Developments in synthetic biology usually bring the conception of individual artificial cells. A key feature of living systems is, however, the interaction between individuals, in which living units can interact autonomously and display a role differentiation such as the case of entities chasing each other. On the other hand, droplets have become a very useful and exciting medium for modern microengineering and biomedical technologies. In this Perspective, we show a brief discussion-outlook of different approaches to recreate predator–prey interactions in both swimmer and crawling droplet systems toward a new generation of synthetic life with impact in both fundamental insights and relevant applications.

Implementing and controlling the basic operating principles of life in synthetic embodiments is one of the grand challenges of modern science. The work toward this ambitious goal has steadily progressed over the past few years, with a focus on scenarios of individual artificial cells.1,2 Going further, the interaction of individuals with each other is a hallmark of life. Evolution even leads to a competition directly through predatory-type behaviors. While there is no single conception of what constitutes a living system, a defining characteristic is, therefore the often non-reciprocal interplay between individuals.3 In other words, living units can process information to recognize and interact autonomously, which implies that during interaction events they display differentiated roles, such as the case of predator and prey (see Fig. 1). An ubiquitous situation of non-reciprocal interactions in nature is indeed that of entities chasing each other, ranging from the micro- to the macroscale; for example, the white blood cells of the immune system actively targeting and engulfing foreign invaders.4 In addition, the energy that drives these motility processes needs to be locally generated, which means that the active minimal units include a transductional chemo-mechanical machinery. A key feature of all forms of life is the fact that the processes occur autonomously by extracting energy from their surroundings at a single unit level and transforming it into local mechanical work.5 

FIG. 1.

Schematic representation of predator–prey interactions. Predatory behaviors are often simply thought of as a non-reciprocal attractive interaction such as a predator targeting and catching a prey. However, the predator–prey behaviors can involve both attractive and repulsive interactions, where the predator is attracted to the prey but at once the prey is repelled by the predator.

FIG. 1.

Schematic representation of predator–prey interactions. Predatory behaviors are often simply thought of as a non-reciprocal attractive interaction such as a predator targeting and catching a prey. However, the predator–prey behaviors can involve both attractive and repulsive interactions, where the predator is attracted to the prey but at once the prey is repelled by the predator.

Close modal

In a separate but related thread, droplets (discrete, very small, fluid volumes) in either bulk fluids or placed over surfaces are currently considered versatile implements in modern microengineering and biomedical applications. The first case involves liquid–liquid interfaces,6,7 while the second case involves the combination liquid–solid–air interfaces,8,9 which provides varied opportunities to design programmable droplet transport. Droplets are also envisaged as elementary active units to harbor life-like features. In recent years, the ability to handle droplets has been significantly improved and impressive developments of droplets as simple machines to attain complex tasks have been achieved.10,11 This prompted dreams of carrying out a next generation of droplet systems that closely mimic life interaction behaviors beyond individual protocells. Attempts to create such “artificial life” will certainly contribute to understand many fundamental questions on how life may have emerged and as well evolved, while elucidating its key design principles. Furthermore, the generation of droplet frameworks that emulate living systems behavior may provide the prospect of novel platforms for a wide variety of applications.

To conclude, it is worth to stress that a minimal droplet system reminiscent of biological interplay behaviors should at least contain sensory-motor abilities that combine role differentiation and transformation of chemical energy into mechanical work. In this Perspective, we discuss the current and rising approaches to how autonomous droplets can influence each other starting from the archetypal predator–prey interactions. Considering the above classification (droplets in liquids/over surfaces), in what follows we characterize self-propelled droplets that move by “swimming” in liquids or “crawling” on surfaces.

