Photocatalytic Microplastics “On‐The‐fly” Degradation via Motile Quantum Materials‐Based Microrobots

Nano/micro‐plastics pollution is an emerging global concern. A variety of biodegradable polymers have been synthesized to enhance the degradation of plastic materials and thereby avoid their accumulation in the environment. However, even biodegradable polymers can accumulate in environments under specific conditions and present a potential hazard. Here, antimony sulfide‐based microrobots decorated with magnetite nanoparticles are designed for microplastics degradation. The propulsion of microrobots is enabled by two independent orthogonal physical modes via magnetic field and via light irradiation. Due to phoretic interactions, the microrobots exhibit affinity toward poly(3‐hydroxybutyrate) (PHB) and poly(lactic acid) (PLA) microplastics, which enables subsequent transport of the microplastics in a transversal rotating magnetic field. The photocatalytic activity of Sb2S3 quantum material provides microrobots with the ability to degrade the microplastics under UV light irradiation in the “on‐the‐fly” regime without the need for any fuel. This proof‐of‐concept work shows efficient capture, transport, and photocatalytic degradation of microplastics and paves the way toward their elimination, especially in water environments.


Introduction
We live in a plastic age. [1,2]Plastics have been employed in daily life since the 20th century and their use increases exponentially with the available technologies. [3]rom a chemical point of view, plastics are polymeric materials; their molecules possess a high molecular weight (up to millions of atomic mass units) as they contain a large number of repeating monomeric units linked by a chemical bond. [4]The design of individual monomeric units, their quantity and organization, ability to polymerize, submolecular orientation, doping, etc. provide these materials with great versatility. [5]owever, a disproportionate research effort of polymers has been devoted to the fate of plastic waste in the environment.The volume of plastic waste is increasing and the images of plastic pollution in nature are widespread. [6]Plastic waste generally decomposes into smaller and smaller parts that are not visible to the human eye and referred to as microplastics (size less than 5 mm) or nanoplastics (size less than 1 μm). [7]10] Considering that plastics are petroleumderived products, the resulting nano/micro-plastics and, eventually, monomer units are not biocompatible and can cause serious consequences to living organisms. [11]t is no wonder that one of the current trends in the plastics research is the search for biodegradable polymers.Biodegradable polymers show not only enhanced degradation abilities typically when exposed to enzymatic activity, but the degradation products are also environmentally friendly materials, such as methane, water, biomass, and carbon dioxide. [12,13]Biodegradable polymers can be divided into two main groups: 1) natural biodegradable polymers that are extracted directly from biomass (polysaccharides, polypeptides, or lipids) or directly from microorganisms, for example, polyhydroxyalcanoates, such as poly(3-hydroxybutyrate) (PHB) and 2) synthetic biodegradable polymers, including a huge variety of aliphatic polyesters such as polylactic acid (PLA) and polycaprolactone (PCL). [14]Specifically, PLA is gaining significant attention due to its strength, stiffness, thermoplasticity, biocompatibility, and good processability.It can be applied in textiles, packaging, electronics, agriculture, and the biomedical industry. [15]Moreover, PLA is one of the most commonly used thermoplastics for 3D printing via fused deposition modeling (FDM); suggesting that PLA-based plastic waste amount will even increase with increasing accessibility of 3D printing technologies. [16,17]It should be noted that successful degradation of PLA requires specific conditions such as high humidity, an oxygen-rich environment, and the presence of microorganisms.Such conditions cannot be provided in all environments and PLA nano/micro-plastics can accumulate for longer periods than desired. [18]][28] Lately, there have been efforts to propel microrobots in multiple modes to improve navigation and enable the performance of tasks in different environments. [29]The nano/micro-robots are usually designed for a specific task, such as substance capture, transfer, delivery, or degradation.[36] Degradation can be performed using specific enzymes immobilized on the surface of microrobots that can cleave polymer chains. [37]In different approaches, microplastics can be degraded photocatalytically using semiconductor-based microrobots with photocatalytic properties. [38,39]etal chalcogenides are at the forefront of inorganic semiconductors research due to their unique optical and electrical properties. [40]They have been applied as photosensitizers for solar cell fabrication, [41,42] phase-changing material for memory devices fabrication, [43,44] for battery devices fabrication, [45,46] and as photocatalysts for water splitting, [47,48] among others.Especially, antimony sulfide (Sb 2 S 3 ), which is one of the prominent quantum materials, is one of the most promising candidates due to its low toxicity, band gap, stability under ambient conditions, and abundant character. [49,50]Our group has previously reported on Sb 2 S 3 -based microrobots that showed great applicability for photocatalytic degradation of food dyes. [51]ere, we present a study of the photocatalytic degradation of biodegradable polymers, namely, PLA and PHB.Degradation was performed by applying Sb 2 S 3 -based microrobots that were designed for dual mode propulsion.The photocatalytic Sb 2 S 3 material not only provided the microrobots with the ability of light-induced propulsion, but also enabled phoretic attraction with microplastics and consecutive photodegradation abilities.Decoration of the Sb 2 S 3 material with magnetite nanoparticles (Fe 3 O 4 NPs) made the microrobots navigable in a magnetic field, allowing them to navigate toward a particular part of the solution and then be retrieved and reused.

