Absolute rate constants for the reactions of primary alkyl radicals with aromatic amines

Article abstract of DOI:10.1021/jo952162x

Hydrogen abstraction from diarylamines (4-X-C₆H₄)₂NH [X = H, CH₃, C₈H₁₇, CH₃O, and Br] by the 2-methyl-2-phenylpropyl radical in n-dodecane solution was investigated by thermolysis of 3-methyl-3-phenylbutanoyl peroxide in the presence of various concentrations of the amines. The reaction is a non-chain process in which the 2-methyl-2-phenylpropyl radical and its rearrangement product, the 2-benzylpropan-2-yl radical, abstract hydrogen from both the solvent and the amine. Cross-disproportionation reactions of the rearranged radical led to the formation of significant amounts of β,β-dimethylstyrene. Rate constants for hydrogen abstraction by the unrearranged, primary alkyl radical from n-dodecane (k^373K = 3.5 × 10³ M^-1 s^-1), diphenylamine (k^373K = 1.3 × 10⁶ M^-1 s^-1), and the substituted diarylamines were determined from the product yields and the known rate constant for the radical rearrangement. From kinetic experiments with AT-deuteriodiphenylamine the deuterium kinetic isotope effect, k_NH/k_ND, was found to be 2.3 at 373 K.

Full text of DOI:10.1021/jo952162x

3778
J . Org. Chem. 1996, 61, 3778-3782
Absolu te Ra te Con sta n ts for th e Rea ction s of P r im a r y Alk yl
Ra d ica ls w ith Ar om a tic Am in es1
Alan Burton, K. U. Ingold,*,2 and J . C. Walton*
School of Chemistry, University of St. Andrews, St. Andrews, Fife KY16 9ST, U.K.
Received December 5, 1995X
Hydrogen abstraction from diarylamines (4-X-C6H4)2NH [X ) H, CH3, C8H17, CH3O, and Br] by
the 2-methyl-2-phenylpropyl radical in n-dodecane solution was investigated by thermolysis of
3-methyl-3-phenylbutanoyl peroxide in the presence of various concentrations of the amines. The
reaction is a non-chain process in which the 2-methyl-2-phenylpropyl radical and its rearrangement
product, the 2-benzylpropan-2-yl radical, abstract hydrogen from both the solvent and the amine.
Cross-disproportionation reactions of the rearranged radical led to the formation of significant
amounts of â,â-dimethylstyrene. Rate constants for hydrogen abstraction by the unrearranged,
primary alkyl radical from n-dodecane (k373K ) 3.5 × 103 M-1 s-1), diphenylamine (k373K ) 1.3 ×
106 M-1 s-1), and the substituted diarylamines were determined from the product yields and the
known rate constant for the radical rearrangement. From kinetic experiments with N-deuterio-
diphenylamine the deuterium kinetic isotope effect, kNH/kND, was found to be 2.3 at 373 K.
Diarylamines find extensive use as radical-trapping
which raises the question we address herein: How fast
is the reaction of diarylamines with carbon-centered
radicals?
