Gelation of Triblock Copolymers in Aqueous ... - Wiley Online Library

Macromolecular Chemistry and Physics

Full Paper

Gelation of Triblock Copolymers in Aqueous Solution through CO2-Triggered Electrostatic Interaction Bing Yu, Weizheng Fan, Yue Zhao* Polymer systems displaying CO2-triggered rheology alteration are widely investigated in recent years, which can change from nonviscous liquid to high-viscosity solution or even gel by passing CO2 through. Among those polymer systems, a previous study of the group shows that a rationally designed ABA-type triblock copolymer with a hydrophilic middle block and two CO2-responsive end blocks can undergo either gel–sol or sol–gel transition triggered by CO2 as a result of increase or decrease in lower critical solution temperature of A blocks, respectively. This paper describes a new and more flexible approach to achieving CO2-induced gelation of polymer aqueous solution at much lower polymer concentrations. It consists in mixing two water-soluble ABA triblock copolymers that have the same B block but different A blocks: for one polymer, the A blocks are a negatively charged polyelectrolyte, and for the other polymer, the A blocks can be turned positively charged under CO2 stimulation. In this way, CO2 bubbling through an aqueous solution of the two constituent triblock copolymers can induce complexation of the oppositely charged A blocks and, as a result, gelation of the solution of mixed triblock copolymers.

1. Introduction Stimulus-responsive polymeric gels, including block copolymer gels, have attracted considerable attention due to their potential applications in biomedical fields such as drug delivery, encapsulation of cells, and tissue-engineering.[1] Interestingly, the gelation mechanism of diblock copolymer is different from that of triblock copoly mer.

Dr. B. Yu, W. Fan, Prof. Y. Zhao Département de chimie Université de Sherbrooke Sherbrooke QC J1K 2R1, Canada E-mail: [email protected] The ORCID identification number(s) for the author(s) of this article can be found under https://doi.org/10.1002/ macp.201700146. Macromol. Chem. Phys. , DOI: 10.1002/macp.201700146

© 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

The diblock copolymer gels are formed through the closepacking of block copolymer micelles, and the gelation usually occurs above a critical concentration when the micellar coronas start to overlap. As a result, the critical gelation concentration of AB diblock copolymers decreases with increasing the size of micelles, and normally is above 20 wt%.[2] By contrast, most of ABA triblock copolymer gels are formed through the bridging chains that interconnect the micelles. In this case, the triblock copolymer is composed of hydrophilic middle B block and hydrophobic end A blocks, and gelation occurs when a sufficient number of micelles are interconnected by bridging chains whose A blocks are located in different micelles. As compared with diblock copolymers, the critical gelation concentration of triblock copolymer is lower (5–10 wt%) and the gels are stronger.[3] In the present study, ABA triblock copolymer is chosen to investigate the carbon dioxide (CO2)-induced block copolymer gelation through a new mechanism.

wileyonlinelibrary.com

DOI: 10.1002/macp.201700146

(1 of 10) 1700146


B. Yu et al.

www.mcp-journal.de

In principle it is possible to use any stimulus to trigger the gelation of an ABA triblock copolymer if the stimulus can enhance the hydrophobicity or crystallization behavior of the A blocks; and a variety of stimuli such as temperature,[4] pH,[5] and light irradiation[6] have already been proved effective in causing gelation of ABA triblock copolymers in aqueous solution. If the cross-linking induced by stimulus is purely physical in nature, the gelation usually can be reversible and a gel–sol transition may occur simply by reversing the relevant stimuli.[7] To date many kinds of stimuli-responsive ABA triblock copolymer gels have been reported. For example, a gelation of azobenzene containing triblock copolymer was realized in an appropriate temperature range after visible light irradiation, and the obtained gels can be dissociated after UV light irradiation.[6b] A reversible gelation of poly(lactic acid) (PLA) containing block copolymer was observed by varying the temperature for crystallization and melting of the PLA blocks, respectively.[8] On the other hand, studies in recent years showed that some stimuli, such as pH and chemical agents, require repeated addition of chemicals into the polymer system and may result in salt accumulation and chemical contamination, while other stimuli, such as heating, mechanical force, and exposure to light, electrical or magnetic field, may be harmful to biological tissue and costly in consuming energy. Compared to these “traditional” stimuli, the CO2 stimulation can overcome the disadvantages, and its stimulation strength can easily be tuned through a continuous gas flow.[9] A number of CO2-switchable surfactants,[10] CO2-tunable vesicles,[11] CO2 wormlike micelles,[12] CO2-responsive microgels,[13] gas sensor,[14] and CO2-responsive membranes[15] have been reported.[16] Moreover, the CO2 gas can also be used as a trigger to induce the gelation.[17] According to a previous study of our group on ABA triblock copolymers, by introducing CO2-responsive comonomer units such as 2-(dimethylamino)ethyl methacrylate (DMAEMA) or methacrylic acid (MAA) into the A blocks of poly[di(methylene glycol) methyl ether methacrylate]-block-poly(ethylene oxide)block-poly[di(methylene glycol) methyl ether methacrylate] (PMEO2MA-b-PEO-b-PMEO2MA), a gel–sol or a sol–gel transition triggered by CO2 bubbling into the aqueous solution could be realized.[18] In that case, the phase transitions were made possible through a shift of the lower critical solution temperature (LCST) of the PMEO2MA blocks due to the CO2-induced protonation of either DMAEMA (weak base) or MAA (weak acid) comonomer units. As a result, the sol–gel or gel–sol transition can occur only when the gas effect can displace the LCST below or above a given polymer solution temperature, respectively. Although interesting, the underlying mechanism of the CO2-induced gelation based on LCST changes may

