Solvolyses of naphthoyl chlorides. Solvent effect and Grunwald-Winstein correlation analyses with YxBnCl scales

Article abstract of DOI:10.1002/poc.549

The solvolyses of 1-naphthoyl (2), 2-naphthoyl (3), 4-methyl-1-naphthoyl (4) and 6-methoxy-2-naphthoyl (5) chlorides in a variety of solvents were studied, and correlation analyses by using the single- and dual-parameter Grunwald-Winstein equations were examined. An excellent linear relationship (R = 0.995) for 4, log (k/k₀) = 0.733Y_xBnCl + 0.269N_OTs, was observed. An S_N1-like mechanism with decreasing extent of nucleophilic solvent participation was found in the solvolysis of 2 and 4.2-Naphthoyl chloride is likely to have a mechanism at the borderline of S_N1-like dissociation and an addition-elimination process. 6-Methoxy-2-naphthoyl chloride shows more S_N1-like character than 3 and is associated with nucleophilic solvent intervention more pronounced than that for 2 and 4. The applicability and the advantages of using the Y_xBnCl scale for different types of substrates are discussed. Copyright

Full text of DOI:10.1002/poc.549

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2002; 15: 750–757
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/poc.549
Solvolyses of naphthoyl chlorides. Solvent effect and
Grunwald±Winstein correlation analyses with YxBnCl scales
†
‡
Kwang-Ting Liu,* Patty Y.-H. Hwang and Hun-I Chen
Department of Chemistry, National Taiwan University, Taipei, Taiwan 106, Republic of China
Received 21 February 2002; revised 21 June 2002; accepted 5 July 2002
ABSTRACT: The solvolyses of 1-naphthoyl (2), 2-naphthoyl (3), 4-methyl-1-naphthoyl (4) and 6-methoxy-2-
naphthoyl (5) chlorides in a variety of solvents were studied, and correlation analyses by using the single- and dual-
parameter Grunwald–Winstein equations were examined. An excellent linear relationship (R = 0.995) for 4, log
(
k/k ) = 0.733Y
‡ 0.269N , was observed. An S 1-like mechanism with decreasing extent of nucleophilic
0
xBnCl
OTs
N
solvent participation was found in the solvolysis of 2 and 4. 2-Naphthoyl chloride is likely to have a mechanism at the
borderline of S 1-like dissociation and an addition–elimination process. 6-Methoxy-2-naphthoyl chloride shows
N
more S 1-like character than 3 and is associated with nucleophilic solvent intervention more pronounced than that for
N
2
and 4. The applicability and the advantages of using the YxBnCl scale for different types of substrates are discussed.
Copyright  2002 John Wiley & Sons, Ltd.
KEYWORDS: naphthoyl chlorides; solvolysis; solvent effect; Grunwald–Winstein correlation analysis
INTRODUCTION
of solvolysis for aromatic acyl chlorides containing
different substituents were suggested from rate–product
8
9
Acyl chlorides are fundamental organic substrates with1
high reactivity towards many kinds of transformations.
Recent examples of physical organic and synthetic
studies illustrate continuing interest in research on this
category of compounds. Concerning the kinetics and
mechanisms of the solvolysis of acyl chlorides, early
work by Hudson and co-workers suggested a dependence
selectivity studies. Non-linear log k vs mY plots were
Cl
10
observed again. On the other hand, the solvent effect on
2
3
the solvolytic reactivity of a series of substituted benzoyl
2c
chlorides (1) was examined using the Y values, YBnCl,
derived from the solvolysis rate constants of a-tert-
butyl(4-methylphenyl)methyl chloride and 2-aryl-2-
11a,b
11
chloroadamantane,
specific to benzylic substrates.
4
of the mechanism on solvent composition. The failure of
An S 1 mechanism was realized in the case of 2,6-
N
employing the single-parameter Grunwald–Winstein
dimethylbenzoyl chloride (1a), and S 1 mechanisms
with various extents of nucleophilic solvent participation
were found in the solvolysis of 1b–d, while different
N
5
equation [Eqn. (1)], with the original Y values, to the
6
solvolysis of 4-nitrobenzoyl chloride and acetyl and
4c
benzoyl chlorides in hydroxylic solvents was reported.
types of non-S 1 reaction mechanisms were suggested
N
2
c
for the parent and deactivated benzoyl chlorides 1e–g.
logꢀk=k † ˆ mY
ꢀ1†
0
The successful application of the YBnCl scale to the
benzoyl system made it desirable to explore the
possibility of employing the YxBnCl scale, the parameter
of solvent ionizing power suitable for the benzylic system
with extended charge delocalization based on the use of
a-tert-butyl(2-naphthyl)methyl chloride as the reference
Mechanistic studies of the solvolysis of benzoyl
chlorides have advanced since the 1980s. Bentley et al.
proposed a limiting S 1 mechanism for the solvolysis of
N
7
4
-methoxybenzoyl chloride, and different mechanisms
12,13
standard,
to the solvolysis of naphthoyl chlorides.
