Synthetic methodology effect on the microstructure and thermal properties of poly(n-butyl acrylate-co-methyl methacrylate) synthesized by nitroxide mediated polymerization

Abstract

In this work, we report the synthesis of poly(n-butyl acrylate-co-methyl methacrylate) copolymers by the nitroxide mediated polymerization (NMP) technique, using N-tert-butyl-N-(1-diethylphosphono-2,2-dimethylpropyl)nitroxide (SG1) as a control agent and 2-methylaminoxypropionic-SG1 alkoxyamine (BlocBuilder®) as the initiator. The copolymers are synthesized either by batch or semi-batch processes and the gradient profile is examined via the determination of the instantaneous fraction of monomer incorporated in the copolymer. A control of the molar mass together with low molar mass distribution (M_w/M_n < 1.4) is observed. The dependence of the copolymer glass transition temperature with conversion was followed by differential scanning calorimetry. The copolymers are investigated by carbon nuclear magnetic resonance and heteronuclear multiple bond correlation (HMBC) NMR sequences to study the effect of the monomer addition mode on the microstructure of copolymers. The thermomechanical properties of gradient copolymers are finally reported to establish the effect of the composition on the mechanical behaviour of the copolymers.

---

Introduction

Free radical polymerization (FRP) is a versatile technique, largely used for industrial manufacturing procedures, which allows polymerization of a wide range of monomers with various functional groups. Nevertheless, due to the spontaneous transfer reactions and irreversible bimolecular terminations existing during FRP, the lifetime of propagating radicals is very short. This induces generally poor control of polymer microstructures with broad molar mass distribution. With the aim of avoiding the disadvantages of free radical synthesis, living controlled radical polymerizations (CRP) were developed in the mid 1990s. They are based on the principle of dynamic equilibration between dormant and active species with decrease of the instantaneous radical concentration. This induces formation of materials with predictable molecular weight and low distribution. Atom Transfer Radical Polymerization (ATRP),^1–4 Radical Addition Fragmentation and Transfer (RAFT)^5–7 and Nitroxide Mediated Polymerization (NMP)^8,9 are the three most successful radical living techniques frequently used to polymerize large families of monomers. These techniques allow the synthesis of a variety of graft, star, block and gradient (co)polymer complex polymeric architectures. Gradient copolymers are a class of materials re-visited by Matyjaszewski et al. in 1996.^10–12 They are characterized by a gradual change of the copolymer composition along the chains. Depending on the nature of the monomers used, two types of gradient copolymers can be differently synthesized: (a) spontaneous gradients obtained by batch process, with monomers exhibiting different reactivity ratios. The more active monomer will be rapidly consumed in the beginning of polymerization, contrary to less one, inducing spontaneous changes in composition of growing macromolecules. (b) Forced gradient copolymers synthesized by so called semi batch process, when monomers with close reactivity ratios are used. In this process, one monomer is continuously added to the other during the course of the polymerization, using a chosen addition rate. ATRP,^13–19 RAFT,^20–22 and NMP^23–31 techniques were heavily investigated to synthesize this new class of copolymers. Indeed, the ATRP technique was mainly applied by Matyjaszewski et al.^13–17 in synthesize of numerous spontaneous gradients using acrylate/methacrylate or styrene/acrylate pairs of monomers. Syntheses of forced gradients were also reported using poly(methyl methacrylate-grad-2(trimethyl silyl oxy)ethyl methacrylate¹⁸ or styrene/acrylonitrile copolymers.¹⁹

On the other hand, the gradient copolymer of methyl methacrylate (MMA) and methacrylate-terminated poly(dimethylsiloxane),²⁰styrene/butyl acrylate²¹ were synthesized by RAFT technique. This same process was applied in the synthesis of gradient copolymers based on complex-radical terpolymerization of styrene, maleic anhydride and N-vinylpyrrolidone initiated by γ-rays.²²

By using the NMP technique, Mignard et al.²³ have reported the synthesis of gradient copolymers of styrene and n-butyl acrylate comonomers. In addition, styrene/4-acetoxystyrene and styrene/4-hydroxystyrene gradients were reported by Torkelson et al.^24,25 Trang et al.²⁶ reported the synthesize of poly(methyl methacrylate-co-N,N-dimethyl acrylamide) gradient copolymers and Karaky et al. reported the successful synthesis of spontaneous and forced gradient copolymers: poly(octadecyl acrylate-grad-methyl acrylate),²⁷ poly(N,N-Dimethylacrylamide-grad-butyl acrylate),^28,29 poly(styrene-grad-methyl acrylate)³⁰ and poly(styrene-grad-butyl acrylate).³¹

The poly(n-butyl acrylate-co-methyl methacrylate) polymers were largely studied by different techniques, conventional radical polymerization,^32,33 RAFT³⁴ and ATRP.^35–39 These two monomers n-BA, MMA present large difference in the monomer reactivity ratios, r_MMA = 2.19 and r_BA = 0.39,⁴⁰ and it has been demonstrated viaATRP^37,38 that this difference leads to the formation of spontaneous gradient copolymers. However, to the best of our knowledge, the NMP technique was never applied in the synthesis of this couple of comonomers. In this contribution, we have been interested in the use of NMP to synthesise n-butyl acrylate/methyl methacrylate gradient copolymers. Batch and semi batch processes were applied using two addition rates. We focused our attention on the control of the polymerization with a study of kinetic and macromolecular features. The instantaneous composition was concomitantly followed by ¹H-NMR.