Results and Discussion
The microrobots were prepared by decorating Sb 2 S 3 microparticles with magnetite nanoparticles (Fe 3 O 4 NPs).In particular, the preparation of the Sb 2 S 3 material was carried out by reacting antimony chloride (SbCl 3 ) with cysteine in the presence of tartaric acid as previously reported. [52]To decorate the resulting microrobots with magnetic particles, Fe 3 O 4 NPs were added to the precursor solution prior to reaction as suggested in Figure 1A.The reaction was carried out in a microwave reactor, which significantly decreased the reaction time to 30 min.For comparison, a standard hydrothermal synthetic approach toward the synthesis of pure Sb 2 S 3 material takes up to 12 h. [52]Moreover, it is worth noting that the microwave-assisted synthetic approach makes eventual mass production easily scalable, low-cost, reproducible, and environmentally friendly. [53]o evaluate the morphology of the resulting microrobots, a scanning electron microscopy (SEM) technique was employed.Figure 2A presents microrobots of urchin-like morphology with an average diameter of 19 ± 5 μm.Elemental mapping by energydispersive spectroscopy (EDS) further confirmed the presence of antimony and sulfur (Figure S1, Supporting Information).Interestingly, no iron signal coming from Fe 3 O 4 NPs was revealed, suggesting that the final concentration of Fe 3 O 4 NPs is in trace amounts.The amount of iron was detected by inductively coupled plasma optical emission spectroscopy (ICP-OES) coupled with electrothermal vaporization (ETV).The iron concentration is of 550 ± 20 ppm and, assuming that iron is bound only in Fe 3 O 4 NPs, the resulting concentration of Fe 3 O 4 within microrobots is ≈0.08 wt%, which was shown to be sufficient for efficient navigation of microrobots in the magnetic field as discussed below.
The crystalline structure of the microrobots was confirmed by recording powder XRD diffractograms (Figure 2B).The crystalline nature suggests that the microwave reactor is suitable for microrobots preparation without requiring any additional annealing steps as typically needed in the case of traditional hydrothermal synthetic approaches. [52]The diffractogram of the microrobots was identical to the diffractogram of pure Sb 2 S 3 material and clearly no influence of Fe 3 O 4 NPs was observed on the resulting crystalline structure of the microrobots.Similarly, no additional diffraction referring to the presence of Fe 3 O 4 NPs was observed in the resulting diffractogram, which is probably caused by its low concentration below 0.1 wt%.To further characterize the microrobots, UV-vis and Raman spectra were recorded.UVvis spectra of microrobots and of pure Sb 2 S 3 material showed an onset of absorption around 770 nm (Figure S2, Supporting Information).Presuming an indirect transition, an optical band gap of 1.61 eV was estimated from the Tauc plot (Figure 2C), which is consistent with previous studies. [49]Raman spectroscopy revealed signals of antisymmetric Sb-S stretching at −297 and −276 cm −1 .The symmetric and antisymmetric S-Sb-S bending signal was detected at −249 and −186 cm −1 , respectively. [44,54]either shifts nor broadening of the peaks was observed when the Raman spectra of microrobots and pure antimony sulfide microparticles were compared (Figure 2D).It must be emphasized that Fe 3 O 4 NPs do not exhibit any specific peaks in the Raman spectrum (Figure S3, Supporting Information) that would be present in the resulting spectrum of microrobots confirming their decoration.To conclude, the presence of Fe 3 O 4 NPs did not influence the optical properties of the microrobots.
Decorating the microrobots with Fe 3 O 4 NPs enabled their navigation in a magnetic field.The magnetic field-induced propulsion of microrobots was studied using a home-made transversal rotating magnetic field as described in a previous report. [55]The  influence of frequency was evaluated by applying the frequency in the range of 0-5 Hz at a fixed amplitude of 3 mT (Figure 3).As expected, the average speed increased with increasing frequency.The propulsion of the microrobots was clear when even 0.5 Hz was applied, showing an average velocity of 3 ± 1 μm s −1 and a linear trajectory (Figure 3B).With increasing frequency up to 3 Hz, the average velocity increased to 9 ± 4 μm s −1 .When the frequency was increased further, the average velocity started to decrease, having reached the highest accessible magnetic torque.At this point, the microrobots were no longer able to rotate in-sync with the magnetic field. [56]An example of a swinging microrobot at the frequency of 3 Hz is shown in Figure 3C.
The microrobots also exhibited light-induced propulsion in a fuel-free environment when irradiated with UV light.However, as Figure 4 demonstrates, the self-propulsion was rather weak, showing the average velocity of 0.6 ± 0.3 μm s −1 when irradiated with UV light.Non-Brownian motion was proved by plotting the mean squared displacement (MSD) of the microrobots (Figure 4C) and comparing it with the MSD curve of the nonirradiated sample.Evidently, the semiconductive properties of  Sb 2 S 3 material enabled light-induced propulsion of the microrobots.This mode of propulsion is independent on the magnetic field-induced propulsion mode; therefore, the microrobots can be propelled in a specific environment in multiple ways, which is necessary in multiple applications. [29]To explore the mechanism of the propulsion deeper, electrochemical properties of microrobots were studied by determining Tafel plots (Figure S4, Supporting Information).The measurements were performed in distilled water under dark conditions and under UV light illumination.A strong shift of about 