Consequently, the solvolysis of naphthoyl chlorides
2
–5 in a variety of solvents was studied. The applicability
*
Correspondence to: K.-T. Liu, Department of Chemistry, National
Taiwan University, Taipei, Taiwan 106.
E-mail: ktliu@ccms.ntu.edu.tw
of the YxBnCl scale in Grunwald–Winstein-type correla-
tion analyses with single- and dual-parameter equations
†
This paper is dedicated to Professor Herbert C. Brown on the occasion
14
[Eqns (1) and (2) ] can be confirmed, and the mechan-
of his 90th birthday.
‡
NSC Undergraduate Research Participant, July 2000 to Febru-
ism of solvolysis may be understood.
ary 2001.
Contract/grant sponsor: National Science Council, Republic of
China.
logꢀk=k † ˆ mY ‡ lN
ꢀ2†
0
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 750–757

SOLVOLYSES OF NAPHTHOYL CHLORIDES
751
solvolysis were monitored conductometrically at appro-
priate temperatures. First-order kinetics were observed
up to at least 80% of the reaction. The rate constants at
2
5°C are shown in Table 1.
The solvolysis of 2 and 3 in acetone–methanol and in
acetonitrile–methanol has also been studied by Yoon et
15
al. For rate constants of methanolysis, the variation of
the two sets of data is within about 10%. Correlation
analyses using the single-parameter Grunwald–Winstein
9
11
BnCl
12
equation [Eqn. (1)] against Y , Y
and YxBnCl
Cl
were carried out. The results for all solvents (All),
nucleophilic solvents (AEM) and iso-dielectric solvents
(
TE) are listed in Table 2. From this table it is clear that,
2c,10
similarly to those reported previously,
poor correla-
tions and scattered data points for log k–YCl plots were
found in all four cases.
With the exception of 2-naphthoyl chloride (3), in
other cases the log k–YxBnCl plots yielded two lines with
16
excellent correlations (R ꢁ 0.99), one for aqueous
acetone, ethanol and methanol (AEM) and the other for
the trifluoroethanol–ethanol (TE) mixtures. The log
k–YxBnCl plot for 2 is shown in Fig. 1 as an example.
The appearance of a downward splitting line for poorly
nucleophilic trifluoroethanol–ethanol mixtures from that
for nucleophilic solvents suggests significant nucleophi-
lic solvent intervention in solvolysis, as was illustrated in
2c,17
recent examples.
Moreover, the difference between
RESULTS AND DISCUSSION
-Naphthoyl chloride (2), 2-naphthoyl chloride (3), 4-
methyl-1-naphthoyl chloride (4) and 6-methoxy-2-
naphthoyl chloride (5) were purified or prepared, and
were solvolyzed in a variety of solvents. The rates of
m
AEM and mTE was found to exhibit various extents of
18
nucleophilic solvent intervention. From Table 2, Dm
plots increased from 4
1
(mAEM À m ) of log k–Y
TE
xBnCl
(0.052) to 2 (0,127) to 5 (0.230). Therefore, regression
analyses of log k values in Table 1 employing the dual-
parameter equation [Eqn. (2)] were carried out. Since the
a
Table 1. Solvolytic rate constants at 25°C
b
Solvent
2
3
4
5
À3
À3
À2
À2
À2
À2
À2
À2
À1c
À4
À3
À3
À2
À4
À3
À3
À3
À3
À3
À3
À2
À2
À4
À4
À3
À3
À3
À3
À3
À3
À4
À3
À2
À2
À4
À3
À3
À2
À2
À3
À2
À3
À2
1
9
8
7
6
1
9
8
7
9
8
7
6
5
1
8
6
4
00E
2.05 Â 10
6.70 Â 10
1.69 Â 10
4.03 Â 10
9.08 Â 10
1.51 Â 10
2.69 Â 10
9.34 Â 10
3.12 Â 10
5.14 Â 10
1.96 Â 10
7.21 Â 10
2.68 Â 10
–
7.30 Â 10
2.08 Â 10
3.37 Â 10
5.58 Â 10
9.33 Â 10
4.55 Â 10
8.54 Â 10
1.49 Â 10
2.13 Â 10
3.37 Â 10
7.58 Â 10
1.43 Â 10
3.28 Â 10
9.24 Â 10
8.96 Â 10
3.29 Â 10
1.13 Â 10
7.38 Â 10
7.00 Â 10
2.37 Â 10
6.56 Â 10
–
7.45 Â 10
2.18 Â 10
5.46 Â 10
1.07 Â 10
2.44 Â 10
5.40 Â 10
1.29 Â 10
2.60 Â 10
5.97 Â 10
–
0E
0E
0E
0E
–
À2
À1c
À1c
00M
0M
4.35 Â 10
1.52 Â 10
5.13 Â 10
–
0M
0M
À3
À3
À2
À2
0A
1.10 Â 10
5.30 Â 10
1.96 Â 10
7.81 Â 10
–
À3
À3
À3
À2
À2
À2
À3
À3
0A
1.05 Â 10
3.16 Â 10
9.94 Â 10
3.44 Â 10
3.33 Â 10
1.59 Â 10
4.74 Â 10
2.24 Â 10
0A
0A
0A
À1c
À1
À2
À3
00T
0T20E
0T40E
0T60E
5.36 Â 10
1.26 Â 10
2.88 Â 10
9.63 Â 10
–
c
1.02
À1c
À2
1.59 Â 10
5.00 Â 10
a
À1
In s
.
