Catalytic oxidation of alkanethiols and dialkyldisulfides to alkanesulfonic acids by H2O2/CH3ReO3 examined by electrospray ionization mass spectrometry

Full text of DOI:10.1002/jms.3325

JMS letters
Received: 30 September 2013
Revised: 4 December 2013
Accepted: 11 December 2013
Published online in Wiley Online Library
(wileyonlinelibrary.com) DOI 10.1002/jms.3325
Catalytic oxidation of alkanethiols and
dialkyldisulfides to alkanesulfonic acids by
H2O2/CH3ReO3 examined by electrospray
ionization mass spectrometry
Dear Sir,
Moreover, MTO has become one of the most widely used cata-
lysts for oxidation reactions involving H2O2, as it exhibits broad
versatility for facilitating the oxidation of a broad range of
functional groups.[6,7] It is now understood that this occurs via
the reversible formation of the mono- and diperoxorhenium
complexes (CH3Re(O2)O2 and CH3Re(O2)2O, respectively), which
allows efficient O atom transfer to a variety of nucleophilic
targets.[6–11]
Work in this lab has recently focused on developing a method by
which mixed self-assembled monolayers (SAMs) of alkanethiols
on gold surfaces may be quantitatively analyzed by electrospray
ionization mass spectrometry (ESI-MS). The approach that is cur-
rently being developed involves the use of hydrogen peroxide
to oxidatively remove these films for direct infusion into the
electrospray source. The rationale for these investigations, as well
as preliminary studies of the experimental conditions necessary
to produce various alkanesulfonates from their alkanethiol
precursors, has been detailed recently.[1] In short, the goal in
the preliminary studies has been the development of an oxida-
tion process that yields the same sulfur oxidation product for
each structurally distinct alkanethiol present. Such an approach
should facilitate the quantification of complex SAMs that consist
of multiple components.
An approach using MTO has therefore appeared to be a prom-
ising route for eliminating disulfide formation in our own studies
and has the added advantage that it involves only a minor
change to our current methodology (the addition of the catalyst).
Since disulfides are intermediate oxidation products between
thiols and sulfonic acids, we decided to examine whether excess
H2O2 in combination with the catalyst MTO would allow us to
quantitatively convert alkanethiol precursors (both short- and
long-chain species) directly to sulfonic acids without getting
trapped in-between with the semi-stable disulfide intermediate.
The effect of alkyl chain length on the efficacy of this process is
critical given that the focus of our methodology will be studies
of mixed alkanethiol SAMs whose components may differ signif-
icantly in chain length.[1] Thus, the research reported here
involves the application of MTO to the oxidation of both short
and long chain ω-mercaptoalkanoic acids and an examination
of the resulting solutions by ESI-MS to determine the sulfur oxida-
tion products. As mentioned above, we have chosen to employ
ω-mercaptoalkanoic acids as the means by which the general
conversion efficiency is assessed since the terminal carboxylic
acid moiety allows the observation of these products by MS. To
the authors’ knowledge, only one other publication has appeared
involving ESI-MS analyses of MTO-containing solutions, the focus
of which centered primarily on the base-catalyzed hydrolysis of
MTO and the resulting decomposition products.[12]
Recent studies in our lab of the oxidation of ω-
mercaptoalkanoic acids by H2O2 have indicated that not only
is the desired alkanesulfonic acid (1) produced by this process,
but that a disulfide-containing product (2) occurs as well:
HO2CðCH2Þ SH þ 3H2O2 → HO CðCH Þ
þ
3H2O
SO3H
2
2
ð Þ 10
10
1
HO2CðCH2Þ SꢀSðCH2Þ10CO2H
10 ð Þ
2 HO2CðCH2Þ10SH þ H2O2 →
2
þ
2H2O
where (1) was observed by negative-ion ESI-MS as the sulfo-
nate, and (2) was observed as the carboxylate monoanion.[2]
The addition of excess H2O2 did not appear to accelerate ox-
idation of the disulfide product to the sulfonate, as the former
seemed to be semi-stable. For the methyl-terminated
alkanethiols employed in our previous work, formation of
the corresponding disulfide product would not have been
detected by MS due to their being electrically neutral.[1] Thus,
the presence of the carboxylic acid moiety at the chain termi-
nus allows one to probe mass spectrally the creation of various
sulfur oxidation products resulting from this procedure. Since
All reagents for these experiments were used as received:
3,3′-dithiodipropanoic acid (3,3′-DTDPA: Acros Organics, 99%),
3-mercaptopropanoic acid (3-MPA: Aldrich, >99%), 11-
mercaptoundecanoic acid (11-MUDA: Aldrich, 95%), pentane-
1-thiol (C5SH: Aldrich, 98%), octane-1-thiol (C8SH: Aldrich,
98.5%), methanesulfonic acid (C1SO3H: Aldrich, 99.5%),
ethanesulfonic acid (C2SO3H: Aldrich, 95%), hydrogen peroxide
(Sigma-Aldrich, 30% w/w) and methyltrioxorhenium(VII)
(CH3ReO3, MTO: Aldrich, 71–76%). Solutions were freshly
our goal is to produce
(alkanesulfonates) upon exposure to H2O2, the presence of a
disulfide product is clearly undesirable.
