Introduction

Polyisobutylenes (PIBs) are nonpolar hydrocarbon polymers produced only by cationic polymerization of isobutylene [1]. Owing to their hydrophobicity, excellent chemical resistance, and biocompatibility, PIBs find applications in many fields including coatings, lubricants, sealants [1], as well as biomaterials for various biomedical applications [2]. Low molecular weight PIBs are also frequently used as building blocks for preparation of various amphiphilic block copolymers and co-networks. The characterization of molecular weights and end-groups of these blocks are essential for the synthesis of polymeric materials with well-defined structure and architecture [2–6]. Mass spectrometry-based analyses, especially those based on soft ionization technique such as matrix-assisted laser desorption ionization (MALDI-MS) [7, 8] and electrospray ionization mass spectrometry (ESI-MS) [9], have become very popular for the determination of molecular weight and end-groups of different polymer classes [10, 11]. However, one of main drawback of these mass spectrometric approaches is that only those polymers that have repeat units and/or functional groups capable of forming adduct ions with certain ions (e.g., H+, Na+, K+, Cl–) or through deprotonation can be analyzed. Nonpolar hydrocarbon polymers with saturated bonds and with no heteroatom are not readily amenable to MALDI-MS and ESI-MS. The development of new ion-sources such as atmospheric pressure photoionization (APPI) [12, 13] and direct analysis in real-time (DART) [14] has paved the way for the analysis of less polar compounds, including nonpolar hydrocarbon polymers. In previous reports, we have shown that APPI-MS is an appropriate method for the characterization of low molecular weight PIBs [15, 16] and other nonpolar polymers [17] through the formation of [M + Cl]– adduct ions. In these experiments, the chloride ions were produced “in situ” from chlorinated solvents. In addition, we have recently demonstrated the capability of DART-MS for the analysis of low molecular weight PIBs [18].

Although DART-MS and especially APPI-MS are well suited for the characterization of PIBs and other nonpolar polymers, these ion-sources are not yet common and thus may not be available for scientists interested in using MS to investigate such nonpolar polymers. Moreover, ESI-sources are available in most mass spectrometric laboratories but they are mainly applied for polar compounds. Based on literature reports, to make the highly nonpolar PIBs amenable to ESI-MS, the presence of any polar group capable of forming adduct ions or deprotonation is necessary [19–21]. Althugh the above-mentioned reports represent important steps towards the successful characterization of highly nonpolar PIB derivatives by ESI-MS, these approaches do not seem to be generally applicable. Here we present a more general approach for the characterization of low molecular weight PIBs by ESI-MS, which is based on the adduct ion formation of PIBs with nitrate and chloride ions. Since the anions (NO3 – and Cl–), as will be shown by means of Density Functional Theory (DFT) calculations, attach to the partially positively charged hydrogen atoms of PIBs, the end-group does not affect significantly the resulting ESI-MS spectra. Thus, using this method, highly nonpolar, low molecular weight PIBs with fully saturated bonds can be analyzed.

Experimental

Chemicals

2-Propanol (HPLC grade), methanol (HPLC grade), ethanol (HPLC grade), acetone (HPLC grade). and sodium nitrate were purchased from VWR International (Leuven, Belgium). Ammonium chloride and ammonium nitrate from Reanal (Budapest, Hungary) were used. Sodium chloride (ACS reagent), 2-butanol (ReagentPlus), n-hexanol (reagent grade), and n-octanol (ACS reagent) were received from Sigma-Aldrich (Taufkirchen, Germany). Dichloromethane (a.r. grade) was purchased from Molar Chemicals (Budapest, Hungary) and distillated before use.

The structures, the number-average molecular weights (Mn), and the polydispersities (Mw/Mn) of the polyisobutylene derivatives determined by gel permeation chromatography (GPC) and mass spectrometry (MS) are compiled in Table 1.

Table 1 The Structures of the Polyisobutylene Derivatives Investigated
Full size table

All of the polyisobutylenes were synthesized by living carbocationic polymerization. The synthetic procedures for the PIB samples can be found in the Supporting Information.