b
Abbreviations of solvents: A = acetone, E = ethanol, M = methanol, T = 2,2,2-trifluoroethanol. Figures shown are percentages by volume in water; 80T20E
indicates T–E 80:20 (v/v) and likewise for 60T40E and 40T60E.
c
Extrapolated from the data at lower temperatures (see text).
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 750–757

7
52
K.-T. LIU, P. Y.-H. HWANG AND H.-I. CHEN
Table 2. Correlation analyses using the single-parameter
equation [Eqn. (1)]
a
b
c
d
Substrate Parameter n (solvent)
R
m (SD )
2
3
4
5
YCl
17 (All)
3 (AEM)
(TE)
17 (All)
3 (AEM)
(TE)
17 (All)
3 (AEM)
(TE)
18 (All)
4 (AEM)
(TE)
18 (All)
4 (AEM)
(TE)
18 (All)
4 (AEM)
(TE)
13 (All)
0 (AEM)
(TE)
13 (All)
0 (AEM)
(TE)
13 (All)
0 (AEM)
(TE)
17 (All)
3 (AEM)
(TE)
17 (All)
3 (AEM)
(TE)
17 (All)
5 (AEM)
(TE)
0.968 0.553 (0.037)
0.871 0.531 (0.090)
0.990 0.448 (0.037)
0.957 0.532 (0.042)
0.987 0.666 (0.033)
0.996 0.470 (0.024)
0.968 0.533 (0.036)
0.993 0.657 (0.024)
0.997 0.530 (0.025)
0.694 0.254 (0.066)
0.794 0.320 (0.071)
0.886 0.201 (0.061)
0.708 0.260 (0.065)
0.962 0.445 (0.037)
0.910 0.215 (0.057)
0.737 0.278 (0.064)
0.982 0.448 (0.025)
0.912 0.243 (0.030)
0.831 0.653 (0.112)
0.764 0.517 (0.163)
0.998 0.620 (0.055)
0.970 0.633 (0.048)
0.973 0.720 (0.061)
0.999 0.583 (0.016)
0.989 0.650 (0.029)
0.996 0.715 (0.022)
0.999 0.663 (0.016)
0.846 0.355 (0.058)
0.870 0.414 (0.071)
0.992 0.314 (0.023)
0.832 0.364 (0.063)
0.994 0.594 (0.020)
0.996 0.328 (0.017)
0.842 0.391 (0.065)
0.992 0.600 (0.018)
0.996 0.370 (0.066)
1
5
YBnCl
YxBnCl
YCl
1
5
1
5
1
5
Figure 1. Plots of log k for 2 against YxBnCl. Data points for
(^) aqueous ethanol (E), (&) aqueous acetone (A), (~)
aqueous methanol (M) and (*) tri¯uoroethanol±ethanol (TE)
YBnCl
YxBnCl
YCl
1
5
1
employed. Therefore, the data in Tables 2 and 3 suggest
the superiority of using YxBnCl over other Y values in the
correlation analysis for the solvolysis of naphthoyl
chlorides.
There are three possible pathways generally consid-
ered for the solvolysis of acyl halides, namely the
5
1
4
YBnCl
YxBnCl
1
4
unimolecular S 1-type dissociation [Eqn. (3)], the bimol-
N
1
ecular synchronous S 2-type reaction [Eqn. (4)] and the
N
4
stepwise addition–elimination reaction [Eqn. (5)], as
YCl
21
have been discussed.
1
5
‡
SOH
YBnCl
‡
RCOXÀ! RC ˆˆ O À! RCOOS ‡ H
ꢀ3†
ꢀ4†
1
‡
À
5
X
YxBnCl
z
À
‡
RCOX ‡ SOHÀ! ‰ TSŠ À! RCOOS ‡ X ‡ H
1
5
À
O
j
a
b
c
d
À
‡
Number of data points.
Abbreviations of solvents as in Table 1.
Correlation coefficient.
RCOX ‡ SOHÀ! RCX À! RCOOS ‡ X ‡ H
ꢀ5†
j
SOH
Standard deviation.