a
single oxidation product
In recent years, several groups have shown that the room tem-
perature oxidation of various alkyl- and aryl-substituted thiols
and disulfides to the corresponding sulfonic acids can be accom-
plished using H2O2 if the oxidation is performed in combination
with the catalyst methyltrioxorhenium(VII) (CH3ReO3, MTO).[3–5]
*
Correspondence to: Brian W. Gregory, Department of Chemistry and Bio-
chemistry, Samford University, Birmingham, AL, 35229-2236, USA. E-mail:
bwgregor@samford.edu
J. Mass Spectrom. 2014, 49, 241–247
Copyright © 2014 John Wiley & Sons, Ltd.

JMS letters
prepared in standard volumetric glassware using HPLC grade
acetonitrile (Baker, 99.9%) or ethanol (Aldrich, 99.5%) as the
solvent. Prior to use, all volumetric glassware were cleaned in
0.5–1.0% v/v aqueous Hellmanex solutions (Hellma Worldwide)
and were extensively rinsed afterward with ultrapure water
(resistivity = 18.2 MΩ-cm) provided by a Millipore Synergy water
system.
In replicating the procedure given by Ballistreri et al., three
stock solutions were created (Table 1).[3,4] In most experiments,
Solution 1 was incubated in a thermostatted bath held at 20 °C.
Once equilibrated, Solution 2 was added, yielding a slightly yel-
low, opaque mixture which was stirred for 1–2 min. Solution 3
was subsequently added, and the resulting solution was stirred
for 5–15 min. Afterwards, aliquots were extracted directly from
this reaction mixture for ESI-MS analysis. Dilution of the extrac-
tions (typically 1000-fold) was often performed prior to MS data
collection, given the high concentrations of both H2O2 and ω-
mercaptoalkanoic acid (or dialkyldisulfide) in the solution
mixture. For experiments involving ω-mercaptoalkanoic acids,
addition of Solution 3 into the mixture containing Solutions 1
and 2 usually yielded a white-ish precipitate, which was
attributed to the formation of the corresponding dialkyldisulfide,
since it is the less soluble oxidation product. It was generally
observed that precipitates formed from the short chain ω-
mercaptoalkanoic acid (3-MPA) re-solublized a short time after
precipitation, indicating swift conversion to the alkanesulfonate.
However, the longer chain ω-mercaptoalkanoic acid (11-MUDA)
required higher thermostat temperatures (50 °C) to remain solu-
bilized in acetonitrile and precipitated out readily at room tem-
perature even after conversion to alkanesulfonate. Experiments
using ethanol as the solvent for 11-MUDA had fewer solubility
problems; however, side reactions with the solvent lead to ester-
ification of the COOH group, yielding two product species in the
ESI-MS spectra.
Negative-ion mass spectra were acquired using
a
ThermoFinnigan LCQ Deca XP MAX quadrupole ion trap mass
spectrometer fitted with an orthogonal ESI source running at room
temperature. The ESI source operated using nitrogen (NM30LA Ni-
trogen Generator, Peak Scientific) as the drying/carrier gas. Mass
spectra were normally acquired over a predetermined m/z range
that encompassed the nominal masses of species of interest
(Table 2) and by coadding 100 individual scans taken under opti-
mized conditions. Direct infusion into the ESI source was performed
at various flow rates (5, 10, 50, 100 or 200 μl/min) with the ion trans-
fer capillary temperature set at 350 °C and the ESI source voltage
set at 4.5 kV. To assist in MS peak assignments, collision-induced
dissociation (CID) experiments were performed using ultrapure
He(g) as the collision gas.
