Abstract
Polybutadiene and hydroxyl terminated polybutadiene (HTPB) were epoxidized using in situ-generated dimethyl dioxirane (DMD) as an oxidant in the presence of nano-TiO2 at 25°C. Reaction time and different percentages of catalyst/Oxone® (w/w) were also examined. The capability of different kinds of double bonds to be epoxidized was studied in detail at various reaction times, and the products were characterized using nuclear magnetic resonance (1HNMR), 13CNMR, and Fourier transform infrared (FT-IR) techniques, with no side reaction being detected. The results indicate that using nano-TiO2 as a catalyst increases the epoxidation yield, especially of cis double bonds.
Introduction
Epoxidation has extensively been studied over the past few decades. The main reason for this interest in the process has been its capability in converting materials such as alkenes and polydienes to a variety of better industrial chemicals. Also, epoxides are one of the most important intermediate groups for the synthesis of oxygen-containing organic compounds both in industry and in academic laboratories. 1
There are several experimental methods for the epoxidation of C=C bonds in the structure of polymers such as polybutadiene and hydroxyl terminated polybutadiene (HTPB): thermal oxidation, 2 oxidation with Vo (acac)2 complex, 3 epoxidation with peroxides and hydro peroxides 4 such as tert-butyl hydro peroxide (TBHP), 5 and epoxidation with peracids 6 such as m-chloroperbenzoic acid (m-CPBA). 7 However, these methods are not without their disadvantages. For example, they bring about several side reactions and require high temperatures to obtain higher conversion rates. Another problem is that such methods degrade the polymer, thus resulting in smaller reaction yields. 8
These drawbacks can be avoided using in situ-generated dimethyl dioxirane (DMD) as the oxidizing agent. 9,10 In situ-generated DMDs are powerful, reactive, and efficient cyclic peroxides, usually obtained from the reaction between acetone and aqueous Oxone® (2KHSO5, KHSO4, and K2SO4) in buffered conditions. 11 Oxone performs simple oxidations and requires no more than readily available cheap reagents in ecofriendly conditions.
Using catalysts in the epoxidation of olefins, especially metal catalysts such as metal salts, oxides, and complexes, have become an important research area in organic synthesis over the last decade. 12 –18 The authors have previously epoxidized polybutadiene and HTPB using in situ-generated DMD with different catalysts: tetra-n-butyl ammonium bromide as a phase transfer catalyst (PTC), 19,20 MoO3, 21 transition metal salts, 22 complex copper (II) complex, 23 and cloisite 30B as a PTC. 24
Another catalyst of interest to researchers over the past few decades has been TiO2, because of the reactive oxygen atoms in it. This catalyzing agent is not only chemically important but also biologically very interesting. 25 Literature reveals that titanium superoxide species are very stable in aqueous as well as gel forms. 26 –28 The remarkable stability of titanium superoxide has spawned extensive studies on the formation and stabilization of reactive oxygen species in TiO2 during UV irradiation in the presence of molecular oxygen, 29 as a result of redox reactions involving the use of hydrogen peroxide and Ti3+ salts, 30 and on titanium-substituted molecular sieves. 31 –33 Most commercial samples of titanium are a mixture of rutile and anatase phases. Titanium in either phase can be used as an active catalyst in selectively epoxidizing many olefins. Among the titanium forms used as catalysts are titanium-substituted zeolites, 34,35 mesoporous molecular sieves, 35 titanium-substituted amorphous silica, 36,37 transition metal oxide-supported TiO2, 38 and titanium peroxy solutions. 39 More recently, indene and cyclooctene were epoxidized using nanocrystalline anatase titanium as a catalyst in the presence of H2O2. 40
This article is a report on the epoxidation of polybutadiene and HTPB using in situ-generated DMD in the presence of nano TiO2.
Experimental
Materials
The HTPB used in this work (average MN: 2840, containing 15% cis, 25% trans, 60% vinyl) was purchased from Chinese Zibo Qilu Chemicals. The polybutadienes used in this study, namely cis polybutadiene (average MW: 2–3 × 106, T g: −102°C, and cis content: 98%) and solid polybutadiene (average MW: 4.2 × 105, T g: −95° C, slab containing 36% cis, 55% trans, and 9% vinyl), were purchased from Aldrich, USA. Nano TiO2 (18 nm) (commercially available under the name of Degussa P-25) was obtained from Degussa Chemical, Germany. Oxone and other auxiliary chemicals were purchased from Merck, Germany and were used as received.
Epoxidation reactions
All reactions took place in a five-necked glassy reactor at 25°C ±0.1°C using the Cole Parmer Polystate temperature regulator (Model 12101-25). In a typical reaction, polymer (0.5 g) and CHCl3 (20 ml) were poured into the reactor and agitated until the polymer was completely dissolved. Then, acetone (10 ml) and 60 ml of potassium bicarbonate solution (5.68 g of KHCO3 in distilled water) were charged to the reactor. Subsequently, 60 ml of the Oxone solution (7.8 g of Oxone in distilled water) was mixed with different quantities of nano TiO2/Oxone (w/w). The mixture was then poured into the reactor and vigorously agitated until we obtained a complete dissolution giving us a clear diphase system. At specified intervals, 25 ml of the mixture was selected for analysis. For each run, the two phases were separated, and the organic phase was dried under vacuum at 55–60°C for 10 h. Finally, the samples were removed and their Fourier transform infrared (FT-IR) (in CHCl3), nuclear magnetic resonance (1HNMR) (in CDCl3), and 13CNMR (also in CDCl3) spectra were obtained.
Results and discussion
Epoxidation reactions
Epoxidation of polybutadiene and HTPB using in situ-generated DMD in the presence of nano TiO2 is shown in Figure 1. In this reaction, the buffer condition is generated by potassium bicarbonate. KHSO4/K2SO4 buffer holds the pH constant at 7–8.5, where the low concentration of H+ and OH– would not lead to the formation of an open ring in DMD or ring epoxy. The reaction between Oxone and acetone in the organic phase leads to the formation of DMD, which is responsible for the epoxidation reaction. The first step in this reaction is the nucleophilic attachment of the HSO5 – anion to the carbonyl group of acetone, followed by the formation of three-membered DMD. 41 It seems that nano TiO2 combines with DMD to form a DMD/Nano TiO2 complex, which at a later stage reacts with polybutadiene and HTPB to create epoxidized polybutadiene and HTPB, respectively (Figure 2).

