pH‐Responsive Side Chains as a Tool to Control Aqueous Self‐Assembly Mechanisms

Abstract pH‐Tunable nanoscale morphology and self‐assembly mechanism of a series of oligo(p‐phenyleneethynylene) (OPE)‐based bolaamphiphiles featuring poly(ethylene imine) (PEI) side chains of different length and degree of hydrolysis are described. Protonation and deprotonation of the PEI chains by changing the pH alters the hydrophilic/hydrophobic balance of the systems and, in turn, the strength of intermolecular interactions between the hydrophobic OPE moieties. Low pH values (3) lead to weak interaction between the OPEs and result in spherical nanoparticles, in which aggregation follows an isodesmic mechanism. In contrast, higher pH values (11) induce deprotonation of the polymer chains and lead to a stronger, cooperative aggregation into anisotropic nanostructures. Our results demonstrate that pH‐responsive chains can be exploited as a tool to tune self‐assembly mechanisms, which opens exciting possibilities to develop new stimuli‐responsive materials.


A. Description of Experimental Techniques
General. All solvents were dried according to standard procedures. Reagents were used as purchased. All starting materials were purchased from Sigma-Aldrich, Roth, ABCR, Merck or Acros Organics and were used as received, unless otherwise noted. 2-Ethyl-2-oxazoline (EtOx) and methyl tosylate (MeOTs) were distilled from barium oxide. 1,4-bis((4ethynylphenyl)ethynyl)benzene (OPE) and PEtOx x -N 3 were synthesized as described in our previous literature report [S1] .

Dynamic Light Scattering (DLS)
Dynamic light scattering (DLS) was performed at a scattering angle of 90° on an ALV CGS-3 instrument equipped with a He-Ne laser operating at a wavelength of 633 nm at 25 °C. The CONTIN algorithm was applied to analyze the obtained correlation functions. For temperature control, the DLS is equipped with a Lauda thermostat. Apparent hydrodynamic radii were calculated according to the Stokes-Einstein equation.

Transmission Electron Microscopy (TEM)
The formed aggregates were analyzed using a transmission electron microscop (TEM) (Zeiss-CEM 902A, Oberkochen, Germany) operated at 80 kV. Images were recorded using a 1k TVIPS FastScan CCD camera. TEM samples were prepared by applying a drop of an aqueous sample solution onto the surface of a plasma-treated carbon coated copper grid.

Atomic Force Microscopy (AFM)
AFM images were recorded on a Multimode® 8 SPM System (AXS Bruker). Silicon cantilevers with a nominal spring constant of 9 Nm -1 and with resonant frequency of ̴ 150 kHz and a typical tip radius of 7 nm (OMCL-AC200TS, Olympus) were employed.
Optical Measurements. Electronic absorption spectra were recorded on a Jasco UV-770-ST UV-Vis/NIR spectrophotometer and emission studies were performed on Jasco FP-8500 spectrofluorometer. All experiments were carried out using quartz cuvettes with optical pathlength of 1 cm. For all measurements, spectroscopic grade solvents (Uvasol) from Merck were used.

Sample preparation:
To investigate the pH-responsiveness of 1a-b and 2a-b the materials were dissolved in the non-selective solvent mixture dimethyl sulfoxide (DMSO)water (1.5:3.5). Afterwards, the samples were dialyzed against a buffer solution (disodium hydrogen phosphate 0.2Mcitric acid 0.1M) and adjusted to a pH value of 7. PEI as a weak polyelectrolyte [S3] would strongly influence the pH and thus no representative and reproducible self-assembly might be investigated. To prevent this, buffer solutions were applied.
Subsequently, one third was taken as such for the investigation of the self-assembly behavior in water. The other two-thirds were divided into two halves and further dialyzed against a buffer solution (glycine 0.1M/sodium chloride 0.1Mhydrochloric acid 0.1M) and adjusted to a pH value of 3 or dialyzed against a buffer solution (glycine 0.1M/sodium chloride 0.1Msodium hydroxide 0.1M) and adjusted to a pH value of 11. A similar concentration of 1 mg mL -1 for all samples was targeted.
DLS measurements with regard to the Stokes-Einstein equation following the assumption that the scattered particles are spherical. Non-spherical particles generally lead to discrepancies in size determinations.
TEM measurements: the samples were drop-casted on carbon-coated TEM grids. To avoid artefacts of the salt from the buffer solutions, 2 drops of pure water were applied to the TEM grid after plotting the sample drop, removing as much of the salt as possible.
Optical studies: All sample were diluted to the required concentration before UV-Vis and emission measurements. Absorption changes were recorded from 200 to 600 nm and emission changes were monitored by exciting at 300 nm and collecting from 330-700nm with a 1cm cuvette at 25 ºC. A hot solution of the samples at 363 K were cooled down to 283 K with a cooling rate of 0.2 K/min. Always freshly prepared buffer solutions were used for checking the reproducibility of the results.

