Use of fluorine kinetic isotope effects in the study of steric effects in nucleophilic aromatic substitution reactions

Article abstract of DOI:10.1021/jo981301i

Leaving group fluorine kinetic isotope effects (F KIEs) have been determined in order to probe the influence of steric effects on the mechanism of the nucleophilic aromatic substitution (S(N)Ar) reaction of 2,4- dinitrofluorobenzene (DNFB) with 2- and 4-m

Full text of DOI:10.1021/jo981301i

9348
J . Org. Chem. 1998, 63, 9348-9350
Use of F lu or in e Kin etic Isotop e Effects in th e Stu d y of Ster ic
Effects in Nu cleop h ilic Ar om a tic Su bstitu tion Rea ction s
J onas Persson and Olle Matsson*
Department of Organic Chemistry, Institute of Chemistry, Uppsala University
P.O. Box 531, S-751 21 Uppsala, Sweden
Received J uly 6, 1998
Leaving group fluorine kinetic isotope effects (F KIEs) have been determined in order to probe the
influence of steric effects on the mechanism of the nucleophilic aromatic substitution (SNAr) reaction
of 2,4-dinitrofluorobenzene (DNFB) with 2- and 4-methylaniline, respectively. The 18F/19F KIE
using isotopically labeled substrate was determined to be 1.0005 ( 0.0030 for 4-methylaniline and
1.0119 ( 0.0037 for 2-methylaniline in DMSO at 30 °C. The large F KIE for 2-methylaniline
suggests rate-limiting leaving group departure for this sterically more hindered nucleophile, whereas
the insignificant F KIE for the less sterically hindered 4-methylaniline indicates rate-limiting
addition of the nucleophile.
Sch em e 1
The mechanism of SNAr reactions has been a subject
of discussion over the years.1 The influence of solvent,
nucleophile, and leaving group are some of the factors
that have been studied in the mechanistic investigations.
The generally accepted mechanism for nucleophilic aro-
matic substitution reactions of substrates activated by
electron-withdrawing groups is an addition-elimination
mechanism first proposed by Bunnett and Zahler.2 This
mechanism involves the formation of a Meisenheimer
type intermediate (Scheme 1).
has been detected by the leaving group F KIE probe.5
We therefore thought it would be of interest to see if the
reported3 change in the rate-limiting step induced by
changing the steric properties of the nucleophile (see
Scheme 2) could be confirmed by determination of the F
KIEs.
The presence or absence of base catalysis, exerted by
the nucleophile itself or by an externally added base, has
played an important role in deciding whether formation
or decomposition of the intermediate complex is rate
limiting.1a Most data available concerns electronic ef-
fects, whereas, the influence of steric effects on the
mechanism are less well investigated.
Resu lts
The synthesis of the DN[18F]B has been reported
earlier.4,5 The previously described kinetic method for
determination of 18F/19F KIEs which is based on HPLC
fractionation and liquid scintillation counting was fol-
lowed with some exceptions.
The average value of the KIE obtained from three
kinetic experiments performed in DMSO at 30 °C was
1.0005 ( 0.0030 for 4-methylaniline and 1.0119 ( 0.0037
for 2-methylaniline. The results from the individual
experiments performed are shown in Table 1.
However, Onyido and Hirst have employed 2-methyl-
aniline and 4-methylaniline as nucleophiles in the reac-
tion with 2,4-dinitrofluorobenzene (DNFB) in DMSO
(Scheme 2). From studies of base catalysis they con-
cluded that a change in the rate-limiting step was
induced by the steric effect of the o-methyl group.3 The
rate of reaction was reduced by a factor of 198 when the
position of the methyl substituent was changed from para
to ortho in the nucleophile.3
Recently we introduced fluorine kinetic isotope effects
(KIEs) as a new mechanistic tool utilizing the accelerator-
produced short-lived radionuclide 18F in combination with
the naturally occurring isotope 19F.4 Furthermore, a
solvent-induced shift in the rate-limiting step for the
reaction of DNFB with piperidine in THF or acetonitrile
Discu ssion
The significant F KIE (1.0119) observed for the reaction
between DNFB and the sterically more hindered nucleo-
phile 2-methylaniline shows that expulsion of the nucleo-
fuge is at least partially rate-limiting. This is in ac-
cordance with the conclusion based on the observation
of a small but distinct curvlinear dependence of reaction
rate on base concentration observed by Onyido and Hirst
for 2-methylaniline.3 A larger value of the F KIE (1.0262)
for the reaction of DNFB with piperidine in THF was
reported earlier.4 This value is close to the maximal
value of 1.032 estimated for total loss of C-F stretching
* To whom correspondence should be addressed. e-mail: ollem@
kemi.uu.se; Fax: +46 18 512524, phone: +46 18 471 3797.
(1) For reviews see: (a) Bernasconi, C. F. MTP International Reviews
of Science, Series 1; pp 33-63; Butterworths: London, 1973, Vol. 3.
(b) Terrier, F. Nucleophilic Aromatic Displacement. The Influence of
the Nitro Group; VCH Publishers: Weiheim, 1991. (c) Vlasov, V. M.
J . Fluor. Chem. 1993, 61, 193-216.
(2) Bunnett, J . F.; Zahler, R. E. Chem. Rev. 1951, 49, 275.
(3) Onyido, I.; Hirst, J . J . Phys. Org. Chem. 1991, 4, 367-371.
(4) Matsson, O.; Persson, J .; Axelsson, B. S.; Långstro¨m, B. J . Am.
Chem. Soc. 1993, 115, 5288-5289.
(5) Persson, J .; Axelsson, S.; Matsson, O. J . Am. Chem. Soc. 1996,
118, 20-23.
10.1021/jo981301i CCC: $15.00 © 1998 American Chemical Society
Published on Web 11/13/1998

