Synthesis and self-assembly of ABC linear triblock copolymers to target CO₂-responsive multicompartment micelles

Abstract

Multicompartment micelles (MCMs) have advantage in medical applications because of their capability of transporting and releasing two incompatible compounds at one time. However, the achievement of stimulus-responsive MCMs, which is a necessary condition for controlled delivery, is rarely assessed. In this study, a series of ABC linear triblock copolymers were synthesized by a stepwise RAFT polymerization method starting from a macromolecular chain transfer agent containing poly(ethylene oxide) and using monomers of 2,2,3,4,4,4-hexafluorobutyl methacrylate and 2-(diethylamino)ethyl methacrylate to construct other two segments. The block lengths were tailored in order to achieve hierarchical CO₂-responsive MCMs. A morphological transition from spheres to MCMs under stimulation of CO₂ is finally observed with two copolymers among this series. The volume fraction of each block in the triblock copolymer was calculated and then depicted in a ternary phase diagram, in which a narrow composition window for the CO₂-responsive MCMs was suggested. These findings will guide the future design and fabrication of MCMs.

---

Introduction

The polymer self-assemblies have proved their versatile applications in various fields¹ such as nano-catalyst,^2,3 bio-mimicry,^4,5 clinical diagnosis⁶ and drug transportation.⁷ To meet a particular utilization, their morphologies should be tailored accordingly. Through unremitting efforts, the common self-assembly geometries involving spheres, vesicles and wormlike micelles have been widely reported and well understood based on the study of diblock copolymers.⁸ However, some hierarchical structures, such as Janus micelles and multicompartment micelles (MCMs) from triblock copolymers are seldom assessed, though they may be used as building blocks to construct higher-level aggregates if the micelles assembled by block copolymer are regarded as the first level. Furthermore, these hierarchical assemblies have bright future in medical applications such as drug delivery.⁹ Taking the MCMs as an example, they might load two or more incompatible medicines simultaneously in different micro-domains, transport to the same site and release at the same time.¹⁰

Given the appealing potential in therapy, a great deal of efforts were made into the fabrication of MCMs, yielding some significant understandings on how to achieve the MCMs. Lodge and co-workers^8,10–15 developed MCMs with miktoarm star triblock copolymers, which contain a water-soluble poly(ethylene oxide) segment, a polymeric hydrocarbon and a perfluorinated polyether. Müller's group^9,16–18 reported MCMs with various ABC linear triblock copolymers. Moreover, Laschewsky's team^19–24 prepared MCMs from ABC triblock copolymers comprising a hydrophilic (A), a lipophilic (B), and a fluorophilic (C) block. Based on these works, it is easy to obtain the first design criteria for fabricating MCMs, i.e., triblock copolymer is the preferred candidate. In addition, the phase segregation of hydro- and fluorocarbons is an effective method to gain the MCMs, which is demonstrated by Lodge⁸ and Laschewsky's group.^8,20 Although these design criteria were recognized, the fabrication of MCMs remains a challenge because of the limited composition window and possible kinetic obstacles.⁹ Furthermore, stimulus-responsive MCMs are rarely reported, except for the redox-responsive multicompartment vesicles developed by Zhang's group²⁵ and CO₂-responsive MCMs fabricated by our group,²⁶ although other stimulus-responsive multicompartment systems including hydrogel²⁷ and nanoparticles²⁸ are also reported.

Keeping the abovementioned design concepts in mind, our laboratory recently developed CO₂-switchable MCMs from a linear ABC triblock copolymer, O₁₁₃F₁₁₀E₂₂₀, composing of a water soluble portion of poly(ethylene oxide) (O), a fluorinated segment of poly(2,2,3,4,4,4-hexafluorobutyl methacrylate) (F) and a CO₂-responsive block of poly-(2-(diethylamino)ethyl methacrylate) (E) (Scheme 1).²⁶ In this polymer, the “O” block is used to stabilize the aggregates in an aqueous solution, the “F” segment works for phase segregation, whereas the “E” block aims to introduce CO₂-sensitiveness. CO₂ is a metabolite of cells with good biocompatibility and membrane-permeability, thus attracting intensive interests in recent years.^4,29–35 Therefore, these CO₂-responsive MCMs have good potential in biomedical applications.

Scheme 1 The synthesis route towards triblock copolymers O_xF_yE_z and schematic of their aggregates as well as morphological transition under stimulation of CO₂.

