Liquid Crystal Polymers as Intrinsically Magnetic and Programmable Soft Materials for Microrobotics Authors: Alexandra Gruzdenko, Ignacio Becerra Osorio, Max de Sain, Albert P H J Schenning, Jaap M J den Toonder, Michael G Debije Published: 2026 ABSTRACT ABSTRACTMagnetic microrobots are emerging as powerful tools for biomedical applications. However, widespread reliance on composite materials to achieve programmed magnetic responsiveness, biomimetic elasticity, and biocompatibility of the microdevices results in complex fabrication protocols involving delicate material and processing optimization. An intrinsically magnetic, programmable, soft, biocompatible, and easily‐processable alternative material for development of magnetic microrobots may be liquid crystal polymers (LCPs). To establish LCPs as magnetic systems, this work studies for the first time in detail the manipulation of pristine LCP microobjects with magnetic torques acting on their intrinsic anisotropic diamagnetism. Rotation of planar LCP microdisks in a rotating magnetic field is characterized to identify the step‐out frequency, the key characteristic of magnetic microrobots defining the regime in which their motion is synchronized with the field. Frequencies up to 0.6 Hz, comparable to rotational frequencies of biological microswimmers, are achieved using readily available fields below 0.3 T. Furthermore, the magnetic properties of LCP microobjects are programmed via photoalignment to adjust microobject orientation relative to the applied field and step‐out frequencies. These findings establish a promising foundation for a new class of advanced microrobots that in the future may combine the magnetic responsiveness of LCPs with their shape‐morphing capabilities.Keywords: diamagnetic manipulation, liquid crystal polymers, magnetic polymers, microrobotics, stimuli‐responsive materials Introduction With advances in microfabrication and materials science, microrobotics continues to rapidly evolve, bringing closer transformative solutions to drug delivery, microsurgery, cell manipulation, and other challenges in the biomedical field [1, 2, 3]. Among microrobot actuation strategies, the magnetic field has received particular attention as a remote and precise control stimulus that can penetrate deeply and non‐invasively into most materials, including biological tissues [4, 5, 6]. Over the past two decades, many ingenious microrobot designs have been proposed: bacteria‐inspired helical swimmers were made from self‐folding magnetic multilayers [7, 8], grown via metal‐vapor deposition [9, 10], and printed with polymers subsequently coated [11, 12] or doped [13] with magnetic materials; sperm‐like robots were constructed using DNA‐linked magnetic particles [14] or electrospun from magneto‐active polymers [15, 16]; magnetic Janus particles achieved surface rolling [17, 18]; even microorganisms were coupled with magnetic “wearables” to create biohybrid systems [19, 20]. The most advanced of these robots manage to combine application‐tailored magnetic responsiveness, elasticity (for agile and safe operation in complex environments), and biocompatibility. To unite these qualities in a single device, existing magnetic microrobots rely on composite materials that, along with the advantages of the constituents, bring inherent challenges to the design and fabrication process. Indeed, multi‐component systems require careful material selection and composition tuning to achieve the synergy of the constituents’ properties. Each component typically adds an extra fabrication step, resulting in complex and labour‐intensive fabrication protocols, especially if sophisticated magnetic profiles are required. As the result, the developed methodologies are often tailored to specific robotic types, hence, offering limited design flexibility. Furthermore, magnetic dopants often pose safety concerns due to corrosion effects and leaching of cytotoxic species [21, 22, 23] and, hence, necessitate protective coatings to prevent their direct contact with biological systems. Thus, a readily processible, single‐component material that is biocompatible, soft and exhibits pronounced and programmable intrinsic magnetic properties is highly desirable but has remained an elusive cornerstone of microrobotics [24]. Among non‐composite materials, liquid crystal (LC) polymers (LCPs) stand out as potential candidates to fulfil