Steric Effects in Nucleophilic Aromatic Substitution
J . Org. Chem., Vol. 63, No. 25, 1998 9349
Sch em e 2
Ta ble 1. Lea vin g Gr ou p F KIEs for th e Rea ction of
DNF B (4 × 10-3 M) w ith 2-Meth yla n ilin e a n d
based on the absence of any detectable base catalysis is
confirmed by the observed leaving group F KIE.
4-Meth yla n ilin e, Resp ectively, in DMSO a t 30 °C. Th e
What is the cause of this change to partially rate-
limiting leaving group expulsion for the sterically more
hindered 2-methylaniline? Steric effects may in principle
be reflected in all elementary reaction rate constants in
Scheme 1. Bernasconi and de Rossi7 have discussed the
importance of steric effects in determining the different
behavior of primary and secondary amines as nucleo-
philes in SNAr reactions. For the present reaction
system, some suggestions were given by Onyido and
Hirst.3 Steric compression in the intermediate may, for
bulky nucleophiles, be relieved by reversion to reactants.
Such a release of steric strain enhances k-18 which tends
to make the addition step less rate limiting (eq 1). The
TS for formation of the intermediate is the same as for
its reversal to reactants and, according to the Hammond
postulate, it is structurally close to the intermediate.
Therefore a reduction of the rate constant for the addition
of the nucleophile to the aromatic substrate, k1, is
expected for formation of a more sterically compressed
intermediate. Steric factors may in principle also affect
the rate for decomposition of the intermediate to prod-
ucts, although it is difficult to estimate the relative effect
of the ortho substituent on the free energy of the
intermediate and the TS for its expulsion of the fluoride.
Thus it seems that the main effect of moving the methyl
substituent from the para to the ortho position is an
increased rate constant for reversal of the intermediate
to starting materials. The significant F KIE observed
for 2-methylaniline as compared to 4-methylaniline
clearly demonstrates a change to rate-limiting nucleofuge
detachment for the sterically more hindered nucleophile.
Con cen tr a tion of An ilin e w a s ca . 5 × 10-3
M
KIE ( std dev for
4-methylaniline
KIE ( std dev for
2-methylaniline
1.0021 ( 0.0041(4)a
0.9983 ( 0.0061(5)a
1.0011 ( 0.0056(3)a
1.0105 ( 0.0053(4)a
1.0110 ( 0.0059(5)a
1.0141 ( 0.0078(5)a
a
The number of kinetic points in each experiment.
vibrational zero point energy in the TS.4 The smaller
value for the 2-methylaniline in DMSO as compared to
piperidine in THF is understandable in terms of the
steady-state rate expression (eq 1) for the mechanism in
k1(k2 + k3[B])
kA )
(1)
k-1 + k2 + k3[B]
Scheme 1. For a fully rate-limiting decomposition of the
intermediate under conditions where a linear dependence
of reaction rate on base concentration is observed (k-1
.
k2 + k3[B]; observed for piperidine in THF at moderate
base concentration6) the observed KIE (k18/k19)obs is
identical to the primary KIE for the expulsion step times
the equilibrium isotope effect for formation of the inter-
mediate. For 2-methylaniline, however, which exhibits
a curvlinear dependence of rate on base concentration,
no simplification of the rate expression is possible (k-1
= k2 + k3[B]), and the observed rate constant is given by
eq 1. The concentration of base used in this investigation
is low (ca. 5 × 10-3 M) and corresponds to the nearly
linear part of the kA vs [B] plot (see Figure 1 in ref 3).
Thus decomposition of the intermediate is expected to
be rate limiting. Assuming a negligible secondary isotope
effect on k1, and a normal (>1) KIE on the decomposition
of the intermediate (k2 + k3[B]), the complex kinetic
situation implies that the observed F KIE will be attenu-
ated as compared to the actual KIE on the decomposition
step. Therefore, a smaller observed F KIE is expected
for 2-methylaniline than for piperidine. The magnitude
of the primary F KIE for the decomposition of the
intermediate will, of course, also be affected by the
position of the TS along the reaction coordinate; an
earlier TS yielding a smaller KIE.
Exp er im en ta l Section
The HPLC analyses were performed on a Beckman HPLC
with a â+-flow detector in series with the UV-detector of the
instrument. The HPLC was equipped with an injector/fraction
collector (Gilson). The HPLC analyses were performed on a
column, 200 × 4.6 mm, packed with Nucleosil RP C-18, 5 µm.
The mobile phase was water and acetonitrile 47:53 (v:v),
isocratic flow 2.00 mL/min. The wavelength used was 254 nm
using 430 nm as a reference.
The radioactive HPLC fractions (usually 4 mL) were col-
lected in scintillation bottles containing 15 mL of scintillation
liquid (Zinsser Quickzint 1). The radioactive counting was
performed using a liquid scintillation counter Beckman LL
6000L. The counting time was usually 1 min.
The F KIE for the reaction of 4-methylaniline is
virtually nil and is thus consistent with rate-limiting
addition of the nucleophile to the substrate (k-1 , k2 +
k3[B]; kA ) k1). In this case the observed F KIE is equal
to the KIE for the addition step, which is a secondary
one and therefore expected to be very small for such a
heavy element as fluorine. Again, the earlier conclusion
DMSO (Aldrich Sure-Seal) was used as bought without
further purification. The anilines were purified by repeated
distillation. The distilled 4-methylaniline was recrystallized
from absolute ethanol. The purity (>98%) was determined
(7) Bernasconi, C. F.; de Rossi, R. H. J . Org. Chem. 1976, 41, 44-
49.
(6) Nudelman, N.; Mancini, P. M. E.; Martinez, R. D.; Vottero, L.
R. J . Chem. Soc., Perkin Trans. 2 1987, 951-954.
(8) Bunnett, J . F.; Garst, R. H. J . Am. Chem. Soc. 1965, 87, 3875-
3878.

