The importance of the ortho effect in the solvolyses of 2,6-difluorobenzoyl chloride

Article abstract of DOI:10.1002/poc.1906

The ortho effect of the chloro substituents in 2,6-dichlorobenzoyl chloride sufficiently hindered attack on the acyl carbon such that an ionization mechanism was observed over the full range of solvents studied. We now compare this behavior with that of 2,6-difluorobenzoyl chloride. The smaller fluoro substituents allow the dominant pathway to be addition-elimination (association-dissociation) in all solvents except those rich in fluoroalcohol, where ionization is dominant. Ranges of operation for both mechanisms had previously been observed for the parent benzoyl chloride but with a wider ionization range than for the 2,6-difluoro derivative. This indicates that, relative to the parent, the electronic destabilizing influence of the fluorines on acyl cation formation outweighs the steric retardation to attack because of the presence of the two ortho-fluorine atoms. An extended (two-term) Grunwald-Winstein equation treatment of the solvolyses of 2,6-difluorobenzoyl chloride is reported. Copyright

Full text of DOI:10.1002/poc.1906

Research Article
Received: 24 March 2011,
Revised: 3 June 2011,
Accepted: 6 June 2011,
Published online in Wiley Online Library: 21 July 2011
(wileyonlinelibrary.com) DOI 10.1002/poc.1906
The importance of the ortho effect in the
solvolyses of 2,6-difluorobenzoyl chloride†
Kyoung-Ho Parka and Dennis N. Kevilla*
The ortho effect of the chloro substituents in 2,6-dichlorobenzoyl chloride sufficiently hindered attack on the acyl
carbon such that an ionization mechanism was observed over the full range of solvents studied. We now compare this
behavior with that of 2,6-difluorobenzoyl chloride. The smaller fluoro substituents allow the dominant pathway to be
addition–elimination (association–dissociation) in all solvents except those rich in fluoroalcohol, where ionization is
dominant. Ranges of operation for both mechanisms had previously been observed for the parent benzoyl chloride
but with a wider ionization range than for the 2,6-difluoro derivative. This indicates that, relative to the parent, the
electronic destabilizing influence of the fluorines on acyl cation formation outweighs the steric retardation to attack
because of the presence of the two ortho-fluorine atoms. An extended (two-term) Grunwald–Winstein equation treat-
ment of the solvolyses of 2,6-difluorobenzoyl chloride is reported. Copyright # 2011 John Wiley & Sons, Ltd.
Keywords: fluorine substituents; Grunwald–Winstein equation; ortho effect; solvolysis
was applied, with a moderate l value of 0.29ꢁ0.07. It was postu-
lated, consistent with earlier studies,[6,11–13] that this l value arose
from a stabilization of a developing carbocation by nucleophilic
solvation. Application of Eqn (2) to other dichlorobenzoyl chlor-
ides showed for the 2,4-derivative, a balance between addition–
elimination (Scheme 1) and ionization (Scheme 2) mechanisms.
Other derivatives (3,4- and 3,5-) with no ortho substituents
showed a wide range of solvents for which solvolyses occurred
by the addition–elimination mechanism, and only in the high
ionizing and low nucleophilic solvents, containing an apprecia-
ble fluoroalcohol component, could evidence for the ionization
pathway be found.
In the present study, the chlorines of 1 are replaced by fluor-
ines to give 2,6-difluorobenzoyl chloride (3). With regard to the
changes in electronic effects in the meta position, the Hammett
s values are essentially identical at 0.34 for fluorine and 0.37 for
chlorine.[14,15] Differences are larger in the para position, where
resonance interactions also come into play with values of 0.06
for fluorine and 0.22 for chlorine,[14,15] indicative of a larger
electron-supplying resonance contribution from the fluorine.
Probably more relevant are the sp + values based on the solvoly-
sis of t-cumyl chloride,[15,16] where values of ꢂ0.07 and 0.11, re-
spectively, are obtained. The difference in values of 0.16 for sp
values and 0.18 for s+p values are, however, essentially identical.
