UvA-DARE (Digital Academic Repository) Separation and ... · MassMass Spectrometry Characterization...

UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl)

UvA-DARE (Digital Academic Repository)

Separation and characterization of functional polymers

Jiang, X.

Link to publication

Citation for published version (APA):Jiang, X. (2004). Separation and characterization of functional polymers.

General rightsIt is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s),other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulationsIf you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, statingyour reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Askthe Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam,The Netherlands. You will be contacted as soon as possible.

Download date: 28 Jan 2020

https://dare.uva.nl/personal/pure/en/publications/separation-and-characterization-of-functional-polymers(82572102-1b6b-4a00-8662-c4e92d7c9de2).html

CHAPTERR 3

Mass-Spectrometricc Characterization of

Functionall Poly(Methyl Methacrylate)

inn Combination with Critica l LC

ABSTRACT T

State-of-the-artt techniques for the mass-spectrometric characterization of synthetic polymers have beenn applied to functional poly(methyl methacrylate) (PMMA), synthesized by reversible addition-fragmentationn chain-transfer (RAFT) polymerization. The polymers were first separated effectively accordingg to functionality by liquid chromatography (LC) at the critical conditions (i.e., almost no influencee of molecular weight on retention). The separated polymers were characterized off-line by matrix-assistedd laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS), andd both off-line and on-line by LC-electrospray-ionization-quadrupole-TOF-MS (LC-ESI-QTOF-MS).. The on-line ESI experiments confirmed a clear baseline separation of the hydroxyl-functional prepolymerss according to the number of hydroxyl groups. Labile end-groups of PMMA, such as the dithioesterr group, were lost in the MALDI-TOF-MS experiments, while they were observed intact inn the ESI-QTOF-MS spectra. This indicates that in the present case ESI is a much softer ionization techniquee than is MALDI . The ESI-MS experiments provided direct evidence that the RAFT polymerss still exhibited living characteristics in the form of the dithio moiety.

Keywords:: Critical liquid chromatography, hydroxyl group, telechelic polymers, RAFT polymers, LC-ESI-MS,, MALDI , fragmentation.

XulinXulin Jiang, Peter J. Schoenmakers, Joost L. J. van Dongen, Xianwen Lou, Vincent Lima and José Brokken-Zijp, Anal.

Chem,Chem, 75 (2003) (2003) 5517-5524.

ChapterChapter 3

3.11 Introductio n Functionall {pre-)polymers can be used as cross-linking components in coating formulations to createe hard, durable films. For example, telechelic prepolymers (containing two hydroxyl end-groups)) with narrowly distributed molecular weights can react with trifunctional isocyanates to formm a well-defined polyurethane network. It is expected that the study of these systems will providee better insights into the relations between the network structure and the properties of cross-linkedd coating systems. Eventually, this should also lead to coatings with better properties. However,, it is very difficult to produce the required telechelic polymers by conventional polymerizationn methods.

Fortunately,, the recent development of controlled/'living' radical polymerization has created possibilitiess for the synthesis of many well-defined polymers with designed architectures and predictablee molecular weights [1,2]. The most important methodologies include nitroxide-mediated radicall polymerization [3], atom-transfer radical polymerization (ATRP) [4], and lately reversible addition-fragmentationn chain-transfer (RAFT) polymerization [5]. Among these, RAFT polymeriza-tionn arguably has the greatest commercial impact, because it only involves organic substances, it hass a high tolerance to impurities, and it can be applied for a wide range of monomers, including acrylicc acid [2]. Therefore, we have selected RAFT polymerization for the preparation of bifunctionall telechelic (mcth)acrylate polymers with a low polydispersity index. RAFTT polymerization relies on the use of thiocarbonylthio-compounds (or dithioesters) of general structuree Z-C(=S)S-R. R is the homolytic leaving group of the RAFT agent; Z is an activating group.. The dithioesters act as reversible addition-fragmentation chain-transfer agents, allowing the formationn of polymers with functional end-groups [5,6]. A simplified mechanism of the RAFT processs is given in scheme 3.1. Initiator-derived primary radicals [I* ] react with monomer units to formm oligomeric (propagating) radicals [Pn*] , which undergo addition to thiocarbonylthio compoundss to form adduct radicals. The resulting species fragments into a polymeric thiocarbonylthioo compound and a homolytic leaving group. The latter ([R*]) is capable of reinitiatingg the polymerization to give a new propagating radical [Pm*] . Equilibrium is then establishedd between all the active propagating radicals [Pn

# and Pm"] and the dormant polymeric thiocarbonylthio-compoundss by way of the intermediate radicals. Most of the polymers obtained at thee end of a RAFT polymerization will contain the leaving group of the RAFT agent, but if initiatorss are used a (small) fraction will contain the initiating radical fragments (I). However, in somee experiments [7,8,9] the amount of initiator-derived polymer was too low to be detected. Destaracc et al, [10] and Schilli et al [11] did observe signals corresponding to polymers with initiatorr end-groups in matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS)) of RAFT polymers.

44 4

MassMass Spectrometric Characterization of Functional PMMA in Combination with Critical LC

Initiatio nn and propagatio n

II * + n M ». Pn ' kv kv

Chai nn Transfe r ^addd £p

1,, ^-add T T

Reinitiatio nn and propagatio n

R** + m M Pm*

Chai nn equilibriu m

Pmm + S ^ S - P n ~ P ^ S ^ S - P n , Pm- S S + P„

zz z

M M

kkpp kp

Schemee 3.1 A simplified mechanism of the RAFT process (involves the addition of monomer, an

initiator,, and the essential ingredient of a reversible transfer agent)

Masss spectrometry plays an increasingly important role in polymer analysis, because of its high sensitivity,, broad dynamic range, specificity, and selectivity [12,13,14,15]. Two significant developmentss in ionization techniques in the late 1980s - MALDI and electrospray ionization (ESI) -- greatly enhanced the applicability of MS for polymer characterization. These techniques enable thee ionization of large nonvolatile molecules with good integrity, thus allowing determination of the weightt of intact polymer molecules by MS. However, to study the heterogeneity of complex polymerr systems at the molecular level, it is vital that MALDI or ESI-MS is combined with a suitablee separation technique [16,17,18,19]. MALDI is the newest and reputedly most promising ionizationn method for synthetic polymers [13]. Thanks to the emergence of MALDI , polymer scientistss have taken a greater interest in MS [13]. In publications relating to the characterization of syntheticc (co-)polymers by mass spectrometry, MALDI is favored as the ionization method, becausee only singly charged ions are observed [15], producing much clearer spectra. However,, only a few references contain MALDI mass spectra of polymers obtained by RAFT polymerizationn [7,8,10,11,20,21]. The spectra obtained in these reports do not provide direct structurall information on, for example, end-groups of the RAFT polymers. Vosloo et al. [20] presentedd a MALDI-TOF-MS spectrum of RAFT polystyrene, which contained several series of peaks.. They stated that it was difficult to obtain useful MALDI-TOF-MS spectra of these polymers. Destaracc et al. [10] published a MALDI-TOF-MS spectrum of RAFT polyvinyl acetate). They observedd initiator-derived and hydrogen-terminated polymers. DAgosto et al. [21] reported the