Sample Preparation

The polyisobutylene samples were dissolved in mixtures of acetone and dichloromethane in volume ratios (V/V) ranging from 0/100 to 90/10 V/V for PIB3, PIB4, and PIB5; 0/100 to 50/50 for PIB1 and PIB2 at a concentration of 0.1 mg/mL, respectively. In some experiments, instead of acetone, 2-propanol, ethanol, 2-butanol, n-hexanol, and n-octanol were used. In order to promote the formation of negatively charged adduct ions 20 μL of saturated solutions of NH4NO3 or NH4Cl in acetone/dichloromethane 50/50 V/V was added to the sample solutions of 1 mL.

Electrospray Quadrupole Time-of-Flight MS/MS (ESI-Q-TOF)

A MicroTOF-Q type Qq-TOF MS instrument (Bruker Daltonik, Bremen, Germany) was used for the MS and MS/MS measurements. The instrument was equipped with an electrospray ion source where the spray voltage was 4 kV. N2 was utilized as drying gas. The drying temperature was 200°C and the flow rate was 4.0 L/min. For the MS/MS experiments, nitrogen was used as the collision gas. The pressure in the collision cell was determined to be 1.2 × 10–2 mbar. The precursor ions for MS/MS were selected with an isolation width of 5 m/z. The mass spectra were recorded by means of a digitizer at a sampling rate of 2 GHz. The mass spectra were calibrated externally using the exact masses of clusters [(NaTFA)n+TFA]– generated from the electrosprayed solution of sodium trifluoroacetate (NaTFA). The spectra were evaluated with the DataAnalysis 3.4 software from Bruker. The sample solutions were introduced either directly into the ESI source with a syringe pump (Cole-Parmer Ins. Co., Vernon Hills, IL, USA) at a flow rate of 10 μL/min or by using a bypass loop for injecting 20 μL sample solution injected into the carrier solvent flow at a flow rate of 100 μL/min (Flow Injection Analysis).

Gel Permeation Chromatography (GPC)

Gel permeation chromatograms were recorded in THF at a flow rate of 0.5 mL/min at 35°C with a Waters chromatograph equipped with four gel columns (4.6 × 300 mm, 4.6 × μm Styragel columns: HR 0.5, 1, 2, and 4), a Waters Alliance 2695 HPLC pump, with a Waters 2414 refractive index detector. To obtain the number-average (Mn) and weight-average molecular weights (Mw) the GPC was calibrated using polystyrene standards of known molecular weights and narrow polydispersities.

Density Functional Theory Calculations

We performed DFT calculations with the B3LYP exchange-correlation functional [22], where 6-31G(d) were the standard split-valence basis sets [23]. Geometry optimizations were carried out in vacuo. Counterpoise calculations were also done to obtain basis set superposition error (BSSE) correction for the anion–ligand interaction. All calculations were done using the Gaussian 09 software package [24]. The lack of imaginary frequencies in vibrational spectral calculations was taken to verify that the calculated stationary points on the potential energy surfaces (PES) represented true minima. The reported relative energies in this work are derived from the calculated zero-point corrected electronic energies of the corresponding structures.

Results and Discussion

ESI-MS of Polyisobutylene Derivatives Recorded in the Negative Ion Mode

As noted above, owing to the very nonpolar nature of PIBs and, thus, lack of effective ionization sites in PIBs capable of forming positively or negatively charged adducts with certain ions, these polymers are very difficult to analyze by ESI-MS. Indeed, in the positive ion mode of ESI-MS no formation of adduct ions were observed when the PIB samples were mixed with salts containing different cations such as Na+, K+, NH4 +, etc. However, in the negative ion mode of ESI-MS formations of [PIB + NO3]– and [PIB + Cl]–, adduct ions were seen when salts of nitrate or chloride were added to the PIB samples. Thus, negatively charged adduct ions of PIBs with different end-groups and/or initiator moieties (listed in Table 1) were detected, as it is shown for PIB1 (chlorine-terminated telechelic polyisobutylene containing aromatic initiator moiety) in Figure 1.

