Reversible Speed Regulation of Self‐Propelled Janus Micromotors via Thermoresponsive Bottle‐Brush Polymers

Abstract This work reports a reversible braking system for micromotors that can be controlled by small temperature changes (≈5 °C). To achieve this, gated‐mesoporous organosilica microparticles are internally loaded with metal catalysts (to form the motor) and the exterior (partially) grafted with thermosensitive bottle‐brush polyphosphazenes to form Janus particles. When placed in an aqueous solution of H2O2 (the fuel), rapid forward propulsion of the motors ensues due to decomposition of the fuel. Conformational changes of the polymers at defined temperatures regulate the bubble formation rate and thus act as brakes with considerable deceleration/acceleration observed. As the components can be easily varied, this represents a versatile, modular platform for the exogenous velocity control of micromotors.

SI-3 stir for 12 h. Triethylamine (2.1 mL, 15.10 mmol) and an excess of propargylamine (1 mL, 15.61 mmol) was added to the mixture and the solution was left to stir overnight. The polymer was removed from the glovebox and filtered by gravitational filtration. The flask and the filter residue were washed with additional THF. The filtrate was concentrated, and the remaining material was placed into a vacuum oven at 40 °C for >48 h. Half of the polymer (2.56 g) was dispersed in ethanol and transferred to a dialysis tube (6-8 kD). Dialysis was carried out in water for 3 days with frequent changes of the water (every hour). The dialysis bag with the polymer was transferred into an ethanol bath for 2 days. This was repeated for the remaining half yielding in total 5.07 g of PPz. 1

Synthesis of SiPPzi by photochemical thiol-yne addition reaction with 3-mercaptopropyl trimethoxysilane:
The polyphosphazene was further functionalized, as reported previously [1] PPz1 (1.05 g) was placed in a round-bottom flask and to this photoinitiator DMPA (26 mg, 0.10 mmol) and ethanol (40 mL) were added. The solution was flushed with Argon for 15 mins. After this, 3mercaptopropyl trimethoxysilane (0.2 mL, 1.0 mmol) was added and the reaction was placed in the UV reactor for 6 h while stirring. NMR spectroscopy of the crude mixture suggested SI-4 quantitative functionalization of the silane. The final silane-derived polymer (SiPPz1) was not further purified and directly attached to the silica particles surface. A similar second thermoresponsive polyphosphazene was prepared, called SiPPz2. Figure S2. Thiol-yne addition of (3-mercaptopropyl)trimethoxysilane to PPz to yield the silane derived thermoresponsive polymer SiPPz1, (n ~ 50), substituents statistically distributed along the backbone.

Lower Critical Solution Temperature (LCST) determination:
>LCST <LCST Figure S3. Illustration of the reversible thermoresponsive behaviour of the bottle brush-like polyphosphazene in water, collapsed above the corresponding LCST.

SI-5
A visual confirmation of the LCST was carried out below and above LCST (at 4 ºC and 25 ºC ( Figure S4) of polymer PPz1 and PPz2 in Milli Q water (1 mg mL -1 ).

Manganese immobilization within bipyridine functionalized microparticles (MOM-Mn):
The manganese bipyridine complex formation within the pores from previous prepared BPycontaining microparticles (see Figure S7) Figure S7. Schematic illustration of the colour change of the mesoporous silica particle after immobilization of Mn 2+ ions within the silica pores through interaction with bypiridine units.

Preparation of Janus micromotors (MOM-Mn-J) by Pickering emulsion:
In order to functionalize one hemisphere of the silica mesoporous microparticles containing  Figure S8). To fabricate the Janus micromotors, a classical grafting-to approach was taken as already reported. [6] The MOM-Mn particles (220 mg) were dispersed in ethanol (20 mL), sonicated for 5 minutes and stirred under N2 atmosphere at room temperature. This suspension was then added to the hot wax mixture, and the polymer SiPPz1 (0.53g in 20 mL EtOH) was added.
The mixture was left to stir for 24 h. The solution was then allowed to cool slowly to room temperature. The reaction mixture was then filtered and the resulting wax-particles were washed with ethanol and acetone (10 mL) and dried at room temperature for more than 12 h (yielding ca. 900 mg of wax-particles). In order to obtain the one-hemisphere functionalized organosilica microparticles (MOM-Mn-J), the wax was then removed (see Figure S8). For this the microparticles were suspended in n-heptane (40 mL) and placed on the orbital shaker

Preparation of fully-grafted micromotors (MSM-Mn-FG):
In order to graft the free 2 nd hemisphere of the silica microparticles with polymer as well, previously dried one-hemisphere functionalized Janus microparticles (MSM-Mn-J, 100 mg) and SiPPz2 (0.25 g) were suspended in EtOH (10 mL) and the mixture was left to stir for 24 h (as shown in Figure S9). After this, the mixture was filtered and the resulting particles were left to dry at room temperature >12 h, yielding the final fully grafted motors MOM-Mn-FG.    Table S3. Quantified EDX sum spectra (without Cu, C) in at% for both hemispheres from the Janus microparticle (half 1 and half 2 relate to the marked areas in Figure S14) EDX sum spectra were obtained by adding up the counts stemming from the microparticles.  Figure   S13. In order to show the elemental distribution, the spectra are quantified pixelwise.
Therefore, the counts collected in each pixel are much lower and the single pixel spectra are highly noisy. The statistical threshold for the quantified mappings had to be lowered.
Nevertheless, changes in concentrations which are not due to thickness changes were observed in the P map for MOM-Mn-J (see Figure S14). This behavior could not be detected for all particles, but the bottom particle featured a Janus appearance. To confirm this, the sum spectra of the two hemispheres were extracted (yellow and red marked regions in Figure   S14). The P peak is clearly only visible in one of the two halfs (see Figure 2b in the main article). The quantified P concentration in the two hemispheres is given in Table S3. Since the concentrations from the functionalization compounds are very low, and EDX is not wellsuited for light elements such as P, N and S, the quantified results do not represent the true concentrations. However, the method shows that the presence of the detected elements matches the functionalization steps and the relative difference of P in a Janus microparticle could be reported, which confirmed the one-sided coverage with PPz.

Motion characterization of the micromotors:
Temperature responsiveness studies: The  Table S4.

SI-18
Mean square displacement (MSD) analysis: This method was used as reported from literature [7]. It computes the weighted average over all MSD curves from the particle trajectories. Weights are taken to be the number of averaged delays in individual curves, which favors short delays. The weighted means was calculated as an approximation, using the next formula: weighted std/(Nfreedom) 1/2 . Trajectories were analyzed with MATLAB, using a package specifically developed for MSD analysis, publicly available: http://www.mathworks.com/matlabcentral/fileexchange/40692-mean-squaredisplacement-analysis-of-particles-trajectories