Synthesis via RAFT polymerization of thermo-and pH-responsive random copolymers containing cholic acid moieties and their self-assembly in water †

Polymer Research Centre, Department of Che Education and Research-Kolkata, PO: BCKV West Bengal, India. E-mail: p_de@iiserkol.a † Electronic supplementary information for MAECA, NMR characterization of m copolymers, FT-IR and ESI-MS spectra, G and TGA thermograms of P(MAECA-co-DM of CD2 at different pHs, uorescenc determination plots, SEM images, and so series. See DOI: 10.1039/c3py01317k Cite this: Polym. Chem., 2014, 5, 1275


Introduction
Bile acids are cholesterol-derived naturally occurring amphiphilic molecules, which are essential for digestion and absorption of fats and fat-soluble vitamins in the small intestine to adjust cholesterol levels in the body. 1 Bile acid derivatives have been used for cholesterol gallstone dissolution, 2 treating primary biliary cirrhosis, 3 antiviral and antifungal properties, gene therapy, radiopharmaceuticals, etc. 4 Supramolecular gels from salts of bile acid derivatives have immense application potential in biomedical and materials science. 5 Driven by the aim to develop bile acid containing polymeric materials, a great deal of research has been dedicated to the incorporation of bile acid residues as cores, pendant groups and end groups, and also in the backbone via polycondensation reactions. 6 For preparation of side-chain bile acid containing polymers, most of the reports used the conventional free radical polymerization (FRP) technique, where uncontrolled molecular weights and broad molecular weight distributions (polydispersity index, PDI) were obtained. Recently, much attention has been devoted to the controlled radical polymerization (CRP) 7 of bile acid based vinyl monomers, where it is possible to achieve targeted molecular weights, narrow molecular weight distributions and well-dened architectures. 8 A comb-shaped cholic acid containing polymer, poly[methacryloyl tri(ethylene glycol) cholanoate], was synthesized by atom transfer radical polymerization (ATRP). 9 Novel biomimetic surfactants based on cholesterol as the hydrophobic head group and poly [2-(methacryloyloxy)ethyl phosphorylcholine] as the hydrophilic segment were prepared by ATRP and their micellar behaviour in an aqueous medium was investigated. 10 A series of multishape memory copolymers with a polynorbornene main chain, and cholic acid and triethylene glycol side groups were synthesized through ring opening metathesis polymerization (ROMP). 11 The glass transition temperature (T g ) of these copolymers was tunable over a temperature range from À58 to 176 C.

Instruments and analysis
The molecular weights and PDIs of polymers were measured by gel permeation chromatography (GPC) using poly(methyl methacrylate) (PMMA) calibration in DMF at 35 C. The GPC system consists of a Waters 515 HPLC pump, Waters 2414 RI detector, and two HSPgel columns. The 1 H NMR study was conducted on a Bruker Avance III 500 spectrometer. Positive mode electrospray ionization mass spectrometry (ESI-MS) was performed on a Q-Tof Micro YA263 high resolution mass spectrometer (Waters Corporation). Hydrodynamic diameter values were determined using a dynamic light scattering (DLS) (Zetasizer Nano ZS, Malvern Instrument Ltd., UK) instrument equipped with a He-Ne laser beam at 658 nm. Polymer solutions (1.0 mg mL À1 ) were ltered through a 0.45 mm syringe lter prior to measurement and each experiment was repeated four times to obtain the average value. UV-vis measurements were done on a Perkin-Elmer Lambda 35 UV-vis spectrophotometer. Emission spectra were recorded using a Horiba Jobin Yvon Fluoromax-3 instrument. The FT-IR spectrum was recorded on KBr pellets using a Perkin-Elmer Spectrum 100 FT-IR spectrometer. The T g values were measured using a Mettler Toledo DSC1 STARe differential scanning calorimeter (DSC) at a 10 C min À1 heating rate in a nitrogen atmosphere. Polymers were cooled from room temperature to À70 C, then heated to +120 C and again cooled to À70 C at 10 C min À1 . The T g was taken from the third segment of the experiment. Thermogravimetric analysis (TGA) was carried out using a Mettler Toledo TGA/SDTA 851e instrument at a heating rate of 10 C min À1 in a nitrogen atmosphere. Atomic force microscopy (AFM) images were recorded on an NT-MDT NTEGRA Prima Scanning Probe Microscope operated in semi-contact mode. An aliquot of sample solution (0.1 mg mL À1 ) was drop cast on a glass slide at room temperature, dried and analyzed by AFM. Scanning electron microscopy (SEM) measurements were performed on a Carl Ziess Sigma SEM instrument.