45 5

ChapterChapter 3

MALDI-TOF-M SS spectrum of poly(N-acryloylmorpholine), synthesized by RAFT polymerization. Theyy described that in the MALDI instrument the RAFT polymer experienced fragmentation in the laserr beam to lose the dithioester moiety. Subsequently, the macroradical formed could abstract a hydrogenn from the matrix, giving rise to hydrogen-terminated polymers. Ganachaud et al. [7] and Schillii et al. [11] reported MALDI mass spectra of RAFT poly(N-isopropylacrylamide). Schilli et al.al. gave a good discussion on the fragmentation of RAFT polymers during ionization. Vana et al, [8]] ascribed the multiple peaks within one repeat unit of RAFT poly(methyl acrylate) (PMA) to the complexx MALDI ionization and fragmentation processes. Although MALDI was initially described ass a "soft ionization" process, a significant extent of fragmentation may occur for MALD1-generatedd ions [22,23]. This fragmentation usually occurs not only for the RAFT polymers mentionedd above, but also for polymers obtained by other controlled/'living' radical-polymerization methods,, such as nitroxide-mediated radical polymerization [24,25,26] and ATRP [27-33]. Nonaka etet al. [32] suggested that the terminal C-Cl group of PMA obtained by ATRP is relatively stable in comparisonn with that of poly(methyl methacrylate) (PMMA) during MALDI-TOF-MS analysis. Musatt et al. [23] analyzed some labile low-molecular-weight polyesteramides by field desorption (FD)) MS, ESI-MS, and MALDI-MS. They found that ESI yielded the least ion fragmentation and samplee decomposition; in other words, they found ESI to be the "softest" ionization method. FD-MSS rated second (less soft) and MALDI-M S third (least soft). Because of the reasonable expectationn that ESI-MS would yield less fragmentation and therefore clearer spectra, we tried to applyy ESI-MS to analyze RAFT polymers and to compare the results with those obtained by MALDI-TOF-MS.. Our objective was to obtain mass spectra containing peaks representative of intactt molecular ions, to unambiguously assign the polymer end-groups.

Inn polymer LC, critical conditions refer to a separation in which retention is independent of the numberr of monomeric units in a homopolymer chain [19]. Under critical conditions, differences in retentionn {i.e. selectivity) are solely caused by the end-groups and by other (functional) groups presentt in the polymer molecules. Critical separations are of great potential interest, but critical conditionss are typically hard to realize and very hard to maintain through extended series of LC experiments.. In a previous study [34], robust critical LC conditions were established for PMMA in normal-phasee LC using mixtures of acetonitrile and dichloromethane as the mobile phase on a bare silicaa column. Baseline separations were obtained for low-molecular-weight RAFT prepolymers, eitherr with or without end-group modification. Separation was thought to be solely based on the numberr of hydroxyl groups present in the molecules. In this work, off-line LC//MALDI-TOF-MS andd both off-line and on-line LC-ESI-QTOF-MS have been used to identify the repeating units and, especially,, the end-groups of the polymers. This should allow us to confirm the critical separation off non-functional, mono-functional, and di-functional polymers.

46 6

MassMass Spectrometry Characterization of Functional PMMA in Combination with Critical LC

3.22 Experimental

3.2.11 Chemicals

Dichloromethanee (DCM), tetrahydrofuran (THF), and acetonitrile (all HPLC grade), were from Rathburnn Chemicals (Walkerburn, Scotland). Polymer sample V37A was synthesized by reversible addition-fragmentationn chain-transfer (RAFT) polymerization using a 2,2'-azobisisobutyronitrile (AIBN)) initiator and a hydroxyl-functional RAFT chain-transfer agent [35]. Sample V37B was obtainedd via end-group modification of V37A, with cleavage of the RAFT activating group [35]. Thee molecular-weight distributions were measured by size-exclusion chromatography (SEC). Calibrationn is based on polystyrene standards and the molecular weights were recalculated using the universal-calibrationn principle and Mark-Houwink parameters [34]. The SEC data of samples V37A andd V37B are summarized in table 3.1. Sample V37A0 is the non-functional fraction of V37A collectedd between 1.7 and 2.2 min, V37A1 is the mono-functional fraction of V37A collected betweenn 2.3 and 3.3 min, V37B0 is the non-functional fraction of V37B (1.7-2.2 min), and V37B1 iss the mono-functional fraction of V37B (2.3-3.3 min; see Fig. 3.1).

Tablee 3.1 The molecular weights (Mm Mp ) and polydispersity indices

(PDI)(PDI) of PMMA samples, measured by SEC.

Samplee name

V37A* *

V37B* *

M„ M„

2,587 7

2,853 3

MMp p

3,754 4

3,678 8

PDI I

1.29 9

1.29 9

Synthesizedd by RAFT polymerization using a 2,2'-azobisisobutyronitrile (AIBN)) initiator and a hydroxyl-functional RAFT chain-transfer agent [35]. Synthesizedd via end-group modification of sample V37A with cleavage of thee RAFT activating group [35].

3.2.22 Equipment

Thee HPLC equipment used was the same as described in the previous paper [34]. The MALDI -TOF-MSS measurements were performed with a Voyager-De Pro instrument (PerSeptive Biosystems,, Framingham, MA, USA) equipped with a 337-nm nitrogen laser. Spectra were acquiredd by summing the data obtained from 200 laser shots in the reflector mode. The laser energy perr pulse was tuned to yield sufficient ionization, while minimizing fragmentation. Absolute intensitiess were similar to the value of 80 uJ reported by Scrivens et al. [36]. a-Cyano-4-hydroxycinnamicc acid (about 20 mg/ml in THF) was used as the matrix. The concentration of the polymerr sample was about 1 mg/ml in THF.

Thee off-line ESI-MS experiments were carried out using a Q-TOF Ultima™ Global mass spectrometerr (Micromass, Manchester, UK) equipped with an atmospheric-pressure-ionization electrosprayy interface. The instrument was calibrated with phosphoric acid in the mass range of 98 too 2058 amu. Polymer samples were dissolved in a 4:1 v/v mixture of dichloromethane/methanol at

47 7

ChapterChapter 3

aa concentration of 0.5 mg/ml. The flow rate of the sample introduced into the electrospray interface wass 20 ul/min. The sampling cone potential was 31 V and the capillary voltage 3.0 kV. The electrospray-sourcee temperature was 80°C and the desolvation temperature 100°C. Mass spectra weree scanned over the range m/z 500-3000 in positive-ion mode. More than 100 scans were summedd to produce the final spectrum.

