2
698 J . Org. Chem., Vol. 62, No. 9, 1997
DeHaan et al.
with toluene, as suggested by the appearance of a yellow
color17 when toluene is added to a colorless solution of
lar interest is the ICR study25 of the same reaction, ArH
+
+ CH
3
OCH
2
, where the electrophile exhibits low sub-
/k ) 1.8 ( 0.1) and statistical
AlCl
3
in nitromethane. We observed no color change with
strate selectivity (k
T
B
benzene, which is consistent with its lower ionization
potential. Although our chromatograms showed none of
the typical photochemical products of the CT structures
benzyl chloride, chlorotoluenes, bibenzyl, methyldiphen-
ylmethanessthis does not prelude its existence.18 How-
positional selectivity, a marked contrast to our solution
phase results. However, it is important to note that
intramolecular isomerization of the intermediate are-
nium ion was also suggested as a possible reason for the
seemingly nonexistent positional selectivity.
1
7
ever, how either complex [ArCH
3
, CH
3
NO
2
, AlCl
3
] or
Finally, the inverse rate order dependence on metal
halides has been reported elsewhere for reactions involv-
ing acyl chlorides: for benzoylation reactions involving
•
+
•-
[ArCH , CH NO , AlCl ] would accelerate toluene
3 3 2 3
chloromethylation is unclear.
Although the zero-order arene dependence in the case
of both chloromethylation with MAC and with CMME
2
6
3 5 3 3
AlCl in nitrobenzene, with SbCl , FeCl and GaCl in
excess benzoyl chloride,14 and with 2,4-dichlorobenzoyl
2
7
prevents meaningful k
rate constants, the competitive method measures k
T
/k
B
determination from absolute
/k
chloride and AlCl in nitromethane. It appears that the
3
T
B
explanation offered by J ensen and Brown14 might apply
for the electrophile-arene reactions from benzene and
toluene reaction products. The results of our competitive
studies are presented in Table 4, and point to a common,
remarkably selective electrophile. While either a chlo-
romethyl methyl ether-aluminum chloride polarized
to our reaction, i.e., the production of an ionic aluminum
-
-
speciessAlCl4 , Al Cl7 sthrough other processes could
2
+
decrease the CH OCH2 concentration and thus inhibit
3
the reaction. Evidence for the production of such species
comes from an 27Al NMR study of solutions of AlCl or
3
+
-
28
adduct or ion pair CH
3
OCH
2
Al
2
Cl
7
or the methoxy-
AlBr in nitromethane, nitroethane, and nitrobenzene.
3
+
27
methyl cation CH
3
OCH
2
would appear to be possibilities,
Tarasov et al. observed narrow (e10 Hz) Al signals
+
-
-
we think the evidence favors CH
phile.
3
OCH
2
as the electro-
characteristic of AlCl4 (-101.9 ppm) and AlBr4 (-79.9
ppm) for all solutions in the 0 to -30 °C range, which
First, since a polarized addition compound between
CMME and AlCl should readily form, one would expect
increased with increasing AlX concentration.
3
3
We hoped NMR studies of AlCl
-MAC and AlCl -
3
3
chloromethylation with CMME to involve arene depen-
dence in the rate-determining step. Second, if either a
polarized complex or ion pair were the electrophile, one
would expect different substrate and positional selectivity
CMME complexation would help elucidate the mecha-
nisms of these reactions. However, even at temperatures
, the 1H and
13
around the freezing point of CD
spectra of the reagents were essentially unchanged by
the presence of AlCl . This is consistent with the heat
of solution of AlBr in nitrobenzene, being some 23 kcal/
mol more exothermic than that of AlBr in ethyl bro-
preferentially coordinates with
3
NO
2
C
with SnCl
as we will argue shortly, inverse rate order dependence
on [AlCl implies solvent-separated ions.
4
, which was not observed (vide infra). Third,
3
29
3
3 0
]
3
The exceedingly high selectivity exhibited in these
chloromethylation reactions requires a diffusionally equili-
brated electrophile of low reactivity. This is consistent
with a methoxymethyl cation in which the resonance
contribution of the methoxy group is important (Taft et
30
mide, and that AlCl
3
31
nitrobenzene over benzoyl chloride. Given our inability
to perform complexometric studies as well as the kinetics
complexities of this system, we switched our efforts to
4
SnCl -catalyzed chloromethylation in dichloromethane.
