Kinetics and mechanism of the reaction of benzyl bromide with copper in hexamethylphosphoramide

Article abstract of DOI:10.1002/kin.20065

The reaction of copper with benzyl bromides in hexamethylphosphoramide has been studied. The kinetic and thermodynamic parameters of the reaction have been obtained. Hammett plots of log (script k sign/script k sign°) vs the substituent constant σ gave good correlations (ρ = 0.15, S _ρ = 0.02, r = 0.954). The structure of the organic group has little effect on the rate of reaction of benzyl bromide with copper. In the absence of atmospheric oxygen, the oxidative dissolution of copper occurred by the mechanism of single-electron transfer with the formation of 1,2-diphenylethane and copper(I) complexes. The stereochemistry and intermediates compound was also investigated. The reaction mechanism is discussed.

Full text of DOI:10.1002/kin.20065

Kinetics and Mechanism
of the Reaction of Benzyl
Bromide with Copper in
Hexamethylphosphoramide
1
2
2
ANATOLY M. EGOROV, SVETLANA A. MATYUKHOVA, ALEXANDER V. ANISIMOV
1
The Department of Natural Sciences, Tula State University, 92 Prospekt Lenina, 300600 Tula, Russian Federation
Department of Chemistry, Moscow State University, 1 Leninskie Gory, 119899 Moscow, Russian Federation
2
Received 29 June 2004; accepted 12 October 2004
DOI 10.1002/kin.20065
Published online in Wiley InterScience (www.interscience.wiley.com).
ABSTRACT: The reaction of copper with benzyl bromides in hexamethylphosphoramide has
been studied. The kinetic and thermodynamic parameters of the reaction have been obtained.
o
Hammett plots of log (k/k ) vs the substituent constant σ gave good correlations (ρ = 0.15,
Sρ = 0.02, r = 0.954). The structure of the organic group has little effect on the rate of reaction
of benzyl bromide with copper. In the absence of atmospheric oxygen, the oxidative dissolution
of copper occurred by the mechanism of single-electron transfer with the formation of 1,2-
diphenylethane and copper(I) complexes. The stereochemistry and intermediates compound
was also investigated.
The reaction mechanism is discussed. ꢀC 2005 Wiley Periodicals, Inc. Int J Chem Kinet 37:
96–305, 2005
2
INTRODUCTION
For the development of optimum conditions for the
target-oriented synthesis of the complex compounds
of copper halides with organic ligands, the kinetics
and mechanism of the reaction of copper with benzyl
bromide in the presence of hexamethylphosphoramide
(HMPA) were studied in details. This work provides
sufficient evidence to exclude many of the transition
states conceivable for the rate-limiting step for reac-
tion of benzyl bromide with copper.
The coordination compounds of copper are in consid-
erable industrial use as selective catalysts in various
chemical processes. The direct oxidative dissolution of
copper metal in systems containing carbon tetrachlo-
ride and coordinating solvent, which occurs under mild
conditions, is one of the methods for the synthesis of
these compounds [1].
Being a destroyer of ozone, carbon tetrachloride is
prohibited for use in chemical industry, and this is why
its substitutes are being sought.
EXPERIMENTAL
Correspondence to: Svetlana A. Matyukhova; e-mail: silver sun
inbox.ru.
Equipment and Analytical Measurements
@
Contract grant sponsor: Russian Federation’s Ministry of Educa-
1H NMR spectra were recorded on a Jeol LTD FX-90 Q
spectrometerusing25–30%solutionsinCDCl3. Chem-
ical shifts are given in ppm relative to tetramethylsilane
tion under research program “The Higher School Fundamental Study
of Natural Sciences and Humanities, The Universities of Russia.”
c 2005 Wiley Periodicals, Inc.
ꢀ

REACTION OF BENZYL BROMIDE WITH COPPER IN HEXAMETHYLPHOSPHORAMIDE
297
as internal standard. The accuracy of chemical shifts
was ± 0.01 ppm.
for 24 h. Just before the experiment, the sample was
mechanically purified to remove the swollen insulating
film, exposed to concentrated nitric acid for 5–10 s, and
washed with water, acetone, and then HMPA.
All organic compounds were obtained from com-
mercial sources.
The purity of commercial samples (Aldrich Chem-
icals Co. Ltd.) of benzyl bromide, 3-methylbenzyl
bromide, 4-methylbenzyl bromide, 3-methoxybenzyl
bromide, 4-fluorobenzyl bromide, 3-fluorobenzyl bro-
mide, and 3-chlorobenzyl bromide was checked by
GLC. Compounds that contained toluene or were less
than 99% pure were purified by low-temperature frac-
tional recrystallization or by fractional distillation in
the case of liquids or by recrystallization from ethanol
in the case of solids.
The ESR spectra were recorded at 77 K on a
Radiopan radiospectrometer in films of copper co-
condensates with benzyl bromide and HMPA (1:50:50)
according to the literature [2,3] at 9 GHz frequency in
the absence of saturation and amplitude broadening.
The copper was evaporated from the corundum cru-
cible at 1400–1440 K during continuous evacuation
−
4
(
10 mmHg).
The low-frequency IR spectra were measured on
a Perkin-Elmer 325 spectrophotometer; the samples
were prepared as suspensions in mineral oil.