When the introduction of substituents is in the ortho position
relative to the reaction center, the “peculiar effects”[17] are largely
because of the steric phenomenon and are considered to
INTRODUCTION
Recently, we reported on the solvolyses of 2,6-dichlorobenzoyl
chloride (1) and isomeric dichlorobenzoyl chlorides.[1] It was pos-
sible to compare the solvolytic behavior with that of the earlier
studied[2,3] 2,6-dimethylbenzoyl chloride (2). Compound 2 was
found, because of the steric hindrance on the attack on the acyl
carbon from the presence of the two ortho-methyl groups, cou-
pled with their electron supplying influence, to undergo solvoly-
sis by an ionization (unimolecular) mechanism.
Because of favorable electronic influences, 2 underwent solvol-
yses very rapidly and, even with a rapid-response conductivity
bridge, rates at 25.0ꢀC could be determined only in ethanol
and 97–70% acetone.[3] The solvolyses of 1 were retarded be-
cause the ortho substituents were now electron-withdrawing,
and elevated temperatures (mainly 55.0ꢀC) were required and
specific rates could now be conveniently determined over the
full range of solvents usually included in studies of solvolytic dis-
placement reactions.[4–7]
In particular, the wide range of solvents studied for the solvo-
lyses of 1 allowed for the application of the original (one-term)
Grunwald–Winstein Eqn (1)[8] and the extended (two-term)
Eqn (2).[5,9] In Eqns (1) and (2), k and ko are the specific rates
(first-order rate coefficients) for the solvolysis of a substrate in a
given solvent and in the standard solvent (80% ethanol), respec-
tively, m is the measure of the sensitivity to changes in solvent
ionizing power (Y),[4–11] l is the measure of the sensitivity to
changes in solvent nucleophilicity (N),[5–7,9] and c is the constant
(residual) term.
* Correspondence to: D. N. Kevill, Department of Chemistry and Biochemistry,
Northern Illinois University, DeKalb, IL 60115–2862, USA.
E-mail: dkevill@niu.edu
logðk=koÞ ¼ mY þ c
(1)
(2)
logðk=koÞ ¼ lN þ mY þ c
†
Abstracted, in part, from the Ph.D. dissertation of Kyoung-Ho Park, Northern
Illinois University, May 2009.
It was found[1] that over the full range of 30 well-chosen sol-
vents, a reasonably good correlation was obtained on applica-
tion of Eqn (1) and this correlation was improved when Eqn (2)
a K.-H. Park, D. N. Kevill
Department of Chemistry and Biochemistry, Northern Illinois University,
DeKalb, IL 60115-2862, USA
J. Phys. Org. Chem. 2012, 25 267–270
Copyright # 2011 John Wiley & Sons, Ltd.

K. H. PARK AND D. N. KEVILL
S
H
S
O
O
C Cl
O-
fast
SOH
SOH
+
-
R
C Cl
R
C OS
O
SOH2
+ Cl
R
R
C Cl
O
+
-
O
Scheme 1. Addition-Elimination (Association-Dissociation) Pathway
slow
-
SOH
SOH
+
R
C
O
Cl
R
C O
R
C
O
R
C OS
O
+
SOH2
O
-
Cl
H
S
Scheme 2. Ionization Pathway
involve, in addition to primary effects, steric hindrance to solva-
tion and secondary steric effects. To evaluate the steric effects
one must be able to quantify the accompanying polar effects.
The earlier attempts involved the assumption that polar effects
at the ortho position would mirror those at the para position
and several examples of the application of this approach have
been discussed.[18,19] Charton has suggested that it is unlikely
that so scales of wide applicability can be developed.[20,21]
Charton developed a steric parameter (υ), which is based on hy-
drogen as the standard (υ=o), with the result that all values are
positive.[22] The halogens have values of 0.27, 0.55, 0.65, and
0.78 for F, Cl, Br, and I, respectively, with the υ value for fluorine
being half-way between those for hydrogen and chlorine.