Figure 1
figure 1

ESI-MS spectrum of PIB1 ionized by NO3 – ions. The numbers at the top of the peaks stand for the total number of isobutylene units (n). The additional series of peaks denoted by asterisk is due to the PIB1 ionized by Cl– ions. The inset shows the measured and the calculated isotopic distributions for [PIB1 + NO3]– adduct ion with a total number of isobutylene units 13. The composition for this adduct ion is C64H120Cl2NO3

Full size image

As seen in Figure 1, the main series is due to the presence of [PIB1 + NO3]– adduct ions and the mass difference between the adjacent peaks is 56, matching the mass of an isobutylene unit. The highest detectable [PIB1 + NO3]– ion appeared at m/z ≈ 2030. In addition, as seen in the inset of Figure 1, the experimental isotopic distribution is also in good agreement with the calculated one. Furthermore, comparisons of the observed and calculated m/z values also support the presence of PIB oligomers with the compositions corresponding to PIB1 (Table 1, line 1). For example, the measured and the calculated m/z values for the [PIB1 + NO3]– adduct ion with a total number of isobutylene units of 13 are 1020.864 and 1020.865, respectively (the composition for this adduct ion is C64H120Cl2NO3). The additional low intensity series is due to the PIB1 ionized by the Cl– ions. The appearance of these low intensity series can be attributed most probably to the ubiquitous presence of Cl– ions in the solvents. It was intriguing to know whether the presence of polar group is essential for the ionization of PIBs. Thus, PIB with similar structure and molecular weight to PIB1 but with exo-olefin end-groups (PIB2) was further studied. It was found that PIB2 can easily be ionized by NO3 – and Cl– ions, too. The ESI-MS spectra of PIB2 recorded in the negative ion mode in the presence of NO3 – and Cl– ions can be found in the Supporting Information (Figure S1, Figure S2). Polyisobutylene with aromatic initiator moiety and hydroxyl end-groups (PIB3) also showed adduct ion formation with NO3 – and Cl– ions under ESI-MS conditions. The negative ion ESI-MS spectra of PIB3 in the acetone/dichloromethane (50/50 V/V) mixture ionized with NO3 – and Cl– ions are compiled in the Supporting information (Figure S3, Figure S4), whereas that of the PIB3 dissolved in ethanol recorded in the presence of Cl- ions is shown in Figure 2.

Figure 2
figure 2

ESI-MS spectrum of PIB3 ionized by Cl– ions. The inset shows the presence of [2M + NaCl2]– adduct ions. The numbers at the top of the peaks stand for the total number of isobutylene units (n). Solvent: ethanol

Full size image

It should be noted that there is no significant difference in the relative ESI-MS distribution when acetone/dichloromethane mixture or ethanol are used as solvents. The effect of the solvent polarity on absolute ESI-MS intensity will be discussed later. Moreover, when ethanol was used, low intensity but well-defined series of peaks appeared in the range of m/z 1800–2900 (Figure 2 inset). These peaks, according to the accurate mass measurement, are due to the formation of [2PIB3 + NaCl2]– ions as shown in Figure 2 inset. Albeit formations of dimers (and trimers, etc.) are frequently observed in ESI-MS, in the case of the PIBs it is unusual because of the weak interactions between the nonpolar polymer chains. Nevertheless, we could detect [2PIB3 + NaCl2]– adduct ions in the MS spectrum. Such kind of adduct ion consisting of highly nonpolar polymer and a complex NaCl2 - components, to the best of our knowledge, has not been described yet.

In the next series of experiments, in order to find out whether the aromatic initiator moiety is necessary for PIBs to be ionized by NO3 – ions, PIB4 with chlorine end-group and without aromatic initiator moiety was subjected to ESI-MS investigations. Figure 3 shows that PIB4 can indeed be ionized by NO3 – ions to obtain ESI-MS spectrum.