1700146 (2 of 10)

limit its application over a wide range of temperatures. In particular, CO2-induced gelation through protonation of MAA units is thermally unstable.[18] To address this issue, triblock copolymers with different structures and cross-linking pathways need to be designed and investigated. In this regard, an aggregation induced by electrostatic interactions may be an interesting choice. Although a number of CO2-switchable viscoelastic fluids[19] and CO2-induced network in nitrogen-containing polymer or hybrid material[20] involving electrostatic interactions have been reported, to the best of our knowledge, no ABA triblock copolymer gels with a critical gelation concentrations lower than 5 wt% was reported. Herein, we present the investigation on a novel type of CO2-triggered ABA triblock polymer gelation based on cross-linking from CO2-induced polyelectrolyte complexation. Compared with the gels formed through hydrophobic interaction, the strength of the physical gels based on electrostatic interaction could be more easily tunable by the amounts of charges in the triblock copolymers. At the same time, as the electrostatic interaction between anionic polymer and cationic polymer can persist over a wide range of temperatures in both water and organic solvents, this CO2-triggered gelation system is of interest for potential application as rheology modifier over a wide range of temperature.

2. Experimental Section 2.1. Materials All chemicals were purchased from Sigma-Aldrich unless otherwise stated. Copper(I) chloride (CuCl, 99.999%), trifluoroacetic acid (TFA, 99%), α-bromoisobutyryl bromide (98%), N,N,N′,N′,N″pentamethyldiethylenetriamine (PMDETA, 99%), poly(ethylene oxide) (PEO, Mn = 20 000 g mol−1), triethylamine (TEA, 99%), anhydrous toluene (99.8%), dichloromethane (DCM, 99%), and tetrahydrofuran (THF, 99%) were used directly without further purification. 2-(Dimethylamino)ethyl methacrylate (DMAEMA, 98%), and tert-butyl methacrylate (tBMA, 98%) were passed through a column of basic aluminum oxide prior to use. PEO macroinitiator (Br–PEO–Br) was synthesized via the esterification reaction of bifunctional PEO with α-bromoisobutyryl bromide in toluene according to the literature.[21] All aqueous solutions of polymers were prepared using ultrapure water from a Millipore Milli-Q system. CO2 (99.998%) and nitrogen (99.998%) were purchased from Praxair Canada Inc. and used as received.

2.2. Synthesis of ABA Triblock Copolymer PMAA-b-PEO-b-PMAA The triblock copolymer PMAA-b-PEO-b-PMAA was obtained by hydrolysis of PtBMA-b-PEO-b-PtBMA which was synthesized through the atom transfer radical polymerization (ATRP) of tBMA with Br-PEO-Br as macroinitiator. The synthetic process

Macromol. Chem. Phys. , DOI: 10.1002/macp.201700146 © 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com

Gelation of Triblock Copolymers in Aqueous Solution through CO2-Triggered Electrostatic Interaction

Macromolecular Chemistry and Physics www.mcp-journal.de

was shown in Scheme S1 (Supporting Information). Br-PEO-Br (20 000 g mol−1, 0.60 g, 0.03 mmol), CuCl (17.8 mg, 0.18 mmol), tBMA (4.26 g, 30 mmol), PMDETA (31.2 mg, 0.18 mmol), and 10 mL THF were added into a 25 mL flask. The reaction mixture was degassed by three-pump-thaw cycles, back-filled with nitrogen and placed in an oil bath equilibrated at 60 °C for desired reaction time. The mixture was then diluted with THF and passed through a column of neutral alumina to remove the metal salt. After precipitation of the THF solution in hexane for three times, the triblock copolymer was collected and then dried under vacuum overnight, yielding a light yellow solid. PtBMA-bPEO-b-PtBMA with different lengths of PtBMA blocks were synthesized similarly but with different reaction times. Then the target triblock copolymer PMAA-b-PEO-b-PMAA (samples PA1-PA3) were prepared through the hydrolysis of the PtBMA-b-PEO-b-PtBMA synthesized above. For example, PtBMA-b-PEO-b-PtBMA (2.0 g) was added into a 50 mL roundbottom flask and dissolved into DCM (20 mL). Then TFA (8.0 mL) was added into the flask, and the reaction mixture was stirred at room temperature for 24 h. After removal of the solvent and excess TFA by rotary evaporator. The obtained polymer was dialyzed against 0.01 mol L−1 NaOH aqueous solution for 1 d and deionized water for another 2 d. Finally, a white solid was obtained after lyophilization, which was the triblock copolymer PMAA-b-PEO-b-PMAA in sodium salt form.