Thepurityofstartingsubstanceswasmonitored, and
the quantitative analysis of organic reaction products
was performed using gas chromatography (GC). The
conditions of GC analysis were described previously
4-Methoxybenzyl bromide and 4-chlorobenzyl bro-
mide were prepared by known methods. The purities of
the products were checked by elemental analysis and
comparison of physical properties with those reported
in the literature [5].
[4].
The reaction products were isolated by prepara-
tive liquid chromatography on a Tsvet-304 chromato-
graph equipped with a UV detector (λ = 254 nm) using
steel column (l = 250 mm, d = 4 mm); Silasorb 600
Synthesis of (+)-R-1-bromo-phenylethane was car-
ried out by interaction of (−)-S-1-phenylethanol with
POBr3 in the presence of pyridine in pentane [6]. Yield
(Chemapol, Czech Republic, particle sizes 15–25 ꢀm)
was used as the sorbent; a 5:1 hexane–ether mixture
was used as the elutriator.
◦
◦
25
◦
◦
74%. B.p. = 86–87 /11 mmHg, [α]D +73.3 (l = 1)
25
Organic reaction products were analyzed on a
Hewlett-Packard GC-MS instrument (HP 5972 mass-
selective detector, HP 5890 chromatograph) using
a capillary column (l = 30 m, d = 0.25 mm) with
a diphenyl (5%) stationary phase supported on
polydimethylsiloxane. The column temperature was
(lit. [6] b.p. = 86–87 /11 mmHg, [α]D +96.3 (l = 1,
1
100%)). H NMR (CDCl3): 1.68 (d, 3H, CH3), 4.86
(dd, 1H,CH), 7.14 (m, 5H, Ph).
Synthesis of (−)-S-1-phenylethanol was carried out
◦
in accordance with [7,8]. Yield 55%. B.p. = 94–95 /
2
5
◦
14 mmHg, [α]D −37.65 (l = 1) (lit. [7,8] b.p. =
◦
◦
◦
25
◦
1
4
0–250 C; the heating rate was 30 /min. Helium was a
94–95 /14 mmHg, [α]D −44.2 (l = 1)). H NMR
(CDCl3): 1.21 (d, 3H, CH3), 3.78 (q, 1H,CH), 7.06
(m, 5H, Ph).
−
1
carrier gas (the carrier-gas flow rate was 1 mL min ).
The injector temperature was 250 C, and the detector
temperature was 280 C.
◦
◦
Synthesis of gaseous DCl was carried out by inter-
action of sulfuric acid-d2 (Aldrich Chemicals Co. Ltd.,
98%) with NaCl in the presence of DCl (37% solution
in D2O, Aldrich Chemicals Co. Ltd.).
3-Methylpyridine (Janssen) was dried with fused
KOH and twice distilled over KOH in an inert gas at-
Specific polarized light plane rotation was measured
◦
on an A-1 EPO automatic polarimeter (δ = 0.01 ).
Inorganic reaction products (copper cations) were
determined by ion chromatography on a Tsvet-3006
chromatograph with the use of Diacat-3 columns
◦
20
(
Elsiko, Moscow; l = 150 mm, d = 3 mm). An aque-
mosphere. b.p. = 143.7–144 C; nD = 1.5067 (lit [9]
◦
20
ous4mMethylenediamine, 5mMcitricacid, and5mM
tartaric acid solution was used as an elutriator. The rate
b.p. = 144.0 C; nD = 1.5068).
All solvents were purified according to standard
procedures [10]. They were freed from dissolved
gases by repeatedly freezing and thawing at a reduced
pressure and stored in ampoules in the absence of
air.
−1
of elution was 15 mL min . The sample volume was
00 ꢀL (after 1:1000 dilution with water).
1
Reagents
Copper powder of >99.99% purity was prepared by the
reduction of CuSO4 (analytical grade) with magnesium
powder (MP-1, Khlorvinil, Ukraine) in an atmosphere
of argon free of oxygen.
A copper wire (GOST 7262-54 Russia, Cu content
of 99.99%) 0.05 mm in diameter was held in HMPA
Reaction of Benzyl Bromide with Copper,
General Procedure
A 1-g portion of Cu powder was dissolved in 10 mL of
HMPA and 5 mL of benzyl bromide in an argon atmo-
◦
sphere at 50 C. After 5 h, the formed white precipitate

298
EGOROV, MATYUKHOVA, AND ANISIMOV
(
which rapidly became green in air) was filtered off
Studies of the Reaction of Active Copper
with Benzyl Bromide and HMPA
in an atmosphere of argon freed from oxygen. The re-
crystallization from 3-MePy afforded 5.26 g (80%) of
yellow crystals, which were stable in air [11]. Analysis:
C18H21N3CuBr (422.01) calculated: Cu, 15.40; Br,
1
1
Studies of the low-temperature reaction of active cop-
per with benzyl bromide were carried out in vacuum
apparatus analogous to that reported previously [16].
