Isolation, X-ray structures, and electronic spectra of reactive intermediates in Friedel-Crafts acylations

Article abstract of DOI:10.1021/jo0501588

Reactive intermediates in the Friedel-Crafts acylation of aromatic donors are scrutinized upon their successful isolation and X-ray crystallography at very low temperatures. Detailed analyses of the X-ray parameters for the [1:1] complexes of different al

Full text of DOI:10.1021/jo0501588

Isolation, X-ray Structures, and Electronic Spectra of Reactive
Intermediates in Friedel-Crafts Acylations†
M. G. Davlieva, S. V. Lindeman, I. S. Neretin, and J. K. Kochi*
Department of Chemistry, University of Houston, Houston, Texas 77004-5003
jkochi@uh.edu
Received January 25, 2005
Reactive intermediates in the Friedel-Crafts acylation of aromatic donors are scrutinized upon
their successful isolation and X-ray crystallography at very low temperatures. Detailed analyses
of the X-ray parameters for the [1:1] complexes of different aliphatic and aromatic-acid chlorides
with the Lewis acids antimony pentafluoride and pentachloride, gallium trichloride, titanium and
zirconium tetrachlorides provide unexpected insight into the activation mechanism for the formation
of the critical acylium carbocations. Likewise, the X-ray-structure examinations of aliphatic and
aromatic acylium electrophiles also isolated as crystalline salts point to the origins of their
electrophilic reactivity. Although the Wheland intermediates (as acylium adducts to arene donors)
could not be isolated in crystalline form owing to their exceedingly short lifetimes, transient (UV-
vis) spectra of benzenium adducts of acylium carbocations with hexamethylbenzene can be measured
and directly related to Wheland intermediates with other cationic electrophiles that have been
structurally established via X-ray studies.
Introduction
identification of such reactive intermediates has largely
relied on transient spectroscopy, especially NMR and IR
techniques as elegantly employed by Olah and co-workers
in their pioneering studies.3-5
We now explore some of the limits to which the
ultimate structural tool of X-ray crystallography can be
employed in the experimental study of these reactive
intermediates insofar as the definitive bond (length/
angle) parameters that are well provided by X-ray
structures can shed additional insight onto their reactiv-
ity.6,7 It is important to recognize, however, that the
requisite isolation of metastable species in crystalline
form poses a difficult experimental problem since those
The commonly used Lewis acid induced or Friedel-
Crafts acylation of various aromatic substrates with
carboxylic-acid halides is one prototypical (textbook)
example of a wide variety of electrophilic aromatic-
substitution processes.1 As such, mechanistic studies
have generally focused on three principal stages of
reaction involving: (I) activation of acyl halide by Lewis
acid, (II) heterolysis of the acyl-halogen bond to generate
the reactive acylium electrophile, and (III) electrophilic
attack on arene donors via cationic σ-adducts.2 The
† This paper is dedicated to George Olah and was presented (in part)
at the Loker Hydrocarbon Institute celebrating “125 Years of Friedel-
Crafts Chemistry”.
(1) See, e.g.: (a) McMurry, J. Organic Chemistry; Brooks/Cole:
Monterey, CA, 1984; p 529 ff. (b) Solomons, T. W. G. Organic
Chemistry, 6th ed.; Wiley: New York, 1996; p 664 ff. (c) Wade, L. J.,
Jr. Organic Chemistry; Prentice Hall: Upper Saddle River, NJ, 1999;
p 757 ff. (d) Streitwieser, A., Jr.; Heathcock, C. H. Introduction to
Organic Chemistry, 2nd ed.; Macmillan: New York, 1981; p 690 ff.
(2) (a) Taylor, R. Electrophilic Aromatic Substitution; Wiley: New
York, 1990. (b) March, J. Advanced Organic Chemistry, 4th ed.;
Wiley: New York, 1992; p 501 ff.
(3) Olah, G. A. Friedel-Crafts Chemistry; Wiley: New York, 1973.
(4) Olah, G. A., Ed. Friedel-Crafts and Related Reactions; Inter-
science: New York, 1964; Vol. III.
(5) For the latest overview, see: Olah, G. A.; Molnar, A. Hydrocarbon
Chemistry, 2nd ed.; Wiley-Interscience: Hoboken, NJ, 2003.
(6) Insofar as X-ray structural parameters reflect molecular electron-
density distributions.7
(7) See, e.g.: Koritsanszky, T. S.; Coppens, P. Chem. Rev. 2001, 101,
1583.
10.1021/jo0501588 CCC: $30.25 © 2005 American Chemical Society
Published on Web 04/12/2005
J. Org. Chem. 2005, 70, 4013-4021
4013

Davlieva et al.
FIGURE 1. Molecular (ORTEP) structure of the [1:1] complex
of benzoyl chloride with antimony pentachloride.
FIGURE 2. Molecular structure of the [2:2] adduct of benzoyl
chloride with titanium tetrachloride.
which are successfully crystallized from solution are not
necessarily the principal intermediates.8 With this caveat
in mind, our aim in this study is to ascertain how the
twin tools provided by (a) X-ray crystallography of
isolable species and (b) the time-resolved electronic
spectroscopy of other intermediates with lifetimes (τ)
<10-8 s can be exploited to provide new insight into the
mechanism of electrophilic aromatic acylation. Experi-
mentally, we employ low-temperature crystallization and
X-ray crystallography at -150 °C in conjunction with
UV-vis spectroscopy to identify species too labile to
isolate in crystalline form.9
FIGURE 3. Molecular structure of the intramolecular Friedel-
Crafts product (hydrindone) as the dimeric titanium-tetra-
chloride complex.
fluoro analogue by slowly adding a CH2Cl2 solution of
p-FC6H4COCl to a SbCl5 solution held at -40 °C. After a
short induction period, colorless crystals of the [1:1]
adduct were collected, washed with cold CH2Cl2, and
Results
To mimic reaction conditions found to be favorable to
Friedel-Crafts acylations, we initially examined the
direct interaction of both aliphatic and aromatic-acid
chlorides in dichloromethane solution with the pure
metal chlorides: SbCl5 and GaCl3 as well as TiCl4 and
ZrCl4, to represent Lewis acids of moderate and limited
reactivity, respectively.