‡
Previous work on the mechanism of solvolysis for
benzoyl chlorides (1) indicated a dependence of the
reaction pathways on the nature of the substitutents,
based on the results of Grunwald–Winstein-type correla-
tion analyses [Eqns (1) and (2)], Hammett-type correla-
tion analysis [Eqn. (6)] and ab initio calculations. The
most reactive and sterically hindered 2,6-dimethyl-
benzoyl chloride (1a) was found to solvolyze with a
1
9
solvent nucleophilicity scale N
was found to be
OTs
20
superior to NT for substrates bearing anionic leaving
groups, such as in the solvolysis of various benzylic and
benzoyl chlorides, only the former scale was used. A
few rate data could not be included owing to the shortage
of N values for the corresponding solvents. Nevertheless,
12
2c
2
c
12
it still covers a wide range of both Y (5.76 for Y
.83 for YBnCl ) and N (3.00 for NOTs ) values, and has
and
xBnCl
11
19
5
limiting S 1 mechanism [Eqn. (3)], whereas the less
N
1
1–15 data points, which make the outcome of a dual-
reactive 2-methyl- (1b), 4-methoxy- (1c) and 4-methyl-
parameter regression analysis acceptable. The results are
given in Table 3.
benzoyl chloride (1d) proceed via an S 1 mechanism
N
with a significant extent of nucleophilic solvent inter-
2c
Table 3 clearly indicates that an excellent correlation
vention. The unsubstituted benzoyl chloride (1e) was
considered to solvolyze with a mechanism at the
borderline of unimolecular dissociation [Eqn. (3)] and
1
6
(
R ꢁ 0.99) can be found only in the case of 4 if the
parameters Y and NOTs were applied. A poor
xBnCl
2c
correlation (R < 0.90) is realized for 3, whereas the
the addition–elimination [Eqn. (5)]. However, for
benzoyl chlorides containing a deactivating substitutent,
16
correlations are satisfactory (R = 0.95–0.98) for 2 and 5
no matter whether the YBnCl or YxBnCl scale was
such as 4-chloro- (1f) and 4-nitro- (1g), an S 2-like
N
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 750–757

SOLVOLYSES OF NAPHTHOYL CHLORIDES
753
Table 3. Correlation analyses using the dual-parameter equation [Eqn. (2)]
a
b
b
Substrate
Parameter
n
R
m (SD )
l (SD )
2
YBnCl,NOTs
YxBnCl,NOTs
YBnCl,NOTs
YxBnCl,NOTs
YBnCl,NOTs
YxBnCl,NOTs
YBnCl,NOTs
YxBnCl,NOTs
14
14
15
15
11
11
15
15
0.986
0.984
0.850
0.844
0.985
0.995
0.964
0.957
0.656 (0.044)
0.648 (0.046)
0.454 (0.082)
0.454 (0.082)
0.745 (0.055)
0.733 (0.031)
0.587 (0.051)
0.568 (0.066)
0.335 (0.069)
0.241 (0.068)
0.505 (0.131)
0.443 (0.125)
0.432 (0.117)
0.269 (0.062)
0.500 (0.081)
0.403 (0.099)
3
4
5
a
Number of data points.
b
Standard deviation.
2
c
mechanism [Eqn. (4)] was found in nucleophilic solvents
and an addition–elimination mechanism [Eqn. (5)] in
trifluoroethanol–ethanol. Similar variation of mechan-
isms would be expected to proceed in the solvlolysis of
naphthoyl chlorides.
the literature also exhibit scattered points (Fig. 2). On
the other hand, a poor correlation (R = 0.924) was
observed for 1e in all solvents by using the dual-
parameter equation [Eqn. (2)] against YBnCl and NOTs.
2c
25
2
c
Furthermore, the plot of logk(3) vs logk(1e) gave
excellent linear relationships (R = 0.997) only in aqueous
acetone, ethanol and methanol (Fig. 3), and showed
significant deviations in 100T and 80T20E. Therefore,
the solvolysis mechanism of 3 in nucleophilic solvents
might be close to that of benzoyl chloride (1e), at the
borderline of unimolecular dissociation [Eqn. (3)] and
the addition–elimination process [Eqn. (5)], as was
suggested. The observed acceleration of solvolysis
rates for 3 compared with 1e in weakly nucleophilic
solvents (100T and 80T20E, Fig. 3) indicates that the
mechanism is closer to the unimolecular dissociation in
the case of 3 than 1e.
Table 1 indicates that 6-methoxy-2-naphthoyl chloride
(5) is slightly more reactive than the unsubstituted
substrate 3 with a rate ratios k(5)/k(3) of 1.3–4.8.