8.90 C, 51.10; H, 5.00; N, 9.53; found Cu, 15.35; Br,
8.87 C, 51.00; H, 5.01; N, 9.77%. IR (mineral oil),
−4
The reagents were evaporated in an evacuated (10
mmHg) reactor (V = 10 L) and condensed on the sur-
face cooled with liquid N2. Active copper was prepared
by evaporating the purified metal from a corundum cru-
cible at 1400–1440 K [17]. The rate of evaporation of
ν = 480 (m, Cu N), 418 (w, Cu N), 348 (s, Cu N),
−1
2
30 (vs, Cu Br), 225 (s, Cu N) cm
Organic products of reaction were extracted with
.
diethyl ether (20 mL). The residues of benzyl bromide
and HMPA, as well as 1,2-diphenylethane and 4,4 -
ditolyl, were detected in the ether solution. The isolated
−
1
ꢁ
Cu was 0.2 mmol min . Benzyl bromide and HMPA
was evaporated at 273–308 K (HMPA:PhCH2Br:Cu =
0–10:10:1). The duration of co-condensation was
yield of 1,2-diphenylethane was 1.37 g (96%). m.p. =
◦
◦
1
2–4 h.
After completion of co-condensation evacuation
5
1–52 C (lit [12] m.p. = 51–52 C). H NMR (CDCl3):
δ = 2.82 (s, 4H, CH2 ), 7.02 (m, 10H, Ph) ppm.
MS (EI, 70 eV): calculated m/z = 182.11 (M), found
was terminated and the reactor was filled with dry pure
Ar (760 mmHg). The samples were heated to 298 K.
At this temperature the film melted and decolorized.
The reaction mixtures were treated with 1% solution
of DCl in D2O (99.5 atom.% D, Aldrich) at 298 K un-
der an atmosphere of pure dry Ar. The liquid phase was
analyzed by GLC. The reaction products were isolated
by preparative liquid chromatography. The results are
given in Table VI.
+
+
m/z = 182 [M] (23), 91 [M/2] (100).
ꢁ
4
,4 -Ditolyl MS (EI, 70 eV): calculated m/z =
+
1
[
82.11 (M), found m/z = 182 [M] (100), 167
+ +
M CH3] (56), 152 [M 2CH3] (15).
Study of the Reaction Kinetics
The reactions of copper with nine benzyl bromides in
HMPA were studied by the resistometric method [13]
according to a published procedure [14] in an atmo-
sphere of water-free argon freed of oxygen.
Studies of the Stereochemistry of
the Reaction under Investigation
Benzene was used as an neutral solvent for the deter-
mination of the kinetic characteristics of the reaction of
copper with benzyl bromide in the presence of HMPA
The reaction of (+)-R-1-bromo-1-phenylethane and
HMPA with copper was studied analogously to the
above procedure. Condensation of DCl at 77 K on the
surface of the samples was carried out according to the
below procedure.
[15].
The reaction was studied in the kinetic mode, as
evidenced by the independence of the rate of copper
dissolution in the test media from the rate of stirring.
Tables I–III summarize the results of this study.
After completion of co-condensation, the samples
were heated to 160 K, kept at this temperature for
20 min, and then cooled to 77 K. DCl (50 g) was
Table I Kinetic and Thermodynamic Parameters of the Oxidative Dissolution of Copper in the Benzyl Bromide–HMPA
System Calculated Using the Langmuir–Hinshelwood Mechanism from the Adsorption of the Reagent and the Solvent
at Different Active Centers of the Metal
5
−2
−1
−1
−1
K2 (l mol )
Parameter
k × 10 (g cm min
)
K1 (l mol
)
T (K)
313
1.3 ± 0.1
3.0 ± 0.1
7.4 ± 0.1
10.1 ± 0.2
17.7 ± 0.2
34.7 ± 0.3
0.998
0.936 ± 0.007
0.664 ± 0.006
0.510 ± 0.006
0.428 ± 0.005
0.370 ± 0.005
0.295 ± 0.005
0.999
0.465 ± 0.006
0.379 ± 0.005
0.293 ± 0.005
0.267 ± 0.005
0.247 ± 0.005
0.203 ± 0.005
0.998
3
3
3
3
3
23
33
38
43
53
ra
EA (kJ mol 1)
−
76.3 ± 4.7
o
−1
)
ꢀ
H
H
(kJ mol
(kJ mol
−26.7 ± 1.5
RBr
o
−1
)
ꢀ
−19.2 ± 1.3
HMPA
a
Sample correlation coefficient.

REACTION OF BENZYL BROMIDE WITH COPPER IN HEXAMETHYLPHOSPHORAMIDE
299
Table II Relative Reactivities of Substituted Benzyl
Bromides toward Copper in HMPA at 80 C
was removed at 190–200 K (100 mmHg). The reaction
mixtures were heated to 298 K, and the composition of
reaction products were isolated by preparative liquid
◦
4
k × 10
−
2
−1
k/ko
chromatography. Yields are given in Table VII.