2.3. Synthesis of ABA Triblock Copolymer PDMAEMA-b-PEO-b-PDMAEMA The triblock copolymer PDMAEMA-b-PEO-b-PDMAEMA was synthesized similarly through the ATRP of DMAEMA with Br-PEOBr as macroinitiator. The synthetic process was also shown in Scheme S1 (Supporting Information). Br-PEO-Br (20 000 g mol−1, 0.60 g, 0.03 mmol), CuCl (17.8 mg, 0.18 mmol), DMAEMA (4.72 g, 30 mmol), PMDETA (31.2 mg, 0.18 mmol), and 10 mL THF were added into a 25 mL flask. The reaction mixture was degassed by three-pump-thaw cycles, back-filled with nitrogen, and placed in an oil bath equilibrated at 60 °C for desired time. The mixture was then diluted with THF and passed through a column of neutral alumina to remove the metal salt. After precipitation of the THF solution in hexane for three times, the triblock copolymer was collected and then dried under vacuum overnight, yielding a light yellow solid. PDMAEMA-b-PEO-b-PDMAEMA with different lengths of PDMAEMA blocks (samples PB1–PB3) were synthesized similarly with different reaction times.

2.4. Characterization 1H

NMR spectra were recorded on a Bruker 300 MHz spectrometer (Bruker, Switzerland) operating at 300 MHz by using CDCl3 and D2O as solvent at 293 K. The chemical shifts were referenced to residual peaks of solvent: CDCl3 (δ = 7.26 ppm) and D2O (δ = 4.81 ppm). The number-average and weight-average molecular weights (Mn and Mw) and the polydispersity index (Ð) were measured by a Waters size exclusion chromatograph (SEC) instrument equipped with a Waters 2414 differential refractometer detector and a Waters 2998 photodiode array detector. Polystyrene (PS) was used as standards to calculate the molecular weight. The SEC


measurements were carried out at 35 °C using a column (Waters Styragel HR4E, 7.8 × 300 mm2, 5 µm beads) and THF as eluent (flow rate: 1.0 mL min−1). The Zeta potential measurements of the polymer aqueous solution were performed using a Zetasizer Nano ZS instrument (Malvern Instruments Ltd., U.K.) equipped with a multi-τ digital time correlation and a 4 mW He–Ne laser (λ = 633 nm) at an angle of 173°. Regularized Laplace inversion (CONTIN algorithm) was applied to analyze the obtained autocorrelation functions. The gelation state of mixed solution of PA and PB was determined by simple vial inversion experiments. Different amounts of PA and PB were dispersed in ultrapure water in proportion with a constant of total concentration. Then different amounts of CO2 were added into the solution filled in a sealed vial with a bubbling rate of 5 mL min−1, and the gelation state of the obtained solution was detected. It was defined that a gelation occurred if the solution can remain in position after the inversion of the vial;[5b] otherwise, it was considered that no gelation occurred.

3. Results and Discussion The schematic in Figure 1 depicts the design. To achieve CO2-induced gelation based on electrostatic interaction, two constituent ABA triblock copolymers are required: one is composed of a hydrophilic B block and two negatively charged A blocks (polyelectrolyte) with essentially no CO2 responsiveness (named as PA); the other one comprises a hydrophilic B block and two CO2-responsive A blocks that can be protonated to be positively charged after CO2 bubbling (denoted as PB). After mixing PA and PB in aqueous solution, gelation is expected to form upon CO2 bubbling that induces positive charges on PB and thus complexation of PA and PB, resulting in a network structure. In this study, as an exemplary system, PA is composed of PEO as the B block and PMAA as A block, while PB has also PEO as B block and PDMAEMA as A block. PDMAEMA is chosen as the CO2-responsive block, in which the tertiary amine groups can readily be protonated on presence of CO2.[22] The synthesis process of PA and PB was shown in Scheme S1 (Supporting Information). PA was synthesized through ATRP with a PEO initiator and tBMA monomers, then the product was hydrolyzed with TFA. Similarly, PB was also synthesized through ATRP with a PEO initiator and DMAEMA monomers. In order to increase the solubility of PMAA and decrease the initial protonation degree of PDMAEMA before CO2 bubbling, unless otherwise specified, all PA samples in this study were neutralized with sodium hydroxide and the initial pH of aqueous solution was adjusted to be ≈9. The compositions and molecular weights of the obtained triblock copolymers were determined from their 1H NMR spectra. The signal at δ = 1.49 ppm was greatly weakened, indicating that the tBMA units were


(3 of 10) 1700146


B. Yu et al.

www.mcp-journal.de

weights calculated by NMR and SEC are close to each other for PA samples. Likewise, the molar content of DMAEMA units in PB as well as the NMR-based molecular weights were calculated by comparing the integral areas of the resonance peaks of OCH2CH2 in PEO (≈3.65 ppm) and N(CH3)2 in DMAEMA (≈2.41 ppm) (Figure S2, Supporting Information). They also increased from PB1 to PB2 to PB3. Before discussing the results on CO2induced gelation, it is noted that as reported previously, polymers bearing carboxylic acid groups can also be CO2responsive in water.[18,23] However, this CO2 responsiveness is dependent on the initial pH of the system, which can be affected by a number of factors such as polymer concentration, content of functional groups and others. Generally, the CO2 induced protonation of carboxylic acid containing polymers occurs at a pH significantly higher than the pKa when the water solubility of their neutral Figure 1. The formation mechanism of the ABA triblock polymer gel based on CO2-trigform is poor.[9b,24] In this study, PMAA gered polyelectrolyte complexation. homopolymer is the A block of the PA triblock copolymer. As the water solubility of PMAA in hydrolyzed into MAA units. By comparing the inteneutral form is good, which changes only a little after gral areas of the resonance peaks of OCH2CH2 in protonation, their CO2 response is very weak and negliPEO (≈3.65 ppm) and C(CH3)3 in tBMA (≈1.49 ppm) (Figure S1, Supporting Information), the average number gible, and this property will be proved further on. of repeating units of PMAA blocks in PA was also calcuThe gelation state under CO2 stimulation for pure PA, lated and the results are shown in Table 1. Both the molar pure PB, and their mixture in aqueous solution were first content of MAA units in the triblock copolymer as well as assessed using samples of PA3, PB3, and their 3:1 (weight the NMR-based molecular weight increased from PA1 to ratio) mixture. The observation results are shown by the PA2 to PA3. The molecular weights of the three precursors photographs in Figure S3 (Supporting Information) and of PA before hydrolysis were also measured by means of Figure 2a. Both PA and PB alone cannot form gel after SEC, and the results are given in Table 1. The molecular injection of CO2 in aqueous solution and equilibration for Table 1. The compositions and molecular weights of PA and PB.