extremely large (0.06 to 0.83). This is consistent with the general
consensus, discussed in the introduction, that fluorine is slightly
more electron-withdrawing than chlorine, coupled with the
Table 1. Specific rates of solvolysis (k) of 2,6-difluorobenzoyl
chloride (3) in a wide variety of hydroxylic solvents at 55.0ꢀC
and a comparison, k(3)/k(1) ratios, with corresponding values
for the solvolyses of 2,6-dichlorobenzoyl chloride (1)
Solventa
103k(sꢂ1 b
)
NT
YdCl
k(3)/k(1)
c
100% EtOH
90% EtOH
80% EtOH
70% EtOH
60% EtOH
50% EtOH
100% MeOH
90% MeOH
80% MeOH
70% MeOH
60% MeOH
50% MeOH
90% Acetone
80% Acetone
70% Acetone
60% Acetone
50% Acetone
100% TFE
97% TFE
90% TFE
80% TFE
70% TFE
60% TFE
50% TFE
80T–20E
60T–40E
40T–60E
6.65ꢁ0.01
15.7ꢁ0.1
24.1ꢁ0.3
34.0ꢁ0.1
48.4ꢁ0.8
66.1ꢁ3.4
37.5ꢁ0.4
70.6ꢁ0.1
112. ꢁ 1.0
159. ꢁ 1.0
211. ꢁ 3.0
308. ꢁ 2.0
3.16ꢁ0.01
8.29ꢁ0.01
14.9ꢁ0.1
28.5ꢁ0.2
42.7ꢁ0.4
0.37 ꢂ2.52 724
0.16 ꢂ0.94 339
0.00
ꢂ0.20
ꢂ0.38
ꢂ0.58
0.00 163
0.78
1.38
2.02
74
35
15
RESULTS
The specific rates of solvolysis were determined by a rapid re-
sponse conductivity technique for 3 in 31 pure and binary sol-
vents at 55.0ꢀC. The specific rates (first-order rate coefficients)
are reported in Table 1. Also listed in the table are the re-
0.17 ꢂ1.17 367
ꢂ0.01
–0.18 212
0.67 108
ꢂ0.06
ꢂ0.40
ꢂ0.54
ꢂ0.75
ꢂ0.35
1.46
2.07
2.70
2.39
56
28
20
[6]
[4,10,11]
quired NT and YCl
values from the literature. The sol-
vents used in the study consisted of ethanol, methanol, and
2,2,2-trifluoroethanol (TFE) and, for each, a series of mixtures
with water. Determinations of the specific rates of solvolysis in
binary mixtures of water with 1,1,1, 3,3,3-hexafluoro-2-propanol
(HFIP) and acetone and in mixtures of TFE with ethanol are also
reported.
Also contained in Table 1 are the ratios, for each solvent, of the
specific rate of solvolysis of 3 relative to that of 1. These values
vary widely from 947 in 80% acetone to 0.062 in 97% HFIP.
ꢂ0.37 ꢂ0.83 947
ꢂ0.42
ꢂ0.52
ꢂ0.70
0.17 329
0.95
1.73
2.81
2.83
2.85
2.90
2.96
3.06
3.16
1.89
0.63
73
33
0.11
0.17
0.32
0.60
0.72
0.82
0.83
1.31
0.389ꢁ0.001 ꢂ3.93
0.371ꢁ0.001 ꢂ3.30
0.934ꢁ0.001 ꢂ2.55
3.56ꢁ0.09
6.10ꢁ0.01
11.9ꢁ0.1
22.2ꢁ0.1
ꢂ2.22
ꢂ1.98
ꢂ1.85
ꢂ1.73
DISCUSSION
The important thing to consider is the establishment of the solvo-
lytic behavior of 3, with two ortho-fluoro substituents. Included
in the study is a comparison with 1 (with larger chlorine
atoms) and with the parent benzoyl chloride (with smaller hydro-
gen atoms in the ortho positions).[23–28] The behavior of 3 for
most solvents was found to be very different to that of 2,6-
dichlorobenzoyl chloride (1). The (k3/k1) specific rate ratios given
in Table 1 for the solvolyses of 3 and 1 are extremely illuminating
in this regard. The compound 1 has been shown to solvolyze by
an ionization mechanism over the full range of solvents.[1] If this
was also the case for the 2,6-difluorobenzoyl chloride (3), a sim-
ilar behavior and a fairly constant ratio would be expected.