Figure 3
figure 3

ESI-MS spectrum of PIB4 ionized by NO3 – ions. The numbers at the top of the peaks stand for the total number of isobutylene units (n). The additional series of peaks denoted by asterisk and symbol “+” are due to the PIB4 ionized by Cl– ions and presence of [PIB4 – HCl + NO3]– adduct ions, respectively. The inset shows the measured and the calculated isotopic distributions for [PIB4 + NO3]– adduct ion with n = 18. The composition for this adduct ion is C72H145ClNO3

Full size image

As shown in Figure 3, the measured isotopic distributions are in line with the calculated ones, and the measured and the calculated m/z values are also in good agreement (e.g., the measured m/z for PIB4 with n = 18 is 1107.093 whereas the calculated for the composition of C72H145ClNO3 is 1107.092). The two additional low intensity series of peaks are due to the presence of [PIB4 + Cl]– and [PIB4 – HCl + NO3]– ions. This latter series is formed by elimination of HCl molecule from PIB4 either under ESI-MS conditions or during the work-up of the polymer sample. However, one of the most challenging issues in ESI-MS analysis is the soft ionization of highly nonpolar hydrocarbon polymers with fully saturated bonds and absence of any heteroatom.

As can be seen in Figure 4, ionization of such polymers can be achieved under ESI-MS conditions as the example of PIB5 shows. PIB5, methylated at both ends, is a fully saturated hydrocarbon polymer and does not contain any heteroatom but it can be effectively ionized by NO3 – ions.

Figure 4
figure 4

ESI-MS spectrum of PIB5 ionized by NO3 – ions. The numbers at the top of the peaks stand for the total number of isobutylene units (n). The additional series of peaks denoted by asterisk and symbol “+” are due to the PIB5 ionized by Cl– ions and presence of the chlorine terminated polyisobutylene ionized by NO3 – ions, respectively. The inset shows the measured and the calculated isotopic distributions for [PIB5 + NO3]– adduct ion with a total number of isobutylene units 15. The composition for this adduct ion is C61H124NO3

Full size image

Both the measured isotopic distribution and the m/z values are in good agreement with the theoretical ones [e.g., for the composition of C61H124NO3 (n = 15) the measured and the calculated masses are 918.957 and 918.959, respectively]. The two additional low intensity series of peaks are due to the PIB5 ionized by Cl– ions and the presence of the small fraction of chlorine terminated polyisobutylene ionized by NO3 – ions. This latter polymer series present in the sample as unreacted starting material from which PIB5 was synthesized by treating the chlorine-terminated PIB with Al(CH3)3.

Dependence of the ESI-MS Intensity on the Solvent Polarity and the Ion-Source Temperature

When pure dichloromethane, which is a good solvent for the low molecular weight PIBs investigated, was used for the ESI-MS experiments, very weak or no ESI-MS signals were obtained. Conversely, when mixtures of dichloromethane and the more polar acetone, 2-propanol, or ethanol were applied, appreciable negative ion ESI-MS signals appeared. Additional experiments were performed using mixtures of dichloromethane and alcohols with longer alkyl chains such as 2-butanol, n-hexanol, and n-octanol. However, no ESI-MS signals were obtained with n-hexanol and n-octanol, while PIBs were detected with low ESI-MS intensities in the case of 2-butanol only. It should be kept in mind, however, that very polar solvents cannot be used because the PIBs owing to their nonpolar character are insoluble in these polar solvents. To study how the ESI-MS intensity depends on the solvent composition, PIBs were dissolved dichloromethane-acetone mixtures in different compositions using constant NH4NO3 concentration. Figure 5 shows the variations of the ESI-MS intensities with the compositions of dichloromethane-acetone mixtures for the PIB3 and PIB4 samples ionized by NO3 – ions.

