Thermodynamics of methanesulfonic acid revisited

Article abstract of DOI:10.1139/v00-134

Recently we reported a study of the thermodynamics of methanesulfonic acid and some of its derivatives. The foundation of these results was a measurement of the heat of reaction of S-methyl thioacetate with aq sodium hypochlorite, leading to methanesulfonic acid. We have reinvestigated this reaction and discovered that contrary to the initial stoichiometry experiments, the stoichiometry under the reaction conditions is not as was believed and that the heat of reaction observed was spuriously high. We have found a new reaction, that of sulfite ion with methyl methanesulfonate, which does allow a clean determination of the heat of formation of methanesulfonic acid. Revised thermodynamic quantities for methanesulfonic acid, methanesulfonyl chloride, and methyl methanesulfonate are reported here.

Full text of DOI:10.1139/v00-134

1
295
Thermodynamics of methanesulfonic acid
revisited
J. Peter Guthrie and Roger T. Gallant
Abstract: Recently we reported a study of the thermodynamics of methanesulfonic acid and some of its derivatives.
The foundation of these results was a measurement of the heat of reaction of S-methyl thioacetate with aq sodium
hypochlorite, leading to methanesulfonic acid. We have reinvestigated this reaction and discovered that contrary to the
initial stoichiometry experiments, the stoichiometry under the reaction conditions is not as was believed and that the
heat of reaction observed was spuriously high. We have found a new reaction, that of sulfite ion with methyl
methanesulfonate, which does allow a clean determination of the heat of formation of methanesulfonic acid. Revised
thermodynamic quantities for methanesulfonic acid, methanesulfonyl chloride, and methyl methanesulfonate are re-
ported here.
Key words: sulfonic acid, heat of reaction, free energy of formation, S 2 reaction.
N
Résumé : On a récemment rapporté une étude sur les propriétés thermodynamiques de l’acide méthanesulfonique et de
ses dérivés. La base de ces résultats était une mesure de la chaleur de réaction du thioacétate de S-méthyle avec de
l’hypochlorite de sodium aqueux conduisant à la formation d’acide méthanesulfonique. On a repris l’étude de cette
réaction et on a trouvé que, contrairement aux expériences initiales de stoechiométrie, cette dernière n’est pas toujours
celle que l’on croyait et que la chaleur de réaction observée était faussement élevée. On a trouvé une nouvelle réaction,
celle de l’ion sulfite avec le méthanesulfonate de méthyle qui permet de faire une détermination propre de la chaleur
de formation de l’acide méthanesulfonique. On rapporte les valeurs révisées des quantités thermodynamiques pour
l’acide méthanesulfonique, le chlorure de méthanesulfonyle et le méthanesulfonate de méthyle.
Mots clés : acide sulfonique, chaleur de réaction, énergie libre de formation, réaction S 2. Guthrie and Gallant
N
1
298
Introduction
ent heat of hydrolysis of methyl trifluoromethanesulfonate in
pH 10 aq borate buffer was –7.85 ± 0.35 kcal mol .
–
1
Recently we reported a study of the thermochemistry and
thermodynamics of methaneulfonic acid and some of its de-
rivatives (1). Two colleagues knowledgeable in this area
pointed out that the heat of formation we reported was in-
consistent with heats of formation of related compounds or
other indirect sources of heat of formation data. To settle the
point we have determined the heat of formation of methane-
sulfonic acid independently, by a reaction of sulfite ion with
methyl methanesulfonate (methyl mesylate). With some em-
barrassment we now report that the initial value we reported
was in error and herewith revise the thermochemistry of
methanesulfonic acid.
−
[
1]
CH -O-SO -CF (1) + (HO) B (soln)
3
2
3
4
−
→
CH -OH(soln) + O S-CF (soln) + (HO) B(soln)
3 3 3 3
The heat of reaction of methyl trifluoromethanesulfonate
with aq sodium sulfite was –11.76 ± 0.12 kcal mol .
–
1
2
−
[
2]
CH -O-SO -CF (1) + SO (soln)
3 2 3 3
−
−
→ CH -SO (soln)+ O S-CF (soln) + (HO) B(soln)
3
3
3
3
3
The heat of reaction of hydroxide with boric acid can be
calculated from the known heats of ionization of boric acid
and water (2) to be –9.97 ± 0.01 kcal mol .