Table 1. Stock solutions used to replicate procedure of Ballistreri et al.a
Studies using a short chain dialkyldisulfide: 3,3′-DTDPA
Solution
Constituents
Solvent
Solvent volume
(ml)
As a starting point, we began by examining the effect of H2O2
and the catalyst MTO (MW = 249.24) on the oxidation of a short
chain carboxylic acid-terminated dialkyldisulfide, following the
approach outlined by Ballistreri et al.[3,4] A short chain disulfide
was used in these first studies in order to minimize or eliminate
complications due to the solubilities of either the disulfide reac-
tant or its product(s). The dialkyldisulfide 3,3′-dithiodipropanoic
1
2
3
6.0 mmol H2O2
acetonitrile
acetonitrile
acetonitrile
1.5
1.0
2.5
6.0 μmol MTO
0.60 mmol dialkyldisulfide
or 1.2 mmol alkanethiol
aSee Refs[3] and [4]
.
Table 2. Mass spectral species of interest by negative-ion ESI-MS
Name
Acronym
m/z
3-MPA derived species
ꢀO2C(CH2)2SH (or HO2C(CH2)2Sꢀ)
3-mercaptopropanoate ion
3,3′- dithiodipropanoate, monoanion;
disulfide monoanion of 3-MPA
3-sulfinopropanoic acid
3-MPA
105.14
209.25
ꢀ
HO2C(CH2)2S-S(CH2)2CO2
3,3′-DTDPA
ꢀ
HO2C(CH2)2SO2ꢀ
————
137.13
153.13
HO2C(CH2)2SO3
3-sulfopropanoic acid
3-SPA
11-MUDA derived species
ꢀO2C(CH2)10SH (or HO2C(CꢀH2)10Sꢀ)
HO2C(CH2)10S-S(CH2)10CO2
11-mercaptoundecanoate ion
11,11′-dithiodiundecanoate, monoanion;
disulfide monoanion of 11-MUDA
11-sulfinoundecanoic acid
11-MUDA
217.35
433.60
11,11′-DTDUA
ꢀ
HO2C(CH2)10SO2ꢀ
————
11-SUDA
11-eSUDA
249.35
265.35
293.40
HO2C(CH2)10SO3
11-sulfoundecanoic acid
ꢀ
C2H5OOC(CH2)10SO3
11-sulfo-ethyl-undecanoate ion
Decoꢀmposition products
HSO4ꢀ
HSO3
bisulfate
————
————
————
97.07
81.07
80.07
bisulfite
ꢀ
•SO3
sulfur trioxide radical anion
MTO decomposition products
185ReO4ꢀ
perrhenate anion, 185Re isotope
perrhenate anion, 187Re isotope
————
————
249.24
251.25
ꢀ
187ReO4
wileyonlinelibrary.com/journal/jms
Copyright © 2014 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2014, 49, 241–247

JMS letters
acid (3,3′-DTDPA) was therefore chosen because its terminal
moieties are ionizable, allowing observation of the resultant
oxidation products by ESI-MS. According to Ballestreri′s method,
the use of H2O2 coupled with MTO as a catalyst should facilitate
conversion of 3,3′-DTDPA to its fully oxidized form, 3-
sulfopropanoic acid (3-SPA):
in solutions containing excess H2O2 but without the catalyst.[2]
Therefore, given the low signal intensity of the parent
monoanion, it is not unexpected that its cleavage product is very
weak. Also observed are the typical alkylsulfonate decomposition
products bisulfite (m/z 81) and bisulfate ion (m/z 97). Additionally,
anions associated with the primary decomposition product of the
catalyst (perrhenate ion, ReOꢀ4 ) are discernible in the data and
have a relative intensity ratio reflecting the two primary Re
isotopes: 185ReO4ꢀ (m/z 249) and 187ReOꢀ4 (m/z 251), where the
natural abundances of 185Re and 187Re are 37.4% and 62.6%,
respectively.[12] Other low intensity species in Fig. 1A remain
unassigned at this time.