Mn (NMR)a) [g mol−1]

Mn (SEC)b) [g mol−1]

Ðb)

Number of repeating units in A blocksa)

Content [mol%] of MAA or DMAEMA in triblock polymera)

PA1

39 000

37 900

1.30

67

14.7

PA2

69 000

69 700

1.19

172

37.8

PA3

106 000

82 500

1.05

301

66.2

c)

–c)

PB1

50 000

–

95

20.9

PB2

64 000

–c)

–c)

140

30.8

c)

–c)

300

65.9

PB3

114 000

–

a)Average molecular weight (M ), number of repeating units in A (PMAA or PDMAEMA) blocks, and content of PMAA or PDMAEMA were n determined from 1H NMR spectra by comparing the integrals of appropriate peaks; b)Average molecular weight (Mn) and polydispersity index (Ð) were determined by SEC analysis using polystyrene as standard and THF as eluent; c)The many amino groups in PDMAEMA are easily adsorbed on the SEC column. As a result, we were unable to determine the molecular weight and Ð of PB samples by SEC.

1700146 (4 of 10)





more than half an hour. For PA or PB, bubbling CO2 in aqueous solution may change the protonation degree of carboxylic acid or tertiary amine groups, which either decreases or increases the number of charges on the A blocks of PA and PB, respectively. The fact that no gelation occurs indicates that either of the two triblock copolymers remains soluble in water under CO2 stimulation. By contrast, their mixture in aqueous solution leads to gelation under CO2 stimulation. When PA and PB are mixed, the main effect of CO2 is to induce positive charges on the A blocks of PB, which then interact with the negative charges brought by the A blocks of PA. The polyFigure 2. Photographs of a) mixture of PA3 and PB3 (3:1 weight ratio) aqueous soluelectrolyte complexation among the A tion (total polymer concentration: 2 wt%) before and after 10 min CO2 bubbling (rate: blocks of the two constituent triblock 5 mL min−1) at 25 °C; b) Mixture of PA3 and PB3 (3:1 weight ratio) aqueous solution (5 wt%) copolymers thus leads to the gelation. before and after injection of 0.25 mL CO2 per mL solution at 25 °C; c) Mixture of PA3 and Importantly, as a result of the triblock PB3 (1:1 weight ratio) aqueous solution (5 wt%) after CO2 stimulation and then cooling or heating to different temperatures. In all cases, Rhodamine B dye (0.01 mg mL−1) was polymer structure and the efficient added to facilitate the visual observation of the state changes. cross-linking caused by the electrostatic interaction, the gelation can occur at a concentration as low as 2 wt%, and only an injection and neutral form, respectively. The two triblock copolyof 0.25 mL CO2 per mL solution can induce the gelation mers are soluble and no gel is formed. After CO2 bubbling, (Figure 2b). ≈92% of tertiary amine groups in PB3 were protonated, Compared with CO2-induced ABA triblock copolymer while only ≈8% of carboxylic acid groups in PA3 were protonated. The main effect of CO2 bubbling thus is to make gelation through LCST shift, which can form only above the LCST of A blocks, gelation based on CO2-triggered PMAA blocks in PA remain anionic and PDMAEMA blocks in PB become cationic, which leads to polyelectrolyte electrostatic interaction can be stable at relatively high complexation and gelation. temperatures. As shown in Figure 2c, even after more The zeta potential changes of PA and PB aqueous soluthan 10 min of N2 bubbling in order to remove CO2, the tions during CO2 bubbling were also measured, and the gel formed by the PA/PB mixture was stable at room temperature, at low temperature (5 °C), and only shrank a results in Figure 3b confirmed the above analysis. The little at 60 °C. However, at 80 °C, as the solubility of CO2 in zeta potential of PA3 changed from −47 to −32 mV after the CO2 bubbling, indicating that PA3 was still heavily water was very limited, most of the tertiary amino groups were deprotonated upon removal of CO2, which reduces negatively charged. While during this process, the zeta potential of PB3 increased from 6 to 40 mV, indicating the number of positive charges on the A blocks of PB and, rise in the number of positive charges on PB. As a result, consequently, results in dissociation of the gel. the electrostatic interaction between PA and PB greatly Under N2 bubbling while heating to 80 °C, the gel was increased during CO2 bubbling. Therefore, the gelation of dissociated within several minutes (Figure 2c). To further confirm that the CO2-induced gelation is driven by the mixed PA/PB aqueous solution is based on the electrostatic interaction of oppositely charged PMAA and CO2-triggered electrostatic interaction between oppoPDMAEMA blocks triggered by protonation of PDMAEMA sitely charged polyelectrolyte A blocks on the two triblock under CO2 bubbling. copolymers, the pH values of PA and PB aqueous solution during CO2 bubbling were monitored separately. As the We then investigated the effect of a number of parameters on the CO2-induced gelation of mixed PA/PB aqueous pKa values of PMAA and PDMAEMA are 5.5[25] and 7.0,[11b] respectively, changes in the protonation degrees of PA solution. As the gelation behavior is governed by elecand PB upon CO2 bubbling could be calculated from the trostatic interaction that is dependent on the molar ratio of positive and negative charges on PDMAEMA and pH values (see the Supporting Information). The results PMAA blocks, respectively, the weight ratio of PA to PB are shown in Figure 3a. As can be seen, at the initial pH should be an important factor. Taking PA3 and PB3 as an ≈9, both PMAA and PDMAEMA (the A blocks) are basiexample, different amounts of PA3 and PB3 were mixed cally deprotonated completely, existing in the anionic



