Changing the ortho/para ratio in aromatic acylation reactions by changing reaction conditions: A mechanistic explanation from kinetic measurements

Article abstract of DOI:10.1021/ja0022066

Kinetic measurements of the acylation of toluene (2a) and p-xylene (2b), side-chain deuterated toluene (2a-d₃), as well as perdeuterated toluene (2a-d₈) and p-xylene (2b-d₁₀) with the aroyl triflate 1 in 1,2-dichloroethane

Full text of DOI:10.1021/ja0022066

J. Am. Chem. Soc. 2001, 123, 3429-3433
3429
Changing the Ortho/Para Ratio in Aromatic Acylation Reactions by
Changing Reaction Conditions: A Mechanistic Explanation from
Kinetic Measurements1
Franz Effenberger* and Andreas H. Maier2
Contribution from the Institut fu¨r Organische Chemie, UniVersita¨t Stuttgart, Pfaffenwaldring 55,
D-70569 Stuttgart, Germany
ReceiVed June 19, 2000. ReVised Manuscript ReceiVed NoVember 9, 2000
Abstract: Kinetic measurements of the acylation of toluene (2a) and p-xylene (2b), side-chain deuterated
toluene (2a-d3), as well as perdeuterated toluene (2a-d8) and p-xylene (2b-d10) with the aroyl triflate 1 in
1,2-dichloroethane reveal a strong dependence of the isotope effect on reaction conditions. In the presence of
trifluoromethanesulfonic acid (HOTf), the second-order rate constants kH/kD observed are in the order of 1.75-
1.94, whereas in the presence of 2,4,6-tri-tert-butylpyridine (4) rate constants kH/kD of 1.14-1.25 are found.
The primary kinetic isotope effects observed correlate with the ortho/para ratio of the acylation of toluene. In
the presence of 4 a relatively high percentage (∼30%) of ortho product is obtained, whereas under acidic
conditions the ratio is only 10%. The correlation between isotope effects and isomer distributions is obviously
due to the rate of deprotonation of the corresponding σ-complex intermediates. Assuming a bent structure for
σ-complexes, the conformation giving deprotonation is preferred in the para σ-complex in comparison with
ortho complex.
Introduction
Scheme 1
In their first fundamental mechanistic investigations of
Friedel-Crafts acylations Olah et al. already supposed that the
ortho/para isomer ratios might depend on the rate of deproto-
nation of the Wheland intermediate.3 Primary kinetic isotope
effects kH/kD in the order of 1.5 to 3.253,4a indicate that the
σ-complex intermediates are relatively stable, and thus their
follow-up reactions can be influenced specifically. These
mechanistic studies were performed in organic solvents with
acylium salts as acylating agents.3 Since acylations of aromatics
normally are carried out either with an excess of Friedel-Crafts
catalyst or in superacidic systems, the deprotonation of the
intermediate σ-complex has not yet been studied kinetically.
Investigations of the deprotonation step to change the ortho/
para ratio specifically by changing reaction conditions have also
not yet been carried out.
In a previous publication1 we have reported on the mechanism
of the acylation of aromatics with carboxylic trifluoromethane-
sulfonic anhydrides (acyltriflates). The extraordinary reactivity
of acyltriflates, also in the absence of Friedel-Crafts catalysts
or strong Brønsted acids, makes it possible to perform acylations
of aromatics even in the presence of a base. Both steps of
aromatic acylations with acyltriflates, the formation of σ-com-
plexes as well as their deprotonation, can therefore be studied.
Scheme 1 represents the model reaction we have investigated
kinetically.1 The acylation of aromatics with acyltriflates was
followed kinetically both in the presence of 2,4,6-tri-tert-
butylpyridine (4) as well as with an excess of trifluoromethane-
sulfonic acid (HOTf).1 Due to steric reasons, the pyridine
derivative 4 does not react with aroylium ions formed by
dissociation of aroyltriflates.1 The rate constants of aroylations
determined in the presence of 4 as well as in the presence of
HOTf refer directly to the corresponding concentrations of
acylium ion 1′.1
(1) Electrophilic Aromatic Substitution. Part 41: for part 40 see:
Effenberger, F.; Eberhard, J. K.; Maier, A. H. J. Am. Chem. Soc. 1996,
118, 12572-12579.
