3D printed self-driven thumb-sized motors for in-situ underwater pollutant remediation

Green fuel-driven thumb sized motors (TSM) were designed and optimized by 3D printing to explore their in-situ remediation applications in rare studied underwater area. Combined with areogel processing and specialized bacteria domestication, each tiny TSM could realize large area pollutant treatment precisely in an impressive half-automatically manner.

Scientific RepoRts | 7:41169 | DOI: 10.1038/srep41169 The basic principle of self-propulsion is that bubbles were generated in the opposite direction of the motion of the TSM, which was driven by the reaction force (F) of the bubbles. The magnitude of the thrust is equal to the rate of change of momentum in the unit time of the fluid flowing through the channel of the thruster 29 Here ρ is fluid density (kg/m 3 ); q is the flow of water through the flow channel (m 3 /s); V j is the velocity of bubbles (m/s); V i is the velocity of water outflow; V s is the velocity of TSM (m/s); R is the resistance of advanced; C R is the resistance coefficient; Ω is wetted surface area (m 2 ), Ω = π D L, D is diameter of TSM (m), L is the length of TSM (m). When F ≧ R, the TSM would moved forward under the action of thrust. The inlet was located to the head of TSM, water flow through the water inlet channel to reach the outlet channel, finally outflow from the outlet. According to the principle of momentum theorem and Bernoulli's equation 30,31 , the flow of water from the outlet can also produce a thrust. According to the above equations, the movement of the TSM can be tailored by several parameters, including external shape and internal structure; the speed of the flow and the velocity of the bubble. For instance, the velocity of the TSM can be improved by the optimization of streamlined shape 32 . On the other hand, their cruise duration is dependent on the drug loading capacity. Figure 2(b,e) shows the autonomous motions of four different 3D designed TSM in the water. A long tail of CO 2 bubbles were ejected from water outlet of each TSM. These bubbles would engender a strong momentum and propelled the TSM forward with a remarkable average speed of 1.29 ± 0.04 cm/s, as summarized in the Table S1 (Movie S1). These different types of design had different velocity and cruise duration. Designs (T1, T2) with superior stream line possessed superior speed. Because design (T3) owned many small holes, which could enhance the reaction of the fuel with water, therefore, it could swim the fastest. On the other hand, due to the quick fuel consumption, this design (T3) couldn't sustain a long time. The shape of design (T4) was wider, which increased the resistance. Its internal structure was also more complex which heavier the body weight, thus the speed was slowed down naturally. Similarly, the effects of bacteria release from TSMs are showed in Figure S2, they mainly depend on the design. As could be seen from the Table S1, design (T1) has the best overall performance compared to others. Hence, it was chosen in the subsequent study for microbial transport. Attributing to the Fe 3 O 4 , the as-obtained TSMs have a good response to the external magnetic field, and their motion could be modulated magnetically. Meanwhile, they could realize the movement way of "stop-and-go" or "stop-turn-and-go" by adjusting the magnetic field direction. As illustrated in Fig. 2g and h, to acquire the track lines of the TSMs movement, four different colored dyes were mixed with the fuel of each TSM respectively. When the fuel reacted with water, these dyes would overflow with the ejected bubbles, subsequently formed visible track lines. With the aid of magnetic field, they could move in a straight line (Fig. 2h).
Considering the print accuracy, currently, Stereo lithography technology (SLA) was chose to build TSM. Admittedly, there is a slight environmental pollution problems existed in photosensitive resin. Accidentally, we found those photosensitive resins formed TSMs were lipophilicity, and could selectively adsorb oil from water. Therefore, in some cases, they could be directly used for the small area oil remediation without any further surface modification. As shown in Figure S3(a-d), the TSM is hydrophobic, but could absorb oil quickly. As illustrated in Figure S3e, with the aid of magnetic guidance, such TSMs could perform "on-the-fly" action to pick up oil droplets. Early researches have been reported the surfaces with hydrophobic properties would hold considerable promise for removal the oil droplets and isolating other hydrophobic targets form waste water samples 33,34 .
However, the wastewater treatment efficiency was still limited by using the lipophilicity of photosensitive resin alone. More efficient materials are needed to improve the practical absorption performance of TSM, especially in the underwater area. Considering the limited load capacity of the TSM, this auxiliary material was expected to have both properties of low-density and high absorption capacity. Ultra light porous structure of aerogel, as one of the hotspots in the current material research, was thus synthesized and utilized for the subsequent study to enhance the adsorption efficiency. Scanning electron microscopy (SEM) depicts the porous structure of the jackfruit-aerogel (Fig. 3a) used in this study. According to the absorption result, this natural derived areogel could absorb over 400 times than its own weight out of oil, 1132 