Analysis of Multiply Charged Poly(ethylene oxide-co-propylene oxide) Using Electrospray Ionization Ion Mobility Spectrometry Mass Spectrometry

Transcription

1 ANALYTICAL SCIENCES FEBRUARY 209, VOL The Japan Society for Analytical Chemistry Analysis of Multiply Charged Poly(ethylene oxide-co-propylene oxide) Using Electrospray Ionization Ion Mobility Spectrometry Mass Spectrometry Kanako ITO, Shinya KITAGAWA, and Hajime OHTANI Department of Life Science and Applied Chemistry, Graduate School of Engineering, Nagoya Institute of Technology, Gokiso, Showa, Nagoya , Japan Poly(ethylene oxide), poly(propylene oxide), and their copolymer (poly(eo-co-po)) were analyzed by electrospray ionization ion mobility spectrometry mass spectrometry (ESI-IMS-MS). ESI produced multiply charged analytes of 2 to 5 Na + additions, and they were separately observed in a 2D map of m/z value vs. drift time. The collision cross-section of the analyte polymers was almost linearly proportional to (molecular weight) 0.644, except for the analytes with 2Na + addition; a nonlinear relation called folding was significantly observed for the analytes with 2Na + addition. An increase in electrostatic repulsion, because of the increase in Na + addition, suppressed the folding of the polymer. Analyses of poly(eo-co-po) with different EO compositions revealed that the copolymer with high EO composition tended to show folding. The separation of highly multiply charged poly(eo-co-po)s with different EO contents by ESI-IMS-MS was successfully demonstrated. Keywords Multiple charges, copolymer, electrospray ionization, ion mobility, mass spectrometry, folding, polymer composition (Received July 9, 208; Accepted September 9, 208; Advance Publication Released Online by J-STAGE September 28, 208) Introduction Ion mobility spectrometry (IMS) is a method for separating ions having the same mass-to-charge (m/z) ratio but different collision cross-sections (CCS) and/or charge numbers based on the difference in ion mobility (or drift time, t D) in a thin gaseous atmosphere, analogous to gel-electrophoresis in the liquid phase. 6 IMS is often combined with MS for determining both t D and m/z values in the gas phase, and this technique is known as ion mobility spectrometry mass spectrometry (IMS-MS). IMS-MS is frequently used for structural characterization and analysis of the conformational dynamics of various compounds. Fundamental studies and applications of IMS-MS are frequently reported for various compounds from small compounds to proteins. 2 5,7 9 In the case of synthetic polymer, in 995, von Helden et al. reported a study of IMS behavior of poly(ethylene oxide) (PEO, polymerization degree, n = 5 to 9) of single Na + addition generated by matrix-assisted laser desorption ionization (MALDI). 9 Further applications of IMS-MS to synthetic polymers have also been reported and most previous studies on synthetic polymers using IMS-MS have focused on homopolymers and their mixtures. 0 9 Meanwhile, copolymers are polymerized from two or more monomer species to control its characteristics, and used generally in various engineering products. In the case of copolymer, molecules having the same or approximately the same molecular weight but different compositions and/or To whom correspondence should be addressed. kitagawa.shinya@nitech.ac.jp arrangements are generally synthesized. Therefore, copolymer analysis in MS is often difficult to interpret due to the complexity of the mass spectrum. To solve this problem, we focused on a potential of IMS-MS for analyses of copolymers. If the CCS of the monomers (CSS per unit mass, strictly) used for synthesis of the copolymer is not the same, the CCS of the copolymer depends on its composition even though their m/z values are the same. In addition, when the copolymer composition and arrangement affect the conformation (or CCS) in a gaseous phase, the difference in the composition/arrangement can be identified in IMS. Furthermore, if the affinity to an attaching ion depends on the monomer unit species, the copolymer composition affects the charge number, which can be identified by IMS. If the charge density attached on a polymer affects the intramolecular interaction electrostatically, the CCS of the polymer is varied. Therefore, it is concluded that IMS-MS has the potential for composition analysis of copolymers. However, the study of IMS-MS for copolymers is insufficient as described above. In this research, PEO, poly(propylene oxide) (PPO), and copolymers of poly(ethylene oxide-co-propylene oxide) (poly(eo-co-po)) were analyzed in IMS-MS for investigating the effect of the charge density on CCS. Two types of triblock copolymers (PEO-b-PPO-b-PEO and PPO-b-PEO-b-PPO) and a random copolymer (poly(eo-r-po)) were analyzed, and the effect of the composition, or EO unit content, on IMS behavior was investigated. In addition to the fundamental study, the composition-based separation of poly(eo-co-po)s in IMS-MS was demonstrated.