(5 of 10) 1700146


B. Yu et al.

www.mcp-journal.de

9.0 80

8.5

PA3 (5 wt%) PB3 (5 wt%)

pH

8.0

60

7.5 40

7.0 6.5

20

6.0 5.5

Zeta potential (mV)

(b)

0

5

10 15 20 25 CO2 bubbling time (min)

30

Degree of protonation (%)

100

0

40 PA3 (0.5 wt %) PB3 (0.5 wt %)

20 0 -20 -40 0

1 2 3 4 CO2 bubbling time (min)

5

Figure 3. a) Variation of pH and protonation degree of PA3 and PB3 as a function of CO2 bubbling time (rate: 5 mL min−1) for their respective aqueous solution (5 wt%). b) Changes in zeta potential of the two solutions upon CO2 bubbling.

in aqueous solution at desired ratios while keeping the total polymer concentration constant from 1% to 10% by weight. The obtained results after 10 min of CO2 bubbling are presented in Figure 4 by indicating the regions where CO2-triggered gelation (sol–gel transition) was observed or not. As can be seen, the critical total polymer concentration for CO2-induced gelation is dependent upon the PA3/PB3 weight ratio. At a weight ratio between 4 and 2, gelation occurred at a low polymer concentration of 2 wt%. Having more or less PA3 in the mixture resulted in increase in the critical polymer concentration. However, at all investigated weight ratios, except the smallest one 0.33, CO2-induced gelation occurred with a total polymer concentration of 5 wt% or below, showing the effectiveness of the gelation mechanism based on polyelectrolyte complexation. The relationship between the gelation state and the molecular weight of PA or PB was also investigated. On one hand, either PA1 or PA2 or PA3 was mixed with the

1700146 (6 of 10)

10 Total Polymer Concentration (wt %)

(a) 9.5

8 Sol-gel transition after CO2 bubbling

6 4 2 0 10

No sol-gel transition after CO2 bubbling

5

2

1

0.5

Weight Ratio of PA and PB

Figure 4. Gelation state “phase diagram” showing the effect of PA3/PB3 weight ratio on the total polymer concentration required for gelation to occur after 10 min CO2 bubbling (rate: 5 mL min−1) at 25 °C. Red: sol–gel transition observed, and black: no sol–gel transition.

same PB sample, PB3, at a constant weight ratio of 1:1. The results in Figure 5a show that longer the PMAA blocks, more efficient is the CO2-induced gelation. With PA1 (with the shortest PMAA block), it took a total polymer concentration of 20 wt% to observe the gelation. By contrast, with PA3 (with the longest PMAA block), the gelation occurred even with 3 wt% of polymers. In the latter case, at higher polymer concentrations (≥15 wt%), the gelation was observed even before CO2 bubbling. This can be explained by the fact that at such a high polymer concentration, chain entanglements are expected for samples of high molecular weight such as PA3, which increases considerably the solution viscosity. Together with electrostatic interaction between existing positive charges on PDMAEMA blocks and anionic PMAA blocks, the gelation before CO2 bubbling is no surprise. On the other hand, mixing PA2 with either of the three PB samples, the results in Figure 5b show that the molecular weight of PB seems to exert little effect on the polymer concentration required for CO2 induced gelation. All of the above CO2-induced gelation experiments were carried out with 10 min CO2 bubbling at a rate of 5 mL min−1, and under this condition the PDMAEMA blocks of PB samples were fully protonated with positive charges (Figure 3). Giving that the electrostatic complexation between PMAA and PDMAEMA blocks is what induces the gelation, the amount of CO2 required for gelation for a given PA/PB mixture would vary depending on such parameters as PA/PB weight ratio and molecular weight of triblock copolymers. To gain more insight into the CO2-induced gelation process, experiments were conducted by injecting different amounts of CO2 into a sealed vial filled with 1 mL of a given PA/PB mixed





www.mcp-journal.de

Total Polymer Concentration (wt %)

(a) 20 Gelation before CO2 bubbling

15 Sol-gel transition after CO2 bubbling

10

5 No sol-gel transition after CO2 bubbling

PA1 PA2 PA3 PA with different molecular weight

Figure 6. Gelation state “phase diagram” showing the effect of PA3/PB3 weight ratio on the amount of CO2 required for gelation to occur at 25 °C (5 wt% polymer).

