The reactivity of the high-energy intermediates formed in the reactions of Group 13 metal atoms and aromatic alkenes

Article abstract of DOI:10.1139/v98-033

Group 13 metal atoms were reacted with aromatic alkenes in a specialized metal atom reactor known as a "rotating cryostat." The nature of the intermediates formed was deduced from a GC-MS study of their hydrolysis and deuterolysis products. The product studies suggest that 2-phenylaluminacyclopropane, cis- and trans-3,4-diphenylaluminacyclopentane, and cis- and trans- 2,4-diphenylaluminacyclopentane are formed when A1 atoms react with styrene, and 2-methyl-2-phenylaluminacyclopropane and 3,4-dimethyl-3,4-diphenylaluminacyclopentane are formed when A1 atoms react with a-methylstyrene. These findings are consistent with the radicals detected in the EPR spectroscopic studies of A1-alkene reaction mixtures prepared under similar conditions. Mechanisms for the formation of the organoaluminium intermediates are discussed. Analogous organogallium intermediates are formed when gallium atoms react with styrene. The reductive coupling of styrene did not occur when In and T1 atoms were used. Only trace quantities of phenylethane were detected in the hydrolyzed reaction mixture.

Full text of DOI:10.1139/v98-033

4
00
The reactivity of the high-energy intermediates
formed in the reactions of Group 13 metal
atoms and aromatic alkenes
Helen A. Joly, Maria Kepes, Natalie Roy, and Jason Prpic
Abstract: Group 13 metal atoms were reacted with aromatic alkenes in a specialized metal atom reactor known as a “rotating
cryostat.” The nature of the intermediates formed was deduced from a GC–MS study of their hydrolysis and deuterolysis
products. The product studies suggest that 2-phenylaluminacyclopropane, cis- and trans-3,4-diphenylaluminacyclopentane,
and cis- and trans- 2,4-diphenylaluminacyclopentane are formed when Al atoms react with styrene, and
2
-methyl-2-phenylaluminacyclopropane and 3,4-dimethyl-3,4-diphenylaluminacyclopentane are formed when Al atoms react
with α-methylstyrene. These findings are consistent with the radicals detected in the EPR spectroscopic studies of Al–alkene
reaction mixtures prepared under similar conditions. Mechanisms for the formation of the organoaluminium intermediates are
discussed. Analogous organogallium intermediates are formed when gallium atoms react with styrene. The reductive coupling
of styrene did not occur when In and Tl atoms were used. Only trace quantities of phenylethane were detected in the
hydrolyzed reaction mixture.
Key words: Group 13 metal atoms, aluminium atoms, organoaluminium intermediates, metal atom reactions.
Résumé : On a fait réagir des atomes de métaux du groupe 13 avec des alcènes aromatiques dans le réacteur à atomes de
métaux connu sous le nom de « cryostat rotatif ». On a déduit la nature des intermédiaires formés en se basant sur une étude
de CG–SM de leurs produits d’hydrolyse et de deutérolyse. Les études sur la nature des produits suggèrent que, lorsqu’on fait
réagir des atomes d’aluminium avec le styrène, il se forme du 2-phénylaluminacyclopropane, des cis- et
trans-3,4-diphénylaluminacyclopentanes et des cis- et trans-2,4-diphénylaluminacyclopentanes; lorsqu’on fait réagir des
atomes d’aluminium avec l’α-méthylstyrène, il se forme du 2-méthyl-2-phénylaluminacyclopropane et du
3
,4-diméthyl-3,4-diphénylaluminacyclopentane. Ces résultats sont en accord avec les radicaux détectés lors d’études RPE de
mélanges provenant de réactions Al–alcènes effectuées dans des conditions semblables. On discute des mécanismes de
formation des intermédiaires organoaluminium. Des intermédiaires organogallium semblables se forment lorsqu’on fait réagir
des atomes de gallium avec le styrène. Le couplage réducteur du styrène ne se produit pas lorsqu’on utilise des atomes d’In et
de Ti. On ne détecte que de très faibles quantités de phényléthane par hydrolyse des mélanges réactionnels.
Mots clés : atomes de métaux du groupe 13, intermédiaires organoaluminium, réactions d’atomes métalliques.
[
Traduit par la rédaction]
tures. The product studies usually involve the hydrolysis or
deuterolysis of the organometallic adducts formed in the metal
atom reaction. The nature of the organometallic intermediate
is inferred from the resulting hydrolysis products. For instance,
when the organoaluminium adduct resulting from the cocon-
densation of Al atoms and propene (5, 6) at 77 K was subjected
to deuterolysis the major products formed were deuterated pro-
pane, 2,3-dimethylbutane, 2-methylpentane, and traces of n-
hexane. The organoaluminium intermediate most consistent
with these products is 1.
Introduction
The reactions of metal atoms with organic substrates provide
interesting routes to the formation of novel organometallic in-
termediates. These organometallic intermediates are useful in
mediating the modification of organic compounds. The reac-
tion of Group 13 metal atoms with alkenes has been the subject
of numerous investigations (1–14, and footnote 3).