Schematic representation of epoxidation reaction.

The mechanism of formation of DMD and DMD/Nano TiO2 complex. DMD: dimethyl dioxirane.
Characterization of the epoxidized polymers
NMR spectra
The chemical shifts (δ) of polybutadiene and HTPB and their epoxidized forms observed in the 1HNMR spectrum are given in Table 1. The findings are similar to the previously reported data. 7 In addition, the assignment of cis, trans, and vinyl epoxy groups in the epoxidized products was based on the data reported by Gemmer, 42 Zuchowska, 43 and Aguiar. 7
Assignment of chemical shifts (δ) of polybutadiene, HTPB, and their epoxidized forms in the 1HNMR spectra.
NMR: nuclear magnetic resonance; HTPB: hydroxyl terminated polybutadiene; VOH: OH–CH2–CH–(CH=CH2), HOH: HO–CH2–CH=CH–CH2–, GOH: OH–CH2–CH=C<.
According to the 1HNMR spectra, cis polybutadiene had two signals at 5.3 ppm and about 2.0 ppm, which were related to the cis double bonds and CH2 protons, respectively (Figure 3(a)). In comparison, epoxidized cis polybutadiene had two additional signals at about 2.9 ppm and about 1.6 ppm, which were related to cis epoxy groups and adjacent CH2 protons, respectively (Figure 3(b)).