(2) Maier, A. H. Dissertation, Universita¨t Stuttgart, 1996.
(3) (a) Olah, G. A.; Kuhn, S. J.; Flood, S. H.; Hardie, B. A. J. Am. Chem.
Soc. 1964, 86, 2203-2209. (b) Olah, G. A.; Lukas, J.; Lukas, E. J. Am.
Chem. Soc. 1969, 91, 5319-5323. (c) Olah, G. A.; Kobayashi, S. J. Am.
Chem. Soc. 1971, 93, 6964-6967.
Acylations under basic conditions proceeded 5 to 10 times
faster than the comparable reactions in the presence of 150 mol
% HOTf (Scheme 1). This result indicates that the deprotonation
(4) (a) Vanhaverbeke, D. Meded. Vlaam. Chem. Ver. 1967, 29, 110-
115; Chem. Abstr. 1968, 68, 86460p. (b) Dowdy, D.; Gore, P. H.; Waters,
D. N. J. Chem. Soc., Perkins Trans. 2 1991, 1149-1159.
10.1021/ja0022066 CCC: $20.00 © 2001 American Chemical Society
Published on Web 03/24/2001

3430 J. Am. Chem. Soc., Vol. 123, No. 15, 2001
Effenberger and Maier
Scheme 2
step contributes to the rate of the overall reaction, and thus
confirms earlier findings3,4 of a primary kinetic isotope effect
in aromatic acylations.
Since variation of reaction conditions, i.e., working in acidic
or basic media, causes also a change of the ortho/para isomer
ratios,1 we have now investigated the rate of the deprotonation
step of the intermediate σ-complexes by kinetic measurements.
Results and Discussion
For a kinetic treatment, the model reaction shown in Scheme
1 is summarized schematically in Scheme 2. The dissociation
step (1 a 1′) is to be considered not rate-determining.1 For k1
, k-1, k2, and k-2 ) 0, which is realistic for electrophilic
aromatic acylations, the general rate law given in eq 1 can be
deduced.5
Figure 1. Reaction profile (a) and determination of rate constants (b)
of the reaction of toluene (2a) and deuterated toluene (2a-d8) with
acylium ion 1′ in DCE at 25 °C in the presence of HOTf.
k2
V ) k1[ArH][1′]
(1)
k-1 + k2
with k2 ) kb[4] or k2 ) ka. From this it follows that if k-1 ≈ k2
2
2
or k-1 . k2 (Scheme 2), the overall reaction rate is influenced
by the deprotonation step. In case of k-1 . k2 there will be a
linear dependence on the deprotonation rate, otherwise the
deprotonation rate affects partially the overall reaction rate. The
magnitude of a primary kinetic isotope effect, as experimental
proof for the influence of the deprotonation step, depends on
the rate of the back reaction (k-1) and the deprotonation (k2) of
the σ-complex.5 With exception of the azo coupling which is
performed in aqueous medium,6 the validity of eq 1 has not
been proved so far.7
Isotope Effects in the Acylation of Toluene and p-Xylene
in the Presence of HOTf and of Base, Respectively. In
general, primary kinetic isotope effects are obtained by deter-
mining the rate constants of reactions of the deuterated and the
corresponding not deuterated compounds.8 This is possible in
the presence of HOTf, where the concentration of acylium ion
1′ corresponds to the concentration of 1 (Scheme 1). In these
cases we were able to calculate kaobs via the integrated second-
Figure 2. Reaction profile and derived initial velocities (Vstart) of the
reaction of acylium ion 1′ with toluene (2a) and deuterated toluene
(2a-d8) in DCE at 25 °C in the presence of base 4.
order rate law9 based on gas chromatographically determined
concentration time data as shown in Figure 1b,a. To check for
side reactions we plotted the concentration of educts ([1 + 1′])
together with the concentration of educts calculated from
determined product concentration ([1 + 1′] ) [1 + 1′]0 - [3]).