antioxidants in lubricating oils exposed to high-temper-
ature service conditions such as the oils used in passenger
cars and jet engines. The increasing severity under
which the new generation of engines are planned to
operate will impose increased thermal and oxidative
stress on the lubricating fluids. In principle, kinetic
modeling studies can provide valuable insights into the
overall behavior of oils at high temperatures. This is
because reliable modeling can pinpoint critical elemen-
tary reactions where a possible chemical modification of
the oil or of an additive would produce a change in the
critical rate constant with a consequent improvement in
the overall stability of the oil/additive system. However,
reliable modeling requires accurate rate constants for all
the elementary reactions known (or thought to be)
involved in these extremely complex systems which are
shown in Scheme 1 in grossly simplified form for the
R• + Ar2NH f RH + Ar2N•
(6)
Resu lts
The radical-clock method7 appeared to be the simplest
procedure for determining k6. Preliminary experiments
using the 5-hexenyl radical clock showed only that this
clock ran “too fast”. Thus, even with 2.4 M diphenyl-
amine at 80 °C, by far the major hydrocarbon product
was methylcyclopentane with the yield of 1-hexene being
too small even for its positive identification. That is, the
rearrangement of the 5-hexenyl radical to the cyclo-
pentylmethyl radical is too fast to compete with hydrogen
abstraction from the amine. We therefore turned to the
much slower neophyl radical clock which was found to
be very suitable for our purposes. Thermolysis of 3-meth-
yl-3-phenylbutanoyl peroxide (1.6 × 10-3 M) in dodecane
at 100 °C yielded tert-butylbenzene, isobutylbenzene, and
â,â-dimethylstyrene, and more importantly, the yield of
tert-butylbenzene relative to the other two products
increased dramatically upon the addition of diphenyl-
amine at concentrations in the range 0.029-0.10 M. The
chemistry which we believe is occurring is portrayed in
Scheme 2 with the products of interest indicated in bold
face.
Sch em e 1
initiation
ln• + RH f R•
(1)
(2)
propagation
R• + O2 f ROO•
ROO• + RH f ROOH + R•
(3)
(1) In partial fulfillment of the requirements for a Ph.D. degree (A.
Burton).
(2) Steacie Institute for Molecular Sciences, National Research
Council of Canada, Ottawa, Ontario, Canada K1A 0R6. This paper is
issued as NRCC No. 39100.
(3) This scheme ignores the complexities introduced by the regen-
eration of Ar2NH from its nitroxide, Ar2NO•, at temperatures g130
°C.4 These complexities can be ignored for our present purposes because
our work has been carried out under oxygen-free conditions.
(4) J ensen, R. K.; Korcek, S.; Zinbo, M.; Gerlock, J . L. J . Org. Chem.
1995, 60, 5396-5400.
(5) (a) Brownlie, I. T.; Ingold, K. U. Can. J . Chem. 1966, 44, 861-
868. (b) Brownlie, I. T.; Ingold, K. U. Can. J . Chem. 1967, 45, 2417-
2425.
termination
ROO• + Ar2NH f ROOH + Ar2N•
ROO• + Ar2N• f nonradical products
(4)
(5)
diarylamine-inhibited autoxidation of a hydrocarbon,
RH.3 Diarylamines are known to be excellent traps for
peroxyl radicals,5,6 reaction 4 (Scheme 1). However,
carbon-centered radicals are also involved in autoxidation
(6) Karpukhin, O. N.; Shlyapintokh, V. Ya.; Zolotova, N. V. Izv.
Akad. Nauk. SSSR, Ser. Khim. 1963, 1722-1727.
(7) Griller, D.; Ingold, K. U. Acc. Chem. Res. 1980, 13, 317-323.
For a more recent review, see: Newcomb, M. Tetrahedron 1993, 49,
1151-1176.
X Abstract published in Advance ACS Abstracts, April 15, 1996.
S0022-3263(95)02162-1 CCC: $12.00 Published 1996 by the American Chemical Society

Reactions of Primary Alkyl Radicals with Aromatic Amines
Sch em e 2
J . Org. Chem., Vol. 61, No. 11, 1996 3779
[PhCMe2CH2C(O)O]2 f 2PhCMe2CH2• + 2 CO2 (7)
kCH
•
PhCMe2CH2• + C12H26
PhCMe2CH2• + Ar2NH
8 P h CMe3 + C12H25 (8)
kNH
8 P h CMe3 + Ar2N• (9)
kr
•
PhCMe2CH2• 98 PhCH2CMe2
(10)
•
PhCH2CMe2 + C12H26 f P h CH2CHMe2 + C12H25
•
(11)
•
PhCH2CMe2 + Ar2NH f P h CH2CHMe + Ar2N•
(12)
(13)
•
PhCH2CMe2 + X• f P h CHdCMe2 + XH
F igu r e 1. Plot according to eq I. Key: Ph2NH, O; Ph2ND, 0;
at [amine] ) 0, product ratio ) 0.33, 4.