Figure 5
figure 5

The dependence of the ESI-MS intensities of the [PIB3 + NO3]– (a) and [PIB4 + NO3]– (b) adduct ions on the composition of the acetone/dichloromethane mixture

Full size image

As it turns out from Figure 5, in pure dichloromethane and at low acetone content (up to 10% V/V) no significant ESI-MS intensities are observed. Further increasing the amount of acetone in the solvent mixture, the ESI-MS intensities start to increase considerably for both PIB3 and PIB4. Moreover, in the case of PIB3, ESI-MS intensities continuously grow up to 90% V/V acetone content, while in the case of PIB4 above 60% V/V acetone content, a slight decrease in the ESI-MS intensity can be observed. The slight decrease in the ESI-MS intensity in the case of PIB4 above 60% V/V acetone content may be attributed to the lower solubility of PIB4. Thus, PIB4 can partly precipitate out in the droplet leaving less “free” PIB4 chains to “escape” into the gas phase after consecutive droplet fissions. The observed ESI-MS intensities versus acetone content curves are discussed in the Supporting Information.

The dependence of the ESI-MS intensity on the temperature of the ion-source was also investigated. Figure 6a shows the changes of the ESI-MS intensities with the ion-source temperatures for the [PIB3 + NO3]– ion formed from solvent mixtures of acetone/dichloromethane and 2-propanol/dichloromethane, whereas in Figure 6b the variations of the relative ESI-MS intensities with the ion-source temperatures are plotted.

Figure 6
figure 6

The dependence of the ESI-MS intensities of the [PIB3 + NO3]– adduct ions formed from acetone/dichloromethane (50/50 V/V) and 2-propanol/dichloromethane (50/50 V/V) solvent mixtures on the ion-source temperature (a) and variation of the ESI-MS intensities with the ion-source temperatures (b). Solvent: acetone/dichloromethane 50/50 V/V)

Full size image

As seen in Figure 6a, raising the ion-source temperature favors the formation of [PIB3 + NO3]– adduct ion (i.e., higher ESI-MS intensities are obtained at higher ion-source temperatures, although a slight decrease with the increasing temperature in the range of 100–150°C was observed for the 2-propanol/dichloromethane mixture. The fact that the ESI-MS intensities increase with the ion-source temperature may be attributed to the more effective desolvation process at higher temperatures. On the other hand, based on Figure 6a, it can also be surmised that PIB3 in 2-propanol/dichloromethane mixtures gives lower ESI-MS intensities at the same ion-source temperatures than in acetone/dichloromethane mixtures, especially in the range of 120–220°C. This finding may be due to the facts that 2-propanol has higher boiling point (82°C) compared with that of acetone (56°C) [25]; thus a lower boiling point allows for a more effective desolvation for the analyte at the same ion-source temperatures. On the other hand, it is likely that difference in the permittivity of the solvent mixtures does not play a significant role in the enhanced ESI-MS intensity obtained from acetone/dichloromethane with respect to 2-propanol/dichloromethane solution as the permittivity of the acetone (21) and 2-propanol (20) are very similar [25]. Furthermore, it is important and is evident from Figure 6b that the ion-source temperature does not affect the resulting ESI-MS intensity distribution in the range of 100–225°C.

Dependence of the Number-Average Molecular Weight on the Collision Cell Voltage

As discussed, NO3 – and Cl– ions can be attached to PIB chains under electrospray conditions to form the corresponding adduct ions. The negatively charged anions are most likely ligated to the partially positively charged hydrogen atoms of PIB chains as we have assumed based our earlier APPI-MS and DART-MS results obtained on related topics. Although in our recent ESI-MS investigations saturated solution of salts (thus, their concentrations were unknown) were used to facilitate the ionization of PIBs, it was found that PIBs with NO3 – ions appeared with higher intensity than PIBs with Cl– ions. Indeed, when NH4NO3 and NH4Cl were dissolved in methanol at known concentrations, mixed in equimolar ratio and added to the solutions of PIBs, the resulting ESI-MS spectrum revealed that [PIB + NO3]– adduct ions occurred with ca. 3–4 times higher intensity than [PIB + Cl]– adduct ions. This finding may highlight the fact that the surface concentration NO3 – ions is higher than that of the Cl– ions. The reason for this is most likely attributable to the higher surface activity of NO3 – ions compared with that of Cl– ions.

