Reversible and Irreversible ElcB Mechanisms in the Hydrolysis of 2,2,2-Trifluoroethanesulfonyl Chloride: Carbanion Intermediates in Aqueous Acid

Article abstract of DOI:10.1021/jo971872v

Hydrolysis of 2,2,2-trifluoroethanesulfonyl chloride (1) is shown to take place by way of the sulfene (CF₃CH=SO₂), formed by (a) an irreversible E1cB process over the pH range 1.8-5 with water acting as the carbanion-forming base in the lower pH range and hydroxide anion at higher pH, and (b) a reversible E1cB reaction in dilute acid.

Full text of DOI:10.1021/jo971872v

808
J . Org. Chem. 1998, 63, 808-811
Rever sible a n d Ir r ever sible ElcB Mech a n ism s in th e Hyd r olysis of
2,2,2-Tr iflu or oeth a n esu lfon yl Ch lor id e: Ca r ba n ion In ter m ed ia tes
in Aqu eou s Acid 1
J ames F. King* and Manjinder Singh Gill
Department of Chemistry, The University of Western Ontario, London, Ontario, Canada N6A 5B7
Received October 10, 1997
Hydrolysis of 2,2,2-trifluoroethanesulfonyl chloride (1) is shown to take place by way of the sulfene
(CF3CHdSO2), formed by (a) an irreversible E1cB process over the pH range 1.8-5 with water
acting as the carbanion-forming base in the lower pH range and hydroxide anion at higher pH,
and (b) a reversible E1cB reaction in dilute acid.
In tr od u ction
lyl)methanesulfonyl chloride, on the other hand, reacts
with water exclusively by attack at the silicon to form
sulfene, and with hydroxide by both silicophilic attack
to form sulfene and by elimination of HCl to give
trimethylsilylsulfene (Me3SiCHdSO2).8 2-Methyl-2-pro-
panesulfonyl chloride, the simplest tertiary alkanesulfo-
nyl chloride, evidently hydrolyzes9 entirely by way of the
tert-butyl cation over the pH range 1-13.
In light of the variety of mechanisms displayed in the
hydrolysis of sulfonyl chlorides it was not possible to
predict with any measure of certainty how the hydrolysis
of 2,2,2-trifluoroethanesulfonyl chloride (“tresyl” chloride,
1) might proceed, and as part of our general study of
sulfonyl transfer mechanisms3-4,6-8 we decided to look
at the hydrolysis of 1.10
Current understanding of the mechanisms of the
hydrolysis of alkanesulfonyl chlorides may be sum-
marized by first describing the following accepted picture
for the hydrolysis of methanesulfonyl chloride.2,3 (a) In
the pH range ∼1 to 6.7, the reaction is a direct nucleo-
philic displacement by attack of water on the sulfonyl
sulfur atom with extrusion of chloride anion (SN2-S
reaction). (b) In the pH range 6.7 to 11.8 the hydrolysis
involves a rate-determining hydroxide-promoted elimina-
tion of HCl (probably E2) to form sulfene (CH2dSO2)
which is then quickly trapped by water. (c) Above pH
11.8 similar formation of the sulfene is followed by very
fast trapping of the sulfene by hydroxide anion. Except
for the transition pH ranges immediately around either
pH 6.7 or 11.8 (in which both competing processes are,
of course, observed) only one process is in evidence, i.e.,
there is no significant competition by minor pathways.3
Most simple alkanesulfonyl chlorides, including ethane-
and butanesulfonyl chlorides, phenylmethanesulfonyl
chloride, 2-methoxyethanesulfonyl chloride,3 and even
cyclopropanesulfonyl chloride4 show essentially the same
mechanisms. A variation of this mechanism in the form
of a reversible E1cB mechanism leading to the sulfene
has been put forward for the hydrolysis of dichlo-
romethanesulfonyl chloride.5
Resu lts a n d Discu ssion
Hydrolysis of 1, though too fast for our apparatus at
25 °C, was conveniently followed by pH-stat at 1 °C. The
observed pH-rate profile for the reaction in 0.05 M KCl,
which is shown in Figure 1, has the familiar “hockey
stick” shape found with the hydrolyses of the simple
alkanesulfonyl chlorides, and which corresponds to the
rate law kobs ) kw + kOH[OH-]. A sample of tresyl-1,1-d2
chloride (6), was prepared as in Scheme 1. The pH-rate
profile for 6, also given in Figure 1, shows the same shape
but with a new feature, namely a kinetic isotope effect
for kw as well as kOH. For 1, kw ) 3.3 × 10-3, whereas
for 6, kw ) 1.50 × 10-3 s-1, (kw)H/(kw)D ) 2.2; similarly
for 1, kOH ) 1.75 × 107, and for 6, kOH ) 3.4 × 106 M-1
There are exceptions to the above picture. 2-Hydroxy-
ethanesulfonyl chloride, for example, reacts principally
by way of â-sultone with no sign of sulfene formation.6
Ethenesulfonyl chloride gives no indication of the cumu-
lated sulfene (CH2dCdSO2) and reacts with hydroxide
anion mostly by direct attack on the sulfur with a small
portion of vinylogous nucleophilic attack to form hy-
droxymethylsulfene (HOCH2CHdSO2).7 (Trimethylsi-
s-1, and (kOH)H/(kOH
) ) 5.4.