In this study, the synthesis and self-assembly of a series of OEF triblock copolymers are discussed, from which we try to understand why the copolymer O₁₁₃F₁₁₀E₂₂₀ with certain chain composition can aggregate into MCMs under the stimulation of CO₂ (Scheme 1) and whether we can achieve MCMs by regulating the block length of these copolymers. Finally, the narrow composition window of MCMs was suggested from a ternary phase diagram of these triblock copolymers (0.34 ≤ f_F ≤ 0.38, Scheme 1), which might guide our future design and development of MCMs.

Experimental

Materials

2-(Diethylamino)ethyl methacrylate (DEA, Aldrich, 99%) and 2,2,3,4,4,4-hexafluorobutyl methacrylate (FMA, Aldrich, 99%) were passed through an activated basic Al₂O₃ column to remove the inhibitors prior to use. 4-(Dimethylamino) pyridine (DMAP), N-(3-dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride crystalline (EDAC), 4,4′-azobis(4-cyanopentanoic acid) (ACPA), and poly(ethylene glycol) methyl ether (PEO₁₁₃, M_n ∼ 5 kg mol⁻¹; PEO₄₅, M_n ∼ 2 kg mol⁻¹) were purchased from Sigma-Aldrich and used as received. All solvents were purchased from Shanghai Chemical Reagent Co., Ltd. and used without further treatment unless otherwise specified. Deionized water (conductivity, κ = 12.8 μS cm⁻¹) used in the experiments was treated by an ultrapure water purification system (Chengdu Ultrapure Technology Co., Ltd., China).

The chain transfer agent (CTA) for reversible addition–fragmentation chain transfer (RAFT) polymerization, 4-cyano-4-thiothiopropylsulfanylpentanoic acid (CTPPA), was synthesised according to a reported procedure.^36,37

Characterization

¹H NMR spectra were obtained at 25 °C on a Bruker AV300 NMR spectrometer (300 MHz). The chemical shifts (δ) were reported in parts per million (ppm) with reference to the internal standard protons of tetramethylsilane (TMS). Dynamic light scattering (DLS) measurement was performed on a Malvern Zetasizer Nano-ZS90 apparatus equipped with a He–Ne laser operated at 633 nm at 25 °C. The scattering angle is fixed at 90° during the experiments. Transmission electron microscopy (TEM) observation was conducted on a Hitachi H600 electron microscope operated at an acceleration voltage of 75 kV. The specimens were prepared by placing one drop of the polymer aqueous solution (1.0 g L⁻¹) onto formvar-coated copper grids. Excess solvent was instantly absorbed by a filter paper. Afterwards, the samples were dipped into the phosphotungstic acid solution (0.2 wt%) and kept for approximately 30 s to stain the micelles. The number-average molecular weight (M_n) of polymers was determined by end group analyses based on ¹H-NMR integration of R-group and UV absorbance of the Z-group of chain transfer agents, the details of which will be described in the following section.

Synthesis of the macromolecular RAFT agent macro-PEO

The macro-PEO is synthesized by an esterification reaction catalysed by EDAC and DMAP. Dichloromethane (DCM) was dried with CaH₂ and then distilled to serve as a solvent for this reaction. A typical procedure is as follows: the chain transfer agent CTPPA (0.55 g, 2.0 mmol), PEO₁₁₃ (5.0 g, 1.0 mmol), EDAC (0.77 g, 4.0 mmol), and DMAP (0.24 g, 2.0 mmol) were dissolved in 50 mL dried DCM, then added into a 100 mL round bottom flask equipped with a magnetic bar and stirred for 48 h at room temperature after deoxygenating by bubbling Ar for 15 min. The reaction mixture was concentrated and precipitated at −70 °C in n-hexane (in a bath of acetone and dry ice mixture) three times, and washed with diethyl ether three times. Finally, a yellow powder was collected after lyophilisation.

Synthesis of the diblock copolymer PEO-b-PFMA

Using the macro-PEO as a macromolecular RAFT agent, the diblock copolymer PEO-b-PFMA (OF) was typically synthesized as follows, taking O₁₁₃F₆₀ as an example: the macro-PEO₁₁₃ (1.0 g, 0.19 mmol), the initiator ACPA (11 mg, 0.038 mmol), the monomers FMA (2.85 g, 11 mmol) and 3 mL of 1,4-dioxane were added into a reaction tube equipped with a magnetic bar, followed by three freeze–pump–thaw cycles. The mixture was reacted at 70 °C with stirring for 48 h. The polymerization was stopped by freezing the reaction tube into liquid nitrogen for more than 5 minutes. Then, the product was obtained after precipitation in n-hexane and lyophilized in a freezing dryer.

