|Home | About | Journals | Submit | Contact Us | Français|
Aromatic sulfides bearing a nitro group undergo sulfur oxidation upon electrospray ionization in the positive-ion mode. For example, 2-nitrophenyl phenyl sulfide, its para nitro isomer, and its chloro and methyl substituted analogs pick up an oxygen atom to afford [M + H + O]+ and [M + Na + O]+ ions upon ESI. Elemental-composition determination and tandem mass spectrometry confirm the reactions. Another oxidation of the sulfur, by the ortho nitro group of the [M + H]+ ions, occurs as intramolecular oxygen-transfer processes, evidenced by characteristic losses of SO, SO2 and SO2H•, the latter yielding the carbazole radical cation, and the generation of the aryl-SO+ product ion. The intramolecular oxidation via oxygen transfer from the nitro group to the sulfur was corroborated by molecular modeling. The results substantiate both inter- and intramolecular oxidation and provide more evidence that care must be taken when analyzing not only methionine-containing peptides but also small sulfides.
Electrochemical processes occurring in electrospray ionization (ESI) continue to generate considerable interest among mass spectrometrists and ion chemists [1-3]. The generation of oxidizing agents, likely OH radicals and oxygen molecules, occur in the spray as a consequence of the electrolysis of water molecules [4, 5]. Electrochemical processes also perturb solution pH and consequently the abundances of positive ions generated in the gas phase . The electrochemically-induced oxidative processes can be advantageous, for example, in the generation of fullerene radical cations in the gas-phase, allowing investigation of their ion/molecule reactions . Other examples include the formation of [M - H]+ ions from easily oxidizable benzofuran derivatives , the generation of higher oxidation states of metal ions in dithiocarbamates , and the production of closed-shell ions in the ESI of stable aromatic radicals . The yield of oxidation products impacts ionization efficiency and become greater when the conditions of ESI are modified so as to exploit the inherent electrochemistry by using a porous flow-through electrode emitter .
The most troublesome oxidation in proteomics occurs where the methionine sulfur in peptides and proteins is converted to a sulfoxide, yielding an [M + H + O]+ at 16 m/z higher than expected [11,12]. Experiments conducted by using 18O labeled water as solvent revealed that the source of oxygen to form the methionine sulfoxide is water . We note that some of the example oxidative processes appear to involve the direct abstraction of an electron [6, 8, 9], whereas others [7, 11] involve a process whereby the oxidation number of functional group is raised . For example, in the case of methionine oxidation, the oxidation number of the functional group containing sulfur is raised by an oxygen transfer.
We report here two different oxidation reactions of the sulfur atoms in the 2-nitrophenyl aryl sulfides (the aryl sulfides 1 to 4 investigated here are given in Scheme 1) observed in these molecules ionized by electrospray. The first class of oxidation of sulfur occurs by oxygen addition, which results from electrochemical oxidation inherent in positive-ion ESI, to form presumably a sulfoxide [M + H + O]+ species. A second, more subtle oxidation of sulfur takes place as an intramolecular process, where upon activation either in the ionization process itself or by collisional activation (CA), the sulfur is oxidized and the nitro group is reduced via oxygen transfer between these moieties. Evidence for this latter process comes from accurate mass measurements, studies of reference compounds, MS/MS, and molecular modeling. The results complement our continuing studies of nitro aromatic compounds . Uniting the two processes are fragmentations that raise the oxidation state of the sulfur atom.
Indeed, aromatic sulfides are susceptible to oxidation under mild conditions on the timescale of minutes under microwave irradiation . Intramolecular oxidation of sulfur via oxygen transfer from a proximal nitro group  occurs for radical cations, leading to surprising fragmentations (e.g., loss of SO2 from protonated 2-nitrophenyl-4-tolyl sulphide in EI MS  and the formation from C6H5SO2+ from protonated 2-phenylthio-3-nitropyridine ). (In addition, oxidation of selenium analogs via hydroxyl transfer from a protonated proximal nitro moiety was reported ). A mechanism of elimination of SO2 involving both oxygen and phenyl migration without further cyclization was proposed and buttressed with rudimentary calculations using DFT theory . The purpose of this research is to see if both inter and an analogous intramolecular oxidation occurs for ESI-generated, closed-shell [M + H]+ ions and to explore the mechanisms for the reactions.