The recovered copolymers were thoroughly characterized by ¹³C-NMR, HMBC and DSC experiments in order to confirm the gradient profile. Finally, the effect of the composition on the thermorheological properties is proposed.

Experimental section

Materials

n-Butyl acrylate (BA, Aldrich, 99%), methyl methacrylate (MMA, Aldrich, 99%) were used as received. The alkoxyamine BlocBuilder® (2-methylaminoxypropionic-SG1) used as the initiator and the nitroxide SG1 (88%) were provided by Arkema Chemicals. All solvents were used without further purification.

Synthesis of P(BA-co-MMA) copolymers

Batch 50/50. In a round flask, we mixed BA (31.1 g, 0.242 mol) and MMA (24.3 g, 0.243 mol) with BlocBuilder® alkoxyamine (0.366 g, 0.960 mmol) and SG1 (5 mol% versusinitiator, 0.0143 g, 4.9 × 10⁻² mmol). The targeted degree of polymerization (DP) at full conversion was equal to 505 (([BA] + [MMA])/[BlocBuilder®] = 505). The mixture was degassed for 30 min then placed into an oil bath thermostated at 115 °C. After 6 h and 30 min, the reaction was stopped by cooling down the polymeric solution to room temperature.

Batch 75/25. In a round flask, we mixed BA (61.4 g, 0.480 mol) with MMA (16.0 g, 0.158 mol), BlocBuilder® alkoxyamine (0.487 g, 1.280 mmol) and SG1 (5 mol% versusinitiator, 0.0188 g, 6.4 × 10⁻² mmol). The targeted degree of polymerization at full conversion was equal to 500. After degassing for 30 min, the flask was placed into an oil bath thermostated at 115 °C and the reaction was stopped after 9 h and 30 min.

Semibatch-2.2. A quantity of MMA (19.2 g, 0.192 mol) was degassed then transferred into a 25 ml Hamilton syringe. BA (24.6 g, 0.192 mol) was mixed with BlocBuilder® (0.292g, 0.767 mmol) and SG1 (5 mol% versusinitiator, 0.0112 g, 3.83 × 10⁻² mmol) in the reaction flask equipped with a septum. The targeted degree of polymerization at full conversion was equal to 500. The mixture was purged under nitrogen flow for 30 min and the round bottom flask was immersed into an oil bath at 115 °C to start the reaction (t = 0 min). The PBA was first polymerized alone for 60 min, then the addition of methyl methacrylate was initiated at a constant addition rate equal to 2.2 mL h⁻¹ using a pump. The total addition period of MMA was 555 min. The copolymerization was stopped after total time of polymerization equal to 720 min.

Semibatch-4.3. The copolymerization was performed in similar conditions than the semibatch-2.2 experiment except that the MMA addition rate was set up at 4.3 mL h⁻¹. After polymerizing PBA for 60 min, the addition time of MMA was 280 min and the total polymerization time of copolymerization BA/MMA equal to 420 min.

For all the polymerizations, samples were withdrawn during the polymerization for further NMR, SEC and DSC analyses. The final recovered copolymers were precipitated twice into a methanol/water (90/10) mixture. The final products were filtered and dried in a vacuum oven until constant weights were reached. Table 1 summarizes the experimental conditions of the copolymerizations.

Table 1 Experimental conditions for batch and semi-batch NMP of n-butyl acrylate copolymerized with methyl methacrylate at 115 °C

Experiment	Final nBA (mol)	Final nMMA (mol)			MMA addition rate (mL h^−1)
Batch 50/50	0.242	0.243	505	0.05	—
Batch 75/25	0.48	0.16	500	0.05	—
Semibatch-2.2	0.192	0.192	500	0.05	2.2
Semibatch-4.3	0.192	0.192	500	0.05	4.3

---

Analytical techniques

The crude polymer solutions were analyzed by size exclusion chromatography (SEC) at 30 °C with THF as eluent at a flow rate of 1 mL min⁻¹. The SEC system was equipped with three Waters Styragel columns HR 0.5, 2, 4 working in series (separation range 1 × 10² – 3 × 10⁶ g mol⁻¹) and a Viscotek ERC 7515-A refractive index detector. Number-average molar mass (M_n) and dispersity (M_w/M_n) were calculated with a calibration derived from PS standards using OmniSEC® software. All polymers samples were prepared at 1 to 5 g L⁻¹ concentrations and filtered through PVDF 0.45 μm filters.