However, large rate enhancements, 100–1000-fold, due
to increasing resonance stabilization in the cationic
transition state by the 6-methoxy group, were found in
the solvolysis of the corresponding a-tert-butyl(6-meth-
‡
logꢀk =k † ˆ ꢀ ꢁ
ꢀ6†
X
H
Among the four substrates investigated, 2-naphthoyl
chloride (3) is the least reactive, as shown from Table 1.
A comparison of its rate constants with those for benzoyl
2c
chlorides indicates that the reactivity lies between those
2c
of 1d and 1e in the more ionizing solvents, i.e. 70E, 60E,
8
0M, 70M, 70A to 50A and 100T to 40T60E, in line with
‡
22
23
the trend of s constants (À0.311 for CH , À0.126,
3
24
À0.135 or À0.18 (K.-T. Liu, unpublished data) for 2-
naphthyl and zero for H). Although the result of
correlation analysis using the single-parameter equation
[
Eqn. (1)] showed similar behavior in nucleophilic
2c
solvents for 1d (m = 0.662, R = 0.990 against YBnCl
)
and for 3 (m = 0.448, R = 0.982 against YxBnCl, Table 2),
the correlation was different between 1d (R = 0.967
2c
against YBnCl and NOTs) and 3 (R = 0.844 against Y
xBnCl
and NOTs, Table 3) if the rate data in all solvents were
considered and Eqn. (2) was used. Logarithmic plots of
rate data for 3 in the present study against that for 1d in
26
oxy-2-naphthyl)methyl chloride (6) versus the parent a-
tert-butyl(2-naphthyl)methyl chloride (7).
12
‡
The s
Figure 2. Plots of log k for 3 against log k for 1d. Symbols as
in Fig. 1
Figure 3. Plots of log k for 3 against log k for 1e. Symbols as
in Fig. 1
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 750–757

7
54
K.-T. LIU, P. Y.-H. HWANG AND H.-I. CHEN
constant for the 6-methoxy-2-naphthyl group was
estimated as À0.57 (K.-T. Liu, unpublished data).
Probably 5 solvolyzed with a mechanism different from
that for 3, but similar to that for 1c and 1d. Although no
good linear relationship was observed for 5 in the dual-
parameter correlation analysis (Table 3), two separate
lines, one for nucleophilic solvents (AEM) and the other
for poorly nucleophilic solvents (TE), were found in the
single-parameter plots (Table 2, Figure 4). Therefore, the
downward splitting of line for data points obtained in TE
(m = 0.370) from that in AEM (m = 0.600), i.e.
Dm = 0.230, suggests the intervention of nucleophilic
solvents in the ionization process of 5 (see above). The
low m values (ꢂ0.6) in Tables 2 and 3 obtained from both
single-parameter [Eqn. (1)] and dual-parameter [Eqn. (2)]
regression analyses also revealed deviations from an SN1
process. The observation of an excellent linear logk(5)–
logk(1d) plot with R = 0.994 and m = 0.933 (Figure 5),
but a poor correlation for the logk(5)–logk(1e) plot (Fig.
Figure 5. Plots of log k for 5 against log k for 1d. Symbols as
in Fig. 1
6
), provides additional evidence for the similarity of
solvolytic mechanisms between 5 and 1d.
Figure 6. Plots of log k for 5 against log k for 1e. Symbols as
in Figure 1
1.5–5.0 in less ionizing but more nucleophilic solvents,
about 10–15 in others, and the largest (ca 60) in the least
nucleophilic trifluoroethanol. On the other hand, the
solvolysis of secondary 1-naphthylmethyl tosylate (8)
was found to be about 10–25-fold more reactive than the
A comparison of k(2) and k(3) (Table 1) reveals that 1-
naphthoyl chloride (2) is more reactive than 2-naphthoyl
chloride (3) in all solvents employed. The rate ratios are
2
7
2-naphthylmethyl analogue 9 in a variety of solvents.
Both 2 and 3 could therefore be proposed to solvolyze, at
least in part, via an ionization mechanism [Eqn. (3)] in
solvents such as 60E, 70M, 100T and some other TE
mixtures. Indeed, Fig. 1 exhibits a splitting of lines, and
Figure 4. Plots of log k for 5 against YxBnCl. Symbols as in Fig.
1
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 750–757

SOLVOLYSES OF NAPHTHOYL CHLORIDES
755
be suggested. The least reactive 2-naphthoyl chloride (3)
is likely to have a mechanism at the borderline of
unimolecular dissociation [Eqn. (3)] and the addition–
elimination process [Eqn. (5)]. The more reactive 6-
methoxy-2-naphthoyl chloride (5) shows more S 1-like
N
character but is associated with significant nucleophilic
solvent participation. Similarly to 5, S 1-like mechan-
N
isms are involved for 1-naphthoyl chloride (2) and the
most reactive 4-methyl-1-naphthoyl chloride (4) and with
decreasing extent of nucleophilic solvent participation
from 5 to 2 to 4, in line with the observed Dm
(
mAEM À m ) of log k–Y
plots discussed pre-
TE
xBnCl
viously.