Sulfides 1 to 4 (Scheme 1) were synthesized from the appropriate chloronitrobenznes and thiophenol by using reported procedures . Purity of the samples was checked by TLC, and the structures were confirmed by NMR, IR, and mass-spectrometric analyses, as described earlier . Carbazole used for this study was purchased from Aldrich Chemical Co. (Milwaukee. WI) and used without further purification.
ESI experiments were conducted by using a Micromass Q-Tof-Ultima (Waters, Manchester, UK) instrument operated in the positive-ion mode. The metallic needle voltage was 3 kV, and the cone voltage was 90 V. The temperatures of the source block and for desolvation were 90 and 150 °C, respectively. The samples were dissolved in 1:1 mixture of acetonitrile and water containing 1% formic acid. The samples were introduced by direct infusion at a flow rate of 10 μL/min. All parameters (i.e., aperture to the TOF, transport voltage, offset voltages) were optimized to achieve maximum sensitivity and a mass resolving power of 15,000 in the “w” mode (full width at half maximum). The CAD experiments were carried out by mass selecting the precursor ion by using the quadruple analyzer and collisionally activating the selected ion in a quadrupole collision chamber that follows; the product ions were recorded by using the time-of-flight analyzer. Collision voltages for fragmenting the precursor ions were in the range of 7 to 10 V. Accurate masses of the product ions were determined by using the precursor ion as the internal mass standard.
Some ESI MS and low-energy MS/MS and MS3 experiments were performed by using a Thermo Finnigan LCQ Advantage or a Thermo Finnigan LCQ Classic 3D ion-trap mass spectrometers (San Jose, CA).
For high-energy experiments, ions were generated by CI (chemical ionization) using methane as the reagent gas. The CI MI (metastable ion) and high-energy CA experiments were conducted on a VG ZAB-T four-sector mass spectrometer of BEBE design . Samples were introduced through a heated direct-insertion probe with source temperature of 200 °C. MS1 is a standard high-resolving power, double-focusing mass spectrometer (ZAB) of reverse geometry. MS2 possesses a prototype Mattauch-Herzog-type design, incorporating a standard magnet and a planar electrostatic analyzer having an inhomogeneous electric field, followed by single-point and array detectors. Ions were accelerated to 8 kV and subjected to collisions at 4 keV in the third field-free region and detected at the single-point detector. Data acquisition and workup were accomplished by using a VAX 3100 work-station equipped with OPUS software (VG, Manchester, England).
Initial scans for candidates and connections on potential energy surfaces for protonated 2-nitropheyl-phenyl sulfide and sulfoxide were performed by using the PM3 [21, 22] semi-empirical algorithm (Spartan '02 for Linux, Wave function, Inc.) for computational efficiency. Additional scans of the potential surface were performed by using density functional theory (B3LYP). Geometric optimization of candidate minima and transition states were performed by using density functional theory (DFT; B3LYP/6-31G(d, p) level), which requires less computational overhead than do formal ab initio methods and yet incorporates dynamic correlation, has little spin contamination [23-25], and usually performs adequately giving proper geometries, energies, and frequencies . DFT was part of the Gaussian 03/98 suites (Gaussian, Inc.) [27, 28]. The optimized minima and transition states were confirmed by vibration frequency analysis. Connections of transition states to minima were analyzed by combination of inspection, projection along normal reaction coordinates, or reaction-path calculations as needed; also discovered by the reaction-path calculations were ion-dipole complexes. Final single-point energies were calculated at the B3LYP/6-311+G(2d,p)//B3LYP/6-31G(d,p) level, scaled thermal-energy corrections for standard conditions derived from the frequency calculations were applied , and reported as relative enthalpies in kcal/mol.