The proton nuclear magnetic resonance (¹H NMR) spectra of the crude polymer solutions were recorded at 25 °C using a Bruker Advanced AM400 spectrometer (400 MHz), in CDCl₃ as solvent, in order to determine the monomer conversions and the copolymer compositions. The copolymer compositions were determined from the relative area of the proton of OCH₃ at 3.5–3.65 ppm corresponding to MMA units and of OCH₂ of BA units at 3.8–4.00 ppm. The copolymer microstructures were also analyzed by carbon nuclear magnetic resonance (¹³C NMR, 100 MHz) and heteronuclear multiple bond correlation (HMBC) NMR sequences in CDCl₃.

Differential scanning calorimetry technique (DSC) was carried out on Q100 from TA instruments, to measure the copolymer glass transition temperature (T_g). The samples taken during the polymerization were precipitated into a methanol/water (90/10) mixture. For each measurement, a quantity of 8–10 mg of copolymer was weighed and scanned at 20 °C min⁻¹, from −90 °C to 140 °C, under dry nitrogen (50 mL min⁻¹).

Thermomechanical analysis has been performed by using a constant strain rheometer (RDA II, Rheometrics) where the real part (G′, storage modulus) and the imaginary part (G′′, loss modulus) of the complex shear modulus G* was measured, in the linear domain, at a fixed circular frequency (1 rad s⁻¹) as a function of temperature with a speed of 1 °C min⁻¹. We have also reported tanδ which is the ratio between G′′ and G′. The measurement of G*(T) allows us to determine the different transitions which can influence the mechanical properties, particularly the glassy transition temperature, T_α, defined by the temperature corresponding to the maximum of tanδ. To minimize the compliance of the apparatus according to the behavior of the polymers at the experimental temperature, plate-plate geometry was used with different plate diameters from 5 mm to 15 mm.

Results and discussion

In the present study, the copolymerization of n-butyl acrylate BA and methyl methacrylate MMA was performed by NMP at 115 °C using Blocbuilder alkoxyamine and free SG1 nitroxide. The polymerization was carried out either by batch or in semi-batch processes to target spontaneous and forced gradient copolymers.

Nitroxide mediated polymerization of methacrylic esters has presented a great challenge for a long time since they have a large activation–deactivation equilibrium constant. In 2005, Charleux et al.⁴¹ presented a theoretical approach to describe the copolymerization kinetics in controlled living copolymerization operating via reversible termination. They theoretically determined the variation of the average activation–deactivation equilibrium constant <K> of the copolymerization and showed that less than 10 mol% of styrene was enough to control the NMP of methyl methacrylate at 90 °C by sufficiently decreasing the value of <K> (<K> ≈ 2 × 10⁻⁹ mol L⁻¹).⁴¹ As reported in Fig. S1 (ESI†), a molar fraction of 75% of n-butyl acrylate is required to reach a similar <K> value of 2 × 10⁻⁹ mol L⁻¹ for its copolymerization with MMA carried out at 115 °C. For the purpose of this work aiming at exploring the properties of poly(BA-co-MMA) copolymers with a minimum content of 25% of MMA, we performed two batch copolymerizations starting with two different initial compositions of the comonomer mixture (BA/MMA: 50/50 and BA/MMA: 75/25) and two semi-batch copolymerizations targeting a final composition of 50/50 in BA/MMA. The <K> value for BA/MMA copolymerization performed at 115 °C starting with a molar fraction of BA in the comonomer mixture of 0.5 is equal to 7 × 10⁻⁹ mol L⁻¹ (see Fig. S1†). The experimental conditions of the four polymerizations are gathered in Table 1.

Fig. 1 exhibits the variation of the logarithmic monomer concentration versus time and the evolution of M_n and the dispersity (M_w/M_n) as a function of conversion for both batch copolymerizations. Theoretical M_n of the copolymer, calculated from eqn (1), are also depicted in Fig. 1.

W_BA and W_MMA are the initial weight fractions of the monomers, X_BA and X_MMA are the monomer conversions determined by proton NMR analysis, m_monomers is the initial mass of monomers and n_BlocBuilder is the initial number of moles of alkoxyamine.

Fig. 1 Batch copolymerization of BA and MAMA. (a) Variation of ln[M]₀/[M] versus time. (b) Evolution of the average molar mass (M_n) and dispersity (M_w/M_n) versus conversion. The black symbols (diamonds) and lines correspond to the Batch 50/50 experiment and the grey symbols (triangles) and lines correspond to the Batch 75/25 experiment. The plain and the dotted lines correspond to the fit of theoretical M_n and the experimental M_n respectively.

The variation of the logarithmic monomer concentration versus time shows a fast initiation followed by a linear evolution (Fig. 1a). The experimental M_n increases linearly versus conversion for both Batch 50/50 and Batch 75/25 experiments, which is characteristic of a controlled polymerization. We can notice that the experimental M_n values match the theoretical ones for the Batch 75/25 whereas M_n values were slightly above M_n,theo for the Batch 50/50 experiment. This reveals the presence of early irreversible termination reactions with an initiating efficiency of approximately 80%, which was expected from the highest <K> value observed for an initial 50/50 comonomer mixture (see Fig. S1 in the ESI†). Nevertheless, the two copolymers exhibit features of controlled systems since no shoulder was observed in the SEC chromatograms of the final samples (see Fig. S2 in ESI†) together with low dispersity (1.2 < M_w/M_n < 1.4).