Substituent
(g cm min
)
σ
◦
(
+)-S-1-Phenylethane-1D (1). B.p. 135–136 C,
m-Cl
m-F
p-Cl
m-OMe
p-F
H
m-Me
p-Me
p-OMe
3.92 ± 0.05
3.89 ± 0.04
3.97 ± 0.05
3.55 ± 0.04
3.65 ± 0.04
3.47 ± 0.03
3.42 ± 0.03
3.34 ± 0.03
3.13 ± 0.03
0.37
0.34
0.23
0.12
0.06
1.129
1.121
1.145
1.023
1.053
1
0.987
0.968
0.902
20
20
1
nD = 1.4954, [ꢁ]D +0.05 (l = 0.1). H NMR
(
7
CDCl3): δ = 1.23 (d, 3H, CH3), 2.62 (q, 1H, CH),
◦
.20 (m, 5H, Ph) ppm. Lit. data: b.p. 135–136 C,
2
0
20
nD = 1.4919 [18], [ꢁ]D +0.81 (l = 0.1) [19].
Mixture of RS-1-phenylethane-1D and phenyl-
0
1
ethane (1 and 4), 1:1. H NMR (CDCl3): δ = 1.20 (d,
−0.07
−0.17
−0.27
3H, CH3)50%, 1.24(t, 3H, CH3)50%, 2.62(q, 1H, CH)
50%, 2.66 (m, 2H, CH ) 50%, 7.19 (m, 5H, Ph) ppm.
2
RR,SS-2,3-Diphenylbuthane (2). B.p. 130–
◦
20
1
1
32 C/7 mmHg, nD = 1.5557. H NMR (CDCl3):
condensed to their surfaces. Then evacuation was
terminated, the reactor was filled with dry pure Ar
δ = 1.17 (d, 6H, 2CH3), 2.76 (m, 2H, CH CH),
◦
7.24 (m, 10H, 2Ph.) ppm. Lit. data: b.p. 130–132 C/7
2
0
(760 mmHg). The samples were heated to 160 K (at
mmHg [18], nD = 1.5557 [19].
RS,RS-2,3-Diphenylbuthane (3). B.p. 144–149 C/
12 mmHg, m.p. 126–127 C. H NMR (CDCl3):
◦
this temperature the film melted and decolorized) and
kept at this temperature for 20 min. The excess of DCl
◦
1
Table III Hammett ρ Values for Reactions Involving Cleavage of Bonds to, or Electron Transfer with, an Benzyl Moiety
Reaction
ρ
Solvent
References
0.15
Benzene + HMPA
This work
[22]
Hal = CB lr
0.40
0.22
1.05
Benzene
Benzene
Benzene
I
0
.312
Benzene
[23]
[24]
0.4
Benzene
0
0
.47
.5
[25]
[26]
0.8
[27]
1.0
[27]
[28]
0
0
.7
[29]
[30]
[31]
.66
Hal = Cl
1.45
DMF
Br
2.8
CH3CN
[32]

300
EGOROV, MATYUKHOVA, AND ANISIMOV
δ = 1.03 (d, 6H, 2CH3), 2.75 (m, 2H, CH CH), 7.24
◦
(
m, 10H, 2Ph.) ppm. Lit. data: b.p. 144–149 C/12
◦
mmHg [18], m.p. 126–127 C [19].
Identification of Radical
Species in Solution
The reaction in the presence of a radical trap was stud-
ied analogously to the general procedures. Dicyclo-
hexyldeuterophosphine (DCPD) was used as the radi-
cal trap [20].
The residues benzyl bromide and HMPA, as well
as 1,2-diphenylethane and ꢁ-deuterotoluene, were de-
tected in the diethyl ether solutions. Table V summa-
Figure 1 The dependence of the oxidative dissolution of
copper in the benzyl bromide-HMPA system on the initial
concentrations of mixture components in the presence of an
indifferent solvent (benzene) at 353 K: 1—CRBr = 0.5 mol
rizes the yields of organic reaction products.
1
ꢁ
-Deuterotoluene H NMR (CDCl3): δ = 2.32 (m,
2
H, CH2 ), 7.15 (m, 5H, Ph) ppm. MS (EI, 70
eV): calculated m/z = 93.07 (M), found m/z = 93
−
1
−1
−1
.
l
, 2—CHMPA = 0.5 mol l , and 3—CHMPA = 2 mol l
+ + +
M] (100), 92 [M H] (93), 91 [M D] (46), 66
M C2H3] (9), 65 [M C2H2D] (11).
[
[
+
+
adsorption of benzyl bromide and HMPA, respectively,
takes place.
In this case, surface coverages derived from the
Langmuir isotherms for the adsorption of individual
components appear in the rate equation, and an expres-
sion for w has the form
RESULTS AND DISCUSSION
The reaction kinetics of the oxidative dissolution of Cu
in the benzyl bromide-HMPA system has been studied
using the resistometric method. This method provides
an opportunity to study various kinetic features of fast
heterogeneous reactions; it is characterized by high ac-
curacy and reproducibility of results [13,21].
k · K
1
· K
2
[PhCH
2
Br][HMPA]
2
[PhCH Br][HMPA]
w = 1
+ K
1
[PhCH
2
Br] + K
2
[HMPA] + K
1
K
2
To determine the kinetic characteristics of the pro-
cess, the reaction was performed in an indifferent sol-
(
4)
−1
vent of benzene (DNSbCl5 = 0.42 kJ mol ) [15].
The Fig. 1 demonstrates the kinetic curves. An in-
crease in the concentration of HMPA from 0.5 to 2
where k = K · N N ; N and N are the numbers of
1
2
1
2
active centers of the metal surface at which the adsorp-
tion of benzyl bromide and HMPA, respectively, takes
place.