AFM measurements:
The aggregates were obtained in an identical way as for UV/Vis studies (slow cooling of the monomer solution). Then, 50-100 µL of solution was spin-coated onto highly oriented pyrolytic graphite (HOPG) with an rpm of 2000. The residual solvent evaporation was confirmed by leaving the samples at room temperature overnight.

SEC measurements:
Size-exclusion chromatography was performed on a Shimadzu system equipped with a SCL-10A system controller, a LC-10AD pump, a RID-10A refractive index detector, a SPD-10A UV detector at 365 nm using a solvent mixture containing chloroform, triethylamine and isopropanol (94:4:2) at a flow rate of 1 mL min -1 on a PSS-SDV-linear M 5 μm column. The system was calibrated with polystyrene, poly(methyl methacrylate) and poly(ethylene oxide) standards.

MALDI-ToF-MS measurements:
MALDI-ToF-MS spectra were measured on an Ultraflex III TOF/TOF (Bruker Daltonics GmbH) equipped with a Nd:YAG laser and a collision cell. All spectra were measured in the positive reflector or linear mode using DHB or DCTB as matrix.
FT-IR measurements: Fourier transform infrared spectroscopy was performed on a FT-IR spectrometer IR Affinity-1 from Shimadzu.

Synthesis of PEtOx x -OPE-PEtOx x
1,4-bis((4-ethynylphenyl)ethynyl)benzene (OPE, 20 mg, 0.06 mmol) and PEtOx x -N 3 (0.13 mmol, 2.2 eq.) were dissolved in 10 mL THF. CuBr (5 eq.) and PMDETA (5 eq.) were added and the corresponding reaction solution was stirred for 1 hour at 50 °C. Subsequently, the solution was diluted with methylene chloride and the copper was removed by washing with water several times until no more coloration of the respective aqueous layer could be observed.

S4
The organic layer was dried over sodium sulfate and the solvent was evaporated under reduced pressure. The residue was dissolved in 10 mL methylene chloride and subsequently the polymer was precipitated in cold diethyl ether. The resulting suspension was allowed to stand at room temperature overnight to get rid of the excess of unconverted homopolymer (soluble in warm ether). Subsequently, the yellowish suspension was centrifuged (5 min, 8000 rpm) and the residue washed with room tempered diethyl ether and dried in vacuo.

Synthesis of partially cleaved (PEtOx m -co-PEI n ) x -OPE-(PEtOx m -co-PEI n ) x (x=12, 17) (1a-b and 2a-b)
In a microwave vial, PEtOx x -OPE-PEtOx x was dissolved in 6M hydrochloric acid (HCl) giving a final amide concentration of 0.48M. The resulting solutions were heated to 100 °C for 110 and 150 min, respectively. After that, the reaction solutions were directly transferred into a Float-ALyzer ® (MWCO: 100-500Da) and dialyzed against water for one day. The pure polymers were obtained after freeze-drying. Figure S5: Comparison of NMR spectra: series of 1 H-NMR (300 MHz, MeOD) spectra corresponding to the hydrolysis after different reaction times of PEtOx17-N3 in 6M HCl at 100 °C. Table S1: Degree of hydrolysis for PEtOx-N3 with DP = 12, 17 after different reaction times in a microwave synthesizer at 100 °C with a constant amide concentration of 0.48M.           Figure S15: Photographs of the aqueous solution of 2b at different pH (left to right -3,7 and 11) and number-weighted DLS CONTIN plots of 2b at pH 3 (black trace), pH 7 (red trace) and pH 11 (blue trace). TEM micrograph of 2b at pH 3 (c) and at pH 11 (d).

C. Supplementary Figures and Table
Figure S16: Cooling curves of 2b at pH 3 (a), pH 7 (b) and pH 11 (c) obtained by monitoring UV-Vis spectral changes at 335 nm. At pH 3, a sigmoidal shape is observed, suggesting the existence of isodesmic self-assembly mechanism (blue fits in a). A cooperative mechanism is found at pH = 11 by fitting to the nucleation-elongation model (red fit in c). At pH = 7, both cooperative (red) and isodesmic (blue) fail to describe the experimental data accurately. Table S4: Thermodynamic parameters of 1b and 2b at different pH values calculated using the isodesmic and nucleation-elongation model (c = 5 × 10 -6 M). High degree of cooperativity was observed for 1b and 2b at pH 11, whereas at pH 3 adequate results were obtained with isodesmic fit.