2 70 ANALYTICAL SCIENCES FEBRUARY 209, VOL. 35 Experimental Reagents PEO (average molecular weight (M w) = 2000, Shodex, Tokyo, Japan), PPO (M w = 2000, Wako, Osaka, Japan), poly(eo-r-po) (M w = 2500, EO unit 70 wt%, Sigma-Aldrich, St. Louis, MO), PEO-b-PPO-b-PEO (M w = 2000, EO unit 0 wt%, Sigma- Aldrich), and PPO-b-PEO-b-PPO (M w = 2000, EO unit 50 wt%, Sigma-Aldrich) were used as analytes. The polymers were dissolved (0. mg/ml each) in a mixture of acetonitrile/water (50/50, v/v) containing 0.0% sodium trifluoroacetate (Wako). Apparatus and procedure Waters Synapt G2 HDMS (Waters, Milford, MA) equipped with an ESI source was used for analyses in the positive mode (capillary voltage: 3.0 kv, sampling cone voltage: 40 V). This apparatus has a T-wave type IMS function (TWIMS). The sample solutions were introduced into the ESI source via direct infusion at a flow rate of 20 μl/min using a microsyringe pump ( Plus, Harvard Apparatus, Holliston, MA). The basic parameters for TWIMS were as follows: bias voltage, 50 V; wave height, 40 V; start velocity, 800 m/s; end velocity, 000 m/s; and recorded m/z range, For the analysis of the mixture of the poly(eo-co-po)s, the parameter set of 36 V, 40 V, 600 m/s, 500 m/s, and was employed. Results and Discussion Analysis of PEO and PPO mixture by ESI-IMS-MS Prior to the analysis of poly(eo-co-po), a mixture of PEO and PPO was analyzed using ESI-IMS-MS, and the result is shown in Fig.. Figure A shows a two-dimensional t D m/z distribution, where the signal intensity is presented as a heat map. As seen in this figure, several groups could be observed in the 2D distribution when a mixture of PEO and PPO was analyzed. The extracted mass spectrum of area # in Fig. A is shown in Fig. B. The mass spectrum indicates area # including the signals of PEO and PPO with various Na + additions, [PEO+3Na] 3+, [PEO+4Na] 4+, [PEO+5Na] 5+, [PPO+2Na] 2+, and [PPO+3Na] 3+. Here, the charge number (z value) was identified from the difference of m/z value between the isotopic peaks, and the polymerization degree and adduct ions species were identified by the exact mass value. The other areas (#2 #6) in Fig. A are also identified from the extracted mass spectra. It was clear that various kinds of ion adducts were formed by ESI and partially separated by IMS. As shown in Fig. A, both Na + and H + additions to the polymer were observed in zones #4, #5, and #6. However, for simplified discussion, we focused only on the analytes with Na + addition in this paper. Figure 2A shows the relationship between molecular weight of the polymer (m/z z m Na z, where m Na is a mass of Na cation) and drift time, without information about the signal intensity. As clearly shown in Fig. 2A, the drift time increased with an increase in the molecular weight, or polymer chain length. In addition, the increase in z reduced the t D, and a curved relationship between the molecular weight and t D was observed in each series. The CCS value, Ω, in TWIMS is given by the following equation: 20 Fig. Analysis of PEO and PPO mixture by ESI-IMS-MS. The 2D heat map of the m/z value and drift time is shown in (A), and the extracted mass spectrum of zone # is shown in (B). The term n in (B) is polymerization degree of polymers. Ω = (8π) 2 6 ze + 2 (8k bt) 2 m I m N 760 P T N A td B () where e, m I, m N, k b, T, P, N, A, and B are the elementary charge, mass of the analyte, mass of the collision gas, Boltzmann constant, temperature of the gas, buffer gas pressure, number density of the gas, correction factor for the electric field parameters, and compensating factor for the non-linear effect of the TWIMS device, respectively. Under certain experimental conditions, Eq. () is transformed into Eq. (2). Ω = A z + m I m N 2 B td (2) where A is a constant depending on the experimental conditions. A = (8π) 2 6 e 760 (8k bt) 2 P T N A (3) In our system, the B value was empirically estimated to be based on the analysis of polyaniline. Therefore, the value proportional to the CCS (relative CCS defined as Ω ) is given by the following equation.