•
(X• ) C12H25•, PhCMe2CH2 , PhCH2CMe2, Ar2N•)
•
Table 1 and are also shown graphically in Figure 1
C12H25• + Ar2NH f C12H26 + Ar2N•
Ar2N• + Ar2N• f Ar 2NNAr 2
(14)
(15)
(squares). The slopes of the plots of [unrearranged
373K
NH(D)
product]/[rearranged products] vs [amine] yield k
/
k3r 73K. The value found for diphenylamine and N-deu-
teriodiphenylamine, viz., 30 and 13, respectively (see final
column in Table 1), yield a deuterium kinetic isotope
effect for reaction 9 of 30/13 ) 2.3.
Two points should be noted about Scheme 2. First, the
reaction is not a chain reaction. That is, any radical-
induced decomposition of the peroxide could be ignored
because esters were not detected among the products.8
Second, Scheme 2 includes only two radical-radical
reactions yielding nonradical products: reaction 13 is
required to explain the formation of â,â-dimethylstyrene;
reaction 15 is required to account for the fact that
tetraarylhydrazines were formed in low yields whereas
N-alkylated diphenylamines (i.e., the products of cross-
combination) could not be detected.
A generally more limited set of experiments was
carried out with four 4,4′-disubstituted diphenylamines.
The substituents were bromine, methyl, methoxyl, and
(mostly) tert, tert-octyl (2-neopentylpropane), the last
being a commercial material, Naugalube 438L. The
results of these experiments are listed in Table 1 together
with the values calculated for k3N7H3(KD)/kr373K
.
Discu ssion
A simple kinetic analysis of Scheme 2 yields:
The original, EPR-measured, Arrhenius parameters for
the neophyl radical clock12 (reaction 10) have been revised
by Franz et al.13 as a result of a careful study of the
n-Bu3SnH/neophyl chloride reaction. The Arrhenius
parameters given by Franz et al.13 for this rearrangement
(viz., log(A/s-1) ) 11.5, E ) 11.82 kcal/mol) were based
on the assumption that the neophyl radical would react
with n-Bu3SnH at the same rate as that measured for
n-alkyl radicals.14 However, it would appear that the
Arrhenius parameters for reaction of the neopentyl
radical with n-Bu3SnH are slightly different from those
for an n-alkyl radical.15 Fortunately, the revised13 and
the “corrected” Arrhenius parameters for the neophyl
rearrangement, viz.,
[unrearranged product]
[rearranged products]
[PhCMe3]
)
[PhCH2CHMe2] + [PhCHdCMe2]
kCH[C12H26] + kNH[Ar2NH]
)
(I)
kr
Thus, a plot of the ratio of the yield of unrearranged
product to rearranged products against the amine con-
centration should give a straight line with a slope equal
to kNH/ kr and an intercept equal to kCH[C12H26]/kr. Five
experiments were carried out in the absence of diphenyl-
amine and the mean value for the ratio: [unrearranged
product]/[rearranged products], viz., 0.33, was combined
with the results of experiments carried out at four
different concentrations of diphenylamine (see Table 1)
to construct the plot (circles) shown in Figure 1.10
Similar experiments were carried out with N-deu-
terated diphenylamine,5,11 and the results are given in
(10) The experimental points show some unexpected curvature
which might be due to analytical errors or to a less than complete
reaction scheme.
(11) Made by exchange of Ph2NH in dodecane with a large excess
of D2O and subsequently including a small quantity of D2O in each
reaction vial. See: Howard, J . A.; Ingold, K. U. Can J . Chem. 1962,
40, 1851-1864.
(12) Maillard, B.; Ingold, K. U. J . Am. Chem. Soc. 1976, 98, 1224-
1226, 4692.