In the reaction with deuterated aromatics in the presence of base,
the concentration of acylium ion 1′ cannot be followed easily.
IR spectroscopy failed in the presence of deuterated aromatics
due to overlapping of the acylium ion band at 2188 cm-1 with
(5) Taylor, R. Electrophilic Aromatic Substitution; Wiley: Chichester,
1990; pp 32-34.
(6) (a) Zollinger, H. AdV. Phys. Org. Chem. 1964, 2, 163-200. (b) Willi,
A. V. Isotopeneffekte bei chemischen Reaktionen; Thieme: Stuttgart, 1983;
pp 74-76.
(7) Willi, A. V. Isotopeneffekte bei chemischen Reaktionen; Thieme:
Stuttgart, 1983; pp 139-142.
(8) Saunders, W. H. In Techniques of Chemistry, 4th ed.; Bernasconi,
C. F., Ed.; Wiley: New York, 1986; Vol. VI, Part 1, pp 565-611.
(9) Bunnett, J. F. In Techniques of Chemistry, 4th ed.; Bernasconi, C.
F., Ed.; Wiley: New York, 1986; Vol. VI, Part 1, pp 183-185.

Ortho/Para Ratio in Aromatic Acylation Reactions
J. Am. Chem. Soc., Vol. 123, No. 15, 2001 3431
Table 1. Isotope Effects kH/kD on the Second-Order Rate Constant kobs in the Acylation of 2a,b with Acylium Ion 1′ in the Presence of HOTf
and 4, Respectively, at 25 ( 0.2 °C in 1,2-Dichloroethane
kH/kD obs
[HOTf]
(M)
[4]
(M)
kaobs(2,2-d)
vstart(2,2-d)
from comparative
reactions
from competitive
reactions
aromatics 2
(M)
[1] (M)
(10-4 M-1 s-1
)
(10-3 M h-1
)
2a, 2a-d8
2b, 2b-d10
2a, 2a-d3
2a, 2a-d8
2b, 2b-d10
0.374
0.325
0.374
0.374
0.340
0.076
0.062
0.086
0.093
0.124
0.141
0.105
0.129
2.91 ( 0.07, 1.50 ( 0.04
4.66 ( 0.5, 2.63 ( 0.5
3.36 ( 0.06, 3.06 ( 0.09
1.94 ( 0.10
1.77 ( 0.05
1.10 ( 0.05
1.19 ( 0.07
1.25 ( 0.08
1.85 ( 0.20
1.75 ( 0.17
0.150
0.200
2.32, 1.95
8.22, 6.64
1.19 ( 0.05
1.14 ( 0.06
Table 2. Distribution of Isomeric Ketones 3a-c in the Acylation
of Toluene (2a) with Anhydride 1 in the Presence of HOTf and 4,
Respectively, at 25 ( 0.2 °C in 1,2-Dichloroethane
reaction. Under the applied reaction conditions a transformation
of the ortho into the para product could not be detected.
Interpretation of the Experimental Results. From eq 1,
written in the form of eq 2, the following deductions for the
ortho/para ratio in the presence of base and acid, respectively,
can be drawn.
isomer ratios (%)
(103 M) (103 M) (103 M) ortho (3a) meta (3b) para (3c)
[1]
[HOTf]
[4]
138.6
93.2
74.0
68.0
85.2
247.5
149.9
29.3 ( 0.2 0.5 ( 0.1 70.1 ( 0.2
28.6 ( 0.2 0.5 ( 0.1 70.9 ( 0.2
8.7 ( 0.3 0.6 ( 0.1 90.5 ( 0.4
9.6 ( 0.2 0.6 ( 0.1 89.8 ( 0.2
10.3 ( 0.2 0.7 ( 0.1 88.9 ( 0.1
141.0
136.8
128.8
1
V ) kobs[ArH][1′]
with kobs ) k
(2)
11 + k-1/k2
In the presence of base k-1/k2 is assumed to be nearly zero,
i.e., only k1 determines the product formation. The ortho/para
ratio is influenced neither by deprotonation (k2) nor by the back
reaction (k-1). The overall rate therefore depends only on the
reaction of the electrophile 1′ with the aromatic compound, and
according to eq 1 nearly the maximal rate is observed. Thus, as
expected, only a small primary kinetic isotope effect results.