I. Isolation and X-ray Structures of Lewis Acid
Complexes with Acyl Halides. A rigorously dry dichlo-
romethane solution of freshly purified benzoyl chloride
was cooled to -10 °C and the CH2Cl2 solution of antimony
pentachloride was slowly added with stirring under a dry
argon atmosphere. The colorless precipitate was collected,
washed with cold dichloromethane, and recrystallized by
the slow cooling of a solution of the complex that was
presaturated at 30-40 °C. A single crystal was isolated
at low temperature and X-ray crystallography was car-
ried out at -150 °C to a uniform precision of 0.3-0.5 pm
(esd). The ORTEP diagram in Figure 1 shows the [1:1]
adduct of C6H5COCl and SbCl5 that is σ-bonded through
the carbonyl oxygen,10 i.e.,
dried in vacuo. The carbonyl stretching band at νCO
)
1561 cm-1 for this crystal was essentially the same as
that of the benzoyl analogue (vide supra) to confirm that
O-complexation had occurred in the manner shown in
Figure 1. The slightly deactivated p-fluorobenzoyl chlo-
ride also afforded the colorless [1:1] adduct of gallium
trichloride that was isostructural with the benzoyl-
chloride complex of SbCl5, as established by the bonding
parameters in Table 1.
The weaker Lewis acid, TiCl4, afforded a series of [2:2]
adducts with p-fluorobenzoyl chloride, the parent benzoyl
chloride, and with the slightly activated p-toluoyl chlo-
ride, in which the Lewis acid exists as a dimeric unit with
each titanium center directly bonded to the carbonyl
oxygen (Figure 2), e.g.,
in the manner found with SbCl5 and GaCl3, as listed in
Table 1.
A similar dimeric Lewis acid unit was also found as
the [2:2] complex of zirconium tetrachloride with the
aliphatic-acid chloride, tert-butylacetyl chloride.
A slightly varied procedure was successfully used for the
(8) When a number of reactive intermediates are extant in solution,
especially under equilibrium control, those which are isolated in crystal
form generally depend on their solvation energy, crystal-lattice energy,
molecular symmetry, etc., and not necessarily on their abundance
(importance).
(9) Rathore, R.; Kochi, J. K. Adv. Phys. Org. Chem. 2000, 33, 193.
(10) For previous X-ray studies of analogous structures, mostly
carried out at higher (room) temperature and lower precision, see: (a)
Chevrier, B.; Le Carpentier, J.-H.; Weiss, R. Acta Crystallogr. 1972,
B28, 2659, 2667. (b) Rasmussen, S. E.; Broch, N. C. Acta Chem. Scand.
1966, 20, 1351. (c) Le Carpentier, J.-H.; Weiss, R. Acta Crystallogr.
1972, B28, 1437, 1442. (d) Chevrier, B.; Le Carpentier, J.-H.; Weiss,
R. J. Am. Chem. Soc. 1972, 94, 5718. (e) Baroni, T. E.; Kolesnichenko,
V.; Seib, L.; Heppert, J. A.; Liable-Sands, L. M.; Yap, G. P. A.;
Rheingold, A. E. Polyhedron 1998, 17, 759.
Interestingly, the direct interaction of the aliphatic
hydrocinnamyl chloride with both titanium and zirco-
nium chloride at -60 °C for several days yielded crystals
of only the intramolecular Friedel-Crafts product as
hydrindone complexed with the Lewis acid in the [2:2]
complex shown in Figure 3.
II. Isolation and X-ray Structures of Acylium
Carbocations. Freshly purified p-methylbenzoyl chlo-
4014 J. Org. Chem., Vol. 70, No. 10, 2005

Reactive Intermediates in Friedel-Crafts Acylations
TABLE 1. X-ray Structural Parameters of Lewis acid Adducts to Aliphatic and Aromatic-Acid Chloridesa
a Diffraction data collected at -150 °C and esd’s in parentheses.
cation as the separated ion pair owing to the wide
interionic separation of 3.29 Å (C‚‚‚ClSbCl5).
Furthermore, the trimethyl analogue, mesitoyl chlo-
ride, also reacted directly with antimony pentachloride
under similar low-temperature conditions to yield the
acylium salt, 2,4,6-trimethylphenyloxocarbonium hexachlo-
roantimonate, as structurally established by X-ray crys-
tallography (Table S1 in the Supporting Information). In
contrast to benzoyl chloride, no crystals of the corre-
sponding Lewis acid complexes (vide supra) were found
when either p-toluoyl or mesitoyl chloride was treated
with antimony pentachloride under the same conditions.
The clear-cut distinction between the formation of the
Lewis acid complex versus the acylium salt, i.e.,
FIGURE 4. ORTEP diagram of p-methylbenzoyl carbocation
ion paired with hexachloroantimonate.
ride was treated directly with antimony pentachloride
in dichloromethane by essentially the same procedure as
described above for the parent benzoyl chloride (see
Experimental Section for details). However, X-ray crys-
tallographic analysis of the colorless crystals at -150 °C
led to the ORTEP diagram in Figure 4 that depicts the
[1:1] salt comprised of the p-methylbenzoyl cation and
the coordinatively saturated hexachloroantimonate coun-
teranion,11 i.e.,
was also observed when acid fluorides were treated with
antimony pentafluoride in dichloromethane solution.
Thus benzoyl fluoride and SbF5 yielded the acylium salt
PhCO+SbF6- that was readily identified by the diagnostic
carbonyl stretching band at νCO ) 2225 cm-1 in the IR
spectrum. Even the aliphatic derivative acetyl fluoride
(11) Davlieva, M. G.; Lindeman, S. V.; Neretin, I. S.; Kochi, J. K.
New J. Chem. 2004, 28, 1568.
Indeed, such an acylium salt represents the free carbo-
J. Org. Chem, Vol. 70, No. 10, 2005 4015

Davlieva et al.
short periods and then diluting the mixture with benzene
and quenching with water. After washing, the organic
extract was directly analyzed for comparison with the
Friedel-Crafts acylation product,15 as well as scrutiny
for any other important (side) product.
The behavior of three prototypical arenessdurene
(DUR), pentamethylbenzene (PMB), and hexamethyl-
benzene (HMB)swas first examined with a series of
aromatic and aliphatic acylium derivatives as the hexaflu-
oroantimonate salts; and the facile conversion of durene
and pentamethylbenzene to the corresponding Friedel-
Crafts ketones (Table S2) was found to occur with each
of these acylium salts under very mild reaction condi-
tions,16 e.g.,
FIGURE 5. Acylium salt CH3CO+SbF6 derived from acetyl
-
fluoride and antimony pentafluoride in dichloromethane.
-
afforded only the acetylium salt CH3CO+SbF6 upon
direct treatment with antimony pentafluoride in dichlo-
romethane solution as shown by the ORTEP diagram in
Figure 5.