D
In the reactions of the simplest alkanesulfonyl chlo-
rides a sizable primary kinetic isotope effect (KIE) was
observed for kOH, but there was no KIE for the kw term,
in accord with the proposed mechanism. It should also
be noted that the rate constants for 1 are faster than the
* Corresponding author: Telephone 519-679-2111 ext 6348; Fax 519-
661-3022; e-mail scijfk@uwoadmin.uwo.ca
(1) Organic sulfur mechanisms. 42. Part 41: King, J . F.; Yuyitung,
G.; Gill, M. S.; Stewart, J . C.; Payne, N. C. Can. J . Chem., in press.
(2) (a) Gordon, I. M.; Maskill, H.; Ruasse, M. F. Chem. Soc. Rev.
1989, 123-151. (b) Kice, J . L. Adv. Phys. Org. Chem. 1980, 17, 65-
181.
(7) King, J . F.; Hillhouse, J . H.; Skonieczny, S. Can. J . Chem. 1984,
62, 1977-1995.
(8) King, J . F.; Lam, J . Y. L. J . Org. Chem. 1993, 58, 3429-3434.
(9) King, J . F.; Lam, J . Y. L.; Dave, V. J . Org. Chem. 1995, 60, 2831-
2834.
(3) King, J . F.; Lam, J . Y. L.; Skonieczny, S. J . Am. Chem. Soc. 1992,
114, 1743-1749.
(10) Part of the work reported in this paper has been presented in
preliminary form at (a) the 23rd Ontario-Quebec Physical-organic
Minisymposium, Hamilton, Ont., Nov 1995, (b) the Fourth Interna-
tional Conference on Heteroatom Chemistry (ICHAC-4), Seoul, Korea,
J uly 1995, cf. King, J . F.; Gill, M. S.; Klassen, D. F. Pure Appl. Chem.
1996, 68, 825-830.
(4) King, J . F.; Lam, J . Y. L.; Ferrazzi, G. J . Org. Chem. 1993, 58,
1128-1135.
(5) Seifert, R.; Zbirovsky´, M.; Sauer, M. Collect. Czech. Chem.
Commun. 1973, 38, 2477-2483.
(6) King, J . F.; Khemani, K. C. Can. J . Chem. 1989, 67, 2162-2172.
S0022-3263(97)01872-0 CCC: $15.00 © 1998 American Chemical Society
Published on Web 01/22/1998

Carbanion Intermediates in Aqueous Acid
J . Org. Chem., Vol. 63, No. 3, 1998 809
Ta ble 1. Selected P seu d o-F ir st-Or d er Ra te Con sta n ts for th e Hyd r olysis of th e Isotop om er s of
2,2,2-Tr iflu or oeth a n esu lfon yl Ch lor id e
substrate
reaction medium (temp, °C)
pH/pDa
method
kobs (s-1
)
1
6
1
1
1
1
6
1
1
H2O, 0.05 M KCl (1.0)
H2O, 0.05 M KCl (1.0)
H2O:CD3CN, 80:20 (4.5)
H2O:CD3CN, 80:20 (4.5)
D2O:CD3CN, 80:20 (4.5)
D2O:CD3CN, 80:20 (4.5)
H2O:CD3CN, 80:20 (4.5)
2 M H2SO4 in H2O:CD3CN, 80:20 (4.5)
2 M D2SO4 in D2O:CD3CN, 80:20 (4.5)
2.5
2.5
2.5
2.5
2.5
1.82
2.5
-
pH-stat
pH-stat
pH-stat
3.67 × 10-3
1.50 × 10-3
9.5 × 10-3
1.0 × 10-2
2.0 × 10-3
1.80 × 10-3
5.5 × 10-3
1.21 × 10-3
2.4 × 10-4
19F NMR
19F NMR
19F NMR
19F NMR
19F NMR
19F NMR
-
a
pH meter reading for solutions with ordinary water; pH meter reading +0.37 for D2O-containing media.