Synthesis of the triblock copolymer PEO-b-PFMA-b-PDEA

Using the diblock copolymer as a macromolecular RAFT agent, the triblock copolymer was typically synthesized as follows, taking O₁₁₃F₆₀E₁₂₀ as an example: the diblock copolymer O₁₁₃F₆₀ (1.0 g, ∼0.05 mmol, verified as O₁₁₃F₅₇ through NMR), ACPA (3 mg, 0.01 mmol), DEA (1.18 g, 6.4 mmol) and 4 mL of 1,4-dioxane were added into a reaction tube equipped with a magnetic bar. After removing oxygen by three freeze–pump–thaw cycles, the reaction mixture was stored at 70 °C for 48 h with stirring. The polymerization was terminated by freezing in liquid nitrogen. Finally, the product was precipitated in n-hexane and lyophilized in a freezing dryer.

Preparation of the micelle solutions

15 mg triblock copolymer was dissolved in 5 mL DMF and stirred for more than three hours, and then dialyzed against deionised water. Five days later, the solution was diluted to 15 mL, so the original micellar solution was obtained with a concentration of 1.0 g L⁻¹. TEM images and DLS tests are taken based on these micellar solutions. Before these tests, the micellar solutions were treated by CO₂ with flow rate 15 mL min⁻¹ for 30 min or by N₂ with same flow rate for 60 min, followed by sealing and equilibrating for 12 h at room temperature.

Results and discussion

Although some design concepts have been recognized, including the triblock structure and involvement of fluorinated segment into copolymers, it is still difficult to determine the block length of the triblock copolymers because the self-assembly morphologies of triblock copolymers could have several possibilities that are difficult to predict, as pointed out by Liu and his coworkers in their latest review.³⁸ Therefore, in order to target the MCMs, we synthesized a series of linear triblock copolymers with several variations. The polymerization route is shown in Scheme 1. From esterification of PEO and chain transfer agent CTPPA, we first produced a macro-initiator macro-PEO₁₁₃, then prepared “F” and “E” blocks by a two-step RAFT technique. To obtain the diblock copolymer, the degree of polymerization of “F” was designed as 60, 100, 180, 360, expecting to obtain copolymers O₁₁₃F₆₀, O₁₁₃F₁₀₀, O₁₁₃F₁₈₀ and O₁₁₃F₃₆₀, respectively. Then, the E block was changed successively to target a series of triblock copolymers. Subsequently, the chain length of PEO was transferred to 45, i.e., macro-PEO₄₅, to enrich this series. Table 1 gives the design and synthesis details of all polymers.

Table 1 Polymerization details with ACPA as initiator under 70 °C in dioxane for 48 h and calculation of f