The ESI mass spectrum of 2-nitrophenyl phenyl sulfide 1 shows an [M + H]+ ion of m/z 232 and another ion of m/z 248, 16 m/z higher (Fig. 1). The measured accurate mass for the ion of m/z 248 is 248.0381, which agrees within 2 ppm of the accurate mass calculated for the [M + H + O]+ ion, C12H10NO3S (Table 1). That ion of m/z 248, [M + H + O]+, is likely the [M + H]+ ion of the sulfoxide generated by an oxygen-transfer process initiated by generation of intermediate oxidizing species at the emitter electrode of the ESI source [4,5]. The ions of m/z 254, 270, and 286 are the analogous [M + Na]+, [M + Na + O]+ likely contaminated with minor isobaric [M + K]+, and [M + K + O]+ ions, respectively, of 1 (Table 1) where Na+ or K+ has replaced the proton as the ionizing agent.
To confirm the structure assigned for the ion of m/z 248, its CAD mass spectrum (Fig. 2a) was compared with that of the authentic sulfoxide of sulfide 1 (Fig. 2b). That standard was produced by a standard oxidation procedure of sulfide 1 by using sodium periodate and introduced to the spectrometer by ESI . The two spectra are virtually identical, confirming the proposed structure produced by ESI-induced electrochemical oxidation of sulfide 1 in the ion source. The elemental compositions of the fragment ions were assigned based on high mass resolving power (~15,000), accurate-mass data (Table 1) obtained in the MS/MS mode, taking the precursor ion as the mass standard.
Calculations by DFT reveal that the preferred site of protonation to form the [M + H]+ of the mono-oxidized species is at the oxygen on the sulfur (Scheme 2), in contrast to one of the nitro-group oxygens as is the case for the [M + H]+ of the unoxidized species . Indeed, the SO site is preferred over the oxygens of the nitro group by > 30 kcal/mol. Major fragment ions, observed in the CAD spectra, are of m/z 217, [M + H - HNO]+; 184, [M + H - SO2]+; 170 [M + H − C6H6]+; and 167 [M + H - (SO2 + OH)]+•. Note that the losses of SO2 and (SO2 + OH•) would require a transfer of an oxygen from the nitro group to the sulfur atom. The overall envelop of fragments is distinct from that observed for the protonated sulfides (Fig. 3), indicating that the sulfoxide, because of a different site for initial protonation, has available to it different routes of fragmentation.
The ESI mass spectra of sulfides 2-4 (Fig. 1b-d, Table 1) show similar features to that for 2-nitrophenyl phenyl sulfide 1 including the ESI-induced oxidation in all cases. The measured accurate masses of the proposed [M + H + O]+ ions (Table 1) agree closely with the expected elemental compositions; also produced by ESI were the adducts, [M + Na]+, [M + Na + O]+, and [M + K + O]+ ions (Table 1). The CAD mass spectra of the ESI-generated [M + H + O]+ ions (Table 2) show that sulfides 1, 3, 4 fragment similarly; whereas, sulfide 2, which has a para nitro group instead, does not undergo the prominent losses of HNO, SO2 and (SO2 + OH•) but instead loses OH• (to give an ion of m/z 231) and NO2•(to give an ion of m/z 202). Thus, the losses of HNO, SO2 and (SO2 + OH•) require the proximal ortho nitro group, and the losses SO2 and (SO2 + OH•) additionally require the transfer of an oxygen, which we view as an intramolecular oxidation. In addition, the [M + H + O]+ ions generated from sulfides 1-3 dissociate by the elimination of a molecule of C6H6 whereas that of sulfide 4, having a methyl group at the 4′- position, expels a molecule of C6H5CH3, consistent with the assignment that they are indeed sulfoxides.