−
1
mol L did not change the shape of curves for the de-
pendence of the reaction rate on the concentration of
benzyl bromide. These facts indicate that the process
occursbytheLangmuir–Hinshelwoodmechanismwith
the adsorption of the reagent and the solvent at different
active centers of the metal surface [14]:
The treatment of the experimental relations (Fig. 1)
using the set of Eqs. (1)–(3) allowed us to determine the
equilibrium constants of benzyl bromide and HMPA
adsorption on the surface of copper (K and K , respec-
1
2
tively) and the rate constant k of the chemical reaction.
The study of the reaction kinetics at different temper-
K1
atures allowed us to find E of the chemical reaction
A
PhCH2Br + S1 ꢀ (PhCH2Br)S1 (1)
ꢁ
and the enthalpies of adsorption of the reactants on the
surface of copper. Table I summarizes the results.
The Langmuir–Hinshelwood scheme for the test
process suggests that the interaction of adsorbed reac-
tant molecules with the metal surface, that is, a surface
chemical reaction, is a rate-limiting step of the reaction.
Relative reactivities of nine substituted benzyl bro-
mides were obtained from the kinetic experiments, as
shown in Table II.
K2
ꢁ
L + S2
ꢀ (L)S2
(2)
(3)
K
(
PhCH2Br)S1 + (L)S2 ꢀ Products
ꢁ
where L is HMPA; K1 and K2 are the equilibrium
constants of benzyl bromide and HMPA adsorption,
respectively; K is the rate constant of the chemical
process; S1 and S2 are the active centers at which the
Figure 2 shows a plot of the logarithm of the rela-
tive rates against the Hammett σ constants, and it will

REACTION OF BENZYL BROMIDE WITH COPPER IN HEXAMETHYLPHOSPHORAMIDE
301
then protonated.
−
¯
•
RHal + e → RHal
¯
•
•
−
RHal → R + RHal
•
−
−
R + e → R
The decomposition of the radical anion is very fast
orevenconcertedincombinationwithelectrontransfer.
The ρ values calculated from the squares slopes of
o
Figure 2 Hammett plots of log (k/k ) vs the substituent
constant σ obtained in this work.
−
the ꢀE1/2 vs. σ plots were ρ = 0.5 from the series of
benzyl chlorides and ρ = 2.8 from the series of benzyl
bromides. The correlation with σ substituent constants
gave no correlation (r = 0.77, Sρ = 0.28, r = 0.78,
Sρ = 0.14) [32].
be seen that there is a reasonably good linear relation-
ship (ρ = 0.15, r = 0.954, Sρ = 0.02). The ρ value ob-
tained here is, however, considerably lower than the
values associated with ionic reactions. This indicates
that the polar effect on the abstraction of bromine atom
is much smaller than in ionic reactions, although this
polar effect is the crucial factor in determining the
relative reactivities.
The Hammett parameters, however, are similar to
those reported by D.D. Tanner (ρ = 0.22, r = 0.95,
Sρ = 0.03) for the reaction of tri-n-butyl tin and benzyl
bromide in benzene [22] (see Fig. 3).
The abstraction of bromine atom from a benzyl bro-
mide by tri-butyltin radical is probably good model
for atom abstraction by a metal surface. For compari-
son abstraction of chlorine atom from substituted ben-
zyl chlorides by triethylgermyl radical has ρ = 0.312
We have considered nine limiting transition states
for the organic moiety summarized by structures 1–9
(Scheme 1) [33]. Here Cu represents a copper atom of
a copper metal surface.
1
: The unperturbed organic halide would represent
the transition state if the overall rate of reaction
was limited by the rate of encounter of this or-
ganic halide, either with the copper surface or
with some other reactive species in solution.
: The organic halide might adsorb at the metal sur-
face in the rate-limiting step.
2
3
: Electron transfer from the copper to the organic
halide, either close to or adsorbed at the copper
surface might be rate limiting.
4–6: The carbon halogen bond might be broken in
the transition state.
[23].
The positive ρ values also have been obtained from
the kinetic studies of a number of alkyl radicals, gener-
ated in the presence of a series of substituted toluenes
24–30]. These data are listed in Table III.
Polarographic reductions of organic halides are well
7
–8: A copper atom at the metal surface (or possi-
bly in solution) might displace halide from car-
bon in a SN2-like reaction or insert into the
carbon halogen bond in concerted process.
[
9
: The rate-limiting step might involve a species
with a covalent carbon copper bond, for exam-
ple, the rate of detachment of an organocopper
species from the copper surface might be rate
determining.
known to be irreversible [31,32] and the overall two-
electron transfer process leads to carbanion which is
Since ρ has a significant positive value, the transi-
tion state does not resemble the starting material, and 1
can be discarded. The magnitude of ρ seems too large
to be compatible with physical absorption, 2. The sign
of the ρ is incompatible with an intermediate benzyl
cation, 5. Since the largest values ρ observed are simi-
lar to those for the tin hydride reduction, and consider-
ably smaller than that estimated for reactions generat-
ing benzyl anions (ρ = 2.8). The magnitude of ρ seems
too small for a fully formed benzyl anion, 4. A rate-
determining step involving an intermediate in which
the carbon copper bond is partially or fully formed
o
Figure 3 Hammett plots of log (k/k ) vs the substituent
constant σ obtained for reaction of tributyltin radical with
benzyl bromides [22].