In order to obtain information on the energetics of [PIB + NO3]– and [PIB + Cl]– adduct ions, the collision energy was varied in the normal MS mode. Note that certain collision energy, typically 5–8 eV, is necessary to transfer the ions into the TOF part of the Q-TOF MS instrument. As seen in Figure 7a, by increasing the collision energy from 8 to 22 eV, considerable shift in the ESI-MS intensity distribution to higher masses can be observed.

Figure 7
figure 7

Shift of the ESI-MS intensity distribution with the collision energy for the [PIB4 + NO3]– adduct ions (a). Variation of the number-average molecular weights (Mn) calculated based on the corresponding ESI-MS intensities of the [PIB4 + NO3]– and [PIB4 + Cl]– adduct ions with the collision energy (b). Solvent: acetone/dichloromethane 50/50 V/V

Full size image

The shift to higher masses with the increasing collision energies can be ascribed to the dissociation of lower mass adduct ions into neutral polymers molecule and anions. The preferential dissociation of lower molecular weight PIB adduct ions to that of higher ones is most likely due to the so-called degrees of freedom (DOF) effect. According to this, at a given collision energy the internal energy shared for fragmentation of the dissociating bond is higher for the lower mass ions compared with that of the higher ones, bringing about that lower mass ions (i.e., with lower DOF value) dissociate in the time frame of MS experiments. In order to rule out that the shifts are not due to the effects such mass dependent scattering of the ions in the collision cell or any other mass discrimination issues, we selected polypropylene glycol (PPG) with similar molecular weight to that of the PIBs cationized by Li+ ions. Li+ ions form adduct with PPG with relative high activation energy for fragmentation [26], thus [PPG + Li]+ adduct ions do not fragment significantly either by losing cation or backbone cleavages in the collision energy range of 8–30 eV. In this experiment, we used the same parameter set as in the negative ion mode by reversing the polarity. According to this experiment, no shift in the ESI-MS intensity distribution was observed, indicating the absence of mass discrimination effect and supporting that the shifts of intensity distribution to higher masses are caused by the DOF effect. On the other hand, from the finding that the PIB adduct ions dissociate at relatively low collision energies, it can be concluded that the activation energy for fragmentation is relatively low. The shift of the ESI-MS intensity distribution as a function of the collision energy can also be modeled by using an Arrhenius type reaction rate constant. The details and the results of this approach can be found in the Supporting Information (Figure S5, Figure S6).

The other consequence of the intensity distribution shift to higher masses with the increasing collision energy is that the number-average molecular weights (Mn) calculated from the corresponding ESI-MS intensities as outlined in the legend to Figure 7 also increases. In Figure 7b, the variation of the Mn values with the collision energy for the [PIB4 + NO3]– and [PIB4 + Cl]– ions are plotted. As seen in Figure 7b, the apparent Mn values increase close to linearly with the collision energy for both [PIB4 + NO3]– and [PIB4 + Cl]– ion and the values of Mn at low collision energy (8–12 eV) are practically the same. However, the apparent Mn values for the [PIB4 + NO3]– increase more steeply than for the [PIB4 + Cl]– adduct ion. This finding indicates that the dissociation energy of the adduct ion into a neutral polymer molecule and an anion is lower for the [PIB4 + NO3]– than for the [PIB4 + Cl]– adduct ion. This observation was further supported by means of energy-dependent collision induced dissociation (CID) MS/MS experiments in which the [PIB4 + NO3]– and the [PIB4 + Cl]– adduct ions with the same number of repeat units were selected simultaneously using a relative wide mass window. It was found, in good agreement with the above mentioned finding, that the signal of the [PIB4 + NO3]– decreased to a greater extent with respect to that of the [PIB4 + Cl]– adduct ion. Similar experiences were obtained for the other PIB derivatives, too, namely dissociation of the [PIB + Cl]– requires higher energy than that of the [PIB + NO3]– ion. Figure representing the variation of the relative intensities of [PIB4 + NO3]- and [PIB4 + Cl]– is available in the Supporting Information (Figure S7).