Monomer synthesis
MAECA was synthesised by esterication condensation reaction of CA and HEMA in the presence of DCC and DMAP (Scheme S1 †). The coupling reaction was carried out in dry THF under a dry N 2 atmosphere. DCC (7.57 g, 36.7 mmol) in 20 mL THF was placed in a 250 mL double necked round bottom ask containing a magnetic stir bar. To this solution, rst CA (15.0 g, 36.7 mmol) in 80 mL THF and then DMAP (0.45 g, 3.67 mmol) in 20 mL THF were added under a N 2 atmosphere. Then, the system was placed in an ice-water bath and HEMA (4.77 g, 36.7 mmol) was added dropwise to the reaction mixture under stirring. The ice-water bath was removed aer 30 min and the reaction mixture was stirred for 28 h at room temperature. The reaction mixture was ltered to separate insoluble N,N 0 -dicyclohexylurea (DCU) and the ltrate was concentrated by rotary evaporation to remove THF. Then, it was dissolved in ethyl acetate and washed with 0.1 N HCl (100 mL Â 3), saturated NaHCO 3 (100 mL Â 3) and brine solution (100 mL Â 2). The organic layer was dried over anhydrous Na 2 SO 4 and the solvent was removed by rotary evaporation. Finally, it was puried by column chromatography using hexanes-ethyl acetate as the mobile phase (2 : 3, v/v) with a yield of 50%. 1

RAFT polymerization of MAECA
Typically, MAECA (0.20 g, 0.38 mmol), CTP (4.28 mg, 15.4 mmol), AIBN (0.504 mg, 3.07 mmol; from the stock solution of AIBN in DMF) and 0.8 g DMF were taken in a 20 mL septa sealed glass vial equipped with a magnetic stir bar. The reaction vial was purged with dry N 2 for 20 min and placed in a preheated reaction block at 70 C. Aer a predetermined time, the polymerization was quenched by cooling in an ice-water bath and exposed to air. The polymer (PMAECA) was precipitated using hexanes. The polymer was redissolved in THF and precipitated from hexanes and this process was repeated 5 times. The precipitate was dried under high vacuum at 40 C for 10 h. Monomer conversions were calculated gravimetrically based on the monomer feed.

Synthesis of statistical copolymers
A typical copolymerization procedure was as follows: MAECA (0.190 g, 0.365 mmol), PEGMA (0.110 g, 0.365 mmol), CTP (3.40 mg, 12.2 mmol), AIBN (0.40 mg, 2.43 mmol; 0.08 g solution of 10.4 mg AIBN in 2.09 g DMF) and DMF (1.2 g) were placed in a 20 mL septa sealed vial equipped with a magnetic stir bar, purged with dry N 2 for 20 min and placed in a preheated reaction block at 70 C for 5 h. The feed ratios of monomers were varied to obtain copolymers of various compositions. Copolymerization reactions were quenched and copolymers were puried as described for the homopolymers.

Determination of LCST
The turbidity of copolymer aqueous solutions (2 mg mL À1 ) was measured using a UV-vis spectrophotometer by monitoring the % transmittance (%T) changes at l ¼ 500 nm. 22 The sample was ltered through a 0.45 mm membrane lter, transferred to a quartz cuvette, placed in a UV-vis spectrometer and warmed from 20 to 80 C. At the measurement temperature the sample was equilibrated for 6 min and the %T value at 500 nm was determined. The LCST was dened as the temperature where a reduction of 50% T of the polymer solution was observed.