Gallium trichloride is also an effective Lewis acid for
the direct production of acylium salts from aliphatic acid
chlorides, as shown by the conversion of acetyl chloride
and propionyl chloride to the corresponding acylium salts,
and the recovered methylarene constituted the remainder
of the material balance, except for small but significant
amount of the transalkylated hexamethylbenzene that
was observed with the aliphatic acetylium and tert-
butylacetylium hexafluoroantimonates. The fully substi-
tuted hexamethylbenzene was recovered intact, with no
other aromatic product being detected; and the isosteric
chloropentamethylbenzene was similarly recovered quan-
titatively. It is thus noteworthy that bromopentameth-
ylbenzene yielded significant amounts of the transalky-
lated hexamethylbenzene (24%) as well as 1,2-dibromo-
tetramethylbenzene (12%) in addition to recovered bro-
mopentamethylbenzene.
RCO+GaCl4 (R ) methyl and ethyl), as judged both by
-
X-ray crystallographic analysis11-13 and diagnostic car-
bonyl stretching bands at νCO ) 2284 and 2282 cm-1
,
respectively.14 By way of contrast, however, the aromatic-
acid chloride p-fluorobenzoyl chloride mainly afforded
only crystals of the Lewis acid adduct that is O-bonded
to GaCl3 with νCO ) 1549 cm-1
.
Titanium and zirconium tetrachlorides as the weakest
Lewis acids examined in this study did not afford acylium
salts when treated directly with either aromatic or
aliphatic-acid chlorides, and only Lewis acid complexes
were isolated (Table 1, entries 2-4). Despite the fact that
the treatment of hydrocinnamoyl chloride directly with
titanium chloride afforded the Friedel-Crafts product
even at -40 °C, no acylium salts could be detectedsand
hydrindone was merely isolated as the crystalline dimeric
titanium and zirconium tetrachloride complexes (Table
1, entries 7 and 8). [The structure of the related aceto-
phenone complex is also included in Table 1 (entry 9) for
comparison.]
Similar treatment of methylarene donors was then
carried out with some of the same acylium salts, but as
the hexachloroantimonate analogues, e.g.,
Indeed, the same trends in Friedel-Crafts reactivity
were observed, but with the exception that significant
amounts of chlorinated byproducts were identified in the
form arising from both side chain and nuclear substitu-
tion (Table S3). Since such electrophilic chlorinations are
known to derive via the pure Lewis acid,17 e.g.,
III. Spontaneous Interaction of Acylium Salts
with Aromatic Donors. Controlled exposure of electron-
rich arenes to the cationic electrophile was carried out
by prior dissolution of the pure acylium salt into dichlo-
romethane (or nitromethane mixture) at low tempera-
ture, followed by the careful introduction of the donor.
[The high reactivity of the acylium electrophile as a “free”
cation was ensured by the use of the poorly coordinating
-
-
we infer that the formation of the acylium salt in eq 4 is
counteranions SbF6 and SbCl6 for these studies.] The
colorless solutions were examined in two ways: first with
respect to the aromatic-substitution product and then by
UV-vis analysis of reactive intermediates, as follows.
A. Friedel-Crafts and other aromatic products were
identified by letting the homogeneous solution stand for
(15) For earlier conversions of acylium salts directly into Friedel-
Crafts ketones, see: (a) Olah, G. A.; Kuhn, S. J.; Flood, S. H.; Hardie,
B. A. J. Am. Chem. Soc. 1964, 86, 2203. (b) Olah, G. A.; Lukas, J.;
Lukas, E. J. Am. Chem. Soc. 1969, 91, 5319. (c) Olah, G. A.; Lin, H.
C.; Germain, A. Synthesis 1974, 12, 895. See also Olah et al. in ref 14.
(16) (a) We were limited in our mechanistic objectives (vide infra)
to these difficult (low-temperature) experimental problems, so it is
important to emphasize the caveat that the GC-MS analyses in Table
S2 (and Table S3) were not required to be highly reproducible. The
semiquantitative results are thus primarily intended to convey the
relative trends in arene conversions to Friedel-Crafts ketones, as well
as arene recovery and byproduct formation under low-temperature
conditions. (b) For the synthetic utility of acylium salts, see the
extensive studies of Olah et al. cited in ref 15.
(12) For X-ray diffraction studies at room temperature, see: (a) Boer,
F. P. J. Am. Chem. Soc. 1968, 90, 6706. (b) Le Carpentier, J. M.; Weiss,
R. Acta Crystallogr. 1972, B28, 1421, 1430. (c) Chevrier, B.; Le
Carpentier, J. M.; Weiss, R. Acta Crystallogr. 1972, B28, 2673. (d)
Chevrier, B.; Le Carpentier, J. M.; Weiss, R. J. Am. Chem. Soc. 1972,
94, 5718.
(13) Hurlburt, P. K.; Rack, J. J.; Luck, J. S.; Dec, S. F.; Webb, J. D.;
Anderson, O. P.; Strauss, S. H. J. Am. Chem. Soc. 1994, 116, 10003.
(14) (a) Olah, G. A.; Kuhn, S. J.; Tolgyesi, W. S.; Baker, E. B. J.
Am. Chem. Soc. 1962, 84, 2733. (b) Olah, G. A.; Tolgyesi, W. S.; Kuhn,
S. J.; Moffat, M. E.; Bastien, I. J.; Baker, E. B. J. Am. Chem. Soc. 1963,
85, 1328. See also: Davlieva, M. G. et al. in ref 11.
(17) (a) Kovacic, P.; Sparks, A. K. J. Am. Chem. Soc. 1960, 82, 5740.
(b) Kovacic, P. In Friedel-Crafts and Related Reactions; Olah, G. A.,
Ed.; Interscience: New York, 1965; Vol. IV. (c) Rathore, R.; Loyd, S.
H.; Kochi, J. K. J. Am. Chem. Soc. 1994, 116, 8414. (d) Rathore, R.;
Hecht, J.; Kochi, J. K. J. Am. Chem. Soc. 1998, 120, 13278.
4016 J. Org. Chem., Vol. 70, No. 10, 2005

Reactive Intermediates in Friedel-Crafts Acylations
reversible, i.e.,
especially when in contact with those arene donors with
no readily replaceable aromatic hydrogens. Indeed, the
reduced antimony trichloride that was formed in the
chlorination (eq 8) was isolated from the mixture and
characterized (X-ray) in crystalline form as the π-arene
complex.18 Otherwise, PMB and DUR afforded respect-
able amounts of Friedel-Crafts acylation product when
exposed to the same acylium hexachloroantimonate salts.