Figure 7. Plots of log k for 4 against YxBnCl. Symbols as in Fig.
1
Furthermore, the present results demonstrate a further
example that the observation of excellent linear correla-
tions in the Grunwald–Winstein-type correlation analysis
using the YxBnCl scale [Eqn. (1)] and YxBnCl and NOTs
scales [Eqn. (2)] as solvent parameters could be regarded
as a criterion for the elucidation of solvolysis mechan-
isms for naphthoyl chlorides 2–5, in addition to other
Table 3 shows a fairly good linear relationship
(R = 0.984) from the dual-parameter correlation with
YxBnCl and NOTs in the case of 2. In other words, 2 is
1
2,13b,c
likely to solvolyze via S 1 mechanisms involving a
systems already reported, such as benzhydryl,
9-
N
12,13a,30
11c
certain extent of nucleophilic solvent intervention,
although it could be partially hindered by the peri-
hydrogen at C-8.
fluorenyl
and
N,N-diphenylcarbamoyl.
Although Kevill and co-workers proposed the use of
the aromatic ring parameter I together with Y and NT
X
Table 1 shows a small increment of the solvolytic
reactivity, 2–8, for 4-methyl-1-naphthoyl chloride (4)
over that for 1-naphthoyl chloride (2). The logk–YxBnCl
plot gave two lines in both cases (Figs 1 and 7), and Table
scales [Eqn. (7)] for studying the solvolytic behavior of
31
benzylic substrates, its inferiority compared with Y
BnX
2c,17a
32
for the solvolysis of benzylic
and benzoyl
derivatives has already been shown in our previous
work. Utilization of YxBnX was also found to be superior
2
reveals a smaller difference in the slopes for 4
(
mAEM À m = 0.052) than that for 2 (0.127). Obviously,
to I and Y in benzhydryl and 9-fluorenyl solvo-
TE
X
2,13,27,33
1
the introduction of a 4-methyl substituent gives rise to
increased stability of the cationic intermediate, probably
due to a resonance effect (10) (Scheme 1), and thus a
decreased extent of nucleophilic solvent intervention in
solvolysis. Since good to excellent linear relationships in
the dual-parameter correlation [Eqn. (2)] were observed
for 4 (R = 0.995) and 2 (R = 0.984), and the NOTs1scale
lyses.
It is therefore desirable to compare these
two approaches in the solvolysis of naphthoyl chlorides.
A regression analysis using Eqn. (7), for example,
yielded less satisfactory results for 2 [n = 15 and
R = 0.955, Eqn. (8)] and 4 [n = 12 and R = 0.981, Eqn.
(9)] than those listed in Table 3. The small magnitudes of
h and their large difference (0.767 vs 1.26) also seem to
be unreasonable for interpreting the contribution of the
naphthalene ring.
7
was defined from the reaction of methyl tosylate, the
intervention of nucleophilic solvents in this study would
28
probably be a kind of ‘participation’ but not ‘nucleo-
29
philic solvation.’ A discussion on the criteria for
nucleophilic solvent participation will be reported else-
where.
logꢀk=k † ˆ mY ‡ lN ‡ hI
ꢀ7†
0
logꢀk=k † ˆ 0:558 Y ‡ 0:670 N ‡ 0:767I ꢀ8†
0
Cl
T
Accordingly, a spectrum of the change in solvolytic
mechanisms for the four naphthoyl chlorides 2–5 could
logꢀk=k † ˆ 0:697 Y ‡ 0:636 N ‡ 1:26I
ꢀ9†
0
Cl
T
Scheme 1
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 750–757

7
56
K.-T. LIU, P. Y.-H. HWANG AND H.-I. CHEN
À2
À2
CONCLUSION
k(0°C) = 1.15 Â 10 , in 80M, k(À10°C) = 1.41 Â 10 ,
À2
À2
k(À5°C) = 2.81 Â 10
and k(0°C) = 4.26 Â 10 , in
À2
À2
From single- and dual-parameter Grunwald–Winstein-
type correlation analyses with YxBnCl, or YxBnCl and NOTs
scales, the solvolytic mechanisms for naphthoyl chlorides
80T20E k(À10°C) = 2.58 Â 10 , k(À5°C) = 4.19 Â
À2
10
and k(0°C) = 8.20 Â 10 , and in 60T40E k
À3
À2
(À5°C) = 8.20 Â 10 ,
k(0°C) = 1.47 Â 10
and
À2
2–5 could be deduced. Along with the increasing trend of
k(5°C) = 2.35 Â 10 . These data were extrapolated to
25°C by the use of an Arrhenius plot. The results at 25°C
are summarized in Table 1.
reactivity from 3 to 5 to 2 and to 4, the mechanism
changes from that at the borderline of S 1-like unimol-
N
ecular dissociation [Eqn. (3)] and the addition–elimina-
tion process [Eqn. (5)] for 3, to a more S 1-like route and
N
involving significant nucleophilic solvent intervention
for 5. A purely unimolecular process is associated with a
decreasing extent of such a participation for 2–4.