In the presence of HOTf, however, the deprotonation rate of
the C-D vibration bands. Since the rate constant Kdiss (Scheme
2) depends on the conversion in the presence of base,1 the
calculation of the concentration of 1′ is not possible. Therefore
we determined the isotope effect kH/kD using the initial velocity
Vstart, where kH/kD equals VH/VD.
Besides these comparative measurements, isotope effects can
also be determined by competitive reactions. We have therefore
investigated the isotope effect of toluene (2a) using mass
spectrometry,8 whereas for xylene (2b) the product ratios were
the σ-complex decreases (ka ≈ k-1), resulting in a primary
2
kinetic isotope effect and a decrease of the overall reaction rate.
That means, a faster back reaction (k-1) and/or a slower
deprotonation (k2) of the ortho σ-complex in the presence of
HOTf favors the para product.
1
determined by H NMR spectroscopy because in mass spec-
trometry different fragmentations of deuterated and not deuter-
ated 3d occur.
Figures 1 and 2 show the results of the kinetic measurements
of acylations of toluene (2a) and perdeuterated toluene-d8 (2a-
d8) with 1′ in the presence of both HOTf (Figure 1) and the
pyridine 4 (Figure 2). The experimental data show good
agreement of isotope effects obtained by comparative and
competitive determination, respectively. The results are sum-
marized in Table 1.
The different course of aromatic acylation reactions with
aroyltriflates in the presence of HOTf and base, respectively,
can be described qualitatively by energy profiles (Figure 3).
The results of our investigations presented in this publication
allow a general interpretation of the ortho/para isomeric ratio
of electrophilic aromatic acylation reactions depending on
reaction conditions.
As can be seen from Table 1, there is a significant primary
kinetic isotope effect in the presence of HOTf, whereas in the
presence of base the rates determined are only in the range of
the secondary isotope effect found for toluene-d3 (2a-d3). Kinetic
isotope effects of 1.75-1.95 obtained in the acylation of 2a,
2a-d8 and 2b, 2b-d10 in the presence of HOTf are comparable
with published data4 for similar reactions.
With these investigations we are also able to confirm
experimentally the supposed relationship3 between primary
kinetic isotope effects and the ortho/para distribution in aromatic
acylations. In Table 2 the isomer ratios of the acylation of
toluene (2a) with triflate 1 depending on the reaction conditions
are listed. As expected, the amount of meta-product 3b is small
and nearly constant independent of reaction conditions. Ap-
proximately 10% ortho substitution product 3a was obtained
in the presence of HOTf, whereas in the presence of base a
significant increase of ortho product to nearly 30% results.
The deprotonation rate of the σ-complex intermediates
obviously influences the product formation in aromatic acyla-
tions. From our experimental results we can exclude the
markedly lower ratio of ortho product 3a in the presence of
acid as a result of reversibility of the acylation for two reasons.
The ortho/para ratio 3a/3c remains constant during the whole
The normal Friedel-Crafts acylation, which requires more
than equimolar amounts of the AlCl3 catalyst, gives besides
G90% of para product only small amounts of ortho product
(1-9%).10 In addition to the factors already discussed in the
literature11 for positional selectivity in aromatic acylations, the
reaction behavior of the intermediate σ-complex has to be
considered according to the equation given in Figure 3b.
In previous investigations12 it was demonstrated that rearo-
matization of Wheland intermediates occurs from a bent
conformation with the leaving group in a quasiaxial position
(Scheme 3). The rate of deprotonation (k2), as well as the rate
of back reaction (k-1) of the σ-complexes, therefore depends
markedly on the preference of conformations R or â.
Due to the bulkyness of the onium complex of the acyl group
with AlCl3, the bent ortho σ-complex prefers the R-conformation
(10) (a) Stock, L. M.; Brown, H. C. AdV. Phys. Org. Chem. 1963, 1,
35-154. (b) Gore, P. H. In Friedel-Crafts and Related Reactions; Olah,
G. A., Ed.; Interscience Publishers: New York, 1964; Vol. III, Part 1, pp
1-381.