(8) The absence of esters can be attributed to the short lifetime of
the peroxide at 100 °C9 and, in most experiments, the presence of an
effective trap for the carbon-centered radicals which might otherwise
have yielded esters and started a chain reaction.
(13) Franz, J . A.; Barrows, R. D.; Camaioni, D. M. J . Am. Chem.
Soc. 1984, 106, 3964-3967.
(14) Chatgilialoglu, C.; Ingold, K. U.; Scaiano, J . C. J . Am. Chem.
Soc. 1981, 103, 7739-7742.
(9) Primary aryl peroxides have k ≈ 10-2 - 10-3 s-1 at 100 °C. See:
Hiatt, R. In Organic Peroxides; Swern, D., Ed.; Wiley-Interscience:
New York, 1971; Vol. 2, Chapter 8.
(15) J ohnston, L. J .; Lusztyk, J .; Wayner, D. D. M.; Abeywickreyma,
A. N.; Beckwith, A. L. J .; Scaiano, J . C.; Ingold, K. U. J . Am. Chem.
Soc. 1985, 107, 4594-4596.

3780 J . Org. Chem., Vol. 61, No. 11, 1996
Burton et al.
Ta ble 1. Rela tive Yield s of P r od u cts Ar isin g fr om th e Th er m a l Decom p osition of 3-Meth yl-3-P h en ylbu ta n oyl P er oxid e
in Dod eca n e a t 100 °C in th e Absen ce a n d P r esen ce of Som e Dip h en yla m in es
[unrearranged prod]/
373K
NH(D)
amine (102 M)
nonec
PhCMe3
0.80
PhCH2CHMe2
(1.0)
PhCHdCMe2
[rearranged prod]a
0.33
k
/k3r 73K b (M-1
)
1.42
(C6H5)2NH
2.9
5.1
6.8
10
2.56
5.72
7.48
17.8
(1.0)
(1.0)
(1.0)
(1.0)
1.64
2.24
2.45
4.02
0.97
1.77
2.17
3.55
30
13
(C6H5)2ND
d
1.8
3.1
4.1
6.2
1.05
1.61
1.96
2.67
(1.0)
(1.0)
(1.0)
(1.0)
0.71
1.06
1.26
1.44
0.61
0.78
0.87
1.09
d
d
d
(4-BrC6H4)2NH
2.3
4.0
5.4
8.1
2.20
2.59
3.73
5.71
(1.0)
(1.0)
(1.0)
(1.0)
3.78
4.22
5.28
8.26
0.46
0.50
0.59
0.62
3.9
53
(4-MeC6H4)2NH
2.2
3.9
5.2
3.35
5.50
8.00
(1.0)
(1.0)
(1.0)
1.34
1.40
1.48
1.43
2.29
3.23
(4-MeOC6H4)2NH
2.6
3.4
5.1
2.65
7.33
30.1
(1.0)
(1.0)
(1.0)
1.27
1.96
7.92
1.17
2.48
3.37
57
(4-C8H17C6H4)2NHe
4.1
5.5
8.2
6.47
8.04
8.93
(1.0)
(1.0)
(1.0)
1.07
1.19
0.80
3.13
3.67
4.96
59
a
b
[PhCMe3]/[PhCH2CHMe2] + [PhCHdCMe2]. Slope of the plot of [unrearranged product]/[rearranged products] vs [amine]. c Mean
of five separate measurements. [PhCMe2CH2D]/[PhCMe3] ) 0.50, 0.53, 0.56, and 0.83 as (C6H5)2ND changes from 1.8 × 10-2 to 6.2 ×
d
10-2 M. e Mixture of isomeric octyl groups, see text.
log(kr/s-1) ) 10.98-10.83/θ
(II)
To obtain slightly more information about reaction 10,
five additional experiments were carried out in the
absence of amine at 80 °C and another five at 60 °C. The
where θ ) 2.3RT kcal/mol, yield essentially the same rate
constant for the rearrangement at 100 °C. That is,
equation II yields k3r 73K ) 4.3 × 104 s-1, whereas Franz
et al.’s13 parameters yield k3r 73K) 3.8 × 104 s-1. Thus,
the overall rate constant for hydrogen abstraction from
dodecane by a primary alkyl radical at 100 °C is
mean rate constant ratios were as follows: (kCH/kr)353K
)
0.51/[C12H26] M-1 and (kCH/kr)333K ) 0.60/[C12H26] M-1
.