B. Kin etic a n d Sp ectr oscop ic Resu lts in Solven t
Dich lor om eth a n e. 1. Ch lor om eth yla tion of Ben -
zen e or Tolu en e w it h Met h oxya cet yl Ch lor id e
CH OCH2+ T CH +OdCH2
3
3
al.19 calculate a CH
OCH
2
+(g) stabilization energy, rela-
3
(
4
MAC) a n d Sn Cl . For the reasons just enumerated, a
+
tive to CH
3
(g), of 66 ( 3 kcal/mol) but which maintains
series of kinetics studies were begun using dichlo-
1
3
its carbonium ion character, as is suggested by C NMR
1
romethane, a solvent of weaker donating ability. Be-
2
0
+ 20
of CH
3
OCH
3
(CH
3
, 59.4 ppm) and CH
3
OCH
2
3
(CH ,
cause AlCl
because SnCl
3
has very limited solubility in this solvent and
is soluble in CH Cl , has been used
8
2.3 ppm; CH
2
, 219.8 ppm). This also fits with Olah’s
4
2
2
2
1
+
finding that the very weak electrophilicity of MeSCH
could be enhanced by lowering the electron-donating
tendency of S through complexation with AlCl Al-
though k /k ratios were not determined, there was “no
detectable amount of meta [product] isomer”, suggesting
2
successfully in other chloromethylation studies and has
119
a fairly abundant isotope, Sn, that has proven useful
3
.
in NMR studies of complexes of Sn compounds with
T
B
(23) For example, see: Broeker, J . L.; Hoffmann, R. W.; Houk, K.
+
similar very high selectivity of the CH
phile.
3
SCH
2
electro-
N. J . Am. Chem. Soc. 1991, 113, 5006. Apeloig, Y.; Karni, M. J . Chem.
Soc., Perkin Trans. 2 1988, 625. Okada, S.; Abe, Y.; Taniguchi, S.;
Yamabe, S. J . Am. Chem. Soc. 1987, 109, 295. Maccoll, A. Org. Mass
Spectrom. 1982, 17, 1. Lossing, F. P. J . Am. Chem. Soc. 1977, 99, 7526.
Beauchamp, J . L.; Dunbar, R. C. J . Am. Chem. Soc. 1970, 92, 1477.
(24) Amyes, T. L.; J encks, W. P. J . Am. Chem. Soc. 1989, 111, 7888.
Cremer, D.; Gauss, J .; Childs, R. F.; Blackburn, C. J . Am. Chem. Soc.
Although alkoxymethyl cations have been postulated
as intermediates in solvolysis reactions,22 most of the
2
2,23
work on these ions has involved the gas phase
or low-
temperature NMR studies in Magic acid.20,24
Of particu-
1
985, 107, 2435. Akhmatdinov, R. T.; Kantor, E. A.; Imashev, U. B.;
Yasman, Ya. B.; Rakhmankulov, D. L. Zh. Org. Khim. 1981, 17, 718.
(25) Dunbar, R. C.; Shen, J .; Melby, E.; Olah, G. A. J . Am. Chem.
Soc. 1973, 95, 7200.
(26) Brown, H. C.; Young, H. L. J . Org. Chem. 1957, 22, 724.
(27) DeHaan, F. P.; Covey, W. D.; Delker, G. L.; Baker, N. J .; Feigon,
J . F.; Ono, D.; Miller, K. D.; Stelter, E. D. J . Org. Chem. 1984, 49,
3959.
(
(
17) Bruggermann, K.; Kochi, J . K. J . Org. Chem. 1992, 57, 2956.
18) Bruggermann, K.; Czernuszewicz, R. S.; Kochi, J . K. J . Phys.
Chem. 1992, 96, 4405.
19) Martin, R. H.; Lampe, F. W.; Taft, R. W. J . Am. Chem. Soc.
966, 88, 1353.
20) The 13C NMR spectrum of CH OCH +
was taken in SbF /SO -
3 2 5 2
(
1
(
FCl at -75 °C. Farcasiu, D.; Horsley, J . A. J . Am. Chem. Soc. 1980,
(28) Tarasov, V. P.; Kirakosyan, G. A.; Randarevich, S. B.; Buslaev,
Yu. A. Koord. Khim. 1984, 10, 487.
(29) Plotnikov, V. A.; Vaisberg, R. G. Zap. Inst. Khim. Akad. Nauk
SSSR 1941, 35, 2405; Chem. Abstr. 1941, 35, 2405.
(30) Lebedev, N. N. J . Phys. Chem. USSR 1948, 22, 1505.
(31) J ensen, F. R.; Brown, H. C. J . Am. Chem. Soc. 1958, 80, 4038.
1
02, 4906. See also: Olah, G. A.; Bollinger, J . M. J . Am. Chem. Soc.
967, 89, 2995.
1
(21) Olah, G. A.; Wang, Q.; Neyer, G. Synthesis 1994, 276.
(
22) Pau, J . K; Kim, J . K.; Caserio, M. C. J . Am. Chem. Soc. 1978,
1
00, 3838.