–
1
Results
−
−
[
3]
Combining these values allows calculation of the heat of
reaction for the convenient hypothetical process, [4], for
(HO) B(soln) + HO (soln) → (HO) B (soln)
Heats of reaction were measured using a Tronac isoperi-
bol calorimeter. The results are given in Table 1. The appar-
3
4
–
1
which ∆Hreaction = +6.06 ± 0.37 kcal mol . From the heats
of formation of the other species involved (see Table 2), we
can now calculate the heat of formation of methanesulfonate
Received September 2, 1999. Published on the NRC Research
Press website on September 26, 2000.
–1
ion in aqueous solution as –149.64 ± 0.70 kcal mol .
1
J.P. Guthrie and R.T. Gallant. Department of Chemistry,
2
−
−
University of Western Ontario. London, ON N6A 5B7,
Canada.
[4]
CH -OH(soln) + SO (soln) → CH + SO (soln)
3 3 3 3
−
+
HO (soln)
1Author to whom correspondence may be addressed.
Telephone: (519) 661-2111 ext. 86352. Fax: (519) 661-3022.
e-mail: peter.guthrie@uwo.ca
The heat of solution of methanesulfonic acid in water was
–
1
–11.40 ± 0.41 kcal mol .
Can. J. Chem. 78: 1295–1298 (2000)
© 2000 NRC Canada

1
296
Can. J. Chem. Vol. 78, 2000
Table 1. Heats of reaction and solution.a
Compound
∆H
Av
b
b
b
b
CH OSO CF + borate buffer
–7.92 , –8.25 , –7.39
–7.85 ± 0.35
–11.76 ± 0.12
–11.40 ± 0.41
–37.05 ± 0.55
3
2
3
c
c
c
c
c
c
CH OSO CF + sulfite
–11.58 , –11.93 , –11.75 , –11.75 , –11.78
3
2
3
d
d
d
d
d
CH SO H
–11.70 –10.69 , –11.51 , –11.68
3
3
e
e
e
e
CH OSO CH + sulfite
–37.89 , –37.11 , –36.80 , –36.38
3
2
3
a
b
c
–1
All at 25°C, units are kcal mol ; heat of reaction unless noted.
In pH 10 aq borate buffer l.
In 0.1 M aq Na SO .
2
3
d
Heat of solution in water.
e
In 0.5 M aq Na SO .
2
3
−
+
[
5]
CH -SO H(1) → CH -SO (soln) + H (soln)
Table 2. Heats of formation.a
3
3
3
3
This permits calculation of the heat of formation of liquid
methanesulfonic acid as –138.24 ± 0.81.
Compound
Phase
∆Hf°
b
H O
l
–68.32 ± 0.01
–112.36 ± 0.01
–97.25 ± 0.02
–266.70 ± 0.59
This value was closer to the previous literature estimates,
2
–
1
b
NaOH
NaCl
Na2SO3
a
soln
soln
l
but still differed by 10 kcal mol from these estimates.
Nagging concerns about this discrepancy, prompted by simi-
lar concerns expressed by a referee reviewing an early ver-
sion of this paper, led us to do further analysis of the results,
and then to do further experiments. From the previous re-
sults (1) we can calculate the heat of hydrolysis for methyl
methanesulfonate. This heat of reaction does not depend on
the absolute value of the heat of formation of methane-
sulfonic acid from our original paper, but only on the heats
of reaction of methanesulfonyl chloride with water or metha-
nol, and this heat of hydrolysis is consistent with the litera-
ture estimates. The heat of hydrolysis for the process
b
b
–1
All at 25°C, heats of formation in kcal mol .
Reference (10).
b
–
1
methanesulfonate, as –7.89 ± 0.85 kcal mol , which is quite
different from the value based on the heats of reaction of
methyl trifluoromethanesulfonate, namely +6.06 ± 0.37 kcal
–
1
mol . In this calculation process [8] minus process [6]
yields process [4].