times diethyl ether, and 1375 times chloroform ( Figure S4). Besides, it is worth mentioning that, after pre-treated with the target reagent, this aerogel would exhibit impressive repellence to other liquid. The selective adsorption ability of TSM was thus significantly enhanced by this interesting "memory" adsorption characteristic, which might had a great significance to remove pollutants precisely in water. To confirm this capacity, chloroform and diethyl ether were chosen for experiment. At first, aerogel was pre-treated with chloroform, for the purpose of selective adsorption of chloroform in underwater area. As depicted in Fig. 3c, the different absorbing effect of areogel toward different solvents (chloroform, diethyl ether and water) could be observed visually. This aerogel had good adsorption capacity of chloroform; however, the absorption of water and diethyl ether was much less. Subsequently, underwater experiment was also conducted by such TSMs. In the presence of chloroform, water and diethyl ether, such TSMs would only absorb chloroform, and avoid water or diethyl ether adsorption (Fig. 3b,c). The relative video information was provided in the supporting information (Movie S2, S3).
Apart from pollutant absorption, it would be more meaningful if the TSM could offer in situ remediation directly in the underwater environment. The entire design of TSM was compact and tiny to realize a quick and convenient usage. However, this characteristic also brought problems in the limited amount of drug loading. If the carrier is conventional water treatment agents (active carbon, quartz sand, polymeric aluminum chloride, etc.), it's obviously difficult to effectively deal with a large area of water pollutants. Therefore, a "self-proliferated agent" concept was selected here, which was realized by using microbial degradation 35,36 . Herein, B.substilis was selected as the loading microbe that used for in situ water remediation. This strain can quickly dominant the underwater environment and effectively degrade petroleum-oil 37 , sulfur, nitrogen and chlorine pollutants 38 . What's more, B.substilis would not damage the human body, but even regulate the intestinal health of people by inhibiting the growth of pathogenic bacteria 39 . It is thus an ideal selection for the purpose of environmental safety. However, the original strain of B.substilis did not own enough degradation efficiency. To get more specialized strain, the experiment of screening and domestication was carried out. The results of biodegradability (Fig. 4) demonstrated that, the original strain were 70.43%, which was lower than domesticated strain (76.82%). More importantly, to achieve this effect, the original strain must be taken almost a month, while the domesticated strain just needed less than one third of that time. Three days at the beginning, two groups of rhodamine-B were degraded to 12.38% and  64.41% treat by original and domesticated strain respectively, and the control group had not significant change. In spite of the load amounts is limited, this self-proliferated "agents" could degrade more than ten thousand times pollutants compared to its initial loading volume. As shown in Figure S5, after seven consecutive days of degradation, the obvious color fading could be seen. Herein, the volume of bacteria liquid was 250 μ L, while the pollutants volume was 3000 mL.
In this study, we have presented the first example of self-propelled 3D printing TSM for in-situ underwater remediation. Instead of traditional H 2 O 2 fuel, such TSMs were propelled by the CO 2 bubbles, which were generated by green chemical reagents -C 4 H 6 O 3 and NaHCO 3 . The new green-driven motion capability would greatly enlarge the application fields, where the use of the traditional peroxide fuel is not convenient or safety. Incorporated with ultralight functional material and microbial acclimation, such 3D printed TSMs, though tiny and compact, have integrated two core capabilities for in-situ underwater remediation: high-efficient selective adsorption, and huge amount degradation. On the current stage, this technology is still in infancy. There are a lot of room for improvement, such as the cruising range; a practical feasible magnetic guidance manner; and how to deal with the pollution of complex components. With the advancement of material science and 3D printing technology, the above problems are expected to be solved gradually. In particularly, we expect to narrow the size of this functional TSM down to the micron level. By that time, these green-driven micromotors might not only apply in underwater area, but also have profound influences on diverse biomedical or industrial applications, such as biological targets isolation, in-vivo targeted drug delivery and anti-thrombotic therapy.

Methods
Fabrication of TSM. TSMs with ellipsoidal-like structure were fabricated by FSL3D Pegasus Touch 3D printer. The TSM was designed by Rhinoceros.

Preparation of aerogels.
Taking fresh jackfruit, removed the cores, the flesh were put into reaction kettle. Reaction kettle was transferred to a tubular furnace for pyrolysis, and then was heated to 180 °C for 12 h. After that, the preliminary aerogel was immerged in absolute ethyl alcohol, until the solution became colorless. Subsequently, this aerogel was put into water to remove the ethyl alcohol. Next, the aerogel was pre-freezed at − 180 °C for 24 h and freeze-dried 12 h in freeze dryer. Finally, the aerogel was obtained.