HOOCðCH2Þ2S–SðCH2Þ2COOH þ 5H2O2 → 2HOOCðCH2Þ2SO3H þ 4H2O
3; 3′-DTDPA
3-SPA
Figure 1A displays the negative-ion ESI-MS spectrum of the
resulting 3,3′-DTDPA + H2O2 + catalyst mixture; nominal masses
for the species of interest are listed in Table 2. The data in this fig-
ure were obtained using a stock solution that was prepared with
0.6026 mmol 3,3′-DTDPA (0.1267 g), 5.96 mmol H2O2 (0.6757 g of
30% w/w solution) and 6.4 μmol MTO (0.0016 g ≅ two to three
small crystals). In this figure, only the fully oxidized form 3-SPA
is observed with reasonable intensity, which occurs as the base
peak (m/z 153). This indicates that the catalyst (in the presence
of H2O2) converts the disulfide precursor to the sulfo-derivative
with high efficiency. Note that neither the parent monoanion of
3,3′-DTDPA (m/z 209) nor 3-mercaptopropanoate ion (3-MPA,
m/z 105) exhibits strong intensities in this figure. The species 3-
MPA is thought to be a disulfide cleavage product of the parent
monoanion of 3,3′-DTDPA. In previous work with ω-
mercaptoalkanoic acids, we have observed such product species
Studies using a short chain ω-mercaptoalkanoic acid: 3-MPA
As a follow-up to the work with 3,3′-DTDPA, another set of
experiments was devised which used the compound 3-
mercaptopropanoic acid (3-MPA). 3-MPA was chosen because it
is the thiol precursor of 3,3′-DTDPA. In replicating the published
procedure with 3-MPA, three stock solutions were again created
following the previous procedure. However, in these studies,
the molar quantity of 3-MPA was doubled compared to that for
3,3′-DTDPA so that approximately the same concentration of
alkylsulfonate would be produced.
Figure 1B displays the negative-ion ESI-MS spectrum of the 3-
MPA + H2O2 + catalyst mixture; nominal masses for the species
of interest are listed in Table 2. The data in this figure were
obtained using a stock solution that was prepared with 1.136
mmol 3-MPA (100.0 μl neat soln), 6.03 mmol H2O2 (0.6840 g of
30% w/w solution) and 6.4 μmol MTO (0.0016 g). Although the
base peak is the sulfo-derivative 3-SPA (m/z 153), a small quantity
of the disulfide product 3,3′-DTDPA (m/z 209) as well as the
sulfino-derivative (m/z 137) appears, suggesting that oxidation
was incomplete at the time the data was taken. As before, resid-
ual signals from the decomposition byproducts bisulfite and
bisulfate, as well as those of MTO, were also observed. Therefore,
Figs. 1A,B together indicate that if the solubilities of reactants,
intermediates or products are not limiting factors, short chain
disulfides and thiol-labeled precursors can be converted directly
to sulfonic acids with high efficiency via peroxide oxidation using
MTO as a catalyst.
Studies using a long chain ω-mercaptoalkanoic acid:
11-MUDA
A third set of experiments was performed to test the oxidation ef-
ficiency of this process using a longer chain ω-mercaptoalkanoic
acid: 11-mercaptoundecanoic acid (11-MUDA). Given the chemical
similarities between 11-MUDA and 3-MPA, catalytic oxidation of
the thiol moiety to a sulfonic acid was expected to proceed in a
similar fashion for both compounds.
In replicating the procedure with 11-MUDA, three stock solu-
tions were again created. In this case, however, the solutions
containing 11-MUDA and H2O2 were mixed first and heated in
a water bath for 20 min at 50 °C; afterwards, the solution was re-
moved from the bath, the catalyst was added and the heating
was continued for another 20 min. While the solution mixture
was outside the bath, a white precipitate spontaneously formed,
which then re-dissolved when the mixture was placed back in the
bath. When the solution mixture (with catalyst) was finally re-
moved from the bath, the precipitate reformed upon cooling.
Initial suspicions focused on the formation of the intermediate
Figure 1. Negative-ion ESI-MS spectra of (A) 3,3′-DTDPA and (B) 3-MPA
following treatment with H2O2 + MTO in acetonitrile, taken at a direct
infusion rate of 50 and 5 μl/min, respectively. Peak assignments are made
based on the expected m/z ratios listed in Table 2.
J. Mass Spectrom. 2014, 49, 241–247
Copyright © 2014 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/jms

JMS letters
as the source of the precipitate: the disulfide monoanion of 11-
MUDA (11,11′-DTDUA). Samples from both the supernatant and
precipitate were subsequently analyzed by ESI-MS to determine
their compositions.