At this point, it is worth comparing the data of Table 2, in which the Mn values of the PIBs determined by GPC and ESI-MS are summarized.

Table 2 The number-average molecular weights (Mn) and the polydispersities (Mw/Mn) of the polyisobutylene derivatives determined by gel permeation chromatography (GPC) and mass spectrometry (MS). The values of Mn and Mw were calculated based on the MS results as: Mn = ΣIiMi/ΣIi and Mw = ΣIiMi 2/ΣIiMi, where Ii and Mi are the relative intensity and the mass of the oligomer with a number of repeat units i, respectively. Mw is the weight-average molecular weight
Full size table

As seen in Table 2 for PIB3 and PIB4, GPC and ESI-MS yielded very similar Mn and Mw/Mn values. However, for the Mn of PIB5, ESI-MS gave higher values than GPC. The higher Mn value determined by ESI-MS can be ascribed to the dissociation of lower molecular weight adduct ions (lower DOF values) in the ion-source and/or during transmission through the collision cell giving rise to an increase in the Mn value. In addition, higher binding energies of longer chains to the anion can also bring about shifts of the Mn to higher values. Furthermore, it can also be concluded from the data of Table 2 that Mn of PIB1 and PIB2 determined by ESI-MS are considerably lower than those determined by GPC. Note that Mw/Mn–s of both PIB1 and PIB2 are higher than 1.3 so these samples represent relatively broad molecular weight distribution from the MS point of view. Thus, due to the mass discrimination effect, MS are less sensitive to the higher mass polymer fraction; hence, the presence of higher molecular weight fraction of the distribution is underestimated based on the ESI-MS intensities. Furthermore, the polydispersities calculated from the MS spectra are much lower than the values determined by GPC. The results show that ESI-MS is capable of estimating the real Mn values of PIBs with low molecular weight and narrow molecular weight distribution. However, the average molecular weights determined from the MS spectra may not be reliable for PIBs with higher molecular weights where solubility and mass discrimination problems can occur.

Density Functional Theory Calculations

DFT calculations were performed to gain insight into the interaction between PIB and small anions such as chloride and nitrate. For these calculations, PIB5 oligomers with different molecular weights including 240, 352, 464, and 576 g/mol corresponding to total numbers of isobutylene units 4, 6, 8, and 10, respectively, were selected. In the calculations, the different conformers were also taken into account. The relative energies for the conformers and the zero-point corrected electronic energies for the various adduct ions are summarized in the Table 3.

Table 3 ZPE corrected relative and zero-point corrected electronic energies for the investigated conformers of the PIB oligomers and adduct ions with anions NO3 – and Cl–. Basis set superposition error (BSSE) corrected energies are collected for the adduct ions
Full size table

According to the data of Table 3, the open straight chain conformation is more favored energetically than the gauche one. The chain-gauche conformation effect was observed for the PIB-n10 and PIB-n6 oligomers but this effect for the smaller PIB-n4 oligomer was negligible. Conversely, the relative energies for the conformers revealed opposite orders when they interacted with the anions because the gauche conformer builds a “cage” for the anions, resulting in a greater number of energetically favorable hydrogen atom-anion interactions.

We have also performed calculations for a third possible conformer of the PIB-n10 oligomer. In this conformation, the oligomer is half opened and there is a gauche region (n = 6) in the middle part of the oligomer. According to these calculations, although all the interactions between the isobutylene units and the anions are energetically preferred, these interactions, because of opposite trends in the relative energies of the oligomers and the adduct ions, are more favorable in the gauche conformation.

