Choice of solvent (MeCN vs H2O) decides rate-limiting step in SNAr aminolysis of 1-fluoro-2,4-dinitrobenzene with secondary amines: Importance of Bronsted-type analysis in acetonitrile

Article abstract of DOI:10.1021/jo701549h

(Chemical Equation Presented) A kinetic study is reported for nucleophilic substitution reactions of 2,4-dinitro-1-fluorobenzene (DNFB) with a series of secondary amines in MeCN and H₂O at 25.0°C. The reaction in MeCN results in an upward curva

Full text of DOI:10.1021/jo701549h

Choice of Solvent (MeCN vs H O) Decides Rate-Limiting Step in
2
SNAr Aminolysis of 1-Fluoro-2,4-dinitrobenzene with Secondary
Amines: Importance of Brønsted-Type Analysis in Acetonitrile
,†
†
,‡
Ik-Hwan Um,* Se-Won Min, and Julian M. Dust*
Department of Chemistry and DiVision of Nano Sciences, Ewha Womans UniVersity,
Seoul 120-750, Korea, and Department of Chemistry and EnVironmental Science, Sir Wilfred Grenfell
College, Corner Brook, Newfoundland and Labrador, A2H 6P9, Canada
jdust@swgc.mun.ca
ReceiVed July 15, 2007
A kinetic study is reported for nucleophilic substitution reactions of 2,4-dinitro-1-fluorobenzene (DNFB)
with a series of secondary amines in MeCN and H O at 25.0 °C. The reaction in MeCN results in an
2
upward curvature in the plot of kobsd vs [amine], indicating that the reaction proceeds through a rate-
limiting proton transfer (RLPT) mechanism. On the contrary, the corresponding plot for the reaction in
H
2
O is linear, implying that general base catalysis is absent. The ratios of the microscopic rate constants
for the reactions in MeCN are consistent with the proposed mechanism, e.g., the facts that k
2
/k-1 < 1
2
and k
and the deprotonation by a second amine molecule becomes dominant when [amine] > 0.01 M,
respectively. The Brønsted-type plots for k /k-1 and k
.84, respectively, which supports the proposed mechanism. The Brønsted-type plot for the reactions in
3 2
/k > 10 suggest that formation of a Meisenheimer complex occurs before the rate-limiting step
1
k
2
1 3
k /k-1 are linear with ânuc values of 0.82 and
0
H
2
O is also linear with ânuc ) 0.52 which has been interpreted to indicate that the reaction proceeds
through rate-limiting formation of a Meisenheimer complex. DNFB is more reactive toward secondary
amines in MeCN than in H O. The enhanced basicity of amines as well as the increased stability of the
2
intermediate whose charges are delocalized through resonance are responsible for the enhanced reactivity
in the aprotic solvent.
,6
Introduction
examples, this substitution has been found useful in synthesis,5
(
7
including improved methods of stereoselective reaction), in
derivatization to extend analytical detection limits, in preparation
of electrophilic derivatives of water-soluble polymers, and in
Nucleophilic aromatic displacement involving electron-
deficient substrates, via the SNAr mechanism,1-4 comprises a
major class of organic transformation. To give but a few
8,9
10-12
some possible environmental remediation protocols.
Fun-
*
Tel.: 82-2-3277-2349; FAX: 82-2-3277-2844 (I.-H.U.). Tel. 1-709-637-
6
200 ext. 6330; FAX: 1-709-639-8125 (J.M.D.).
(4) Terrier, F. Nucleophilic Aromatic Displacement: The Influence of
the Nitro Group; Feuer, H., Ed.; Organic Nitro Chem. Ser.; VCH: New
York, 1991.
†
Ewha Womans University.
Sir Wilfred Grenfell College.
‡
(
1) Bunnett, J. F. J. Chem. Educ. 1974, 51, 312-315 and references
(5) Beugelmanns, R.; Singh, G. P.; Bois-Choussy, M.; Chastanet, J.; Zhu,
J. J. Org. Chem. 1994, 59, 5535-5542.
therein.
(2) Miller, J. Nucleophilic Aromatic Substitution; Elsevier: Amsterdam,
(6) Khalafi-Nezhad, A.; Zare, A.; Parhami, A.; Rad, M. N. S.; Nejabat,
G. R. Can. J. Chem. 2006, 84, 979-985.
1
968.
(3) (a) Buncel, E.; Crampton, M. R.; Strauss, M. J.; Terrier, F. Electron
(7) Snyder, S. E.; Carey, J. R.; Shvets, A. B.; and Pirkle, W. H. J. Org.
Chem. 2005, 70, 4073-4081.
Deficient Aromatic- and Heteroaromatic-Base Interactions; Elsevier: New
York, 1984. (b) Buncel, E. The Chemistry of Functional Groups. Supplement
F. The Chemistry of Amino, Nitro and Nitroso Compounds; Patai, S., Ed.;
Wiley: London, 1982.
(8) Dust, J. M.; Harris, J. M. J. Polym. Sci. Chem. A 1990, 28, 1875-
1886.
(9) Dust, J. M.; Secord, M. D. J. Phys. Org. Chem. 1995, 8, 810-824.