Using the correct dodecane molarities at each tempera-
ture16 and the kr values derived from eq II, we obtain
353K
CH
333K
CH
k
) 2.3 x 103 M-1 s-1 and k
) 1.05 × 103 M-1 s-1
.
Although rate constants measured at three temperatures
covering such a small temperature range cannot yield
reliable Arrhenius parameters, the “best” line through
our three points gives,
373K
CH
k
) 0.33 × 4.3 × 104/[C12H26] M-1 s-1 (III)
where [C12H26] is the molar concentration of neat do-
decane at this temperature which we estimate16 to be 4.05
M. That is,
log (kCH/M-1 s-1) ) 7.1-6.1/θ
(VI)
373K
CH
where error limits are not given (for obvious reasons).
The expected pre-exponential factor for this class of
k
) 1.42 × 104/4.05 ) 3.5 × 103 M-1 s-1 (IV)
reactions is 108.5(0.5 M-1 s-1 18
.
Since hydrogen abstraction must occur predominantly at
the alkane’s 10 methylene groups, we obtain
Hydrogen atom abstraction from diphenylamine is, as
would be expected, very much faster than abstraction
from dodecane, i.e.,
k3C7H3K/(CH2) ) 3.5 × 102 M-1 s-1
(V)
373K
NH Ph2NH
(k
)
) 30 × 4.3 × 104 ) 1.3 × 106 M-1 s-1
Unfortunately, there would appear to be no data in the
(VII)
373K
CH
literature with which the present value of k
/CH2
might be compared.17
Experiments at 80 and 60 °C with four different concen-
trations of diphenylamine in each case yielded kNH/kr )
38 M-1 and 43 M-1, respectively. Hence, via eq II,
(16) Densities for dodecane were obtained from Table 20d in:
Selected Values of Physical and Thermodynamic Properties of Hydro-
carbons and Related Compounds; Rossini, F. D., Ed.; Carnegie Press:
Pittsburgh, PA, 1953. These densities are 0.6900, 0.7048, and 0.7198
g/mL at 100, 80, and 60 °C, respectively, corresponding to dodecane
molarities of 4.05, 4.14, and 4.23 M at these three temperatures. We
thank a reviewer for steering us to this data source.
(17) There are no kinetic data for any similar reaction in solution
listed by: Lorand, J . P. In Landolt Bo¨rnstein, New Series; Radical
Reaction Rates in Solution; Fischer, H., Ed.; Springer-Verlag: Berlin,
1984; Vol. 13a, Chapter 2.
353K
NH
333K
NH
(k
)
) 7.2 × 105 M-1 s-1 and (k
)
) 3.2 ×
Ph2NH
Ph2NH
105 M-1 s-1. Once again, these data are not expected to
yield reliable Arrhenius parameters. The “best” line
gives,
(18) Benson, S. W. Thermochemical Kinetics, 2nd ed.; Wiley-Inter-
science : New York, 1976.

Reactions of Primary Alkyl Radicals with Aromatic Amines
log (kNH ) 11.2-8.6/θ (VIII)
J . Org. Chem., Vol. 61, No. 11, 1996 3781
of the rate constants for hydrogen abstraction from 3- or
4-substituted and 4,4′-disubstituted diphenylamines by
peroxyl radicals at 65 °C were very much better cor-
related by Σσ+ (F+ ) -0.89) than by Σσ.5b The correlation
of the rates of hydrogen abstraction by electrophilic
peroxyl radicals with Σσ+ is readily understood in terms
of polar contributions to the transition state, i.e.