330 mV in the positive direction occurred when the sample was irradiated with UV light, suggesting the generation of photovoltage. [57]Assuming that UV light interacts with the semiconductive material of the microrobots, electrons and holes are generated.They can recombine or eventually migrate to the surface and react with water or oxygen molecules to form superoxide and hydroxyl radicals, causing a localized electrolyte gradient.The formation of •OH radicals has been successfully demonstrated in the previous study on Sb 2 S 3based microrobots by using ethylenediaminetetraacetic acid as a scavenger. [51]The photodecomposition on the non-symmetrical microrobots induces a non-uniform electrolyte gradient that can propel the microrobots via self-diffusiophoresis as in this case. [58]s expected, the addition of hydrogen peroxide (H 2 O 2 ) caused an increase in the mean velocity due to its photocatalytic decomposition under UV irradiation.H 2 O 2 interacts with holes to produce protons and oxygen, and electrons to form hydroxide anions.The formation of an ionic gradient is stronger and propulsion is significantly enhanced compared to the non-fueled experiment. [59]As Figure 4A shows, the addition of 0.5 wt% H 2 O 2 fuel increased the average velocity about two times compared to the non-fueled microrobots only under UV irradiation.The average velocity also increases with an increasing concentration of fuel.The average velocity increased up to 1.6 ± 0.2 μm s −1 when using 2.5 wt% H 2 O 2 fuel.The MSD curves (Figure 4D) also showed the non-Brownian motion of the fueled microrobots.It is worth noting that the propulsion of the microrobots was not observed under dark conditions even if the fuel was applied.Figure 4B demonstrates typical trajectories of a non-specific shape, which is most likely caused by the non-symmetric shape of the microparticles. [32]aking into account the toxic character of H 2 O 2 [60] and its possible issues with large-scale applications, further experiments were carried out without any fuel; however, the propulsion abilities of the microrobots were not as significant as when the fuel was applied.
Observing the microrobots for an extended period using an optical microscope, schooling behavior was observed as the microrobots tended to form agglomerates over time (Figure 4E).After UV irradiation was applied on the aggregates, the microrobots spread out and were observed as individuals.When the UV irradiation was switched off, the agglomerates of microrobots were formed again.This phenomenon was already observed in previous work with light-active semiconductors such as TiO 2 [59] or organic semiconductors. [33]Light irradiation causes the formation of radicals and ionic species (such as H + , •OH, and •O 2 − ), and strong electrostatic repulsions occur among the microrobots, forcing the microrobots to spread within the sample. [33,58,59]After the irradiation is switched off, the generated ionic species diffuse and form an electro-osmotic flow that forces the microrobots to agglomerate again. [58]aking these interesting properties into account, we applied the microrobots as photocatalysts for microplastics degradation.Because microplastics are present in the environment in a form of nano/micro-particles that typically shows limited solubility in an aqueous environment, their collection and degradation are challenging.As microplastics, microparticles of the aliphatic polyester derivatives PHB and PLA were chosen.Both PHB and PLA are gaining a significant attention due to their non-petrochemical origin, which makes them "environmentally friendly."Although PHB and PLA are considered biodegradable polymers, their decomposition in nature leads to the formation of microplastics that can remain in the environment for long periods.For example, the degradation rate of biodegradable plastics is comparable to their petrochemical analogs in the marine environment. [61]Specifically, biodegradation of PLA requires very specific conditions to take place, such as high humidity, an oxygen rich environment, elevated temperature, and even the presence of microorganisms. [18]he adsorption of microplastics on microrobots is necessary to enable successful degradation. [38]Therefore, the interaction of microrobots with PHB and PLA microplastics was studied in the next step.The PHB microplastics were of bacterial origin, spherical, of an average size 100-1000 nm, and tended to aggregate into clusters (Figure S5, Supporting Information).Microplastics of PLA were prepared by alcoholyzing crystalline PLA granulate followed by precipitating the polymer in deionized water.The resulting microparticles were of irregular morphology and ranged in size from 1 to 100 μm (Figure S6, Supporting Information).First, the zeta-potential of microrobots and individual microplastics was measured to determine the colloidal stability and potential affinity among the microparticles.The zeta-potential of the microrobots was determined to be −32 ± 5 mV, suggesting moderate stability in colloidal solutions.Indeed, as discussed above, the microrobots were dispersible in an aqueous environment and formed colloidal solutions.The zeta-potential of the PHB and PLA dispersions was of −32.6 ± 1.0 and −37.2 ± 0.6 mV, respectively.Because of the negative values of the zeta-potential of both microrobots and microplastics, there are no strong electrostatic interactions binding microplastics and microrobots together.However, due to photocatalytic activity, microrobots induce phoretic interactions and microplastics tend to agglomerate around them. [62,63]Figure 5 clearly shows that microplastics gather around clustering microrobots over time.After the sample is exposed to UV irradiation, the microplastics are repelled as a result of electrostatic forces generated upon the formation of ionic species.Repulsion is reversible as the local electric field induces electro-osmotic flow that moves the microplastics toward microrobots after the UV irradiation is turned off. [64]It is worth noting that this phenomenon was stronger in the case of employing PHB microplastics, which could possibly be due to their smaller size compared to PLA microplastics.To distinguish the microrobots from the microplastics, the borders of microrobots are colored in red in Figure 5.The original raw micrographs are shown in Figure S7, Supporting Information.