Moreover, the presence of significant amounts of both
nuclear and side chain chlorination products indicated
that electrophilic aromatic chlorination and acylation
were more or less competitive, i.e.,
FIGURE 6. Benzenium structure of the product of Friedel-
Crafts alkylation of hexamethylbenzene with FC6H4COCl‚
GaCl3 in dichloromethane.
To evaluate the reactivity of gallium trichloride as a
weak Lewis acid, the [1:1] complex with p-FC6H4COCl
(see Table 1, entry 5) was treated directly with hexam-
ethylbenzene. After the dark red solution was allowed
to stand for 10 days at -70 °C, it deposited only yellow
crystals of the Friedel-Crafts (alkylation) salt derived
from the solvent, i.e.,
and the benzenium structure of this σ-adduct was verified
by X-ray crystallography (Figure 6) together with its
characteristic UV-vis absorption (λmax ) 400 nm, ꢀ ) 6.7
× 103 M-1 cm-1).
FIGURE 7. UV-vis absorption spectra observed at -70 °C
upon the addition of hexamethylbenzene to CH2Cl2 solutions
of acylium hexafluoroantimonates (as indicated).
Thus the full recovery of FC6H4COCl pointed to the
efficient reversion of the [1:1] complex to free gallium
trichloride (but not to the acylium salt), i.e.,
pentamethylbenzene, or durene under precisely the same
conditions employed in the product studies (vide supra),
with the exception that the systems were never allowed
to rise above the lowest practicable temperature of -70
°C.
Dichloromethane solutions of pure acylium hexafluo-
roantimonates are colorless and show no significant
absorption in the 350 to 800-nm spectral region. How-
ever, upon the addition of hexamethylbenzene under
rigorously anhydrous conditions, the CH2Cl2 solutions
turned yellow and pronounced new bands were observed
with λmax ≈ 400 nm (Figure 7).
which was independently confirmed by the isolation of
the same benzenium salt from treatment of hexameth-
ylbenzene in dichloromethane solution with pure gallium
trichloride. Moreover, the analogous reversion of the
acylium salt is also indicated by the isolation of the same
crystalline salt (Figure 6) when hexamethylbenzene was
treated with the aliphatic tert-butylacetyl tetrachloro-
gallate at -70 °C for 12 days. Although no evidence of
Friedel-Crafts acylation could be detected, it is note-
worthy that the treatment of the aliphatic hydrocin-
namoyl chloride with gallium trichloride at -70 °C for 2
days led to colorless crystals of the cyclized hydrindone
complex, i.e.,
The results in Table 2 show that the new absorptions
were essentially invariant with acylium structure, inde-
pendent of whether aliphatic (R ) CH3, (CH3)3CCH2) or
aromatic (4-CH3, 2,4,6-(CH3)3, 4-F, or 2,6-F2) derivatives
were employed. The same insensitivity of λmax was
observed with changes in solvent polarity (CH3COCl,
PrNO2) and counteranion (SbCl6-). It is noteworthy that
no changes in the absorption spectra of acylium cations
were observed upon the addition of the sterically hin-
dered hexaethylbenzene (HEB).19,20 Similarly, only very
weak (indeterminate) bands were detected when either
the structure of which was established by X-ray crystal-
lography (Table 1).
B. Transient (UV-Vis) spectra of acylium electrophiles
were measured in the presence of hexamethylbenzene,
(18) (a) Hubig, S. M.; Lindeman, S. V.; Kochi, J. K. Electronic
submission to the Cambridge Structural Database, 2000, CCDC
138873. (b) Schmidbaur, H.; Nowak, R.; Schier, A.; Wallis, J. M.; Huber,
B.; Mueller, G. Chem. Ber. 1987, 120, 1829.
J. Org. Chem, Vol. 70, No. 10, 2005 4017

Davlieva et al.
TABLE 2. Transient UV-Vis Spectra Attendant upon
the Addition of Various Acylium Cations to Different
Aromatic Donors in Dichloromethane Solution
to elucidate the three critical stages (I-III described
above in the Introduction) for Friedel-Crafts acylation.
Stage I: Activation of Acid Chlorides by Lewis
Acids. The preferential and facile coordination at the
carbonyl-oxygen center of acid chlorides by all of the
different Lewis acids, including antimony pentachloride,
gallium trichloride, titanium, and zirconium tetrachlo-
rides used in this study, is established by the consistent
set of X-ray structures presented in Figures 1-3 and
Table 1, independent of whether monomeric or dimeric
Lewis acid moieties are involved.
absorption, λmax (nm)
acylium
CH3CO+
HMB
HEB
PMB
405a
405b
405c
400d
397
no band
no band
(CH3)3CCH2CO+
4-FC6H4CO+
no band
e
no band
400
no band
no band
2,4,6-(CH3)3C6H2CO+
4-CH3C6H4CO+
2,6-F2C6H3CO+
404
The mechanism of the Lewis acid effect on acid chloride
can be gleaned from the scrutiny of the small structural
changes that is allowed by the X-ray structural param-
eters in Table 1 obtained at low (-150 °C) temperatures.24a
Thus upon Lewis acid coordination, all the acid chlorides
suffer significant elongation of the CdO bond from 1.18
to 1.22 Å (av) and shortening of the C-Cl bond from 1.80
to 1.72 Å, together with a slight contraction of the C-C
bond to the aliphatic or aromatic group.24b Such trends
when evaluated on the basis of Pauling’s bond-length/
bond-order relationship25 indicate the increased electron
deficiency that is induced at the carbonyl center by the
Lewis acid coordination is compensated by π-electron
donation from chlorine, resulting in the strengthening of
the C-Cl bond (as well as increased conjugation of the
alkyl/aryl group). This important step leading to the
extensive electron polarization of the acid chloride must
either be separate from or lie prior to the activation of
the C-Cl bond. However, the singular absence of any
crystallographic evidence for the formation of structures
involving Lewis acid coordination to the chlorine center
indicates that the concentration of such an isomeric
adduct is not likely to be mechanistically significant, and
thus relegated to a weak intermediate or a transition-
state structure. This dilemma can be circumvented if the
predominant acyl-chloride complex undergoes a facile
intramolecular (direct) chlorine transfer to the complexed
Lewis acid moiety, e.g.,
400
395
a In dichloromethane at -70 °C, with acylium carbocations
taken as hexafluoroantimonate salts, unless otherwise noted. b In
-
AcCl solution. c In PrNO2 solution. d CH3CO+SbCl6 salt. e Very
weak band appeared at ∼410 nm.
