ISSN 15600904, Polymer Science, Ser. B, 2013, Vol. 55, Nos. 3–4, pp. 116–121. © Pleiades Publishing, Ltd., 2013. Original Russian Text © A.E. Tarasov, O.M. Ol’khova, Ya.I. Estrin, G.V. Lagodzinskaya, E.R. Badamshina, Yu.M. Mikhailov, 2013, published in Russian in Vysokomolekulyarnye Soedineniya, Ser. B, 2013, Vol. 55, No. 3, pp. 330–335.

POLYMERIZATION

Chain Termination in Polymerization of Substituted Oxetanes in the Presence of Boron Trifluoride Etherate A. E. Tarasov*, O. M. Ol’khova, Ya. I. Estrin, G. V. Lagodzinskaya, E. R. Badamshina, and Yu. M. Mikhailov Institute of Problems of Chemical Physics, Russian Academy of Sciences, pr. Akademika Semenova 1, Chernogolovka, Moscow oblast, 142432 Russia *email: [email protected] Received July 2, 2012; Revised Manuscript Received September 17, 2012

Abstract—It has been shown that the polymerization of oxetanes with azidomethyl substituents initiated by boron trifluoride etherate in the absence and the presence of ethylene glycol proceeds via chain termination with fluorine atom transfer. This reaction results in the formation of a polymer that is monofunctional with respect to hydroxyl groups and contains a fluorine atom at one of the chain ends. With the use of 19F NMR spectroscopy, the numberaverage functionality of polymer with respect to fluorine atoms was studied. The methods of suppressing the aforementioned reaction, whose intensity decreases during a decrease in the poly merization temperature and an increase in the ethylene glycol concentration, were considered. In the absence of ethylene glycol, the chain termination with fluorine atom transfer is the main reaction of chainpropaga tion restriction. DOI: 10.1134/S1560090413030056

Oligooxetanediols based on 3,3bis(azidome thyl)oxetane (BAMO) and 3azidomethyl3methy loxetane (AMMO) can be used as precursors for the synthesis of polyurethane thermoelastoplastics. To obtain the optimal molecular mass in polyaddition reactions, oligomers of oxetanes with azidomethyl substituents of a given molecular mass and a function ality with respect to hydroxyl groups close to two must be used. The synthesis of oligooxetanediols meeting these requirements calls for understanding the pat terns of development of the molecular mass and molecularmass distribution as well as the mechanism of end functional groups formation during cationic polymerization of azidocontaining oxetanes. In spite of a large body of research on the synthesis of polymers and copolymers of substituted oxetanes [1–3], the mechanism of their polymerization is poorly known. As was shown previously [4], ethylene glycol acts as a chaintransfer agent; thus, its concentration deter mines the molecular masses of the forming oligomers. At the same time, it was suggested that other reactions of chain termination occur during AMMO polymer ization with boron trifluoride etherate (BTE) in the presence of ethylene glycol. However, this was not cor roborated in [4]. In [4], it was suggested that, in the course of poly merization of oxetanes with azidomethyl substituents initiated by BTE, the chain termination with fluorine

atom transfer can occur via a mechanism similar to that observed for epoxy compounds [5, 6]. O+ kterm

CH2 CH2 O F

CH2 CH2

n

CH2 CH2 O

O BF3

−

CH2 CH2 O BF2 n+1

The aim of this study was to establish the presence of the analogous chaintermination reaction in the case of polymerization of oxetanes with azidomethyl substituents as well as to determine the intensity of this reaction and to find the methods of its suppression. EXPERIMENTAL AMMO and BAMO monomers were synthesized at the Institute of Organic Chemistry, Russian Acad emy of Sciences (Moscow). The monomer purity was evaluated via reversedphase chromatography, IR spectroscopy, and elemental analysis. The impurity contents in AMMO and BAMO were ~2 and <1%, respectively. 1,2Dichloroethane (ZAO Ekos1) and THF (KhimMed) used as solvents, as well as ethylene glycol (LabTekh), were purified through common proce dures [7, 8]. BTE (Acros Organics) was used without further purification.