(a)

Volume of CO2 bubbling (mL)

Total Polymer Concentration (wt %)

(b) 20 Sol-gel transition after CO2 bubbling

15

10

5


PB1 PB2 PB3 PB with different molecular weight

aqueous solution at a polymer concentration of 5 wt%, and observing the gelation state. Figure 6 shows the results obtained with PA3 and PB3 at various weight ratios. As can be seen, the amount of CO2 required for gelation increased with decreasing the weight ratio, i.e., with increasing the amount of PDMAEMA blocks, which is no surprise. With PA3/PB3 = 3/1, only 0.25 mL of CO2 was enough to induce the gelation. By contrast, at the used polymer concentration (5 wt%), no gelation was observed for PA/PB = 1/3 even with 2 mL of CO2. Likewise, the effect of triblock copolymer molecular weight on the required amount of CO2 for gelation was also investigated under otherwise the same conditions. Presented in Figure 7a are the results obtained with PB3


Sol-gel transition after CO2 bubbling

4 3 2 1 0


PA1 PA2 PA3 PA with different molecular weight

(b)

Volume of CO2 bubbling (mL)

Figure 5. Gelation state “phase diagram” showing the effect of molecular weight of one triblock copolymer in the PA/PB mixture (1:1 weight ratio) on the total polymer concentration required for gelation to occur after 10 min CO2 bubbling (rate: 5 mL min−1) at 25 °C: a) Solutions of PB3 with PA of various molecular weights (PA1–PA3), and b) Solutions of PA2 with PB of different molecular weights (PB1–PB3).

5

2.0 Sol-gel transition after CO2 bubbling

1.5 1.0 0.5 0.0


PB1 PB2 PB3 PB with different molecular weight Figure 7. Gelation state “phase diagram” showing the effect of molecular weight of one triblock copolymer in the PA/PB mixture on the critical amount of CO2 required for gelation to occur at 25 °C: a) Solutions of PB3 with PA of various molecular weights (10 wt% polymers, PA:PB = 1:1), and b) Solutions of PA3 with PB of different molecular weights (10 wt% polymers, PA:PB = 1:3).


(7 of 10) 1700146


B. Yu et al.

www.mcp-journal.de

Figure 8. Schematic illustrations for: a) How the PA/PB weight ratio changes the number ratio of negative and positive charges after CO2 bubbling, and the effect on gelation; and b) How the molecular weight of negatively charged PA (PMAA block length) affects the complexation (cross-linking) with positively charged PB (PDMAEMA block) induced by CO2 bubbling.

mixed with PA samples of varying molecular weights (PA1–PA3) at a weight ratio of 1:1. The critical amount of CO2 increased sharply with decreasing the molecular weight of PA. While 0.5 mL of CO2 was sufficient for gelation with PA3, 5.0 mL was required for PA2, and no gelation was observed for PA1. The results obtained with PA3 mixed with PB samples of different molecular weights, shown in Figure 7b, are similar. But similar to the effect on the polymer concentration required for gelation (Figure 5b), the effect seems smaller than changing the molecular weight of PA. Considering the uncertainty in the values of triblock copolymer composition and molecular weight determined with 1H NMR and SEC, no calculation about the numbers of carboxylic acid and tertiary amine groups or negative and positive charges was provided for a given aqueous solution. Nevertheless, on the basis of the results by changing the PA/PB weight ratio and using samples of different molecular weights, some features can be qualitatively extracted about the present mechanism for triblock copolymer gelation through CO2-triggered electrostatic interaction. First, the effect of PA/PB weight ratio on the efficiency of CO2-induced gelation, as revealed by the minimum polymer concentration, reflects the necessary right balance of positive and negative charges in the system. As schematized in Figure 8a, with increasing the weight ratio of PA/PB, the number ratio of negative charges over positive charges after CO2 bubbling

1700146 (8 of 10)

increases. It is easy to picture that efficient complexation of the PMAA and PDMAEMA bocks on PA and PB, and thus gelation of the solution, should occur when the negative and positive charges have similar numbers, while a large excess in either positive charges (low PA/PB ratio) or negative charges (high PA/PB ratio) would only generate dangling polymer chains that are soluble in aqueous solution and no help for gelation. As a result, the most efficient gelation, with the lowest critical polymer concentration, is observed at a certain intermediate PA/PB weight ratio. However, one can imagine that the actual numbers of charges that interact to form the complex may be different from the total numbers. Therefore, the PA/PB weight ratio corresponding to the lowest critical polymer concentration does not necessarily mean the same numbers of positive and negative charges in the solution. In addition to the relative numbers of positive and negative charges, the relative numbers of PA and PB chains carrying those charges are also an important factor to consider, which accounts for the observed effect of molecular weight of PA or PB, especially PA, on the efficiency of CO2-induced gelation of the aqueous solution with mixed triblock copolymers. Figure 8b is a schematic illustration of a plausible explanation for the effect of molecular weight of PA. At a given PA/PB weight ratio, a higher molecular weight of PA means longer PMAA blocks, but in a smaller number. In other words, with increasing the molecular weight of PA, the number of negative charges





per PMAA block increases. Therefore, in the mixed PA and PB solution, upon CO2 bubbling, as PDMAEMA blocks of PB become positively charged, a longer PMAA block can interact with more PDMAEMA blocks through electrostatic interaction. This likely gives rise to stronger complexing domain with more connecting chains and acting as more stable cross-linking point for the network structure. In principle, any factors the favor effective complexing of the end blocks of the two triblock copolymers would help CO2-induced gelation.