1HNMR spectra of cis polybutadiene (a), epoxidized cis polybutadiene (b), solid polybutadiene (36% cis, 55% trans, and 9% vinyl) (c), epoxidized solid polybutadiene (d), HTPB (15% cis, 25% trans, and 60% vinyl) (e), and epoxidized HTPB (f) after 1 h of reaction. NMR: nuclear magnetic resonance; HTPB: hydroxyl terminated polybutadiene.
Similarly, solid polybutadiene (containing 36% cis, 55% trans, and 9% vinyl) underwent epoxidation. In this polybutadiene, the signals at 5.3 ppm and 5.4 ppm were pertained to cis and trans double bonds, respectively (Figure 3(c)). In contrast, epoxidized solid polybutadiene had two additional signals at about 2.9 ppm and 2.7 ppm, which were pertinent to cis epoxy and trans epoxy groups, respectively (Figure 3(d)).
Furthermore, the unepoxidized and epoxidized forms of HTPB (15% cis, 25% trans, and 60% vinyl) were similar to solid polybutadiene in terms of the chemical shift of cis and trans double bonds and their epoxides in the 1HNMR spectra (Figure 3(e) and (f)).
Of course, we did not observe vinyl epoxy in the 1HNMR, which was due to the lower reactivity of vinyl bonds in comparison to cis and trans double bonds. 42 This indicates that vinyl double bonds are rarely epoxidized.
The 13CNMR spectrum was used to characterize epoxy group carbons. The chemical shifts (δ) of polybutadiene and HTPB and their epoxidized forms observed in the 13CNMR spectrum are presented in Table 2. The 13CNMR spectra of cis polybutadiene and epoxidized cis polybutadiene are shown in (Figure 4(a) and (b)). The signal of cis double bonds appeared at 129 ppm. As a result of epoxidation, an additional signal was obtained at 56 ppm, which turned out to pertain to cis epoxy. In solid polybutadiene (36% cis, 55% trans, and 9% vinyl) (Figure 4(c) and (d)) and HTPB (Figure 4(e) and (f)), cis and trans double bonds had their signals at 128–134 ppm, but as the epoxidation reaction progressed, additional signals appeared at 56 ppm and 58 ppm, which were pertinent to cis and trans epoxy, respectively. These findings echoed the results obtained by Gemmer. 42

13CNMR spectra of cis polybutadiene (a), epoxidized cis polybutadiene (b), solid polybutadiene (36% cis, 55% trans, and 9% vinyl) (c), epoxidized solid polybutadiene (d), HTPB (15% cis, 25% trans, and 60% vinyl) (e), and epoxidized HTPB (f) after 1 h of reaction. NMR: nuclear magnetic resonance; HTPB: hydroxyl terminated polybutadiene.
Assignment of chemical shifts (δ) of polybutadiene, HTPB, and their epoxidized forms in the 13C NMR spectra.
NMR: nuclear magnetic resonance; HTPB: hydroxyl terminated polybutadiene; VOH: OH–CH2–CH–(CH=CH2); HOH: HO–CH2–CH=CH–CH2–; GOH: OH–CH2–CH=C<.
Focusing on the 115–120 ppm and 140–150 ppm regions in the 13CNMR spectra, we found that vinyl carbons remained unchanged due to their low electron density. This depicts that the vinyl microstructures reacted more slowly than did the cis and trans stereoisomers. The results agreed with past research. 42,43
In brief, by observing the 1HNMR spectra in the range of δ = 9–16 ppm, we found no signal in the region, a fact which means no side reactions such as the formation of H-CO of aldehydes or H-OCO of acids occurred. Likewise, the 13CNMR spectra of the epoxidized products showed no signal in the region of δ = 170–220 ppm. This suggests that carbonyl groups of aldehydes or of acids were not formed.
FT-IR spectra
Figure 5 displays the FT-IR spectra of cis polybutadiene, solid polybutadiene, HTPB, and the epoxidized forms of these polymers at different reaction times. As for the unepoxidized polymers, we observed a series of sharp signals at about 960, 911, and 725 cm− 1, which were due to trans, vinyl, and cis microstructures, respectively. 20,44

FT-IR spectra for cis polybutadiene, solid polybutadiene, HTPB, and their epoxides at different reaction times: 0 h (a), 1 h (b), 2 h (c), and 24 h (d). FT-IR: Fourier transform infrared; HTPB: hydroxyl terminated polybutadiene.
A great majority of cis and trans double bonds were epoxidized in comparison with only a small number of vinyl double bonds. It is worth noting that the broad signal observed at 3400–3600 cm−1 was related to the O-H bond of HTPB and remained intact in the epoxidation process. Later on, we focused on the expected symmetrical stretching and frequent breathing opening of the epoxy ring, which are usually observed at about 1,259–1,271 cm−1 and 813–834 cm−1, respectively.
To summarize, FT-IR spectral results depict the successful epoxidation of polybutadiene and HTPB. Moreover, no significant side reactions, such as ring opening and gelation, were seen in this epoxidation process, a fact confirmed by the lack of signals at the 1,000–1,100 cm−1 and 1,700–1,750 cm−1 regions in the FT-IR spectra.
Determination of epoxidation yields
The epoxidized polybutadienes and HTPB were characterized using 1HNMR. Putting signal area integrations at 2.7–3.0 ppm and at 4.9–5.5 ppm in the equation cited elsewhere,
7
we can calculate the double bond conversion as follows:
where A 2.7–3.0 and A 4.9–5.5 represent the integrated area under the signals at 2.7–3.0 ppm and at 4.9–5.5 ppm, respectively.
As mentioned earlier, cis polybutadiene, solid polybutadiene, and HTPB were epoxidized using in situ-generated DMD, with no or different quantities of nano TiO2 as a catalyst. The epoxidation yields of the three polymers at different reaction times are shown separately in Figures 6 to 8.