F igu r e 1. pH-rate profiles for the hydrolysis in 0.05 M KCl
at 1.0 °C of trifluoroethanesulfonyl chloride (1) (circles and
solid line) and trifluoroethanesulfonyl-d2 chloride (6) (squares
and broken line). The points are experimental and the lines
calculated from the equation log kobs ) log(kw + kOH[OH-])
taken with following rate constants derived from nonlinear
least-squares fitting of the points: for 1, kw ) 3.3 × 10-3 s-1
and kOH ) 1.75 × 107 M-1 s-1; for 6 kw ) 1.50 × 10-3 s-1 and
kOH ) 3.4 × 106 M-1 s-1
.
Sch em e 1
corresponding values for other alkanesulfonyl chlorides3
(at 25 °C), which is consistent with the intervention of
different mechanisms in the hydrolysis of 1.
F igu r e 2. 19F NMR spectra of the reaction product as a
function of time for the hydrolysis of 1 in an 80:20 mixture of
2 M D2SO4 and CD3CN at 4.5 °C.
Further information was obtained by following the
hydrolysis and deuteriolysis of 1 and 6 by 19F NMR
spectrometry; to obtain concentrations of 1 (or 6) suf-
ficiently high for convenient determination of the spectra,
an 80:20 mixture of H2O:CD3CN (or D2O:CD3CN) was
used as solvent. For all of the reactions above pH 1.8,
Interestingly, the solvent KIE on going from H2O:CD3-
CN to D2O:CD3CN is large, kH/kD ) 5.0.
A new feature appeared in the 19F NMR spectra
obtained by following the reaction of 1 in 2 M D2SO4 in
D2O (plus CD3CN), shown in Figure 2. As is readily
evident, each isotopomer of tresyl chloride and the
tresylate anion is easily distinguished. As with the
spectra observed at higher pH’s, there is only the smallest
trace of CF3CH2SO3-, again in accord with the interme-
diacy of the sulfene, CF3CHdSO2. What is new in Figure
2, however, is the set of peaks due to the isotopomers of
tresyl chloride, CF3CHDSO2Cl (doublet) and CF3CD2SO2-
Cl (singlet), in addition to the expected hydrolysis
-
the product was the monodeuterated anion CF3CHDSO3
with a trace of the anion corresponding to no H-D
-
exchange (CF3CH2SO3 or CF3CD2SO3-) in accord with
the intermediacy of the sulfene, CF3CHdSO2; there was
no sign of H-D exchange in the starting material. Table
1 lists the rate constants (obtained at 4.5 °C) by 19F NMR,
as well as a pH-stat determination, along with two of the
earlier pH-stat results (at 1 °C in 0.05 M KCl) for
comparison; the 19F NMR and pH-stat kobs values are
identical within experimental uncertainty. Comparison
of the rate constants for 1 and 6 (in H2O:CD3CN)
indicates a primary kinetic isotope effect (KIE) of 1.8.
-
products, CF3CHDSO3 and CF3CD2SO3-; a control ex-
periment showed the complete absence of H-D exchange
in the tresylate anion under these conditions.

810 J . Org. Chem., Vol. 63, No. 3, 1998
King and Gill
Sch em e 2
These observations are most simply accounted for by
assuming formation of a discrete carbanion by reaction
with H2O (or D2O), and its reversal (as required by
microscopic reversibility) by H3O+ (or D3O+). In other
words (a) the reaction in 2 M D2SO4 (and presumably
that in 2 M H2SO4 as well) is a reversible E1cB process
leading to the sulfene, and (b) that in the pH range 2 to
∼4.3, in which no H-D exchange in 1 or 6 is seen, is an
irreversible ElcB reaction; in both of these cicumstances
the carbanion is generated by attack of H2O (or D2O).