Product analysis and (or) electron paramagnetic resonance
EPR) spectroscopy have been used to study the intermediates
(
formed between Al atoms and alkenes at cryogenic tempera-
1
Received November 12, 1996.
2
A1
A1
H.A. Joly, M. Kepes, N. Roy, and J. Prpic. Department of
Chemistry and Biochemistry, Laurentian University, Sudbury,
ON P3E 2C6, Canada.
1
1
Revision received January 27, 1998.
Author to whom correspondence may be addressed.
2
The reactive intermediates formed when Al atoms are re-
acted with propene in an adamantane matrix at 77 K have been
studied by EPR spectroscopy (7, 8). Three radical species are
Telephone: (705) 675-1151, ext. 2333. Fax: (705) 675-4844.
E-mail: HJOLY@Nickel.laurentian.ca
J.A. Howard, H.A. Joly, and B. Mile. Unpublished results.
3
Can. J. Chem. 76: 400–406 (1998)
© 1998 NRC Canada

Joly et al.
401
thought to contribute to the overall EPR spectrum, namely,
π-allyl aluminium hydride (2), the allyl radical (3), and a di-
methylaluminacyclopentane (4), formed upon cyclodimeriza-
tion of propene.
bond. We present the results of a product study involving the
Group 13 organometallic adducts formed when Group 13 met-
al atoms are reacted with styrene and α-methylstyrene at cryo-
genic temperatures.
H
Experimental
Me
A1
H
H
CH
2
CH CH2
A1
H
Preparation of the reaction mixtures
A specialized metal atom reactor known as a rotating cryostat
H
H
Me
(15) was used to prepare the reaction mixtures. The device
consisted of a stainless steel drum held in the centre of a reac-
2
3
4
–
6
tion chamber that was maintained at ca. 10 Torr (1 Torr =
33.3 Pa). During operation of the cryostat the liquid nitrogen
In a few cases, Al atoms react with alkenes, e.g., ethene
1–4), CD CHTCH and CD CDTCD (14), to form mono-
ligand π-complexes, Al[alkene].
,3-Butadiene reacts with Al atoms to form two major
1
(
3
2
3
2
filled drum rotated at 2000 rpm while the alkene and the metal
vapour were introduced through portholes in the outer cham-
ber. The metal atom vapour (generated by resistively heating
Al wire, Ga, In, or Tl pellets in a tungsten basket fixed between
two electrodes of a furnace), condensed on the cold surface of
the rotating drum and was bombarded with the alkene at a
pressure of 0.1 Torr. Reaction was believed to occur at the
interface of the two reactants. After approximately 8 min, the
current supplied to the furnace was decreased to 0 A and the
alkene was allowed to condense onto the reaction mixture for
an additional 2 min. Next, distilled water or deuterium oxide
was introduced through a porthole at a pressure of 1 Torr for
1
paramagnetic species detected by EPR spectroscopy (9), a σ-
bonded aluminacyclopentene and aluminium-substituted allyl
radicals. It would appear that conjugated systems undergo in-
tramolecular cyclodimerization more easily than intermolecu-
lar cyclodimerization, which was observed for the simple
alkenes (3, 4, 7, 8) discussed above.
Two groups of workers have investigated the interaction
between Al atoms and benzene by EPR spectroscopy (10, 11)
and have shown that an Al–benzene molecular complex forms.
A good computer simulation of the Al–benzene species in a
neon matrix (10) at 4 K was obtained, assuming an interaction
3
min. The above sequence was repeated once. The reaction
mixture was transferred at 77 K and under reduced pressure
into a sample tube fitted with a rotoflow stopcock. The sample
was slowly warmed to room temperature by keeping the sam-
2
1
between the ground state Al atom (3s , 3p ) and only two of
the six protons of the benzene ring. This observation is consis-
tent with the formation of a π-complex, resulting from the
ple tube in a Dewar containing liquid N until the coolant had
2
dative bonding of the semi-occupied Al 3p orbital with two
x
antibonding p orbitals on two adjacent carbon atoms. In an-
evaporated. Next, approximately 1 mL of distilled water was
added to the sample tube. Samples that appeared to contain
excess metal were first treated with a few drops of 5% HCl.
Dichloromethane (ca. 1 mL) was added to the aqueous phase
to extract the organic compounds. Molecular sieves were
added to dry the dichloromethane layer, which was sub-
sequently analyzed by gas chromatography – mass spectrome-
try.
π
other study (11), the Al–benzene complex isolated in adaman-
tane was warmed to 220 K in the cavity of the EPR
spectrometer. The EPR spectrum in this case was best inter-
preted as the interaction of the Al atom with six equivalent
hydrogen nuclei. This led Howard et al. (11) to propose the
formation of an η -complex in which the Al atom lies along
6
the C6v axis of the benzene ring.