116

CHAIN TERMINATION IN POLYMERIZATION OF SUBSTITUTED OXETANES

The azidocontaining oxetanes were polymerized in dichloroethane under the action of BTE in the pres ence of ethylene glycol in a thermally controlled glass reactor equipped with a stirrer. The monomer, ethyl ene glycol, and 80% solvent were placed into the reac tor, the reagents were stirred and kept at 8°С for 10 min, and a catalyst solution in the residual solvent was added. In all experiments, the initial concentra tions of AMMO and BAMO were as high as 4 and 3.5 mol/L, respectively. In the series of experiments with the ethylene glycol concentration changing from 0.2 to 1.0 mol/L, the BTE concentration was as high as 0.2 mol/L, and in the series of experiments with the BTE concentration varying from 0.1 to 0.3 mol/L, the concentration of ethylene glycol was 0.2 mol/L. The model samples were polymerized at a catalyst concentration comparable with that of a monomer and in the absence of ethylene glycol for the purpose of increasing the intensity of chain termination and, hence, the numberaverage functionality with respect to fluorine for the analyzed polymer. The pristine con centrations of AMMO and BAMO were 0.52 and 0.45 mol/L, the BTE concentration was 0.13 mol/L, and the polymerization temperature was 8°С. The polymerization procedure was identical to that described above for the experimental series with changing concentrations of ethylene glycol and BTE. The model samples were synthesized at 30°С not only at a catalyst concentration comparable to the monomer concentration but also in the absence of ethylene glycol. In BAMO polymerization, [BAMO] = 0.53 mol/L and [BTE] = 0.089 mol/L, and in AMMO polymerization, [AMMO] = 0.55 mol/L and [BTE] = 0.088 mol/L. NMR spectroscopy measurements were performed on a Bruker Avance III 500 MHz spectrometer (oper ating frequencies of 470 MHz (19F) and 500 MHz (1H), with tetramethylsilane as an internal standard for 1H). For analysis, a polymer solution in CDCl3 (~10 wt %) was prepared. The GPC study of oligomers was conducted on a Waters 200 chromatograph equipped with three Styra gel columns (pore sizes of 200, 500 and 1000 Å) and a refractometric detector. THF was used as an eluent at an elution rate of 1.2 mL/min and a column tempera ture of 25°С. In analyses, 0.2–0.3 wt % polymer solu tions were used (an inputloop volume of 2 mL). The MMD parameters of oligooxetanediols were deter mined with the use of a universal calibration curve in coordinates of elution time and calculated van der Waals volume of samples, including the reference samples [9, 10]. GPC curves of polymers were obtained on a Waters GPCV 2000 chromatograph equipped with two MIXEDC PLgel 5 µm Styragel columns. A refractometer, a viscometer, and a Wyatt Dawn Heleos II lightscattering gage were used as detec tors. THF was used as the eluent at an elution rate of 1.0 mL/min, a column temperature of 35°С, and a POLYMER SCIENCE

Series B

Vol. 55

Nos. 3–4

117

polymersolution concentration ranging from 10 to 15 g/L. The application of the lightscattering gage allows determination of the absolute values of poly mer MMD parameters. Reversedphase chromatograms of polymers were recorded on a Waters 2414 chromatograph equipped with a Diasphere110C18 column. A Waters 2414 refractometer and a PDA 496 spectrophotometer were employed as detectors with a methanol–water equal volume mixture used as an eluent at an elution rate of 0.7 mL/min and a column temperature of 35°С. For analysis, monomer solutions with concentrations ranging from 10 to 15 g/L were prepared. RESULTS AND DISCUSSION As was shown previously [4], during polymeriza tion of AMMO, along with chain termination, other reactions of chainpropagation restriction should occur. We believe that the polymerization of azido containing oxetanes can be accompanied by chain ter mination via a mechanism similar to that of epoxy polymerization in the presence of BTE. During poly merization of oxetanes, the chaintermination reac tion can occur via the following scheme. F F B O F

−

N3

+

O N3

O N3

n

F B O

O

O

F N3

N3

n

F N3

According to this mechanism, owing to hydrolysis of the –ОBF2 group during removal of traces of the catalyst, a polymer that is monofunctional with respect to hydroxyl groups and contains a fluorine atom at one of the chain ends must be formed. The 19F NMR spectrum displays a triplet with a chemical shift in the range –200 to –250 ppm relative to CFCl3 due to the fluorine atom located in the CH2 ⎯F group at the chain end with a spin–spin inter action constant of J ~ 50 Hz [11, 12]. As can be seen in Fig. 1, the spectrum of the BAMO oligomer exhibits a triplet with a shift of –236 ppm and J = 49 Hz. For AMMO, the triplet has a shift of –232 ppm and J = 48 Hz. The fact that the above triplet is due to the signal of the fluorine atom (which is located at the chain end and bound with a carbon atom), rather than to the cat alyst signal, is confirmed by its retention in the sample spectrum even after thorough washing of the polymer from catalyst traces with water and by the values of the 2013

118

TARASOV et al.