It is also evident from the data of Table 3 that there are no significant differences in the zero-point corrected electronic energies for the formations of adduct ions PIB-n6-chloride-gauche, PIB-n8-chloride-gauche, and PIB-n10-chloride-gauche (ca. 60 kJ/mol). Similar observation but lower zero-point corrected electronic energy values can be established for the adduct ions PIB-n6-nitrate-gauche, PIB-n8-nitrate-gauche, and PIB-n10-nitrate-gauche (ca. 47 kJ/mol). However, the zero-point corrected electronic free energies for the formations of adduct ions PIB-n4-chloride-gauche and PIB-n4-nitrate-gauche are lower than their larger size homologues (n6, n8, and n10) with values of ~15.0 kJ/mol and ~10.0 kJ/mol, respectively. Furthermore, PIB-n6 and PIB-n8, because of the similar alignment and the number of the hydrogen atom–anion interactions involved in the adduct ion formation, show a very similar result to that obtained for the half opened conformer of PIB-n10. Figure 8 shows the structures of conformers of PIB-n10 adduct ions with bonds H···Cl and H···O within bond lengths of 3 Å.

Figure 8
figure 8

The structures of the chain and the gauche conformers of PIB-n10-chloride and PIB-n10-nitrate adduct ions with bonds H···Cl and H···O within bond lengths of 3 Å

Full size image

In the PIB-n10-chloride adduct ion there are 10 H···Cl bonds available for the gauche conformer, whereas there are only 4 are for the chain conformer. Although in the PIB-n10-nitrate adduct ion the average H···O distances are ~30–40 pm shorter, the gauche conformer consists of a more complicated bond-system including 16 H···O bonds, while the nitrated chain conformer (PIB-n10-nitrate-chain) is similar to the chlorinated one (PIB-n10-chloride-chain).

On the other hand, the gauche conformer of PIB-n10 enwraps the anions fully. Thus, PIB-n10 may be regarded as a critical, lowest sized oligomer to obtain the highest stabilization effect for the adduct ion formation. Evidently, for the higher sized homologues, neither the zero-point corrected electronic energy for the adduct ion formation nor the “wrapping ability” of the PIB oligomer chains towards the anions will change. It is also important to note that the zero-point corrected electronic energies for adduct ion formations are lower with the chloride ion than with nitrate in every case by a value of ~10–15 kJ/mol. This finding is in line with the experimental results obtained by collision induced studies, namely, higher collision energies are needed for the dissociations of [PIB + Cl]– ions than those of the [PIB + NO3]– ions.

Conclusions

It was demonstrated with examples of PIB derivatives having different end-groups, the presence and absence of aromatic initiator moiety, that PIBs are capable of adduct ion formation with anions such as NO3 – and Cl– under electrospray conditions. This finding extends the capability of ESI-MS to the analysis of nonpolar polymers and makes this method valuable when other ion-sources that are more appropriate for the investigation of nonpolar compounds such as APPI-MS are not available. However, according to our experience, to obtain ESI-MS spectra of good quality in the negative mode for PIBs, solvent mixtures containing a polar solvent including acetone or 2-propanol should be applied. Another important conclusion of our work gained from the energy-dependent CID studies (in the normal MS and MS/MS modes) is that the dissociation energy for the [PIB + NO3]– and [PIB + Cl]– ions may be relatively low, and this finding was further supported by DFT calculations. Thus, low molecular weight fraction of the adduct ions may decompose during the transmission, while the higher molecular weight part can survive owing to DOF effect, causing an apparent increase in the value of Mn. Thus, it is advisable that in the Q-TOF instrument the collision voltage should be kept as low as possible. On the other hand, PIBs with higher polydispersities (Mw/Mn > 1.3) can also be analyzed by ESI-MS, but in that case the Mn value determined based on the corresponding ESI-MS intensities are lower than those obtained by GPC. Based on our energy-dependent CID MS and MS/MS investigations, it was also shown that bond dissociation energy for the [PIB + Cl]– is higher than for the [PIB + NO3]– ion being in line with the results of DFT calculations. Moreover, in spite of this fact, PIBs with NO3 – ion gave ESI-MS spectra of higher quality than with Cl– ion. Furthermore, the anion attachment technique presented for PIBs in this report may be applicable to other nonpolar polymers as well.