2
1
Often reactions of ground-state Ga atoms (4s , 4p ) mimic
the reactions of Al atoms. For instance, Ga atoms have also
been found to react with alkenes. The Ga–ethylene π-complex
is the major paramagnetic species formed when Ga atoms are
reacted with ethylene, in both argon (12) at 4 K and adaman-
tane (13) at 77 K. A second paramagnetic species, detected in
the Ga–ethylene reaction in adamantane (13), has been tenta-
tively assigned to gallacyclopentane. Ga atoms react with 1,3-
butadiene (9, and footnote 3) yielding 1-substituted allyl,
MCH CHCHCH . There is no evidence for the cheleotropic
Sample analysis
A Varian Saturn II gas chromatograph – mass spectrometer
operating in electron impact mode at 70 eV was used to ana-
lyze the hydrolysis products. The gas chromatograph was fit-
ted with a Varian 1093 septum-equipped programmable
injector (SPI) and an SPB-1 nonpolar fused silica column
(30 m × 0.25 µm film thickness, Supelco). Ultrapur helium
was used as the carrier gas with a column head pressure of 12
psi (1 psi = 6.9 kPa). The SPI injector was programmed to hold
at 40°C for 0.10 min at which time the temperature was ram-
ped from 40 to 280°C at a rate of 200°C/min. After an initial
2 min hold period the column temperature was increased from
40 to 140°C at 5°C/min, then from 140 to 200°C at 2°C/min
with a 5 min final hold period. The transfer line coupling the
GC to the MS was maintained at 260°C whereas the tempera-
ture of the ion trap manifold was kept at 220°C. The electron
impact mass spectral scans were recorded from m/z 40 to m/z
350 amu. The hydrolysis products were identified by compar-
ing the retention times and mass spectral data to those of
authentic samples.
2
2
addition of Ga leading to the formation of gallacyclopentene
as was observed for Al atoms. An EPR study of a Ga atoms
matrix isolated in benzene at 77 K (14) indicated that a trapped
atom or weak Ga–benzene complex forms that is unlike the
Al[C H ] complex found in the Al–benzene reaction. In addi-
6
6
tion, samples of Ga atoms in benzene, annealed in the cavity
of the EPR spectrometer, yielded well-resolved spectra of cy-
•
7
clohexadienyl radicals, C H .
6
Based on the studies involving the reaction of Group 13
metal atoms with alkenes, styrene presents two sites for reac-
tion with the metal atoms, the benzene ring and the double
©
1998 NRC Canada

4
02
Can. J. Chem. Vol. 76, 1998
Table 1. Hydrolysis products of the organoaluminium adducts
formed in the reaction of Al atoms with styrene at 77 K.
yielded 1,3-diphenyl-1-butene (bp 126–130°C, 3 Torr). Hy-
drogenation (17) of the alkene with H and PtO in ethanol
2
2
resulted in the formation of the desired product. 1,3-Diphenyl-
butane (2 g, 18% yield) was collected between 115 and 120°C
a
Entry
Relative amounts of the hydrolysis products
EtBz
2,3-DPB
2,3-DPB
1,3-DPB
1
at 2 Torr. H NMR (200 MHz, CDCl ) δ: 7.11–7.39 (m, 10 H),
3
2.60–2.78 (m, 1 H), 2.48–2.59 (m, 2H), 1.86–2.01 (m, 2 H),
dl
meso
+
1
.30 (d, 3H). GC–MS: 210 (30.3, [M] ), 133 (2.9), 119 (11.5),
1
2
3
26
43
51
21
28
18
31
26
14
30
21
2.5
3.4
0.0
0.0
9.4
4.0
7.7
9.9
b
106 (50.9), 105 (100), 92 (34.6), 91 (75.9), 79 (22.8), 77 (29.6),
1 (18.7).
6
9
5
39
b
6
9
e
b
64
29
Results and discussion
c
4
53
b
6
2
d
e
b
Reaction of Al atoms with styrene and α-methylstyrene
Aluminum atoms were reacted with styrene under cryogenic
temperatures. The ratio of the metal:styrene was approxi-
mately 1:10. Water was added to the reaction mixture. This
causes hydrolysis of the C—Al bonds and replacement of Al
with a hydrogen atom. The main hydrolysis products detected
by GC–MS analysis were phenylethane (EtBz), dl-2,3-
diphenylbutane (dl-2,3-DPB), meso-2,3-diphenylbutane
(meso-2,3-DPB), and 1,3-diphenylbutane (1,3-DPB), Table 1.
Confirmation of these assignments was obtained by comparing
the mass spectral data and retention times to those obtained for
authentic samples.
5
70
a
Based on AC /ΣAC × 100 where AC is the area counts
i
i
i
for product i, obtained from the electronic integration of the
peaks in the total ion current chromatogram.
b
Relative amounts of the isomeric diphenylbutanes.
c
D O was substituted for H O in the hydrolysis reaction.
2
2
d
Cyclohexane was deposited as an inert matrix (vapour
pressure 1 Torr) along with Al and styrene (vapour pressure
.1 Torr). No water was added to the reaction mixture until
0
after the mixture had been annealed to room temperature.
e
EtBz not determined for these cases.
From the entries 1, 2, and 3 in Table 1 it can be seen that
the relative amount of the hydrolysis products formed varied
from experiment to experiment, e.g., more phenylethane was
formed in the second experiment than in the first (cf. 43% vs.
