Fragmentation Studies of Fulvic Acids Using Collision Induced

Mar 5, 2009 - ... Amy M. McKenna , Ryan P. Rodgers , Alan G. Marshall , and William T. ... Steven C. Habicht , Mingkun Fu , Ryan C. Shea , Allen B. Mo...
0 downloads 0 Views 347KB Size
Download PDF
Anal. Chem. 2009, 81, 2688–2694

Fragmentation Studies of Fulvic Acids Using Collision Induced Dissociation Fourier Transform Ion Cyclotron Resonance Mass Spectrometry Matthias Witt,*,† Jens Fuchser,† and Boris P. Koch‡,§ Bruker Daltonik GmbH, Fahrenheitstrasse 4, D-28359 Bremen, Germany, Alfred Wegener Institute for Polar and Marine Research, Am Handelshafen 12, D-27570 Bremerhaven, Germany, and University of Applied Sciences, An der Karlstadt 8, D-27568 Bremerhaven The complex natural organic matter standard Suwannee river fulvic acid (SRFA) was analyzed by negative ion mode electrospray ionization Fourier transform ion cyclotron resonance mass spectrometry (ESI FTICR MS) using onresonance collision induced dissociation (CID) of single ultrahigh resolved mass peaks in the ICR cell. Molecular formula assignment of precursor masses resulted in exactly one molecular formula for each of the peaks. Analyses of the corresponding fragment spectra and comparison to different standard substances revealed specific neutral losses and fragmentation patterns which result in structures consisting of a high degree of carboxyl- and fewer hydroxyl groups. The comparison of fragmented mass peaks within different pseudohomologous series (CH2-series, and CH4 vs O exchange) suggested structurally based differences between these series. CID FTICR MS allowed isolating single mass peaks in a very complex natural organic matter spectrum. Subsequently, fragmentation gave structural insights into this material. Our results suggest that the structural diversity in complex humic substances is not as high as expected. Humic substances are a complex and polydisperse mixture of degraded organic compounds which are essentially derived from plant material. The structure of these compounds is of high interest due to their important role in the transport and deposition of organic and inorganic compounds in soil and water. Knowledge about the molecular structure of humic substances is beneficial for the understanding of global organic carbon fluxes. Despite their important role in global element cycles, the molecular structure of these compounds is largely unknown.1 Fulvic acids are a subfraction of humic substances which are soluble in water * Corresponding author. Phone: +49 421 2205 268. Fax: +49 421 2205 107. E-mail: [email protected]. † Bruker Daltonik GmbH. ‡ Alfred Wegener Institute for Polar and Marine Research. § University of Applied Sciences. (1) Denman, K. L.; Brasseur, G.; Chidthaisong, A.; Ciais, P.; Cox, P. M.; Dickinson, R. E.; Hauglustaine, D.; Heinze, C.; Holland, E.; Jacob, D.; Lohmann, U.; Ramachandran, S.; da Silva Dias, P. L.; Wofsy, S. C.; Zhang, X. Couplings Between Changes in the Climate System and Biogeochemistry. In Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change; Solomon, S., Qin, D., Manning, M., Chen, Z., Marquis, M., Averyt, K. B., Tignor, M., Miller, H. L., Eds.; Cambridge University Press: Cambridge, U.K., 2007.