ESI-protonated 2-nitrophenyl phenyl sulfide 1 produces, upon collisional activation, fragment ions of m/z 215, 214, 202, 186, 184 168, 167, 154, 125, 123 (Fig. 3) and m/z 109 (not shown). The ions of m/z 215 and 214 arise by the losses of an OH radical and H2O, respectively; we described the mechanisms of these fragmentations earlier . The unusual ions of m/z 202, 168 and 167 are formed by eliminations of NO•, SO2 and SO2H•, respectively, from the [M + H]+. The product ions of m/z 154, 125, 123, and 109 are C6H4NO2S+, C6H5SO+, C6H5NO2+, and likely C6H5S+, respectively. The isobaric fragment ions of m/z 184 arise by the expulsion of either SO or the elements of H2NO2 from the [M + H]+ ion; selection of m/z 234 as the [M + H]+ ion, using the native 34S isotope, reveals both are major losses. The m/z 186 ion arises from the sequential loss of H2O and CO from the [M + H]+ ion, as previously reported . Furthermore, MS3 studies of the product ions of m/z 184 and 168 reveal further dissociations by losses of OH and H radicals, respectively, to generate m/z 167. Moreover, the CAD mass spectrum of the m/z 167 fragment is virtually identical to that of the radical cation of carbazole, indicating that m/z 167 product ion is indeed the carbazole radical cation and is formed in violation of the even-electron ion rule. Accurate mass measurements of the fragment ions, obtained under high-mass resolving power as described in Experimental Section, verify the formulae of the product ions and losses in the eliminations (Table 3).
For comparison, we produced protonated 2-nitrophenyl phenyl sulfide 1 by CH4 CI (Experimental); this ionization does not produce the [M + O + H]+ ion. The MI spectrum of the CI-generated [M + H]+ ion is shown in Fig. 4; the CAD spectrum of the same precursor is similar but shows enhanced abundances of lower m/z fragment ions (not shown). No matter the ionization mode, CAD produces the same series of fragment ions but in somewhat different abundances. In addition, it appears that the CH4 CI ions are produced with more internal energy that the ESI produced ions.
The abundances of ions from the losses of NO•, SO, SO2, and SO2H• are summarized in Table 3 for the CAD spectra for the ESI-produced [M + H]+ ions of sulfides 1-4. We note that losses of SO, SO2, and SO2H• are absent in the CAD spectra for the ESI-produced [M + H]+ ion of sulfide 2, which has a para rather than an ortho nitro group. Consistent also with oxidation of S are the aryl-SO+ ions formed upon CAD of the [M + H]+ ions of sulfides 1, 3, 4. On this basis, we surmise that the expulsions of SO, SO2, and the SO2H radical from the [M + H]+ ions arise after transfer of one or two oxygen atoms from the nitro group to the sulfur atom rather than by an abstraction in the final elimination step. In addition, given that the m/z 167 ion is the tricyclic carbazole radical cation, it is likely that cyclization takes place as a consequence of this intramolecular oxidation of sulfur, and is driven by the internal energy of the ions. The structure of the m/z 184 ions, which arise by losses of both SO and the elements of NO2H2 (likely H2O + NO), were unresolved with respect to model ions because of lack of necessary precursor ion resolving power. To elucidate further the mechanism of the necessary rearrangements and subsequent eliminations of SO, SO2 and SO2H radical, we carried out molecular orbital calculations by using DFT, as described in Experimental.
We discovered in scanning the potential energy surface at both low level and DFT that there exists a web of potential routes for the intramolecular oxidation of the sulfur by the nitro group and the formation of characteristic product ions. We selected for presentation the ‘best’ of the potential routes based on lowest required energy to surmount the greatest transition-state barriers. We report the calculated enthalpies of formation for the intermediates, products and transition states, relative to that of the previously reported [M + H]+ ion  in kcal/mol.