Supporting Information Supporting Information is available from the Wiley Online Library or from the author.

4. Conclusions We have described a new approach to achieving CO2induced gelation of ABA-type triblock copolymers in aqueous solution. Instead of using CO2 to render the end A blocks insoluble in water through a shift of LCST as reported previously,[18] here CO2 is used to induce positive charges on the A blocks of an ABA triblock copolymer mixed with another ABA triblock copolymer whose A blocks bear negative charges. Gelation occurs as a result of the CO2-triggered electrostatic interaction between oppositely charged groups, i.e., complexation of the A blocks on the two constituent triblock copolymers. To demonstrate the validity of the approach, we synthesized two triblock copolymers, namely, PMAA-b-PEO-b-PMAA and PDMAEMA-b-PEOb-PDMAEMA, with the same water-soluble PEO middle block, PMAA for negatively charged end blocks prior to CO2 bubbling, and PDMAEMA for positively charged end blocks under CO2 stimulation. Despite the soft gels prepared using the two triblock copolymers due to the long middle block (PEO MW: 20 000 g mol−1), we were able to investigate the CO2-induced gelation behavior of the system by varying a number of parameters such as the weight ratio of the two triblock copolymers and their molecular weight. The obtained results suggested that an appropriate balance between the numbers of positive and negative charges as well as the numbers of chains carrying those charges were key to gelation. As compared to micelle-based gelation of ABA triblock copolymers, the present mechanism relying on electrostatic interaction has some significant advantages in terms of efficiency of using CO2 to induce gelation of triblock copolymers in aqueous solution. As showed, under certain preparation conditions, the total polymer concentration required for gelation could be as low as 2 wt%, and the necessary amount of CO2 could be very small. Moreover, gels obtained from electrostatic interaction could remain stable over a wider range of temperatures as compared to those gels based on LCST shift. Another feature, which has not been exploited in the present study but will be subject of future work, is that this approach provides much flexibility in designing CO2-responsive triblock copolymer gels. Not only the polyelectrolyte end blocks, like


PMAA, can be changed, but also the water-soluble middle block, like PEO, can be made to have different lengths for the two constituent triblock copolymers, or chosen to be different polymers. Using this approach, a large number of CO2-responsive or switchable hydrogels can be envisaged and offer possibilities for application as rheology modifier useful over a wide range of temperatures.

Acknowledgements: This work was financially supported by the Natural Sciences and Engineering Research Council of Canada (NSERC) and le Fonds de recherche du Quebec: Nature et technologies (FRQNT). B.Y. acknowledges the Merit Scholarship for Foreign Students awarded by FRQNT and the China Scholarship Council. Y.Z. is a member of the FRQNT-funded Center for SelfAssembled Chemical Structures (CSACS) and Centre québécoissur les matériauxfonctionnels (CQMF). Conflict of Interest: The authors declare no conflict of interest. Received: March 19, 2017; Revised: April Published online: ; DOI: 10.1002/macp.201700146

17,

2017;

Keywords: block copolymer gels; CO2-switchable polymers; stimuli-sensitive polymers

[1] a) A. S. Hoffman, Adv. Drug Delivery Rev. 2002, 54, 3; b) E. Ruel-Gariépy, J.-C. Leroux, Eur. J. Pharm. Biop harm. 2004, 58, 409; c) C. d. l. H. Alarcon, S. Pennadam, C. Alexander, Chem. Soc. Rev. 2005, 34, 276; d) N. A. Peppas, J. Z. Hilt, A. Khademhosseini, R. Langer, Adv. Mater. 2006, 18, 1345; e) C. He, S. W. Kim, D. S. Lee, J. Controlled Release 2008, 127, 189. [2] J. P. A. Fairclough, A. I. Norman, Ann. Rep. Prog. Chem., Sect. C: Phys. Chem. 2003, 99, 243. [3] a) Q. T. Pham, W. B. Russel, J. C. Thibeault, W. Lau, Macro molecules 1999, 32, 5139; b) M. E. Seitz, W. R. Burghardt, K. T. Faber, K. R. Shull, Macromolecules 2007, 40, 1218; c) K. A. Erk, K. R. Shull, Macromolecules 2011, 44, 932. [4] a) Y. Kitazawa, T. Ueki, L. D. McIntosh, S. Tamura, K. Niitsuma, S. Imaizumi, T. P. Lodge, M. Watanabe, Macromolecules 2016, 49, 1414; b) C. C. Hall, C. Zhou, S. P. O. Danielsen, T. P. Lodge, Macromolecules 2016, 49, 2298; c) B. Hu, W. Fu, B. Zhao, Macromolecules 2016, 49, 5502; d) S. E. Kirkland, R. M. Hensarling, S. D. McConaughy, Y. Guo, W. L. Jarrett, C. L. McCormick, Biomacromole cules 2008, 9, 481; e) T. Vermonden, N. A. M. Besseling, M. J. van Steenbergen, W. E. Hennink, Langmuir 2006, 22, 10180. [5] a) L. Lauber, O. Colombani, T. Nicolai, C. Chassenieux, Macro molecules 2016, 49, 7469; b) Y. Ma, Y. Tang, N. C. Billingham, S. P. Armes, A. L. Lewis, Biomacromolecules 2003, 4, 864. [6] a) R. Tamate, T. Ueki, Y. Kitazawa, M. Kuzunuki, M. Watanabe, A. M. Akimoto, R. Yoshida, Chem. Mater. 2016, 28, 6401; b)