)
Ph2NH
which has a preexponential factor well outside the
“normal” range.18 If the preexponential factor is assigned
the (probably) more reasonable value of 108.5 M-1 s-1 the
activation energy would be ca. 4.1 kcal/mol.
Despite our uncertainties regarding the true temper-
ature dependence of reaction 9, there can be no doubt
that this reaction has a substantial activation energy.
Furthermore, hydrogen abstraction from diphenylamine
• •
+•
•
[Ar N-H OOR T Ar N-H••Oh OR T
2
2
•
Ar N••H-OOR]q
by a primary alkyl radical is certainly fast at tempera-
2
333K
NH
tures ranging from 60 °C (k
≈ 3 × 105 M-1 s-1) to
Such transition state stabilization is not available for
abstractions by primary alkyl radicals which are neither
strongly electrophilic nor strongly nucleophilic.23
373K
100 °C (k
≈ 1.3 × 106 M-1 s-1). For comparison, rate
NH
constants for hydrogen abstraction from diphenylamine
by peroxyl radicals, reaction 4, have been reported as
(k3438K
)
) 4 × 104 M-1 s-1 5 and (k4333K
)
4.4 × 104
Ph2NH
Ph2NH
Con clu sion
M-1 s-1 6
. Thus, these peroxyl radical reactions are slower
Hydrogen atom abstraction from diarylamine antioxi-
dants by carbon-centered radicals is a reaction which
cannot be ignored in kinetic modeling studies of inhibited
hydrocarbon autoxidations at elevated temperatures,
particularly at oxygen pressures less than those existing
at sea level.24
than hydrogen abstraction by primary alkyl radicals.
There is also excellent experimental evidence indicating
that the preexponential factors and activation energies
for reaction 4 are unusually low.19 Thus, hydrogen
abstraction from aromatic amine antioxidants by alkyl
radicals is certainly competitive with, and may frequently
dominate, abstraction by peroxyl radicals at the elevated
temperatures encountered in many practical situations,
particularly in the case of gas turbine lubricant oils in
jet aircraft operating at high altitudes (and hence at low
oxygen pressures, see reaction 2). The fact that reaction
9 is faster than reaction 4 can be attributed to the greater
exothermicity of the former reaction (ca. ∼12 kcal/mol
Exp er im en ta l Section
Ma ter ia ls. The 4,4′-disubstituted diphenylamines were
synthesized by literature procedures from readily available
commercial materials and had physical properties which
agreed with literature reports.
4,4′-Dibr om od ip h en yla m in e.27 Diphenylamine + benzoyl
chloride f N-benzoyldiphenylamine + Br2 f dibromodiphen-
ylamine (mp 104-6 °C (lit.27 mp 105.5-7 °C); 1H NMR (60
MHz, CDCl3) δ 5.6 (1H, s), 6.8 (4H, d, J ) 10 Hz), 7.3 (4H, d,
J ) 10 Hz)).
for a primary alkyl radical).
373K
NH
The deuterium kinetic isotope effect (k
)
/
Ph2NH
373K
(k
)
) 2.3 (vide supra) which may be compared
Ph2NH
ND
with (k3438K
)
NH/(k338K
)
) 3.0 for hydrogen abstrac-
Ph2
Ph2ND
4,4′-Dim eth yld ip h en yla m in e.27 4-Methylaniline + acetic
anhydride f N-acetyl-4-methylaniline + 4-bromotoluene f
dimethyldiphenylamine (mp 75-6 °C (lit.27 mp 82-83 °C); 1H
NMR (200 MHz, CDCl3) δ 6.2 (1H, s), 3.0 (6H, s), 7.7 (8H, m)).