3 ANALYTICAL SCIENCES FEBRUARY 209, VOL Table Average chain length per unit Na +, n/z, for PEO, PPO, and poly(eo-co-po) a Polymer 2Na + 3Na + 4Na + PEO 2 25 FD b (6 25) PPO 8 SFD (5 8) Poly(EO-r-PO) 2 23 (EO unit 70 wt%) FD (8 23) PPO-b-PEO-b-PPO 2 24 (EO unit 50 wt%) FD (6 24) PEO-b-PPO-b-PEO 2 22 (EO unit 0 wt%) SFD (6 24) 8 9 SFD b (5 9) FD (7 22) 9 9 SFD (6 9) b a. In the case of copolymers, the sum of the number of EO and PO units was used for calculating the n/z value. b. FD: folding, SFD: slight folding, : no folding. morphology when the molecular weight of the polymer was the same. On the other hand, the Ω value of the analytes with 2Na + addition was significantly small in the region of molecular weight >500 due to folding, whereas the Ω value for molecular weight <500 would be the same as that for 3 to 5 Na + additions. Figure 2C depicts the relationship between the polymerization degree and Ω. The Ω value for PPO is larger than that for PEO at the same n. The side methyl groups in PPO enhanced the molecular volume and/or suppressed the intramolecular interaction of the PPO main chain. Fig. 2 Effect of molecular weight on (A) drift time and (B) relative CCS, Ω, for PEO (red) and PPO (blue) with multiple Na + additions. The relationship between polymerization degree and relative CCS is also shown in (C). Ω z + m I m N td ( Ω ) (4) The relationships between Ω and the molecular weight value of PEO/PPO are shown in Fig. 2B. A nearly linear relationship between Ω and the molecular weight of the polymers is seen, i.e., the CCS is proportional to the molecular weight of PEO/PPO, except [PEO+2Na] 2+ and [PPO+2Na] 2+. For [PEO+2Na] 2+ and [PPO+2Na] 2+, nonlinear relationships are clearly observed (a slightly curved relationship is also seen for [PEO+3Na] 3+ in the higher molecular weight region). This nonlinear relationship between molecular weight and CCS is referred to as folding, which is often observed in the analysis of polymers by IMS.,2 23 The folding behavior is discussed in further detail in the next section. As shown in Fig. 2B, the Ω values for 3, 4, and 5 Na + additions were approximately the same for the same molecular weight for each polymer. That is, the charge number, or number of Na + additions, did not significantly affect the polymer Effect of charge number on CCS Winter et al. reported that folding was caused by the extension of the chain length during the analysis of polylactide by IMS. 22 As shown in Fig. 2, folding of both PEO and PPO was observed for their 2Na + adducts. However, there was no notable folding in the cases of 3 to 5 Na + additions. Since the morphology of multiply charged polymers in the gaseous state will be affected by electrostatic interactions, 22 the charge density of PEO and PPO was evaluated. The average chain length per unit Na + addition, n/z, was calculated as a measure of charge density. For example, 2Na + addition was observed for PEO with the polymerization degree ranging from 24 to 50, and the n/z value was calculated as 2 to 25. Table shows the n/z values of PEO and PPO for 2, 3, and 4 Na + additions. Moreover, folding is cited in terms of three magnitudes (folding, slight folding, and no folding). The n/z value for the folding region is also determined, as summarized in Table. In the case of [PEO+2Na] 2+, folding was observed for n = 32 50, as shown in Fig. 2C, and the n/z ranged from 6 to 25. As clearly shown in Table, the charge density on the polymer increased with an increase in the z value, i.e., a higher z value produced a lower n/z value. Interestingly, folding was observed in the region with a relatively large n/z, or in the low charge density region (n/z range: 5 to 25). This phenomenon suggested that a high charge density (n/z < 5) would result in the electrostatic repulsions, thus preventing folding. As shown in Fig. 2C and Table, the folding effect was more clearly observed in PEO rather than in PPO, suggesting that the methyl side group in PPO inhibits the intramolecular interaction to induce folding. Analysis of poly(eo-co-po) by ESI-IMS-MS In this section, we describe the analysis of different types of poly(eo-co-po) using ESI-IMS-MS. Figure 3A shows the