By extension the reaction at higher pH (>4.3), i.e., the
kOH term, must also be an irreversible ElcB reaction
giving the sulfene, with the only difference being that
the carbanion is formed by attack of hydroxide anion. The
nature of the sulfene trapping step has not been studied,
though trapping by water would seem to be the most
likely route to the tresylate anion under (most of) the
conditions of these experiments. The mechanism of the
low pH hydrolysis (kw term) is shown in Scheme 2.
F igu r e 3. Quantities of the isotopomers of trifluorethane-
sulfonyl chloride (measured as a fraction of the total original
1) as a function of time in the hydrolysis of 1 in an 80:20
mixture of 2 M D2SO4:CD3CN at 4.5 °C. The points were
obtained from 19F NMR spectra (shown partly in Figure 2) and
the lines from computer simulation of the mechanism shown
in Scheme 2 using k1 ) 3.5 × 10-4 s-1, k3:k4:k5 ) 1:3:1, KIE )
3 for trapping of the sulfene, [D2O] ) 2 M, and deuterium
content of the D2O, 99.8 atom % excess D.11
Computer simulation11 based on this mechanism was
carried out with the following parameters: k1 3.5 × 10-4
s-1; k3:k4:k5 1:3:1; KIE in sulfene trapping 3; [D3O+] 2 M;
99.8 atom % excess D. As my be seen from Figures 3
and 4, the correspondence between prediction and the
results of the 19F NMR experiment is good, except
perhaps for the relatively large amount of undeuterated
tresylate anion (CF3CH2SO3-) found over what is pre-
dicted by the simulation. Simulation using a larger value
for the protium concentration in the D2SO4 than that
indicated by the suppliers gives a very good fit to the
points for CF3CH2SO3-. Though this could simply mean
that there is more protium in the D2SO4-D2O solution
than expected from the label, it is also conceivable that
the H-D exchange processes involving the carbanion,
CF3ChHSO2Cl, may be very fast relative to the mixing
processes and the concentration of D2OH+ in the vicinity
of any carbanion may well be higher than that corre-
sponding to true equilibration of all species, with the
result that protiation of the carbanion may be much
greater than expected on the basis of complete mixing
before sulfene trapping.
F igu r e 4. Amounts of the isotopomers of trifluoroethane-
sulfonate anion in the same reaction of 1 in D2SO4:CD3CN at
4.5 °C; further details are in the caption to Figure 3.
exchange in recovered sulfonyl chloride after partial
reaction in D2O:dioxane. This behavior was quite dif-
ferent from that seen by the same authors with meth-
anesulfonyl or chloromethanesulfonyl chlorides, which
showed the normal reactions described earlier. Like all
studies of sulfonyl chloride hydrolysis carried out up to
that time (and later), their experiments were conducted
without pH control, and hence most of the reactions must
be regarded as having taken place in aqueous media of
somewhat varying acidities. We estimate that in the
experiment which led to recovered Cl2CDSO2Cl, 90% of
the reaction took place with [D3O+] in the range 0.02 to
0.22 M. Under their conditions the reversible E1cB
reaction is evidently operating, but the mechanism of
hydrolysis under less acidic conditions is simply not
known. Further study is in order, not only of dichlo-
romethanesulfonyl chloride, but also of both the large
solvent isotope effect and the effect of acid on the rate of
ionization of 1, and we hope to be able to report on these
points at a later date.
As was noted in the Introduction, there is one report
of a sulfonyl chloride hydrolysis by way of an E1cB
process. Seifert, Zbirovsky, and Sauer5 concluded that the
hydrolysis of dichloromethanesulfonyl chloride in diox-
ane:water mixtures takes place via the reversible E1cB
reaction. The key pieces of evidence were (a) rate
suppression by added acid and (b) the observation of H-D
(11) For the simulation program see Gill, M. S., Ph.D. Thesis, The
University of Western Ontario, April, 1996; pp 201-202; Chem. Abstr.
1997, 127, 148846s.