Evolutions of the kinetics and macromolecular features of the semi-batch copolymerizations of BA and MMA are displayed in Fig. 2. An increase of M_n with conversion and narrow molar mass distributions with M_w/M_n values below 1.5 are observed. The SEC traces of the copolymers are reported in Fig. S2 (ESI†) and shows symmetrical peaks. On the other hand, the logarithmic monomer concentrations gradually increase with time and should follow eqn (2).

Fig. 2 Semi-batch copolymerization of BA and MMA. (a) Variation of ln[M]₀/[M] versus time. (b) Evolution of molar mass (M_n) and dispersity (M_w/M_n) versus conversion. The squares correspond to the Semibatch-2.2 experiment and the triangles correspond to the Semibatch-4.3 experiment.

Indeed, the non-linearity of the curve can be explained by the semi-batch process as the continuous addition of MMA into the comonomer mixture induces a change of the MMA comonomer fraction (f_MMA) which impacts directly the <k_p><K> product, with <k_p> the average rate constant of propagation of the copolymerization (see eqn (3) from ref. 41). Fig. 3 depicts the theoretical evolution of the <k_p><K> product with the MMA fraction in the comonomer mixture, hence predicting the non linear variation of ln[M]₀/[M] versus time (see also Fig. S3 in the ESI†). The numerical values used for plotting eqn (3) are reported in ref. 42–44 and note 45.

Fig. 3 Predicted values of the product of the average activation-deactivation equilibrium constant with the average propagation rate constant <k_p> <K>versus the molar fraction of methyl methacrylate in the comonomer mixture, for nitroxide-mediated copolymerization of BA and MMA at 115 °C.

The variation of instantaneous fraction of monomer (F_inst) as a function of the apparent normalized chain length has been largely used to assess the gradient profile of synthetic copolymers.^27,46–48 The instantaneous fraction of BA in the copolymers was calculated from eqn (4), where Δ(%conv) denotes the monomer conversion difference in a time interval.

The apparent normalized chain length is calculated by dividing the M_n value of copolymer, recovered at given time and to the final M_n value of the recovered final copolymer. Fig. 4 illustrates the results obtained for the four BA/MMA copolymerizations experiments. In the case of both batch experiments (Batch 50/50 and Batch 75/25), the BA instantaneous fraction increases with the apparent normalized chain length, which is in accordance with the values of reactivity ratios (r_MMA = 2.19 and r_BA = 0.39).⁴⁰ Since MMA is more reactive than BA, thus it is firstly incorporated and preferentially consumed. This induces the enrichment of the copolymer in BA at the end of the copolymerization, and consequently the formation of instantaneous gradient copolymers.

Fig. 4 Evolution of the instantaneous fraction of BA (F_inst) for synthesized P(BA-co-MMA) copolymers as a function of the apparent normalized chain length: (◆) Batch 50/50, (●) Batch 75/25, (□) Semibatch-2.2 and (△) Semibatch-4.3.

For Semibatch-2.2 and Semibatch-4.3 experiments, BA monomer was first polymerized in bulk for one hour up to 10% conversion (DP ≈ 50) before applying a semi-batch process by incorporating MMA at two different addition rates. As shown by Fig. 4, from the moment the addition of MMA started, we first observed a decrease in the instantaneous fraction of BA in the copolymer with an increase of the apparent normalized chain length until reaching a plateau of F_inst,BA. Hence, gradient copolymers by semi-batch synthesis were also obtained, but with gradient shape different to that obtained by batch synthesis.

Several research groups used the nuclear magnetic resonance (NMR) technique for determining the copolymer microstructure, particularly the sequence distribution and tacticity.^36,49–53 In the present study, we have been interested in investigating the microstructure of the different synthesized copolymers by NMR. The aim was to compare the copolymer microstructure according to the gradient profile (Fig. 4) which depends on the polymerization process (batch or semi-batch). In 2001, According to Aerdts’ method,⁴⁹ Madruga et al.³⁶ reported the study of stereochemical arrangement of monomers in poly(BA-co-MMA) copolymers synthesized by ATRP and FRP techniques. They reported seven distinguishable signals in the carbon carbonyl region of the ¹³C NMR spectra. Their relative intensities changed with the composition of monomer in the copolymer and depended on both the diad tacticity (racemic (r) or meso (m)) and the proportion of the different diads and triads sequences. Four of these signals were assigned to sequences and configurations of MMA (M)-centered triads. The three others corresponded to BA (B)-centered triads.³⁶

The carbonyl area of the copolymers synthesized in the present work was analyzed by carbon NMR. Fig. 5 shows the corresponding NMR spectra in the carbonyl regions of both Batch 75/25 and Semibatch-2.2 and Fig. S4 in the ESI† displays the NMR spectra of Semibatch-4.3 copolymer. The signals of the MMA (M) centered triads are assigned to (A) MMM (rr), (B) MMM (mr + rm), BMM (mr + rr), (C) MMM (mm), BMM (rm + mm), MMB (mr + mm) and BMB (mm + rm + mr), and finally (D) BMB (rr) triads. The signals of the BA (B) centered triads are assigned to (E) MBM, (F) MBB and BBB triads (G).