2688

Analytical Chemistry, Vol. 81, No. 7, April 1, 2009

at low pH. Suwannee river fulvic acid (SRFA) is a well studied reference of humic substances provided by the International Humic Substances Society (IHSS). SRFA is known as an extremely complex mixture of organic compounds and mainly consists of thousands of different CxHyOz compounds.2-5 Because of the high molecular diversity of this mixture, the isolation of single compounds using conventional techniques such as reverse-phase liquid chromatography,6,7 capillary electrophoresis,8 or other separation methods like isoelectric focusing9 or size exclusion chromatography10-12 is time-consuming if successful at all. Fulvic acids can be analyzed with mass spectrometry using electrospray ionization (ESI).13-15 The elemental composition of single compounds of humic substances can be analyzed using ultrahighresolution mass spectrometry, e.g., Fourier transform ion cyclotron resonance (FTICR) mass spectrometry.16-25 This analytical (2) Stenson, A. C.; Marshall, A. G.; Cooper, W. T. Anal. Chem. 2003, 75, 1275– 1284. (3) Averett,R. C.; Leenheer, J. A.; McKnight, D. M.; Thorn, K. A. U.S. Geological Survey Water-Supply, Paper 2373; 1994; pp 13-19. (4) Benner, R. Chemical Composition of Isolated Fractions of DOM. In Biogeochemistry of Marine Dissolved Organic Matter; Hansell, D. A., Carlson, C. A., Eds.; Academic Press: San Diego, CA, 2002. (5) Leenheer, J. A.; Wershaw, R. L.; Brown, G. K.; Reddy, M. M. Appl. Geochem. 2003, 18, 471–482. (6) Saleh, F. Y.; Chang, D. Y. Sci. Total Environ. 1987, 62, 67–74. (7) Fievre, A.; Solouki, T.; Marshall, A. G.; Cooper, W. T. Energy Fuels 1997, 11, 554–560. (8) Garrison, A. W.; Schmitt, P.; Kettrup, A. Water Res. 1995, 29, 2149–2159. (9) Schmitt, P.; Garrison, A. W.; Freitag, D.; Kettrup, A. Water Res. 1997, 31, 2037–2049. (10) Chin, Y.-P.; Aiken, G.; O’Loughlin, E. Environ. Sci. Technol. 1994, 238, 1853–1858. (11) Wagoner, D. B.; Christman, R. F.; Cauchon, G.; Paulson, R. Environ. Sci. Technol. 1997, 31, 937–941. (12) Klaus, U.; Pfeiffer, T.; Spiteller, M. Environ. Sci. Technol. 2000, 34, 3514– 3520. (13) Persson, L.; Bastviken, D.; Alsberg, T.; Tranvik, L.; Odham, G. Int. J. Environ. Anal. Chem. 2005, 13, 15–27. (14) Leenheer, J. A.; Rostad, C. E.; Gates, P. M.; Furlong, E. T.; Ferrer, I. Anal. Chem. 2001, 73, 1461–1471. (15) Koch, B. P.; Ludwichowski, K.-U.; Kattner, G.; Dittmar, T.; Witt, M. Mar. Chem. 2008, 111, 233–241. (16) Sleighter, R. L.; Hatcher, P. G. Mar. Chem. 2008, 110, 140–152. (17) Reemtsma, T.; These, A.; Linscheid, A.; Leenheer, J.; Spitzy, A. Environ. Sci. Technol. 2008, 42, 1430–1437. (18) Stenson, A. C.; Landing, W. M.; Marshall, A. G.; Cooper, W. T. Anal. Chem. 2002, 74, 4397–4409. (19) Kujawinski, E. B.; Behn, M. D. Anal. Chem. 2006, 78, 4363–4373. (20) Kujawinski, E. B. Environ. Forensics 2002, 3, 207–216. (21) Kim, S.; Kaplan, L. A.; Benner, R.; Hatcher, P. G. Mar. Chem. 2004, 92, 225–234. 10.1021/ac802624s CCC: $40.75  2009 American Chemical Society Published on Web 03/05/2009