No.	Polymer designed	RAFT agent (g)	Monomer (g)	ACPA^b (mg)	Product composition	Mn,NMR (kg mol^−1)	Mn,UV (kg mol^−1)	Yield (%)	fO	fF	fE
a “O” represents PEO block; “F” represents PFMA block; “E” represents PDEA block; the subscript is the designed polymerization degree.b [CTA or macro-CTA]/[ACPA] = 5.c Lost too much from post-polymerization treatment to calculate.
1	O113F60^a	Macro-PEO113, 1.0	FMA, 2.85	11	O113F57	19.5	19.1	78	—	—	—
2	O113F60E120	Entry 1, 1.0	DEA, 1.18	3	O113F57E55	29.7	29.8	61	0.19	0.45	0.36
3	O113F60E150	Entry 1, 1.0	DEA, 1.47	3	O113F57E114	40.6	40.7	40	0.13	0.32	0.54
4	O113F60E220	Entry 1, 0.5	DEA, 1.10	2	O113F57E121	41.9	40.1	15	0.13	0.31	0.56
5	O113F60E300	Entry 1, 0.5	DEA, 1.50	2	O113F57E201	56.8	52.5	27	0.09	0.23	0.68
6	O113F100	Macro-PEO113, 1.0	FMA, 4.75	11	O113F83	26.0	25.4	73	—	—	—
7	O113F100E120	Entry 6, 1.0	DEA, 0.96	2	O113F83E53	35.8	35.9	59	0.16	0.55	0.29
8	O113F100E150	Entry 6, 1.0	DEA, 1.20	2	O113F83E110	46.4	42.9	55	0.12	0.42	0.46
9	O113F100E220	Entry 6, 1.0	DEA, 1.75	2	O113F83E202	63.4	51.3	52	0.08	0.30	0.62
10	O113F100E300	Entry 6, 1.0	DEA, 2.39	2	O113F83E290	79.8	60.2	60	0.07	0.24	0.69
11	O113F180	Macro-PEO113, 1.0	FMA, 8.55	11	O113F110	32.8	32.0	90	—	—	—
12	O113F180E220	Entry 11, 1.0	DEA, 0.90	1	O113F110E192	68.3	50.1	54	0.08	0.37	0.55
13	O113F180E300	Entry 11, 1.0	DEA, 1.22	1	O113F110E212	72.0	54.7	50	0.08	0.35	0.57
14	O113F180E400	Entry 11, 1.0	DEA, 1.63	1	O113F110E281	84.8	59.1	30	0.06	0.30	0.64
15	O113F360	Macro-PEO113, 0.5	FMA, 8.55	5	O113F178	49.8	48.6	58	—	—	—
16	O113F360E300	Entry 15, 2.0	DEA, 1.95	2	O113F178E135	74.8	55.3	—^c	0.07	0.57	0.36
17	O45F30	Macro-PEO45, 0.5	FMA, 1.65	12	O45F28	9.0	9.1	78	—	—	—
18	O45F30E60	Entry 17, 0.3	DEA, 0.39	2	O45F28E42	16.8	16.2	—	0.13	0.39	0.48
19	O45F30E100	Entry 17, 0.3	DEA, 0.65	2	O45F28E55	19.1	19.2	—	0.11	0.33	0.56
20	O45F60	Macro-PEO45, 0.5	FMA, 3.30	12	O45F55	15.8	15.5	81	—	—	—
21	O45F60E80	Entry 20, 0.4	DEA, 0.40	1	O45F55E62	27.2	26.4	—	0.08	0.50	0.42
22	O45F60E150	Entry 20, 0.4	DEA, 0.75	1	O45F55E104	35.0	34.8	—	0.06	0.39	0.55

---

¹H NMR spectroscopy was used to confirm the chemical structure of the produced polymers. Taking the designed polymer O₁₁₃F₆₀E₁₂₀ as an example, only 3.63 ppm, which belongs to the chemical shift of the protons in “O” chain (–CH₂CH₂O–), is observed in the spectrum of macro-PEO₁₁₃ (upper, Fig. 1). In the designed diblock copolymer O₁₁₃F₆₀, the chemical shifts at 4.73–5.09 ppm (–CHFCF₃) and 4.13–4.62 ppm (–COOCH₂CF₂–) correspond to the “F” block appear (middle, Fig. 1), indicating the introduction of “F” block. After the second RAFT polymerization, the characteristic peak at 3.92–4.10 ppm (–COOCH₂CH₂N(CH₂CH₃)₂) corresponding to “E” segment appears (bottom, Fig. 1), supporting the production of the triblock copolymer. According to the integration, the degree of polymerization can be calculated according to the following equations:

where δ_a, δ_b and δ_d represent the integration of peak a, b and d, respectively; DP_O, DP_F and DP_E are the polymerization degrees of the block “O”, “F” and “E”, respectively. Thus, the real structure of the product is O₁₁₃F₅₇E₅₅ rather than O₁₁₃F₆₀E₁₂₀, indicating that the conversion of DEA is very low. The effort to improve the conversion by increasing the reaction time finds no improvement. Even so, this conversion is comparable with the previous report by Laschewsky's group when the fluorinated polymers work as macro-initiators (<60%).²⁴ One reason might be the lipophobic feature of “F” chain, which decreases the reactivity of diblock macro-initiators. The other reason might be the low concentration of DEA monomers in the second RAFT reaction because the diblock micro-initiators need much solvent to dissolve.

Fig. 1 ¹H NMR spectra of the diblock precursor O₁₁₃F₆₀ and corresponding triblock copolymer O₁₁₃F₆₀E₁₂₀.