In the aqueous fluoroalcohols, favoring ionization, the ratios
are below unity for TFE (0.11) and all the binary mixtures of
water with TFE or HFIP that were studied, but the range is not
0.624ꢁ0.005 ꢂ1.76
1.66ꢁ0.01
3.39ꢁ0.01
5.35ꢁ0.01
3.46ꢁ0.01
2.48ꢁ0.02
6.84ꢁ0.01
ꢂ0.94
12
53
ꢂ0.34 ꢂ0.48
20T–80E
0.08 ꢂ1.42 262
97% HFIP
90% HFIP
70% HFIP
ꢂ5.26
5.17
4.31
3.83
0.062
0.103
0.294
ꢂ3.84
ꢂ2.94
aPrepared on a v/v basis at 25.0ꢀC with other component wa-
ter, except that T–E represents TFE–ethanol mixtures and
TFE–H20 and HFIP–H20 mixtures are prepared on a w/w basis.
bWith associated standard deviation.
cSolvent nucleophilicity values from Ref.
dSolvent ionizing power values from Ref.
.
[6]
[4,10,11]
.
wileyonlinelibrary.com/journal/poc
Copyright # 2011 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2012, 25 267–270

ORTHO EFFECT IN THE SOLVOLYSES OF 2,6-DIFLUOROBENZOYL CHLORIDE
operation of the mechanism of Scheme 2 as the dominant mech-
anism for both 1 and 3 in these solvents.
the 47 solvents were divided into 32 proceeding by the ioniza-
tion mechanism (l=0.47ꢁ0.03; m=0.79ꢁ0.02; correlation coeffi-
cient (R) of 0.990) and 12 proceeding by the bimolecular
mechanism (l=1.27ꢁ0.29; m=0.46ꢁ0.07; R value of 0.917), with
three borderline solvents showing intermediate behavior.[28] For
3, the relative importance of the two pathways, in terms of the
number of solvolyses assigned to each, is reversed, with etha-
nol–water, methanol–water, acetone–water, and TFE–ethanol
solvents assigned to the addition–elimination pathway. For
these 20 solvents, values are obtained of l=1.80ꢁ0.18; m=
0.66ꢁ0.06; R value of 0.940. The plot is shown in Fig. 2. For the
nine TFE–H20 and HFIP–H20 solvents the ionization pathway
dominates, with values of 0.88ꢁ0.10 for l; 1.12ꢁ0.14 for m,
and with an R value of 0.964. This plot is shown in Fig. 3.
In ethanol (724) and methanol (367) and binary mixtures with
water of these two alcohols and acetone, the values are above
unity and they fall as the water component is increased, reaching
values of 15 for 50% EtOH, 20 for 50% MeOH, and 33 for 50% ac-
etone. For 3, these values are best explained by an addition–
elimination mechanism (Scheme 1) accompanying the ionization
mechanism in these solvents, in contrast to the operation of only
the ionization mechanism in the presence of the two ortho-
chloro-substituents of 1. Accordingly, the drop off in the specific
rate values as one goes to the less ionizing and more nucleo-
philic solvents will be much less for 3 than for 1, leading to the
observation of increased (k3/k1) ratios.
A more sophisticated approach involves the application of
the Grunwald–Winstein equation[8,9] (Eqns (1) and (2)). Applying
Eqn (1) to the data in Table 1 leads to what at first appears
to be a scatter plot (Fig. 1). There are, however, fairly linear
plots for each individual binary system. This is in stark con-
trast to the corresponding plot for 1 where a good linear plot
was obtained[1] with all solvents included, with a slope of 0.55ꢁ
0.04 and a correlation coefficient of 0.943. The behavior for sol-
volyses of 3 is typical for what is observed when a bimolecular
mechanism, involving nucleophilic attack by a solvent molecule,
is operative for all or a large fraction of the solvents.[6,29–32]
When the extended (two-term) Grunwald–Winstein equation
is applied, a poor correlation is again obtained when all 31 sol-
vents are included. The l value is 0.80ꢁ0.10, the m value is
0.42ꢁ0.07, the c value is 0.07ꢁ0.10 and the multiple correlation
coefficient is at a low value of 0.839. One can divide up the sol-
volyses into two groups, as was achieved previously for the ma-
jority of monosubstituted benzoyl chlorides studied.[28] One
group is dominated by the addition–elimination pathway
(Scheme 1) and the other by the ionization pathway (Scheme 2).