−
−
[
6]
CH SO OCH (1) + HO (soln) → CH SO (soln)
From the new value for the heat of reaction of process [4],
we derive a heat of formation of methanesulfonate ion in
aqueous solution as –163.79 ± 1.04 kcal mol , and hence a
3
2
3
3
3
+
CH OH(soln)
3
–1
–
1
–1
is –27.31 kcal mol , while for the analogous reaction of
methyl trifluoromethanesulfonate
value of –152.39 ± 1.12 kcal mol for liquid methanesul-
fonic acid. This requires revision of all the thermodynamic
quantities dependent upon the heat of formation of methane
sulfonic acid, in Table 4 of ref. (1). The revised values are
given in Table 3 below.
−
−
[
7]
CF SO OCH (1) + HO (soln) → CF SO (soln)
3
2
3
3
3
+
CH OH(soln)
3
The stoichiometry of the reaction of methyl thioacetate
with hypochlorite was reexamined. In the original work it
had been demonstrated by NMR analysis that the only prod-
ucts were acetate and methanesulfonate, in equimolar amounts.
This experiment involved reaction with full strength, 5%,
hypochlorite solution, to allow concentrations of products
we calculate from the results given above that the heat of hy-
–
1
drolysis is –17.82 kcal mol . That the heat of hydrolysis of
the ester of the stronger acid should be less than that of the
ester of the weaker acid seems wildly unlikely.
We turned next to measurement of the heat of reaction of
sulfite ion with methyl methanesulfonate. With 0.1 M so-
dium sulfite in water the reaction was not quite fast enough
for convenient calorimetry, but at 0.5 M sodium sulfite the
reaction was fast enough to give a well-defined post reaction
trace and thus allow good determination of the heat of reac-
tion. For the process:
1
high enough for analysis by 200 MHz H NMR spectros-
copy. However there was a difference in concentration be-
tween the stoichiometry determinations and the calorimetry,
since full strength 5% sodium hypochlorite was used for
–
stoichiometry determinations (ca. 0.67 M OCl , 0.11 M
thioester), and 15:35 sodium hypochlorite:water was used
–
2
−
for calorimetry (ca. 0.28 M OCl , 0.0054 M thioester). The
[
8]
CH -O-SO -CH (1) + SO (soln)
3 2 3 3
distilled water could significantly lower the hydroxide con-
centration of this very weakly buffered solution, and thus
have affect the relative rates of hydroxide dependent and
hypochlorite dependent reactions.
−
→
2CH -SO (soln)
3 3
the heat of reaction was –37.05 ± 0.55. This is much more exo-
thermic than the corresponding value from experiments with
–1
methyl trifluoromethanesulfonate (–11.76 ± 0.12 kcal mol ).
Again it seems wildly unlikely that the more reactive ester
would have the smaller heat of reaction. These experiments
with methyl methanesulfonate, also reported in Table 1, per-
mitted calculation of the heat of reaction for the hypothetical
process [4], based on the heat of reaction of methyl
When the analyses were repeated at lower concentrations
of hypochlorite, and consequently lower concentrations of
1
products, using 500 MHz H NMR spectroscopy, we found
that the amount of methane sulfonate was less than the
amount of acetate. This result requires that there be a pro-
cess consuming methanethiol equivalents which does not lead
©
2000 NRC Canada

Guthrie and Gallant
1297
Table 3. Thermodynamic quantities for compound discussed in this paper.a
Compound
CH SO H
∆Hf(l)
∆Hvapb
∆Hf(g)
S°(g)b
∆Gf(g)
∆Gtb
∆Gf(aq)
–152.39 ± 1.12
–101.20 ± 1.20
–138.63 ± 1.24
17.97 ± 2.0
11.50 ± 0.28
13.96 ± 0.70 –124.67 ± 1.43 82.33 ± 2.0
–134.42 ± 2.29 74.03 ± 2.0
–89.70 ± 1.21 74.76 ± 2.0
–113.29 ± 2.37 –12.73 ± 0.79
–72.80 ± 1.35
–96.30 ± 1.55
–126.02 ± 2.50
–75.58 ± 1.36
–101.57 ± 1.84
3
3
CH SO Cl
–2.97 ± 0.27
–5.27 ± 0.18
3
2
CH SO CH
3
3
3
a
–1
–1
–2
All at 25°C, enthalpies and free energies are in kcal mol , entropies are in cal deg mol , standard states are the ideal gas at 1 atm and 1 M aqueous
solution with an infinitely dilute reference state. Standard entropies in the gas-phase and free energies of transfer (from the gas at 1 atm to 1 M aqueuos
solution) were calculated as described in the text. Free energies of formation in the gas-phase are calculated from the corresponding enthalpies of
formation and the standard entropies, and free energies of formation in aqueous solution are calculated from the corresponding free energies of formation
in the gas and free energies of transfer.
b
Reference (1).
to methanesulfonate, and thus implies that the measured heat
of reaction will be spuriously high.