From the ¹H NMR spectra analyses, one can figure out the molecular structure and the number-average molecular weight (M_n) of all polymers in this series (Table 1). This analysis actually is based on the R group (PEO chain here) of the CTA. Nevertheless, gel permeation chromatography (GPC) might not be a preferred method here because of the possible complex mutual interactions in triphilic polymers, which make the polymer difficult to be molecularly dispersed in the eluent.²⁴ Laschewsky and co-workers studied the end-group analysis based on Z-group of CTA via UV-vis spectroscopy, finding it a good alternative method.³⁹ Herein, we also investigate the M_n based on the UV absorbance of Z-group (Fig. S1, see test details in ESI†). The results are shown in Table 1. Compared with the results from ¹H NMR spectra, they are comparable when the M_n is lower than 40 kg mol⁻¹. However, the M_n data from UV procedure are lower than that from NMR when the M_n is higher than 40 kg mol⁻¹. This mismatch might be caused by the possible loss of Z-groups at the end of polymers.²⁴ Moreover, this impact might become more significant for polymers with higher M_n.

To confirm the CO₂-responsiveness of the triblock copolymer in an aqueous solution, DLS was used to test the hydrodynamic diameter (D_h) of the self-assembles of O₁₁₃F₅₇E₅₅ before and after treatment with CO₂. Before bubbling CO₂, the D_h of the aggregate is approximately 115 nm, but it increases to 295 nm upon the stimulation of CO₂ (Table 2, entry 1 and Fig. 2a). Interestingly, the D_h returns back to around 106 nm after removing CO₂ by bubbling N₂ into the solution, indicating the size expansion under the stimulus of CO₂ is reversible. Subsequently, TEM was employed to visualize the morphology of the aggregate. As shown in Fig. 2b, the assemblies appear as spherical micelles with light core inside (hydrophobic domain) and darker shell outside (hydrophilic domain) that is stained by hydrophilic phosphotungstic acid. The micelle has an average diameter of 80 nm, which is lower than that of DLS data caused by the drying procedure during TEM sample preparation and invisible PEO coronas in TEM image.⁸ However, beyond our expectation, the aggregates remain as spheres except for a slight size expansion to around 95 nm after the treatment of CO₂, as shown in Fig. 2c, implying no morphological transformation under stimulation. Afterwards, it was found that the aggregates of O₁₁₃F₅₇E_x (x = 55, 114, 121, 201) series are all spherical micelles without significant morphological change before (a1, b1, c1 of Fig. S2, ESI†) and after bubbling CO₂ (a2, b2, c2 of Fig. S2, ESI†) rather than MCMs, though some of them show collapsed vesicles such as O₁₁₃F₅₇E₁₁₄ after treatment of CO₂ (a2 of Fig. S2, ESI†).

Table 2 DLS data of aggregates from some triblock copolymers

Polymer	Before bubbling CO2		After bubbling CO2		After removing CO2
	Dh (nm)	PDI	Dh (nm)	PDI	Dh (nm)	PDI
O113F57E55	115	0.08	295	0.07	106	0.09
O113F83E202	141	0.07	328	0.11	122	0.10
O113F110E212	122	0.09	396	0.02	141	0.06

---

Fig. 2 DLS data (a) and TEM image ((b) before bubbling CO₂; (c) after bubbling CO₂) of triblock copolymer O₁₁₃F₅₇E₅₅ (corresponding to the designed O₁₁₃F₆₀E₁₂₀) in an aqueous solution. The polymer concentration is fixed at 1.0 g L⁻¹. The specimens for TEM were stained with 0.2 wt% phosphotungstic acid aqueous solution.