4 72 ANALYTICAL SCIENCES FEBRUARY 209, VOL. 35 Fig. 3 Analysis of PEO-b-PPO-b-PEO (EO unit 0 wt%) by ESI- IMS-MS. The 2D heat map of the m/z value and drift time is shown in (A), and the extracted mass spectrum of zone # is shown in (B). typical results for PEO-b-PPO-b-PEO (EO unit 0wt%) in a 2D heat map. The extracted mass spectrum for zone # in Fig. 3A is shown in Fig. 3B. Similar to Fig. A, the charge number was identified from the m/z difference between the isotope peaks and both the additive type and the polymer composition were identified from the exact mass value. In zone #, the copolymers corresponding to 2, 3, and 4 Na + additions were observed and marked by diamonds, circles, and triangles, respectively. The number of EO and PO units were described for each peak as [N EO, N PO], e.g. [5, 23] and [3, 25] for m/z = and , respectively. Figure 4A illustrates the relationship between Ω and the molecular weight of PEO-b-PPO-b-PEO with 2, 3, and 4 Na + additions. Slight folding was observed only for 2Na + addition, similar to Fig. 2A. The copolymers of PPO-b-PEO-b-PPO (EO unit 50 wt%) and poly(eo-r-po) (EO unit 70 wt%) also analyzed by ESI-IMS-MS, and their results for 2 to 4 Na + additions are shown in Figs. 4B and 4C, respectively. Similar to the case of the PEO and PPO homopolymers shown in Fig. 2, the increase in z, or Na + addition, inhibited the folding of the polymer. The average chain length per unit Na + addition, n/z, was calculated as a measure of the charge density, akin to both PEO and PPO. Since the analyte is a copolymer, the sum of the number of EO and PO units is used to calculate the n/z value. As shown in Table, the increase in z led to an increase in the charge density on the polymer, similar to the case of PEO and PPO homopolymers. The folding effect was not observed for the analytes with 4Na + addition due to electrostatic repulsion. In the case of 2Na + addition, the folding effect was observed for all copolymers. Interestingly, the folding effect was notable in Fig. 4 Effect of molecular weight on relative CCS, Ω, for (A) PEOb-PPO-b-PEO (EO unit 0 wt%), (B) PPO-b-PEO-b-PPO (EO unit 50 wt%), and (C) poly(peo-r-ppo) (EO unit 70 wt%) with different Na + additions. the order of poly(eo-r-po), PPO-b-PEO-b-PPO, and PEO-b- PPO-b-PEO. This order was consistent with that for the EO content in the copolymers. Therefore, the folding behavior of poly(eo-co-po) copolymer depends on its EO unit content. At the same molecular weight, the polymer chain length, or total monomer unit number, increased with the increase in EO content, because mass of EO is smaller than that of PO. The decrease in the charge density due to the increased chain length will be effective for folding. Composition-based separation of poly(eo-co-po) by ESI-IMS-MS At the same molecular weight, the drift time for PEO with 3Na + addition was almost the same as that for PPO, as shown in Fig. 2A. In the case of 2Na + addition, the difference in drift time between PEO and PPO was found only in the region with the folding phenomenon. On the other hand, in the case of 4Na + addition, the drift time for PEO was slightly larger than that for PPO in the entire region. That is, the drift time of PEO was different from that of PPO under high Na + addition conditions. Therefore, composition-based separation of poly(eo-co-po)

5 ANALYTICAL SCIENCES FEBRUARY 209, VOL Conclusions ESI-IMS-MS was applied to the analysis of multiply charged poly(eo-co-po)s. The effect of charge number on the drift time and folding behavior was similar to that reported for homopolymers. 0 9 The folding behavior was dominated by the charge density of the multiply charged polymer. The EO content in the copolymer affected the folding behavior, that is, higher EO content resulted in a more notable folding effect. When the molecular weight was the same, in the case of poly(eo-co-po), the EO content dominated the polymer main chain length or polymerization degree, that is, a higher EO content resulted in a longer main chain and a decrease in the charge density. The difference in the main chain length or charge density would be the key factor influencing the folding behavior of multiply charged polymers. In addition, the side chain also affected the CCS of the polymer. It is reasonable to conclude that the