these requirements. Indeed, LCPs [25, 26] are biocompatible [27, 28, 29, 30, 31], soft, anisotropic diamagnetic polymers consisting of aligned aromatic rod‐like monomers (Figure 1A) possessing an unusually large magnetic susceptibility anisotropy [32, 33] that enables their efficient manipulation with magnetic torques, the primarily actuation mode of magnetic microrobots. Since the magnetic anisotropy is coupled with the molecular order, the LCP magnetic properties can be readily programmed with microscale‐resolution via well‐established photo‐ [34] and field‐induced [35, 36] alignment techniques, something that is not possible for common dia‐ and paramagnetic systems (e.g., graphite, silica, nylon, polystyrene, and hydrogels). Furthermore, LCPs are processable via major additive micromanufacturing methods [37], including direct laser writing (DLW) [36, 38] and digital light processing (DLP), [39], which provide the flexibility of structural design and set LCPs apart from alternative single‐component materials, including certain dia‐ and paramagnets (e.g., graphite, cellulose, and nylon) and especially inorganic ferromagnetic materials (e.g., iron, nickel, and cobalt). Finally, LCPs offer the prospective combination of the magnetic properties with exceptional shape‐morphing [25, 40] that may position these materials as a versatile platform for single‐material, multifunctional, and multiresponsive soft microrobots capable of locomotion and task‐performing via orthogonal external stimuli including magnetic fields, light [41, 42, 43, 44, 45], and chemical cues [46, 47]. (A). Schematic of the liquid crystal polymer architecture and LC monomers used in this work. (B). Proof‐of‐concept experiment demonstrating magnetic torque‐driven manipulation of LCP microobjects; a planarly aligned LCP microdisk is placed in the liquid bulk and subjected to a rotating magnetic field; the disk starts rotating, compelled to align the director with the field vector. (C). Schematic of the fabrication process of planar LCP microdisks (only one of the disks is shown). (D). Electron and polarized optical microscopy images of one of the prepared planarly aligned LCP microdisks. E. Magnetic setup: two permanent magnets flank a pool hosting the LCP disk submerged in the sodium polytungstate (SPT) aqueous solution and create a uniform, uniaxial field across the sample; a home‐built setup (Figure S3) is used to revolve the magnets and create the rotating magnetic field. However, LCPs remain overlooked as intrinsically magnetic materials [48] and are generally considered to require dopants to gain magnetic properties [49, 50, 51, 52, 53]. To unlock the potential of pristine LCPs as alternatives to magnetic composites, we present a proof‐of‐concept study demonstrating that the anisotropic diamagnetism of LCPs can be effectively used to manipulate these materials at the microscale. Focusing on magnetic torque‐driven actuation, we fabricate planarly aligned LCP microdisks using DLP and study their rotation in a rotating magnetic field. We characterize different regimes of the particle motion both theoretically and experimentally and use the acquired insights to measure the step‐out frequency of the LCP disks, the key characteristic of magnetic microrobots defining the frequency range where their motion is fully synchronized with the external field. The field strength dependence of the step‐out frequency is measured in the range of readily available magnetic fields below 0.3 T, and a maximum value of ∼0.6 Hz is achieved at ∼250 mT, comparable to tail‐beating frequencies of sperm cells. Finally, we demonstrate the programmability of LCP magnetic properties by using photoalignment to tailor the orientation of the geometrical axes of LCP microobjects relative to the applied magnetic field as well as to alter their step‐out frequencies. Results and Discussion Theoretical Principles of Magnetic Manipulation of Liquid Crystal Polymer Microobjects Before demonstrating magnetic manipulation of LCP microobjects (also later referred to as ‘microparticles’), we first introduce the related theoretical principles. The anisotropic diamagnetism of liquid crystal polymers is described by the volume magnetic susceptibility tensor where χ ∥ and χ ⊥ are the susceptibility components parallel and orthogonal to the liquid crystal director, n , characterizing the average orientation of the