Synthesis of standard samples
Preparation of meso- and dl-2,3-diphenylbutane
The meso and dl-2,3-diphenylbutane were synthesized by a
method suggested by Kochi (16). Acetone (5 mL) in 50 mL of
anhydrous diethyl ether was added dropwise to the Grignard
reagent prepared from magnesium turnings (4 g, 0.16 mol) and
26%). However, the ratio of dl-2,3-, meso-2,3-, and 1,3-
diphenylbutanes remains relatively constant for the three ex-
periments, i.e., 11.2:4.5:1. In all cases, the relative amount of
dl-2,3-DPB (69 ± 4%) in the mixture is greater than that of
meso-2,3-DPB (27 ± 4%), which in turn is greater than the
1-bromo-1-phenylethane (7.4 mL, 0.057 mol). The reaction
mixture was next hydrolyzed and the meso- and dl-2,3-
diphenylbutane recovered by ether extraction. Removal of the
ether with the aid of a rotary evaporator resulted in the partial
crystallization of a solid. The solid and mother liquor were
separated. Recrystallization of the solid from methylcyclohex-
ane – petroleum ether yielded 0.48 g (7%) of meso-2,3-
1,3-DPB (6 ± 4%).
Deuterolysis of the reaction mixture, entry 4, Table 1, re-
sulted in compounds that eluted at retention times similar to
those observed for the hydrolysis products. However, in each
case, the mass spectral data indicated deuterium incorporation.
For instance, the m/z of the main mass spectral fragments (and
the percentage abundance relative to the base peak) observed
for phenylethane were 106 (27%), 91 (100%), 77 (9%), 65
(12%), and 51 (25%). The fragment at m/z 106 is the molecular
ion, while the base peak with a m/z value of 91 corresponds to
diphenylbutane melting from 119 to 121°C (lit. (16) mp
1
1
24–126°C). H NMR (200 MHz, CDCl ) δ: 7.19–7.35 (m, 10
3
H), 2.70–2.90 (m, 2 H), 1.08 (d, 6 H). GC–MS: 209 (0.1, [M –
+
+
6 5 6 5 3
+
1
] ), 133 (2.4, [M – C H ] ), 105 (100, [C H CHCH ] ), 104
(
30.8), 91 (7.2), 79 (17.5), 77 (16.7), 63 (2.7), 51 (8.5).
The dl isomer (0.2 g, 3.5%) was recovered from the mother
+
the [C7H7] ion obtained from the rupture of the benzylic bond.
1
liquor. H NMR (200 MHz, CDCl ) δ: 7.20–7.60 (m, 10 H),
3
1
(
In the deuterolysis experiment, the m/z of the main mass
spectral fragments for the compound eluting at a similar reten-
tion time as that of phenylethane were 108 (37%), 92 (100%),
77 (2%), 65 (7%), and 51 (4%). This is consistent with the
incorporation of two deuterium atoms into the phenylethane
molecule because the molecular ion has an m/z of 108 and the
3
.05–3.20 (m, 2 H), 1.40 (d, 6 H). GC–MS: 209 (0.12, [M –
+
+
+
] ), 133 (1.4, [M – C H ] ), 105 (100, [C H CHCH ] ), 104
6
34.0), 91 (5.1), 79 (19.0), 77 (20.1), 63 (2.9), 51 (8.9).
5
6
5
3
Preparation of 1,3-diphenylbutane (16, 17)
+
A mixture containing 6.30 mL (0.054 mol) of acetophenone in
diethyl ether was added dropwise to a Grignard reagent pre-
pared by adding 7.38 mL (0.054 mol) of 1-bromo-2-
phenylethane in 50 mL of sodium-dried diethyl ether to 3 g
fragment with an m/z of 92 suggests that the [C H D] ion
7
6
formed in the fragmentation of the molecule. It has been con-
cluded, therefore, that one of the products formed on deutero-
lysis of the reaction mixture is phenylethane-1,2-d . A similar
2
(
H SO , 1,3-diphenyl-3-butanol was recovered from the reac-
tion mixture by ether extraction followed by removal of the
solvent on a rotary evaporator. Potassium bisulfate (14.7 g,
0.08 mol) was added to the alcohol and the mixture was heated
at 160–170°C for 30 min. Distillation of the resulting liquid
0.25 mol) of magnesium turnings. After hydrolysis with 10%
analysis of the mass spectral data of the remaining deuterolysis
products suggests that dl-2,3-DPB-1,4-d , meso-2,3-DPB-1,4-
2
4
2
d , and 1,3-DPB-1,4-d are formed. The mass spectral data of
2
2
the isomeric diphenylbutanes and their deuterated analogues
are presented in Table 2.
The deuterium atoms in the deuterolysis products mark
©
1998 NRC Canada

Joly et al.
403
Table 2. Mass spectral data for the isomeric diphenylbutanes and
their deuterated analogues.
Scheme 2.