1

2 −146

−148

−150

−152

−235.5

−236.0

−236.5 δF, ppm

Fig. 1. 19F NMR spectra of the BAMO oligomer obtained at [BAMO] = 0.45 mol/L and [BTE] = 0.13 mol/L (1) prior to and (2) after the removal of catalyst traces.

1

2

4.4

4.3

4.2

δH, ppm

Fig. 2. Fragments of the 1H NMR spectra of (1) BAMO oligomers ([BAMO] = 0.45 mol/L, [BTE] = 0.13 mol/L) and (2) AMMO oligomers ([AMMO] = 0.52 mol/L, [BTE] = 0.13 mol/L).

chemical shift and spin–spininteraction constant typical of this type of bond. During the presence of the fluorine atom at the chain end, the 1H NMR spectrum should exhibit also a signal due to the СН2–F group as a doublet in the range of 4.4–4.5 ppm [13]. In fact, the 1H NMR spec trum of the BAMO oligomer displays a doublet in the region of 4.36 ppm with an interaction constant iden tical to that for the triplet in the 19F NMR spectrum (Fig. 2); for AMMO, the doublet has a shift of 4.26 ppm.

The application of 19F NMR spectroscopy, unlike NMR spectroscopy, to determine the quantitative fluorine content in a polymer requires the preliminary calibration of the spectrometer. The signal due to pro tons of the СН2–F group in the 1H NMR spectrum (4.36 for BAMO and 4.26 ppm for AMMO), which is distinct of the shifts of signals due to the rest of the protons (AMMO, δН: 0.95 (s, 3H, –CH3), 3.21 (q, 4H –CH2O–), 3.26 (s, 2H, –CH2N3) ppm; BAMO, δН): 3.33 (s, 4H, 4H –CH2O–), 3.34 (s, 4H, –CH2N3) ppm), allows determination of the equivalent mass

1H

POLYMER SCIENCE

Series B

Vol. 55

Nos. 3–4

2013

CHAIN TERMINATION IN POLYMERIZATION OF SUBSTITUTED OXETANES

119

Table 1. Fluorine contents in AMMO oligomers obtained at [AMMO] = 4.0 mol/L Experiment number

[EG], mol/L

[BTE], mol/L

Mn

Mw

fn

F, %

1

0.5

0.2

1300

1500

>0.01

~1.00

2

0.2

0.2

2100

2800

>0.01

~0.40

3

0.0

0.2

15600

42600

1.64

8.63

4

0.2

0.3

2500

3200

0.13

2.92

of a polymer molecule for each fluorine atom in the polymer. The numberaverage functionality of the AMMO oligomers with respect to fluorine, fn; the fraction of fluorine used in the chaintermination reaction rela tive to the total fluorine content in the catalyst; and MMD parameters Mn and Mw are given in Table 1. Because of weak signals due to protons of the СН2 ⎯F group in spectra 1 and 2 (Fig. 3), in Table 1, only the appraised values of fn and the fluorine fraction used in chain termination are given. However, it fol lows from the data that, in the absence of ethylene gly col, chain termination is the main reaction of chain propagation restriction, and in the presence of ethyl ene glycol, the intensity of this reaction decreases by

several orders of magnitude. On the basis of these data and with allowance for differences in the calculated and measured molecular masses of oligooxetanes pre pared in the absence of ethylene glycol, it may be con cluded that the presence of ethylene glycol results in suppression of chain termination and that the molec ular mass of the resulting oligomer is mainly deter mined by chain transfer to ethylene glycol. The numberaverage functionality of the BAMO oligomers with respect to fluorine, fn; the fraction of fluorine used in the chaintermination reaction rela tive to the total fluorine content in the catalyst; and MMD parameters Mn and Mw are given in Table 2. For BAMO, because of weak signals due to protons of the СН2–F group (Fig. 4), in Table 2 (experiments

d

d

a 4 3 2

c

b

1 4.3

4.2 δH, ppm

4 3 2

1 3.7

3.5

3.3

3.1

1.05

0.90 δH, ppm

Fig. 3. 1H NMR spectra of AMMO oligomers. The numbering of the spectra corresponds to the experiment numbers in Table 1. POLYMER SCIENCE

Series B

Vol. 55

Nos. 3–4

2013

120

TARASOV et al.