4,4′-Dim eth oxyd ip h en yla m in e.27 4-Methoxyaniline +
acetic anhydride f N-acetyl-4-methoxyaniline + 4-bromo-
anisole f dimethoxydiphenylamine (mp 99-100 °C (lit.27 mp
100-102 °C); 1H NMR (60 MHz, CDCl3) δ 3.9 (7H, s), 7.0 (8H,
m)).
4
tion by peroxyl radicals.5a Analyses of the deuterium
content of tert- and isobutylbenzene gave results consis-
tent with Scheme 2. That is, all the isobutylbenzene
contained one atom of deuterium (reaction 12) in each
experiment but not all of the tert-butylbenzene contained
one deuterium (see footnote d of Table 1).21
The data for the 4,4′-disubstituted diphenylamines (see
Table 1) reveal a significant polar effect on reaction 9.
N-Deu ter iod ip h en yla m in e was prepared from triply re-
crystallized diphenylamine (0.21 g, 1.24 mmol) in n-dodecane
(10 mL) to which was added 99.8% D2O (5 mL), and the liquids
were stirred for 1 h. For the kinetic runs, aliquots of the
dodecane solution were removed together with some D2O to
ensure that dedeuteration by adventitious H2O did not occur.
3-Meth yl-3-p h en ylbu ta n oyl p er oxid e was synthesized
from neophyl chloride via 3-methyl-3-phenylbutanoic acid.
Neophyl chloride (21 g, 0.125 mol) and magnesium turnings
(3.2 g, 0.13 mol) in dry ether (100 mL) were refluxed for 10 h
under a dry N2 atmosphere, following which dry CO2 was
bubbled through the mixture for 5 h. Acidification (dilute
HCl), extraction with ether (3 × 100 mL), and removal of most
of the ether was followed by addition of 5 M NaOH (200 mL).
Two electron-withdrawing p-bromine atoms (σ/Br )
373K
NH
+0.26)22 reduce (k
) by a factor of 7.7 relative to the
unsubstituted amine, and two electron-releasing methyl
(σ/Me ) -0.14),22 methoxyl (σ/MeO ) -0.12),22 and
373K
NH
tert,tert-octyl (σ/Me3C ) -0.15)22 groups increase k
by slightly less than a factor of 2. A Hammett plot of
373K
log (k
/M-1 s-1) against Σσ yields a reasonable cor-
NH
relation with F ≈ -1.4. For comparison, the logarithm
(19) For example:20 amine, log (A/M-1 s-1), E (kcal/mol): R-naph-
thylamine, 3.9, 1.0; phenyl-R-naphthylamine, 5.1, 1.0; â-naphthyl-
amine, 5.0, 2.3; aniline, 6.6, 5.0.
(20) Howard, J . A.; Furimsky, E. Can. J . Chem. 1973, 51, 3738-
3745. Chenier, J . H. B.; Furimsky, E.; Howard, J . A. Can. J . Chem.
1974, 52, 3682-3688.
(23) Certainly this appears to be true for methyl radicals. See:
Wong, M. W.; Pross, A.; Radom, L. J . Am. Chem. Soc. 1993, 115,
11050-11051.
(21) At first sight it appeared that it would be possible to calculate
373K
CH
k3N7D3K/k
from these data since a simple kinetic analysis which
(24) Hydrogen atom abstraction from the very good phenolic anti-
oxidant R-tocopherol (vitamin E) by a primary alkyl radical is also a
rapid reaction, k343K ) 1.7 × 106 M-1 s-1 in benzene.25 As would be
expected, less effective phenolic antioxidants would appear to be less
reactive to carbon-centered radicals, e.g.,26 k298K(Me3C• + PhOH) ) 14
373K 373K
ignores reaction 13 yields:
k
/kCH ) ([PhCMe2CH2D]/[PhCMe3])
ND
([C12H26]/[(C6H5)2ND]). However, with [C12H26] ) 4.05 M and kC37H3K
)
3.5 × 103 M-1 s-1, the values calculated for kN37D3K range from 1.9 × 105
M-1 s-1 to 3.9 × 105 M-1 s-1. Such values are considerably smaller
M-1 s-1, k298K(Me3C• + 2,6-(Me3C)2C6H3OH) ) 93 M-1 s-1
.
