Tuning the properties of hydrogen-bonded block copolymer worm gels prepared via polymerization-induced self-assembly

Polymerization-induced self-assembly (PISA) is exploited to design hydrogen-bonded poly(stearyl methacrylate)-poly(benzyl methacrylate) [PSMA-PBzMA] worm gels in n-dodecane. Using a carboxylic acid-based RAFT agent facilitates hydrogen bonding between neighboring worms to produce much stronger physical gels than those prepared using the analogous methyl ester-based RAFT agent. Moreover, tuning the proportion of these two types of end-groups on the PSMA chains enables the storage modulus (G′) of a 20% w/w worm gel to be tuned from ∼4.5 kPa up to ∼114 kPa. This is achieved via two complementary routes: (i) an in situ approach using binary mixtures of acid- and ester-capped PSMA stabilizer chains during PISA or (ii) a post-polymerization processing strategy using a thermally-induced worm-to-sphere transition to mix acid- and ester-functionalized spheres at 110 °C that fuse to form worms on cooling to 20 °C. SAXS and rheology studies of these hydrogen-bonded worm gels provide detailed insights into their inter-worm interactions and physical behavior, respectively. In the case of the carboxylic acid-functionalized worms, SAXS provides direct evidence for additional inter-worm interactions, while rheological studies confirm both a significant reduction in critical gelation concentration (from approximately 10% w/w to 2–3% w/w) and a substantial increase in critical gelation temperature (from 41 °C to 92 °C). It is remarkable that a rather subtle change in the chemical structure results in such improvements in gel strength, gelation efficiency and gel cohesion.

S2 spectroscopy analysis indicated a mean degree of esterification of 97% (see Figure S1) while THF GPC studies indicated an M n of 5 500 g mol -1 and an M w /M n of 1. 10.

Synthesis of HOOC-PSMA 11 -PBzMA 65 diblock copolymer worms via RAFT dispersion polymerization of BzMA.
The synthesis of HOOC-PSMA 11  T21s (307 µL of a 1.0 % v/v solution in n-dodecane, 0.0128 mmol; CTA/T21s molar ratio = 5.0) and additional ndodecane (4.98 mL) were weighed into a 14 mL glass vial. This vial was sealed and the reaction mixture degassed with nitrogen for 30 min with stirring. The vial was then placed in an oil bath pre-set to 90 °C. According to 1 H NMR spectroscopy studies, 97% BzMA monomer conversion was achieved within 5 h. THF GPC analysis indicated an M n of 15 600 g mol -1 and an M w /M n of 1.11 for the final PSMA 11 -PBzMA 65 worms. The same synthetic protocol was also used to prepare PSMA 11 -PBzMA 65 diblock copolymer worms with either 3:1 or 1:3 binary mixtures of HOOC-PSMA 11 and H 3 COOC-PSMA 11 macro-CTAs. 1 H NMR spectroscopy. All 1 H NMR spectra were recorded in either CDCl 3 or CD 2 Cl 2 using a 400 MHz Bruker Avance-400 spectrometer with 64 scans being averaged per spectrum.
Gel permeation chromatography (GPC). GPC was used to assess the molecular weight distributions (MWDs) of the two PSMA 11 homopolymer precursors and the five diblock copolymers prepared using the binary mixture approach summarized in Scheme 1b. An Agilent 1260 Infinity system comprising an HPLC pump, two PL gel 5 µm (30 cm) Mixed C columns and a refractive index detector was run at 30 °C. The GPC eluent was HPLC-grade THF S3 containing 2.0% v/v triethylamine and 0.05 % w/v butylhydroxytoluene (BHT) at a flow rate of 1.0 mL min -1 . A series of twelve near-monodisperse poly(methyl methacrylate) standards with M p values ranging from 625 to 2 480 000 g mol -1 were used for calibration. Varian Cirrus GPC software provided by the instrument manufacturer was used to analyze the chromatograms.
Transmission electron microscopy (TEM). TEM studies were conducted using a Philips CM 100 instrument operating at 100 kV and equipped with a Gatan 1 k CCD camera. A single droplet of each 0.10% w/w diblock copolymer dispersion was placed on a carbon-coated copper grid, allowed to dry and then exposed to ruthenium(VIII) oxide vapor for 7 min at 20 °C prior to analysis. This heavy metal compound acted as a positive stain for the core-forming PBzMA block so as improve electron contrast. The ruthenium(VIII) oxide was prepared as follows. Ruthenium(IV) oxide (0.30 g) was added to water (50 g) to form a black slurry; addition of sodium periodate (2.0 g) with stirring produced a yellow solution of ruthenium(VIII) oxide within 1 min. 2