pentamethylbenzene or durene was added under the
same (low-temperature) conditions.21 Such a spectral
behavior accompanying the exposure of different acyl
cations to hexamethylbenzene is strongly reminiscent of
the formation of σ-adducts as benzenium cations with
other cationic electrophiles such as E+ ) H+, CH3+, NO2
,
+
and Br+ summarized recently,22 i.e.,
Although this characteristic absorption band experiences
a noticeable red shift with strong electrophiles such as
E+ ) NO2 and Br+, it will be rather invariant among
+
the HMB adducts of the family of carbon-centered cations
such as E+ ) CH3+, ClCH2+, CH3C+dO, and PhC+dO,
which are unlikely to exert appreciably different pertur-
bations of the cyclohexadienyl chromophore (especially
at its nodal position) owing to their essential carbo-
cationic character.23
Discussion
The successful isolation of reactive intermediates in
crystalline form for X-ray analysis must be reconciled
with the mechanism of the prototypical electrophilic
substitution of aromatic donors. As such, let us discuss
how the availability of (static) X-ray structures coupled
with transient (UV-vis) spectral analysis can be used
Indeed, this concerted rearrangement is consistent with
the competition between the formation of acylium cation
and the [1:1] complex in eq 5, as shown by its sensitivity
to the acid-chloride structure. For example, the minor
replacement of benzoyl chloride with the slightly more
electron-rich methyl-substituted analogue is sufficient to
lead from the isolation of the crystalline [1:1] complex of
(19) (a) Iverson, D. J.; Hunter, G.; Blount, J. F.; Damewood, J. R.,
Jr.; Mislow, K. J. Am. Chem. Soc. 1981, 103, 6073. (b) Hunter, G.;
Iverson, D. J.; Mislow, K.; Blount, J. F. J. Am. Chem. Soc. 1980, 102,
5942. (c) Hunter, G.; Weakley, T. J. R.; Mislow, K.; Wong, M. G. J.
Chem. Soc., Dalton Trans. 1986, 577.
(24) (a) The precision/accuracy of the earlier X-ray diffraction data
collected at room temperature10,12 was limited to 1-1.5 pm (largely
owing to thermal motion) in contrast to 0.3-0.5 pm from our low-
temperature data.11 (b) The bond lengths for CdO and C-Cl of 1.181
and 1.804 Å that were determined in hydrocinnamoyl chloride at -150
°C generally fall in the range (1.16-1.21 and 1.75-1.83 Å) of those in
other acid chlorides typically reported by: Leser, J.; Rabinovich, D.
Acta Crystallogr. 1978, B34, 2253, 2257, 2260, 2264. Fukushima, S.;
Ito, Y.; Hosomi, H.; Ohba, S. Acta Crystallogr. 1998, B54, 895. Sandor,
R. B.; Foxman, B. M. Tetrahedron 2000, 56, 6805.
(20) Rathore, R.; Lindeman, S. V.; Kochi, J. K. J. Am. Chem. Soc.
1997, 119, 9393.
(21) (a) The corresponding arenium cations with E+ ) H+ also have
similar absorptions in this region but are generally blue-shifted by
about 5-10 nm. (b) Mamatyuk, V. I.; Rezvukhin, A. I.; Detsina, A. V.;
Buraev, V. I.; Isaev, I. S.; Koptyug, V. A. Zh. Org. Khim. 1973, 9, 2429.
(c) Koptyug, V. A. Top. Curr. Chem. 1984, 122, 1.
(22) Hubig, S. M.; Kochi, J. K. J. Org. Chem. 2000, 65, 6807.
(23) (a) The relative acceptor properties of the carbocations can be
judged by differences in their reduction potentials.23d See: (b) Wayner,
D. D. M.; McPhee, D. J.; Griller, D. J. Am. Chem. Soc. 1988, 110, 132.
(c) Lund, T.; Wayner, D. D. M.; Jonsson, M.; Larsen, A. G.; Daasbjerg,
K. J. Am. Chem. Soc. 2001, 123, 12590 and references therein. (d)
See: Mulliken, R. S.; Person, W. B. Molecular Complexes; Wiley-
Interscience: New York, 1969.
(25) (a) Pauling, L. Nature of the Chemical Bond; Cornell University
Press: Ithaca, NY, 1960; p 239. (b) According to Pauling, the bond
order (n) is related to the bond distance ri by the following relation-
ship: ni ) exp[2.303(r1 - ri)/K], where r1 is the single-bond distance,
ri is the observed distance, and K is an empirical constant. For the
carbonyl bond, r1 ) 1.43 Å and K ) 0.75. For the bond to alkyl, r1
)
1.48 Å and Κ ) 0.50, which includes the correction for the change in
covalent radius with hybridization.
4018 J. Org. Chem., Vol. 70, No. 10, 2005

Reactive Intermediates in Friedel-Crafts Acylations
respectively. Such bond-length variations reflect in-
creased electron donation from the aliphatic and aromatic
moieties to compensate for the electron-deficient (car-
bocationic) center with sp1-hybridized linear configura-
tion for the CR-C-O angle close to â ) 180°. Indeed,
aromatic acyl cations show π-π conjugation between the
carbocationic center and the attached benzenoid ring,
which increases progressively with aryl donicity so that
the shortening of the single (conjugated) C-C bond of
FIGURE 8. ORTEP diagrams of (A) propionyl and (B)
p-fluorobenzoyl cations showing linear bindings of CO with â
≈ 180°.
rCR down to 1.37 Å matches the corresponding elongation
of the carbonyl triple bond with rCO of up to 1.125 Å [and
tracks the decreasing IR stretching frequency (νCO)].
Quantitative analysis of the bond-length changes with
the aid of the Pauling bond-length/bond-order relation-
ship25 leads to the conclusion that the total bond order
at the carbonyl-carbon site is rather invariant with
substituents in the aromatic ring.11 Such an unexpected
result predicts the intrinsic electrophilic reactivities of
benzoyl cations to be largely unaffected by nuclear
substituents.