Substituent effects enhancing resonance stabilization of
the cationic transition state might be responsible for the
increasing reactivity and the changing mechanisms. The
present results also demonstrate the wide applicability
and the advantage of using the YxBnCl scale to elucidate
the mechanism of solvolysis for different types of
substrates.
Acknowledgements
We are indebted to the National Science Council,
Republic of China, for financial support of this research.
REFERENCES
1
. (a) Patai S (ed). The Chemistry of Acyl Halides. Wiley: Chichester,
972; (b) Patai S (ed). The Chemistry of Acid Derivatives,
Supplements B, vol. 2. Wiley: Chichester, 1992; (c) Katrizky AR,
Meth-Cohn O, Rees CW (eds). Comprehensive Organic Func-
tional Groups Transformations, vol. 5. Elsevier Science: Oxford,
995, Section 5.01.
. (a) Chao I, Chen K-W, Hwang T-S, Liu K-T. J. Phys. Org. Chem.
1
1
EXPERIMENTAL
2
2
4
8
001; 14: 591; (b) Garcia-Riu L, Leis JR. Chem. Commun. 2000;
55; (c) Liu K-T, Chen H-I. J. Chem. Soc., Perkin Trans. 2 2000;
93; (d) Koo IS, Kim J-S, An SK, Yang K, Lee I. J. Korean Chem.
Spectra. Proton and carbon-13 NMR spectra were
recorded on a Bruker Model DMX-300 instrument. IR
spectra were measured on a Nicolet MANGA-IR 550
spectrometer.
Soc. 1999; 43: 527; Chem. Abstr. 2000; 132: 107619e; (e) Koo IS,
Lee SI, An SK, Yang K, Lee I. Bull. Korean Chem. Soc. 1999; 20:
1
451; Chem. Abstr. 2000; 132: 236663p.
3
4
. (a) Kodomari M, Nagaoka T, Furusawa Y. Tetrahedron Lett. 2001;
4
2: 3105; (b) Kabalka GW, Malladi RR, Tejedor D, Kelley S.
Materials. Spectral-grade or reagent-grade solvents
Tetrahedron Lett. 2000; 41: 1001.
. (a) Hudson RF, Wardill JE. J. Chem. Soc. 1950; 1729; (b) Hudson
RF, Archer BJ, J. Chem. Soc. 1950; 3259; (c) Hudson RF, Brown
DA. J. Chem. Soc. 1953; 883.
(
Merck)3 were purified following conventional
4
methods for kinetic studies. Doubly deionized water
was used to prepare aqueous solvent mixtures for
solvolytic studies. Commercially available naphthoyl
chlorides (2 and 3) were purchased from TCI. 4-Methyl-
5. Grunwald E, Winstein S. J. Am. Chem. Soc. 1948; 70: 846.
6
7
8
. Swain CG, Mosely RB. J. Am. Chem. Soc. 1955; 77: 3727.
. Bentley TW, Koo IS, Norman SJ. J. Org. Chem. 1991; 56: 1604.
. (a) Bentley TW, Carter GE, Harris HC. J. Chem. Soc., Perkin
Trans. 2 1985; 983; (b) Bentley TW, Koo IS. J. Chem. Soc., Perkin
Trans. 2 1989; 1385; (c) Bentley TW, Shim CS. J. Chem. Soc.,
Perkin Trans. 2 1993; 1659; (d) Bentley TW, Shim CS. J. Chem.
Soc., Perkin Trans. 2 1993; 2351.
1
-naphthoyl chloride (4) was synthesized from 1-bromo-
4
-methylnaphthlene (Aldrich) to 4-methylnaphthalene-1-
35
carboxylic acid and then by treating with thionyl
chloride. 6-Methoxy-2-naphthoyl chloride (5) was
36
9. Bentley TW, Llewellyn G. Prog. Phys. Org. Chem. 1990; 17: 121.
0. Bentley TW, Harris HC. J. Chem. Soc., Perkin Trans. 2 1986; 619.
11. (a) Liu K-T, Sheu H-C. J. Org. Chem. 1991; 56: 3021; (b) Liu K-T,
Chang L-W, Yu D-G, Chen P-S, Fan J-D. J. Phys. Org. Chem.
1997; 10: 879; (c) Liu K-T, Chen H-I, Lin Y-S, Jin B-Y. J. Phys.
Org. Chem. 2000; 13: 322.
2. Liu K-T, Lin Y-S, Tsao M-L. J. Phys. Org. Chem. 1998; 11: 223.
3. (a) Liu K-T, Lin Y-S, Tsao M-L. Org. React. 1997; 31: 59; (b) Liu
K-T, Chuang C-S. J. Chin. Chem. Soc. 2001; 48: 1047; (c) Liu
K-T, Chuang C-S, Jin B-Y. J. Phys. Org. Chem. 2002; 14: 21.
