173.
(34) Hansch, C.; Leo, A.; Taft, R. W. Chem. Rev. 1991, 91, 165-
195.
4
5
82-490 (Translated from Izv. Akad. Nauk SSSR, Ser. Khim. 1989,
49-557).

Reactivity of Tris(trimethylsilyl)silane
J . Org. Chem., Vol. 66, No. 19, 2001 6321
F igu r e 5. Intersection of Si-H and N-H bond potential
curves for the parabolic model.
Ta ble 5. Ca lcu la ted Ee, ∆He, a n d bir e Va lu es for th e
•
Rea ction s of Ar 2N Ra d ica ls w ith (TMS)3SiH
Ar2N• E, kJ /mol Ee, kJ /mol ∆He, kJ /mol bire, (kJ /mol)1/2
3
4
5
6
7
26.3a
30.9b
31.7b
37.4
42.0
42.8
45.3
36.5
-21.1
-13.9
-12.3
-5.0
11.01
11.27
11.29
11.27
10.88
3
F igu r e 4. Linear regression analysis of kobs/2[(TMS) SiH] vs
[
Ar
2
NH]/[2] for amines (4-Br-C
6
H
4
)
2
NH (9), (4-Me-C
6
H
4
)
2
-
b
34.2
NH (b), (4-MeO-C
at 364.2 K.
6
H
4
)PhNH (0), and (4-MeO-C
6
H
4 2
) NH (O)
25.4b
-20.6
a
Taken from Table 2. b Calculated from the k5 values reported
•
-1 -1
Ta ble 4. Kin etic Da ta for th e Rea ction of Ar 2N Ra d ica ls
w ith (TMS)3SiH a t 364.2 K
in Table 4 by assuming log(A/M
s ) ) 5.3.
energy E (eq 11). The coefficients b
i
) πν
i
(2µ
)1/2 and b
)
i f
D°(Ar2NH),a
k5K6,
-1
Ar2N•
10-2
M
s-1
K6, 10
a
-3
k5, M-1 s-1
kJ /mol
1/2
πν
f f
(2µ )
characterize the relationship between the po-
tential energy and the vibration amplitude of the Si-H
3
4
5
6
7
0.95 ( 0.04
2.48 ( 0.43
2.93 ( 0.14
11.43 ( 0.14
1.57 ( 0.01
0.283
3.41
5.21
46.6
0.343
33.5
7.27
5.63
2.45
45.8
364.7
357.5
355.9
348.6
364.2
2
and N-H bonds, respectively, whereas 2b is the force
constant of a bond.3 The coefficient r
7a
is the distance
e
between the minimum points of the two potential curves
Figure 5). The relationship between these parameters
f
may be represented as shown in eq 12, where R ) b /b .
(
a
From ref 16.
i
differences in the bond dissociation energies of amines,
3
5
D°(Ar
of Ar
magnitude higher than that in the analogous reaction
2
NH). It is interesting to note that the reactivity
•
2
3
N radicals toward (TMS) SiH is ca. 2 orders of
2
0a
2
with PhCMe H, whereas the dissociation energies of
the corresponding Si-H and C-H bonds are very similar.
Sem iem p ir ica l Ca lcu la tion s Ba sed on th e P a r a -
bolic Mod el. One of us has introduced in recent years a
parabolic model for estimating activation energies of
7
,18
hydrogen abstraction reactions.