1
0.1021/jo701549h CCC: $37.00 © 2007 American Chemical Society
Published on Web 10/20/2007
J. Org. Chem. 2007, 72, 8797-8803
8797

Um et al.
damental studies of nucleophilic aromatic substitution, therefore,
continue to attract attention.13-17
SCHEME 1
In general, the SNAr mechanism involves a first step in which
attack of a nucleophile on an electron-deficient aromatic or
heteroaromatic either gives an anionic σ-bonded adduct, com-
monly termed a Meisenheimer complex (MC), or proceeds
through a transition-state modeled on the MC. If the position
of nucleophilic attack is suitably 1-X substituted (i.e., to give
MC-1), expulsion of the leaving group in a second step gives
the displacement product. On the other hand, although attack
at an unsubstituted position, to give MC-3 or -5, usually is
unproductive, reaction at such a position with an appropriate
nucleophile that itself bears a leaving group and can undergo
â-elimination with the ring proton provides entry into the
vicarious nucleophilic substitution (VNS)1
8,19
reaction. The
factors that stabilize MC and correspondingly influence regi-
oselectivity in SNAr/VNS displacement form, therefore, one
8,9,20-23
consideration in examining these reaction systems.
Reaction of neutral nucleophiles, such as amines, with
electron-deficient aromatics is at once both more complex and
also more mechanistically interesting, particularly when medium
effects are included in the study. These two themes constitute
the focus of the current paper where 2,4-dinitro-1-fluorobenzene,
Sanger’s reagent, that has proven so useful in labeling protein
residues, is a typical electron-deficient substrate and a series
of structurally related secondary amines comprise the set of
nucleophiles as shown below.
substituted aniline, DNA, or alternatively, base-catalyzed depro-
tonation of MC-1-Z (the k3 step in Scheme 1) to yield a new
amino-Meisenheimer complex, MC-1, that loses fluoride ion
to give the same new aniline product. (An alternative pathway
(not shown in Scheme 1) involves proton transfer from nitrogen
to fluorine in MC-1-Z, with the possible intermediacy of solvent
to relay the proton, leading to expulsion of HF and direct
formation of DNA in a single step. (Bernasconi, C. F. MTP
Int. ReV. Sci. Org. Chem. Ser. 1, 1973, 3, 33-63). The amine
concentration-dependent behavior (at higher amine concentra-
tions) found in the current study suggests that although this
process may compete with the k2 path shown in Scheme 1, it
cannot compete with the k3 path under these conditions. The
discussion of the mechanism and, notably, of the importance
of solvent choice in determining the rate-determining step is
otherwise unchanged and this alternative will not be discussed
further. We thank a referee for calling our attention to this
alternative.)
2
4
Here, the putative first-step in the SNAr process (Scheme 1)
leads to formation of a zwitterionic MC-1-Z from which two
competitive processes for decomposition have been postu-
lated: expulsion of the fluoride ion leaving group (k2, where k1
may become rate-limiting) followed by rapid proton loss from
Although in theory a reaction may also occur at the
+
20
the protonated product dinitroaniline, DNAH , to give the new
unsubstituted C-3 and C-5 positions, attack at these sites with
the secondary amines involved in the current study would not
lead to a stable final product, and any Meisenheimer complexes
that would arise from such attack would simply constitute a
reservoir of DNFB and the relevant secondary amine. In water
or in an aprotic polar solvent such as dimethyl sulfoxide
(
10) Fant, F.; De Sloovere, A.; Mathijsen, K.; Marl e´ , C.; El Fantroussi,
S.; Verstraete, W. EnViron. Pollut. 2001, 111, 503-507.
11) Brunelle, D. J.; Singelton, D. A. Chemosphere 1985, 14, 173-181
and references therein.
12) Balakrishnan, V. K.; Dust, J. M.; vanLoon, G. W.; Buncel, E. Can.
J. Chem. 2001, 79, 157-173.
13) Consiglio, G.; Frenna, V.; Guernelli; Macaluso, G.; spinelli, D. J.
Chem. Soc., Perkin Trans 2 2002, 971-975.
14) Crampton, M. R.; Emokpae, T. A.; Isanbor, C. Eur. J. Org. Chem.
007, 1378-1383.
15) (a) Crampton, M. R.; Emokpae, T. A.; Isanbor, C.; Batsanov, A.
S.; Howard, J. A. K.; Mondal, R. E. Eur. J. Org. Chem. 2006, 1222-1230.
b) Asghar, B. H. M.; Crampton, M. R. Org. Biomol. Chem. 2005, 3, 3971-
978.
16) Crampton, M. R.; Emokpae, T. A.; Howard, J. A. K.; Isanbor, C.;
Mondal, R. J. Phys. Org. Chem. 2004, 17, 65-70.
17) (a) Goumont, R.; Terrier, F.; Vichard, D.; Lakhdar, S.; Dust, J. M.;
(
(
(DMSO) or acetonitrile (MeCN), Scheme 1 is sufficient
(
representation of the pathways leading to SNAr displacement,
but the situation is more complex in aprotic nonpolar solvents
such as benzene, ethyl acetate, or tetrahydrofuran25 where
deprotonation of a zwitterionic Meisenheimer complex like MC-
(
2
(
2
6
1
-Z may involve an amine dimer competing with free amine
(
3
27
or two molecules of free amine acting in concert. Mixed ethyl
acetate-chloroform media have been used to probe mechanistic
changeover in 2,6-dinitro-1-fluorobenzene/secondary amine
(
(
2
8
Buncel, E. Tetrahedron Lett. 2005, 46, 8363-8366. (b) Moutiers, G.; Le
Guevel, E.; Cannes, C.; Terrier, F.; Buncel, E. Eur. J. Org. Chem. 2001,
279-3284.
systems.