In other experiments we measured the concentration of
hypochlorite by UV before and after adding S-methyl thio-
acetate, and found that the amount of hypochlorite used was
greater than required for the simple stoichiometry previously
assumed.
as was observed. Sulfite is potentially an oxygen nucleophile
instead of a sulfur nucleophile, but as an oxygen nucleophile
it would be expected to have a Swain–Scott n value, 3.8,
2
−
similar to that of HPO which has a very similar pK . This
4
a
implies a reaction rate only one twentieth that of the sulfur.
In any case NMR analysis of the product solution (for an ex-
periment done in D O) showed only traces of additional
2
Thus it seems clear that the basis for the initial calorime-
try was incorrect, despite apparently clear evidence from a
stoichiometry experiment carried out before these measure-
ments were performed.
products. There was about 2% of a peak which might be
methanol, and an even tinier trace of a peak to lower field.
An amount of competing hydrolysis or reaction at oxygen of
sulfite amounting to no more than this would have an effect
on the heat of reaction less than the observed experimental
error.
NMR analysis of the product solution from the reaction of
sulfite with methyl methanesulfonate shows only the peak
expected for methanesulfonate ion, with no other products
amounting to more than 0.5% of the total.
Why then are the results of reactions with methyl
trifluoromethanesulfonate so out of line with expectations?
The only explanation we have been able to devise is that the
reactions are fast and exothermic, and so may have caused
significant amounts of volatilization of ester before reaction,
leading to anomalously low measured heats of reaction.
Trifluoromethanesulfonic acid is 4 pKa units stronger than
methanesulfonic acid (5), so an initial guess using a
Brønsted exponent of 0.5 suggests that the reactions of
methyl trifluoromethanesulfonate would be 100 times faster
than those of methyl methanesulfonate. Reactions of meth-
anesulfonyl chloride with aqueous hydroxide could not be
studied because of spattering from the exothermic reaction,
leading to obvious scatter and unreacted droplets (1). This
reaction could only be studied in solutions buffered at
pH 10, and this experience guided our use of buffer for the
hydrolysis reaction of methyl trifluoromethanesulfonate.
Methyl trifluoromethanesulfonate boils at 95°C (6), and thus
would volatilize before water, unlike methanesulfonyl chlo-
ride, bp 159 (7); this may explain the difference in behavior.
Discussion
It seems clear that we were mistaken in thinking that we
knew the stoichiometry of the reaction used to determine the
heat of formation of methanesulfonic acid. The distressing
thing is that the stoichiometry had been determined by direct
analysis for methanesulfonate and acetate, admittedly under
slightly different conditions than were used for the actual
calorimetry. The difference, amounting to a factor of 3.3 in
hypochlorite concentration, had not seemed important but
obviously was. In addition, the logical consistency of the re-
sults had been checked by a γ plot (3) (admittedly a novel
extrapolation of the idea of a γ plot, but one which seemed
reasonable). Clearly, this too was misleading. Work in prog-
ress suggests that γ is indeed relevant to these equilibria, but
is not the whole story.
We do not know what reactions could have led to con-
sumption of methanethiolate equivalents so that they did not
become methanesulfonate, but it is tempting to speculate
that under the alkaline reaction conditions elimination of
+
+
HCl from an initial S Cl species could lead to a CH =S
2
species which would quickly release formaldehyde, and
leave a sulfide equivalent. Oxidation of sulfide to sulfate is
strongly exothermic.