By contrast, structural analysis of aliphatic acyl cations
indicates that σ-π hyperconjugation in the acetyl cation28
can contribute only about 60% and results in the rather
short value of rCR ∼ 1.42 Å for this bond type. The
effectiveness of σ-π hyperconjugation is reduced in the
isobutyryl (R ) isopropyl) cation owing to the availability
of only a single CR-H bond. Significantly, the bond-length
changes are insufficient to complete the bond-order
requirement at the carbonyl-carbon in the isobutyryl
cation, resulting in its significant electron deficiency. As
such, the electrophilic reactivities of the aliphatic acyl
cations are expected to increase in the following order:
R ) CH3 < CH3CH2 < (CH3)2CH.
benzoyl chloride in eq 1 as opposed to the isolation of the
acylium salt of the toluoylium cation in eq 4, when
antimony pentachloride is simply added to the dichlo-
romethane solution of benzoyl chloride and p-toluoyl
chloride, respectively, under otherwise identical reaction
conditions.
The conceptual interchangeability between the Lewis
acid complex and the acylium salt (eq 15) demands that
(i) the formation of the acylium salt involving the
unimolecular rearrangement of the [1:1] complex and
then (ii) the sequestration of the acylium cation via the
microscopic-reverse process involving chlorine transfer
from the SbCl6- counterion proceed via the same transi-
tion state, i.e.,
Otherwise, crystallographic studies would provide no
further insight into the mechanism of the Lewis acid
activation of the acyl chloride.26
Stage III: Electrophilic Addition of Acyl Cations
onto Aromatic Donors. Product studies establish the
facile interaction between electron-rich arene donors and
acyl cations under rather mild conditions to effect
Friedel-Crafts acylation (Tables S2 and S3) at rather
low temperatures. However, it is the competitive side
reactions that provide information as to the reactive
intermediates involved. Thus the transalkylation prod-
ucts hexamethylbenzene from pentamethylbenzene and
the trans-chlorination products chloropentamethylben-
zene and R-chlorohexamethylbenzene from both hexa-
methylbenzene and pentamethylbenzene as well as chlo-
ropentaethylbenzene from pentaethylbenzene and R-
chlorodurene from durene suggest that benzenium ad-
ducts are involved as Wheland intermediates (consider,
for example, eq 11). Moreover, the transient (UV-vis)
spectra of the benzenium adducts of acyl cations observed
in Figure 7 confirm the presence of such electrophilic
adducts when hexamethylbenzene is the arene donor.
Likewise, the absence of such absorption bands when
either pentamethylbenzene or durene are exposed to acyl
cations under the same conditions (Table 2, columns 3
and 4) is consistent with rapid proton loss from the
Wheland intermediate,29 e.g.,
Stage II: Structure/Reactivity of Acylium Elec-
trophiles. Aliphatic and aromatic acyl cations (RCO+
and ArCO+) independently synthesized from the corre-
sponding carboxylic-acid halides and Lewis acids (or an
alternative procedure with silver (I) salts3) can be struc-
turally characterized as free carbocations by isolating
them as crystalline salts with low-nucleophilic counter-
anions such as SbF6-, SbCl6-, or GaCl4-. X-ray analyses
at low temperatures (-150 °C) afford crystallographic
parameters sufficient to detect subtle changes in acyl
structures.11,24 Thus the series of aliphatic [with R ) CH3,
CH3CH2, and (CH3)2CH] and X-substituted benzoyl [with
X ) 4-CH3, 2,4,6-(CH3)3, 4-CH3O, and 4-F] structures are
characterized by carbonyl groups that are linearly bonded12
(Figure 8) and show significant triple-bond character as
follows from C-O bond lengths of rCO ∼ 1.10Å (av)27 and
enhanced carbonyl stretching frequences of ∼2300 cm-1
(av).11
Furthermore, the adjacent single bonds attached to the
carbonyl groups in both aromatic and aliphatic acyl
cations are also shortened to rCR ∼ 1.37 and 1.42 Å,
(26) (a) It should be noted, however, that Friedel-Crafts acylation
is often carried out in the presence of excess Lewis acid;3,5 and further
activation of the acylium cation by “superelectrophilic solvation” as
described by Olah and co-workers26b is possible, particularly with
regard to solvation of the oxygen nonbonded electron pairs to induce
acylium bending (vide infra). Unfortunately, the X-ray results do not
address this possibility. (b) Olah, G. A.; Klumpp, D. A. Acc. Chem. Res.
2004, 37, 211. Olah, G. A.; Laali, K. A.; Wang, Q.; Prakash, G. K. S.
Onium Ions; Wiley: New York, 1998.
(28) Originally proposed by Boer12a but based on an erroneously
short value of rCO ) 1.385 ( 0.016 Å. Our redetermination11 confirms
that this bond is still exceedingly short (1.419 ( 0.004 Å) as compared
to the standard C(sp3)-C(sp) bond length of 1.47 Å (Allen, F. H.;
Kennard, O.; Watson, D. G.; Brammer, L.; Orpen, A. G.; Taylor, R. J.
Chem. Soc., Perkin Trans. 2 1987, S1)
(27) Relative to the bond length of 1.07 Å estimated for a C≡O triple
bond as described by Davlieva et al. in ref 11 (footnote 18).
(29) Olah, G. A. Acc. Chem. Res. 1971, 4, 240.
J. Org. Chem, Vol. 70, No. 10, 2005 4019

Davlieva et al.
three critical stages in Friedel-Crafts acylation of arene
donors. First, the spontaneous association of the acid
chloride and Lewis acid occurs at the oxygen center of
the carbonyl moiety, and analysis of the X-ray param-
eters of the [1:1] complex shows the elongation of the
carbonyl bond accompanied by the tightening of the
(single) bond to chlorine. Although the latter may seem
to represent passivation of the acyl chloride, facile
(concerted) rearrangement of the Lewis acid complex by
the intramolecular delivery of chlorine leads directly to
the critical acylium electrophile, as depicted in eq 16.