Carbanion Intermediates in Aqueous Acid
J . Org. Chem., Vol. 63, No. 3, 1998 811
min. The reaction mixture was stirred for 3.5 h; workup with
dichloromethane followed by washing with ice cold water and
HCl (10%), drying, and evaporation of the organic phase
followed by fractional distillation gave 4 (2.8 g, 62%) as a clear
Though carbanion formation with water (rather than
hydroxide ion) acting as the base is well-precedented in,
for example, the ionization in water of such very strong
carbon acids as 2-(dicyanomethylene)-1,1,3,3-tetracyano-
propane12 and tris(methylsulfonyl)methane,13 the hy-
drolysis of 1 is to the best of our knowledge the first
clearly demonstrated example of reversible and irrevers-
ible E1cB reactions induced by water. We were fortunate
in this instance to be able to induce H-D exchange under
acidic conditions, and hence to observe the alteration in
the reaction from (E1cB)irr to (E1cB)rev, thereby providing
an easy way to distinguish the (E1cB)irr and E2 processes.
liquid. 1H NMR δ 3.15 (s), 13C NMR δ 38.0, 63.6 (4 × 5, J CF
)
38.1 Hz, J CD ) 23.4 Hz), 121.2 (q, J ) 277.5 Hz), 19F NMR δ
-74.47 (s). A solution of sodium thiocyanate (3.6 g, 44.4 mmol)
in DMF (25 mL) was heated till the head temperature reached
150 °C (6 mL of DMF collected). To the cooled reaction mixture
was added 4 (2 g, 11.1 mmol), and the reaction mixture slowly
distilled over a period of 2 h. The distillate was taken up in
pentane (100 mL), and the solution was washed with water
(4 × 50 mL). The pentane layer was dried over anhydrous
MgSO4 and the solvent carefully distilled off to give 5 (600
mg, 37.8%) as a light yellow liquid; 19F NMR δ -66.7 (s).
Conversion of 5 to 6 followed the preparation of 1; 5 (520 mg,
3.63 mmol) gave 6 (355 mg, 53%) as a clear liquid. The 1H
NMR spectrum showed only signals due to the presence of
traces of DMF; 19F NMR δ -62.36 (s).
Exp er im en ta l Section
Laboratory procedures, spectra determination, and pH-stat
rate measurements were carried out as previously described,14
except that a Neslab ULT-80 low-temperature bath circulator
was used for rate measuurements at 1.0 °C. 2,2,2-Trifluoro-
ethanesulfonyl chloride (1) was either purchased from Aldrich
Chemical Co. Inc. and used as supplied, or prepared by the
following variation on the literature15 procedure. Aqueous 6
M HCl (20 mL) in an ice bath was saturated with Cl2, and
CF3CH2SCN15 (1.81 g, 12.8 mmol) was quickly added to the
stirred solution. The reaction mixture was stirred for 8 min
with a constant flow of Cl2. Workup, followed by distillation
under reduced pressure, gave 1 (1.28 g, 53%) as a clear liquid.
1H NMR δ 4.43 (q, J ) 8 Hz, 2H); 19F NMR δ -62.09 (t, J )
8 Hz); reported 1H NMR (neat) δ 4.37 (q, J ) 9 Hz, 2H).16
Authentic 2,2,2-trifluoroethanesulfonic acid was prepared by
stirring a mixture of 1 (283 mg, 1.55 mmol) and water (5 mL)
for 1.5 h and then removing the water under reduced pressure
P r od u cts of Deu ter iolysis of 1 a n d Hyd r olysis of 6. In
D2O. A sample of 1 (10 mg, 0.055 mmol) in CD3CN:D2O, 20:
80, after 15 min showed 1H, 13C, and 19F NMR signals
-
assignable almost entirely to monodeutrated CF3CHDSO3
,
with a small amount (e2%) of CF3CH2SO3-. Similar spectra
-
were obtained for the reaction in pH 2-7 range; CF3CHDSO3
1H NMR (D2O/DSS) δ 3.86 (qt, J ) 9 Hz, J ) 2.3 Hz, 1H);
13C NMR (D2O/DSS) δ 54.9 (qt, J ) 30 Hz, J ) 21 Hz), 125.3
(q, J ) 275.5 Hz); 19F NMR (D2O/CFCl3) δ -63.49 (d, J ) 9
Hz). In H2O. Similar reaction of 6 (10 mg, 0.054 mmol) in
CD3CN:H2O, 20:80, mixtures in pH range 2-7 after 15 min
:
showed signals assignable mostly to CF3CHDSO3- and a trace
-
of CF3CD2SO3
.