Most laboratories conceive artificial cells as lipid vesicles containing complex individual machineries. However, synthetic cells do not need to look like natural cells. In fact, using a system with only a small number of molecular components, an interacting protocell ecosystem that evinces a simple form of predatory behavior was successfully achieved, in which one protocell type preys upon another.12 The killing interaction process is established by charging coacervate microdroplets (predators) with proteases that promote obliteration of the protein–polymer membrane of proteinosomes (preys). The motion is driven by electrostatic attraction due to opposite surface charges, with subsequent binding and dissembling. This approach highlights the design of droplets displaying predatory behaviors to reflect the features of real ecosystems and provides new directions for the development of colloidal objects as interacting smart systems. Later on, the same group reported a system of three protocells,13 which interacted in sequence to exhibit a response-retaliation behavior, which is a relevant step toward the programming of complex mesoscale dynamics.

The directed motion and chasing behavior of multiple self-propelling microdroplets that specifically migrate from one to another can also emerge in oil emulsions of carefully balanced composition.14 The predatory behavior in this case involves both attractive and repulsive interactions. The predator is attracted to the prey but at the same time the prey is repelled by the predator (see Fig. 1). The motion of the oil droplets in an aqueous solution of surfactants is driven by the micelles-mediated oil exchange between different entities, where a Marangoni flow develops from local changes of the droplets interfacial tension. In the same work, the authors also demonstrated multibody interactions. Actually, these systems are intensely investigated to develop synthetic cellular models with single and collective dynamics. The driving force for oil droplet propulsion might have different origins (surface reactions, self-assembly, solubilization), but normally involves the generation of directional Marangoni flows.15 The proposed system is, therefore, a powerful platform for developing droplet consortia that mimic interactions of living systems. This flexible platform indeed illustrates the design of a droplet population with motility-induced self-organization driven by non-reciprocal chemotactic interactions.

The transport and manipulation of droplets on planar substrates is of practical use for a variety of important applications, including microfluidic liquid handling, thermal management, and material delivery to biochemical analysis.16,17 In that context, droplet manipulation strategies based on external stimuli (electric, magnetic, luminic) or target surfaces having a physicochemical anisotropy have been reported.18–21 The finding of droplet frameworks to autonomously execute cellular life-like features such as predator–prey interactions on surfaces provides fresh scenarios for many top emerging interdisciplinary fields. In line with this innovative route, engineered tunable chasing interactions were attained between binary droplets stabilized by evaporation-induced surface tension gradients on a high-energy surface.22 The droplet propulsion is driven by the vapor emitted by neighboring droplets, which leads to local variations of the surface tension and the associated Marangoni stress. The vapor-mediated interplay leads to interactive droplets moving toward each other in intriguing patterns. A wide range of interesting motility-based interactions was recorded such as one droplet bouncing-off another and one droplet chasing another in a loop. Following this work, the vapor-induced motion of different liquid droplets was carefully studied: it was shown that the droplet motion can be induced even in the absence of the Marangoni effect due to the gradient of the evaporation rate23 and that the droplets propulsion can be effectively controlled by varying the droplet volume and the inter-droplet distance.24 The interaction and motion of droplets on planar surfaces can also be driven by a small increase in the contact angle of the binary droplet due to the surface tension gradients generated where the precursor layers from different droplets meet.25 This driving force between droplets arising at the overlapping zone of the precursor films had been early reported;26 moreover, a couple of droplets following escaper and pursuer behavior was described.

We recently show that nanoporous coatings offer a versatile fashion to mediate reconfigurable droplet communication via messages through a surface to autonomously execute complex interaction behaviors (see Fig. 2).27 Peculiar nanopore imbibition at the thin film level (steady-state wetted annular region in drop periphery)28,29 acts as an emitter of surface chemical messages between droplets with catalytic complementarity. Droplets can indeed become intelligently interactive (stimulus-response role-differenced autonomous operation) when lying on a nanoporous thin film surface. When a H2O2-drop border gets in contact with the wetted ring of a catalyst (KI) droplet, the catalytic peroxide decomposition takes place, which locally decreases the liquid surface tension. The generated surface tension gradient yields a Marangoni stress on the air–liquid interface, which pulls the H2O2-drop contact line outwards and originates a drop projection in a protrusion morphology. The progress of the reaction preserves the surface tension difference and the protrusion keeps growing, until eventually a bridge is formed and the droplets come in full contact (see Fig. 2). The platform also supports the generation of diverse dynamic couplings between the droplets. It is interesting to highlight that the mechanical droplet responses bear resemblance to immune system cell functions such as pseudopod emission and even phagocytic actions. In a minimal framework, the droplets experience selective attacking behaviors after perceiving the presence of the victim drops. Such active droplets represent interacting protocells with primitive responses and abilities that are analogous to living systems.