302
EGOROV, MATYUKHOVA, AND ANISIMOV
Scheme 1
Scheme 2
(
9) is unlikely, both because this transition state bears a
able worsening of the resolution of the ESR spec-
tra. Paramagnetic species that appeared in the benzyl
bromide–HMPA atomic copper systems do not differ
from those formed when the compact copper film is
formed.
DCPD provides an opportunity not only to detect
radical intermediates but also to determine their con-
centration in solution [20].
In the dehalogenation of benzyl bromide with cop-
per in HMPA in the presence of DCPD, the compo-
sition of reaction products changed depending on the
copper:DCPD molar ratio (Table V). At the copper :
DCPD molar ratio equal to 1:5, the formation of 1,2-
diphenylethane was not observed (Table V), but an
equivalent amount of ꢁ-deuterotoluene was detected
instead:
large negative charge on carbon and would be expected
to show a large value for ρ and because cleavage of
carbon bromine bond appears to be irreversible.
It is more difficult to differentiate between reactions
leading to the three remaining limiting structures—
electron transfer from copper to benzyl bromide gen-
erating the radical anion 3, bromine abstraction by
copper forming an benzyl radical 6, and insertion of
copper directly in to the carbon bromine bond of the
adsorbed benzyl bromide 7—on the basis of available
evidence.
It is well known that ESR study of organic reaction
is the most reliable evidence for the existence of para-
magnetic particles as intermediates and one of the most
convenient methods for their detection.
Our study showed that the ESR spectrum of the co-
condensate of copper with benzyl bromide and HMPA
DCPD
nPhCH2Br + nCu + mL −−− →[CunLmBrn] + PhCH2D
(1:50:50) at 77 K is a triplet of quartets with whole
width about 50 G and g-factor of 2.002. The parameters
of the ESR spectrum of benzyl radical are presented in
Table IV.
where L = HMPA.
The analysis of reaction products in the presence of
radical traps and in their absence indicates that the pro-
cess occurs by a radical mechanism via the formation
of the benzyl radical (Scheme 2).
A comparison shows that the ESR spectrum of the
co-condensates of copper with benzyl bromide and
HMPA coincides with that of benzyl radical.
When excess of benzyl bromide and HMPA (1:1)
is condensed on the compact copper film (the thick-
The results of this work indicate that the dehalo-
genation process occurs at the surface of copper by a
radical mechanism via the formation of benzyl radi-
cals, which undergo recombination and isomerization
in solution.
−4
ness of the film is about 10 mm), losses of UHF
power in the sample increase due to an increase in
the electric conductivity, which results in consider-
Table IV Parameter of Spectra ESR for Benzyl Radical in Solid Matrices (a, G)
H
CH2
H
O
H
H
g-factor
a
(G)
a
(G)
a
(G)
a (G)
p
References
m
2
2
.002 ± 0.001
16.4 ± 0.9
16.4 ± 0.5
16.5
16.5 ± 1
5.5 ± 0.6
5.5 ± 0.5
6
5.5 ± 0.5
–
5.5 ± 0.6
5.5 ± 0.5
6
5.5 ± 0.5
This work
[4]
.002 ± 0.001
–
–
–
–
–
[34]
[35]

REACTION OF BENZYL BROMIDE WITH COPPER IN HEXAMETHYLPHOSPHORAMIDE
303
Table V Product Composition in the Oxidative
Dissolution of Copper in Benzyl Bromide-HMPA
Systems in the Absence and in the Presence of
Radical Traps
Table VI The Composition of Reaction Products being
Produced by Acidolyses of Organocopper Compounds
Yields of
PhCH2CH2Ph
2
Cu:PhCH Br:HMPA T (K) (%)
Yields of
PhCH2Db
(%)
a
Yield (mol%)
a
DCPD /Cu
Ph CH2Db
Ph CH2 CH2 Phc
1:10:0
:10:0
1:10:5
1:10:5
1:10:10
1:10:10
160
298
160
298
160
298
98
99
93
97
91
96
2
1
7
3
9
4
Entry
(mol/mol)
1
1
2
3
4
0
1
2
5
0
69
81
99
100
31
19
1
a
a
ꢁ
Admixture of dicyclohexylphospine 1%.
Admixture of toluene 1%.
Admixture of 4,4 -ditolyl < 0.1%.
Admixture of 4,4 -ditolyl < 0.01%.
b
c
b
Admixture of toluene 2%.
ꢁ
were precipitated in the molecular beam mode on the
surface cooled with liquid N2 afforded white films. Af-
ter thawing (273 K) and treatment of the samples with
DCl, optically inactive 1-phenylethane-1D (1) and four
other unlabeled hydrocarbons 2, 3, 4, and 5 were de-
tected in the reaction mixtures (Scheme 3).
It is well known that recombination of optically ac-
tive 1-phenylethyl radicals proceeds within a solvent
cage 15 times more slowly than inversion and leads
to partial retention configuration of the recombination
products [37].