However, esterification of the carboxylate group via reaction with
the ethanol was anticipated and thought possibly to lead to an
additional MS peak from the ethyl ester of 11-MUDA: 11-sulfo-
ethyl-undecanoate ion (11-eSUDA).
Negative-ion ESI-MS spectra of both the supernatant (Fig. 2A)
and precipitate (Fig. 2B) were obtained from the 11-MUDA +
H2O2 + catalyst mixture in acetonitrile. The data in these figures
were obtained using a stock solution that was prepared with
0.6068 mmol 11-MUDA (0.2621 g), 6.07 mmol H2O2 (0.6882 g of
30% w/w solution) and two to three small crystals of the MTO cat-
alyst. For the supernatant, data were obtained following a 1000-
fold dilution in acetonitrile; for the solid, a small portion was
dissolved in ethanol immediately prior to data collection. The
product species 11-SUDA (m/z 265) was found to be the predom-
inant product species in both the supernatant and precipitate, in-
dicating that its solubility in acetonitrile is quite limited. A
negligibly small quantity of the disulfide 11,11′-DTDUA appeared
only in the spectra obtained from the solid.
An additional experiment was performed using 11-MUDA
dissolved in ethanol (rather than acetonitrile) in order to verify
its utility as a solvent for this process. Ethanol is more polar than
acetonitrile and was expected to solubilize the parent compound
and its oxidation products more easily. Furthermore, it is an ESI-
friendly solvent commonly used in MS studies of anions.
As expected, mass spectral results from the reagent solution
above indicated the presence of a small amount of 11-SUDA
(m/z 265) and a large amount of 11-eSUDA (m/z 293) (Fig. 3).
Despite the reaction with ethanol, the conversion efficiency
for producing only the sulfonic acid product again appeared
to be quite high. In order to verify the structure of 11-eSUDA,
CID was performed on the peak observed at m/z 293. Using
40% collision energy (He gas) on the isolated m/z 293 peak,
two CID peaks were observed, with the primary one located
at m/z 265 (11-SUDA) and a minor one at m/z 247 (data not
shown). The apparent mass difference between the parent
and primary collision product peaks was 28, corresponding to
a mass loss Δm = (–CH2CH3 + H) and identifying the feature
at m/z 293 as the ethyl ester of 11-SUDA (m/z 265). It is
important to note that esterification via side reaction with
alcoholic solvents is only problematic for COOH-terminated
alkanethiols and dialkyldisulfides. Such products are not
expected to form from compounds bearing –CH3, –NH2, –OH
or a number of other terminal functionalities.
Alkylsulfonate conversion efficiencies relative to prior work
Given the apparent effectiveness of the catalyst MTO for facilitat-
ing the peroxide-based oxidation of both thiols and disulfides, it
became necessary to examine the relative conversion efficiencies
of our previous approach without the catalyst to this new
procedure with the catalyst. To do this, a series of stock solutions
containing the catalyst were prepared in a manner that would
yield reagent concentrations comparable to those described in
the experiments above. These stock solutions would then be
used to generate a sequence of serially diluted solutions from
which the calibration data would be obtained.[1] Table 3 displays
the content of these various solutions. The initial stock solutions
(labeled O1 and O2) were used to create two precursor solutions
(A1 and D1, Table 3), from which the sequentially diluted series
would be created. The serial dilution scheme (Table 3) was
Figure 2. Negative-ion ESI-MS spectra of the (A) supernatant and (B)
precipitate taken from an 11-MUDA sample containing H2O2 and MTO
in acetonitrile, taken at direct infusion rate of 20 μl/min. The spectrum
for the precipitate was obtained after dissolving a small portion of the
solid in ethanol immediately prior to data collection. Peak assignments
are made based on the expected m/z ratios listed in Table 2.
Figure 3. Negative-ion ESI-MS spectra of 11-MUDA following treatment
with H2O2 + MTO in ethanol, taken at direct infusion rate of 20 μl/min. The
stock solution was prepared using the same quantities of reagents as in Fig. 2.