Thee on-line LC-ESI-QTOF-MS experiments were carried out with an HPLC system consisting of twoo Shimadzu LC-10ADvp pumps (high-pressure gradient system to prepare the mobile phase in-

situ).situ). A post-column addition of methanol at a flow rate of 20 (il/min was delivered by another Shimadzuu LC-10ADvp pump to enhance the ionization efficiency. The concentration of the polymerr injected was about 1 mg/ml in DCM. Care was taken to avoid breakthrough peaks [37]. Thee other ESI operation parameters were the same as in the off-line process described above. Tandem-MSS (ESI-MS/MS) experiments were carried out on the selected precursor ions by collision-inducedd dissociation (CID) using helium as collision gas and a collision energy of about 500 V to obtain the fragmentation-ion spectrum.

48 8


3.33 Results and discussion

Inn a previous study [34] PMMA prepolymers were successfully separated according to hydroxyl functionalityy under the critical LC conditions. These conditions were used to analyze the functional RAFTT polymers. The representative chromatograms of two low-molecular-weight functional-polymerr samples are shown in Fig. 3.1. It is relatively straightforward to collect fractions (identified inn Fig. 3.1) and to perform off-line mass spectrometry.

V37A1 1

A A V37A00 11

II I I MV37B1 1

V 3 7 B 00 / \ V37B2

ll 1 1 1

V37A A

V37B B

ll 1

00 1 2 3 4 5 6

Timee (min)

Fig.. 3.1 Separation of functional RAFT polymers according to the number of OH groups. Evaporative

light-scatteringg detector, mobile phase 40% acetonitrile in dichloromethane, home-packed

silicaa column (150 mm x 4.6 mm i.d.; 3-(xm particles; 100-A pore size), flow rate 1 ml/min.

3.3.11 Comparison of MALDI-TOF-M S and ESI-QTOF-MS

Becausee low-molecular-weight materials yield very high MALDI signals, even if present at very loww concentrations, the complete spectra are not very useful. This is indicated in Appendix A Fig. Ap-1,, where spectra are shown across a broader range, especially for sample V37B1. Fig. 3.2 shows thee most useful range of the MALDI-TOF-MS spectra of four different fractions. It can be seen fromm the spectra in Fig. 3.2 that two different series were observed with a repeating unit of 100.1 amu,, which clearly corresponds to a single methyl methacrylate (MMA) monomeric unit, in all of thee mass spectra for the fractions V37A0, V37B0, V37A1, V37B1. The main series of peaks result fromm adducts with one sodium cation (Na+). The minor series (seen in Fig. 3.3a and Appendix A Fig.. Ap-3a) can be assigned to the potassium-cation (K+) series (with mass increments of 16 amu relativee to the Na+ series). The results are summarized in table 3.2. In addition, we were unable to assignn another minor series (with masses 25 amu lower than those of the main peaks) for V37A0 (Fig.. 3.3a), for which there were many possibilities. There is another minor series (with masses 16

49 9

ChapterChapter 3

amuu lower than those of the main peaks) that can be seen in Fig. 3.2 for V37B1. These peaks are nott likely to be due to the loss of an OH group, as they are not observed for any of the other structures,, which contain the same OH moiety. Also, a difference of 16 amu is not expected to resultt from the loss of a hydroxyl group. More likely, these peaks may originate from oligomeric speciess BIS undergoing cyclisation to yield oligomers with structure BIRi (see table 3.2 for abbreviations).. Such a mechanism would be analogous to the loss of BrCH3 followed by cyclisation off the final two methyl methacrylate monomer repeat units to yield a lactone end-group, as demonstratedd by Borman et al. [30] for PMMA prepared by ATRP.

'm^^* '' J L i Xi 1,, J^LJULJUL

mmm<mmm< JiwJuAnLiJLJjNJUUL

V37A0 0

kLkL i J*. J>I»I>LI JLJJ*. JL,.^LiLi J V37B0 0

V37A1 1

JL L

M. . Ik. .

V37B1 1

I L L 1111111111 i i 11 i i 111111111 i i 11111111 i i 1111111111111 i n i i i i n 1111 i i 111

11000 1300 1500 1700 1900 2100 2300 2500

m/z z

Fig.. 3.2 MALDI-TO F mass spectra for the first two fractions of sample PMMA V37A and V37B.

(V37A0)) Non-functional fraction of V37A collected at 1.7-2.2 min; (V37A1) Mono-

functionall fraction of V37A (2.3-3.3 min); (V37BO) Non-functional fraction of V37B (1.7-2.2

min).. (V37B1) Mono-functional fraction of V37B (2.3-3.3 min). LC separation conditions

weree the same as in Fig. 3.1, except a flow rate of 0.5 ml/min.

50 0


Tablee 3.2

Sample e name e

V37A0 0

V37A1 1

V37B0 0

V37B1 1

Structure e (abbr.1) )

AOH H

AOR R

A1H H

AI R R

BOH H

BOH H

BOS S

B1H H

BIRi i B1H H

BIS S

Structurall assignment of the peaks displayed in the MALDI-TO F and ESI-

QTOFF mass spectra of the PMMA samples reported in this work.

Slope2 2

100.07 7

100.05 5

100.06 6

100.06 6

100.03 3

100.06 6

100.05 5

100.02 2

100.05 5 100.06 6

100.05 5

Intercept11 End-group, mh Exper.. Calc.

92.055 69.06 69.06

unknown n 244.033 221.04 221.03

136.188 113.19 113.08

288.055 265.06 265.06

92.066 69.07 69.06

92.166 69.17 69.06

124.377 101.38 101.03

136.400 113.41 113.08

120.099 97.10 97.05 136.055 113.06 113.08

168.066 145.07 145.06

Representativee ions (m/z ) Exper.44 Calc.5 (n)

1492.96 6 1509.00 1509.00 1467.93 3 1644.79 9 983.98 8

1537.08 8 1553.00 0 1688.84 4 1006.01 1 1045.19 9 1492.38 8 1508.41 1 1092.52 2 907.82 2 1124.56 6 973.83 3 1536.89 9 1552.80 0 1520.82 2 1036.55 5 1030.04 4 961.16 6 968.49 9 946.00 0 1005.22 2

1492.75(14) ) 1508.86(14) )

unknown n 1644.76(14) ) 983.95(17) ) 1536.78(14) ) 1552.89(14) ) 1688.78(14) ) 1005.97(17) ) 1045.17(28) ) 1492.75(14) ) 1508.86(14) ) 1092.55(10) ) 907.94(19) ) 1124.52(10) ) 973.96(18) )

1536.78(14) ) 1552.89(14) ) 1520.75(14) ) 1036.55(9) ) 1030.03(19) ) 961.16(27) ) 968.477 (8) 945.96(17) ) 1005.16(28) )

Associated d ion n

Na* *

unknown n Na ^

(2Na)2+ +

Na* *

Na+ +

(2Na)2+ +

(3Na)3~ ~ Na' ' K+ +

Na* * (2Na)2* *

Na+ +

(2Na)2t t

Na+ +

K+ +

Na+ +

Na* * (2Na)2, ,

(3Na)3+ +

Na" " (2Na)2* * (3Na)J+ +

Ionization n method d

MALD I I

ESI I

MALD I I

ESI I

MALD I I

ESI I

MALD I I

ESI I

Abbreviations: : CH, ,

AOH,, BOH: C H 3 - C [MMAJ^H

CN N

CH33 s

A 0 R :: CH3 C [MMA]„- S C

CN N

CH3 3

BOS:: C H3 - C - [ M M A } n SH

CN N

22 The molar mass of the monomer calculated from peakk series in MS, see Appendix A. Thee total molar mass of polymer-chain end-groups

andd associated cations calculated from peak series inn MS, see Supporting Info. 44 Experimental m/z value of end-groups: intercept minuss associated cation. Theoreticallyy calculated m/z value.

