Changes in post-source decay fragmentation behavior ... - Google Sites

O. Laine et al., Eur. J. Mass Spectrom. 9, 195–201 (2003)

195

Changes in PSD Fragmentation Behavior of Poly(Methyl Methacrylate) using MALDI-ToF MS O. Laine et al., Eur. J. Mass Spectrom. 9, 195–201 (2003)

Changes in post-source decay fragmentation behavior of poly(methyl methacrylate) polymers with increasing molecular weight studied by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry Olli Laine*,a University of Joensuu, Department of Chemistry, PO Box 111, 80101 Joensuu, Finland

Sarah Trimpin, Hans Joachim Räder and Klaus Müllen Max-Planck-Institut für Polymerforschung, Ackermannweg 10, 55128 Mainz, Germany

In order to investigate the systematic changes in fragmentation behavior of poly(methyl methacrylate) (PMMA) with increasing molecular weight, alkali–metal cationized PMMA 20-mer, 60-mer and 100-mer were selected for post-source decay (PSD) fragmentation study by matrix-assisted laser desorption/ionization time-of-flight (MALDI-ToF) mass spectrometry. PMMA polymers were cationized with lithium, potassium and cesium cations to explore the influence of the cation size on the fragmentation behavior of the polymers. All PMMA polymers could be fragmented by MALDI-PSD and fragmentation of the MALDI ionized synthetic polymer of molecular weight 10 kDa is reported here for the first time. It was shown that an increasing molecular weight of the PMMA chain required an increase in the size of the cation to improve the intensity and the number of the fragments in the PSD spectrum. Some instrumental parameters had to be optimized prior to a successful PSD analysis of the largest PMMA polymers. Keywords: matrix-assisted laser desorption/ionization, time-of-flight mass spectrometry, poly(methyl methacrylate), polymers, post-source decay

Introduction

5

Matrix-assisted laser desorption/ionization (MALDI) mass spectrometry has become an important tool in the structural characterization of synthetic polymers. Nowadays, it is commonly used to provide information on molecular weight distributions as well as on the end-groups of polymers.1–4 End-group analysis requires that the peaks of differently terminated polymer chains be separated in a MALDI spectrum. In addition, it is necessary that the compounds used for the polymer synthesis are known or the details of the polymer composition have been defined by some other method. Determination of the end-groups of unknown polymers may be assisted by chemical treatment of a

Current address: National Public Health Institute, Department of Environmental Health, PO Box 95, 70701 Kuopio, Finland. E-mail: [email protected].

DOI: 10.1255/ejms.541

the sample. However, more detailed information is obtained by means of mass spectrometric fragmentation techniques 6–14 such as collision-induced dissociation (CID) and post6,8,15–19 source decay (PSD) which also give structural information on the entire polymer chain. For instance, MALDI mass spectrometry combined with CID or PSD fragmentation methods has been used to give structural information for poly(alkyl methacrylate)s,6–8,12 poly14,18 15 6,19 amides, polycarbonates, polyesters, poly(ethylene gly6,10,11 col), poly(ethylene oxide)-b-poly(p-phenylene 17 8,9,13,16 ethynylene) and polystyrene. Time-of-flight 6,8,9,14–19 6–13 (ToF) and hybrid magnetic sector-ToF mass specrometers have been used to perform CID and PSD experiments. The heaviest synthetic polymer that has been fragmented by using MALDI-CID on a hybrid magentic-sector 9 ToF instrument is polystyrene 5 kDa, whereas polystyrene 6.5 kDa was fragmented when the MALDI-PSD studies were 16 performed on a ToF mass spectrometer. The high-mass limit of the mass range that can be measured with a magnetic-sector

ISSN 1356-1049

© IM Publications 2003

196

Changes in PSD Fragmentation Behavior of Poly(Methyl Methacrylate) using MALDI-ToF MS

instrument is determined by the magnetic field but for a simple ToF instrument there is not such an instrumental limit.