Peak assignments are made based on the expected m/z ratios listed in Table 2.
wileyonlinelibrary.com/journal/jms
Copyright © 2014 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2014, 49, 241–247

JMS letters
Table 3. Composition of stocka and precursorb solutions, and serial dilution scheme
Solution
Constituents
Solvent
Volume used (μl)
CnSH conc. (M)
Stock O1
C8SH (neat solution)
C5SH (neat solution)
H2O2 (neat solution)
MTO (solid)
acetonitrile
195
60
0.2214
0.0948
870
———
3 crystals
60
———
Stock O2
C5SH (neat solution)
H2O2 (neat solution)
MTO (solid)
acetonitrile
0.0948
870
———
3 crystals
125
———
Precursor A1
Stock Solution O1
ethanol
ethanol
2.768×10ꢀ3 (C8SH)
1.185×10ꢀ3 (C5SH)
1.185×10ꢀ3 (C5SH)
Precursor D1
Stock Solution O2
625
Serial dilution scheme
Final concentrations (mM)
Solution
C1
Mixture
C8SH
C5SH
1.185
1.185
1.185
1.185
1.185
1.185
1.185
1.185
1.185
1.185
5.0 ml A1 + 5.0 ml D1
5.0 ml C1 + 5.0 ml D1
5.0 ml C2 + 5.0 ml D1
5.0 ml C3 + 5.0 ml D1
5.0 ml C4 + 5.0 ml D1
5.0 ml C5 + 5.0 ml D1
5.0 ml C6 + 5.0 ml D1
5.0 ml C7 + 5.0 ml D1
5.0 ml C8 + 5.0 ml D1
5.0 ml C9 + 5.0 ml D1
ethanol
ethanol
ethanol
ethanol
ethanol
ethanol
ethanol
ethanol
ethanol
ethanol
1.384
C2
0.692
C3
0.346
C4
0.173
C5
0.0865
0.0433
0.0216
0.0108
0.0054
0.0027
C6
C7
C8
C9
C10
aStock solutions O1 and O2 each had a total volume of 5.0 ml.
bPrecursor solutions A1 and D1 had total volumes of 10.0 and 50.0 ml, respectively.
devised so that the resulting solutions were nearly identical in
composition compared to those in our previous publication.[1]
Using a 20-μl injection loop, each sample was analyzed eight to
ten times by ESI-MS and the integrated peak areas for octane-1-
sulfonate (C8SO3ꢀ: m/z 193) and pentane-1-sulfonate (C5SO3ꢀ: m/z
151) were evaluated, following our previous approach.[1] It is
important to note that in using these methyl-terminated
alkanethiols, the formation of dialkyldisulfides would not be
observed directly by ESI-MS (given their electroneutrality). From the
MS data, two sets of scatter plots were generated: (1) the logarithm
of the integrated signal of C8SO3ꢀ as a function of logarithm of the
C8SH concentration (Fig. 4A) and (2) the logarithm of the integrated
ion ESI-MS spectra obtained from two methanolic, unbuffered
alkylsulfonate mixtures consisting of: (1) two highly soluble short
chain species (Fig. 5A: 0.491 mM C1SO3ꢀ + 0.466 mM C2SO3ꢀ) or
(2) one short chain and one longer chain species (Fig. 5B: 0.122
mM C2SO3ꢀ + 0.114 mM C8SO3ꢀ). The short chain alkylsulfonates
(C1SO3ꢀ and C2SO3ꢀ) were prepared from commercial sources of
methanesulfonic acid and ethanesulfonic acid, respectively,
rather than using the H2O2/MTO oxidation procedure described
above. Note that in both of these figures, not only do the mono-
mers appear (C1SO3ꢀ: m/z 95; C2SO3ꢀ: m/z 109; C8SO3ꢀ: m/z 193),
but also their corresponding H+-bound homodimers and
heterodimers. In Fig. 5A, the H+-bound homodimers result in
two (2M+1) peaks occurring at m/z 191 and m/z 219, whereas
the H+-bound heterodimer results in a (M1+M2+1) peak at m/z
205. In Fig. 5B, the H+-bound homodimers result in two (2M+1)
peaks occurring at m/z 219 and m/z 388, whereas the H+-bound
heterodimer results in a (M1+M2+1) peak at m/z 303. Under the
conditions employed in these figures, the homo- and
heterodimer MS intensities are as large (or larger) than those of
their parent monomers, which has a detrimental effect on the in-
tensities of the corresponding monomers. In the case where there
is a significant difference in chain length (C2SO3 + C8SO3 mix-
ture, Fig. 5B), the intensities of both the C8SO3ꢀ homodimer and
the heterodimer are comparable to that for the C8SꢀO3ꢀ monomer
and are significantly more intense than the C2SO3 homodimer.