The formation of optical inactive 2, 3 with the
ratio of 2:3 equals 1.03 along with 4 and 5 both
shows that the reaction occurs according to the radi-
cal mechanism, and recombination and disproportion-
ation of 1-phenylethyl radical can proceed after they
enter solution.
Theformationofracemic1-phenylethane-1-Dinthe
course of reaction of (+)-R-1-bromo-1-phenylethane
with copper is reliable evidence for this suggestion
and may proceed by radical pair recombination at
a slower rate than that of 1-phenylethyl radical in-
version on the copper surface within the solvent
cage.
At temperatures above 273 K, benzyl cuprates are
known to thermally decompose to give their respec-
tive homocoupled products in good yield along the
elemental copper [36]. At lower temperatures, these
organocoppercompoundshavegreaterstabilitybutwill
react quickly with benzyl bromide present to produce
the corresponding coupled products [36].
The low-melting point of HCl made it possible to
detect the reaction products both before (159 K) and
after (298 K) defrosting of the sample and to observe
unstable at 298 K copper organic compounds whose
acidolysis proceeds via the equation:
PhCH2Cu + DCl → PhCH2D + CuCl + Cun−1
n
The composition of reaction products being pro-
duced by acidolysis of organocopper compounds ob-
tained by the co-condensation of copper with benzyl
bromide is listed in Table VI.
The result obtained leads us to conclude that coordi-
nating solvent stabilizes the formation of organocopper
compounds. This conclusion agrees well with data in
work [36].
We have investigated the reaction of optically active
(
+)-R-1-bromo-1-phenylethane with copper in HMPA
Condensation of DCl at 77 K on the surface
of the samples, which were prepared by heating
co-condensates of copper with (+)-R-1-bromo-1-
phenylethane and HMPA, afforded white films. After
completion of condensation of DCl, the samples were
heated to 160 K (at this temperature films melted and
were decolorized), kept at this temperature for 20 min,
and then heated to 298 K.
The composition of the reaction products being pro-
duced by acidolyses of samples obtained is listed in
Table VII.
The optical purity of 1-phenylethane-1D (6 ± 1%)
does not exceed that for 1-phenylethane recombination
within a solvent cage.
to seek experimental evidence of optical activity reten-
tion at the asymmetry center in the course of formation
of organocopper compounds.
The main cause of racemization of organocopper
compounds is stereochemical instability of C Cu bond
and can arise just after organocopper formation. The
configuration of organocopper reagents must be fixed
at the moment of its formation. It can be achieved by
carrying out this reaction in the presence of deuterium
chloride that reacts vary rapidly with organocopper
compounds.
The co-condensation of a copper vapor with (+)-
R-1-bromo-1-phenyletane and HMPA (1:10:10) that

304
EGOROV, MATYUKHOVA, AND ANISIMOV
Table VII Yield of Products (%) for Studying
Stereochemistry of Reaction under Investigation
The yield of organocopper compounds depends on
reaction conditions.
At temperatures above 273 K, benzyl cuprates are
known to thermally decompose to give their respective
homocoupled products [36]:
Yields (%)
1
2
3
4
5
Acidolysis was carried out at 160 K
43.3 41.7 6.0
Acidolysis was carried out at 298 K
48.4 46.6 1.5
˙
PhCH2Cu → PhCH2 + Cu
–
2
9.0
1.5
At lower temperatures, these organocopper compounds
have greater stability but will react quickly with benzyl
bromide [41]:
The results obtained suggest that the limiting
step in the course of the reaction (+)-R-1-bromo-1-
phenylethane with copper is the dissociative electron
transfer on a copper surface.
PhCH2Cu + PhCH2Br → PhCH2CH2Ph + CuBr
Recombination and isomerization of the benzyl
radicals proceed in solution according to the Scheme 5
The results of this work indicate that the dehalo-
genation process occurs at the surface of copper by a
radical mechanism [38–40] via the formation of benzyl
radicals, which undergo recombination and isomeriza-
tion in solution. The adsorption of the reagent and the
solvent occurs by the Langmuir–Hinshelwood mech-
anism at different active centers of the metal surface
in accordance with the Scheme 4. The limiting step
of the reaction of benzyl bromides with copper is the
dissociative electron transfer on a metal surface.
(
for X = H).
The complex compounds of copper(I) with organic
ligands in solutions form di- or tetramers [42]:
CuLBr + Cun−1Ln−1Brn−1 → CunLnBrn
CunLnBrn + (n − m)L → [CunLmBrn] ↓
where L = HMPA; 2 ≤ n ≤ 4; 2 ≤ m ≤ 4.
Scheme 3
Scheme 4
Scheme 5

REACTION OF BENZYL BROMIDE WITH COPPER IN HEXAMETHYLPHOSPHORAMIDE
305
BIBLIOGRAPHY
21. Nifontova, G. A.; Echmaev, S. B.; Sikorenko, Yu. B.;
Lavrent’ev, I. P. Zh Fiz Khim 1998, 72, 147–151.
22. Blackburn, E. V.; Tanner, D. D. J Am Chem Soc 1980,
102, 692–697.
23. Sakurai, H.; Mochida, K. J Organomet Chem 1972, 42,
339–343.