CH3 3

A1H,, B1H: HO-(CH2)3-C-[MMA] nH

CN N

C H33 S

H C H C H ^CC [MMA] n S- C

CN N

CH3 3

H C M C H^^ C {MMA} n SH

CN N

CH3 3

AIR: :

BIS: :

BIRi: :

C H33 CH2 CH3

HCMCHjb-CC {MMAJ n.2-CH2 C C CDCD *-U2L.rl 3

CNN a

51 1

ChapterChapter 3

Surprisingly,, the MALDI-TOF mass spectrum of the fraction V37B0 is very similar to the one of V37A0,, and V37B1 is roughly identical to V37A1. Yet, the end-groups of the polymers are known too be different. The UV (ultraviolet) spectrum of sample V37A showed strong absorbance with a maximumm at 300 nm in dichloromethane, which is ascribed to the RAFT dithioester moiety [S=C(C6H5)S-].. The UV spectrum of sample V37B, which was prepared by removing the RAFT dithioesterr moiety via end-group modification of the RAFT polymer [35], showed no UV absorbancee at 300 nm.

Panelss a and b of Fig. 3.3 compare the MALDI and ESI spectra for non-functional PMMAs V37A0 inn the mass range of 1,400 to 2,050 amu. The ESI-QTOF mass spectrum is different from the MALDI-TO FF spectrum for the same sample. End-groups calculated from the MALDI spectrum of V37A00 are different from those obtained from the ESI spectrum. In the former case the total end-groupp mass is consistent with hydrogen and initiator moiety [(CH.^CCN-] as the end-groups. In the latterr case the end-groups of V37A0 can be assigned to the RAFT dithioester moiety and the initiatorr moiety, which correspond to the non-functional structure. The fragments lost in the MALD II experimental process account for 152 amu.

Wee also compare the ESI and MALDI spectra for mono-functional PMMA V37A1 (see Appendix AA Fig. Ap-3). Again, the ESI mass spectrum is different from that obtained by MALDI for the fractionn V37A1. The total end-group mass calculated from the MALDI spectrum of V37A1 is differentt (152 amu lower) than that obtained from the ESI spectrum. From the ESI spectra we can concludee that sample V37A1 does contain PMMA with the RAFT dithioester moiety and the leavingg group of the RAFT agent [HO(CH2)3CH3CCN-], which together result in a mono-functional structuree (one hydroxyl group).

Theree are also differences between MALDI-M S and ESI-MS spectra for samples V37B0 and V37B11 (see discussion below). The end-groups observed in ESI-MS are different from those observedd in MALDI-MS, as shown in table 3.2.

Inn conclusion, the weakly bonded end-groups were cleaved in the MALDI-M S experimental process,, while the polymers were observed intact in ESI-MS. This indicates that in the present case ESII is a much softer ionization technique than MALDI . This is in agreement with the results obtainedd by Musat et at. [23].

3.3.22 Off-lin e LC//ESI-QTOF-M S and MS2

Fig.. 3.4a shows an expanded portion of the ESI spectrum for V37A1 in the range of 980 to 1100 amu.. It can be seen clearly from Fig. 3.4a and table 3.2 that singly charged ions, doubly charged ionss (difference between isotopic peaks 0.5 amu) and triply charged ions (isotopic peaks separated byy 0.33 amu) occur in the ESI spectrum. These different series of peaks arise from the same series off PMMA (same end-groups).

52 2


1492.966 1593.08

1693.11 1

(a)) MALDI (V37A0)

1793.14 4

1893.30 0 1993.36 6

JLJU i i

100 0

II I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I

14000 1500 1600 1700 1800 1900 2000

1444.688 m/z (b) ESI (V37A0)

% %

0 0 i i

1445.69 9

1545.73 3 1544.70 0

1446.72 2

1485.24 4

1644.81 1

II 1645.79

1546.75 5

1585.33 3

1745.84 4 1746.85 5 1844.95 5

1647.79 9

i k.iLjL,üJ^L,,,.iilJl,uu iJ,,u^ilr,. t f,i

1845.977 1945.98

1846.966 1946.94 2045.99 9

14000 1500 1600 1700 1800 1900 2000 m/z z

Fig.. 3.3 Comparison of MALD I and ESI mass spectra for non-functional RAFT polymer (V37A0). (a)) MALDI-TOF-MS; (b) ESI-QTOF-MS.

53 3

ChapterChapter 3

100 0 yöa.43 3 (a)ESI(V37Al) )

1088.51 1

989.45 5

1006.52 2 1006.01 1

\ \ 990.46 6

1056.52 2 1007.01 1

1007.51 1 1045.54 4

0 ) U É É

980 0

il^ll11,lt.,iV» l ..ia-rf i-' ' umuÊlih, umuÊlih,

1078.89 9

AA l m "tu i iHlllÊM

1089.50 0

1090.51 1

^^.^ulii.lLukm.Uj^r', ^^.^ulii.lLukm.Uj^r', m/z z 1000 0 1020 0 1040 0 1060 0 1080 0 1100 0

100 0 335.19 9 (b)ESI-MS2(V37Al) )

% %

235.13 3 934.50 0

336.20 0

435.23 3

935.54 4

936.48 8

1088.49 9

L JJ o . m/z z 400 0 6000 800 1000 1200 1400

(+Na+) )

CH, ,

(+Na+)) (+Na+) 2355 335 934

CH33 ! CH3 ! CH3 CH2f H „

CC CR?-6^tCH2-C.^5CH2-C- - S-C-

(c) )

nO-iCHnO-iCH22)T)TCC^ CH2—y- - ^ n2 — y — ^ n 2 — y — 75

CNN C = 0 C = 0 C = 0 C = 0

OCH, , OCH33 OCH3 OCH, ,

Fig.. 3.4 ESI-QTOF mass spectra of mono-functional RAFT polymer (V37A1). (a) Enlarged part of

ESI-QTOF-MS;; (b) ESI-MS2 selected at 1088 amu; (c) the proposed fragmentation of the 8-

merr in ESI-CID-MS2.