1
prepared from the corresponding acid (Aldrich) and
37
thionyl chloride. The IR and NMR spectra were found
to be in accord with the assigned structures. All chlorides
were freshly purified prior to kinetic measurements.
1
1
Kinetic measurements. Rate constants were measured
by a conductimetric method at least in duplicate. The
À4
14. Winstein S, Grunwald E, Jones HW. J. Am. Chem. Soc. 1951; 73:
700.
conductivity cells containing solution of about 1 Â 10 –
2
À5
1
 10 M were placed in a thermostat with a tempera-
15. Yoon SK, Uhm TS, Sung DD. J. Korean Chem. Soc. 1980; 24: 347.
1
1
6. Jaffe HH. Chem. Rev. 1953; 53: 191.
ture variation of Æ0.02°C. The error for the measurement
of k was Æ3%.
7. (a) Liu K-T, Hou I-J. Tetrahedron 2001; 57: 2343; (b) Liu K-T,
Duann Y-F, Hou S-J. J. Chem. Soc., Perkin Trans. 2 1998; 2181;
(c) Bentley TW, Llewellyn G, Ryu ZH. J. Org. Chem. 1998; 63:
Most of the rate constants were monitored at 25°C. For
those measured at low temperatures the values are as
4
654.
À1
À2
18. Liu K-T, Chen P-S. J. Chem. Res. (S) 1994; 310.
follows (in s ): for 2 in 100T k(À10°C) = 2.53 Â 10 ,
1
2
9. Fujio M, Susuki T, Goto M, Tsuji Y, Yatsugi K, Saeki Y, Kim SH,
Tsuno Y. Bull. Chem. Soc. Jpn. 1994; 67: 2233.
0. Kevill DN, Anderson SW. J. Org. Chem. 1991; 56: 1845.
21. Kivinen A. In The Chemistry of Acyl Halides, Patai S (ed). Wiley:
Chichester, 1972; 177–230.
22. Hansch C, Leo A, Taft RW. Chem. Rev. 1991; 91: 165.
À2
À2
k(À5°C) = 3.44 Â 10 and k(0°C) = 6.70 Â 10 , and
À3
À2
in 70M k(À5°C) = 9.58 Â 10 , k(0°C) = 1.79 Â 10
À2
and
k(5°C) = 3.33 Â 10 ;
for
4
in
90M
and
À3
À3
k(À10°C) = 3.46 Â 10 , k(À5°C) = 6.30 Â 10
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 750–757

SOLVOLYSES OF NAPHTHOYL CHLORIDES
757
2
3. Allen AD, Girdhar R, Jansen MP, Mayo JD, Tidwell TT. J. Org.
Chem. 1986; 51: 1324.
6303; (b) Kevill DN, D’Souza MJ. J. Chem. Soc., Perkin Trans. 2
1995; 973.
2. Liu K-T. J. Chem. Soc., Perkin Trans. 2 1996; 327.
3. Liu K-T, Lin Y-S. Tetrahedron Lett. 1997; 38: 1419.
4. Armarego WLF, Perrin DD. Purification of Laboratory Chemicals
2
4. Tsuji Y, Fujio M, Tsuno Y. Bull. Chem. Soc. Jpn. 1990; 63: 856.
5. Chen H-I. PhD Thesis, National Taiwan University, 1999, Chapt.
3
3
3
2
3.
2
2
2
6. Lin Y-S. PhD Thesis, National Taiwan University, 1997, Chapt. 3.
7. Liu K-T, Hsu H-Y, Yang J-S. Tetrahedron Lett. 1992; 33: 3327.
8. Bentley TW, Bowen CT, Parker W, Watt CIF. J. Am. Chem. Soc.
(
4th edn). Butterworth-Heinemann: London, 1996.
3
3
5. Blatt AH (ed). Org. Synth., Coll. Vol. 1943; 2: 425.
6. Eissenstat MA, Bell MR, D’Ambra TE, Alexander EJ, Daum SJ. J.
Med. Chem. 1995; 38: 3094.
7. (a) Fries K, Schimmelschmidt K. Chem. Ber. 1925; 58B: 2835; (b)
Leigh WJ, Frendo DT, Klawunn PJ. Can. J. Chem. 1985; 63: 2131.
1
979; 101: 2486.
2
3
3
9. Richard JP, Toteva MM, Amyes TL. Org. Lett. 2001; 3: 2225.
3
0. Liu K-T, Lin Y-S. J. Chin. Chem. Soc. 2000; 47: 71.
1. (a) Kevill DN, Ismail NHJ, D’Souza MJ. J. Org. Chem. 1994; 59:
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 750–757