PhCH CH2
a
Compound
dl-2,3-DPB
m/z for the main mass spectral fragments
+
133 (1.1, [M – C H ] ), 105 (100,
6
5
+
6 5 3 8 8
+
[
C H CHCH ] ), 104 (33.6, [C H ] ), 91
+
(
4.9, [C H ] ), 79(20.0), 77(22.6)
7 7
PhCH CH2
A1
+
dl-2,3-DPB-1,4-d2
meso-2,3-DPB
135 (0.7, [M – C H ] ), 106 (100,
6
5
+
+
[
C H CHCH D] ), 105 (49.0, [C H D] ),
6
5
2
8
7
+
9
1 (6.7, [C H ] ), 79(15.0), 77(19.7)
7 7
+
105 (100, [C H CHCH ] ), 104 (29.8,
6
5
3
+
+
[
C H ] ), 91 (5.6, [C H ] ), 79(19.1),
7
8
8
7
7
7(21.1), 63(2.6), 51(9.3)
+
PhCHCH A1
2
meso-2,3-DPB-1,4-d2
106 (100, [C H CHCH D] ), 105 (42.0,
2
6
5
+
+
[
C H D] ), 91 (8.9, [C H ] ), 79(17.4),
8
7
7
7
7
7(20.6), 63(4.6), 51(14.9)
+
PhCH CH
2
+
1
,3-DPB
210 (25.0, [M] ), 133(1.1, [M – C H ] ), 119
6 5
+
(
10.7, [M – C H ] ), 106 (49.5), 105 (100,
7 7
+
8 9 7 7
+
[
C H ] ), 92 (31.8), 91 (74.0, [C H ] ), 79
PhCHCH2
A1
PhCHCH2
A1
PhCHCH2
(
27.7), 77 (33.7), 51 (18.1)
+
+
1
,3-DPB-1,4-d2
212 (23.0, [M] ), 135 (9.5, [M – C H ] ),
6 5
+
CH CH
2
1
20 (10.5, [M – C H D] ), 107 (36.8), 106
6
7
+
(
100, [C H D] ), 105 (48.4), 93 (37.9), 92
8
Ph
9
+
(
57.9), 91 (33.7, [C H ] ), 79 (30.5), 77
7 7
(35.8), 51 (20)
a.
The numbers in parentheses represent the relative abundance of the
fragment.
PhCHCH
2
PhCHCH2
A1
PhCHCH2
A1
CH CH
2
Scheme 1.
Ph
H
H
H
H
H
H
H
Ph
H
H
H
A1
A1
A1
tion (hfi) of 4.0 G, and with four equivalent hydrogen atoms
with an hfi of 2.5 G. The two sets of four protons are consistent
with aluminacyclopentane where the four protons on the two
carbons adjacent to the Al have a 4 G hfi while the four protons
on the β carbons have a 2.5 G hfi.
H
H
Ph
H
H
H
H
H
H
Ph
5
Ph Ph
6
Ph
7
A possible mechanism for the formation of the aluminacy-
clopentane adducts involves the three-step process shown in
Scheme 2. In the first step, Al atoms add in a concerted fashion
to styrene to give 2-phenylaluminacyclopropane. The C—Al
bond dissociates in a second step to give the most stable alu-
mina–carbon radical. Then, the addition of a second molecule
of styrene results in the formation of a diradical. To explain
the formation of both 2,3- and 1,3-diphenylbutane, two differ-
ent diradical intermediates must form, i.e., attack of the β-carb-
on of styrene by the β-alumina carbon radical results in the
diradical that is the precursor of 2,3-diphenylbutane, whereas
attack of the α-carbon of styrene results in the diradical re-
sponsible for the formation of 1,3-diphenylbutane. In the third
step, a bond forms between the two radical centres to form the
aluminacyclopentane intermediates. The fact that the dl isomer
is favoured over the meso isomer suggests that ring closure is
stereoselective. One could argue that the transition state energy
for ring closure is lower when the gauche interactions between
D O
2
D O
2
D O
2
dl-DPB-1,4-d2
meso-DPB-1,4-d2
1,3-DPB-1,4-d2
where Al was attached in the original organometallic interme-
diates. The formation of the dideuteriodiphenylbutanes sug-
gests that two C—Al bonds have been ruptured and replaced
in the hydrolysis process by deuterium atoms. The deuterolysis
of the aluminacyclopentane intermediates 5, 6, and 7 would
yield the dideuteriodiphenylbutanes observed in the study,
Scheme 1.
Aluminacyclopentanes (3, 4, 7, 8) have been observed spec-
troscopically in the Al atom reactions of alkenes. In the reac-
tion of Al atoms with ethylene (3, 4), a paramagnetic
mononuclear aluminium species with EPR parameters similar
to those reported for aluminacyclopentene (9) was detected. In
the phenyl groups and the CH groups are minimized, as is the
2
case for structure Ia, Fig. 1. As a result, it is reasonable to
expect that the formation of Ia will be favoured kinetically and
that the dl isomer will form more readily than the meso isomer.