technique provides mass accuracies better than 0.5 ppm and a resolving power of more than 300 000 at m/z 400 needed for mass peak separation of very complex mixtures like natural organic matter. Therefore, FTICR mass spectrometry has become an important technique to study the elemental composition of humic substances. Nevertheless, the exact structure of these compounds is largely unknown. The high degree of oxygen incorporated in these organic molecules is predominantly bound in hydroxyl and carboxyl groups.4,5 The fragmentation of these molecules using collision induced fragmentation (CID), infrared multiphoton dissociation (IRMPD), or other fragmentation methods in the FTICR MS analyzer cell can give important insight into the possible structures of the ionized molecules. So far, it was only possible to isolate SRFA ions in a 1 Da mass window.18 Mass differences of peaks in complex humic material usually are as low as 36.4 mDa (mass difference of CH4 and O), 15.3 mDa (exact mass difference between C4 and O3), or even lower.26 Therefore, MS/ MS experiments in a 1 Da window include several parent ions biasing the structural interpretation of single compounds. Nevertheless, parent and product ions were linked together due to exact mass differences by FTICR mass spectrometry.18 In this study, we were able to isolate single mass peaks with very small mass difference to its nearest neighbors. Isolation was performed in the ICR cell using single shot ejections after broadband ejection. Fragmentation of four different standard compounds facilitated the interpretation of the fragment spectra and structure proposals. EXPERIMENTAL SECTION A total of 0.1 mg of SRFA (IHSS, catalog no. 2S101F) was dissolved in 1 mL of 50% MeOH/water (v/v) for electrospray ionization in negative ion mode. The mass analysis was performed with a Bruker Apex ultra Fourier transform ion cyclotron resonance mass spectrometer (Bruker Daltonics, Billerica, MA) equipped with a 9.4 T refrigerated actively shielded superconducting magnet (Bruker Biospin, Wissembourg, France). All sample solutions were continuously infused into an electrospray ion source equipped with an ion funnel guide using a syringe pump at a flow rate of 120 µL h-1 using negative ion mode. The precursor masses were preisolated with an external quadrupole (window of 20 mass units) and stored in a collision cell consisting of a hexapole for continuous accumulation of selective ions (CASI) for 1 s to increase the S/N of the parent ion. These preselected ions were transferred to the Infinity ICR cell27 and isolated with a 1 Da window applying correlated harmonic excitation field (CHEF)28,29 isolation. With the excep(22) Kim, S.; Kramer, R. W.; Hatcher, P. G. Anal. Chem. 2003, 75, 5336–5344. (23) Koch, B. P.; Witt, M.; Engbrodt, R.; Dittmar, T.; Kattner, G. Geochim. Cosmochim. Acta 2005, 69, 3299–3308. (24) Hertkorn, N.; Benner, R.; Frommberger, M.; Schmitt-Kopplin, P.; Witt, M.; Kaiser, K.; Kettrup, A.; Hedger, J. I. Geochim. Cosmochim. Acta 2006, 70, 2990–3010. (25) Kujawinski, E. B.; Hatcher, P. G.; Freitas, M. A. Anal. Chem. 2002, 74, 413–419. (26) Hertkorn, N.; Frommberger, M.; Witt, M.; Koch, B. P.; Schmitt-Kopplin, Ph.; Perdue, E. M. Anal. Chem. 2008, 80, 8908–8019. (27) Caravatti, P.; Allemann, M. Org. Mass Spectrom. 1991, 26, 514–518. (28) Heck, A. J. R.; De Koning, L. J.; Pinkse, F. A.; Nibbering, N. M. M. Rapid Commun. Mass Spectrom. 1991, 5, 406–414. (29) de Koning, L. J.; Nibbering, N. M. M.; van Orden, S. L.; Laukien, F. H. Int. J. Mass Spectrom. Ion Processes 1997, 165/166, 209–219.

Figure 1. Pulse sequence for CID experiment in the ICR: (a) ion accumulation in the source hexapole, (b) ion accumulation in the collision cell with Q preselection, (c) transfer ions to the ICR cell, (d) CHEF isolation, (e) single shot isolation, (f) pulse gas, (g) on resonance patent ion excitation for fragmentation, (h) pumping delay, (j) ion excitation for detection, and (k) ion detection.