Determination of critical aggregation concentration (CAC) by uorescence spectroscopy 23
Pyrene was used as a hydrophobic uorescent probe to measure the CAC of copolymers in aqueous medium. A stock pyrene solution in acetone was added to different 50 mL volumetric asks. Then, 4.0 mL of aqueous solution containing different concentrations of copolymers was added to the asks to make the nal pyrene concentration in each ask 10 À6 mol L À1 . These pyrene containing asks were kept open for 24 h at room temperature for the complete evaporation of acetone. Fluorescence spectra of these solutions were recorded (at excitation wavelength 339 nm) from 360 to 450 nm. The uorescence intensity ratio of I 383 /I 373 was plotted against the polymer concentration. The CAC was obtained from the intersection of two tangent plots of the intensity ratio I 383 /I 373 versus the polymer concentration.

Synthesis of MAECA
MAECA was synthesised by a single step coupling reaction of commercially available CA and HEMA in the presence of DCC and DMAP at room temperature (Scheme S1 †). The molecular structure of MAECA was identied by 1

RAFT polymerization of MAECA
CTAs have an obvious inuence during the controlled RAFT polymerization of vinyl monomers. It was reported that both trithiocarbonates and dithioesters can be used for the RAFT polymerization of methacrylates with good control over both molecular weight and PDI. 24   where number average molecular weights determined from GPC (M n,GPC ) are much higher compared to the theoretical molecular weights (M n,theo ) based on monomer conversion. Also, for these two CTAs we noticed broad molecular weight distributions. The polymers from the reactions with CDP and CTP exhibited narrow PDIs (<1.35) and the M n,GPC values match well with the corresponding M n,theo values. We selected CTP for further investigations because with this CTA, it was easy to determine number average molecular weights by NMR (M n,NMR ) from the comparison of phenyl chain end protons to the main chain protons (vide infra).
To  Table  1). Note that in the high molecular weight region of the GPC traces, a small shoulder was observed for all the PMAECA homopolymers, which could be due to the partial recombination of propagating polymer chains. Also, formation of aggregates during the GPC experiments in polar DMF solvent may give rise to a shoulder in higher molecular weight regions. The inuence of the initiator concentration was also studied in which the molar feed ratio of CTP to initiator was varied from 1 : 0.1 to 1 : 0.3, while the [MAECA]/[CTP] ratio was maintained at 25 : 1. Generally, the rate of polymerization depends on the availability of radicals for initiation and Table 1 shows that the polymerization rate increases with decreasing [CTP]/[AIBN] ratio, as expected.
The structure of PMAECA was conrmed by 1 H NMR spectroscopy. In the 1 H NMR spectrum of PMAECA in Fig Table 1, which match reasonably well with the M n,theo values indicating PMAECA chain ends with the fragments of CTP as expected according to the typical RAFT mechanism.
To further demonstrate the retention of dithioester CTA moieties at the polymer chain ends, the chain extension experiment was carried out using PMAECA-macroCTA. PMAECA (M n,GPC ¼ 6500 g mol À1 , PDI ¼ 1.14) was used for the polymerization of MAECA at 70 C in DMF at [MAE-CA] : [PMAECA-macroCTA] : [AIBN] ¼ 50 : 1 : 0.2. We obtained 77% monomer conversion aer 260 min and GPC traces of PMAECA-b-PMAECA shied toward higher molecular weight side aer the chain extension (Fig. S4 †). The M n,GPC ¼ 41 000 g mol À1 for PMAECA-b-PMAECA is somewhat higher than M n,theo ¼ 26 550 g mol À1 (PDI ¼ 1.48) and GPC trace is bimodal indicating bimolecular terminations. However, the M n,NMR ¼ 30 100 g mol À1 value is close to the theoretical value with $88% CTA chain ends. This result suggests that most of the chain ends of PMAECA are capped with the fragments of CTP and could be used further for block copolymer synthesis.
To investigate the effect of cholate moieties on the solution behaviour of copolymers containing PEGMA (a thermoresponsive monomer) or DMAEMA (a dual thermo-and pH-responsive monomer) units, several RAFT copolymerizations of MAECA with PEGMA and DMAEMA were carried out using   of PEGMA or DMAEMA hydrophilic units in the copolymer chain. Solubilities of all the polymers in various organic solvents and water are listed in Table S1. † To name copolymers, P(MAECA-co-PEGMA/DMAEMA), we followed the following rules ( Table 2): two letters CP and CD indicate MAECA-PEGMA and MAECA-DMAEMA series, respectively; 1, 2, 3, 4, 5 and 6 stand for the different feed compositions of the MAECA monomer.
The 1 H NMR spectra of copolymers were recorded in DMSOd 6 to determine % of co-monomer units in the copolymer chain by integration of the resonance signals related to each monomeric unit (Fig. 2). The 1 H NMR spectrum of PPEGMA (Fig. 2C) showed resonance signals at 1.61-1.  protons at 3.98 ppm from DMAEMA units). The compositions of the copolymers determined by 1 H NMR are given in Table 2. In all copolymers, we observed characteristic resonance signals in the aromatic region ( Table 2. The M n,GPC and PDI values were determined for all copolymers from the GPC analysis ( Table 2). The P(MAECA-co-PEGMA) copolymers showed unimodal RI traces with a small peak at higher molecular weight side due to the formation of dead chains via bimolecular termination reactions (Fig. S6A †). Substantial bimolecular terminations were noticed in the case of the P(MAECA-co-DMAEMA) copolymer series (Fig. S6B †). Although M n,NMR values match well with the M n,theo values predicted from stoichiometry and conversion (Table 2), M n,GPC values are somewhat higher compared to the M n,theo values and the discrepancy may be mostly due to the different hydrodynamic volume between copolymers and PMMA. Differences between M n,NMR and M n,theo values are higher in the case of P(MAECA-co-DMAEMA) copolymers since we observed higher bimolecular terminations (Fig. S6B †) in this series compared to the P(MAECA-co-PEGMA) copolymers (Fig. S6A †). Copolymerization of two monomers with different properties allows the synthesis of different products by variations in the relative amounts of the two monomer units in the copolymer chain. Since knowledge of reactivity ratios allows prediction of copolymer compositions at a particular monomer feed ratio, the reactivity ratios of MAECA (r MAECA ), PEGMA (r PEGMA ) and DMAEMA (r DMAEMA ) were determined for the copolymerization of MAECA with PEGMA and DMAEMA. To determine reactivity ratios, we have used the extended Kelen-Tüdös (extended K-T) method, 27 which is applicable for living controlled copolymerization at higher conversions. 28 Using monomer feed compositions, co-monomer contents in the copolymer (determined by NMR in Table 2) and conversion data, the reactivity ratios were determined by the extended K-T method. The r MAECA ¼ 1.62 and r PEGMA ¼ 0.71 for the MAECA-PEGMA series, and r MAECA ¼ 1.35 and r DMAEMA ¼ 0.65 for the MAECA-DMAEMA system were obtained, respectively. These results indicated random distribution of monomer units in both the copolymerization series and the copolymer will be moderately enriched with MAECA units. 29 Although all three monomers are methacrylate derivatives, the reactivity of the bulky MAECA monomer is almost double compared to PEGMA or DMAEMA. A similar trend of the reactivity ratio was observed for the copolymerization of tert-butyl methacrylate with the bulky rst-generation dendronized monomer carrying an ethylene glycol-based dendron. 30