The importance of solvent choice in determining the nature
of the mechanism of aminolysis with electron-deficient sub-
3
(
(
(
18) Makosza, M.; Sienkiewicz, K. J. Org. Chem. 1990, 55, 4979-4981.
19) Makosza, M.; Kwast, A. Eur. J. Org. Chem. 2004, 2125-2130.
20) For a review: Buncel, E.; Dust, J. M.; Terrier, F. Chem. ReV. 1995,
(25) Hirst, J.; Onuoha, G. N.; Onyido, I. J. Chem. Soc., Perkin Trans. 2
1988, 971-974.
9
5, 2261-2280.
(21) Manderville, R. A.; Dust, J. M.; Buncel, E. J. Phys. Org. Chem.
(26) Akinycle, E. T.; Onyido, I.; Hirst, J. J. Chem. Soc., Perkin Trans.
2 1988, 1859-1861.
1
996, 9, 515-528.
(
22) Dust, J. M.; Manderville, R. A. Can. J. Chem. 1998, 76, 662-671.
(27) Banjoko, O.; Otiono, P. J. Chem. Soc., Perkin Trans. 2 1981, 399-
(23) Buncel, E.; Dust, J. M.; Manderville, R. A.; Tarkka, R. M. Can. J.
402.
Chem. 2003, 81, 443-456.
24) Sanger, F. Biochem. J. 1945, 39, 507-515.
(28) Mancini, P. M. E.; Fortunato, G. G.; Vottero, L. R. J. Phys. Org.
Chem. 2005, 18, 336-346.
(
8798 J. Org. Chem., Vol. 72, No. 23, 2007

Brønsted-Type Analysis in Acetonitrile
strates such as DNFB cannot be overemphasized.29-31 Moreover,
for synthetic utility a solvent should not only promote reactivity
but be readily removed (recycled); in this case, MeCN is to be
preferred over DMSO. Further, a recent calculational study
-
confirmed that the intermediate MC-1 (DNFB + N3 ) to have
enhanced stability in aprotic solvents (MeCN, DMSO) relative
to protic ones (H2O, EtOH).32
There is a similarity between stepwise nucleophilic attack at
CdO of esters and SNAr displacement in that both involve: (1)
2
initial addition with rehybridization of the C-center from sp to
3
sp to give a tetrahedral intermediate and (2) elimination of a
2
leaving group in a second-step to regenerate the sp center. In
both general addition-elimination systems, either the first or
second step may be rate-limiting. However, in SNAr displace-
ment, the process involves loss of aromaticity in step 1 and
rearomatization in step 2; the importance of electron withdraw-
ing substituents that effectively delocalize negative charge in
the MC has been highlighted.1
-4
Brønsted analysis has previously been found to be a useful
tool to determine mechanism in CdO and related ester
3
3-36
17b,37,38
systems,
though less commonly used for SNAr.
We
extend our study now to SNAr aminolysis with DNFB in MeCN,
a synthetically useful solvent. This study is possible now because
pKa values of the secondary amines used in this study in MeCN
have only recently become available.39
FIGURE 1. Plots of kobsd vs [HNRR′] for the reactions of DNFB with
morpholine in MeCN and in H
line for the reaction in MeCN was calculated by eq 1.
2
O (inset) at 25.0 ( 0.1 °C. The solid
The results will be discussed in terms of comparison of
Brønsted-type slope parameters (ânuc) for the reactions in
acetonitrile and water. The utility of the Brønsted analysis in
assigning the mechanism and particularly the rate-limiting step
will be discussed.
substrate concentration. All of the reactions obeyed first-order
kinetics over 90% of the total reaction. No spectroscopic
evidence was found for formation of nonproductive MC-3 or
MC-5 adducts. Pseudo-first-order rate constants (kobsd) were
calculated from the equation ln(A∞ - At) ) -kobsdt + C. It is
estimated from replicate runs that the uncertainty in the rate
constants is less than ( 3 %. The kobsd values with the reaction
conditions are summarized in Tables S1-S12 in the Supporting
Information.
Results and Discussion
The kinetic study was performed under pseudo-first-order
conditions with the concentration of amines in excess over the
(
(
(
29) Parker, A. J. Chem. ReV. 1969, 69, 1-32.
30) Cox, B. G.; Parker, A. J. J. Am. Chem. Soc. 1973, 95, 408-410.
31) Buncel, E.; Stairs, R.; Wilson, H. The Role of the SolVent in
Chemical Reactions; Oxford University Press: Oxford, 2003.
The plot of kobsd vs [HNRR′] for the reaction of DNFB with
morpholine in MeCN curves upward as a function of increasing
amine concentration (Figure 1). A similar result has been
obtained for reactions with all the other amines studied in MeCN
(
32) Acevedo, O.; Jorgensen, W. L. Org. Lett. 2004, 6, 2881-2884.