MALDI-ToF mass spectrometry

Experimental section

Mass spectrometric studies were performed with a Bruker Reflex II time-of-flight mass spectrometer (Bruker–Franzen Analytik, Bremen, Germany) equipped with a nitrogen laser (λ = 337 nm) and a microchannel-plate detector. Measurements were carried out in reflectron mode using delayed extraction. The length of the flight tube was 145 cm and the delay time 50 ns. Acceleration voltage was 28.5 kV, extraction voltage 20 kV and lens voltage 10 kV. Detector voltage was increased with the increasing size of PMMA studied; for example, the voltage was –1.5 kV for the 20-mer, –1.6 kV for the 60-mer and –1.7 kV for the 100mer. The precursor ions were selected by the ion gate situated in the linear tube after the ion source. The width of the massselection window was m/z 50 for the PMMA 20-mer, m/z 130 for the 60-mer and m/z 230 for the 100-mer. To collect PSD data, the reflectron voltage was decreased in 15 steps from 30 to 0.95 kV. The segments acquired for each reflectron voltage were pasted together to produce a complete MALDI-PSD spectrum. Pasting and calibration of the segments was carried out using the FAST software from Bruker. The spectrum of each segment was obtained by combining data from 200 laser shots. The segments were calibrated with ACTH-clip 18–39 according to the Bruker Reflex II User manual. The intensities of the fragment ion peaks in a segment were decreased towards the high-mass end of the segment especially in the PSD spectra of PMMA 60-mer and 100-mer.

Materials

MALDI sample preparation

Poly(methyl methacrylate) standards PMMA 2500, PMMA 7100 and PMMA 13100 were obtained from Polymer Standards Service (Mainz, Germany). The MALDI matrix, 1,8,9-trihydroxyanthracene (dithranol), and trifluoroacetate salts LiTFA, KTFA and CsTFA were purchased from Aldrich (Steinheim, Germany). HPLC-grade tetrahydrofuran (THF) used for GPC-fractionation was obtained from Fluka (Deisenhofen, Germany).

PMMA standards, trifluoroacetate salts and dithranol matrix were all separately dissolved in THF. The concentra–1 tions of the PMMA solutions were 3–4 mg mL , those of the –1 –1 salts 5–6 mg mL and that of dithranol 20 mg mL . The PMMA sample fractionated by GPC was dried under vacuum and then dissolved in THF. Separate solutions of PMMA, salt and matrix were mixed at a volume ratio of 1 : 1 : 4 just before the measurements and approximately 0.5 µL of the mixture was applied to the target plate and allowed to air-dry.

(1)

Our goal was to study changes in the PSD fragmentation behavior of poly(methyl methacrylate) (PMMA) polymers (1) with increasing molecular weight by means of MALDIToF mass spectrometry. We also aimed to determine the upper molecular weight limit for this technique and expand it up to 10 kDa. The PMMA polymers selected for the fragmentation study had molecular weights of about 2 kDa, 6 kDa and 10 kDa. The precursor ions were generated from PMMA standards with average molecular weights of 2500, 7100 and 13100. Salts of three different alkali metals (Li, K and Cs) were added to the PMMA samples in order to explore the influence of the adduct cation size on the fragmentation behavior of the PMMA polymers. In addition, fractionation by gel-permeation chromatography (GPC) was used to increase the concentration of a PMMA chain of 10 kDa in the MALDI sample. Instrumental parameters such as detector voltage and the width of the mass-selection window of the ion gate were varied to improve the performance of the PSD measurements.

GPC-fractionation

The poly(methyl methacrylate) standard PMMA 13100 was fractionated by GPC in order to increase the concentration of the PMMA 100-mer. Fractionation was carried out with a Gilson GPC system (Abimed Analysentechnik, Langenfeld, Germany) equipped with three Jordi-Gel DVB columns of respective pore sizes 100, 1000 and 10000 Å and a Gilson refractive index detector. THF was used as an eluent and the PMMA standard was dissolved in THF at a concentration of 30 mg mL–1. Injection volume of the sam–1 ple was 1 mL and the flow rate was 1 mL min . Six PMMA fractions of about 1 mL were collected and normal MALDI spectra were run from every fragment. The fraction where the amount of PMMA 100-mer was greatest was selected for the PSD study.