However, there is some concern about the fact that the di-
radical, which is proposed to be responsible for the formation
addition, the M = 1/2 line of the organoaluminium species
I
resolved, upon annealing the Al–ethylene sample, to give a
quintet of quintets from the interaction of the unpaired electron
with four equivalent hydrogen atoms with a hyperfine interac-
©
1998 NRC Canada

4
04
Can. J. Chem. Vol. 76, 1998
Fig. 1. Newman projections representing conformations of the
trans (Ia, Ib) and cis (IIa, IIb) 3,4-diphenylaluminacyclopentane.
Scheme 3.
PhCH CH2
H
Ph
Ph
Ph
CH
CH
2
CH
2
H
H
A1
A1
A1
CH
2
2
H
Ph
PhCH CH2
Ia
Ib
A1
Ph
Ph
Ph
H
CH
CH
2
Ph
CH
2
A1
A1
H
CH
2
2
H
H
PhCHCH A1
2
IIa
IIb
PhCH CH2
of 1,3-diphenylbutane, has a primary radical centre. The rela-
tive stabilities of the primary versus the benzyl radical centre
can be estimated by comparing bond dissociation energies. For
instance, the bond dissociation energy of the secondary C—H
bond of phenylethane (18) is 357.3 kJ/mol whereas the bond
dissociation energy of the primary C—H bond of ethane (18)
is 422.8 kJ/mol. The benzyl radical is therefore significantly
more stable than the primary radical centre. Considering that
the 1,3 isomer represents anywhere from 3.4 to 9.4% of the
reaction mixture, one would expect that the difference in en-
ergies of the intermediates leading to formation of the 1,3 and
H
H
H
C
Ph
CH A1
H
H
H
H
Ph
Ph
Ph
CH A1
2
CH A1
2
Ph
2
Ph
CH2
CH2
H
2
,3 isomers would only be of the order of 4.184 kJ/mol and
H
H
H
H
H
H
Ph
H
H
A1
A1
A1
not of the order of 422.8 – 357.3 or 65.5 kJ/mol. Therefore, it
is unlikely that the Al atom reaction of styrene proceeds via
the stepwise mechanism proposed above.
H
Ph
H
H
H
H
H
H
H
H
Shusterman et al. (19) have suggested that dimethylger-
mylene produced by the thermal decomposition of 7,7-di-
methyl-1,4,5,6-tetraphenyl-7-germabenzonorbornadiene adds
to (E)-2-deuteriostyrene by a mechanism comparable to that
shown in Scheme 2. These workers found that only the cis and
trans isomers of 2,5-dideuterio-1,1-dimethyl-3,4-diphenylger-
macyclopentane were formed. There was no evidence for the
formation of 1,1-dimethyl-2,4-diphenylgermacyclopentane.
In addition, unlike the Al atom reaction, the cis and trans iso-
mers were formed in a 1:1 ratio.
Ph
Ph Ph
H
Ph
drolysis of the organoaluminium intermediates gave two main
products. The compound that eluted first had a retention time
and mass spectrum identical to those of an authentic sample
of 2-phenylpropane. The m/z values of the main mass spectral
fragments (and the percentage abundance relative to the base
peak) observed for the second compound were 119 (100%),
103 (7.2%), 91 (57.5%), 77 (11.9%), 65 (3.3%), and 51 (6.2%).
These data are consistent with 2,3-dimethyl-2,3-diphenylbu-
Alternatively, the Al–styrene reaction could proceed by a
two-atom insertion into the 2-phenylaluminacyclopropane,
Scheme 3. In this process, the β-alumina radical, PhCH-
tane. The base peak has an m/z of 119 and is assigned to
+
+
6 5 3 2 7 7
[C H C(CH ) ] . The peak at an m/z of 91 is due to [C H ]
and forms as a result of the elimination of C H from
CH Al, is trapped by styrene in a concerted manner. The fact
2
that three isomers of diphenylbutane were observed suggests
that the radical character of the α and β carbons in the transi-
tion state is not very advanced, i.e., the transition state is early.
The observation that the reaction mixtures always contain
greater amounts of the dl isomer can be explained in terms of
a steric effect. The transition state energy is lowest for the
structure where the phenyl rings are trans. The phenylethane
forms from the hydrolysis of 2-phenylaluminacyclopropane.
The incorporation of two deuterium atoms into the
phenylethane, at the α and β carbons, is consistent with this
interpretation.
2
C H C(CH ) ] . A molecular ion was not observed, as was
4
+
[
6 5 3 2
the case for the symmetric compound, 2,3-diphenylbutane.
The 2-phenylpropane can be explained by the hydrolysis of
2-methyl-2-phenylaluminacyclopropane, formed by the con-
certed addition of Al atoms to α-methylstyrene. The 2,3-di-
methyl-2,3-diphenylbutane is formed upon hydrolysis of
3,4-dimethyl-3,4-diphenylaluminacyclopentane. Support for
this structure comes from the deuterolysis experiments, which
indicate that deuterium is incorporated in the methyl groups
(at C1 and C4), Table 3. The aluminacyclopentane intermediate
could form in one of at least two ways. The first reaction path-
way involves the addition of the β-alumina-α-methylstyrl radical,
PhC(CH )CH Al, to α-methylstyrene yielding the diradical,
Seyferth and co-workers (20, 21) suggested that hex-
amethylsilirane transfers Me Si to styrene, giving 1,1,2,2,3,3-
2
hexamethyl-5-phenylsilacyclopentane by a two-atom insertion
comparable to that proposed for the Al–styrene reaction.