(33) (a) Jencks, W. P. Chem. ReV. 1985, 85, 511-527. (b) Stefanidis,
D.; Cho, S.; Dhe-Paganon, S.; Jencks, W. P. J. Am. Chem. Soc. 1993, 115,
1
6
6
1
650-1656. (c) Stefanidis, D.; Jencks, W. P. J. Am. Chem. Soc. 1993, 115,
045-6050. (d) Berg, U.; Jencks, W. P. J. Am. Chem. Soc. 1991, 113,
997-7002. (e) Murray, C. J.; Jencks, W. P. J. Am. Chem. Soc. 1990, 112,
880-1889.
(Figures S1-S4 in the Supporting Information). Such upward
curvature is typical for reactions that proceed through a rate-
(
34) (a) Castro, E. A. Chem. ReV. 1999, 99, 3505-3524. (b) Castro, E.
15
limiting proton transfer (RLPT) mechanism. Accordingly, one
A.; Aliaga, M.; Santos, J. G. J. Org. Chem. 2005, 70, 2679-2685. (c) Castro,
E. A.; Aguayo, R.; Bessolo, J.; Santos, J. G. J. Org. Chem. 2005, 70, 3530-
can suggest that the reactions in MeCN proceed through two
central intermediates (a zwitterionic adduct MC-1-Z and its
deprotonated form MC-1) as shown in Scheme 1. In contrast,
the plots for the corresponding reactions in H2O are linear
passing through the origin in all cases (e.g., the inset of Figure
3536. (d) Castro, E. A.; Aguayo, R.; Bessolo, J.; Santos, J. G. J. Org. Chem.
2
005, 70, 7788-7791. (e) Castro, E. A.; Gazitua, M.; Santos, J. G. J. Org.
Chem. 2005, 70, 8088-8092.
(35) (a) Lee, I.; Sung, D. D. Curr. Org. Chem. 2004, 8, 557-567. (b)
Oh, H. K.; Lee, J. M.; Sung, D. D.; Lee, I. J. Org. Chem. 2005, 70, 3089-
3
093. (c) Oh, H. K.; Oh, J. Y.; Sung, D. D.; Lee, I. J. Org. Chem. 2005,
1
for the reaction with morpholine). The linear plot implies that
7
0, 5624-5629. (d) Oh, H. K.; Park, J. E.; Sung, D. D.; Lee, I. J. Org.
Chem. 2004, 69, 3150-3153.
the rate-limiting deprotonation process by a second amine
molecule (i.e., the k3 step in Scheme 1) is absent for the reactions
in H2O.
(36) (a) Um, I. H.; Akhtar, K.; Shin, Y. H.; Han, J. Y. J. Org. Chem.
2
007, 72, 3823-3829. (b) Um, I. H.; Park, Y. M.; Fujio, M.; Mishima, M.;
Tsuno, Y. J. Org. Chem. 2007, 72, 4816-4821. (c) Um, I. H.; Kim, E. Y.;
Park, H. R.; Jeon, S. E. J. Org. Chem. 2006, 71, 2302-2306. (d) Um, I.
H.; Lee, J. Y.; Ko, S. H.; Bae, S. K. J. Org. Chem. 2006, 71, 5800-5803.
Determination of Microscopic Rate Constants. On the basis
of the kinetic result and the mechanism proposed in Scheme 1,
one can express the pseudo-first-order rate constant (kobsd) for
the reactions in MeCN as eq 1, in which [HNRR′] represents
the concentration of amine. Equation 1 can be simplified as eq
(e) Um, I. H.; Shin, Y. H.; Han, J. Y.; Mishima, M. J. Org. Chem. 2006,
7
1, 7715-7720. (f) Um, I. H.; Hwang, S. J.; Baek, M. H.; Park, E. J. J.
Org. Chem. 2006, 71, 9191-9197.
(37) Dixon, J. E.; Bruice, T. C. J. Am. Chem. Soc. 1972, 94, 2052-
2
5
056.
(38) Bordwell, F. G.; Hughes, D. L. J. Am. Chem. Soc. 1986, 108, 5991-
2
under the assumption, k-1 . k2 + k3[HNRR′]. Thus, one can
997.
expect that the plot of kobsd/[HNRR′] vs [HNRR′] is linear if
(39) Spillane, W. J.; McGrath, P.; Brack, C.; O’Byrne, A. B. J. Org.
Chem. 2001, 66, 6313-6316.
the reaction proceeds as in Scheme 1.
J. Org. Chem, Vol. 72, No. 23, 2007 8799

Um et al.
2
kobsd ) (k k [HNRR′] + k k [HNRR′] )/
1
2
1 3
(
k-1 + k + k [HNRR′]) (1)
2 3
kobsd/[HNRR′] ) Kk + Kk [HNRR′], where K ) k /k (2)
-1
2
3
1
In fact, as shown in Figure 2, the plot of kobsd/[HNRR′] vs
HNRR′] is linear for the reaction with morpholine up to ca.