Results and discussion PSD fragmentation mechanisms of PMMA

The fragment ions observed here were similar to those reported in earlier studies on poly(alkyl methacrylate) poly6–8 mers where both CID and PSD techniques were employed. The structures of the fragment ions and their m/z values are presented in Scheme 1. The radical ions A and B are the con8 sequence of direct cleavages of the polymer backbone whereas the ions C, D, E and F are supposed to be the result 7 of rearrangement processes.

O. Laine et al., Eur. J. Mass Spectrom. 9, 195–201 (2003)

197

Scheme 1.

Figure 1. PSD spectra of the (a) [M + Li]+, (b) [M + K]+ and (c) [M + Cs]+ molecular ions of the PMMA 20-mer.

PSD fragmentation of PMMA 20-mer

and [M + K] molecular ions (Figure 2). Similar fragment + ions of type A were also found in the [M + Cs] spectrum but with very low intensity. High abundance of the low-mass fragments in the PSD and CID spectra of PMMA has already been observed in several studies before.6–8 The m/z values of the most abundant fragment ions were found to be about a tenth of the m/z value of the precursor ion, as well. The series of fragment ions which originate from rearrangement reactions were seen in the [M + Li]+ and [M + K]+ spectra above m/z 500. Figure 3 presents those high-mass fragments from + the [M + Li] precursor ion. The fragment ions were type C, D, E and F ions as in previous fragmentation studies on 7,8 poly(alkyl methacrylate)s. All previous findings concerning the PMMA 20-mer were obtained repeatedly from different PSD runs.

+

+

+

+

PSD spectra of the [M + Li] , [M + K] and [M + Cs] molecular ions of the PMMA 20-mer are presented in Figure 1. Series of fragment ions were observed only when + + [M + Li] and [M + K] ions of the 20-mer were selected as precursor ions whereas an unexpected isolated peak at m/z + 1913 dominated the PSD spectrum of the [M + Cs] molecular ion. The lithium cationized polymer produced more abundant fragment ions than did potassium cationized chains. Usually, the intensities of the fragment ion peaks have to be considered with caution but in this case the varying intensities were obvious. These results agree with experimental and theoretical studies on PMMA fragmentation where lithium was found to possess the highest binding energy of alkali–metal ions for the low-mass PMMA fragments.8,20 Type A and B radical cations which consist of 1–2 PMMA monomer units were the most abundant fragment + ions at the low-mass end of the PSD spectra of the [M + Li]

PSD fragmentation of PMMA 60-mer +

The low-mass PSD fragments of the [M + Li] , + + [M + K] and [M + Cs] molecular ions of the PMMA 60mer are presented in Figure 4. PSD fragment ions were

198

Changes in PSD Fragmentation Behavior of Poly(Methyl Methacrylate) using MALDI-ToF MS

width of the window used for the selection of the PMMA 20mer was not large enough to pass an adequate amount of the PMMA 60-mer through the ion gate. The isolation of the 60mer succeeded well but the fragments from the precursor ion could not be observed in the PSD spectrum. When the width of the window was enlarged from m/z 50 to more than m/z 100, the intensity of the precursor ion signal was high enough that the fragment ions began to appear. The width of the mass-selection window was then so large that the cationized neighbor oligomers with 1–2 PMMA monomer units less than in the chosen precursor ion were also seen in the PSD spectrum. However, fragments originating from these neighbor oligomers were not expected to appear in the spectrum because the intensities of those fragment ions were much lower than that of their precursor ion. The envelopes of the low-mass fragment ions were mostly similar with lithium and potassium since they ranged from m/z 300 to m/z 1100 and the most abundant fragment ions were found at about m/z 600. Based on repeated PSD runs, it was not possible to conclude whether lithium or