54 4


Too verify the obtained structure, multiple-stage MS (ESI-MS2) experiments were carried out on selectedd precursor ions with helium as collision gas. This technique takes advantage of the ion-trap mechanismm and allows for the fragmentation of a selected precursor ion by collision-induced dissociationn (CID) and results in a product-ion spectrum. Fig. 3.4b shows a representative MS2

spectrumm of the precursor ion with mass-to-charge ratio m/z 1088 corresponding to n = 8 (V37A1) withh a collision voltage of 50 V. The RAFT dithioester moiety (152.98 amu) was observed to split offf easily. The resulting macroradical (the polymer chain without the RAFT group) loses an additionall hydrogen to give rise to the most abundant peak at m/z 934.50. Also a series of fracture ionss with m/z 135, 235, 335, and 445 were observed. These correspond to PMMA chains with the RAFTT leaving group and a sodium cation. The fragmentation scheme is shown in Fig. 3.4c. Thee ESI-CID-MS2 spectrum (Appendix A Fig. Ap-4) for the m/z=1044 (w=8) molecular-ion peak of V37A00 shows similar results. The RAFT dithio moiety (152.98 amu) was again observed to split offf easily. An additional hydrogen is abstracted from the polymer chain, to give rise to the most abundantt peak at m/z 890.63. A series of fracture ions with m/z 191, 291, 391, etc. were observed, whichh correspond to PMMA prepolymers with an initiator moiety as an end-group. Al ll our results can be explained from the RAFT polymerization mechanism (Scheme 1). Our results providee direct evidence that the RAFT polymers still present "latent living" (or dormant) characteristicss in the form of the dithioester end-group. It can be used for further-chain-extension if neww batches of initiator and monomer are added. We can conclude that the dithio moiety very easily splitss off from the RAFT polymer, as shown in MALDI-TOF-MS as well as in ESI-CID-MS2

experiments. .

Wee proved directly that the RAFT polymer V37A contained not only a fraction V37A1 featuring thee leaving group of the RAFT agent, but also a fraction V37A0 featuring initiator moiety. Therefore,, an initiator with a radical fragment (I) that has the same structure as the leaving group of thee RAFT agent should be used to synthesize RAFT polymers with a well-defined functionality. Basedd on this knowledge, a new hydroxy 1 -functional initiator was used to obtain a well-defined mono-functionall RAFT polymer with a narrow molecular-weight distribution (see references 34, 35). .

Inn order to characterize sample V37B, fractions V37B0 and V37B1 were also subjected to ESI-QTOF-MS.. For the di-functional fraction of V37B2 collected at 3.3-4.3 min, highly complex ESI-MSS spectra were obtained, which will not be discussed in the present paper. Fig. 3.5a shows an enlargedd part of the ESI spectrum for mono-functional polymer V37B1 in the mass range of 935 to 10400 amu. Despite the rather low ionization efficiency, which is apparent from the relatively high base-linee noise, the qualitative information from this spectrum is clear. It can be seen from Fig. 3.5a andd table 3.2 that singly, doubly and triply charged ions are all present in the spectrum of V37B1, (somewhat)) enhancing the complexity. There are two dominant structures B1H and BIS (see table 3.22 and Appendix A table Ap-1). The ESI-CID-MS2 spectrum for the m/z 1036 (n=9) molecular-ion peak,, which is shown by way of example in Fig. 3.5b, provides further evidence for the structure of B1H.. A series of fracture ions was identified with m/z 135, 235, 335, and 445, which correspond to PMMAA polymers with the RAFT leaving group as a chain-end. The most abundant peak at m/z 210

55 5

ChapterChapter 3

andd the associated series correspond to hydrogen as another end-group in sample V37B1. The fragmentationn scheme is shown in Fig. 3.5c (see reference 36). The analysis of sample V37B0 via ESI-MSS is similar and the results are also summarized in table 3.2.

100--(a)ESI(V37Bl) )

937.48 8

9400 950 960 970 9900 1000 1010 1020 1030 1040

100--210.05 5

% % 135.09 9 235.13 3

335.20 0

(b)) ESI-MS2 (V37B1)

1036.56 6

435.277 535.27

- -LL,—L ii LL.I.L,L.,L..I I.L,L.1„

1038.57 7

1000 200 300 400 500 600 700 800 900 1000 1100

(C) ) (+Na+) ) 235 5

(+Na+) ) 335 5

CH3 3

HCHCH^C-- CH2

CN N

CH, , CH, , CH, ,

(+Na+) ) 210 0

!! "CH, CH, ,

- C - - C H , - C-- CH, , -C—-fcCHff-- C- CHjC H

C =0 0

OCH3 3

c=o o OCH3 3

c=o o OCH3 3

c=oo c=o OCH33 OCH3

Fig.. 3.5 ESI-QTOF mass spectra of mono-functional polymer V37B1. (a) Enlarged section (935-1040 amu); (b)) ESI-MS2 with 1036 as the parent ion; (c) the proposed fragmentation of the 9-mer in ESI-CID-MS2

56 6

MassMass Spectrometry Characterization of Functional PMMA in Combination with Critical LC

3.3.33 On-line LC-ESI-QTOF-M S

Inn order to further confirm the hydroxy 1-based separation, we used on-line LC-ESI-QTOF-MS to identifyy the repeating units and, especially, the end-groups. Two narrow-bore columns (2 x 150 mm xx 1.0 mm i.d) were used, as these were compatible with the ESI interface without post-column splittingg of the effluent (LC eluent flow rate 0.1 ml/min). However, a mobile phase containing 40% acetonitrilee in dichloromethane could not be used to elute the PMMA samples from the two columns.. The critical solvent composition for the two small columns was about 5% higher in acetonitrilee concentration compared with the conventional column (150 mm * 4.6 mm i.d).

Ann example of the summed mass chromatogram of sample V37A using (near) critical conditions (45%% acetonitrile in dichloromethane) is shown in Fig. 3.6. Fig. 3.6a shows the summed chromatogramm of masses set at 244+100.06/7 and 288+100.06/7 (n is an integral number, which representss the degree of polymerization of the polymer, from 0 to 22 in the present case). These two seriess correspond to non-functional (244 series, Fig. 3.6c) and mono-functional (288 series, Fig. 3.6b)) RAFT polymers, respectively (see Appendix A). It can be seen from Fig. 3.6 that the non-functionall polymer (no OH group) eluted earlier (2.17 min.) than the mono-functional polymer (one OHH group; 2.67 min). This confirms the base-line separation of polymers according to functionality ass seen in Fig. 3.1.

Thee mass spectra of the two peaks at elution times of 2.2 and 2.6 min (Appendix A Fig. Ap-5) are similarr to those in Fig.s 3a and Ap-3a. The only observed difference in these mass spectra is that nextt to the sodium-cation (Na+) series also a high-intensity potassium-cation (K+) series (with a masss difference of 16 amu) was observed in the on-line mass spectra. This is probably due to traces off potassium present in the mobile phase.

On-linee MS monitoring of elution profiles of individual oligomeric species can also facilitate a betterr understanding of LC separation mechanisms [16]. For example, thanks to the high mass accuracyy of mass spectrometry [18], on-line LC-ESI-QTOF-MS can be used to investigate the effectt of the molecular weights of the polymer on the elution behavior. Fig. 3.7 shows an example off mass chromatograms for various selected masses for sample V37A. It can be seen that the retentionn times of PMMA oligomers with different molecular weights were almost identical. However,, it also can be seen from Fig. 3.7 that the retention time of the polymer increased slightly withh increasing molecular weight (from 2.53 to 2.85 minutes for mono-functional PMMAs with n increasingg from 0 to 22). The retention time increased from 2.10 to 2.24 minutes for non-functional PMMAss with n between 0 and 20 (Appendix A Fig. Ap-6). This indicates a slight adsorption effect off PMMA, so that the LC conditions are not perfectly critical. We therefore speak of near-critical conditions.. The original RAFT agent (n =0) was also observed in the ESI-QTOF-MS spectra of the low-molecular-weightt RAFT polymers.