CCS of PPO is larger than that of PEO when the polymerization degree is the same, as shown in Fig. 2C. On the other hand, when the molecular weight is the same, the CCSs of PEO and PPO depended on the charge number, as shown in Fig. 2B. In the case of z = 4, the CCS for PEO is slightly larger than that for PPO. This behavior was exploited for the successful compositionbased separation of poly(eo-co-po)s, as demonstrated in Fig. 5. In the case of poly(eo-co-po), the differences in structure, molecular weight, and chemical property between EO and PO are not very large. When the differences between the properties of the monomer units are significant, composition-based separation by ESI-IMS-MS will become easy. Further investigation in this regard is ongoing in our laboratory. References Fig. 5 Separation of PEO-b-PPO-b-PEO (EO unit 0 wt%) and poly(eo-r-po) (EO unit 70 wt%) by ESI-IMS-IMS. The 2D heat map of the m/z value and drift time is shown in (A), and the extracted mass spectra of zones # and #2 are shown in (B) and (C), respectively. copolymers may be achieved by ESI-IMS-MS analysis with highly charged analytes. Figure 5 shows the analytical results for the mixture of PEOb-PPO-b-PEO (EO unit 0 wt%) and poly(eo-r-po) (EO unit 70 wt%) in a 2D heat map. The analytical conditions were optimized for observing the highly charged analytes. Under these conditions (Fig. 5A), only two zones were observed, in contrast to Fig., and this will be suitable for achieving simple separation based on the difference in polymer composition by IMS-MS. The extracted mass spectra from regions # and #2 in Fig. 5A are shown in Figs. 5B and 5C, respectively. Zone # included the poly(eo-co-po) of approximately 75% EO content with 4 and 5 Na + additions and zone #2 included that of 5 20% EO content with 3 and 4 Na + additions. Therefore, it is concluded that zone # included poly(eo-r-po) (EO unit 70 wt%) while zone #2 included PEO-b-PPO-b-PEO (EO unit 0 wt%). That is, poly(eo-co-po) copolymers with different EO content were successfully separated by ESI-IMS-MS.. A. B. Kanu, P. Dwivedi, M. Tam, L. Matz, and H. H. Hill Jr., J. Mass Spectrom., 2008, 43,. 2. B. T. Ruotolo, J. L. P. Benesch, A. M. Sandercock, S.-J. Hyung, and C. V. Robinson, Nat. Protoc., 2008, 3, C. Uetrecht, R. J. Rose, E. van Duijn, K. Lorenzen, and A. J. R. Heck, Chem. Soc. Rev., 200, 39, S. Armenta, M. Alcala, and M. Blanco, Anal. Chem. Acta, 20, 703, C. Lapthorn, F. Pullen, and B. Z. Chowdhry, Mass Spectrom. Rev., 203, 32, F. Lanucara, S. W. Holman, C. J. Gray, and C. E. Eyers, Nat. Chem., 204, 6, I. D. G. Campuzano and P. D. Schnier, Int. J. Ion Mobility Spectrom., 203, 6, M. T. Jafari, B. Rezaei, and H. Bahrami, Anal. Sci., 208, 34, G. von Helden, T. Wyttenbach, and M. T. Bowers, Int. J. Mass Spectrom. Ion Process., 995, 46 47, S. Trimpin, M. Plasencia, D. Isailovic, and D. E. Clemmer, Anal. Chem., 2007, 79, S. Trimpin and D. E. Clemmer, Anal. Chem., 2008, 80, G. R. Hilton, A. T. Jackson, K. Thalassinos, and J. H. Scrivens, Anal. Chem., 2008, 80, J. N. Hoskins, S. Trimpin, and S. M. Grayson, Macromolecules, 20, 44, C. A. Scarff, J. R. Snelling, M. M. Knust, C. L. Wilkins, and J. H. Scrivens, J. Am. Chem. Soc., 202, 34, E. C. Hidalgo, J. F. García, and J. F. de la Mora, Anal.

6 74 ANALYTICAL SCIENCES FEBRUARY 209, VOL. 35 Chem., 203, 85, D. Morsa, T. Defize, D. Dehareng, C. Jérôme, and E. D. Pauw, Anal. Chem., 204, 86, W. Zhang, M. Quernheim, H. J. Räder, and K. Müllen, Anal. Chem., 206, 88, Q. Duez, F. Chirot, R. Liénard, T. Josse, C. M. Choi, O. Coulembier, P. Dugourd, J. Cornil, P. Gerbaux, and J. D. Winter, J. Am. Soc. Mass Spectrom., 207, 28, S. Kokubo and P. Vana, Macromol. Chem. Phys., 207, 28, D. P. Smith, T. W. Knapman, I. Campuzano, R. W. Malhan, J. T. Berryman, S. E. Radford, and A. E. Ashcroft, Eur. J. Mass Spectrom., 2009, 5, J. Gidden, T. Wyttenbach, A. T. Jackson, J. H. Scrivens, and M. T. Bowers, J. Am. Chem. Soc., 2000, 22, J. D. Winter, V. Lemaur, R. Ballivian, F. Chirot, O. Coulembier, R. Antoine, J. Lemoine, J. Cornil, P. Dubois, P. Dugourd, and P. Gerbaux, Chem. Eur. J., 20, 7, Q. Duez, T. Josse, V. Lemaur, F. Chirot, C. M. Choi, P. Dubois, P. Dugourd, J. Cornil, P. Gerbaux, and J. D. Winter, J. Mass Spectrom., 207, 52, 33.