As microplastics tend to gather around microrobots, the possibility of transporting the microplastics in a magnetic field was tested.As expected, when the magnetic field was applied, the  microrobots tended to push the microplastics that were attached to their surface (Figure 6 and Video S1, Supporting Information).The original raw micrographs without the border coloring of microrobots are shown in Figure S8, Supporting Information.Clearly, the ability of orthogonal dual-mode propulsion of microrobots opens possibilities for the capture and navigation of the microplastics simultaneously.
Because the microplastics were apparently influenced by the ionic and radical species, that is, OH radicals, [51] generated by microrobots under UV irradiation, the ability to degrade the microplastics photocatalytically was studied.A typical sample for degradation was prepared by dispersing microplastics in deionized water and then treating the suspension with microrobots.It should be noted that no hydrogen peroxide was applied as fuel.The sample was then exposed to UV irradiation for 7 h and the degree of degradation process monitored by gel permeation chromatography (GPC).Considering the non-homogeneous nature of the suspension due to the presence of microplastics, only the status of the sample before and after the whole degradation process was monitored.Figure 7 shows the successful degradation of microplastics using microrobots.The non-degraded microplastics were also characterized by determining the average molecular weight (labeled as "Reference samples").For comparison, the control experiment was performed by exposing the microplastics to UV irradiation for 7 h without the presence of microrobots.As Figure 7A,C shows, there is a drop after irradiating microplastics with a UV lamp compared to a non-treated reference sample.The partial degradation of the control sample is most likely induced by the UV irradiation and the presence of oxygen that typically lead to oxidation of PLA and to UV-or thermal-induced degradation. [65]However, degradation with microrobots is clearly more efficient because of the formed radical species that induce radical degradation. [66]Considering that the degradation was performed at normal conditions in the presence of oxygen and under UV irradiation that caused local heating of the solution, various degradation mechanisms (oxidation, thermal degradation, radical-induced degradation) are expected to take place. [65,66]The non-specific mechanism of the degradation is also supported by monitoring the change in polydispersity.The decrease in polydispersity after degradation of microplastics suggests non-specific random cleavage of polymer chains leading to the formation of polymers of different chain lengths.During the course of the study, changes in pH were monitored before and after degradation.The pH value dropped from 6.5 to 3.1 and from 5.5 to 3.2 in the case of PHB and PLA, respectively, suggesting that the concentration of acidic carboxylic groups increased with the cleavage of polyester groups.
To eventually increase the efficiency of microplastics degradation, the degradation experiment of PHB microplastics was performed in rotating magnetic field by placing the photodegradation cell on a magnetic stirrer during the procedure.It was expected that enabling more efficient locomotion of the microrobots would enhance the degradation efficiency as the average velocities in the magnetic field (Figure 3A) are higher in comparison to the average velocities only under UV light irradiation (Figure 4A).Indeed, the molecular weight of PHB microplastics dropped from 1.6 × 10 4 to 1.1 × 10 4 g mol −1 after the photodegradation procedure by repeating the experiment without any external magnetic field and in the magnetic field, respectively.However, it should be noted that both resulting values are within the standard deviation of typical photodegradation experiments that showed mean value of 1.2 × 10 4 ± 0.1 × 10 4 g mol −1 (Figure 7A), and therefore, the positive effect of the magnetic stirring cannot be confirmed with absolute certainty.
In the last step, the microrobots were isolated after the PHB photodegradation experiment and recycled to demonstrate their reusability.To isolate the microrobots, water was evaporated and PLA was dissolved in chloroform.The resulting colloidal solution was centrifuged and the microrobots were collected for a next cycle of the photodegradation experiment.The microrobots indeed showed ability to photocatalytically degrade PHB microplastics after the recyclation procedure.The molecular weight of PHB polymers was of 8.5 × 10 4 g mol −1 after reusing the microrobots for photocatalytic degradation which is shown to be less efficient than when using unused microrobots (1.2 ± 0.1 × 10 4 g mol −1 ).However, comparing the results to the control experiments when no microrobots were applied (1.3 ± 0.1 × 10 5 g mol −1 ), the microrobots are still able to photocatalytically degrade microplastics even after the recyclation process.