Next, Al atoms were added to α-methylstyrene. The hy-
3
2
PhC(CH )CH AlCH (CH ) C_ _ P_ _h _. _T _h _e _ d_ i_r _a d_ i_ c _a _l _r _i n_ g_ __c loses in
3
2
2
3
a subsequent step to give PhC(CH )CH AlCH (CH )CPh. The
3
2
2
3
©
1998 NRC Canada

Joly et al.
405
a
Table 3. Mass spectral data for the hydrolysis and deuterolysis
products of the Al–α-methylstyrene reaction mixture.
Table 4. Relative amounts of the isomeric
diphenylbutanes found for the organoaluminium
adducts formed in the reaction of Ga atoms with
styrene at 77 K.
m/z for the main mass spectral
a
fragments
Compound
-Phenylpropane
a
+
Entry Relative amounts of the hydrolysis products
2
2
119 (16.3, [M – 1] ), 105 (100,
+
+
[
M – CH ] ), 91 (8.1, [C H ] ),
3
2,3-DPB
2,3-DPB
1,3-DPB
7
7
7
7 (20.5), 63 (4.7), 51 (16.9)
+
dl
meso
-Phenylpropane-1,2-d2
121 (12.2, [M – 1] ), 107 (99, [M
+
1
2
3
4
5
6
60.8
67.3
65.8
67.7
58.5
52.5
23.1
21.7
22.1
21.5
26.6
28.5
16.1
11.0
12.1
10.8
14.9
19.9
–
CH ] ), 106 (100, [M –
3
+
+
CH D] ), 92 (7.6, [C H D] ),
2
7
6
7
7 (13.8), 63 (5.6), 51 (18.4)
+
2
2
,3-Dimethyl-2,3-DPB
119 (100, [C H C(CH ) ] ), 103
6 5 3 2
b
c
(
(
7.2), 91 (57.5), 77 (11.9), 65
3.3), 51 (6.2)
a
+
Based on AC /ΣAC × 100 where AC is the area counts
i i i
,3-Dimethyl-2,3-DPB-1,4-d2
120 (100, [C H C(CH )(CH D)] ),
6
5
3
2
for product i, obtained from the electronic integration of the
1
7
04 (5.2), 92 (32.7), 91 (64.8),
7 (15.0), 65 (4.5), 51 (9.3)
peaks in the total ion current chromatogram.
b
D O was substituted for H O in the hydrolysis reaction.
2
2
a
The numbers in parentheses represent the relative abundance of the
c
Cyclohexane was deposited as an inert matrix (vapour
pressure 1 Torr) along with Ga and styrene (vapour pressure
.1 Torr). No water was added to the reaction mixture until
after the sample had been annealed to room temperature.
fragment.
0
β-alumina-α-methylstyrl radical could form by the ring open-
ing of 2-methyl-2-phenylaluminacyclopropane.
Alternatively, the β-alumina-α-methylstyrl radical could
insert into an α-methylstyrene molecule by a concerted two-
atom insertion, as was suggested for styrene, yielding 3,4-di-
methyl-3,4-diphenylaluminacyclopentane. We are not able to
tell by which mechanism the reaction proceeds. However, it is
intere _s _t i_n _g _ t_o _ _n _o _t e_ _t h_ a_t _t _h _e _r e_ _i s__ no evidence for the formation
of PhC(CH )CH AlC(CH )PhCH , suggesting that if the proc-
noting that there is tentative EPR evidence for the formation
of gallacyclopentane when Ga atoms are reacted with ethene
(13).
Indium and thallium atoms are not very reactive towards
styrene. We were only able to detect traces of phenylethane in
the hydrolyzed reaction mixtures. Klabunde and co-workers
(22) found the reactivity order of Group 13 metal atoms/clus-
ters for the reaction with methyl bromide to be Al > Ga > In
> Tl. They argued that the reactivity displayed by the atoms
would be influenced by the heats of vaporization (which are
also in the order Al > Ga > In > Tl) in that the reactivity would
depend on the energy put into the system during vaporization.
Hauge et al. (23) have reported a similar reactivity dependency
in Group 13 atom – water reactions. Whereas aluminium at-
oms were found to react spontaneously with water to form
HAlOH, the corresponding Ga and In products were only
formed after photolysis of the reaction matrix and no insertion
product could be detected for the case of Tl and water.
3
2
3
2
ess were concerted the radical character of the transition state
is advanced.