This suggests that for alkylsulfonate mixtures having significantly
different chain length components, the impact on MS intensities
would be more deleterious for the longer chain species (presum-
ing their ionization efficienꢀcies are coꢀmparable). Such would likely
be the case for the C5SO3 + C8SO3 mixtures examined above
and as shown in Fig. 4.
ꢀ
signal of C5SO3 as a function of solution number (C1, C2, etc.)
(Fig. 4B). Note that C5SH (converted to C5SO3ꢀ) acts as an internal
standard, and therefore its concentration across solutions C1 to
C10 should not vary; consequently, its signal stays relatively con-
stant (Fig. 4B). Compared to our previous approach without the
catalyst, a tenfold increase in the absolute C8SO3ꢀ signal was ob-
served across the range of concentrations investigated (Fig. 4Aꢀ),
whereas a 100-fold increase appeared for the absolute C5SO3
signal (Fig. 4B). The large increases in MS signal for both species
are consistent with the qualitative data discussed above and indi-
cate significantly greater alkanethiol-to-alkanesulfonate conver-
sion efficiencies.
ꢀ
ꢀ
It is notable, however, that there is an ordeꢀr-of-magnitude
difference in signal increase observed for C5SO3 compared to
C8SO3ꢀ. This difference is likely due to chain-length-dependent
association phenomena in the electrospray source via the forma-
tion of H+-bound alkylsulfonate adducts (dimers, trimers, etc.),
the occurrence of which has been documented previously.[1] As
examples of this association behavior, Fig. 5 displays negative-
J. Mass Spectrom. 2014, 49, 241–247
Copyright © 2014 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/jms

JMS letters
Figure 5. Negative-ion ESI-MS spectra of (A) 0.491 mM methanesulfonate
(C1SO3ꢀ: m/z 95) + 0.466 mM ethanesulfonate (C2SO3ꢀ: m/z 109) and (B)
ꢀ
0.122 mM C2SO3 (C2SO3ꢀ: m/z 95) + 0.114 mM octane-1-sulfonate
(C8SO3ꢀ: m/z 193) in methanol, both taken at a direct infusion rate of 200
μl/min. The various alkylsulfonate monomers, as well as H+-bound homo-
and heterodimers, are labeled according to their chain lengths.
resulting from MTO-catalyzed oxidation of short- and long-chain
alkanethiols and dialkyldisulfides by hydrogen peroxide. This has
been accomplished in negative-ion mode by employing carboxylic
acid-terminated derivatives of these species. In all cases, MTO-
catalyzed oxidation of both thiols and disulfides proceeds with
high efficiency to the corresponding sulfonic acid products.
Measurements when MTO is used (compared to without) indi-
cate that shorter chain alkylsulfonates exhibit a significantly
greater increase in their integrated MS intensity compared to
longer chain species following oxidation. This has been
attributed to the formation of H+-bound alkylsulfonate adducts
in the electrospray ꢀsource. These adducts decrease the MS
signals for all CnSO3 monomers but are more deleterious for
longer-chain species.
Figure 4. Integrated peak areas observed by ESI-MS for (A) octane-1-
sulfonate (C8SO3ꢀ: m/z 193) and (B) pentane-1-sulfonate (C5SO3ꢀ: m/z
151) in a sequence of serially diluted solutions containing both species.
Points represent the average integrated area measured eight to ten
times from each solution using a 20-μl injection loop. Note that C5SH
(converted to C5SO3ꢀ) acts as an internal standard, and therefore its con-
centration across solutions C1 to C10 should remain constant. See Table 3
for a list of the concentrations of both species in each solution. ESI
source voltage = 4.5 kV; ion transfer capillary temperature = 350 °C.
Acknowledgements
Furthermore, previous studies have shown that the MS intensi-
ties of these species depend on the ion transfer capillary temper-
ature in a manner that is sensitive to their masses.[1] In this work,
the intensities of H+-bound alkylsulfonate adducts have been
characterized by their having an optimal capillary temperature
which increases with the adduct mass; decreasing intensities
above the optimal temperature for a given adduct appear to be
preferentially driven by intermolecular dissociative processes.[1]
Such a mechanism will tend to favor a greater recovery in the sig-
nal intensity for shorter chain monomers in comparison to the
longer chain varieties. At the capillary temperature employed in
the present work (350 °C), thermally driven adduct dissociation
is therefoꢀre expected to benefit the integrated signal intensities
This work was supported by a Major Research Instrumentation
grant from the National Science Foundation (Grant No. 0619217)
and by Samford University through the John R. Sampey, Jr.