Then, the aggregates of O₁₁₃F₈₃E_x (x = 53, 110, 202, 290) series were also checked, showing as spherical micelles without morphological change before and after the stimulation of CO₂. Taking O₁₁₃F₈₃E₂₀₂ as an example, the aggregates have diameter of 141 nm and then expand to 328 nm under stimulation of CO₂ (Table 2, entry 2); however, the self-assemble morphology remains as sphere (d1 and d2 of Fig. S2, ESI†). Fortunately, a different trend arises in the series O₁₁₃F₁₁₀E_x (x = 192, 212, 281). The aggregates of polymer O₁₁₃F₁₁₀E₂₁₂ can transform from spheres (Fig. 3c) to MCMs in which the “hamburgers” and “reverse hamburgers” are easily distinguished (Fig. 3d). The D_h of the aggregate of O₁₁₃F₁₁₀E₂₁₂ increases from 122 to 396 nm after exposure to CO₂ and recovers to around 141 nm when CO₂ is removed by bubbling N₂ (Table 2, entry 3). The polydispersity (PDI) of the MCMs is 0.02 (Table 2 entry 3), which is quite narrow compared with reported data (0.02–0.15).²⁴ This size expansion should be ascribed to the protonation of “E” block and chain stretching under the electrostatic repulsion.³³ We should point out that the formation of CO₂-switchable MCMs from this polymer (O₁₁₃F₁₁₀E₂₁₂) has been demonstrated in our previous work using a bunch of characterization techniques, including TEM with two staining methods, cryogenic transmission electron microscopy (cryo-TEM), atomic force microscopy (AFM), and scanning electron microscopy (SEM) as well as dissipative particle dynamics (DPD) simulation.²⁶ Furthermore, the mechanism was also interpreted.²⁶ Briefly, in the absence of CO₂, the phase segregation of the “O”, “E” and “F” block is incomplete, thus resulting in spheres; in the presence of CO₂, the phase segregation is targeted because of the protonation of “E” block, thus yielding MCMs. Interestingly, another polymer O₁₁₃F₁₁₀E₁₉₂ in this series, possessing same chain length of F block, appears as spherical micelle in water without CO₂ (Fig. 3a) but transforms into MCMs, including “hamburgers” and “clovers”, under the treatment of CO₂ (Fig. 3b, see more TEM images from Fig. S3†). However, when the chain length of “E” block increases to 281, i.e., O₁₁₃F₁₁₀E₂₈₁, majority of the aggregates are spherical even under treatment of CO₂ (Fig. 3e and f) without any trace of MCMs (the reason will be discussed further down). Subsequently, our effort of increasing the fluorinated block “F” produces polymer O₁₁₃F₁₇₈E₁₃₅, but yielding no MCMs under the stimulation of CO₂ (e1 and e2 of Fig. S2, ESI†). We also changed the starting macro-initiator to PEO₄₅ and prepared O₄₅F_xE_y (x = 28, y = 42, 55 and x = 55, y = 62, 104) series, finding no MCMs in their aqueous solution (f1 and f2 of Fig. S2, ESI†).

Fig. 3 TEM images of triblock copolymer O₁₁₃F₁₁₀E₁₉₂ (a and b), O₁₁₃F₁₁₀E₂₁₂ (c and d) and O₁₁₃F₁₁₀E₂₈₁ (e and f) in an aqueous solution. (a), (c) and (e) are the samples before CO₂ bubbling; (b), (d), and (f) are the samples after CO₂ treatment. The numbers in the image indicate different types of MCMs: (1) “hamburgers”, (2) “clovers”, (3) “reverse hamburgers”.

Based on abovementioned results, herein, we try to comprehend what type of polymer structures tend to aggregate into MCMs. For triblock copolymers, the ternary phase diagram is a good tool to understand the morphology–molecular structure relationship.^11,40 Thus, the volume fraction (f) was calculated (Table 1) and then depicted in a ternary phase diagram (Fig. 4). The calculation of f is obtained according to the following equation, taking f_O as an example:

where ρ_O = 1.15 g mL⁻¹,³³ ρ_E = 1.19 g mL⁻¹,³³ ρ_F = 1.35 g mL⁻¹ (using density of FMA monomers supplied by Sigma-Aldrich); M_O = 5000 g mol⁻¹, M_F = DP_F × M_FMA, M_E = DP_E × M_DEA; M_FMA = 250.4 g mol⁻¹, M_DEA = 185.26 g mol⁻¹.

Fig. 4 Ternary phase diagram for the triblock copolymers in an aqueous solution as a function of composition. f_O, f_F and f_E are the volume fractions of the O, F and E blocks, respectively.

In our research scale, the values of f_O are small and lie in a narrow range between 0.06 and 0.19 since the O block used here is very short (M_O = 5 kg mol⁻¹ or 2 kg mol⁻¹). f_F and f_E have wider variation (0.24 ≤ f_F ≤ 0.57, 0.36 ≤ f_E ≤ 0.69) so their values should provide more opportunities for the formation of MCMs and thus will be discussed more. As shown in Fig. 4, the two polymers O₁₁₃F₁₁₀E₁₉₂ (f_O = 0.08, f_F = 0.37, f_E = 0.55) and O₁₁₃F₁₁₀E₂₁₂ (f_O = 0.08, f_F = 0.35, f_E = 0.57), which target MCMs appear as red stars in the diagram. First, we found that four different compositions (two blue triangles and two black squares) share similar f_E value with the two red stars between f_E = 0.54 and 0.58 (in the region pointed by triple arrows), but no MCMs were observed in their aqueous solution, implying f_E is not the determining factor for the formation of MCMs. Then, checking along the f_F axis, one can find an extremely narrow region (pointed by double arrows) from approximately f_F = 0.34 to 0.38, which is occupied only by the two stars, i.e., polymers O₁₁₃F₁₁₀E₁₉₂ and O₁₁₃F₁₁₀E₂₁₂, suggesting that the formation of MCMs depends on the f_F value. The finding of this composition window (f_F = 0.34 to 0.38, pointed by double arrows) help us understand why only two polymers can transform to MCMs. More importantly, it might be helpful for the future development of MCMs. The rigid and super-hydrophobic feature of “F” block may account for the crucial roll of f_F for the formation of MCMs, which is helpful for the phase segregation in self-assemblies.⁴¹

