Reconfigurable artificial microswimmers with internal feedback

Self-propelling microparticles are often proposed as synthetic models for biological microswimmers, yet they lack the internally regulated adaptation of their biological counterparts. Conversely, adaptation can be encoded in larger-scale soft-robotic devices but remains elusive to transfer to the colloidal scale. Here, we create responsive microswimmers, powered by electro-hydrodynamic flows, which can adapt their motility via internal reconfiguration. Using sequential capillary assembly, we fabricate deterministic colloidal clusters comprising soft thermo-responsive microgels and light-absorbing particles. Light absorption induces preferential local heating and triggers the volume phase transition of the microgels, leading to an adaptation of the clusters’ motility, which is orthogonal to their propulsion scheme. We rationalize this response via the coupling between self-propulsion and variations of particle shape and dielectric properties upon heating. Harnessing such coupling allows for strategies to achieve local dynamical control with simple illumination patterns, revealing exciting opportunities for developing tactic active materials.

The article presents a dumbbell-shaped particles as a self-propelling system which alters its motion via change in properties on one of the lobes in response to external stimuli. The lobes of particles are made of polystyrene and poly-NIPAM, which induces a selective thermoresponsivenss where only one lobe shrinks upon increasing the temperature. Authors used AC electric field to drive the propulsion, and external light to induce a phase transformation in the poly-NIPAM lobe which results in changes in the motion trajectory of the particles. The article presents few interesting findings, but all the effects observed are well known and established. The unique part of the articles in combining these effects, which is non-trivial. My major criticisms come from the overselling of the presented idea and lack of scientific discussion in the article. Therefore, I cannot recommend its publication in present form. My comments/suggestions are as follows: 1. While authors try to make the claim about the autonomy of the particles, I believe it is still a combination of two external stimuli one driving the propulsion and one (sort of) controlling the characteristics of propulsion (which has been previously demonstrated). Therefore, abstract and introduction should be toned down, and authors should refocus these sections on the science and minimize overselling the model system. 2. The authors demonstrate that dumbbell particles invert the direction of swimming from upon increasing the incident light intensity. Why do authors believe that the inversion of the propulsion direction is the result of change in bulk K'' of the microgel and PS? It is known that in AC electric field, the primary contribution to the net polarization (and corresponding fluid flow) originates from the electrical double layer. Since authors state in line 67, that the zeta potential of microgel doubles upon increasing temperature, I would expect the particles to propel with PS facing forward not as reported. Additionally, there is significant change in the volume of counter ionic double layer upon shrinking of microgel, those effects should also be incorporated in the model and corresponding discussion. A clear explanation of these aspects based on the relative surface potential, volume of double layer and polarization of PS and microgel halves is necessary. 3. Was the temperature constant upon the application of the field? And how was it monitored and controlled? The AC-electric fields tend to generate Joule heat, especially with low counter ion concentration (high resistance), which increases the temperature of the dispersion significantly. Given the nature of experiments, it is critical to clarify the effect of field on the dispersion temperature. 4. What is the role of fluid flow at the surface of the experimental chamber? Such flows are known to influence the motion of colloidal particles in AC electric field (as shown by Prof. Ning Wu's work). 5. How do authors know if the motion presented in Fig. 5b is a helix or curvilinear motion confined to a plane? What is the origin of such motion, and what are the effect of gravity, electric field gradients and surface flows on the motion? Reviewer #2 (Remarks to the Author): The authors describe the assembly and behavior of artificial microswimmers made from a combination of polystyrene (PS) and polyNIPAM particles and propelled by induced-charge electrophoresis. A shape change and response to the external driving AC field can be triggered by light, through local heating of the PS particle and subsequent collapse of the polyNIPAM particle. This experimental setup is clever and effective. The only minor comment is that I would not consider these particles as "autonomous" microswimmers because they rely on the external field.
The shape-change induced adaption of the motion is conceptually very exciting, although I find the observed change in magnitude and direction of the speed or rotation direction slightly underwhelming. Still, it is a clear effect and the authors demonstrate that this can be used to accumulate particles in a predefined area. The results are sound and well documented, the manuscript is written in a clear manner. I recommend publication in Nature Communication in its current state.
Reviewer #3 (Remarks to the Author): The manuscript by Alvarez et al describes a new type of colloidal motors that changes their propulsion speed under AC electric field when the particles are illuminated, as their shape is altered. The work is built upon many previous work where dielectric particles of various shapes can have different response to the electric field, which drive the particle to swim, and where particle shape can shift by some stimuli. The combination of the two seems to work very well in this work, creating a new system that possesses an internal feedback mechanism, resembling examples found in biology. The cool part of the idea is that, although PNIPAM is known to change shape according to temperature, such change here is induced by a nearby fluorescent particle that transfers energy harvested from light. Another point I do not appreciate before reading the paper is that upon changing shapes, the PNIPAM particle's dielectric constant and zeta potential have changed to such an extent that the EHD flow would reverse its direction. This is a key point related to the reversal of the motor's swimming direction. The data presentation and argument are convincing, yet a few of my comments need to be addressed before I would fully support its publication.
1. The energy transfer from the PS to the PNIPAM and raising local temperature is the key concept that makes the system work. I am wondering how this can be verified, and how to rule out the possibility that the whole illuminated area is heated, which does not seem to be uncommon. Some sort of control experiments should be designed: for instance, if PS without dye molecules can be used and under same illumination condition, the propulsion speed of the motor is not changed.
On a related note, in the experiment, is the base temperature of the system controlled, or the experiment is conducted under room temperature? It seems unlikely that, if the experiment is conducted at a low temperature, say 20 degree C, the same fluorescent non-radiative transfer will still be able to heat the PNIPAM above 32 degree C. Will the Au coating on the device contribute to some extent the temperature of the system? 2. As fluorescent particles are used, I am wondering if dye bleaching would cause some inconsistence/problems? 3. ICEO/EHD. The authors introduce the particle propulsion mechanism using ICEO but later focus the discussion on EHD. Although the two describe the same physics, both considering the effect of electric field on ionic charges. However, the two are generally considered different electrokinetic mechanism in the context of micromotors. In papers by Wu. N and Velev. O, these are differently treated, with ICEO describing Janus particle with a gold lobe whereas EHD describes pure dielectric dimers. Their frequency responses also seem different. The authors can consider this point for revision so these terms may be aligned with previous papers.
On a related note, did the authors try a different AC frequency? Will the conclusion still be valid? 4. In supplementary video 3, there seems to be quite a fraction of particles that do not respond/swim. Can the author give an explanation? 5. Conceptually, will simply change the system temperature considered a sort of internal feedback? Temperature is also an environment parameter that bacteria/particles can feel, then they change shape and response with a different motility.
We report here a detailed response to all questions and comments from the reviewers. The        However, the strategy that we report is in principle applicable to systems where no external in-72 tervention is needed, i.e. for chemically-powered particles where reconfiguration is triggered by 73 spontaneous temperature changes. Nonetheless, the use of external stimuli affords much better 74 control in this early development phase and we are working towards extending our strategies 75 to closed systems. We therefore regard our findings as a necessary step to take us closer to the 76 realization of autonomous microswimmers.