FIG. 2.

The predatory behavior found in living systems can emerge from a simple nanopore-supported two-droplet framework. The H2O2-droplet (right) senses and subsequently responds with a macroscopic morphological shift and thereby specifically attacking a neighboring KI droplet (left). The droplet interplay stems from the feedback between nanopore-driven underlying communication and chemical activity (KI-catalyzed H2O2 decomposition) as both signaling and power source. Reproduced with permission from Pizarro et al., Nat. Commun. 13, 3047 (2022). Copyright 2022 Springer Nature.

FIG. 2.

The predatory behavior found in living systems can emerge from a simple nanopore-supported two-droplet framework. The H2O2-droplet (right) senses and subsequently responds with a macroscopic morphological shift and thereby specifically attacking a neighboring KI droplet (left). The droplet interplay stems from the feedback between nanopore-driven underlying communication and chemical activity (KI-catalyzed H2O2 decomposition) as both signaling and power source. Reproduced with permission from Pizarro et al., Nat. Commun. 13, 3047 (2022). Copyright 2022 Springer Nature.

Close modal

After all the above discussions, it is evident that, by analogy to living cells, artificial cells must be designed to convert chemical energy into kinetic energy, namely, to achieve autonomous movement. Of course, motility is not the only relevant functionality in artificial cells,2,30 but one of the most challenging to realize, as it involves the integration of environment signal processing and energy transduction. Here, we focus on the autonomous locomotion; hence, we leave aside the externally actuated mechanisms to produce motion. We concentrate our attention on the most used strategy for artificial swimmers: the Marangoni effect.31 Indeed, the vast majority of cases here reported involve Marangoni flows as the mechanism to convert localized chemical changes into fluid transport. The effect, schematically shown in Fig. 3, can be summarized as follows: given an interface separating two fluids (either air–liquid or liquid–liquid), if the surface tension varies from point to point, a tangential force per unit area (Marangoni stress) arises on the interface, which drags fluid toward the region of larger surface tension.32 The generation of a surface tension gradient is due to local variation in the concentration of dissolved species and/or temperature. Therefore, the displacement of mobile interfaces ultimately depends on the heat and mass transfer processes that take place in the system: for example, the transport of oil between droplets of differing composition, in the case of Ref. 14 [Fig. 3(a)], or the consumption of reagents and increase of reaction products, in the case of Ref. 27 [Fig. 3(b)]. Another aspect of the problem is the region where the hydrodynamic effect occurs: for example, the Marangoni flow may develop at the droplet surface (liquid–liquid interface), such as in the case of Ref. 14 [Fig. 3(a)], or at the droplet periphery (liquid–air interface), such as in Refs. 25 and 27 [Fig. 3(b)]. In any case, the relative displacement of fluid produces the droplet propulsion.

FIG. 3.

Highly schematic representations of the droplet self-propulsion by Marangoni flow: (a) free droplet immersed in a liquid and (b) sessile droplet in the triple phase (air–liquid–solid) system. In both diagrams, is the local surface tension, the red arrows stand for the interfacial Marangoni stress, the blue arrows represent the consequent streamlines, and the green arrows denote the resulting direction of the droplet movement. In the case of droplets immersed in a fluid (a), the Marangoni stress develops along the liquid–liquid interface, when a surface tension gradient is generated across the droplet. The asymmetric surfactant coverage may have different origins, such as surface reactions, self-assembly, or solubilization.14 The associated Marangoni stress generates fluid convection inside and around the droplet, which yields a net droplet displacement.15 In the case of sessile droplets (b) laying on high-energy surfaces (relatively low contact angle), precursor films grow around the contact line, and Marangoni flow develops when a surface tension difference arises between the contact line and outermost part of the precursor film. The variation of the surface tension may be due to a selective evaporation on binary droplets25 or the progress of a chemical reaction with reagents that come from neighbor droplets.27 Here, the Marangoni stress takes place on the air–liquid interface, which pulls the droplet contact line and drives the self-propulsion. It should be noted that the drawings are deliberately out of scale for the purposes of illustration; in the real systems, while the size of droplets is on the length scale of micrometers, both the streamlines in the limiting layer around the droplet (a) and the fluid velocity profile in the precursor film (b) expand gaps in the length scale of nanometers.