A listing of some typical correlation values for mono-substituted
and di-substituted benzoyl chloride has recently been pre-
sented.[1] For the parent benzoyl chloride (ortho hydrogens),
1.5
50M
60M
1.0
70M
80M
50E
60E
70E
90M
50A
0.5
0.0
100M
60A
70A
80E
90E
80A
100E
-0.5
-1.0
-1.5
-2.0
20T-80E
40T-60E
60T-40E
R = 0.940
l/m = 2.73
80T-20E
-1.5
-2.0
-1.0
-0.5
0.0
0.5
1.0
1.80 NT + 0.66 YCl
Figure 2. Plot of log (k/ko) for the solvolyses of 2,6-difluorobenzoyl chlo-
ride (3) in 20 pure and binary solvents at 55.0ꢀC against (1.80NT+0.66YCl).
Solvents: See caption of Fig. 1
3.5
50M
0.2
60M
70M
50TFE
0.0
3.0
2.5
2.0
1.5
1.0
0.5
0.0
80M
70E
90M
50E
-0.2
60TFE
60E
60A
100M
90E
-0.4
50A
80E
50TFE
60TFE
70TFE
-0.6
-0.8
-1.0
-1.2
-1.4
-1.6
-1.8
-2.0
70HFIP
70A
80TFE
97HFIP
90TFE
100E
70HFIP
70TFE
20T-80E
80A
80TFE
90TFE
90HFIP
97HFIP
90HFIP
40T-60E
90A
60T-40E
80T-20E
100TFE
97TFE
3
R = 0.964
l/m = 0.79
97TFE
-3
-2
-1
0
1
2
4
5
6
YCl
-2.0 -1.8 -1.6 -1.4 -1.2 -1.0 -0.8 -0.6 -0.4 -0.2 0.0
0.88 NT + 1.12 YCl
Figure 1. Plot of log (k/ko) for the solvolyses of 2,6-difluorobenzoyl chlo-
ride (3) in 31 pure and binary solvents at 55.0ꢀC against YCl values. Sol-
vents: E, EtOH–H2O; M, MeOH–H2O; A, Acetone–H2O; TFE, TFE–H2O;
HFIP, HFIP–H2O; T–E, TFE–EtOH
Figure 3. Plot of log (k/ko) for the solvolyses of 2,6-difluorobenzoyl chlo-
ride (3) in nine binary solvents at 55.0ꢀC against (0.88NT+1.12YCl). Sol-
vents: See caption of Fig. 1
J. Phys. Org. Chem. 2012, 25 267–270
Copyright # 2011 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

K. H. PARK AND D. N. KEVILL
[6] D. N. Kevill, in Advances in Quantative Structure-Property Relation-
ships. Ed.: M Charton, JAI Press, Greenwich, CT, 1996, pp. 81–115.
[7] D. N. Kevill, S. W. Anderson, Journal of Organic Chemistry 1991, 56,
1845–1850.
[8] E. Grunwald, S. Winstein, Journal of the American Chemical Society
1948, 70, 846–854.
[9] S. Winstein, E. Grunwald, H. W. Jones, Journal of the American Chem-
ical Society 1951, 73, 2700–2707.
[10] D. N. Kevill, Z. H. Ryu, Int. Journal of Molecular Science 2006, 7,
451–455.
CONCLUSIONS
The reduction in the size of the halogen substituents in going
from 2,6-dichlorobenzoyl chloride (1) to 2,6-difluorobenzoyl
chloride (3) has a profound effect upon the mechanism of solvo-
lyses in the commonly used hydroxylic pure and binary solvents.
For 1, an ionization mechanism over the full range of the solvents
was previously established and it was proposed that this was ob-
served by default, despite the acyl cation-destabilizing influence
of the chlorine substituents, because of an appreciable steric hin-
drance towards the approach of the nucleophile to the acyl carbon.
In contrast, for 3, with the chlorine substituents being replaced
by considerably smaller fluorines, there is a reduced hindrance to
the approach of the nucleophile. It is now observed that the
addition–elimination pathway is dominant in all solvents, except
those rich in fluoroalcohol (TFE or HFIP). For these solvents, the
ionization pathway continues to be the dominant one.
[11] D. N. Kevill, M. J. D’Souza, Journal of Chemical Research (S) 1993,
174–175.
[12] D. N. Kevill, M. J. D’Souza, Journal of Chemical Research 2008,61–66.