Second, its isolation and X-ray crystallography establish
the linear binding of the cationic carbonyl (-C+dO)
center to both alkyl (R) and aryl (Ar) groups. Detailed
analyses of the X-ray parameters point to the electro-
philic reactivity of such linear aliphatic acylium carbo-
cations that increases in the order R ) CH3 < CH3CH2
< (CH3)2CH, whereas it is predicted to be relatively
insensitive to substituent changes in Ar. Third, the
enhanced electrophilic reactivity of both aliphatic and
aromatic acylium carbocations is confirmed by Friedel-
Crafts acylations of methylarenes that occur readily even
at very low temperatures. Although the short-lived
Wheland intermediates could not be isolated in crystal-
line form despite these drastic conditions, transient
electronic (UV-vis) spectra of cationic acyl adducts can
be observed with the strong hexamethylbenzene donor,
and structurally related to other benzenium adducts
established in previous studies.35
Our failure to obtain the X-ray structures of any Wheland
intermediates under the reaction conditions of Friedel-
Crafts acylation indicates their lifetimes are too short to
isolate in crystalline form relative to the free acyl cation
and even such electron-rich and polyalkylated arene
donors as hexamethylbenzene. However, if this electro-
philic addition/reversion process is considered in the
context of the acylium structures illustrated in Figure
8, the direct addition of the linear structure requires its
simultaneous bending in order to achieve the structure
of the acyl adduct in the benzenium state. As such, the
reorganization energy required for the simultaneous
bending of ∆â ≈ 60° will exact a sizable penalty on the
activation barrier. Indeed, we faced a similar mechanistic
conundrum in the very rapid addition of the nitronium
(NO2+) cation during electrophilic aromatic nitration
involving an analogous bending from linear to highly bent
(NO2) structures.30 In this case, theoretical (molecular-
orbital) studies predicted the existence of a contiguous
pair of π/σ structures, i.e.,
in which the bending penalty is separated from the
activation barrier for conversion to the σ-adduct (Whe-
land intermediate).31,32 Due cognizance of the same
reorganization problem in Friedel-Crafts acylation then
leads to a similar 2-step process, i.e.,
Experimental Section
Materials. Acetyl chloride, acetyl fluoride, propionyl chlo-
ride, tert-butylacetyl chloride, benzoyl chloride, p-toluoyl chlo-
ride, p-anisoyl chloride, p-fluorobenzoyl chloride, hydrocin-
namoyl chloride, 2,4,6-mesitoyl chloride, and pentamethyl-
benzoyl chloride were prepared from the corresponding car-
boxylic acids by treatment with oxalyl chloride; the Lewis acids
were used as received: SbCl5, SbF5, GaCl3, TiCl4, and ZrCl4.
Solvents were purified according to published procedures.36
Hexamethylbenzene was additionally sublimed, but penta-
methylbenzene and durene were used without additional
purification. Chloropentamethylbenzene was prepared as de-
scribed by Aitken et al.37
Syntheses of [1:1] Crystalline Addition Complexes. All
the Lewis acid adducts were extremely moisture sensitive, and
necessitated the synthetic preparations and physical measure-
ment to be carried out with the rigorous exclusion of air and
water. Schlenk flasks and an inert atmosphere glovebox with
purified argon and high-vacuum techniques were routinely
employed. In every case, the acyl halides were freshly distilled
prior to use. Dichloromethane and hexane were additionally
passed through an activated molecular sieves column prior to
use. Benzoyl chloride:antimony pentachloride com-
plexes: Freshly distilled benzoyl chloride (1.4 g, 10 mmol) was
dissolved in 10 mL of dry CH2Cl2 under an argon atmosphere,
and the solution was cooled to -10 °C. Under careful stirring,
a solution of 2.99 g of SbCl5 (10 mmol) in 10 mL of dry
dichloromethane was added. After 15 min of continued stirring,
the white, crystalline precipitate was collected, washed with
cold CH2Cl2, and dried under vacuum. Single crystals suitable
for X-ray analysis were obtained by slow cooling of the dry
CH2Cl2 solution initially saturated at 40 °C. p-Fluorobenzoyl
chloride:antimony pentachloride complexes: Antimony
Preliminary DFT calculations support π-acyl/arene struc-
tures as a stationary point along the potential-energy
surface but require more extensive validation.33 And from
a more related structural perspective, the X-ray struc-
tures of π-arene complexes with planar carbocations are
established;34,35 and they have been directly related to a
wide variety of π-arene complexes held together by
Mulliken charge-transfer forces.9,22
Summary and Conclusions
Isolation and X-ray crystallography of some of the
important reactive intermediates provide insight into the
(30) (a) Schofield, K. Aromatic nitration; Cambridge University
Press: Cambridge, UK, 1980. (b) Olah, G. A.; Malhotra, R.; Narang,
S. C. Nitration; VCH: New York, 1989.
(31) Gwaltney, S. R.; Rosokha, S. V.; Head-Gordon, M.; Kochi, J. K.
J. Am. Chem. Soc. 2003, 125, 3273.
(32) Esteves, P. M.; de Carneiro, J. W.; Cardoso, S. P.; Barbosa, A.
G. H.; Laali, K. K.; Rasul, G.; Prakash, G. K. S.; Olah, G. A. J. Am.
Chem. Soc. 2003, 125, 4836.
(33) (a) Gwaltney, S. R. Unpublished results. (b) The incursion of
such a cationic π-complex (or benzonium ion) was proposed as a
precursor step in the general mechanism of electrophilic aromatic
substitution by Olah29 following Mulliken’s theoretical classification
of intermolecular interactions. (c) Mulliken, R. S. J. Am. Chem. Soc.
1950, 72, 600. Mulliken, R. S. J. Am. Chem. Soc. 1952, 74, 811.
(34) Takahashi, Y.; Sankararaman, S.; Kochi, J. K. J. Am. Chem.
Soc. 1989, 111, 2954.
(36) Perrin, D. D.; Armarego, W. L. F.; Perrin, D. R. In Purification
of Laboratory Chemicals, 2nd ed.; Pergamon: New York, 1980.
(37) Aitken, R. R.; Badger, G. M.; Cook, J. W. J. Chem. Soc. 1950,
331.
(35) Hubig, S. M.; Lindeman, S. V.; Kochi, J. K. Coord. Chem. Rev.
2000, 200, 831.