9350 J . Org. Chem., Vol. 63, No. 25, 1998
Persson and Matsson
1
using H NMR (Varian Unity 400) and GC (Varian 3400 gas
reaction point was performed as described earlier4,5 using the
HPLC UV-detector and liquid scintillation data.
chromatograph). A solution of the desired aniline, 6 × 10-3
M, was prepared and thermostated. When the synthesis of
DN[18F]B was complete, 0.200 mL of the DNFB solution,
containing the DN[18F]B, was transferred via a syringe to the
reaction vial containing 1 mL of the aniline solution. The
reaction clock was started, and the solution was distributed
to 3-6 reaction vials using a thermostated syringe. The vials
were capped and placed in the thermostat. The amount of the
reaction solution distributed to the different vials was deter-
mined by weight. The quenching of a reaction point was
performed by addition of acid to the reaction vial followed by
immediate freezing in liquid nitrogen. After quenching, each
sample was injected five times on the HPLC and the DNFB
fraction was collected in scintillation bottles containing 14 mL
of scintillation liquid. The calculation of the KIE for each
Ackn owledgm en t. We thank Professor Bengt Lång-
stro¨m and the staff at Uppsala University PET-center
for providing the radioactive fluorine and for putting
the excellent facilities at our disposal. We thank Dr.
Svante Axelsson for assistance and Dr. Anita Husse´nius
for valuable comments on the manuscript. Olle Matsson
is indebted to Professor Franc¸ois Terrier, Universite´ de
Versailles, for his hospitality when this manuscript was
written. The project is supported by the Swedish
Natural Science Research Council.
J O981301I