Fig. 5 ¹³C NMR spectra showing the carbonyl region of (a) Batch 75/25 and (b) Semibatch-2.2 copolymers as function of time and apparent normalized chain length.

The comparison between the intermediate ¹³C NMR spectra recorded at different polymerization time intervals for the Batch 75/25 experiment, as shown in Fig. 5, reveals an increase in the relative intensity of BBB triad signal versus time, accompanied with a decrease in the relative intensities of (A), (B), (C) and (E) signals. This confirms the enrichment of the copolymer chain in BA, as observed in Fig. 4. In contrast, the series of ¹³C NMR spectra for the Semibatch-2.2 copolymer present a decrease in the relative intensity of BBB triad signal versus time, together with a gradual increase in the relative intensities of (A), (B), (C) and (E) signals (see Fig. 5). These results confirm the gradual incorporation of MMA units in the copolymer chains. Similar results are observed for the Semibatch-4.3 copolymer (see Fig. S4 in the ESI†).

Two dimensional NMR (2D NMR) experiments are very useful for the investigation of polymer microstructures^50–53 considering different techniques especially Heteronuclear Single Quantum Coherence (HSQC), Total Correlated Spectroscopy (TOCSY) and Heteronuclear Multiple Bond correlation (HMBC).

Brar et al.⁵³ have extensively studied the microstructure of spontaneous poly(MMA-g-BA) copolymers produced by ATRP with a complete assignment of the peaks by 2D HSQC, 2D TOCSY and 2D HMBC experiments. In this work, we have been interested in the characterization of the synthesized copolymers by 2D-HMBC NMR spectroscopy in the area of the α-methyl, methylene and methyne protons coupled with carbonyl carbons, in order to illustrate the formation of MMMM or BBBB tetrads in copolymers, not observed by ¹³C NMR. We focused our attention on the crosspeaks observed at: (1) 177.24/0.94 ppm assigned to the coupling of carbonyl carbons with α-methyl proton in the triad MrMmM. The three crosspeaks (2) 177.24/1.91 ppm, (3) 177.42/1.86 ppm and (4) 177.00/1.37 ppm, assigned to the coupling of the carbonyl carbon with the methylene protons of the MrMmMrM tetrads. Also, the crosspeaks at (5) 177.05/0.81 ppm, characteristic of the coupling of α-methyl protons with carbon carbonyle in MrMB triad, (6) 177.05/1.81 ppm assigned to the correlation of methine proton of triad MMB with MrMB of the carbonyl carbon. (7) 174.90/2.19 ppm, assigned to BBM triads of CH. The three crosspeaks (8) 174.64/1.33, (9) 174.42/1.85 and (10) 174.42/1.36 correspond to BBmBB tetrads. As depicted in Fig. 6, by comparing between HMBC spectra of Batch 75/25, at 45 min, 300 and 570 min, we notice the appearance of crosspeaks (8), (9) and (10), reflecting the enrichments of chains in BBmBB tetrads. Fig. 7 shows results obtained for Semibatch-2.2. After 60 min, only crosspeaks characteristic of polybutyl acrylate are observed, since it was firstly homopolymerized. At 300 min, we notice the formation of MrMB and MrMM triads. At 720 min, the cross-peaks (2), (3) and (4), characteristics of the tetrads MrMmMrM are observed, confirming the enrichment of polymeric chains in MMA units.

Fig. 6 2D HMBC NMR spectra for Batch75/25 in CDCl₃ (as function of time, apparent normalized chain length).

Fig. 7 2D HMBC NMR spectra for Semibatch-2.2 in CDCl₃ (as function of time, apparent normalized chain length).

The four poly(BA-co-MMA) copolymers were analyzed by differential scanning calorimetry to follow the glass transition temperature (T_g) of the copolymer as a function of the apparent normalized chain length (see Fig. 8) in order to compare with the gradient profile. Indeed, many studies have reported that depending on the nature of monomers and synthesis conditions, gradient copolymers may present different thermal behaviours.^{25,28,38,54–56} Torkelson et al.²⁵ reported the presence of one broad T_g in poly(styrene-grad-4-hydroxystyrene), the temperature transition width being as large as 65–85 °C. A microphase separation was highlighted in other gradient copolymers by the presence of two glass transition temperatures, as reported by Karaky et al.^28,29 in the poly(N,N-dimethylacrylamide-grad- butyl acrylate) gradient copolymer.

Fig. 8 Evolution of the glass transition temperatures (T_g) as a function of apparent normalized chain length in synthesized poly(BA-co-MMA) copolymers: (◆) Batch 50/50, (●) Batch 75/25, (□) Semibatch-2.2 and (△) Semibatch-4.3.