Proposed mechanism for intramolecular oxidation of sulfur by two sequential oxygen transfers is shown on Scheme 3. The first step involves the transfer of an OH group from the nitro group to the sulfur atom, affording intermediate I (A3). This first step is presumably rate limiting in that it presents the greatest energetic barrier, TS(1-2), in the scheme. The second round of oxygen transfer involves a stable intermediate, A6, which decomposes followed by cyclization ultimately leading to the tricyclic intermediate II (A9), an energetically favorable form (Δ2Hf = -37.1 kcal/mol) in which both oxygens are now attached to the sulfur, and the nitro group has been consequently reduced to a secondary amine.
The route to the expulsions of SO2 and the SO2H radical starts from intermediate II (Scheme 4) and proceeds through the metastasis of a carbon-sulfur bond to a carbon-carbon bond yielding a tricyclic intermediate with a five-membered ring, A13. We propose that this intermediate, which features a long C-S bond of 2.54 Å, can decompose to yield the heterocyclic fragments of m/z 168 and 167 (protonated carbazole and a carbazole radical cation, respectively, the latter being experimentally verified). In addition, the intermediate I, formed by the transfer of one oxygen atom from the nitro group to sulfur, is also the starting point for the formation of the m/z 125 ion (C6H5SO+), as shown in Scheme 5. A critical and significant feature of this route to the product ion involves sequential NO• migrations that are more energetically favorable than any H or OH migration we discovered. Additionally, the exit channel features a prominent ion-neutral complex between the phenyl-SO+ product ion and nitrosobenzene.
Proposed route for SO loss is analogous to that for SO2 loss and involves an intramolecular transfer of oxygen from the protonated nitro moiety to the sulfur via a cyclic intermediate, A20, which is just a shoulder on the transfer pathway (Scheme 6). This is followed by cyclization and proton transfer to form a energetically favorable tricyclic intermediate, A23 (Δ2Hf = -29.0 kcal/mol). The expulsion of SO commences from intermediate A23 via the metastasis of a carbon-sulfur bond to a carbon-carbon bond, yielding a tricyclic intermediate with a five-membered ring, A24, which decomposes by SO loss (Scheme 6) through and ion-dipole complex. We note that in all the proposed schemes, the maximum barrier at a transition state is ~ 40 kcal/mol, which is also consistent with previously proposed mechanisms for other fragmentations of the these sulfides (14).
ESI of 2-nitropheyl phenyl sulfide, the chloro and methyl substituted analogs, and its para nitro isomer all undergo oxygen atom addition in an ESI source during ionization. The products formed are sulfoxides, as characterized by tandem mass spectrometric experiments and accurate mass measurements. Additionally, those ESI-protonated aromatic sulfides containing ortho nitro groups also show oxidation of sulfur by internal oxygen transfer. Evidence includes the characteristic expulsions of SO, SO2 and the SO2H radical, and the production of aryl-SO+. We view this chemistry as a second class of oxidation, an intramolecular oxidation of the sulfur atom with reduction of the nitro group via transfer of oxygen atoms. The mechanisms for transfer and consequent fragmentation were substantiated by molecular modeling using DFT theory.
J.T.M and M.G. thank the KSCSTE for financial assistance and Principal, S. H. College, Thevara for providing infrastructure. Research at WU was supported by the National Centers for Research Resources of the NIH, Grant P41RR000954. This work made use of the Washington University Computational Chemistry Facility, supported by NSF grant CHE-0443501.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Joseph T. Moolayil, Department of Chemistry, Sacred Heart College, Thevara, Cochin, Kerala, India 682013.
M. George, Department of Chemistry, Sacred Heart College, Thevara, Cochin, Kerala, India 682013.
Daryl Giblin, Department of Chemistry, Washington University, One Brookings, Drive, St.Louis, MO, 63130 USA.
Michael L. Gross, Department of Chemistry, Washington University, One Brookings, Drive, St.Louis, MO, 63130 USA.