(11) Taylor, R. Electrophilic Aromatic Substitution; Wiley: Chichester,
1990; pp 483-490.
(12) Effenberger, F.; Reisinger, F.; Scho¨nwa¨lder, K. H.; Ba¨uerle, P.;
Stezowski, J. J.; Jogun, K. H.; Scho¨llkopf, K.; Stohrer, W.-D. J. Am. Chem.
Soc. 1987, 109, 882-892.

3432 J. Am. Chem. Soc., Vol. 123, No. 15, 2001
Effenberger and Maier
It is now also possible to interpret unusual ortho/para ratios
we have observed and published earlier. The benzoylation of
toluene with benzoyl triflate affords a markedly enhanced
percentage of ortho product (up to 30%) as could be demon-
strated by both the benzoylation with benzoyl triflate in the
presence of a base (27% ortho product)13 and the conversion
of toluene with benzoyl chloride and catalytic amounts (5 mol
%) of trifluoromethanesulfonic acid.14
Experimental Section
1
General. H NMR spectra were recorded on a Bruker AC 250 F
(250 MHz) in CDCl3 as solvent and TMS as internal standard. Mass
spectrometry was performed on a Varian MAT 711. Gas chromatog-
raphy was performed with a Carlo Erba Fractovap 4160 with FID and
on column injector, 0.45 bar hydrogen, 20 m column, phases PS086
or SDPE08, temperature program 40 °C isotherm per 1 min, heating
rate 10 °C/min, and 300 °C end temperature. 1,2-Dichloroethane (DCE)
was purified as described in the literature15 but freshly distilled under
Ar from CaH2. Toluene (2a) and p-xylene (2b) were distilled over a
Spaltrohr column, dried (CaH2), and freshly distilled. Trifluoromethane-
sulfonic acid (3 M) was twice fractionally distilled. Tri-tert-butylpy-
ridine (4) was prepared according to ref 16. Deuterated toluene (2a-
d3) was purchased from Cambridge Isotope Laboratories, toluene 2a-
d8 from Merck, and p-xylene 2b-d10 from Aldrich, and all compounds
2-d were used without further purification.
Preparation of a Solution of Triflate 1. A solution of 4-(2,2-
dimethylpropoxy)-2,3,5-trimethylbenzoic acid chloride1 in dichloro-
ethane (DCE) (10 mL) was added to a stirred suspension of AgOTf
(1.1 equiv) in DCE (10 mL). After 30 min, precipitated AgCl was
filtered off and washed with 2 mL of DCE. The combined filtrates
were transferred to a 25 mL graduated flask under Ar atmosphere which
was filled up with DCE. The concentration was then determined by
GC from at least two samples.
Determination of Isotope Effects. (a) For comparative reactions,
two aliquots (10 mL) of the same solution of 1 in DCE containing
HOTf (concentrations in Table 1) were each taken and transferred to
a 25 mL graduated flask which was filled up with DCE to 23 mL.
After standing at 25 °C for 1 h, to one aliquot was added the deuterated
and to one the not deuterated aromatic compound 2. The flasks were
filled up with DCE to 25 mL, and the reaction was started by shaking.
Samples were taken at different time intervals, worked up as described
under (b) and analyzed by GC according to ref 1. (b) For competitive
reactions, a solution of 1 in DCE, containing HOTf and 4, respectively,
was transferred to a 25 mL graduated flask which was filled up with
DCE to 22 mL (for concentrations see Table 1). After the mixture was
left to stand at 25 °C for 1 h, both the respective deuterated and not
deuterated aromatic compound 2 were added (1 mL each, Table 1).
The flask was filled up with DCE to 25 mL, and the reaction was started
by shaking. To the samples taken after different time intervals was
added aqueous Na2CO3 solution (5%), and the reaction mixture was
stirred for 1 h. After being extracted with DCE, the combined organic
phases were dried (MgSO4), and the solvent was removed in vacuo. In
the case of toluene, the reaction was evaluated by mass spectrometry
(EI, 70 eV) by using the ratio M+H/MD+ (MH+ ) 324; MD+ ) 331) as a
measure for the isotope effect. In the case of xylene, the product ratio
was determined by NMR spectroscopy.