Conclusion
In summary, we have designed photoactive Sb 2 S 3 -based microrobots propelled via two orthogonal physical modes: magnetic field and light irradiation.The possibility to navigate microrobots using a magnetic field enables their navigation toward a specific location, whereas light-induced propulsion enables the photodegradation of pollutants in an "on-the-fly" regime in a nonfueled environment.Eventually, light-induced propulsion could be enhanced by employing H 2 O 2 as fuel.Moreover, Sb 2 S 3 material provided the microrobots with the ability to phoretically interact with biodegradable PHA and PLA microplastics, and, ad-ditionally, induce degradation of the microplastics.This proofof-concept study shows that Sb 2 S 3 -based microrobots show great potential in the capture, transport, and photodegradation of microplastics.

Experimental Section
Materials: Hydrogen peroxide (30 wt%), Fe 3 O 4 NPs, tartaric acid, antimony chloride, and cysteine were purchased from Merck.Ethanol and chloroform were purchased from Penta.PLA was purchased from Total Corbion and PHB was purchased from TianAn.
Preparation and Characterization of Microrobots: Magnetite nanoparticles (Fe 3 O 4 NPs, 7 mg) were added to the tartaric acid solution (20 mL, 0.18 mol L −1 ) and a colloidal solution was formed.The solution was sonicated for 10 min.Following the addition of 0.5 mmol SbCl 3 , the solution was sonicated for 30 min.Finally, 0.7 mmol cysteine was added and the solution sonicated for 10 min.The precursor solution was transferred to the microwave reactor (Anton Paar, Monowave 400) and reacted for 30 min at 180 °C while magnetically stirred (600 rpm).The resulting black precipitate was collected by centrifugation (4000 rpm, 4 min) and washed with distilled water once and with ethanol three times.Purified microrobots were dried in a vacuum oven at 40 °C overnight.The Sb 2 S 3 microparticles were prepared by the same procedure without adding magnetite nanoparticles to the precursor solution.
The morphology of the microrobots was studied using a SEM (Tescan, MIRA 3 XMU) equipped with an EDX detector (Oxford Instruments).An X-ray powder diffractometer (Rigaku SmartLab 3 kW) was used using a Cu K radiation source to determine the crystalline structure of the samples.The iron content was determined by ICP-OES combined with electrothermal vaporization (ETV) using the ETV system 4000c (Spectral Systems) and the Spectro Arcos spectrometer (Spectro Analytical Instruments).The resulting iron concentration was calculated using emission signals at 218.7 and 373.5 nm.Optical characterization of the microrobots was performed by recording reflectance spectra using a UV-vis spectrophotometer (Jasco V-750) equipped with an integrating sphere (ISV-922).Furthermore, an IR-Raman spectrometer (Bruker RFS 100) with a Nd:YAG laser source (1064 nm, 100 mW) was used to obtain Raman spectra.The zeta-potential was determined using a Zetasizer Ultra (Malvern Panalytical Ltd.).Measurements for the Tafel plots were made using a potentiostat (Metrohm, Autolab) equipped with a light-emitting diode source (LedEngin Inc., 365 nm, 700 mA, radiant flux 3.3 W).The sample was drop-casted on the ITO substrate.The measurement was performed in demineralized water using a Pt counter electrode and Ag/AgCl reference electrode.
Tracking of Microrobots: The magnetically induced locomotion of the microrobots was studied using an inverted optical microscope (Nicon, Eclipse Ts2R) equipped with a digital camera (Basler ace acA1920-155uc).The microscope was coupled with a home-made setup for a rotating transversal magnetic field.The videos for tracking were recorded for 20 s at 25 fps at 10× magnification and speed was calculated using ImageJ software with a TrackMate plugin.The typical trajectories were plotted by NIS software.
Light-induced locomotion was observed using an inverted optical microscope (Nicon, Eclipse Ti2) equipped with a digital camera (Hamatsu, C13440).A light source (CoolLed, pE-300 lite) was used to illuminate the sample and filter cubes were applied to select the UV range of 361-389 nm (463 mW cm −2 ).The videos for tracking were recorded for 10 s at 25 fps at 40× magnification.The typical trajectories were plotted using the NIS software.Videos were processed using ImageJ software with the TrackMate plugin [67] and the average speed and MSD were calculated using Python code. [68]Hydrogen peroxide was eventually added as fuel to determine its influence on the propulsion under UV irradiation.The H 2 O 2 concentration was varied in the range 0-2.5 wt%.