57 7

ChapterChapter 3

37AJ_C_POSTCOL1 1

100 0 2.17 7 (a) )

% % 2.67 7

0.000 1.00

37A_LC_POSTCOL1 1

100--

% %

2.000 3.00

2.67 7

Time e

0 0

4.000 5.00

(b)) 288 series

37A_LC_POSTCOL1 1

100n n

% %

2.17 7 (c)(c) 244 series

1.00 0 1.500 2.00 2.50 0 3.00 0 3.50 0 Time e

4.00 0

Fig.. 3.6 LC-ESI-QTOF-MS summed mass chromatogram of sample 37A. (a) Summed masses set at

244+100.1nn and 288+100.1n (n: 0-22); (b) 288 series; (c) 244 series (see explanation in text and

Appendixx A Fig. Ap-3). Mobile phase: 45% acetonitrile in dichloromethane, two home-packed

silicaa columns (150 mm x 1 mmi.d.; 3-urn particles; 100-A pore size), flow rateO.10 ml/min.

58 8


Fig.. 3.7 Representative (near-critical) LC x MS mass chromatograms for various selected masses of

mono-functionall PMMA from sample V37A. LC conditions as in Fig. 3.6.

3.44 Conclusions

MALDI-TOF-M SS and on-line LC-ESI-QTOF-MS were performed to analyze hydroxyl-functional PMMAA samples synthesized by RAFT polymerization. The end-groups of a non-functional fraction containingg the RAFT dithioester moiety [S=C(C6H5)S-] and the initiating-radical fragment [(CH3)2CCN-]] could be assigned directly from the ESI spectrum. However, the dithioester moiety wass lost in the MALDI experimental process. It was also observed to split off easily in collision inducedd dissociation (ESI-CID-MS ) experiments, which further verified the structure of the non-functionall fraction of the RAFT polymer. Similar results were obtained with ESI-MS for the mono-funtionall fraction of the RAFT polymer, which was shown to contain the leaving group of the RAFTT agent [HO(CH2)3CH3CCN-] and the RAFT dithioester moiety. The latter was not observed inn the MALDI-M S spectrum. These results provide direct evidence that the RAFT polymers still presentt latent living character through the presence of the dithioester moiety. This property can be usedd for further chain extension with addition of fresh monomer and fresh initiator. The structures off non-functional and mono-functional fractions of a derivatized RAFT polymer were also establishedd and discussed. All these results are in agreement with a contemporary interpretation of thee RAFT polymerization process and the derivatization reaction. The active (weakly bonded) end-groups,, such as the dithioester moiety, were lost in the MALDI-M S experiments, while the

59 9

ChapterChapter 3

polymericc molecules were observed intact in the ESI-MS spectra. This indicates that in the present casee ESI is a much softer ionization method than MALDI .

On-linee LC-ESI-QTOF-MS was used to provide further evidence for the structure of the end-groups andd to study the mechanism of the LC separation. Specifically the effect of the polymer molar mass onn the elution behavior was studied. The results confirmed the base-line separation of the functional prepolymerss according to the number of hydroxy 1 groups. They also revealed a slight adsorption effectt of PMMA under the experimental (near-critical) conditions.

Acknowledgments s

Thiss project is funded by the Dutch Polymer Institute (DPI project 205). We thank Aschwin van der Horstt and Fiona Fitzpatrick (University of Amsterdam) and Gabriëlle van Benthem and Prof. Rob vann der Linde (Eindhoven University of Technology) for their cooperation. One of the authors acknowledgess Dr. Xinghua Guo of AMOLF (Macromolecular Mass Spectrometry & Ion Physics Group,, FOM Institute for Atomic and Molecular Physics, Amsterdam) for helpful discussions. Helpfull discussions with Profs. A.M.Skvortsov and A.A.Gorbunov from St.Petersburg (Russia) weree made possible by INTAS (project nr. INTAS-OPEN-2000-0031).

References s

[I ]] K. Matyjaszewski, Ed., Controlled/Living Radical Polymerization, Progress in ATRP, NMP, and RAFT, ACS

Symp.. Ser. Vol.768, American Chemical Society: Washington, DC, 2000.

[2]] J. T. Lai, D. Filla, R. Shea, Macromolecules, 35 (2002) 6754.

[3]] D. Solomon, G. Waverty, E. Rizzardo, W. Hill , P. Cacioli, U.S. Patent 4,581,428, 1986.

[4]] K. Matyjaszewski, S. Coca, S. G. Gaynor, M. Wei, B. E. Woodworth, Macromolecules, 30 (1997) 7348.

[5]] J. Chiefari, Y. K. Chong, F. Ercole, J. Krstina, J. Jeffery, T. P. T. Le, R. T. A. Mayadunne, G. F. Meijs, C. L. Moad,

G.. Moad, E, Rizzardo, S. H. Thang, Macromolecules, 31 (1998) 5559.

[6]] J. Chiefari, R. T. A. Mayadunne, C, L. Moad, G. Moad, E. Rizzardo, A. Postma, M. A. Skidmore M.A.; S. H.

Thang,, Macromolecules, 36 (2003) 2273.

[7]] F. Ganachaud, M. J. Monteiro, R. G. Gilbert, M. R. Dourges, S. H. Thang, E. Rizzardo, Macromolecules, 33 (2000)

6738. .

[8]] P. Vana, L. Albertin, L. Barner, T. P. Davis, C. Barner-Kowollik, J. Polym. Sci. A, 40 (2002) 4032.

[9]] C. Ladavière, N. Dörr, J. P. Clavere, Macromolecules, 34 (2001) 5370.

[10]] M. Destarac, D. Charmot, X. Franck, S. Z. Zard, Macromol. Rapid Commun., 21 (2000) 1035.

[ I I ]] C. Schilli, M. G. Lanzendörfer, A. H. E. Muller, Macromolecules, 35 (2000) 6819.

[12]] J. H. Scrivens, A. T. Jackson, Int. J. Mass Spectrom., 200 (2000) 261. [13]] G. Montaudo, R. P. Lattimer , Eds. Mass Spectrometry of Polymers, CRC press: Boca Ratin, FL, USA, 2002,

pp.10,, 149-180,419-521.

[14]] S. D. Hanton, Chem. Rev., 101 (2001) 527.

[15]] C. N. McEwen, P. M. Peacock, Anal. Chem., 74 (2002), 2743.

[16]] R. Murgasova, D. M. Hercules, Anal. Bioanal. Chem., 373 (2002) 481.

[17]] P. J. Schoenmakers, C. G. de Koster, LC-GC Europe, 15 (2002)) 38.