As depicted in Fig. 8, we noticed that the two batch copolymers exhibit a unique T_g intermediate between those of two PBA and PMMA homopolymers (T_g, PBA ≈ −54 °C; T_g,PMMA ≈ 100 °C for DP > 100). The characteristics of the final copolymers are reported in Table 2. For both Batch 75/25 and Batch 50/50 experiments, the value of the T_g remains constant with the increase of the apparent normalized chain length despite the enrichment in BA highlighted in Fig. 4 and 5. Similar results were reported by Madruga et al. when studying poly(BA-co-MMA) spontaneous gradient copolymers obtained by ATRP.³⁸ It has been reported that T_g values remained constant with the increase of monomer conversion. This was explained by the presence of two opposite factors which can simultaneously affect the glass transition temperature values. The first one is the continuous increase in molar mass with conversion inducing an increase in T_g of copolymers, while the second one is a composition variation with enrichment in BA causing a decrease in T_g. These two effects compensate each other, and therefore the T_g values appear constant. The same explanation was also reported for poly(allyl methacrylate-g-butyl acrylate) copolymers⁵⁶ and remains valid in the case of our synthesized Batch 50/50 and Batch 75/25 copolymers. In contrast, the Semibatch-2.2 copolymer exhibits a large T_g equal to −3.2 °C and a second weak transition at 82 °C (see Fig. S5 in ESI†). In the case of the Semibatch-4.3 copolymer, a first T_g equal to −0.6 °C was observed together with a second very weak transition at 56 °C (see Fig. S5 in the ESI†).

Table 2 Characteristics of synthesized copolymers

Experiment	Overall monomer conversion (%)	BA molar fraction in the copolymer (%)	M n (g mol^−1)	M w/Mn	T g ^a(°C)	T g th ^b(°C)	T α ^c(°C)
a Experimental Tg measured by DSC. b Theoretical Tg calculated from eqn (5). c Experimental Tg measured by rheology.
Batch50/50	70	37	48600	1.24	22	20	31.6
Batch75/25	90	73	55800	1.33	−22	−25	n.a.
Semibatch-2.2	91	56	61900	1.32	−3.2 (82)	−6.2	1.6
Semibatch-4.3	88	46	55100	1.29	−0.6 (56)	6.8	24.5

---

In addition, the increase of T_g values with the apparent normalized chain length confirms the gradient profile of the both Semibatch copolymers (see Fig. 8). The increase of T_g values confirmed the incorporation of MMA along the copolymer chains and the production of forced poly(BA-grad-MMA) gradient copolymers. The two weak glass transition temperatures at 82 °C and 56 °C only observed in final products are not reported in this figure. Table 2 gathers the experimental T_g values obtained for the final copolymers and measured by DSC and rheology. The difference between the two T_g values is a classical feature.⁵⁷ In the same table, we collected the theoretical T_g values determined by the Fox equation (see eqn (5) which is commonly applied to predict glass transition temperatures, especially in statistical copolymers.

W_BA and W_MMA are the weight fractions of BA and MMA in the copolymer, T_g,BA and T_g,MMA are the T_g values of the two PBA and PMMA homopolymers, respectively considered equal to −47 and 105 °C.⁵⁸ We noticed that the theoretical values of T_g are close to the experimental values of T_g determined by DSC analysis or by rheology and present a similar evolution. However, this equation does not take into account the organization of monomers along the chains and the microstructure of copolymers as it depends only on the weight fractions of each monomer. Consequently, the second weak glass transition temperature observed in the forced gradient copolymers, synthesized by semi-batch process, could not be predicted by the Fox equation.

We have reported the thermomechanical analysis of Batch 50/50, Semibatch-2.2 and Semibatch-4.3 in the Fig. 9. The three gradient copolymers exhibit similar global behaviour particularly at higher temperatures. The glass-transition temperature is clearly affected by the composition as reported in Table 1. Each sample exhibits a narrow rubbery zone in the intermediate temperatures domain (tanδ < 1) just after the transition region. This particularity is certainly due to the molar mass of PBA domains which present a limit for entanglements (M_wPBA part ≈ 2 × Me).⁵⁸ Finally, all samples exhibit a terminal domain at relatively low temperatures (from 55 °C to 75 °C lower than T_g of PMMA) in accordance with non-segregation observed by AFM investigation. This last point is still underway and will have to be evaluated by a spectromechanical analysis which is relevant to explore the segregation of block copolymer.⁵⁹

Fig. 9 Temperature dependence of the storage modulus (G′ – left axis) and tan δ (right axis) for Batch 50/50 (37% mol BA with M_n = 48 600 g mol⁻¹, I_p = 1.24, tan δ = 32 °C), Semibatch-2.2 (56% mol BA with M_n = 61 900 g mol⁻¹, I_p = 1.32, tan δ = 12.3 °C) and Semibatch-4.3 (46% mol BA with M_n = 55 100 g mol⁻¹, I_p = 1.29, tan δ = 24.5 °C).