Epoxidation yields (in %) of cis polybutadiene in the presence of no, 1%, and 3% nano TiO2 as a catalyst at different reaction times.

Epoxidation yields (in %) of solid polybutadiene in the presence of no, 1%, and 3% nano TiO2 as a catalyst at different reaction times.

Epoxidation yields (in %) of HTPB in the presence of no, 1%, and 3% nano TiO2 as a catalyst at different reaction times. HTPB: hydroxyl terminated polybutadiene.
As it is clear in Figure 6, using nano TiO2 significantly increased the epoxidation yield of cis polybutadiene. More specifically, after 24 h of reaction time, the epoxidation yield was 31%, 67%, and 85% in the presence of no, 1%, and 3% nano TiO2, respectively.
It can be seen in Figure 7 that nano TiO2 did increase the epoxidation yield of solid polybutadiene but not so much as it increased the epoxidation yield of cis polybutadiene. To be more precise, after 24 h of reaction time, the epoxidation yield was 69%, 74%, and 78% in the presence of no, 1%, and 3% nano TiO2, respectively.
Figure 8 shows that the epoxidation yield of HTPB increased only minimally as a result of using nano TiO2. Specifically speaking, after 24 h of reaction time, the epoxidation yield was 29%, 30%, and 33% in the presence of no, 1%, and 3% nano TiO2, respectively. This can be attributed to the high percentage of vinyl double bonds in this polymer (15% cis, 25% trans, and 60% vinyl), as they are reluctant to react with DMD. Indeed, the vinyl microstructure reacted significantly less than did the cis and trans microstructures. This low reactivity is probably a result of the lower nucleophilicity of the vinyl group. 43
A closer look at Figures 6 to 8 reveals that nano TiO2 is more effective on the epoxidation of cis double bonds than it is on the epoxidation of trans and vinyl double bonds in such a way that the higher the percentage of cis bonds in a polymer, the higher the epoxidation yield in the presence of nano TiO2 will be. Additionally, epoxidation underwent a progressive decline in rate with the passage of time, a fact which may be due to the inductive effects of the epoxy groups created on the polymer.
Furthermore, cis polybutadiene was epoxidized in the presence of different percentages of nano TiO2 and normal TiO2 so as to see how differently the two catalysts affected the epoxidation yield. The results (Table 3) showed that both catalysts increased the epoxidation yield of the polymer, with the stronger effect coming from nano TiO2, especially when it is made up to 3% of the solution. This can be a result of the large ratio of surface area to volume in nano TiO2, which increased the epoxidation yield of double bonds.
Effects of nano TiO2 and normal TiO2 on the epoxidation of cis polybutadiene at 25°C.
Conclusion
Polybutadienes and HTPB in various microstructure configurations were epoxidized with different percentages of cis, trans, and vinyl double bonds, using in situ-generated (DMD) as an oxidant and nano TiO2 as a catalyst. The oxidability of polymers and the effect of varying quantities of nano TiO2 on the epoxidation process were examined after various reaction times by means of 1HNMR, 13CNMR, and FT-IR spectra. Nano TiO2 demonstrated a stronger effect on the epoxidation of double bonds, especially cis bonds. Also, according to the results obtained in this study, relative reactivity depends on the chain microstructure of double bonds as shown in the following relationship: cis > trans > vinyl. A final consideration is that we successfully epoxidized such industrial polymers as polybutadienes and HTPB, using in situ-generated DMD in the presence of nano TiO2.
Footnotes
Funding
The authors thank the Imam Khomeini International University (IKIU) for the financial support of Dr. Alavi Nikje with grant number of D/1/55920.