Thermal analysis
The thermal stability was studied by TGA ( Fig. 3A and S7A †), which suggested the high stability of PMAECA (up to 320 C) towards thermal decomposition. The thermal stability of P(MAECA-co-PEGMA) and P(MAECA-co-DMAEMA) copolymers decreased with the decrease of MAECA content in the copolymer. The homopolymers PMAECA, PPEGMA and PDMAEMA displayed a T g of 148, À59 and 18 C, respectively. The T g value for PDMAEMA matches well with the previous report. 31 Very low T g for PPEGMA was observed due to the presence of exible ethylene glycol units in the side chains and this T g value is close to the previously reported T g ¼ À60 C. 32 Interestingly, PMAECA showed very high T g because of the side chain bulky cholate moieties with -OH groups. Although, high T g values (e.g., 192 C) were reported for polymers containing CA moieties, 33 we observed T g ¼ 148 C for PMAECA due to the presence of an additional ethylene oxide group before the CA moiety, which added some exibility to the polymer chains. For the copolymers, Fig. 3B and S7B † show a single T g value in between the corresponding T g values for the homopolymers (Table 2), which conrmed homogeneous nature of the copolymers and MAECA forms a copolymer in every composition with PEGMA or DMAEMA. The Fox equation 34 was used to calculate the theoretical T g values (T g,theo ) of copolymers in the bulk state and T g,theo values are plotted in Fig. 3C against the MAECA composition in the copolymer. Experimental T g values (T g,expt ) from the DSC study are also plotted in Fig. 3C as a function of the weight fractions of MAECA in two copolymer series, and we observed an excellent correlation between T g,expt and T g,theo .