Conclusion

In this work, copolymers of poly(butyl acrylate-grad-methyl methacrylate) were synthesized by nitroxide mediated polymerization, applying both the batch and the semi batch processes. The obtained materials were characterized by H-NMR and SEC analysis. It has been found that criteria for controlled radical polymerization were verified, with controlled molecular weight and narrow dispersities. The gradient profiles of synthesized materials were verified by the study of variation of instantaneous fractions of butyl acrylateversus apparent normalized chain length in copolymers. On the other hand, ¹³C-NMR spectroscopy and hetero nuclear multiple bond correlation experiments (HMBC) were also applied to study the microstructure of synthesized copolymers and the distribution of BA, MMA units along polymeric chains. The evolution of glass transition temperatures with total conversion of monomers during polymerization were determined using the DSC technique, and permitted us to confirm the gradient profile of copolymers. In addition, thermomechanical analysis of these materials demonstrates that the physical properties depend on the synthesis method which has an effect on the final composition of gradient copolymer and corresponding molecular weight.

Acknowledgements

NC thanks the Algerian Ministry of high education and scientific research (MESRS-PNE program) for funding. The authors are grateful to Arkema for providing the SG1 nitroxide and BlocBuilder® alkoxyamine. The authors would like to thank G. Clisson for his help with SEC.