[
0.025 M. The plot for the reaction with 1-formylpiperazine (see
Figure S5 in the Supporting Information) is also linear up to
ca. 0.01 M, indicating that the above assumption is valid for
the reactions with these weakly basic amines in the low
concentration region. However, as shown in the inset of Figure
2
, the plot for the reaction with piperidine exhibits a downward
curvature as the amine concentration increases beyond 0.01 M.
A similar downward curvature is obtained for the reaction
with piperazine (see Figure S6 in the Supporting Information),
indicating that the above assumption is invalid for the reactions
with the strongly basic amines when the concentration of these
amines increases highly. This argument is in accord with the
idea that k-1 decreases with increasing amine basicity, and the
term k3[HNRR′] becomes larger with increasing the concentra-
tion of amines. When the concentration of amines is high
enough, then k2 , k3[HNRR′] and eq 1 can be reduced to eq 3.
As shown in Figure 3, the plot of [HNRR′]/kobsd vs 1/[HNRR′]
for the reaction with piperidine is linear in the region where
the amine concentration exceeds ca. 0.01 M but exhibits a
downward curvature as amine concentration decreases. A similar
result is obtained for the other amines studied, indicating that
the assumption k2 , k3[HNRR′] is valid only when [HNRR′]
FIGURE 2. Plots of kobsd/[HNRR′] vs [HNRR′] for the reactions of
DNFB with morpholine and piperidine (Inset) in MeCN at 25.0 (
0
.1 °C.
>
0.01 M.
[
HNRR′]/kobsd ) 1/k + 1/Kk [HNRR′]
(3)
1
3
Therefore, 1/k1 and 1/Kk3 values have been extracted from
the intercept and slope of the linear part of the curved plots,
respectively. More reliable values of k1, k2/k-1, and k3/k-1 have
been determined through the nonlinear least-squares fitting of
eq 1 to the experimental data by using the 1/k1 and 1/Kk3 values
obtained above as input values. The values of k1, k2/k-1, and
k3/k-1 determined in this way are summarized in Table 1. The
k3/k2 ratios which were calculated from the k3/k-1 and k2/k-1
ratios are also included in Table 1.
As shown in Table 1, the k1 value for the reaction with
piperidine is in good agreement with kN value (e.g., 290-320
-1 -1
M s ) reported previously for the same reaction performed
at 30 °C in MeCN.40 Nudelman et al. found the kN value to be
insensitive to the piperidine concentration. This is because the
concentration of piperidine employed in these studies ranged
4
0
up to 0.0025 M for the reaction at 30 °C or 0.005 M at 15 °C.
In such low concentration regimes, the contribution of the k3-
2
[
HNRR′] term to kobsd should be negligible (see eq 1). In fact,
as shown in Figure S1, the plot of kobsd vs amine concentrations
for the reaction of DNFB with piperidine appears to be linear
up to ca. 0.008 M of piperidine but exhibits an upward curvature
as the concentration of piperidine increases further.
2
It is apparent from Table 1 that k2/k-1 < 1 but k3/k2 > 10
FIGURE 3. Plot of [HNRR’′]/kobsd vs 1/[HNRR′] for the reactions of
DNFB with piperidine in MeCN at 25.0 ( 0.1 °C. Insert highlights
the linear region of the plot.
for reactions with all the amines studied in MeCN. The fact
that k2/k-1 < 1 indicates that formation of a Meisenheimer
complex occurs before the rate-limiting step. Besides, k3/k2 >
2
amine concentration is high enough (e.g., [HNRR′] > 0.01 M).
Thus, the microscopic rate constants determined above can
account for the nonlinear plots shown in Figures 2 and 3 (and
also Figures S5-S6 in the Supporting Information).
1
0 implies that the k3 process becomes dominant when the
(
40) Nudelman, N. S.; Mancini, P. M. E.; Martinez, R. D.; Vottero, L.
R. J. Chem. Soc., Perkin Trans. II 1987, 951-954.
8800 J. Org. Chem., Vol. 72, No. 23, 2007

Brønsted-Type Analysis in Acetonitrile
TABLE 1. Summary of Microscopic Rate Constants for the Reactions of DNFB with Alicyclic Secondary Amines in MeCN at 25.0 ( 0.1 °Ca
amine
pKa
k1/M-1s-1
k2 /k-1
Kk2 / M-1s-1
k3 /k-1 / M-1
Kk3 / M-2s-1
k3 /k2 / M-1
1
2
3
4
5
.
.
.
.
.
piperidine
piperazine
1-(2-hydroxyethyl) piperazine
1-formylpiperazine
morpholine
18.8
18.2
17.6
17.0
16.0
380 ( 3
0.293
0.137
0.182
0.180
0.0400
111
50.3
42.5
32.0
32.3
5.74
19000
16700
1340
359
172
310
176
179
144
394 ( 3
54.0
7.63
2.00
0.712
41.9 ( 1.5
11.1 ( 0.5
17.8 ( 0.7
102
a
The pKa data in MeCN were taken from ref 39.
TABLE 2. Summary of Apparent Second-Order Rate Constants
kN) for the Reactions of DNFB with Alicyclic Secondary Amines in
H2O at 25.0 ( 0.1 °C
By analogy to CdO ester aminolysis that can give a
zwitterionic tetrahedral intermediate analogus to MC-1-Z, one
might expect that k2 is independent of the amine basicity since
there is little or no electron donation from the cationic amine
moiety of MC-1-Z to exert the push to expel the leaving
(
amine
pKa
kN / M-1s-1
1
2.