60 0


[18]] M. W. F. Nielen, F. A. Buijtenhuijs, Anal. Chem., 71 (1999) 1809.

[19]] C. Keil, E. Esser, H. Pasch, Macromolecular Materials and Engineering, 286 (2001) 161.

[20]] J. J.Vosloo, D. de Wet-Roos, M. P. Tong, R. D. Sanderson, Macromolecules, 35 (2002) 4894.

[21]] F. D'Agosto, R. Hughes, M.-T Charreyre,C. Pichot, R. G. Gilbert, Macromolecules, 36 (2003) 621.

[22]] H. Pasch, A. Deffieux, R. Ghahary, M. Schapacher, L. RiqueLurbet, Macromolecules, 30 (1997) 98.

[23]] D. Musat, H. Henderickx, G. Kwakkenbos, R. van Benthem, C, G. de Koster, R. Fokkens, N. M. M. Nibbering, J. Am.. Soc. Mass Spectrom., 11 (2000), 218.

[24]] C. B. Jasieczek, D. M. Haddleton, A. J. Shooter, A. Buzy, K. R. Jennings, R. T. Callagher, Polym. Prepr., 37 (1996)845. .

[25]] E. Beyou, P. Chaumont, F. Chauvin, C. Devaux, N. Zydowicz, Macromolecules, 31 (1998) 6828.

[26]] M.-A.Dourges, B. Charleux, J.-P. Vairon, J.-C. Blais,G. Bolbach, J.-C. Tabet, Macromolecule, 32 (1999) 2495.

[27]] S. Coca, C. B. Jasieczek, K. L. Beers,K. Matyjaszewski, J. Polym. Sci. A, 36 (1998) 1417.

[28]] K. Matyjaszewski, Y. Nakagawa, C. B. Jasieczek, Macromolecules, 31 (1998) 1535.

[29]] K. Matyjaszewski, M. Jo, H. J. Psik, D. A. Shipp, Macromolecules, 32 (1999) 6431.

[30]C.. D. Borman, A. T. Jackson, A. Bunn, A. L. Cutter, D. J. Irvine, Polymer, 41 (2000) 6015.

[31]E.. L. Grognec, J. Claverie, R. Poli, J. Am. Chem. Soc, 123 (2001) 9513.

[32]] H. Nonaka, M. Ouchi, M. Kamigaito, M. Sawamoto, Macromolecules, 34 (2001) 2083.

[33]] H.-Q. Zhang, X.-L. Jiang, R. van der Linde, Polymer, 45 (2004) 1455.

[34]] X.-L. Jiang, L. Lima, P. J. Schoenmakers, J. Chromatogr. A, 1018 (2003) 19.

[35]] V. Lima, X.-L. Jiang, J. Brokken-Zijp, B. Klumperman, B. R. van der Linde, P. J. Schoenmakers, J. Polym. Sci. A, submitted. .

16]] J. H. Scrivens, A. T. Jackson,, H. T. Yates, M. R. Green, G. Critchley, J. Brown, R. H. Bateman, M. T. Bowers, J. Gidden.,, Int. J. Mass Spectrom. Ion Process, 165/166 (1997) 363.

37]] X.-L. Jiang, A. van der Horst, P. J. Schoenmakers, J. Chromatogr. A, 982 (2002) 55.

61 1

Appendixx A

Additionall ESI-MS spectra, assignment, mass chromatograms as well as explanatory texts are

providedd in this appendix.

V37A0 0

V37B0 0

V37A1 1

V37B1 1

5000 1500 2500 3500 4500

m/z z

Fig.. Ap-1 MALDI-TO F mass spectra for the first two fractions of sample PMMA V37A and V37B.

(V37A0)) Non-functional fraction of V37A collected at 1.7-2.2 min; (V37A1) Mono-

functionall fraction of V37A (2.3-3.3 min); (V37B0) Non-functional fraction of V37B (1.7-

2.22 min). (V37B1) Mono-functional fraction of V37B (2.3-3.3 min). LC separation

conditionss were the same as in figure 1, except a flow rate of 0.5 ml/min.


Thee polymer end groups can be calculated from a polymeric series in a mass spectrum using equationn 1

MMm m (1) ) ^peakk H xiK/ mon0mer " ^t end group "^ ^Wcounteric

wheree Mpeak is the molar mass value of the selected peak, Mmonomer the molar mass of the monomer,

nn the (integral) number of monomer repeat units, Afe„d group the total molar mass of chain end groups

(includingg the initiating group and the end-capping group), and Mcomterion the molar mass of the

counterionn attached in the ionization process.

Plott Afpeak vs. n should yield a straight line. The slope of the line represents the value of MmommeT.

Thee intercept of the line is the sum of Mendgroup and Afcounten0n. Figure SI-1 shows an example of the

calculationn of the total end-group mass (Meni group). Al l the calculation results and structures are

summarizedd in table 2.

00 2 4 6 8 10 12 14 16 18 20 22 24 26

Numberr of monomeric units

Fig.. Ap-2 Representative calculation of sum of the two end-groups and the counterion

inn sample V37A1 from MALDI-TOF-MS and ESI-QTOF-MS data.

Fig.. Ap-4a shows an expanded portion of the ESI spectrum of V37A0 in the range of 930 to 1050

amu.. It can be clearly seen from Fig. Ap-4a that singly charged ions exist next to doubly charged

ionss based on the isotopic peaks separated by 0.5 amu.

Fig.. Ap-4b displays the MS2 spectrum for the m/z 1044 (« = 8) molecular ion peak. The

fragmentationn scheme is shown in Fig. Ap-4c.

63 3

ChapterChapter 3

(a)MALDI(V37Al) )

1637.07 7

1737.20 0 1837.22 2

ii i i i i i i i i i i i i i i i i i 1 1 i i i i i i i i i i i i i i i i i i • i ' ' ' ' i ' ' i '

14000 1450 1500 1550 1600 1650 1700 1750 1800 1850 1900

100 0 1488.74 4 m/zm/z (b)ESI(V37Al)

1406.71 1

1488.69 9

1456.75 5

1407.72 2

1408.23 3

1489.70 0

1588.78 8

1506.77 7 1689.84 4

1557.311 1589.77 1688.83. :.

1507.82 2

1508.33 3

1657.40 0

1656.78 8

* *

1789.90 0

1788.88 8

1707.98 8

L i t d h * ^ ^

,1 8 0 7-455 1889.97

1857.51 1

1808.466 |

iMim iMim „ I M M B ^ H -- Mij I I . M 1 m/z 14000 1450 1500 1550 1600 1650 1700 1750 1800 1850 1900

Fig.. Ap-3 Comparison of MALDI and ESI mass spectra for mono-functional RAFT

polymerr (V37A1). (a) MALDI-TOF-MS; (b) ESI-QTOF-MS.