Table 2. Fluorine contents in BAMO oligomers obtained at [BAMO] = 3.5 mol/L Experiment number

[EG], mol/L

[BTE], mol/L

1

0.5

0.2

800

2

0.2

0.2

3

0.0

4 5

Mn

fn

F, %

1000

>0.01

~1.16

2100

3700

>0.01

~0.47

0.2

57000

318000

0.74

1.21

0.2

0.3

2000

3800

>0.01

~0.31

0.2

0.1

1500

2700

>0.01

~1.29

1, 2, 4, 5), only appraised values of fn and the fluorine fraction used in chain termination are given. However, here, as in the case of AMMO, the conclusion can be made that, in the absence of ethylene glycol, chain ter mination is the main reaction of chainpropagation restriction. In the presence of ethylene glycol, this reaction becomes less intense, as is the case for AMMO, albeit to a lesser extent. Moreover, the change in BTE concentration has practically no effect on the intensity of the above reaction. The scheme of chain termination for BAMO is similar to that for AMMO. It was shown that the intensity of reactions result ing in termination of the active center increases with

Mw

temperature, but the mechanism of this process has not been elucidated [14]. It may be suggested that this process is accomplished through the abovementioned scheme, because, during polymerization of BAMO model samples, the numberaverage functionality increases from 0.14 to 0.65 as temperatures changes from 8 to 30°С. In the case of polymerization of AMMO, the functionality with respect to fluorine increased from 0.30 to 0.40 during the same tempera ture rise. During polymerization of BAMO and AMMO in the absence of ethylene glycol, the isothermal charac ter of the process was not attained. This fact is corrob orated by the numberaverage functionality with

c 5 4 3 2 1

4.40

4.32

b a

δH, ppm 5

4 3

2

1 3.6

3.4

3.2 δH, ppm

Fig. 4. 1H NMR spectra of BAMO oligomers. The numbering of the spectra corresponds to the experiment numbers in Table 2. POLYMER SCIENCE

Series B

Vol. 55

Nos. 3–4

2013

CHAIN TERMINATION IN POLYMERIZATION OF SUBSTITUTED OXETANES

respect to fluorine for the samples, which is much higher than the values obtained during polymerization in the presence of ethylene glycol. In the case of AMMO polymerization in particular, the reaction mass heats up to the boiling temperature of the sol vent, despite the cooling of the reactor with a flow of water at 8°С. ACKNOWLEDGMENTS The authors are grateful to D.V. Vinogradov (Insti tute of Organic Chemistry, RAS) for kindly providing the monomers, to E.O. Perepelitsyna for assistance with GPC measurements of polymers, and to G.A. Estrina for reversedphasechromatography measurements (Institute of Problems of Chemical Physics, RAS). REFERENCES 1. Chi Zhang, Jie Li, and Yunjun Luo, Propellants, Explosives, Pyrotechnics 37, 235 (2012). 2. T. Saton, M. Tamaki, T. Taguchi, H. Misaka, Nguyen To Hoai, R. Sakai, and T. Kakuchi, J. Polym. Sci., Part A: Polym. Chem. 49, 2353 (2011). 3. Y. Shintani and H. Tsutsumi, J. Power Sources, 195, 2863 (2010).

POLYMER SCIENCE

Series B

Vol. 55

Nos. 3–4

121

4. A. E. Tarasov, Ya. I. Estrin, O. M. Ol’khova, V. P. Lody gina, and E. R. Badamshina, Polymer Science, Ser. B 52, 144 (2010). 5. P. H. Plesch, The Chemistry of Cationic Polymerization (Pergamon, New York, 1963; Mir, Moscow, 1966). 6. S. G. Entelis, G. V. Korovina, and A. I. Kuzaev, Vysokomol. Soedin., Ser. A 13, 1438 (1971). 7. A. Weissberger, E. Proskauer, J. Riddick, and E. Toops, Organic Solvents. Physical Properties and Methods of Purification (Wiley, New York, 1955; Inostrannaya Lit eratura, Moscow, 1958). 8. Laboratorni technika organicke chemie, Ed. by B. Keil (Prague, 1963; Mir, Moscow, 1966). 9. A. I. Kuzaev, in Heterophase Network Polymers: Synthe sis, Characterization and Properties (Taylor and Francis, London, 2002), p. 27. 10. A. I. Kuzaev, Vysokomol. Soedin., Ser. A 22, 1146 (1980). 11. A. J. Gordon and R. A. Ford, The Chemist’s Companion. A Handbook of Practical Data, Techniques and Refer ences (Wiley, New York, 1972; Mir, Moscow, 1976). 12. H. Gunther, NMR Spectroscopy. An Introduction (Wiley, Chichester, 1980; Mir, Moscow, 1984). 13. http://willson.cm.utexas.edu/Teaching/Chem318N/ Files/exam1_sp05.pdf (application date: 11.11.2011). 14. P. Dreyfuss and M. P. Dreyfuss, Polym. J. (Tokyo) 8, 81 (1976).

2013