.
piperidine
piperazine
1-(2-hydroxyethyl) piperazine
morpholine
1-formylpiperazine
piperazinium ion
11.22
9.82
9.38
8.36
7.98
5.68
9.37
5.48
1.43
1.07
0.444
0.0163
4
1,42
group.
However, k3 would be little influenced by amine
3
4
.
.
basicity. This is because a more basic amine tends to deprotonate
MC-1-Z more rapidly, but it becomes a weaker acid in the
zwitterionic intermediate and would hold the proton more
strongly.4 As a result of this compensatory effect, the k3/k2 ratio
is expected to be insensitive to amine basicity. In fact, as shown
in Table 1, the k3/k2 ratio remains nearly constant except for
the reaction with piperazine, which exhibits a larger k3/k2 ratio
than others. The larger k3/k2 ratio obtained for the reaction with
piperazine can be ascribed to the fact that piperazine has two
basic sites to deprotonate. This argument can be further
supported from the linear plot of log k3/k2 vs pKa with a slope
close to zero when k3 and pKa values were statistically corrected
using p (numbers of protons which can be deprotonated from
the conjugate acid of the amine) and q (numbers of nucleophilic
sites of the amine), i.e., p ) 2 (except p ) 4 for piperazinium
ion) and q ) 1 (except q ) 2 for piperazine)44 (Figure S7 in
Supporting Information).
5.
6.
3
from the slope of the linear plots of kobsd vs amine concentration
and summarized in Table 2.
kobsd ) k [HNRR′], where k ) k k /(k + k2) (4)
N
N
1
2
-1
As shown in the Table 2, the kN value for the aqueous
reactions decreases as the basicity of the amines decreases. The
effect of amine basicity on reactivity is illustrated in Figure 5.
The statistically corrected Brønsted-type plot using p and q is
linear, indicating that the reaction proceeds without changing
the rate-limiting step or mechanism on changing the amine
basicity over a range of 5.5 pKa units. Crampton et al. have
recently reiterated that formation of a Meisenheimer complex
Brønsted-Type Treatment (MeCN). As shown in Figure
is the rate-limiting step for SNAr reactions in which general
4
, Brønsted-type plots for Kk2 and Kk3 values exhibit good
1,16a
base catalysis is absent.
Since the plots of kobsd vs [HNRR′]
linear correlations when Kk2, Kk3, and pKa are statistically
corrected by using p and q. It is noted that the slope for Kk2 is
practically identical to that for Kk3 (i.e., âKk2 ) 0.82 and âKk3
for the current reactions in H2O are linear, general base catalysis
by a second amine molecule is definitely absent. Thus, one can
propose the reactions of DNFB in H2O proceed through rate-
limiting formation of MC-1-Z (Scheme 1).
The above argument is consistent with the magnitude of ânuc
values. The ânuc value determined for the reactions in H2O is
)
0.84), which is consistent with the preceding argument that
k2 and k3 are insensitive to the basicity of amines. The magnitude
of these âKk2 and âKk3 values will receive our scrutiny
subsequently.
0
.52 (Figure 5), which is comparable to those reported for
reactions of 2,4-dinitrohalobenzenes with primary amines in H2O
i.e., ânuc varies from 0.42 to 0.45 and 0.52 as the halogen atom
Medium Effect on Reaction Mechanism: Comparison of
Brønsted-Type Treatment (H2O vs MeCN). It is apparent
from the inset of Figure 1 that the effect of medium is
significant, since the plot of kobsd vs [HNRR′] for the reaction
performed in H2O is linear passing through the origin. The linear
plot of kobsd vs amine concentration clearly indicates that the
deprotonation process found for the reactions in MeCN is absent
for the reactions in H2O (i.e., the k3 step in Scheme 1).
The fact that the plot of kobsd vs amine concentration passes
through the origin for the reaction in H2O suggests that the
contribution of hydroxide and/or water to the kobsd value is
negligible. Thus, the pseudo-first-order rate constant (kobsd) can
be expressed as eq 4. The apparent second-order rate constants
(
37,38
changes from F to I and Cl, in turn).
However, the ânuc value
obtained for the reaction of DNFB in H2O is much smaller than
that found for the corresponding reactions in MeCN, i.e., âKk3
)
0.84 or âKk2 ) 0.82 (Figure 4).
The large âKk3 or âKk2 values found for the reactions in MeCN
are in accord with the RLPT mechanism, in which bond
formation between the amine nucleophile and the electrophilic
site of DNFB is fully advanced in the rate-limiting transition
state. On the contrary, the smaller ânuc value shown in Figure
5
can account for the proposal that the reactions of DNFB with
the amines in H2O proceed through rate-limiting formation of
MC-1-Z, in which bond formation is not much advanced.
Halogen Atom Effect. The reaction of DNFB with piperidine
in MeCN was suggested to proceed through a rate-limiting
formation of a Meisenheimer complex on the basis of the kinetic
result that DNFB is much more reactive than 2,4-dinitrochlo-
(kN) for the reactions of DNFB in H2O have been determined
(
41) Gresser, M. J.; Jencks, W. P. J. Am. Chem. Soc. 1977, 99, 6970-
980.