Research Fund. Acknowledgment is also made to the Donors of
the American Chemical Society Petroleum Research Fund for sup-
port of this research (Grant No. 49516-UR5).
Yours,
Austin B. Murphy, Chi N. Duong, Karmalita K. Crenshaw,
Savannah K. Hartman, William F. Barrett IV,
Alyssa R. Whitehead, Andrew J. Lampkins†
and Brian W. Gregory*
ꢀ
for C5SO3 monomers compared to the C8SO3 monomers. Cur-
rent efforts are directed toward minimizing or eliminating associ-
ation phenomena in these systems by optimizing the solution pH
and composition as well as rate of infusion.
In summary, we have been able to use ESI-MS to examine the
sulfur oxidation products and alkylsulfonate conversion efficiencies
a
Department of Chemistry and Biochemistry, Samford
University, Birmingham, AL 35229-2236, USA
† Current address: Department of Chemistry, University
of Evansville, Evansville, IN, 47722, USA
wileyonlinelibrary.com/journal/jms
Copyright © 2014 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2014, 49, 241–247

JMS letters
[7] M. Crucianelli, R. Saladino, F. De Angelis. Methyltrioxorhenium catal-
ysis in nonconventional solvents: A great catalyst in a safe reaction
medium. Chem. Sus. Chem. 2010, 3, 524.
[8] W. A. Herrmann, R. W. Fischer, W. Scherer, M. U. Rauch.
Methyltrioxorhenium(VII) as catalyst for epoxidations. Structure of
the active species and mechanism of catalysis. Angew. Chem. Intl.
Ed. Eng. 1993, 32, 1157.
References
[1] R. Roberts, J. A. Driver, D. M. Brown, S. H. Amin, B. W. Gregory. Hydro-
gen peroxide-induced oxidation of mixtures of alkanethiols and
their quantitative detection as alkanesulfonates by electrospray ion-
ization mass spectrometry. Anal. Chem. 2011, 83, 9605.
[2] A. B. Murphy, C. N. Duong, K. K. Crenshaw, W. F. Barrett, IV, B. W. Gregory.
Unpublished results.
[9] S. Yamazaki, J. H. Espenson, P. Huston. Equilibria and kinetics of the
reactions between hydrogen peroxide and methyltrioxorhenium in
[3] F. P. Ballistreri, G. A. Tomaselli, R. M. Toscano. Selective and mild
aqueous perchloric acid solution. Inorg. Chem. 1993, 32, 4683.
[10] J. H. Espenson, M. M. Abu-Omar. In Electron Transfer Reactions: Inor-
ganic, Organometallic, and Biological Applications, Advances in
Chemistry Series, Vol. 253 S. S. Isied(Ed.). American Chemical Society:
Washington, DC, 1997, Chp. 7, pp. 99–134.
[11] M. M. Abu-Omar, P. J. Hansen, J. H. Espenson. Deactivation of
methylrhenium trioxide-peroxide catalysts by diverse and compet-
ing pathways. J. Am. Chem. Soc. 1996, 118, 4966.
oxidation of thiols to sulfonic acids by hydrogen peroxide catalyzed
by methyltrioxorhenium. Tet. Lett. 2008, 49, 3291.
[4] F. P. Ballistreri, G. A. Tomaselli, R. M. Toscano. Selective oxidation re-
actions of diaryl- and dialkyldisulfides to sulfonic acids by CH3ReO3/
hydrogen peroxide. Tet. Lett. 2009, 50, 6231.
[5] Y. Wang, J. H. Espenson. Oxidation of symmetric disulfides with
hydrogen peroxide catalyzed by methyltrioxorhenium(VII). J. Org.
Chem. 2000, 65, 104.
[12] J. H. Espenson, H. Tan, S. Mollah, R. S. Houk, M. D. Eager. Base hydrolysis of
methyltrioxorhenium. The mechanism revised and extended: A novel ap-
plication of electrospray mass spectrometry. Inorg. Chem. 1998, 37, 4621.
[6] C. C. Romão, F. E. Kühn, W. A. Herrmann. Rhenium(VII) oxo and imido com-
plexes: Synthesis, structures, and applications. Chem. Rev. 1997, 97, 3197.
J. Mass Spectrom. 2014, 49, 241–247
Copyright © 2014 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/jms