(9 of 10) 1700146


B. Yu et al.

www.mcp-journal.de

T. Ueki, Y. Nakamura, R. Usui, Y. Kitazawa, S. So, T. P. Lodge, M. Watanabe, Angew. Chem., Int. Ed. 2015, 54, 3018. [7] J. Madsen, S. P. Armes, Soft Matter 2012, 8, 592. [8] H. Mao, G. Shan, Y. Bao, Z. L. Wu, P. Pan, Soft Matter 2016, 12, 4628. [9] a) Q. Yan, R. Zhou, C. Fu, H. Zhang, Y. Yin, J. Yuan, Angew. Chem., Int. Ed. 2011, 50, 4923; b) A. Darabi, P. G. Jessop, M. F. Cunningham, Chem. Soc. Rev. 2016, 45, 4391. [10] a) P. Liu, W. Lu, W.-J. Wang, B.-G. Li, S. Zhu, Langmuir 2014, 30, 10248; b) Q. Zhang, G. Yu, W.-J. Wang, B.-G. Li, S. Zhu, Macromol. Rapid Commun. 2012, 33, 916. [11] a) B. Yan, D. Han, O. Boissiere, P. Ayotte, Y. Zhao, Soft Matter 2013, 9, 2011; b) Q. Yan, Y. Zhao, J. Am. Chem. Soc. 2013, 135, 16300. [12] a) Z. Chu, C. A. Dreiss, Y. Feng, Chem. Soc. Rev. 2013, 42, 7174; b) H. Liu, W. Wang, H. Yin, Y. Feng, Langmuir 2015, 31, 8756; c) H. Liu, S. Lin, Y. Feng, P. Theato, Polym. Chem. 2017, 8, 12. [13] a) L. Lei, Q. Zhang, S. Shi, S. Zhu, Langmuir 2015, 31, 2196; b) M. Li, L. Lei, Q. Zhang, S. Zhu, Macromol. Rapid Commun. 2016, 37, 957. [14] a) H. Wei, J. Zhang, N. Shi, Y. Liu, B. Zhang, J. Zhang, X. Wan, Chem. Sci. 2015, 6, 7201; b) T. Huang, Z. Hou, Q. Xu, L. Huang, C. Li, Y. Zhou, Langmuir 2017, 33, 340; c) Q. Yan, W. Sang, Chem. Sci. 2016, 7, 2100. [15] H. Che, M. Huo, L. Peng, T. Fang, N. Liu, L. Feng, Y. Wei, J. Yuan, Angew. Chem., Int. Ed. 2015, 54, 8934. [16] a) S. Lin, P. Theato, Macromol. Rapid Commun. 2013, 34, 1118; b) Q. Yan, Y. Zhao, Chem. Commun. 2014, 50, 11631.

1700146 (10 of 10)

[17] a) Y. Zhang, Y. Feng, J. Wang, S. He, Z. Guo, Z. Chu, C. A. Dreiss, Chem. Commun. 2013, 49, 4902; b) C. Zhao, J. Zhang, G. Yuan, C. C. Han, RSC Adv. 2013, 3, 9645; c) Y. Zhang, Z. Chu, C. A. Dreiss, Y. Wang, C. Fei, Y. Feng, Soft Matter 2013, 9, 6217; d) D. Nagai, A. Suzuki, T. Kuribayashi, Macromol. Rapid Commun. 2011, 32, 404. [18] D. Han, O. Boissiere, S. Kumar, X. Tong, L. Tremblay, Y. Zhao, Macromolecules 2012, 45, 7440. [19] a) H. Lu, Z. Huang, L. L. Yang, S. Dai, J. Dispersion Sci. Technol. 2015, 36, 252; b) X. Su, M. F. Cunningham, P. G. Jessop, Chem. Commun. 2013, 49, 2655; c) Y. Zhang, Y. Feng, Y. Wang, X. Li, Langmuir 2013, 29, 4187. [20] a) Y. Zhang, Z. Zhou, Y. Shen, Q. Zhou, J. Wang, A. Liu, S. Liu, Y. Zhang, ACS Nano 2016, 10, 9036; b) J. Zhang, D. Xu, J. Guo, Z. Sun, W. Qian, Y. Zhang, F. Yan, Macromol. Rapid Commun. 2016, 37, 1194; c) D. Nagai, A. Suzuki, Y. Maki, H. Takeno, Chem. Commun. 2011, 47, 8856; d) T. Yu, K. Wakuda, D. L. Blair, R. G. Weiss, J. Phys. Chem. C 2009, 113, 11546. [21] J. W. Woodcock, R. A. E. Wright, X. Jiang, T. G. O’Lenick, B. Zhao, Soft Matter 2010, 6, 3325. [22] D. Han, X. Tong, O. Boissière, Y. Zhao, ACS Macro Lett. 2012, 1, 57. [23] E. Ceschia, J. R. Harjani, C. Liang, Z. Ghoshouni, T. Andrea, R. S. Brown, P. G. Jessop, RSC Adv. 2014, 4, 4638. [24] B. A. Abel, M. B. Sims, C. L. McCormick, Macromolecules 2015, 48, 5487. [25] V. Kozlovskaya, O. Zavgorodnya, J. F. Ankner, E. Kharlampieva, Macromolecules 2015, 48, 8585.