[13] T. W. Bentley, G. E. Carter, Journal of the American Chemical Society
1982, 104, 5741–5747.
[14] O. Exner, in Correlation Analysis in Chemistry, Recent Advances
(Eds., N. B. Chapman, J. Shorter), Plenum Press, New York, NY,
1978, Chap. 10.
[15] J. Shorter, Correlation Analysis of Organic Reactivity, Research Studies
Press, New York, NY, 1982, Chap. 3.
[16] H. C. Brown, Y. Okamoto, Journal of the American Chemical Society
1958, 80, 4979–4987.
[17] Ref [15], p. 112.
EXPERIMENTAL
[18] J. Shorter, in Advances in Linear Free Energy Relationships (Eds., N.
B. Chapman, J. Shorter), Plenum Press, New York, NY, 1972, chap. 2.
[19] J. E. Leffler, E. Grunwald, Rates and Equilibria of Organic Reactions,
Wiley, New York, NY, 1963, p.225.
[20] M. Charton, Journal of the American Chemical Society 1969, 91,
6649–6654.
[21] M. Charton, Progress in Physical Organic Chemistry 1971, 8, 235–317.
[22] M. Charton, Progress in Physical Organic Chemistry 1981, 13,
119–252.
[23] I. Lee, I. S. Koo, S. C. Sohn, H. H. Lee, Bulletin of the Korean Chemical
Society 1982, 3, 92–98.
[24] I. Lee, D. D. Sung, T. S. Uhm, Z. H. Ryu, Journal of Chemical Society
Perkin Transaction 2 1989, 1697–1701.
[25] T. W. Bentley, I. S. Koo, Journal of Chemical Society Perkin Transaction
2 1989, 1385–1392.
[26] T. W. Bentley, I. S. Koo, S. J. Norman, Journal of Organic Chemistry
1991, 56, 1604–1609.
2,6-Difluorobenzoyl chloride (3, 99%) was used as received from
Sigma-Aldrich Chemical Company (St. Louis, Missouri 63178, USA).
Solvents were purified as previously described.[7] The kinetic
experiments were carried out in a conductivity cell containing
5.00mL of the appropriate solvent to which 1mL of a 1.00M stock
solution of 3 in acetonitrile was added (substrate concentration
of 2ꢃ10ꢂ4M). Data were collected until a constant conductivity
value was observed, after which the data were subjected to curve
fitting using the ORIGIN 3.0 (Microsoft Corporation, Redmond,
Washington 98052, USA) program. The monitoring procedure
and the procedure for calculation of the first-order rate coeffi-
cients (specific rates) were as previously reported.[33]
[27] K.-T. Liu, H.-I. Chen, Journal of Chemical Society Perkin Transaction 2
2000, 893–898.
[28] D. N. Kevill, M. J. D’Souza, Journal of Physical Organic Chemistry
2002, 15, 881–888.
[29] A. Streitwieser, Solvolytic Displacement Reactions, McGraw-Hill, New
York, NY, 1962.
REFERENCES
[1] K.-H. Park, D. N. Kevill, Journal of Physical Organic Chemistry 2011,
24, in press (D01 10.1002/poc.1851)
[2] M. L. Bender, M. C. Chen, Journal of the American Chemical Society
1963, 85, 30–36.
[30] P. R. Wells. Linear Free Energy Relationships, Academic Press, New
York, NY, 1968.
[31] A. H. Fainberg, S. Winstein, Journal of the American Chemical Society
1956, 78, 2770–2777.
[32] P. R. Wells, Chemistry Review 1963, 63, 171–219.
[33] H. J. Koh, K. L. Han, H. W. Lee, I. Lee, Journal of Organic Chemistry
1998, 63, 9834–9839.
[3] T. W. Bentley, H. C. Harris, I. S. Koo, Journal of Chemical Society Perkin
Transaction 2, 1988, 783–789.
[4] T. W. Bentley, G. Llewellyn, Progress in Physical Organic Chemistry
1990, 17, 121–158.
[5] F. L. Schadt, T. W. Bentley, P. v. R. Schleyer, Journal of the American
Chemical Society 1976, 98, 7667–7674.
wileyonlinelibrary.com/journal/poc
Copyright # 2011 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2012, 25 267–270