373K
ND
than the value which can be calculated for (k
)
from the data
(25) Evans, C.; Scaiano, J . C.; Ingold, K. U. J . Am. Chem. Soc. 1992,
Ph2ND
in Table 1, viz. 5.6 × 105 M-1 s-1. We attribute the lower than
114, 4589-4593.
“expected” [PhCMe2CH2D]/[PhCMe3] ratios to reaction 13, X• ) PhCMe2-
(26) Ruegge, D.; Fischer, H. Int. J . Chem. Kinet. 1989, 21, 703-
•
CH2
.
714.
(22) Exner, O. In Conformational Analysis in Chemistry; Chapman,
N. B., Shorter, J ., Eds.; Plenum: New York, 1978; Chapter 10.
(27) Chen, M. M.; D’Adamo, F. A.; Walter, R. I. J . Org. Chem. 1961,
26, 2721-2727.

3782 J . Org. Chem., Vol. 61, No. 11, 1996
Burton et al.
1
The resulting aqueous solution was washed with ether (3 ×
70 mL) and then acidified, and the 3-methyl-3-phenylbutanoic
acid was extracted with ether (yield 11.4 g, 50%; mp 55 °C
lit.28 mp 56-7 °C); 13C NMR (50 MHz, CDCl3) δ 178.5, 149.0,
128.7, 126.5, 125.9, 48.5, 36.0, 28.9; 1H NMR (200 MHz, CDCl3)
δ 7.2-7.5 (5H, m), 2.65 (2H, s), 1.5 (6H, s)). This acid (1.78 g,
10 mmol) was dissolved in dry ether (20 mL), and oxalyl
chloride (3 g, 24 mmol) was added slowly (anhydrous condi-
tions in an ice bath). After the solution was refluxed for 2 h,
the solvent and excess oxalyl chloride were removed under
vacuum. The resulting clear liquid plus 20 mL of ether was
cooled in an ice bath, and hydrogen peroxide (0.62 g of 27.5%
aqueous solution, 5 mmol H2O2) and pyridine (1 g, 13 mmol)
were added with stirring for 2 h at temperatures below 2 °C.
The product was washed first with dilute HCl and then with
saturated Na2CO3 followed by drying over anhydrous Na2SO4.
No attempt was made to concentrate all the product to record
a yield for fear of its decomposition. However, a portion was
evaporated to obtain NMR spectra on the crude product: 13C
NMR (50 MHz, CDCl3) δ 167.2, 149.0, 128.7, 126.5, 125,9, 44.4,
36.0, 28.9; H NMR (200 Mz, CDCl3) δ 7.2-7.5 (5H, m), 2.7
(2H, s), 1.5 (6H, s).
Exper im en tal P r ocedu r e. n-Dodecane was passed through
alumina prior to use. Standard solutions of the amine and
the peroxide were made up in this solvent. Aliquots of the
amine and peroxide solutions were further diluted with
dodecane in glass vials which were degassed by freeze, pump,
thaw cycles and were then sealed under a few Torr of N2.
Reactions were carried out five at a time in a Pye-Unicam GC
oven with the samples generally being left in the oven for
sufficient time (up to a few days) to ensure complete reaction.
The relative yields of the three hydrocarbon products were
determined (using authentic standards) by GC (5 m, OV 101
at 10% loading, N2 carrier gas, 95 °C, FID). Retention times
were ca. 1 h, and the column required N2 purging for 1 h at
250 °C after each analysis.
Ack n ow led gm en t. A.B. thanks Exxon Chemical for
a studentship. J .C.W. and K.U.I. thank N. A. T. O. for
a travel grant.
(28) Bordwell, F. G.; Almy, J . J . Org. Chem. 1973, 38, 575-579.
J O952162X