24. Zavitsas, A. A.; Hanna, G. M. J Org Chem 1975, 40,
3782–3783.
25. Henderson, R. W.; Ward, R. D., Jr. J Am Chem Soc 1974,
96, 7556–7557.
26. Pryor, W. A.; Davis, W. H., Jr. J Am Chem Soc 1974,
96, 7557–7558.
27. Davis, W. H., Jr.; Pryor, W. A. J Am Chem Soc 1977,
99, 6365.
28. Pryor, W. A.; Davis, W. H., Jr.; Stanley, J. P. J Am Chem
Soc 1973, 95, 4754.
29. Henderson, R. W. J Am Chem Soc 1975, 97, 213.
30. Tanner, D. D.; Henriquies, R.; Reed, D. W. Can J Chem
1979, 57, 2578–2584.
1
2
3
4
5
6
7
8
9
. Garnovskii, A. D.; Kharisov, B. I.; Gokhon-Zorrila, G.;
Garnovskii, D. A. Usp Khim 1995, 64, 215–236.
. Sergeev, G. B.; Smirnov, V. V.; Zagorsky, V. V.
J Organomet Chem 1980, 209, 9–20.
. Sergeev, G. B.; Smirnov, V. V.; Zagorsky, V. V. Vesthik
Moskovskogo Universiteta 1980, 21, 257–261.
. Egorov, A. M.; Anisimov, A. V. Appl Organomet Chem
1
995, 9, 285–289.
. Aizpurua, J. M.;Lecea, B.;Palomo, C. CanJChem1986,
4, 2342–2347.
. Dauben, H. J.; McCoy, L. L. J Am Chem Soc 1959, 81,
404–5409.
6
5
. Giondo, G. M.; Gregorio, F. D.; Palladino, N.; Marconi,
W. Tetrahedron Lett 1973, 34, 3195–3198.
. Egorov, A. M.; Anisimov, A. V. J Organomet Chem
1
994, 479, 197–198.
. Potekhin, A. A. (Ed.). Svoistva Organicheskikh
Soedinenii. Spravochnik (The Properties of Organic
Compounds: A Handbook); Khimiya: Leningrad, 1984.
31. Streitwieser, A., Jr.; Perrin, C. J Am Chem Soc 1964, 86,
4938–4942.
1
0. Gordon, A. J.; Ford, R. A. The Chemist’s Companion: A
HandbookofPracticalData, TechniquesandReferences.
Wiley: New York, 1972.
32. Tanner, D. D.; Plambeck, J. A.; Reed, D. W.; Mojelsky,
T. M. J Org Chem 1980, 45, 5177–5183.
33. Rogers, H. R.; Hill, C. L.; Fujiwara, Y.; Rogers, R. J.;
Mitchell, H. L.; Whitesides, G. M. J Am Chem Soc 1980,
102, 217–226.
1
1
1
1
1. Goher, M. A. S. Acta Chim Acad Sci Hung 1979, 99,
3
07–317.
2. Caubere, P.; Moreau, J. Tetrahedron 1969, 25, 2469–
486.
3. Zhukov, S. A.; Lavrent’ev, I. P.; Nifontova, G. A. React
Kinet Catal Lett 1977, 7, 405–412.
34. Buben, N. Y.; Tolkachev, V. A.; Chkheidze, I. I. Kinet
Katal 1963, 4, 683–687.
2
35. Tolkachev, V. A.; Chkheidze, I. I.; Buben, N. Y. Dokl
Akad Nauk SSSR 1962, 147, 683–687.
36. Ginah, F. O.; Donovan, T. A., Jr.; Suchan, S. D.;
Pfennig, D. R.; Ebert, G. W. J Org Chem 1990, 55, 584–
589.
4. Sergeeva, V. P.; Kondin, A. V.; Alyasov, V. N.; Shul’pin,
G. P.; Maslennikov, V. P. Zh Obshch Khim 1990, 60,
2
268–2272.
1
5. Kondin, A. V.; Tabachkova, T. V.; Alyasov, V. N.;
Maslennikov, V. P. Metallorgan Khim 1992, 5, 993–
37. Green, F. D.; Berwick, M. A.; Stowell, J. C. J Am Chem
Soc 1970, 92, 867.
1
000.
38. Curran, D. P. Synnthesis 1988, 489–513.
39. Matyjaszewski, K. Chem—Eur J 1999, 5, 3095–3102.
40. Qiu, J.; Matyjaszewski, K. Macromolecules 1997, 30,
5643–5648.
1
1
6. Timms, P. L. J Chem Educ 1972, 49, 782.
7. Moskovits, D.; Ozin, G. A. Cryochemistry; Wiley:
New York, 1976.
1
1
2
8. Cohen, H. L.; Wright, G. F. J Org Chem 1953, 18, 432.
9. Khalaf, A. A.; Roberts, R. M. J Org Chem 1966, 31, 926.
0. Ashby, E. C.; Oswald, J. J Org Chem 1988, 53, 6068–
41. Beletskaya, I. P.; Artamkina, G. A.; Reutov, O. A.
Usp Khim 1976, 45, 661–694.
42. Rostovshchikova, T. N. Doctoral (Chem.) Dissertation,
Mos. Gos. Univ., Moscow, 1998.
6
076.