P r oced u r e for Hyd r olysis Ra te Mea su r em en ts. Aque-
ous KCl (50 mL, 0.05 M) solution was brought to 1 °C and the
desired pH using 0.1 M H2SO4 or 0.1 M NaOH. To the above
stirred solution was added a solution of 1 (or 6) (3.72 mg, 0.02
mmol) in THF (50 µL), and the pH of the solution was kept
constant with NaOH (0.1 M). The rate of hydrolysis was
monitored by recording the volume of the titrant (NaOH)
added with time. Plots of ln(V∞ - Vt) versus time were
constructed from the volume of sodium hydroxide titrant
added. The pseudo-first-order rate constants were then
obtained from the slopes of straight lines; the results are
shown in Figure 1 and listed in Table S1 (see also Table 1).
The pH-stat measurements at 4.5 °C were carried out similarly
except that a mixture of CH3CN (10 mL) and aqueous KCl
solution was kept at 4.5 °C in a Haake FJ constant temper-
ature circulating bath, and a solution of 1 (3.11 mg, 0.017
mmol) in dry CH3CN was used for each run.
1
leaving tresic acid as a clear liquid (202 mg, 80%); H NMR
(D2O/DSS) δ 3.88 (q, J ) 10 Hz, 2H); 19F NMR δ -63.4 (t, J )
10 Hz); reported16 1H NMR (D2O) δ 3.87 (q, J ) 10 Hz, 2H).
2,2,2-Tr iflu or oeth a n esu lfon yl-1,1-d 2 Ch lor id e (6). Ben-
zyl trifluoroacetate (2) (15 g, 73.5 mmol) from a literature
preparation17 (in 80% yield) was added dropwise to a slurry
of LiAlD4 (2 g, 48 mmol) in dry ether (80 mL) under N2, over
a period of 40 min, and the mixture was stirred for 1 h at rt.
Excess reducing agent was destroyed by dropwise addition of
water, followed by addition of 20% H2SO4 (50 mL). The
reaction mixture was extracted with dichloromethane (200
mL). Fractional distillation using a Vigreux column (12 cm)
gave fractions that contained 3 (5 g, 67%) with a little diethyl
ether. 1H NMR δ 3.5 (br, s, OH); 19F NMR δ -77.86 (s).
Triethylamine (3.08 g, 30.4 mmol) was added to a stirred
solution of 3 (2.7 g, 26.5 mmol) in dry ether (75 mL) in an ice
bath, and methanesulfonyl chloride (3.34 g, 29 mmol) was
added dropwise from an addition funnel over a period of 15
Ack n ow led gm en t. We thank the Natural Sciences
and Engineering Research Council of Canada for finan-
cial support of this study.
(12) Middleton, W. J .; Little, E. L.; Coffman, D. D.; Engelhardt, V.
A. J . Am. Chem. Soc. 1958, 80, 2795-2806.
(13) Schwartzenbach, G.; Felder, E.; Helv. Chim. Acta 1944, 27,
1701-1711.
Su p p or tin g In for m a tion Ava ila ble: Rate constants for
the hydrolysis of 1 and 6 at 1 °C and of 1 in H2O:CH3CN 80:
20 at 4.5 °C (1 page). This material is contained in libraries
on microfiche, immediately follows this article in the microfilm
version of the journal, and can be ordered from the ACS; see
any current masthead page for ordering information.
(14) King, J . F.; Gill, M. S. J . Org. Chem. 1996, 61, 7250-7255.
(15) Crossland, R. K.; Wells, W. E.; Shiner, V. J ., J r. J . Am. Chem.
Soc. 1971, 93, 4217-4219.
(16) Bunyagidj, C.; Piotrowska, H.; Aldridge, M. H. J . Org. Chem.
1981, 46, 3335-3336.
(17) Oliverio, V. T.; Sawicki, E. J . Org. Chem. 1955, 20, 363-367.
J O971872V