molecules, respectively [32, 33]. Both χ ∥ and χ ⊥ are negative due to the diamagnetic nature of the material, and |χ ∥ | < |χ ⊥ | since mesogens generate a field‐resisting current more efficiently when the field is orthogonal to the benzene rings. In the presence of an external magnetic field, B , an LCP microparticle acquires a magnetic moment, m . Since magnetic susceptibilities of diamagnetic materials are very small (∼ −10−6−10−5), the weak magnetic fields produced by individual LC monomers do not lead to appreciable demagnetising effects [54]. Therefore, the orientation and magnitude of the induced magnetic moment are independent of the sample shape, in contrast to ferro‐, ferri‐, and superparamagnetic materials, and can be expressed using a simplified equation where V is the particle volume, μ 0 is the vacuum permeability, χ m is the volume magnetic susceptibility of the surrounding medium, and B is the external magnetic field. The interaction of m with B enables manipulation of LCPs via both a magnetic force, F m , and a magnetic torque, T m , that can be derived from the magnetic potential energy, Um , of the system by taking its spatial and angular derivatives: where θ is the angle between the magnetic field and director, and with i being a unit vector parallel to m × B . The magnetic force pulls the LCP particle toward low‐field regions while the torque acts to align the director with the field vector to minimize the energy of the system. This work focuses on magnetic torque‐driven manipulation due to its wider applicability in microrobotic propulsion. Proof‐of‐Concept Experiment To study the behaviour of LCP microobjects under the action of external magnetic torques, an experiment is proposed where an LCP microdisk with planar molecular alignment in the particle plane is placed in the bulk of a liquid and subjected to a magnetic field rotating at frequency fmf (Figure 1B). Compelled to align its director with the external field vector, the disk is expected to start rotating together with the field. The motion of the disk is governed by two torques, the magnetic torque given by Equation 1 and the viscous torque where R is the resistance coefficient that considers both the geometry and size of the object, η is the dynamic viscosity of the surrounding liquid, and fd is the frequency of the disk rotation. Given the small size of the disk, the motion occurs in the low Reynolds number regime, meaning the inertia can be neglected and the torque balance describing the disk rotation takes the following form: The LCP disk moves at the same frequency as the field if the magnetic torque, defined by the offset angle, θ, between n and B during the disk rotation, is sufficiently strong to counteract the viscous resistance associated with the particle motion at fd = fmf . The maximum frequency at which such synchronized motion is possible is called the step‐out frequency, fso , and corresponds to θ = 45° that maximizes the magnetic torque. Following Equation 3 considered at θ = 45°, fso takes the form of The step‐out frequency marks the transition from the synchronized motion, called the subcritical regime in this work, to the supercritical regime in which the disk can no longer keep up with the field due to excessive viscous forces. It defines the frequency range in which a potential device can be reliably controlled and is an important characteristic of magnetically driven robots [15, 55]. In the following sections, the proposed experiment is used to first demonstrate the possibility of magnetic manipulation of LCP microparticles and then study their rotation in the sub‐ and supercritical regimes to eventually identify fso . LCP Microdisk Fabrication and Magnetic Setup The planarly aligned LCP microdisks were prepared using a mixture of LC monoacrylate monomer C6BP (80 wt %) and LC diacrylate cross‐linker RM257 (20 wt %) with an addition of 1 wt % photoinitiator and 0.5 wt % inhibitor (Figure 1A). The mixture consisting of only LC reactive molecules was chosen to ensure high concentration of benzene rings, maximizing the anisotropy of the magnetic susceptibility. The material was used to fill a ∼75 µm cell prepared with rubbed polyimide (PI) alignment layers at 80°C (i.e. in the isotropic phase of the mixture; T NI ≈ 53°C (Figure S1)) and allowed to slowly cool to the nematic phase at room temperature to promote the uniaxial planar orientation