Solution behaviour of side-chain cholate copolymers
Since PPEGMA is a thermoresponsive polymer and PDMAEMA is both temperature and pH-sensitive, we decided to study solution properties of all copolymers at different temperatures and solution pHs. Both thermo-and pH-responsive properties were determined by recording %T using UV-vis spectroscopy at 500 nm. Fig. 4A shows typical %T against temperature curves for CP1, CP2, CP3 and PPEGMA. The homopolymer PPEGMA showed LCST at 72 C at 2 mg mL À1 concentration, while the LCST of 68 C was reported for this polymer. 35 The LCST of a polymer depends on molecular weight, solution concentration, chain end structure, etc., 36 and the above difference could originate from these effects. For the CP1, CP2 and CP3 copolymers, as expected, LCST values decrease with the increase of the molar fraction of the hydrophobic MAECA content in the copolymer. The LCST values were found to be 61, 56 and 42 C for CP1, CP2 and CP3, while MAECA contributions in the copolymer chains were 5.5, 12 and 23%, respectively. Signicant cloudiness of CP4 was observed for the aqueous solution of CP4 even at close to freezing temperature of water and this copolymer did not illustrate a distinct LCST. The phase transition of CP4 may be troubled by the restricted movement of the side chain ethylene oxide groups due to the surrounding aggregation of hydrophobic components. 37 Also, high aggregation tendency of cholate moieties leads to the breakage of hydrogen-bonding from solvated PEGMA units. Note that the LCST values of PPEGMA and copolymers were independent of pH in the studied range 5.5 to 8.5, conrming that PEGMA and MAECA are not pH-dependent units. 38 In the P(MAECA-co-DMAEMA) series, only CD1 and CD2 were soluble in water, where MAECA units in the copolymer chains were 5.9 and 11.8%, respectively. In order to determine the pH responsiveness of PDMAEMA, CD1 and CD2, their aqueous solutions were analyzed as a function of pH at 25 C. PDMAEMA showed an LCST of 52 C at pH 7.4 and pH-responsiveness at pH 12.9 at 25 C as a result of complete deprotonation, and these values match well with the literature. 39 The phase transition pH is reduced with the increasing MAECA content in the copolymer and at 25 C we determined pH-responsiveness at 12.8 and 8.2 for CD1 and CD2, respectively (Fig. 4B). These results indicate that the hydrophobic MAECA units facilitate the aggregation of DMAEMA units in these copolymers at lower pH. The LCST of CD1 and CD2 was studied at different pHs and we determined hydrophilic to hydrophobic transitions for CD1 at 73, 63, 53, 44, 37 and 33 C at pH values 7.4, 7.6, 7.8, 8.0, 8.2 and 8.4, respectively (Fig. 4C). Similarly, the LCSTs of 54, 43 and 31 C were determined for CD2 at pH 7.4, 7.6 and 7.8, respectively (Fig. S8 †). The transmittance decreased slowly for CD2 (Fig. S8 †) at all pHs, indicating disturbed deprotonation of tertiary amine groups with increasing cholate moiety in the copolymer. The turbidity of polymers in aqueous solution depends on pH, involving hydrophobic interactions between cholate moieties and electrostatic repulsion between protonated amino groups. Amino groups are protonated below the responsive pH at a particular temperature, resulting in electrostatic repulsion among them and increased solubility. However, the above results indicate that the LCST can be tuned at a particular solution pH by employing different molar factions of monomer units in the copolymer. The above phase transitions were reversible indicating uninterrupted protonation and deprotonation of the tertiary amine functional groups in the DMAEMA moiety.