Figure 2. Low-mass fragment ions from the (a) [M + Li]+, (b) [M + K]+ and (c) [M + Cs]+ molecular ions of the PMMA 20mer.

observed with all the alkali–metal cations used for the cationization of the PMMA 60-mer. The key factor in obtaining a PSD spectrum from the PMMA 60-mer was to have a precursor-ion signal intensity as high as possible. The

Figure 3. High-mass fragment ions from the [M + Li] molecular ion of the PMMA 20-mer. +

Figure 4. Low-mass fragment ions from the (a) [M + Li]+, (b) [M + K]+ and (c) [M + Cs]+ molecular ions of the PMMA 60mer.

O. Laine et al., Eur. J. Mass Spectrom. 9, 195–201 (2003)

199

potassium produced more abundant fragment ions. The most abundant low-mass fragments were type A and B ions accompanied by a few less-abundant C ions. In the PSD spectrum of the [M + Cs]+ molecular ion, two pairs of intense fragment ions were type A and B ions. The low-mass fragments dominate the PSD spectra, even though they possess lower binding energy towards alkali–metal cations than do the high-mass fragments; this has previously been explained on the basis of charge-induced fragmentation of the PMMA chain.17 Four clear series of fragment ions were seen in the + + [M + Li] and [M + K] spectra above m/z 2500. Two series of the high-mass fragments were again type C and D ions but a definite identification of other fragment ion peaks was not possible because the resolution was insufficient to distinguish ions of type A and B from type E and F. However, based on earlier studies, these two series of fragment ions 7,8 were more likely type E and F than type A and B. Highmass fragment ion series also seemed to be present in the + [M + Cs] spectrum but the peaks could not be identified at all. PSD fragmentation of PMMA 100-mer

Fragmentation of the synthetic polymer of 10 kDa ionized by MALDI is reported here for the first time. PSD fragmentation of the PMMA 100-mer was much more difficult to accomplish than the fragmentation of smaller PMMA polymers. A much wider mass-selection window of the ion gate was needed for the isolation of the precursor ion than for the PMMA 60-mer; in this case, it was enlarged to m/z 230. PSD fragments were observed after enlargement of the mass-selection window with all alkali metals used for cationization of the PMMA 100-mer, but their signal intensity was still relatively weak. The results could be improved significantly when the poly(methyl methacrylate) standard PMMA 13100 was first fractionated by GPC and PSD spectra were then measured from the fractionated sample. Both these procedures increased the yield of the isolated molecular ion that was passed through the ion gate and this was found to be a prerequisite for obtaining PSD spectra from the PMMA 100-mer. The low-mass PSD fragments of the [M + Li]+, + + [M + K] and [M + Cs] molecular ions of the PMMA 100mer are presented in Figure 5. Envelopes of low-mass fragments were observed with all alkali metals used for cationization of the PMMA 100-mer. The widths of the envelopes were again approximately the same with all alkali metals; in this case, they ranged from m/z 400 to m/z 2000. However, the intensity of the fragment ion peaks increased with the size of the alkali–metal adduct so the highest abundance of fragment ions was observed in the [M + Cs]+ spectrum. The m/z values of the most abundant fragment ions were around m/z 1000 which means that they were once more about one-tenth of the m/z of the precursor ion. As in the case of the PMMA 60-mer, the low-mass fragment ions were mostly type A and B ions. Series of peaks that could

Figure 5. Low-mass fragment ions from the (a) [M + Li]+, (b) [M + K]+ and (c) [M + Cs]+ molecular ions of the PMMA 100mer.