FIG. 3.

Highly schematic representations of the droplet self-propulsion by Marangoni flow: (a) free droplet immersed in a liquid and (b) sessile droplet in the triple phase (air–liquid–solid) system. In both diagrams, is the local surface tension, the red arrows stand for the interfacial Marangoni stress, the blue arrows represent the consequent streamlines, and the green arrows denote the resulting direction of the droplet movement. In the case of droplets immersed in a fluid (a), the Marangoni stress develops along the liquid–liquid interface, when a surface tension gradient is generated across the droplet. The asymmetric surfactant coverage may have different origins, such as surface reactions, self-assembly, or solubilization.14 The associated Marangoni stress generates fluid convection inside and around the droplet, which yields a net droplet displacement.15 In the case of sessile droplets (b) laying on high-energy surfaces (relatively low contact angle), precursor films grow around the contact line, and Marangoni flow develops when a surface tension difference arises between the contact line and outermost part of the precursor film. The variation of the surface tension may be due to a selective evaporation on binary droplets25 or the progress of a chemical reaction with reagents that come from neighbor droplets.27 Here, the Marangoni stress takes place on the air–liquid interface, which pulls the droplet contact line and drives the self-propulsion. It should be noted that the drawings are deliberately out of scale for the purposes of illustration; in the real systems, while the size of droplets is on the length scale of micrometers, both the streamlines in the limiting layer around the droplet (a) and the fluid velocity profile in the precursor film (b) expand gaps in the length scale of nanometers.

Close modal

In summary, although the generation of Marangoni flows is the common mechanism, each one of the discussed systems has a particular chemical process to generate the required energy for locomotion. It is worth to remark that another artificial swimmers (e.g., micro and nano-motors build on Janus particles) also displaying autonomous locomotion have been developed,33 but not necessarily react or move toward a neighbor particle, in contrast to the droplet pair interactions illustrated along this Perspective. Nevertheless, coupling the locomotion of micro and nano-motors to chemical communication systems is quickly advancing.34 

Finally, it is worth to mention that, to our best knowledge, there are no scientific reports relating the Marangoni effect to the propulsion of individual cells in the natural world. In spite of that, it is known that some of the arthropods that are known to “walk on the water” are also able to “surf” the water by exploiting the Marangoni stress.35 These small insects generate surface tension gradients by locally modifying the chemical composition of the water surface. The remarkable physicochemical mechanism has inspired researchers to develop miniaturized aquatic robots by mimicking the mechanism: the small robots make use of self-powered microfluidic pumps to secret surfactants that induce the Marangoni stress.36 Furthermore, it is well-understood in biomicrofluidics that microorganisms are predestined to swim at very low Reynolds numbers, where inertia is totally irrelevant and viscous flows have to be generated by irreversible and asymmetric forces.37 The bioinspired design of artificial micro-swimmers essentially follows the same strategy, integrating functional microstructures to drive mechanical propulsion,38 including soft, shape-changing, responsive materials.39 Instead, the above-mentioned artificial “surfer” exploits the local generation of surface tension gradients, which drive interfacial (Marangoni) flows that propel themselves through the surrounding liquid. Notably, this is precisely the case of self-propelled droplets (both “swimmers” and “crawlers”) that we are discussing in the present perspective.