Small-angle X-ray scattering (SAXS). SAXS patterns were recorded at a synchrotron source (Diamond Light
Source, station I22, Didcot, UK) using monochromatic X-ray radiation (X-ray wavelength λ = 0.999 Å, with scattering vector q ranging from 0.0018 to 0.15 Å -1 , where q = 4π sin θ/λ and θ is one-half of the scattering angle) and a 2D Pilatus 2M pixel detector (Dectris, Switzerland). A glass capillary of 2 mm diameter was used as a sample holder and measurements were conducted on 1.0% w/w dispersions in n-dodecane. Scattering data were reduced and normalized with water being used for the absolute intensity calibration utilizing standard routines available at the beamline 3 and were further analyzed using Irena SAS macros for Igor Pro. 4 Oscillatory rheology. All rheology measurements were recorded using a TA Instruments AR-G2 rheometer equipped with a 40 mm 2° aluminum cone and a variable temperature Peltier plate. For angular frequency sweeps, the storage (G') and loss (G") moduli were measured from 0.1 to 100 rad s -1 at 20 °C using a fixed strain amplitude of 1.0 %. For temperature sweeps, G' and G" were measured at a heating rate of 2 °C min -1 , and at an angular frequency of 10 rad s -1 and a fixed strain amplitude of 1.0%.
Fourier transform infrared (FT-IR) spectroscopy. All FT-IR spectra were recorded in absorbance mode using a Thermo Scientific Nicolet IS10 FT-IR spectrometer equipped with a Diamond ATR 'Golden Gate' accessory. A baseline was recorded before analyzing 50% w/w solutions of the HOOC-PSMA 11 and H 3 COOC-PSMA 11 homopolymer precursors in n-dodecane at 20 °C. In each case, 1024 scans were averaged per spectrum. Omnic software provided by the instrument manufacturer was used to analyze the IR spectra. No attempts were made to include solvent effects in these calculations. Calculations on molecular clusters are affected by the basis set superposition error (BSSE). For energy, geometry optimization (gradients) and vibrational (Hessian) calculations on molecular clusters, the BSSE was corrected using a suite of programmes, [6][7][8][9] implementing the counterpoise correction procedure (CP) as proposed by Boys and Bernardi 10 within GAMESS.  In order to support the hypothesis of the formation of carboxylic acid dimers and also to guide spectroscopic measurements, we performed high-level quantum chemical calculations (MP2/6-31G(d,p,)++, CP-corrected) on two model systems. In these calculations, acetic acid is used as a proxy for the carboxylic acid-functionalized polymer chains and methyl acetate is employed as a model for the methyl-ester-capped polymer chains.
As expected, the most stable conformation for acetic acid is characterized by strong hydrogen-bonded dimers (see Scheme 3 in the main manuscript). Given its C i inversion symmetry, the vibrational exclusion rule applies to this structure, thus vibrations are either IR-active (anti-symmetric) or Raman-active (symmetric). The binding energy, ΔE el (essentially the dissociation energy at 0 K) is calculated to be -56.2 kJ mol -1 (and ΔE 0 = -50.2 kJ mol -1 including the vibrational zero-point energy). These results are consistent with prior theoretical and experimental studies on acetic acid dimers. For example, B3LYP/6-31G** calculations by Chocholoušová and co-workers suggest ΔE el = -66.3 kJ mol -1 , although this is a somewhat lower level theory. 11 Similarly, an experimental study of acetic acid dimers in the gas phase by Jaffe and Rose indicated ΔH o (298 K) = -60.3 kJ mol -1 . 12 The ester carbonyl band is split into an IR-active anti-symmetric stretch at 1787.5 cm -1 and an IR-inactive (but Raman-S12 active) symmetric stretch at 1753.9 cm -1 . The wavenumber for the IR-active component is red-shifted by -29.9 cm -1 compared to the monomer. However, this shift is expected to be smaller at 298 K owing to hot bands. 9 The most stable equilibrium configuration of the methyl acetate dimer also involves hydrogen-bonded dimers (see Figure S9). However, hydrogen bonds between C=O and H-C are much weaker than those involving O-H or N-H groups. [6][7][8][9] The calculated binding energy is ΔE el = -12.5 kJ mol -1 , and ΔE 0 = -11.1 kJ mol -1 (including vibrational zero-point energies). To a first approximation, the ester carbonyl stretching vibration is split into an IR-active antisymmetric stretching at 1789.0 cm -1 , and an IR-inactive (but Raman-active) symmetric stretching at 1782.0 cm -1 . The wavenumber shift of the IR-active component compared to the monomer is only -5.0 cm -1 , which reflects the much weaker hydrogen-bonding in this case. In practice, this shift will be even smaller at 298 K owing to hot bands. 9 The Raman spectrum should exhibit a more pronounced shift of -12.0 cm -1 (but again this shift is expected to be smaller at 298 K owing to hot bands).