have represented fragment ions were found at m/z values above 5000 but the identification of those peaks was not possible because of insufficient resolution. Similar findings concerning the PMMA 100-mer were observed in all repeated PSD runs. A higher abundance of cesium cationized fragments than of the lithium and potassium cationized fragments in the PSD spectra of the PMMA 100-mer cannot be explained by the binding energies of alkali–metal cations for the PMMA fragments. Molecular modeling calculations and CID experiments on PMMA20 and poly(ethylene tere21 phthalate) oligomers, which are about same size as the most abundant fragments from the 100-mer, have clearly proved that the attachment of those oligomers to a lithium cation is much more energetically favorable than attachment to a cesium cation. However, several studies have shown that, when the length of the PMMA chain increases, the interaction of the chain with the larger cations also strengthens.22–25 It is important that the alkali–metal adduct ion of the PMMA is stable enough because ejection of the alkali–metal cation

200

Changes in PSD Fragmentation Behavior of Poly(Methyl Methacrylate) using MALDI-ToF MS

from the precursor ion is a competitive reaction for dissocia20 tion of the PMMA chain. Clear evidence of the stability of the precursor ions could have been obtained if the peaks of bare alkali–metal ions had been seen in our PSD spectra. Unfortunately, because of our measuring procedure, the lower end of the m/z scale of the PSD spectra was limited so that lithium and potassium cations could not be observed. Nevertheless, the absolute intensities of the alkali–metal adduct ions increased with the cation size when the normal MALDI spectra were measured using PMMA 13100 which illustrates the higher stability of [M + Cs]+ ions compared with other adduct ions.

Acknowledgments Financial support from the Neste Research Foundation is gratefully acknowledged. Warm thanks are given to Laurence Przybilla for fruitful discussions and Ali Rouhanipour for technical assistance.

References 1. 2. 3.

Conclusions 4. PSD fragmentation of PMMA polymers with different chain lengths was studied by MALDI-ToF mass spectrometry. The alkali–metal cationized PMMA 20-mer, 60-mer and 100-mer were selected as precursor ions. Any influence of the cation size on the fragmentation behavior was investigated by using lithium, potassium and cesium salts for cationization of the PMMA samples. PSD fragmentation succeeded with all the PMMA polymers studied which is the first time that this has been reported for the PMMA 60-mer and 100-mer. Our studies showed that an increasing chain length of the PMMA polymer required an increasing size of the adduct cation to enhance the abundance of the fragment ions in the PSD spectrum. The best PSD results were obtained when the PMMA 20-mer was cationized with lithium and the 100-mer with cesium whereas it was not possible to decide between lithium or potassium as the best cation for the PMMA 60mer. The advantage of different cations was postulated to be a consequence of the strength of interaction of the cations with the isolated PMMA molecules and the PSD fragments. The comparison of the fragment ion intensities obtained from precursor ions at different molecular weights revealed that the masses of the most intense fragments shift to higher values with an increasing size of the precursor ion and have a mass which is about 10% of the mass of the precursor ion. In addition, some instrumental parameters were found to be important in the PSD study of high-mass PMMA polymers. Detector voltage, as well as the width of the massselection window of the ion gate, have to be increased with an increasing size of the PMMA polymer studied. In addition, GPC-fractionation of the PMMA sample improved the PSD results obtained from the PMMA 100-mer. A key factor, in general, was to increase the yield of the precursor ion that passed through the ion gate and also to improve the detection sensitivity for the fragment ions which had much lower intensities than the precursor ion in the PSD spectra.

5.

6.

7.

8.

9.

10.

11.

12. 13.

14. 15. 16. 17.