In the past few decades, droplet microfluidics have demonstrated to be an eminent resource for designing novel devices for a vast number of fields and applications. How much more will management of droplets be relevant for technology in the future and which areas of research will become even more prominent for applications of droplets? Although it is only possible to hypothesize about how the future will unfold, in vogue research trends already give us an indication of where droplet microfluidics could be headed in the future. Some of the most interesting developments in droplet management concern the exploration of interacting droplets. The idea of autonomous operation and role differentiation in interactive droplets represent in a better way the conceptualization of mimicking living organisms, taking into account that the interaction between individuals is an essential attribute of life. The presented routes of chasing behaviors in both swimmer and on-surface droplet schemes serve as experimental frameworks to investigate biomimetic interaction dynamics. These powerful platforms are not only a scenario for cell prototypes, but also for the development of self-organized complex soft machines from minimal molecular complexity. Macroscale interaction phenomena emerge in cell systems when transforming energy in the microscale to produce moving forces. Many cell sensory-motor interactions are indeed driven by active stresses arising from a localized activity.40,41 It is interesting to note that the origin of the droplet interplay dynamics reminiscent of cell interaction behavior can be usually traced back to a rudimentary way of active stress (Marangoni-stress response) triggered by a physicochemical transductional inter-droplet communication. These inanimate systems with motility-induced dynamics are narrowing the gap between biology and soft matter science, and advancing the next generation of synthetic constructs with life-like emergent behaviors. Therefore, there are vast opportunities for droplet microfluidic optimization and innovation, since droplet-based systems serve as toolkits to engineer life behaviors, enabling broader fundamental insights and relevant applications.

This work was supported by the Agencia Nacional de Promoción de la Investigación, el Desarrollo Tecnológico y la Innovación (No. PICT 2020-01822).

The authors have no conflicts to disclose.

Claudio L. A. Berli : Conceptualization (equal); Writing – original draft (equal); Writing – review & editing (equal). Martín G. Bellino: Conceptualization (equal); Funding acquisition (lead); Writing – original draft (equal); Writing – review & editing (equal).

Data sharing is not applicable to this article as no new data were created or analyzed in this study.