Figure 2. (a) Full mass spectrum of SRFA in negative ion mode and (b) a 30 Da section of the mass spectrum.

tion of the target ion for fragmentation, all remaining mass peaks were ejected out of the ICR cell by single shots. With this technique, ions are excited with their resonance frequency using a long and, therefore, sharp pulse of 10-20 ms. Very narrow excitation pulses are needed to avoid the excitation of other ions being close to the ions which are supposed to be ejected.30 Nitrogen was pulsed into the ICR cell to a final pressure of about 10-7 mbar in the analyzer cell. The isolated parent masses were excited with their resonance frequency for 400 µs using a peakto-peak voltage of 3-5 V. A pump delay of 3 s was applied to regain a lower pressure (∼10-10 mbar) prior to mass detection. The ions were then excited with a frequency sweep and (30) McIver, R. T.; Hunter, R. L.; Baykut, G. Int. J. Mass Spectrom. Ion Processes 1989, 89, 343–358.

Analytical Chemistry, Vol. 81, No. 7, April 1, 2009

2689

Figure 3. Parent ion isolation with quadrupole preselection, CHEF isolation followed by single shot ejection in the ICR cell: (a) section of the full mass spectrum; isolated parent masses at (b) m/z 365.051, (c) m/z 365.088, (d) m/z 365.124, and (e) m/z 365.160.

Figure 4. In-cell CID fragmentation spectra of parent ions at (a) m/z 365.051, (b) m/z 365.088, (c) m/z 365.124, and (d) m/z 365.160.

detected for about 0.55 s in the mass range m/z 153-2000. An overview on the pulse sequence for the CID experiment is displayed in Figure 1. The size of the acquired data set was 2 MW for the mass spectrum and 1 MW for MS/MS spectra. The mass accuracy of the parent and fragment ions was generally better than 1 ppm. The data were zero filled once before sine apodization. Between 100 and 200 scans were added for one mass spectrum. Mass spectra were externally calibrated with deprotonated arginine clusters [(Arg)n - H]- (n ) 1-5) in the mass range m/z 173-869. The mass spectrum of SRFA was internally recali2690

Analytical Chemistry, Vol. 81, No. 7, April 1, 2009

Figure 5. In-cell CID fragmentation spectrum of the parent ion at m/z 365.088 (C17H17O9-) with annotation of fragment losses.

brated with a known homologous series C10H10O8(CH2)n, C10H6O9(CH2)n, and C12H6O10(CH2)n (n ) 0-8). The molecular formula calculation was performed with DataAnalysis 4.0 (Bruker Daltonik GmbH, Bremen, Germany). RESULTS AND DISCUSSION The mass spectrum of Suwannee river fulvic acids (SRFA) in electrospray negative ion mode shows several thousand singly charged peaks between m/z 200 and m/z 800 with a maximum of the peak distribution at around m/z 350 (Figure 2). For each of the four nominal masses m/z 363, 365, 367, and 381, four individual mass peaks were analyzed by on-resonance CID using

Figure 6. In-cell CID fragmentation spectra of compounds with similar elemental formula to those measured in the SRFA sample: (a) myricetin (C16H18O9), (b) chlorogenic acid (C15H10O8), (c) 1,4,5,8-naphthalenetetracarboxylic acid (C16H18O10), and (d) 2-(4-2,2-dicarboxy-ethyl)-2,5-dimethoxybenzyl)-malonic acid (C14H18O8).

Figure 7. Relative intensities of fragments of studied precursor masses at (a) m/z 363, (b) m/z 365, (c) m/z 367, and (d) m/z 381 normalized to the fragment [M- H - CO2]-. Analytical Chemistry, Vol. 81, No. 7, April 1, 2009