Conclusions

A series of linear triblock copolymers were designed and synthesized using a stepwise control radical polymerization method. Their aggregates show size expansion under the stimulation of CO₂ and recovers after removing CO₂ by bubbling N₂. However, the morphologies for most polymers remain spherical even under the exposure of CO₂. Only triblock copolymers in a narrow composition window (0.34 ≤ f_F ≤ 0.38) transform from spherical micelles to MCMs after reaction with CO₂, including O₁₁₃F₁₁₀E₁₉₂ and O₁₁₃F₁₁₀E₂₁₂. This composition window might help us pave a way for the design and preparation of MCMs in the future. Considering the medical potential of MCMs⁴² and the fine biocompatibility of CO₂,^43–46 these CO₂ responsive MCMs might seek potential in therapeutic applications.

Acknowledgements

The authors would like to thank the financial support from the National Natural Science Foundation of China (21273223) and the open funding from State Key Laboratory of Polymer Materials Engineering (sklpme2014-2-06).

References

1. Y. Mai and A. Eisenberg, Chem. Soc. Rev., 2012, 41, 5969–5985.
2. D. M. Vriezema, P. M. Garcia, N. Sancho Oltra, N. S. Hatzakis, S. M. Kuiper, R. J. Nolte, A. E. Rowan and J. van Hest, Angew. Chem., Int. Ed., 2007, 39, 7378–7382.
3. P. Broz, S. Driamov, J. Ziegler, N. Ben-Haim, S. Marsch, W. Meier and P. Hunziker, Nano Lett., 2006, 6, 2349–2353.
4. Q. Yan and Y. Zhao, J. Am. Chem. Soc., 2013, 135, 16300–16303.
5. H.-J. Choi and C. D. Montemagno, Nano Lett., 2005, 5, 2538–2542.
6. C. Sanson, O. Diou, J. Thévenot, E. Ibarboure, A. Soum, A. Brûlet, S. Miraux, E. Thiaudière, S. Tan, A. Brisson, V. Dupuis, O. Sandre and S. Lecommandoux, ACS Nano, 2011, 5, 1122–1140.
7. F. Ahmed, R. I. Pakunlu, A. Brannan, F. Bates, T. Minko and D. E. Discher, J. Controlled Release, 2006, 116, 150–158.
8. Z. B. Li, E. Kesselman, Y. Talmon, M. A. Hillmyer and T. P. Lodge, Science, 2004, 306, 98–101.
9. A. H. Gröschel, F. H. Schacher, H. Schmalz, O. V. Borisov, E. B. Zhulina, A. Walther and A. H. E. Müller, Nat. Commun., 2012, 3, 710.
10. T. P. Lodge, A. Rasdal, Z. B. Li and M. A. Hillmyer, J. Am. Chem. Soc., 2005, 127, 17608–17609.
11. Z. B. Li, M. A. Hillmyer and T. P. Lodge, Langmuir, 2006, 22, 9409–9417.
12. Z. B. Li, M. A. Hillmyer and T. P. Lodge, Macromolecules, 2006, 39, 765–771.
13. C. Liu, M. A. Hillmyer and T. P. Lodge, Langmuir, 2009, 25, 13718–13725.
14. C. I. Huang, C. H. Liao and T. P. Lodge, Soft Matter, 2011, 7, 5638–5647.
15. A. O. Moughton, M. A. Hillmyer and T. P. Lodge, Macromolecules, 2012, 45, 2–19.
16. B. Fang, A. Walther, A. Wolf, Y. Xu, J. Yuan and A. H. E. Müller, Angew. Chem., Int. Ed., 2009, 48, 2877–2880.
17. F. Schacher, A. Walther, M. Ruppel, M. Drechsler and A. H. E. Muller, Macromolecules, 2009, 42, 3540–3548.