1.
H.
Seo
and
H.
Lee
, “
Recent developments in microfluidic synthesis of artificial cell-like polymersomes and liposomes for functional bioreactors
,”
Biomicrofluidics
15
,
021301
(
2021
).
2.
B. C.
Buddingh
and
J. C. M.
van Hest
, “
Artificial cells: Synthetic compartments with life-like functionality and adaptivity
,”
Acc. Chem. Res.
50
,
769
(
2017
).
3.
T.
Vicsek
and
A.
Zafeiris
, “
Collective motion
,”
Phys. Rep.
517
,
71
(
2012
).
4.
N. K.
Björkström
,
B.
Strunz
, and
H. G.
Ljunggren
, “
Natural killer cells in antiviral immunity
,”
Nat. Rev. Immunol.
22
,
112
(
2022
).
5.
G.
Popkin
, “
The physics of life
,”
Nature
529
,
16
(
2016
).
6.
A. B.
Theberge
,
F.
Courtois
,
Y.
Schaerli
,
M.
Fischlechner
,
C.
Abell
,
F.
Hollfelder
, and
W. T.
Huck
, “
Microdroplets in microfluidics an evolving platform for discoveries in chemistry and biology
,”
Angew. Chem., Int. Ed.
49
,
5846
(
2010
).
7.
M. J.
Zhang
,
P.
Zhang
,
L. D.
Qiu
,
T.
Chen
,
W.
Wang
, and
L. Y.
Chu
, “
Controllable microfluidic fabrication of microstructured functional materials
,”
Biomicrofluidics
14
,
061501
(
2020
).
8.
R.
Malinowski
,
I. P.
Parkin
, and
G.
Volpe
, “
Advances towards programmable droplet transport on solid surfaces and its applications
,”
Chem. Soc. Rev.
49
,
7879
(
2020
).
9.
W.
Lyu
,
M.
Yu
,
H.
Qu
,
Z.
Yu
,
W.
Du
, and
F.
Shen
, “
Slip-driven microfluidic devices for nucleic acid analysis
,”
Biomicrofluidics
13
,
041502
(
2019
).
10.
A.
Testa
,
M.
Dindo
,
A. A.
Rebane
,
B.
Nasouri
,
R. W.
Style
,
R.
Golestanian
,
E. R.
Dufresne
, and
P.
Laurino
, “
Sustained enzymatic activity and flow in crowded protein droplets
,”
Nat. Commun.
12
,
6293
(
2021
).
11.
I.
Cazimoglu
,
M. J.
Booth
, and
H. A.
Bayley
, “
Lipid-based droplet processor for parallel chemical signals
,”
ACS Nano
15
,
20214
(
2021
).
12.
Y.
Qiao
,
M.
Li
,
R.
Booth
, and
S.
Mann
, “
Predatory behaviour in synthetic protocell communities
,”
Nat. Chem.
9
,
110
(
2017
).
13.
Y.
Qiao
,
M.
Li
,
D.
Qiu
, and
S.
Mann
, “
Triggerable protocell capture in nanoparticle-caged coacervate microdroplets
,”
Angew. Chem., Int. Ed.
58
,
2
(
2019
).
14.
C. H.
Meredith
,
P. G.
Moerman
,
J.
Groenewold
,
Y.-J.
Chiu
,
W. K.
Kegel
,
A.
Van Blaaderen
, and
L. D.
Zarzar
, “
Predator–prey interactions between droplets driven by non-reciprocal oil exchange
,”
Nat. Chem.
12
,
1136
(
2020
).
15.
P.
Dwivedi
,
D.
Pillai
, and
R.
Mangal
, “
Self-propelled swimming droplets
,”
Curr. Opin. Colloid Interface Sci.
61
,
101614
(
2022
).
16.
Q.
Zhang
,
S.
Feng
,
L.
Lin
,
S.
Mao
, and
J. M.
Lin
, “
Emerging open microfluidics for cell manipulation
,”
Chem. Soc. Rev.
50
,
5333
(
2021
).
17.
P.
Dak
,
A.
Ebrahimi
,
V.
Swaminathan
,
C.
Duarte-Guevara
,
R.
Bashir
, and
M. A.
Alam
, “
Droplet-based biosensing for lab-on-a-chip, open microfluidics platforms
,”
Biosensors
6
,
14
(
2016
).
18.
J.
Hartmann
,
M. T.
Schür
, and
S.
Hardt
, “
Manipulation and control of droplets on surfaces in a homogeneous electric field
,”
Nat. Commun.
13
,
289
(
2022
).
19.
J.
Zhang
,
X.
Wang
,
Z.
Wang
,
S.
Pan
,
B.
Yi
,
L.
Ai
,
J.
Gao
,
F.
Mugele
, and
X.
Yao
, “
Wetting ridge assisted programmed magnetic actuation of droplets on ferrofluid-infused surface
,”
Nat. Commun.
12
,
7136
(
2021
).
20.
K.
Ichimura
,
S. K.
Oh
, and
M.
Nakagawa
, “
Light-driven motion of liquids on a photoresponsive surface
,”
Science
288
,
1624
, (
2000
).
21.
A. J.
Mazaltarim
,