Self-assembly behaviour of random copolymers
The self-assembly of random copolymers to different morphologies such as micelles, vesicles, and other shapes in a selective medium is not so common compared to the block copolymers. Since amphiphilic random copolymers contain hydrophobic cholate side chains and hydrophilic PEGMA or DMAEMA units, herein we studied self-assembly characteristics of CP1, CP2, CP3 and CP4 from the P(MAECA-co-PEGMA) series, and CD1 and CD2 from the P(MAECA-co-DMAEMA) category in aqueous medium by uorescence spectroscopy using pyrene as a uorescent probe. Fig. S9A † shows the emission spectra of pyrene as a function of CP2 concentration, where the intensity of uorescence gradually increased with increasing CP2 concentrations indicating solubilisation of pyrene in the hydrophobic microdomains of the aggregates. The critical aggregate concentration (CAC) of polymers was measured by the standard procedure from emission spectra of pyrene from the plot of I 383 /I 373 versus the logarithm of the copolymer concentration, which is shown for CP2 in Fig. S9B (see Fig. S10 † for CD2). The CAC values of 3.3, 3.2 and 2.3 mg L À1 were determined for CP2, CP3 and CP4, respectively. Among the three copolymers from the P(MAECA-co-PEGMA) series, aggregation tendency gradually improved with increasing MAECA contents in the polymer main chain. Interestingly, CP1 was unable to increase the uorescence intensity of pyrene with increasing polymer concentration, indicating the lack of aggregation to form hydrophobic microdomains due to the overall high hydrophilicity of CP1. In the P(MAECA-co-DMAEMA) series, CD1 and CD2 gave CAC values of 41 and 16 mg L À1 , respectively. Here too, aggregation probability increases with increasing MAECA components in the copolymer. Note that the CAC values are much smaller than those of low molecular weight surfactants and comparable with other polymer micelle-like aggregates. Furthermore, self-assembly behaviour of CP2, CP4 and CD2 was investigated by 1 H NMR spectroscopy in two different solvents, D 2 O and DMSO-d 6 . The DMSO-d 6 is a good solvent for both MAECA and PEGMA/DMAEMA units in the copolymer, whereas D 2 O does not solubilise the hydrophobic MAECA moiety. The 1 H NMR spectra of CP2 (Fig. 5A), CP4 (Fig. S11 †) and CD2 (Fig. 5B) in DMSO-d 6 show all characteristic peaks from both the co-monomers, while the resonance signal in D 2 O disappeared at 0.60 ppm from the methyl group of hydrophobic cholate units. These results revealed that CP2, CP4 and CD2 formed higher order structure through the arrangement of hydrophilic and hydrophobic components in the copolymers in aqueous media. In the aggregates the hydrophilic units were uncovered to the outer surface and hydrophobic parts were favoured to the core, and as a result we did not observe the 0.60 ppm peak from hydrophobic cholate moieties in the core in the NMR spectra of CP2, CP4 and CD2 in D 2 O.
In order to study the aggregation behaviour in aqueous solution, the hydrodynamic diameter (D h ) of various copolymers was determined by DLS at 25 C with the sample concentration of 1 mg mL À1 (w/v). CP2 (2.8 and 95 nm), CP3 (3.0 and 95 nm), and CP4 (4.6 and 96 nm) displayed two peaks, one at around 3-5 nm and another peak at approximately 95 nm (Fig. 6A). The smaller size in the range of 3 to 5 nm might be attributed to the size of a single chain conformation. Larger size in the range of 95-96 nm suggests formation of aggregated particles. CP1 showed only one peak in Fig. 6A with a D h value of 2.7 nm, conrming its unimeric structure with insufficient amount of hydrophobic units in the polymer chain for selfassembly. This result conrmed the above pyrene encapsulation studies, where CP1 was unable to increase the uorescence intensity of pyrene with increasing polymer concentration, due to the lack of self-assembly formation capabilities. Similarly, we   measured D h values for CD1 at pH 7 and CD2 at pH 5 (Fig. 6B). Both CD1 and CD2 showed single size distribution with D h values of $100 nm, indicating their self-assembly in water to yield multi-molecular aggregates.