42) (a) Castro, E. A.; Ureta, C. J. Org. Chem. 1990, 55, 1676-1679.
b) Castro, E. A.; Ibanez, F.; Santos, J. G.; Ureta, C. J. Org. Chem. 1993,
6
(
(
5
8, 4908-4912.
43) Um, I. H.; Seok, J. A.; Kim, H. T.; Bae, S. K. J. Org. Chem. 2003,
8, 7742-7746.
44) Bell, R. P. The Proton in Chemistry; Methuen: London, 1959.
40
robenzene (DNCB) in the aprotic solvent. Nudelman et al.
(
have found that DNFB is 375 times more reactive than DNCB
toward piperidine in MeCN. It is well-known that fluoride ion
6
(
J. Org. Chem, Vol. 72, No. 23, 2007 8801

Um et al.
FIGURE 4. Brønsted-type plots for the reactions of DNFB with
alicyclic secondary amines in MeCN at 25.0 ( 0.1 °C. The assignment
of numbers is as given in Table 1.
FIGURE 5. Brønsted-type plot for the reactions of DNFB with
alicyclic secondary amines in H
of numbers is given in Table 2.
2
O at 25.0 ( 0.1 °C. The assignment
is a poorer nucleofuge than chloride ion in an aprotic solvent
such as MeCN.40 Thus, the fact that DNFB is much more
reactive than DNCB in the aprotic solvent led them to conclude
-
-
that the leaving group (F or Cl ) departs after the rate-limiting
4
0
step. However, the current results (i.e., the upward curvature
in the plots of kobsd vs [HNRR′] and the microscopic rate
constants shown in Table 1) clearly indicate that formation of
a Meisenheimer complex occurs before the rate-limiting step.
Thus, the fact that DNFB is more reactive than DNCB in MeCN
cannot be an unambiguous measure of which step is rate-
limiting.
To investigate the origin of the high reactivity of DNFB
compared with DNCB, the reaction of the latter compound with
piperidine in MeCN has been performed in this study. As shown
in Figure 6, the plot of kobsd vs [HNRR′] is linear passing through
-
1
-1
the origin with a slope (kN) of 0.558 M s . Such a linear
plot is consistent with the report that the reaction of DNCB
with piperidine in MeCN proceeds through formation of an
intermediate in the rate-limiting step.45 Accordingly, the kN for
the reaction of DNCB with piperidine in MeCN represents the
rate constant for the amine-attack process (i.e., the k1 in
Scheme 1).
Since kN ) k1 for the reaction of DNCB in MeCN, one can
-
1 -1
compare the kN for the reaction of DNCB (0.558 M s ) with
FIGURE 6. Plots of kobsd vs [HNRR′] for the reactions of piperidine
with DNFB (b) and with DNCB (O) in MeCN at 25.0 °C.
-
1 -1
the k1 for that of DNFB (i.e., 380 M s in Table 1). The ratio
of these rate constants, k1(DNFB)/kN(DNCB) is 681. However,
this result does not indicate that DNFB is always 681 times
more reactive than DNCB. This argument is evident from Figure
amine concentration, which is not possible if the reactions of
DNFB and DNCB proceed via the same mechanism. Thus, one
can attribute the enhanced reactivity of DNFB compared with
DNCB in MeCN to the enhanced electrophilicity of the C-1
site of the former; it is expected to have a more electrophilic
site than the latter on the basis of the electronegativity of F vs
Cl.
6
, i.e., the plot of kobsd vs [HNRR′] is curved upward for the
reaction of DNFB but linear for the corresponding reaction of
DNCB. It is noteworthy that the ratio of the observed-rate
constants, kobsd(DNFB)/kobsd(DNCB) is highly dependent on the
Medium Effect on Reactivity. It is well-known that the rate
of reactions between neutral molecules decreases on changing
(
45) Martinez, R. D.; Mancini, P. M. E.; Vottero, L. R. J. Chem. Soc.,
Perkin Trans. II 1986, 1427-1431.
8802 J. Org. Chem., Vol. 72, No. 23, 2007

Brønsted-Type Analysis in Acetonitrile
the medium from H2O to a dipolar aprotic solvent such as
MeCN. In fact, we have recently shown that the amines used
in this study exhibit a similar or even decreased reactivity in
microscopic rate constants determined for the reactions in MeCN
-
1
2
(e.g., k2/k < 1 and k3/k2 > 10 ) account for the curvature
found in the plots of kobsd vs [HNRR′] and support the proposed
mechanism. (3) The Brønsted coefficients obtained in this study
(âKk2 ) 0.82 and âKk3 ) 0.84 in MeCN and ânuc ) 0.52 in
H2O) are also consistent with the proposed mechanisms. (4)
DNFB is significantly more reactive than DNCB in MeCN,
indicating that the fluorine atom in DNFB is more effective
than the chlorine atom in DNCB in enhancing the electophilicity
of the C-1 reaction site.
aminolysis of carboxylic esters on changing the medium from
H2O to MeCN,4
3,46
although these amines become more basic
39
in the aprotic solvent by 7-9 pKa units. Unlike aminolysis of
carboxylic esters, the current SNAr reactions exhibit higher
reactivity in the aprotic solvent than in H2O. Here the k1 value
in Table 1 for the reaction in MeCN is 17 (morpholine) to 72
(piperazine) times larger than the kN value for the corresponding
reaction in H2O (Table 2).