J. J.
Bowen
,
J. M.
Taylor
, and
S. A.
Morin
, “
Dynamic manipulation of droplets using mechanically tunable microtextured chemical gradients
,”
Nat. Commun.
12
,
3114
(
2021
).
22.
N. J.
Cira
,
A.
Benusiglio
, and
M.
Prakash
, “
Vapour-mediated sensing and motility in two-component droplets
,”
Nature
519
,
446
(
2015
).
23.
X.
Man
and
M.
Doi
, “
Vapor-induced motion of liquid droplets on an inert substrate
,”
Phys. Rev. Lett.
119
,
044502
(
2017
).
24.
Y.
Wen
,
P. Y.
Kim
,
S.
Shi
,
D.
Wang
,
X.
Man
,
M.
Doi
, and
T. P.
Russell
, “
Vapor-induced motion of two pure liquid droplets
,”
Soft Matter
15
,
2135
(
2019
).
25.
B.
Majhy
and
A. K.
Sen
, “
Evaporation-induced transport of a pure aqueous droplet by an aqueous mixture droplet
,”
Phys. Fluids
32
,
032003
(
2020
).
26.
P.
Bahadur
,
P. S.
Yadav
,
K.
Chaurasia
,
A.
Leh
, and
R.
Tadmor
, “
Chasing drops: Following escaper and pursuer drop couple system
,”
J. Colloid Interface Sci.
332
,
455
(
2009
).
27.
A. D.
Pizarro
,
C. L. A.
Berli
,
G. J. A. A.
Soler-Illia
, and
M. G.
Bellino
, “
Droplets in underlying chemical communication recreate cell interaction behaviors
,”
Nat. Commun.
13
,
3047
(
2022
).
28.
M.
Mercuri
,
K.
Pierpauli
,
M. G.
Bellino
, and
C. L. A.
Berli
, “
Complex filling dynamics in mesoporous thin films
,”
Langmuir
33
,
152
(
2017
).
29.
R.
Gimenez
,
G. J. A. A.
Soler-Illia
,
C. L. A.
Berli
, and
M. G.
Bellino
, “
Nanopore-enhanced drop evaporation: When cooler or more saline water droplets evaporate faster
,”
ACS Nano
14
,
2702
(
2020
).
30.
N. A.
Yewdall
,
A. F.
Mason
, and
J. C. M.
van Hest
, “
The hallmarks of living systems: Towards creating artificial cells
,”
Interface Focus
8
,
20180023
(
2018
).
31.
L. E.
Schriven
and
C. V.
Sternling
, “
The Marangoni effects
,”
Nature
187
,
186
(
1960
).
32.
R. F.
Probstein
,
Physicochemical Hydrodynamics: An Introduction
(
John Wiley & Sons
,
2005
).
33.
H.
Su
,
C. A.
Hurd Price
,
L.
Jing
,
Q.
Tian
,
J.
Liu
, and
K.
Qian
, “
Janus particles: Design, preparation, and biomedical applications
,”
Mater. Today Bio
4
,
100033
(
2019
).
34.
L.
Wang
,
S.
Song
,
J.
van Hest
,
L. K. E. A.
Abdelmohsen
,
X.
Huang
, and
S.
Sánchez
, “
Biomimicry of cellular motility and communication based on synthetic soft-architectures
,”
Small
16
,
1907680
(
2020
).
35.
J. W. M.
Bush
and
W. L.
Wu
, “
Walking on water: Biolocomotion at the interface
,”
Annu. Rev. Fluid Mech.
38
,
339
369
(
2006
).
36.
B.
Kwak
,
S.
Choi
,
J.
Maeng
, and
J.
Bae
, “
Marangoni effect inspired robotic self-propulsion over a water surface using a flow-imbibition-powered microfluidic pump
,”
Sci. Rep.
11
,
17469
(
2021
).
37.
E. M.
Purcell
, “
Life at low Reynolds numbers
,”
Am. J. Phys.
45
,
3
11
(
1977
).
38.
S.
Palagi
and
P.
Fischer
, “
Bioinspired microrobots
,”
Nat. Rev. Mater.
3
,
113
124
(
2018
).
39.
G.
Yan
,
A. A.
Solovev
,
G.
Huang
,
J.
Cui
, and
Y.
Mei
, “
Soft microswimmers: Material capabilities and biomedical applications
,”
Curr. Opin. Colloid Interface Sci.
61
,
101609
(
2022
).
40.
M. S. E.
Peterson
,
A.
Baskaran
, and
M. F.
Hagan
, “
Vesicle shape transformations driven by confined active filaments
,”
Nat. Commun.
12
,
7247
(
2021
).
41.
D. S.
Seara
,
V.
Yadav
,
I.
Linsmeier
,
A. P.
Tabatabai
,
P. W.
Oakes
,
S. M.
Tabei
,
S.
Banerjee
, and
M. P.
Murrell
, “
Entropy production rate is maximized in non-contractile actomyosin
,”
Nat. Commun.
9
,
4948
(
2018
).