P.O. Danis, D.E. Karr, F. Mayer, A. Holle and C.H. Watson, Org. Mass Spectrom. 27, 843 (1992). U. Bahr, A. Deppe, M. Karas, F. Hillenkamp and U. Giessmann, Anal. Chem. 64, 2866 (1992). G. Montaudo, M.S. Montaudo, C. Puglisi and F. Samperi, Rapid Commun. Mass Spectrom. 9, 453 (1995). A.K. Chaudhary, G. Critchley, A. Diaf, E.J. Beckman and A.J. Russel, Macromolecules 29, 2213 (1996). O. Laine, H. Österholm, H. Järvinen, K. Wickström and P. Vainiotalo, Rapid Commun. Mass Spectrom. 14, 482 (2000). A.T. Jackson, H.T. Yates, J.H. Scrivens, G. Critchley, J. Brown, M.R. Green and R.H. Bateman, Rapid Commun. Mass Spectrom. 10, 1668 (1996). A.T. Jackson, H.T. Yates, J.H. Scrivens, M.R. Green and R.H. Bateman, J. Am. Soc. Mass Spectrom. 8, 1206 (1997). J.H. Scrivens, A.T. Jackson, H.T. Yates, M.R. Green, G. Critchley, J. Brown, R.H. Bateman, M.T. Bowers and J. Gidden, Int. J. Mass Spectrom. Ion Processes 165/166, 363 (1997). A.T. Jackson, H.T. Yates, J.H. Scrivens, M.R. Green and R.H. Bateman, J. Am. Soc. Mass Spectrom. 9, 269 (1998). A.R. Bottrill, A.E. Giannakopulos, C. Waterson, D.M. Haddleton, K.S. Lee and P.J. Derrick, Anal. Chem. 71, 3637 (1999). A.R. Bottrill, A.E. Giannakopulos, A. Millichope, K.S. Lee and P.J. Derrick, Eur. J. Mass Spectrom. 6, 225 (2000). C.D. Borman, A.T. Jackson, A. Bunn, A.L. Cutter and D.J. Irvine, Polymer 41, 6015 (2000). A.T. Jackson, A. Bunn, L.R. Hutchings, F.T. Kiff, R.W. Richards, J. Williams, M.R. Green and R.H. Bateman, Polymer 41, 7437 (2000). R. Murgasova and D.M. Hercules, J. Mass Spectrom. 36, 1098 (2001). L. Przybilla, H.J. Räder and K. Müllen, Eur. Mass Spectrom. 5, 133 (1999). R.J. Goldschmidt, S.J. Wetzel, W.R. Blair and C.M. Guttman, J. Am. Soc. Mass Spectrom. 11, 1095 (2000). L. Przybilla, V. Francke, H.J. Räder and K. Müllen, Macromolecules 34, 4401 (2001).

O. Laine et al., Eur. J. Mass Spectrom. 9, 195–201 (2003)

18. I. Fournier, A. Marie, D. Lesage, G. Bolbach, F. Fournier and J.C. Tabet, Rapid Commun. Mass Spectrom. 16, 696 (2002). 19. O. Laine, T. Laitinen and P. Vainiotalo, Anal. Chem. 74, 4250 (2002). 20. J. Gidden, A.T. Jackson, J.H. Scrivens and M.T. Bowers, Int. J. Mass Spectrom. 188, 121 (1999). 21. J. Gidden, T. Wyttenbach, J.J. Batka, P. Weis, A.T. Jackson, J.H. Scrivens and M.T. Bowers, J. Am. Soc. Mass Spectrom. 10, 883 (1999). 22. J. Spickermann, K. Martin, H.J. Räder, K. Müllen, H. Schlaad, A.H.E. Müller and R.-P. Krüger, Eur. Mass Spectrom. 2, 161 (1996).

201

23. B. Thomson, K. Suddaby, A. Rudin and G. Lajoie, Eur. Polym. J. 32, 239 (1996). 24. D. Dogruel, R.W. Nelson and P. Williams, Rapid Commun. Mass Spectrom. 10, 801 (1996). 25. A.T. Jackson, H.T. Yates, W.A. MacDonald, J.H. Scrivens, G. Critchley, J. Brown, M.J. Deery, K.R. Jennings and C. Brookes, J. Am. Soc. Mass Spectrom. 8, 132 (1997). Received: 27 March 2003 Revised: 16 May 2003 Accepted: 16 May 2003 Web Publication: 10 July 2003