This semiempirical
method has also been applied for calculating kinetic
parameters of reaction of atoms and radicals with a
variety of silicon hydrides and is herein extended to the
reactions with Ar N radicals. The parabolic model
2
describes the transition state of hydrogen abstraction as
the point of intersection between two potential curves
For reaction 5, the following parameters were used:
11
1/2
-1
R ) b
.5hN
i
/b
c(ν
f
) 0.640, b
- ν
i
) 2.756 × 10 (kJ /mol)
m ,
•
36
0
A
i
f
) ) -7.4 kJ /mol, 0.5hN
A
cν - 0.5RT ) 11.1
i
37b
kJ /mol. The activation energies (E) were estimated by
-1
-1
assuming log(A/M
experimental results reported in Table 2. ∆H
are obtained from eqs 10, 11, and 12, respectively.
s
) ) 5.3, in analogy with the
(
unperturbed) of the stretching vibrations (harmonic)
associated with cleaving Si-H and forming N-H bonds
Figure 5). Along with the enthalpy (∆H ) and activation
energy (E ) of the reaction, this model uses the coef-
ficients b , b , and r . ∆H is the energy difference between
e
, E , and
e
i e
b r
(
e
•
Table 5 summarizes the calculated data for the five Ar
radicals described in the above section.
2
N
e
i
f
e
e
The b
reaction. The b r
i
r
e
value is a general characteristic of a certain
values of the five reactions in Table 5
the minimum points of the two parabolic curves and is
related to the experimental reaction enthalpy ∆H (eq 10).
i e
e
E is the difference between the minimum point of the
Si-H parabolic curve and the point of intersection of the
two curves and is related to experimental activation
(
37) (a) Zero vibration frequency (ν) and reduce mass (µ) of the bond
-1 -1
are expressed in cm and g mol , respectively. The subscripts i and
f refer to the cleaving (Si-H) and the forming (N-H) bonds, respec-
tively. (b) The following parameters are used: ν Si-H ) 2100, ν N-H
)
3
A
337, ν O-H ) 3600, T ) 364.2 K, h ) Planck constant, N ) Avogadro
(35) Zavitsas, A. A.; Fogel, G.; Halwagi, K. E.; Donnaruma Legotte,
constant, c ) light speed.
P. A. J . Am. Chem. Soc. 1983, 105, 6960-6962. Zavitsas, A. A.; Pinto,
(38) Seetula, J . A.; Feng, Y.; Gutman, D.; Seakins, P. W.; Pilling,
M. J . J . Phys. Chem. 1991, 95, 1658-1664.
J . A. J . Am. Chem. Soc. 1972, 105, 7390-7396.
(
36) Denisov, E. T. Russ. Chem. Bull. 1998, 47, 1274-1279 (Trans-
(39) Kalinovski, I. J .; Gutman, D.; Krasnoperov, L. N.; Goumri, A.;
Yuan, W.-J .; Marshall, P. J . Phys. Chem. 1994, 98, 9551-9557.
lated from Izv. Akad. Nauk, Ser. Khim. 1998, 47, 1311-1316).

6
322 J . Org. Chem., Vol. 66, No. 19, 2001
Varlamov et al.
dissociation energies,12 the values of ∆H
) 55.2 kJ /mol, and b ) 11.72 (kJ /mol)
obtained. Using these values together with the appropri-
ate bond dissociation energies of silanes and applying eqs
) 2.2 kJ /mol,
1/2
Ta ble 6. Ca lcu la ted Activa tion En er gies (E, k J /m ol) for
e
•
•
th e Rea ction of (4-X-C6H4)2N (3-7) a n d Ar O (1) Ra d ica ls
E
e
i
r
e
are
w ith a Va r iety of Silicon Hyd r id es
D°(Si-H),
kJ /mol
silane
3
4
5
6
7
1
1
0, 13 and 11, the activation energies for hypothetical
a
H4Si
384.1 ( 2.0 40.0 43.1 43.8 47.2 40.2 57.5
hydrogen abstraction from a variety of silanes by the
radicals 1 can be obtained (Table 6, last column).