of LC molecules (Figure 1C, i). Thanks to the supercooling behaviour of C6BP, the sample remained in this state for at least ∼1 h without crystallization. In the next step, arrays of LCP microdisks were DLP‐printed using a digital micromirror device (DMD), projecting a UV mask of the disks into the sample (Figure 1C, ii). The disk diameter of ∼100 µm was chosen for convenient sample manipulation, though smaller LCP objects can also be prepared. Upon completing the photopolymerization, the cell was opened, and the unreacted monomers were washed away with a developer solution, leaving the LCP microdisks on one of the cell substrates (Figure 1C, iii)). The final disks had well‐defined shapes with an average height of ∼76 µm and an average diameter of ∼107 µm, slightly enlarged due to radical diffusion during photocuring (Figure 1D, top). The planar alignment was well preserved after photopolymerization as evidenced by the strong birefringence of the sample when observed through crossed polarizers: the disk appeared bright when n was at 45° to the polarizers and dark in the orthogonal configuration (Figure 1C, bottom). It must be noted that slight distortion in the molecular orientation was present in the system due to monomer diffusion toward polymerized regions occurring during microprinting (Note S1 and Figure S2) [56, 57, 58]. The implications of these distortions are discussed later in the text. To study the behaviour of LCP disks in a rotating magnetic field, the disks were suspended within the bulk of a ∼23 wt % sodium polytungstate (SPT) aqueous solution (matching the density of the LCP) filled inside a small pool (Figure 1E). To prevent the disks from sticking to the pool walls and ease breaking of the surface tension necessary to submerge the particles, 0.1 mM of surfactant Triton X‐100 was added to the SPT solution. The pool was positioned between two permanent magnets, creating a uniaxial and uniform magnetic field across the sample (Figure S3); the field strength could be adjusted by changing the size of the magnets and/or distance between them (Table S1). To achieve rotation of the field vector, a home‐built setup was used to revolve the magnets around the pool (Note S2; Figure S4). The setup was placed on a microscope stage to observe rotation of the disks (Figure S5). Note that at all stages of the experiment, extreme care was taken to minimize the presence of ferromagnetic materials in all tools to ensure only the isolated magnetic response of LCPs was observed (see the Experimental Section). To enable comparison between the experimental results and theoretical predictions, the unknown parameters appearing in Equation 4 for fso were characterized. The magnetic anisotropy of the LCP, Δχ = χ ∥ − χ ⊥, was measured to be ∼1.3·10−6 using the SQUID‐VSM technique (Figure S6), which is comparable to previously reported values for LC materials [32]. The SPT solution was characterized with a rheometer and confirmed to behave as a Newtonian liquid with the viscosity, η, of 1.15 mPa·s (Figure S7). The resistance coefficient, R, was determined by approximating the disks with ellipsoids inscribed within the particles and using the equation derived in previous work [59]. These and other key parameters of the experiment are summarized in Table 1. TABLE 1Key characteristics of the LCP disks and aqueous SPT solutionLCP diskAqueous SPT solution Diameter [µm] Height [µm] Resistance coefficient, R [·10−12 m3] χ⊥ [·10−6] χ∥ [·10−6] Δχ [·10−6] Density [g cm−3] Viscosity, η [mPa·s] 107763.2−9.5−8.21.31.231.15 Key characteristics of the LCP disks and aqueous SPT solution Validation of Magnetic Manipulation of LCP Microdisks To prove that magnetic manipulation of LCPs is possible, a planarly aligned microdisk was suspended in the bulk of the SPT solution and subjected to a rotating magnetic field (∼165 mT; ∼0.07 Hz). Once the field started rotating, the disk followed and rapidly reached the frequency of the applied field, signifying the subcritical regime (Figure 2A; Movie S1: Part 1). To confirm that the response of the disk was caused by the anisotropic magnetic properties of the LCP, the coalignment of its director with the field vector in a static magnetic field was verified. The intensity of light passing through the disk was observed through crossed polarizers as a function of the angle γ between the field and polarizer (Figure 2B). The angle was