64 4


100 0 944.40 0

% %

(a)) ESI (V37A0) )

934.45 5

^ V T - V V

1044.46 6

945.41 1

946.42 2

'W*fcmlUl.wnkW W j i . 1 * *

1045.46 6

1046.48 8

984.48 8

9400 960 980 1000 10200 1040 m/z z

100-, , 291.22 2 (b)) ESI-MS2 (V37A0)

191.15 5

890.63 3

890.S S

292.23 3

391.30 0 392.29 9

891.66 6

892.64 4

893.777 1044.69

. .. ( m/z z

(+Na+)) (+Na+) 1911 291

CH H

2000 400 600 800 1000 1200 1400

(+Na+) ) 890890 w

CH33 | CH3 CH2^ H

CH2-C—^CH2—C—)5CH2—C—rr S-

c=oo c=o c=o

133 C H3

CH33 - C CH2—C—

CNN C = 0

OCH33 OCH3 OCH, OCH3

Fig.. Ap-4 ESI-QTOF mass spectra of non-functional RAFT polymer (V37A0). (a) Enlarged part of ESI-QTOF

MS;; (b) ESI-MS2 selected at 1044 amu; (c) the proposed fragmentation of the 8-mer in ESI-CID-MS2.

65 5

ChapterChapter 3

Tablee Ap-1 Assignments of the peaks in the ESI-QTOF-MS spectrum of V37B1 (Fig. 3.6a).

Entry y

1 1

2 2

3 3

4 4

5 5

6 6

7 7

8 8

9 9

10 0

11 1

12 2

13 3

n n

8 8

9 9

18 8

19 9

27 7

28 8

29 9

8 8

17 7

18 8

26 6

27 7

28 8

= = 1 1

1 1

2 2

2 2

3 3

3 3

3 3

1 1

2 2

2 2

3 3

3 3

3 3

m/zm/z (calculated )

936.49 9

1036.55 5

980.00 0

1030.03 3

961.16 6

994.51 1

1027.86 6

968.47 7

945.96 6

995.99 9

938.46 6

971.81 1

1005.16 6

m/zm/z (experimental)

936.47 7

1036.55 5

979.99 9

1030.04 4

961.16 6

994.52 2

1027.89 9

968.49 9

946.00 0

996.04 4

938.49 9

971.83 3

1005.22 2

Structuree symbol

B1H H

BIS S

Thee calculated value of B1H: m/z = [112.08 (HO(CH2),CH,CCN-) + n * 100.06 (PMMA

backbone)) + 1.01 (-H) + z x 22.99 (Na)r*]/z, where n refers to the degree of polymerization, and m

too the number of charges (Na) on the polymers.

Thee calculated value of BIS: m/z = [112.08 (HO(CH2),CHiCCN-) + „ x 100.06 (PMMA

backbone)) + 32.98 (-SH) + z x 22.99 (Na)""]/z, where n refers to the degree of polymerization, andd m to the number of charges (Na') on the polymers

Fig.. Ap-5 shows ESI-MS spectra of the non-functional peak eluted at 2.2 min (a) and mono-functionall peak eluted at 2.6 min (b) for sample V37A. Two different series were observed with a repeatingg unit of 100.1 Da (PMMA unit). They result from the counter-ions, sodium cation (Na+) andd potassium cation (K+) respectively, with mass difference of 16 amu. The calculated mass of the endd groups (MetKi gr0up= 221 amu) at an elution time of 2.2 min corresponds to polymeric structures withh (CH3)2CCN as one end group and SCSC^Hs as the other end group (Fig. Ap-5a). The calculatedd mass of the end groups (A/Cnd group 265 amu) at an elution time of 2.6 min corresponds to structuress with HO(CH2)3CH3CCN as one end group and SCSC6H5 as the other end group (Fig. Ap-5b).. These confirm the results obtained in off-line ESI-QTOF-MS and are in agreement with our understandingg of the RAFT polymerization process. On this basis, the summed chromatograms selectedd masses of 244+100.06« and 288+100.06/7 are shown in Fig. 3.6.

66 6


3 7 A _ L C _ P O S T C O L 11 61 (2.203) l n nn 744.3203

(A)) 100

644.2567 7

544.2175 5

444.1572 2

l l l u u

944.4395 5

1044.4894 4

TOFF MS ES + 359 9

m m

1144.5431 1

1244.6063 3

1245.6201 1

1444.7405 5

1445.7393 3

mm mm MÉÉ É QQ l l I I i l i lHIH H i l l m U M M M J — — I M ^ — — W ^ ^ — p ^ P — P g fftlZ

4000 600 800 1000 1200 1400 1600

IB ) )

37A_ _

100 0

L C _ P 0 S T C 0 L 11 73 (2.635) 688.28311 804.3303

% %

TOFF MS ES + 310 0

588.2396 6

504.1628 8

488.1905 5

888.3942 2

904.3944 4 988.4592 2

1104.5181 1

oo Lull]i l l l l ^ ^

1188.5891 1

1288.6332 2

1404.6772 2

1405.6771 1 1605.8210 0

4000 600 800 1000 1200 1400 1600 m/z z

Fig.. Ap-5 ESI-MS spectra of the non-functional peak eluted at 2.2 min (a) and mono-

functionall peak eluted at 2.6 min (b) from sample V37A.

67 7

ChapterChapter 3

Fig.. Ap-6 illustrates the effect of the molecular mass on retention time for non-functional PMMAs. Thee selected mass-to-charge ratio is shown in the second line of the top-right corner. The value shownn in the curve is the retention time.

37A_LC_POSTCOL1 1

100 0 2.24 4

T T

TOFF MS ES+ 2246 6

350 0

37A_LC_POSTCOL1 1

100 0

% % 0 0

37A_LC_POSTCOL1 1

100 0

%% i-

224 4 TOFF MS ES+

1945 5 532 2

2.20 0 TOFF MS ES+

1645 5 959 9

37A_LC_POSTCOL1 1

100 0

% %

2.20 0 TOFF MS ES+

1345 5 1.77e3 3

37A_LC_POSTCOL1 1

100 0

% %

2.20 0 TOFF MS ES+

1145 5 2.60e3 3

0 0 37A_LC_POSTCOL1 1

100 0 2.17 7 TOFF MS ES+

944 4 4.29e3 3

37A_LC_POSTCOL1 1

100 0

% %

2.13 3 TOFF MS ES+

644 4 6.80e3 3

37A_LC__POSTCOL1 1

100 0

% %

2.10 0 TOFF MS ES+

344 4 2.41e3 3

0 jj _ _

37A_LC_POSTCOL1 1

100 0 2.10 0

% % 0 0 0.00 0

TOFF MS ES+ 244 4 665 5

1.00 0 2.00 0 3.00 0 4.00 0 Time e

5.00 0

Fig.. Ap-6 Representative (near-critical) IX x MS mass chromatograms for various

selectedd masses of non-functional PMMA from sample V37A. Mobile phase:

45%% acetonitrile in dichloromethane, two home-packed silica columns (150

mmm x 1 mm i.d.; 3-um particles; 100-A pore size), flow rate 0.10 ml/min.

68 8

UvA-DARE (Digital Academic Repository) Separation and ... · MassMass Spectrometry Characterization...

Documents

Transcript of UvA-DARE (Digital Academic Repository) Separation and ... · MassMass Spectrometry Characterization...