One can account for the contrasting medium effects found
for the current reactions and aminolysis of esters in terms of
the structures of their intermediates. It has generally been
understood that aminolysis of carboxylic esters proceeds through
a zwitterionic intermediate as illustrated in Structure I, in which
the negative and positive charges are mainly localized on the
O or the N atom. Water molecules can stabilize such charge
localized species through H-bonding interaction. However,
H-bonding interaction is absent in MeCN. Furthermore, there
would be large electronic repulsion between the negative charge
of the intermediate I and the negative dipole end of MeCN.
This argument accounts for the fact that the reactivity of amines
toward esters decreases on changing the medium from H2O to
MeCN, although amines become 7 and 9 pKa units more basic
in the aprotic solvent.43
Experimental Section
Materials. 2,4-Dinitrofluorobenzene and alicyclic secondary
amines were of the highest quality available. MeCN was distilled
over P O and stored under nitrogen. Doubly glass-distilled water
2 5
was further boiled and cooled under nitrogen just before use.
Kinetics. The kinetic study was performed using a UV-vis
spectrophotometer for slow reactions (t1/2 > 10 s) or a stopped-
flow spectrophotometer for fast reactions (t1/2 e 10 s) equipped
with a constant temperature circulating bath. The reactions were
followed by monitoring the appearance of N-(2,4-dinitrophenyl)-
amines at a fixed wavelength corresponding the maximum absorp-
tion (λmax, e.g., 379 nm for N-2,4-dinitrophenylpiperidine).
Typically, the reaction was initiated by adding 3 µL of a 0.02
M DNFB stock solution in MeCN by a 10 µL syringe to a 10 mm
UV cell containing 2.50 mL of the reaction medium and amine.
2
The amine sock solution of ca. 0.2 M for the reactions in H O was
prepared in 25.0 mL volumetric flask under nitrogen by adding 2
equiv of amine to 1 equiv of standardized HCl solution to obtain
a self-buffered solution. Transfers of solutions were carried out by
means of gastight syringes. All reactions were carried out under
pseudo-first-order conditions in which amine concentrations were
at least 50 times greater than the substrate concentration. The kinetic
conditions and pseudo-first-order rate constants are summarized in
Tables S1-S12 in the Supporting Information.
Product Analysis. N-(2,4-dinitrophenyl)amine was identified as
one of the products by comparison of the UV-vis spectra at the
end of the reactions with the authentic sample. For example, ꢀ )
On the other hand, the negative charge on the intermediate
of the current SNAr reaction is highly delocalized through the
resonance interaction as illustrated in the resonance structures
IIa and IIb.47 Such charge delocalized species are not solvated
strongly in H2O and would not experience significant desolva-
tion on changing the medium from H2O to MeCN. This
argument together with the enhanced basicity of amines, explains
the enhanced aminolytic reactivity in MeCN as compared to
H2O in the present SNAr reaction systems.
-
1
-1
15800 M cm at 379 nm for N-2,4-dinitrophenylpiperidine in
MeCN.
Acknowledgment. We are grateful for the financial support
from the Korea Research Foundation (KRF-2005-015-C00256).
J.M.D. thanks Sir Wilfred Grenfell College (SWGC Principal’s
Research Fund) for generous support. Professor D. R. Parkinson
and Ms. K. Rice (SWGC) are thanked for helpful discussions.
Conclusions
Supporting Information Available: Plots of kobsd vs [HNRR′]
for the reactions of DNFB with piperidine, piperazine, 1-(2-
hydroxyethyl)piperazine, and 1-formylpiperazine (Figures S1-S4).
Plots of kobsd/[HNRR′] vs [HNRR′] for the reaction of DNFB with
The current study has allowed us to conclude the following:
(1) The effect of medium on reactivity and reaction mechanism
is significant for the current SNAr reactions; the reaction of
DNFB in MeCN proceeds through an RLPT mechanism, while
the one in H2O proceeds through a Meisenheimer complex (MC-
1
-formylpiperazine and piperazine (Figures S5 and S6). Plot of log
/k vs pK for the reactions of DNFB with amines in MeCN
Figure S7). The kinetic conditions and results for the reactions of
k
3
2
a
(
1
-Z) with its formation being the rate-limiting step. (2) The
DNFB with amines in MeCN and in H O (Tables S1-S11). The
2
kinetic conditions and results for the reactions of DNCB with
(
46) Um, I. H.; Jeon, S. E.; Seok, J. A. Chem.-Eur. J. 2006, 1237-
piperidine in MeCN (Table S12). This material is available free of
charge via the Internet at http://pubs.acs.org.
1
2
243.
(
47) (a) Capon, B.; Rees, C. W. Ann. Rep. Prog. Chem. 1963, 59, 207-
54. (b) Hirst, J. J. Phys. Org. Chem. 1994, 7, 68-79.
JO701549H
J. Org. Chem, Vol. 72, No. 23, 2007 8803