Degradation of Microplastics:
The PHB microplastics were used as received.The PLA microplastics were prepared by alcoholysing PLA granulate (50 g) in ethyl lactate (200 g) with para-toluenesulfonic acid (1g) at 154 °C under microwave irradiation for 15 min.The resulting solution was precipitated in deionized water.The polymer particles were sieved over 0.2 mm sieves (70 Mesch) after filtration and vacuum drying (50 °C, 1 kPa).
A typical degradation experiment was performed by dispersing the polymer in distilled water (0.6 mg mL −1 ) in a sonication bath.Following the addition of microrobots, a concentration 0.4 mg mL −1 was reached, and the resulting suspension was vortexed for 15 min.The sample of a total volume of 10 mL was exposed to UV irradiation (LedEngin Inc., 365 nm, 700 mA, 4.1 W) for 3, 5, and 7 h.As no significant or rather small drop in the molecular weight was observed after 3 and 5 h of exposure of the sample to UV irradiation, the photodegradation experiments were performed for 7 h each time.Finally, water was evaporated in a nitrogen flow at 60 °C and the solid material was dissolved in chloroform at 70 °C.A control experiment was conducted under the same conditions without the presence of microrobots in the sample.To recycle the microrobots, the photocatalytic reaction mixture was processed by evaporating water and dissolving microplastics in chloroform at 70 °C.The resulting colloidal solution was centrifuged to isolate the microrobots that were dried in vacuum overnight before using them for repeated photocatalytic experiment.