Worm-like micelle SAXS model
Programming tools within the Irena SAS Igor Pro macros 3 were used to implement the scattering models.
In general, the intensity of X-rays scattered by a dispersion of nano-objects [as represented by the scattering cross-section per unit sample volume, (q)] can be expressed as: where is the form factor, is a set of k parameters describing the structural morphology, is the distribution function, S(q) is the structure factor and N is the number density of nano-objects per unit volume expressed as: where is the volume of the nano-object and is its volume fraction within the dispersion. It is assumed that S(q) = 1 at the sufficiently low copolymer concentrations used in this study (1.0% w/w).
The worm-like micelle form factor for Equation S1 is given by: where is the radius of gyration of the coronal steric stabilizer block (in this case, PSMA 11 ). The X-ray scattering length contrasts for the core and corona blocks are given by and respectively. Here, , and are the X-ray scattering length densities of the core block ( = 10.38 x 10 10 cm -2 ), corona block ( = 9.24 x 10 10 cm -2 ) and n-dodecane solvent ( = 7.32 x 10 10 cm -2 ), respectively. and are the volumes of the core block ( ) and the corona block ( ), respectively. The selfcorrelation term for the worm core cross-sectional volume-average radius is: where 2 ( , ) = [ 2 1 ( ) ] 2 S5 S14 and is the first-order Bessel function of the first kind, and a form factor for self-avoiding semi-1 ( , , ) flexible chains represents the worm-like micelles, where is the Kuhn length and is the mean contour length. A complete expression for the chain form factor can be found elsewhere. 6 The mean aggregation number of the worm-like micelle, , is given by: where is the volume fraction of solvent within the worm-like micelle cores, which was found to be zero in all cases. The possible presence of semi-spherical caps at both ends of each worm is neglected in this form factor.
A polydispersity for one parameter ( ) is assumed for the micelle model, which is described by a Gaussian distribution. Thus, the polydispersity function in Equation S1 can be represented as: where is the standard deviation for . In accordance with Equation S2, the number density per unit volume for the worm-like micelle model is expressed as: where is the total volume fraction of copolymer in the worm-like micelles and is the total volume of ( 1 ) copolymer in a worm-like micelle .