To further characterize the aggregates and obtain direct visualization of their size and morphology, we investigated thin lms of the random copolymers by imaging techniques such as AFM and SEM. The morphology of CP2, CP3 and CP4 copolymers was found to be spherical, with an average size of 78, 71 and 76 nm, respectively (Fig. 7). These copolymers formed micellar structures in water due to the aggregation of the hydrophobic cholate component towards the core and hydrophilic PEGMA units constitute the corona of the micelles. The AFM result for CP4 was also supported by SEM analysis, where we observed $78 nm size with spherical structure (Fig. S12 †). Similarly, CD1 gave globular shape in the SEM image with a diameter of $82 nm (Fig. S12 †). Note that sizes estimated from the AFM/SEM study are slightly lower than DLS data as reported elsewhere. 40 Also, the observed diameters by DLS and AFM analyses are much higher than the contour length of the corresponding copolymers, indicating some another sort of selfassemblies. 41 Nile red encapsulation using copolymer aggregates The encapsulation of nile red, a model hydrophobic dye, by various water soluble copolymers was spectroscopically investigated. Typically, 10 mM nile red stock solution was prepared in THF, and 100 mL aliquots were added to 4 mL of the polymer solutions in deionised water (1.0 mg mL À1 , w/v). The solution was stirred for 24 h at room temperature to eliminate THF, and then it was ltered through a 0.45 mm syringe lter to remove insoluble nile red. 42 At this point, the aqueous solution of CP1 was colourless, which further demonstrates the inability of CP1 to self-assemble into higher order structure. However, CP2, CP3 and CP4 (Fig. 8A) solutions were pink coloured, which were studied by UV-vis and uorescence spectroscopy. Fig. 8B shows UV-vis spectra of nile red loaded CP2, CP3 and CP4, where we observed l max values of 568, 563 and 560 nm for CP2, CP3 and CP4, respectively. Also, absorbance of copolymers gradually increased with increasing MAECA content in the copolymer. The uorescence emission spectra in Fig. 8C suggest relatively low uorescence intensity for CP2 (l max ¼ 628 nm) and CP3 (l max ¼ 625 nm) compared to CP4 with a l max of 622 nm. The uorescence intensity of CP4 is roughly 10 times higher than CP3. These results indicate that CP4 has more hydrophobic environment due to the presence of more number of MAECA units. Since the P(MAECA-co-PEGMA) copolymer series is thermoresponsive and the P(MAECA-co-DMAEMA) copolymers are both thermo-and pH-sensitive, external stimuli such as thermo-and pH-induced controlled release of hydrophobic molecules (e.g., nile red and doxorubicin) from the polymeric micelles are under investigation. 43 Conclusions RAFT polymerization has been used to design bile acid containing thermo-, and dual thermo-and pH-responsive random copolymers. The random distribution of co-monomers in the copolymer chains was conrmed by the reactivity ratios of two monomers for the two systems (MAECA with PEGMA and DMAEMA), and further supported by a single T g value of copolymers which varies smoothly with composition between the T g values of the respective homopolymers. The thermal stability of copolymers increased with increasing MAECA contents in copolymer. The LCST can easily be tuned for the P(MAECA-co-PEGMA) series by varying the co-monomer composition of copolymers. For P(MAECA-co-DMAEMA) copolymers, the LCST can be adjusted between 31 and 73 C by changing the pH and co-monomer contents in the copolymer. 1 H NMR and uorescence spectroscopy indicated self-assembly from random copolymers in water to micellar structures with a spherical shape, which were characterized by AFM and SEM studies, while aggregation was dependent on hydrophilic PEGMA or DMAEMA to hydrophobic MAECA contents in the copolymers. Cholic acid based copolymers were capable of encapsulating hydrophobic nile red molecules in aqueous solutions. Further studies are in progress to achieve bile acid containing random copolymer architectures with high drug loading capacity, and controlled release capability at different pHs and temperatures.