77
2 Finally, we remark that the particle reconfiguration is not externally guided during the experi-78 ments but it is rather encoded during their fabrication; the particles are designed to reconfigure 79 along a specific pathway during synthesis and they spontaneously follow this path upon exposure 80 to the right stimulus. The inclusion of different responsive components during fabrication will 81 enable the future the design of active particles with more complex spontaneous reconfiguration 82 pathways. 83 We have now revised the abstract and introduction and moved a discussion on the above-84 mentioned topics to a dedicated section in the conclusions. We hope that our revisions satis-85 factorily address the concern of the reviewer.  We have now included additional data on this point in the Supplementary Information (Fig.S4).

127
The temperature remains constant during the experiments as the sample cell is in contact with 128 the large heat sink constituted by the microscope operated in a temperature-controlled labora-129 tory. In order to support this statement, the global temperature of the dispersion is measured   We thank the reviewer for identifying our erroneous nomenclature. At the applied AC field 161 frequencies and amplitudes, the motion of the L-shapes is confined to a plane and exhibits a 162 curvilinear/trochoid-like motion. [1,2] The origin of this motion is to be found in the existence can have different response to the electric field, which drive the particle to swim, and where 202 5 particle shape can shift by some stimuli. The combination of the two seems to work very well in 203 this work, creating a new system that possesses an internal feedback mechanism, resembling ex-204 amples found in biology. The cool part of the idea is that, although PNIPAM is known to change 205 shape according to temperature, such change here is induced by a nearby fluorescent particle 206 that transfers energy harvested from light. Another point I do not appreciate before reading 207 the paper is that upon changing shapes, the PNIPAM particleâȂŹs dielectric constant and zeta 208 potential have changed to such an extent that the EHDflow would reverse its direction. This is  On a related note, in the experiment, is the base temperature of the system controlled, or 241 the experiment is conducted under room temperature? It seems unlikely that, if the experiment 242 is conducted at a low temperature, say 20 degree C, the same fluorescent non-radiative transfer 243 will still be able to heat the PNIPAM above 32 degree C. Will the Au coating on the device con-244 tribute to some extent the temperature of the system? to generate the data shown in Figs. S12 and S13. shows that no significant bleaching is observed up to 500 s (interval between frames of 5 sec-266 onds for 500 s). In the experiments such illumination power density is used for up to 2 minutes.

267
The stability of the dye is also supported by the consistent recovery of initial velocities during  lations are now reported in Fig. S14. Essentially, the propulsion reversal is maintained but the 292 overall propulsion speed rapidly decays to zero for frequencies approaching 10 4 Hz. We have 293 now added a comment in the main text and refer to the data in the Supplementary Information. The particles that do not respond or swim might not have a microgel attached (fabrication defect) 299 and therefore not generate an asymmetric EHD flow, or they may simply stick to the substrate 300 due to local imperfections of the silica coating. 301 302 5. Conceptually, will simply change the system temperature considered a sort of internal 303 feedback? Temperature is also an environment parameter that bacteria/particles can feel, then 304 they change shape and response with a different motility.

306
As we now discuss in the conclusions section of the manuscript, we expect that a conceptually 307 similar behavior is to be expected in a system where temperature variations are not externally in-308 duced, e.g. as we do by illumination, but stem from internal, spontaneous changes which are dy-309 namically coupled to self-propulsion. Finally, we remark that we consider temperature changes 310 as a stimulus, either internally generated or externally induced, but the feedback emerges from 311 the particle reconfiguration, which we control during particle synthesis, and which proceeds 312 spontaneously upon the sensing of the temperature stimulus. I thank the authors for carefully addressing my concerns and comments. The manuscript is much improved and now provides sufficient discussion on the scientific principles driving the propulsion. I believe the manuscript will generate interesting discussion on the topic and will be of interest to wide audience in active colloids community.
Reviewer #3 (Remarks to the Author): I have examined the revised manuscript. The authors have addressed my previous comments and concerns in details with new experiments, extended discussions, and supplemented figures and movies etc. I have also looked through other reviewers' comments, which the authors have responded to properly, especially with the added discussion on the mechanism accounting for the reversal of the velocity of the swimmer.
I am satisfied with the current version and therefore recommend its publication.