4020 J. Org. Chem., Vol. 70, No. 10, 2005

Reactive Intermediates in Friedel-Crafts Acylations
pentachloride (1.2 g, 4 mmol) was dissolved in 15 mL of dry
dichloromethane under an argon atmosphere, and the solution
was cooled to -40 °C. Into the stirred solution was introduced
p-fluorobenzoyl chloride (0.58 g, 4 mmol) in 5 mL of dichlo-
romethane. After 15 min of continued stirring, the white
crystalline precipitate was collected, washed with cold CH2-
Cl2, and dried in vacuo. Aromatic acid-halide (benzoyl-,
p-fluorobenzoyl chloride):titanium(IV) chloride com-
plexes: Titanium(IV) chloride (1.89 g, 10 mmol) was dissolved
in 10 mL of dry CH2Cl2 under an argon atmosphere. Under
careful stirring, 5 mL of dichloromethane solution containing
1.4 g (1.44 g, 10 mmol) of freshly distilled benzoyl (p-
fluorobenzoyl) chloride were introduced. After 15 min of
continued stirring the solution was covered with a layer of
n-hexane (15 mL), and after 3-5 days standing at -40 °C
colorless crystals suitable for X-ray analyses were collected
from the trap wall near the phase borderline. Hydrocin-
namoyl chloride:titanium(IV) chloride complex: Tita-
nium(IV) chloride (1.89 g, 10 mmol) was dissolved in 10 mL
of dry CH2Cl2 under an argon atmosphere. Under careful
stirring, 5 mL of dichloromethane solution containing 1.68 g
(10 mmol) of hydrocinnamoyl chloride was added. Stirring was
continued for 15 min. Crystals suitable for X-ray study were
obtained by slow evaporation of the dichloromethane solution
(12 h) under an argon atmosphere. tert-Butylacetyl chlo-
ride:zirconium(IV) chloride complex: Zirconium(IV) chlo-
ride (2.33 g, 10 mmol) was suspended in 5 mL of dry
dichloromethane under an argon atmosphere. Under careful
stirring, a solution of 1.34 g (10 mmol) of tert-butylacetyl
chloride in 10 mL of dry dichloromethane was added. A yellow
suspension was obtained, which turned to a dark yellow clear
solution after 30 min of continued stirring. Single crystals
suitable for X-ray analysis were obtained by slow evaporation
under an argon atmosphere over a period of 6 h. Hydrocin-
namoyl chloride:zirconium(IV) chloride complex: Zirco-
nium(IV) chloride (2.33 g, 10 mmol) was suspended in 5 mL
of dry dichloromethane under an argon atmosphere. Under
careful stirring, a solution of 1.68 g (10 mmol) of hydrocin-
namoyl chloride in 10 mL of dry dichloromethane was added.
After 30 min of continued stirring a yellow suspension was
obtained. The solution was covered with a layer of dry
n-hexane (15 mL) and allowed to stand at -70 °C. Well-formed
crystals of the complex were collected at the solvent interface
after 12 days. p-Fluorobenzoyl chloride:trichlorogallate
complex: p-Fluorobenzoyl chloride (1.44 g, 10 mmol) was
dissolved in 10 mL of dry CH2Cl2 under an argon atmosphere.
Under careful stirring, 5 mL of dichloromethane solution
containing 1.76 g (10 mmol) of GaCl3 was added. Stirring was
continued for 15 min. Crystals suitable for X-ray study were
obtained by slow evaporation of the dichloromethane solution
(2 h) under an argon atmosphere.
Electronic (UV-Vis) Spectra of Acylium Adducts to
Aromatic Donors. A spectroscopic study were carried out on
a HP 8453 diode-array or Cary 5 spectrophotometer in a Dewar
equipped with quartz lens, and the temperature was adjusted
with an ethanol-liquid nitrogen bath. Measurements were
carried out in the quartz (0.1 to 1.0 cm path length) spectro-
scopic cells fitted with a sidearm and equipped with a Teflon
valve with Viton O-rings. The sample of the acylium salt
(typically 3-5 mg, as the crystalline salt with hexafluoroan-
timonate or hexachloroantimonate anion) was placed into a
spectroscopic cell and the sample of aromatic donor (e.g.,
hexamethylbenzene, typically 10 mg) was placed in the
sidearm. Dichloromethane (3-4 mL) was added to dissolve the
acylium salt within the cell. [All these operations were carried
out under argon atmosphere in the glovebox to avoid any
contact of acylium salt with air or moisture and carefully dried
solvent was used.] The tightly closed cell was transferred from
the glovebox and placed in Dewar cooled to -70 °C, and
absorption spectrum was measured. After measurement of the
spectrum of pure acyl cation, the solution was transferred to
the sidearm of a UV-vis cell to dissolve aromatic donor (at
low temperature), and then back to the cell, and the spectrum
of the solution containing both acylium cation and aromatic
donor was measured immediately. Then the solution were
warmed and cooled back to -70 to -80 °C, and spectra were
measured at room temperature and low-temperature again.
The solutions of the pure oxocarbonium cations are colorless
and do not show any noticeable absorption bands in the 350-
800 nm range of electronic spectra. However, these solutions
turned yellow upon addition of the hexamethylbenzene and
an absorption band appeared at λmax ≈ 400 nm (see Figure 7).
X-ray Crystallography. The diffraction data for the crys-
talline compounds were collected at -150 °C on a diffracto-
meter with a CCD detector using Mo KR radiation (λ ) 0.71073
Å). In all cases, a semiempirical (SADABS) absorption correc-
tion was applied.38 The structures were solved by direct
methods and refined by a full-matrix least-squares procedure
with the SHELXTL suite of programs.39 The structural data
on the compounds listed in Table 1 are on deposit (as CCDC
261514-261523) and can be obtained from the Cambridge
Crystallographic Data Center, U.K.
Acknowledgment. We thank S. V. Rosokha for
technical assistance with the spectral measurements,
V. Zaitsev for some of the product studies and syntheses,
and the R. A. Welch Foundation and National Science
Foundation for financial support.
Supporting Information Available: X-ray structural
parameters of aliphatic and aromatic acylium salts (Table S1);
product analysis from Friedel-Crafts acylation carried out
with acylium hexafluoroantimonates (Table S2); product analy-
sis from Friedel-Crafts acylation carried out with acylium
hexachloroantimonates (Table S3) and typical examples 1-4.
This material is available free of charge via the Internet at
http://pubs.acs.org.
Reaction of Acylium Salts with Aromatic Compounds.
The reactions were carried out in dry dichloromethane solution
in which the aromatic donors and the acylium salts are both
soluble at the low temperatures used. The appropriate acylium
salt (0.05 mmol) in 1 mL of dry dichloromethane solution was
cooled and added with stirring into 0.05 mmol of the aromatic
donor. After continuous stirring, the solvent was removed, and
2 mL of benzene was added. The resulting mixture was washed
with water, and the organic phase was separated, dried over
MgSO4, and analyzed by GC/MS techniques (Tables S2 and
S3, Supporting Information).
JO0501588
(38) G. M. Sheldrick, SADABS (Ver. 2.03); Bruker/Siemens Area
Detector Absorption and Other Corrections, 2000.
(39) G. M. Sheldrick, SHELXTL, v. 6.12, Program Library for
Structure Solution and Molecular Graphics; Bruker Analytical X-ray
Systems: Madison, WI, 2000.
J. Org. Chem, Vol. 70, No. 10, 2005 4021