Figure 1 .
Figure 1.Schematic illustration of the photocatalytic degradation of microplastics.A) One-pot fabrication of microrobots using a microwave reactor.B) Creation of microplastics during the degradation process of plastic waste and chemical structure of polylactic acid (PLA) and poly(3-hydroxybutyrate) (PHB).C) Light-driven propulsion of microrobots and their photocatalytic properties toward the degradation of microplastics.

Figure 2 .
Figure 2. Characterization of microrobots.A) SEM image of urchin-like microrobots.B) Powder XRD characterization of microrobots and comparison of their diffractograms with diffractograms of unmodified Sb 2 S 3 microparticles and Fe 3 O 4 NPs.C) Tauc plot calculated from UV-vis spectra of microrobots and unmodified Sb 2 S 3 microparticles.D) Raman spectra of microrobots and unmodified Sb 2 S 3 microparticles.

Figure 3 .
Figure 3. Navigation of microrobots in a transversal rotating magnetic field.A) Velocity of microrobots in a transversal rotating magnetic field at 3 mT.B) Typical trajectories of microrobots in a transversal rotating magnetic field at 3 mT at different frequencies; duration: 20 s.C) Swinging microrobot in a transversal rotating magnetic field at 3 mT and 3 Hz.

Figure 4 .
Figure 4. Locomotion analysis of light-driven microrobots.A) Velocity of microrobots and its dependence on UV irradiation and fuel concentration.B) Typical trajectories of fueled microrobots under UV irradiation, trajectories were observed for 10 s.C,D) MSD of microrobots in the presence of 0 and 2.5 wt% H 2 O 2 , respectively, under UV irradiation and dark conditions.E) Collective behavior induced by UV light of microrobots in the presence of 2.5 wt% H 2 O 2 .

Figure 5 .
Figure 5. Affinity of microrobots toward PHB and PLA microplastics.Influence of UV irradiation over time.The borders of microrobots are colored in red to differentiate them from microplastics.Scale bar: 50 μm.

Figure 6 .
Figure 6.Affinity of microrobots toward PHB and PLA microplastics.Collection and transport of microplastics using microrobots.The borders of microrobots are colored red to differentiate them from microplastics.Scale bar: 50 μm.

Figure 7 .
Figure 7. Degradation of microplastics.Monitoring the degradation of PHB (A,B) and PLA (C,D) using light-propelled microrobots (sample) and UV irradiation (control).The reference shows the characteristics of non-treated polymers.