The new calorimetric results reported here are for an in-
herently simpler reaction. When methyl trifluoromethane-
sulfonate was added to a 1 M solution of Na SO in D O, to
The present approach to the heat of formation of
methanesulfonic acid, though free of the problems which af-
fected the hypochlorite oxidation procedure, is less general,
since it requires that there be a rapid nucleophilic displace-
2
3
2
give a final concentration of organic material of 0.2 M, the
1
H NMR spectrum showed essentially only one peak, at the
chemical shift expected for methanesulfonate ion. The only
nucleophiles in the solution are water, hydroxide, and sulfite
ion. Hydroxide is at quite low concentration and therefore
should be of negligible significance. The Swain–Scott n
value for sulfite ion is 5.1, (4), while that for water is 0.0.
Thus it seems highly reasonable that the only reaction would
be that of sulfite attacking methyl trifluoromethanesulfonate
ment. Examination of the literature showed that the S 2 re-
N
actions of CH X were not likely to be fast enough for
3
calorimetry without an extremely good leaving group, such
as trifluoromethanesulfonate or methanesulfonate. For alkyl
groups other than methyl, the reaction becomes slower, and
side reactions become increasingly likely as the group gets
larger.
©
2000 NRC Canada

1
298
Can. J. Chem. Vol. 78, 2000
The values reported in the present paper for the heats of
the cooperation of Dr. G. Shaw, of the Departments of Bio-
chemistry and Chemistry at this university.
formation of liquid methanesulfonic acid, methyl
methanesulfonate, and methanesulfonyl chloride are much
closer to the values in the earlier literature than those in our
original paper. The values we now report are respectively
Acknowledgments
We thank the Natural Sciences and Engineering Research
Council of Canada for financial support of this work. We
thank Dr. G. Shaw, Departments of Biochemistry and Chem-
istry, for recording the 500 MHz NMR spectra. We thank
Professors John Bartmess and Joel Liebman for helpful dis-
cussions which provoked the work reported here.
–
152, –139, and –101 as opposed to –178, –164, and –127
–
1
(
1) kcal mol ; Liebman(8) gave –147, –135, and –101 kcal
–
1
mol .
Experimental
References
Materials
1
. J.P. Guthrie, A.R. Stein, and A.P. Huntington. Can. J. Chem.
6, 929 (1998).
S-Methyl thioacetate was a fresh bottle from Lancaster
Chemicals, and was used without further purification. NMR
spectroscopy showed no impurities. Sodium hypochlorite as
a 5% solution in water was Malinckrodt analytical reagent
grade. Methyl trifluoromethanesulfonate was a fresh bottle
from Aldrich, used without further purification. H and
NMR spectra each showed only one peak with the expected
chemical shift. Methyl methanesulfonate was taken from a
7
2. J.W. Larson and L.G. Hepler. In Solvent solute interactions.
Vol. 1. Edited by J.F. Coetzee and C.D. Ritchie. Marcel
Dekker, New York. 1969.
3. E.G. Sander and W.P. Jencks. J. Am. Chem. Soc. 90, 6154
(1968).
4. C.G. Swain and C.B. Scott. J. Am. Chem. Soc. 75, 141 (1953).
5
6. Beilstein. Handbuch der Organischen Chemie. 3(4) 34.
7
8
1
19
F
. J.P. Guthrie. Can. J. Chem. 56, 2342 (1978).
1
fresh bottle (Aldrich 99%). The H NMR spectrum showed
. Beilstein. Handbuch der Organischen Chemie. 4(2) 525.
. J. Liebman. In The chemistry of sulphonic acids, esters, and
their derivatives. Edited by S. Patai and Z. Rappoport. John
Wiley and Sons Ltd., Chichester. 1991. pp. 283.
. S.P. Datta and A.K. Grzybowski. In Biochemist’s handbook.
Edited by C. Long. E. & F.N. Spon Ltd., London. 1961.
pp. 40.
only the two peaks expected for methyl methanesulfonate.
Boric acid, potassium chloride, sodium hydroxide, and so-
dium sulfite were analytical reagent grade. pH 10 borate
buffer was made up following ref. (9).
9
Methods
Reaction calorimetry was carried out as previously de-
scribed (1). H NMR spectra were recorded on Varian Gem-
ini 200 or Unity 500 MHz spectrometers, the latter through
10. D.D. Wagman, W.H. Evans, V.B. Parker, R.H. Schumm, I.
Halow, S.M. Bailey, K.L. Churney, and R.L. Nuttall. J. Phys.
Chem. Ref. Data. Suppl. 2, 11, (1982).
1
©
2000 NRC Canada