b
MeSiH3
Me3SiH
Et3SiH
PhSiH3
PhMeSiH2
Me3SiSi(H)Me2 378
TMS)3SiH
MeS)3SiH
386
40.8 43.9 44.7 48.1 41.0 58.4
c
397.4 ( 2.0 45.9 49.3 50.0 53.7 46.1 63.8
b,d
398
377
382
46.1 49.6 50.3 54.0 46.4 64.1
37.0 40.0 40.7 43.9 37.2 54.3
39.1 42.2 42.9 46.2 39.3 56.5
37.4 40.4 41.1 44.4 37.7 54.7
27.4 29.9 30.5 33.2 27.6 43.8
32.8 35.5 36.2 39.1 33.0 49.6
Exp er im en ta l Section
b
b
Ma ter ia ls. Tris(trimethylsilyl)silane (Fluka AG) was used
as received. Secondary aromatic amines (compounds 3-7) and
2,4,6-tri-tert-butylphenol (2) were purified by silica gel column
chromatography with UV-detection. Chlorobenzene was passed
prior to use in a glass column (inner diameter 3 cm) through
the following layers (layers in cm indicated in parentheses):
b,d
b,d
b,d
(
(
351
366
a
From ref 38. b From ref 1. c From ref 39. d From ref 13.
1
(
5 wt % H
7); KOH (10); Al
Tr i-ter t-bu tylp h en oxyl Ra d ica l (1). The stable radical 1
2
SO
4
on SiO
2 2 4 2 3 2
(3); SiO (2); KMnO on Al O (5); CaCl
2 3
O
(40).
are very close to each other indicating that all these
reactions belong to one class and the difference in the
activation energies is related only to the enthalpy of the
was prepared from the corresponding phenol by oxidation. A
0.15 M solution of tri-tert-butylphenol in chlorobenzene was
slowly (1 drop/min) passed by means of an argon flow through
a glass column (length 5 cm, inner diameter 0.8 cm) filled with
reaction. Therefore, the average value of 11.14 ( 0.17
kJ /mol)1 can be used to calculate the activation param-
/2
(
2
a mixture of PbO and glass wool in order to increase the
eters of the radicals 3-7 with a variety of organosilicon
hydrides. For practical purposes, eq 11 can be written
permeability. At the outlet of the column, the solution of
radical 1 was diluted to 1.5-3 mM under argon and used for
the kinetic experiments. The solution was stored under argon
immediately prior to use. Such a solution was stable in the
dark for several months and a solution of 0.2 mM kept at 383.2
K for 10 min was consumed of e1%.
as eq 13. ∆H
propriate bond dissociation energies, allow to calculate
the corresponding E values, which are transformed into
e
values, obtained from eq 10 using ap-
e
E through eq 11. In Table 6 are reported 45 calculated
activation energies for hydrogen abstraction from a
variety of silanes by the aminyl radicals 3-7.
Kin etic Mea su r em en ts. The disappearance of radical 1
in the presence of (TMS) SiH or (TMS) SiH/Ar NH was fol-
3 3 2
lowed spectrophotometrically either at 400 or 625 nm where
3
2
it absorbs with extinction coefficients of 2 × 10 and 4 × 10
-1
-1
M
cm , respectively. The real concentration of the reactants
was calculated taking into account the thermal expansion of
-
1
the solvent (0.001 deg ). The experiments were carried out
under Ar in a thermostated 8.5 mL cell-reactor having 2 cm
optical path using a Specord UV-vis spectrometer. The
temperature (( 0.2 K) was controlled with a standard variable-
temperature accessory. In all experiments, the kinetic curves
of radical 1 disappearance show a short and fast decay at the
initial phase followed by a long and slow process. The later
The above-described parabolic model has also been
applied to reaction 2. The following parameters were
21
time profile was used for kinetic investigation.
1
1
1/2
-1
used: R ) b
i
/b
) ) -8.9 kJ /mol, 0.5hLν
Taken the activation energy of 43.8 kJ /mol for
f
) 0.590, b
i
) 2.756 × 10 (kJ /mol)
m ,
Ackn owledgm en t. This work was carried out within
the scientific cooperation joint project CNR-RAS.
0
.5hL(ν
i
- ν
f
i
- 0.5RT ) 11.4 kJ /
3
7a
mol.
reaction 2 (Table 2) and using the appropriate bond
J O010408U