2691

the pulse sequence given in Figure 1. Masses in the range between m/z 360-390 have been selected for fragmentation analysis due to their relatively high peak intensities. Isolation of precursor masses from the selected group at m/z 365 is shown in Figure 3. The full pattern of SRFA at m/z 365 is depicted in Figure 3a with more than 10 mass peaks at one nominal mass. Four major peaks with mass differences of only 36 mDa were isolated. These mass differences result from the exact mass difference between CH4 and O (Figure 3b-e). The parent masses were isolated maintaining >70% of the precursor ion intensity. The fragment spectra of the precursor masses with the identical nominal masses of m/z 365 are shown in Figure 4. The precursor molecules show similar neutral losses resulting in the same fragment mass pattern. However, because of the exact mass measurements of the fragment ions, the peaks differ from one spectrum to the next by the same mass difference as the precursor masses (36 mDa). The calculated neutral losses are shown in Figure 5 for the parent mass at m/z 365.088 which corresponds to the molecular formula C17H17O9-. The neutral loss of CO2 was observed as most abundant. Therefore, most likely several carboxyl groups are present in the molecular structure of this precursor ion. Nevertheless, also neutral losses of H2O, CO, and CH4 are observed. The observation of decarboxylation and parallel loss of water of fulvic acid compounds is in agreement with fragmentation studies by low-resolution mass spectrometry of all isobaric precursor ions at one nominal mass.2,31-33 On the basis of the fragmentation pattern, possible structures can be determined. To decide which kind of chemical structures are most likely present in SRFA, commercially available model compounds were studied. The fragmentation spectra of four different oxygen rich compounds with an elemental composition similar to the studied precursor mass peaks of SRFA (CxHyOz) are shown in Figure 6. The fragment spectrum of the model compound 4 shown in Figure 6d is most similar to the studied precursors in SRFA. The other spectra show different fragmentation patterns and abundances. Higher abundances occurr due to a predominant bond cleavage of weak ester bonds (Figure 6a). Fewer and less abundant fragments are detected where the ions are stabilized by aromatic cores or where several bond cleavages are necessary to form the fragments (parts b and c of Figure 6). Therefore, the aromatic structure and the functional groups of model compound 4 containing carboxy and methoxy groups is most likely similar to the structure of the studied fulvic acid compounds in SRFA. The spectrum is dominated by the loss of CO2 and, to a smaller extend, the loss of water. Nevertheless, it should be pointed out that structural isomers exist in fulvic acids having exactly the same mass and molecular formula. Therefore, the fragments can stem from different chemical structures and isomers. A comparison of relative intensities of the observed fragments of all studied masses is shown in Figure 7. The intensities are normalized on fragment [M -H CO2]- which is the most intense fragment peak in all the MS/ MS spectra. The decrease of oxygen in the molecular formula (31) These, A.; Winkler, M.; Thomas, C.; Reemtsma, T. Rapid Commun. Mass Spectrom. 2004, 18, 1777–1786. (32) Reemtsma, T.; These, A.; Springer, A.; Linscheid, M. Environ. Sci. Technol. 2006, 40, 5839–5845. (33) Reemtsma, T.; These, A.; Venkatachari, P.; Xia, X.; Hopke, P. K.; Springer, A.; Linscheid, M. Anal. Chem. 2006, 78, 8299–8304.

2692

Analytical Chemistry, Vol. 81, No. 7, April 1, 2009

Figure 8. Possible fragmentation pathway by collision induced dissociation of model structure of compound C17H17O9- (m/z 365.088).

results in the reduction of stepwise neutral loss of CO2 which confirms that oxygen in SRFA is predominantly bound in carboxyl groups. A possible parent ion structure which is in agreement with the observed fragments including the fragmentation pathways and the fragment structures of the precursor ion C17H17O9- is shown in Figure 8. The neutral loss of CO2 is mainly observed for molecular structures with carboxyl groups. Loss of water typically indicates the presence of hydroxyl groups. However, the fragmentation of the model structure shown in Figure 8 is only an example. Several other molecular structures are possible which would correspond to the observed fragmentation spectrum. Nevertheless, the observed successive loss of CO2 verifies that carboxyl groups are very common in fulvic acid molecules. These results correlate well with the MS/MS results by Reemtsma et al.17 using a triple quadrupole mass spectrometer and Stenson et al.18 using FTICR infrared multiphoton dissociation with a 1 Da isolation window by stored waveform inverse Fourier transform (SWIFT) isolation. This study showed for the first time that CID MS/MS spectra can be acquired from isolated single mass peaks of Suwannee river fulvic acid. Because of the similar fragmentation pattern of the studied peaks at m/z 365 with the molecular formula C16H13O10-, C17H17O9-, C18H21O8-, and C19H25O7-, similar core structures of these molecules are very likely functionalized with mainly carboxyl and hydroxyl groups.