Figure 3. Energy profile of the reaction of acylium ion 1′ with
aromatics in the presence of base 4 and HOTf, respectively.
Scheme 3
Isolation of Deuterated Anhydrides 3-d from Comparative
Reactions. To the remaining reaction mixture was added aqueous Na2-
CO3 solution (5%). After being stirred for 2 h, the reaction mixture
was extracted with diethyl ether. The combined extracts were washed
with water, dried (MgSO4), and concentrated. The crude products were
chromatographed on silica gel with petroleum ether/dichloromethane
(1:1) and recrystallized from MeOH/H2O.
(Scheme 3), resulting in a decreasing deprotonation rate k2,
whereas the back reaction becomes more probable.12 In the para
σ-complex, for steric reasons the â-conformation should be
preferential and thus deprotonation (k2) under rearomatization
will be favored against the back reaction (k-1). In terms of the
(13) Braun, E. Dissertation, Universita¨t Stuttgart, 1987.
(14) Effenberger, F.; Sohn, E.; Epple, G. Chem. Ber. 1983, 116, 1195-
1208.
(15) Gutmann, R. Dissertation, Universita¨t Stuttgart, 1981.
(16) Dimroth, K.; Mach, W. Angew. Chem., Int. Ed. Engl. 1968, 7, 460-
461.
equation given in Figure 3b this results in (k2/k-1 para
k-1 ortho. Therefore the para isomer is formed in a higher ratio
as expected considering only the values of k1.
)
> (k2/
)

Ortho/Para Ratio in Aromatic Acylation Reactions
J. Am. Chem. Soc., Vol. 123, No. 15, 2001 3433
Data for 3a-d3: Colorless crystals, mp 95-96 °C (MeOH/H2O); 1H
NMR (CDCl3) δ 1.12 (s, 9H, C(CH3)3), 2.13 (s, 3H, CH3), 2.23 (s, 3H,
CH3), 2.24 (s, 3H, CH3), 3.40 (s, 2H, CH2), 6.95 (s, 1H, Ar-H6), 7.24
(AA′, J ) 8.0 Hz, 2H, Ar-H2′,6′), 7.70 (BB′, J ) 8.0 Hz, 2H, Ar-
H3′,5′). MS (EI, 70 eV) m/z (%) 327 (27) [M+], 312 (3) [M+ - CH3],
309 (20) [M+ - CD3].
Data for 3a-d7: Colorless crystals, mp 95.5-96.5 °C (MeOH/H2O);
1H NMR (CDCl3) δ 1.12 (s, 9H, C(CH3)3), 2.13 (s, 3H, CH3), 2.23 (s,
3H, CH3), 2.24 (s, 3H, CH3), 3.40 (s, 2H, CH2), 6.95 (s, 1H, Ar-H6).
Anal. Calcd for C22H21D7O2 (331.4): C, 79.93; H, 6.38; D, 4.22.
Found: C, 79.89; H, 6.48; D, 4.32. MS (EI, 70 eV) m/z (%) 331 (26)
[M+], 316 (10) [M+ - CH3], 313 (19) [M+ - CD3].
Data for 3d-d9: Colorless crystals, mp 78-79 °C (MeOH/H2O);
1H NMR (CDCl3) δ 1.11 (s, 9H, C(CH3)3), 2.20 (s, 3H, CH3), 2.24 (s,
3H, CH3), 2.25 (s, 3H, CH3), 3.39 (s, 2H, CH2), 6.94 (s, 1H, Ar-H6).
Anal. Calcd for C23H21D9O2 (347.4): C, 79.52; H, 6.09; D, 5.18.
Found: C, 79.39; H, 5.96; D, 5.10. MS (EI, 70 eV) m/z (%) 347 (26)
[M+], 332 (5) [M+ - CH3], 329 (24) [M+ - CD3].
JA0022066