18. A. Walther, C. Barner-Kowollik and A. H. E. Muller, Langmuir, 2010, 26, 12237–12246.
19. A. Laschewsky, Curr. Opin. Colloid Interface Sci., 2003, 8, 274–281.
20. S. Kubowicz, J.-F. Baussard, J.-F. Lutz, A. F. Thünemann, H. von Berlepsch and A. Laschewsky, Angew. Chem., Int. Ed., 2005, 44, 5262–5265.
21. H. v. Berlepsch, C. Bottcher, K. Skrabania and A. Laschewsky, Chem. Commun., 2009, 2290–2292.
22. K. Skrabania, A. Laschewsky, H. von Berlepsch and C. Bottcher, Langmuir, 2009, 25, 7594–7601.
23. K. Skrabania, H. von Berlepsch, C. Bottcher and A. Laschewsky, Macromolecules, 2010, 43, 271–281.
24. J. N. Marsat, M. Heydenreich, E. Kleinpeter, H. V. Berlepsch, C. Bottcher and A. Laschewsky, Macromolecules, 2011, 44, 2092–2105.
25. P. Shi, Y. Qu, C. Liu, H. Khan, P. Sun and W. Zhang, ACS Macro Lett., 2015, 5, 88–93.
26. H. Liu, Y. Zhao, C. A. Dreiss and Y. Feng, Soft Matter, 2014, 10, 6387–6391.
27. C. Zhou, M. A. Hillmyer and T. P. Lodge, J. Am. Chem. Soc., 2012, 134, 10365–10368.
28. P. Shi, Q. Li, X. He, S. Li, P. Sun and W. Zhang, Macromolecules, 2014, 47, 7442–7452.
29. Q. Yan, R. Zhou, C. Fu, H. Zhang, Y. Yin and J. Yuan, Angew. Chem., Int. Ed., 2011, 50, 4923–4927.
30. Q. Yan, J. Wang, Y. Yin and J. Yuan, Angew. Chem., Int. Ed., 2013, 52, 5070–5073.
31. Z. R. Guo, Y. J. Feng, S. He, M. Z. Qu, H. L. Chen, H. B. Liu, Y. F. Wu and Y. Wang, Adv. Mater., 2013, 25, 584–590.
32. P. G. Jessop, S. M. Mercer and D. J. Heldebrant, Energy Environ. Sci., 2012, 5, 7240–7253.
33. H. Liu, Z. Guo, S. He, H. Yin, C. Fei and Y. Feng, Polym. Chem., 2014, 5, 4756–4763.
34. H. Liu, W. Wang, H. Yin and Y. Feng, Langmuir, 2015, 31, 8756–8763.
35. Z. R. Guo, Y. J. Feng, Y. Wang, J. Y. Wang, Y. F. Wu and Y. M. Zhang, Chem. Commun., 2011, 47, 9348–9350.
36. A. J. Convertine, D. S. W. Benoit, C. L. Duvall, A. S. Hoffman and P. S. Stayton, J. Controlled Release, 2009, 133, 221–229.
37. X. W. Xu, A. E. Smith, S. E. Kirkland and C. L. McCormick, Macromolecules, 2008, 41, 8429–8435.
38. I. W. Wyman and G. Liu, Polymer, 2013, 54, 1950–1978.
39. K. Skrabania, A. Miasnikova, A. M. Bivigou-Koumba, D. Zehm and A. Laschewsky, Polym. Chem., 2011, 2, 2074–2083.
40. F. Schacher, J. Yuan, H. G. Schoberth and A. H. Müller, Polymer, 2010, 51, 2021–2032.
41. H. Ito, T. Imae, T. Nakamura, M. Sugiura and Y. Oshibe, J. Colloid Interface Sci., 2004, 276, 290–298.
42. A. H. Gröschel and A. H. Müller, Nanoscale, 2015, 7, 11841–11876.
43. Q. Yan and Y. Zhao, Angew. Chem., Int. Ed., 2013, 52, 9948–9951.
44. S. Lin and P. Theato, Macromol. Rapid Commun., 2013, 34, 1118–1133.
45. J. Y. Quek, P. J. Roth, R. A. Evans, T. P. Davis and A. B. Lowe, J. Polym. Sci., Part A: Polym. Chem., 2013, 51, 394–404.
46. H. Yin, A.-L. Bulteau, Y. Feng and L. Billon, Adv. Mater. Interfaces, 2016, 3, 1500623.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra18826e

---