Figure 9. Relative intensities of fragments of studied homologous precursor masses with (a) 8, (b) 9, (c) 10, and (d) 11 oxygen atoms normalized to the fragment [M - H - CO2]-.

Since one molecular formula can represent a multitude of different structural isomers, the fragmentation pattern can also be a result of a mixture of several structural isomers. However, comparison of the similar fragmentation patterns of all four analyzed ions points to analogue structural features. These features can either be created by one dominating structure or by comparable mixtures of compounds on each parent molecular formula. These findings contradict the expectation that masses which are relatively similar in their mass (e.g., 36 mDa for CH4 vs O exchange) are chemically less similar than those molecules having a larger mass difference (e.g., 14 Da for CH2 homologues).26 However, different core structures are possible due to various double bond equivalents (DBE). For the studied molecular formulas (C16H13O10- to C19H25O7-) DBE decreased from DBE ) 10 to DBE ) 7. The results also indicate that not all oxygen atoms can be lost by decarboxylation, a part is lost as water or bound in the core structure. The loss of water after two or three neutral losses of carbon dioxide is mentionable. Therefore, at least one oxygen atom is probably bound as a hydroxyl group. The comparison of compounds in the same homologous group with a difference of just a methylene group is of major interest. These compounds should fragment nearly identical, because they likely have the same core structure and the same functional group only with a difference of one methylene group in one side chain. The relative fragment

intensities of parent ions with 8, 9, 10, and 11 oxygen atoms and a difference of one methyl group is shown in Figure 9. The relative fragment intensities of compounds with a difference of one methylene group are indeed very similar indicating that these molecules are most likely compounds of the same homologous series. CONCLUSION Fragmentation pathways of single isolated mass peaks of SRFA with exactly one molecular formula were studied for the first time by collision induced dissociation FTICR mass spectrometry. The structure of compounds in complex mixtures like natural organic matter, which cannot be separated by chromatographic methods or electrophoresis, were analyzed by FTICR mass spectrometry. Mainly stepwise losses of CO2 and H2O were observed of all precursor masses even with different numbers of double bond equivalents. Therefore, it can be concluded that similar structures with the presence of carboxyl and hydroxyl groups are present in this fulvic acid standard. In recent studies of natural organic matter, it has been shown that the diversity of molecular formulas in NOM covers a large proportion of the chemical feasible formulas. Our results imply that the structural diversity is lower. Peaks with very small mass differences in the millidalton range can be isolated and fragmented by collision induced fragmentation in the analyzer cell of a FTICR mass spectromAnalytical Chemistry, Vol. 81, No. 7, April 1, 2009

2693

eter necessary for structure analysis. Therefore, FTICR in-cell MS/MS is a straightforward method for studies of molecular structures of natural organic matter from various sources or other complex mixtures like crude oils or aerosols. Coupling the presented method to a preceding polarity-based chromatography will help to separate molecules with identical mass formulas and different molecular structures and by that supply additional analytical information on the chemical structure of the target molecules.15

2694

Analytical Chemistry, Vol. 81, No. 7, April 1, 2009

ACKNOWLEDGMENT The authors would like to thank Dr. Philippe Schmitt-Kopplin from the Helmholtz Zentrum Munich for providing the Suwannee river fulvic acid sample.

Received for review December 12, 2008. Accepted February 13, 2009. AC802624S