Reaction of other Group 13 metal atoms with styrene
The study was extended to include the reactions of several
other Group 13 metal atoms, namely, Ga, In, and Tl, with
styrene. In addition to phenylethane, dl-2,3-, meso-2,3-, and
1,3-DPB, hydrolysis of the Ga–styrene reaction intermediates
resulted in the formation of compounds with retention times
and mass spectral data identical to those of authentic samples
of 1-phenylethanone and styrene oxide. A compound with
mass spectral data consistent with 1-phenylethanol was also
detected. (The formation of 1-phenylethanone, styrene oxide,
and 1-phenylethanol will be the subject of a separate publica-
tion.) The ratio of the diphenylbutane isomers in the mixture is
similar to that observed for the Al atom reaction in that the [dl]
Acknowledgements
The authors wish to express thanks to Drs. K.C. Westaway,
W. Rank, and J.A. Howard for useful discussions. The com-
ments made by the referees are also greatly appreciated. The
authors gratefully acknowledge the financial support provided
by the Natural Sciences and Engineering Research Council of
Canada in the form of an operating grant (H.A.J.), an equip-
ment grant (H.A.J.), and a student summer fellowship (N.R.).
Financial support from the Laurentian University research
fund is also acknowledged.
>
[meso] > [1,3], Table 4. Also, the relative amounts of the dl
and meso compounds are similar. However, there appears to be
more of the 1,3-DPB formed in the Ga–styrene hydrolysis re-
action. The deuterolysis of the Ga–styrene reaction mixture
yielded phenylethane-1,2-d2 (78%), meso-2,3-DPB-1,4-d2
(
(
13%), dl-2,3-DPB-1,4-d2 (5.8%), and 1,3-DPB-1,4-d2
3.2%).
Phenylethane and the isomeric diphenylbutanes are the hy-
drolysis products of 2-phenylgallacyclopropane, cis- and
trans-3,4-diphenylgallacyclopentane, and cis- and trans-2,4-
diphenylgallacyclopentane. It would seem likely that the
mechanism for the formation of the organogallium intermedi-
ates is comparable to that suggested for the reaction of Al
atoms with styrene as the same product ratios and deuterium
incorporation were observed for both cases. It is also worth
References
1
. P.H. Kasai and D. McLeod, Jr. J. Am. Chem. Soc. 97, 5609
(1975).
2. P.H. Kasai. J. Am. Chem. Soc. 104, 1165 (1982).
3. J.H.B. Chenier, J.A. Howard, and B. Mile. J. Am. Chem. Soc.
109, 4109 (1987).
©
1998 NRC Canada

4
06
Can. J. Chem. Vol. 76, 1998
4
. J.A. Howard, B. Mile, J.S. Tse, and H. Morris. J. Chem. Soc.
Faraday Trans. 1, 83, 3701 (1987).
15. A.J. Buck, B. Mile, and J.A. Howard. J. Am. Chem. Soc. 105,
3381 (1983).
5
6
. P.S. Skell and L.R. Wolf. J. Am. Chem. Soc. 94, 7919 (1972).
. M.J. McGlinchey and P.S. Skell. Angew. Chem. Int. Ed. Engl.
1
1
6. J.K. Kochi. J. Org. Chem. 28, 1960 (1963).
7. M. Casey, J. Leonard, B. Lygo, and G. Procter. Advanced prac-
tical organic chemistry. Blackie Academic and Professional,
New York, 1993. pp. 236–237.
1
. M. Histed, J.A. Howard, H. Morris, and B. Mile. J. Am. Chem.
4, 1975 (1975).
7
8
9
Soc. 110, 5290 (1988).
1
8. Handbook of chemistry and physics. 76th ed. CRC Press, New
York. 1995–1996. pp. 9–51.
. J.A. Howard, H.A. Joly, B. Mile, M. Histed, and H. Morris.
J. Chem. Soc. Faraday Trans. 1, 84, 3307 (1988).
. J.H.B. Chenier, J.A. Howard, J.S. Tse, and B. Mile. J. Am.
Chem. Soc. 107, 7290 (1985).
19. A.J. Shusterman, B.E. Landrum, and R.L. Miller. Organometal-
lics, 8, 1851 (1989).
1
0. P.H. Kasai and D. McLeod, Jr. J. Am. Chem. Soc. 101, 5860
1979).
1. J.A. Howard, H.A. Joly, and B. Mile. J. Am. Chem. Soc. 111,
094 (1989).
2. P.M. Jones and P.H. Kasai. J. Phys. Chem. 92, 1060 (1988).
3. J.A. Howard, H.A. Joly, and B. Mile. J. Phys. Chem. 96, 1233
2
2
2
0. D. Seyferth and D.C. Annarelli. J. Organomet. Chem. 117, C51
1976).
1. D. Seyferth, D.P. Duncan, and S.C. Vick. J. Organomet. Chem.
25, C5 (1977).
2. Y. Tanaka, S.C. Davis, and K.J. Klabunde. J. Am. Chem. Soc.
04, 1013 (1982).
(
(
1
8
1
1
1
1
(
1992).
4. J.A. Howard, H.A. Joly, B. Mile and R. Sutcliffe. J. Phys. Chem.
5, 6820 (1991).
1
23. R.H. Hauge, J.W. Kauffman, and J.L. Margrave. J. Am. Chem.
Soc. 102, 6005 (1980).
9
©
1998 NRC Canada