References and notes

1. W. A. Braunecker and K. Matyjaszewski, Prog. Polym. Sci., 2007, 32, 93.
2. J. Qiu, B. Charleux and K. Matyjaszeski, Prog. Polym. Sci., 2001, 26, 2083.
3. V. Coessens, T. Pintauer and K. Matyjaszeski, Prog. Polym. Sci., 2001, 26, 337.
4. M. Kamigaito, T. Ando and M. Sawamoto, Chem. Rev., 2001, 101, 3689.
5. J. Chiefari, Y. K. B. 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 and S. H. Thang, Macromolecules, 1998, 31, 5559.
6. G. Moad, E. Rizzardo and S. H. Thang, Aust. J. Chem., 2005, 58, 379.
7. G. Moad, E. Rizzardo and S. H. Thang, Aust. J. Chem., 2006, 59, 669.
8. J. C. Hawker, A. W. Bosman and E. Harth, Chem. Rev., 2001, 101, 3661.
9. B. Charleux and J. Nicolas, Polymer, 2007, 48, 5813.
10. D. Greszta and K. Matyjaszewski, Polym. Prepr. (Am. Chem. Soc., Div. Polym. Chem.), 1996, 36, 569.
11. T. Pakula and K. Matyjaszewski, Macromol. Theory Simul., 1996, 5, 987.
12. D. Greszta and K. Matyjaszewski, Polym. Prepr. (Am. Chem. Soc., Div. Polym. Chem.), 1996, 37(2), 569.
13. M. J. Ziegler and K. Matyjaszewski, Macromolecules, 2001, 34, 415.
14. S. C. Hong, J. F. Lutz, Y. Inoue, C. Strissel, O. Nuyken and K. Matyjaszewski, Macromolecules, 2003, 36, 1075.
15. S. Qin, J. Saget, J. Pyun, S. Jia, T. Kowalewski and K. Matyjaszewski, Macromolecules, 2003, 36, 8969.
16. H. Lee, K. Matyjaszewski, S. Yu and S. S. Sheiko, Macromolecules, 2005, 38, 8264.
17. D. Neugebauer, M. Theis, T. Pakula, G. Wegner and K. Matyjaszewski, Macromolecules, 2006, 39, 584.
18. H. G. Börner, D. Duran, K. Matyjaszewski, M. Silva and S. S. Sheiko, Macromolecules, 2002, 35, 3387.
19. D. Greszta, K. Matyjaszewski and T. Pakula, Polym. Prepr. (Am. Chem. Soc., Div. Polym. Chem.), 1997, 38, 709.
20. H. Shinoda, K. Matyjaszewski, L. Okrasa, M. Mierzwa and T. Pakula, Macromolecules, 2003, 36, 4772.
21. X. Sun, Y. Luo, R. Wang, B. Li, B. Liu and S. Zhu, Macromolecules, 2007, 40, 849.
22. Z. Hu and Z. Zhang, Macromolecules, 2006, 39, 1384.
23. E. Mignard, T. Leblanc, D. Bertin, O. Guerret and W. F. Reed, Macromolecules, 2004, 37, 966.
24. M. K. Gray, H. Zhou, S. T. Nguyen and J. M. Torkelson, Macromolecules, 2004, 37, 5586.
25. J. Kim, M. M. Mok, R. W. Sandoval, D. J. Woo and J. M. Torkelson, Macromolecules, 2006, 39, 6152.
26. N. Trang, T. Phan, S. Maiez-Tribut, J. Pascault, A. Bonnet, P. Gerard, O. Guerret and D. Bertin, Macromolecules, 2007, 40, 4516.
27. K. Karaky, G. Clisson, G. Reiter and L. Billon, Macromol. Chem. Phys., 2008, 209, 715.
28. K. Karaky, C. Derail, G. Reiter and L. Billon, Macromol. Symp., 2008, 267, 31.
29. K. Karaky, C. Pouchan, J. Desbrières and L. Billon, Macromolecules, 2007, 40, 458.
30. K. Karaky, E. Pere, C. Pouchan, H. Garay, A. Khoukh, J. Desbrières, J. Francois and L. Billon, New J. Chem., 2006, 30, 698.
31. K. Karaky, E. Pere, C. Pouchan, J. Desbrières, C. Derail and L. Billon, Soft Matter, 2006, 2, 770.
32. P. Bujak, N. Henzel and M. Matlengiewicz, Int. J. Polym. Anal. Charact., 2008, 13, 149.
33. E. L. Madruga and M. Fernández-García, Macromol. Chem. Phys., 1996, 197, 3743.
34. X. Zhang, J. Gao, Y. Luo and B. Li, Acta Polym. Sin., 2008, 1, 83.
35. S. V. Arehart and K. Matyjaszewski, Polym. Prepr. (Am. Chem. Soc., Div. Polym. Chem.), 1999, 40, 458.
36. J. L. de la Fuente, M. Fernández-García, M. Fernández-Sanz and E. L. Madruga, Macromolecules, 2001, 34, 5833.
37. M. J. Ziegler and K. Matyjaszewski, Macromolecules, 2001, 34, 415.
38. M. Fernández-García, J. L. De La Fuente, M. Fernández-Sanz and E. L. Madruga, Macromol. Rapid Commun., 2001, 22, 1046.
39. A. Nese, J. Mosnáček, A. Juhari, J. A. Yoon, K. Koynov, T. Kowalewski and K. Matyjaszewski, Macromolecules, 2010, 43, 1227.
40. G. Sebastian Roos, Axel H. E. Müller and K. Matyjaszewski, Macromolecules, 1999, 32, 8331.
41. B. Charleux, J. Nicolas and O. Guerret, Macromolecules, 2005, 38, 5485.
42. S. Beuermann and M. Buback, Prog. Polym. Sci., 2002, 27, 191.
43. G. S. Ananchenko, M. Souaille, H. Fischer, C. Le Mercier and P. Tordo, J. Polym. Sci., Part A: Polym. Chem., 2002, 40, 3264.
44. P. Lacroix-Desmazes, J. F. Lutz, F. Chauvin, R. Severac and B. Boutevin, Macromolecules, 2001, 34, 8866.
45. The numerical values used for plotting eqn (3) are as follows: r_MMA = 2.19, r_BA = 0.39 (ref. 40), k_p, MMA (115 °C) = 2575 L mol⁻¹s⁻¹ (ref. 42), k_p, BA (115 °C) = 82 245 L mol⁻¹s⁻¹ (ref. 42), K_MMA (100 °C) = 1.30 × 10⁻⁷ mol⁻¹ L⁻¹ (ref. 43), K_BA (112 °C) = 4.34 × 10⁻¹¹ mol⁻¹ L⁻¹ (ref. 44).
46. K. Matyjaszewski, M. J. Ziegler, S. V. Arehart, D. Greszta and T. Pakula, J. Phys. Org. Chem., 2000, 13, 775.
47. M. Yu. Zaremski, D. I. Kalugin and V. B. Golubev, J. Polym. Sci., 2009, 51, 103.
48. M. Kryszewski, Polym. Adv. Technol., 1998, 9, 244.
49. A. M. Aerdts, A. L. German and G. P. M. van der Velden, Magn. Reson. Chem., 1994, 32, S80.
50. A. S. Brar, G. Singh and R. Shankar, J. Mol. Struct., 2004, 703, 69.
51. A. S. Brar and Puneeta, J. Appl. Polym. Sci., 2006, 99, 2694.
52. A. S. Brar, S. Hooda and R. Kumar, J. Polym. Sci., Part A: Polym. Chem., 2003, 41, 313.
53. A. S. Brar and S. Kaur, J. Polym. Sci., Part A: Polym. Chem., 2005, 43, 1100.
54. C. L. H. Wong, J. Kim and J. M. Torkelson, J. Polym. Sci., Part B: Polym. Phys., 2007, 45, 2842.
55. M. M. Mok, J. Kim and J. M. Torkelson, J. Polym. Sci., Part B: Polym. Phys., 2008, 46, 48.
56. R. Paris and J.l. de la Fuente, J. Polym. Sci., Part B: Polym. Phys., 2007, 45, 1845.
57. G. Marin, Ph. Vandermaesen and J. Momornicki, J. Adhes., 1991, 35, 23.
58. N. Jullian, F. Léonardi, B. Grassl, J. Peyrelasse and C. Derail, Appl. Rheol., 2010, 20(3), 33685.
59. N. Jullian, L. Rubatat, J. Peyrelasse, P. Gerard and C. Derail, J. Rheol., 2011, 55(2), 379.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c1py00066g

---