incremented in 5° steps, and the disk was allowed sufficient time to finish its motion in response to the field change. The intensity reached local maxima at γ = 45°, 135°, 225°, and 315°, indicating that the director indeed aligned with the field as expected from the anisotropic diamagnetism of LCPs. (A). Planar LCP microdisk in a ∼165 mT rotating magnetic field moving at the same frequency as the field (∼0.07 Hz) after stabilization of the particle rotation. (B). Normalized intensity of light passing through a planar LCP disk observed through crossed polarizers as a function of the angle γ between the static magnetic field and polarizer. Eight disks were analysed to measure the average intensity values presented in the graph. The inset images show one of the disks (∼100 µm in diameter) at some of the angles studied. (C). Schematics and polarized images of LCP planar, LCP radial, and SU8 microdisks and their response to the applied magnetic field of ∼365 mT rotating at ∼0.01 Hz. It must be noted that even brief contacts with ferromagnetic materials can cause contamination of samples and significantly affect their magnetic properties [60]. To confirm that only the LCP properties were responsible for the disk rotation, control tests were carried out using LCP disks with axially symmetric radial alignment and disks made of isotropic diamagnetic photoresist SU8 (see Experimental Section for fabrication detail), both of which should be non‐responsive to a rotating magnetic field. The particles were subjected to the maximum possible field of ∼365 mT rotating at a low frequency of ∼0.01 Hz so that samples would have sufficient time to respond in the case of even minor contamination. The majority of the tested planar particles (14 out of 16, i.e. 88%) rotated in response to the applied field while only 2 out of 14 radial disks (i.e. ∼14%) and 2 out of 16 SU8 disks (i.e. ∼13%) showed the same behaviour, indicating the contamination level in the experiment was low (Figure 2C; Movie S2). To gain more insight into the origin of the ferromagnetic contamination, the radial disks that rotated in the applied field were retrieved from the SPT solution and studied with EDX elemental analysis. The EDX maps did not reveal apparent ferromagnetic (i.e. containing Gd, Fe, Co, or Ni) inclusions, and the spectra collected over the area of the disks confirmed their absence to the precision of the instrument detection limit (∼0.1 wt %) (Figure S8). Based on this analysis, it was concluded that trace amounts of contaminants are likely responsible for the response of these samples. Finally, the non‐responsiveness of some of the planar LCP disks was most likely related to the more significant alignment distortion in those particular samples. Overall, the observed rotation of the planar LCP disks combined with the results of the light intensity measurements and control tests prove that magnetic fields can be used to manipulate undoped LCP microobjects using only the anisotropic magnetic properties of these materials. In the next sections, different regimes of the disk motion and the efficiency of the magnetic torque‐driven manipulation in terms of the step‐out frequency are studied. Rotation of LCP Microdisks in the Sub‐ and Supercritical Regimes To experimentally determine the step‐out frequency, the subcritical (fmf ≤ fso ) and supercritical regimes (fmf > fso ) of the disk rotation were studied first. The angle α swept by the particle director during its rotation was measured as a function of time for different frequencies of the ∼165 mT magnetic field (Figure 3A, B). The resulting curves divided into two groups. The first group (blue plots in Figure 3A) represented an overall linear increase in α at the same angular speed as that of the applied field and corresponded to the subcritical regime. In this regime, immediately after the magnetic field started rotating, the angle θ between the director and the field became non‐zero, and the disk, in the absence of inertia, instantaneously acquired rotational speed fd (t) determined by the current torque balance, Tm (θ(t)) = Tv (fd (t)) (Figure 3C, left, Movie S1: Part1). θ kept increasing until it reached its final value, θ f , at which Tm (θ f ) = Tv (fmf ) and the disk entered the regime of uniform rotation, spinning at fmf (Figure 3B, left). Such synchronized rotation of the field and LCP particle persisted with increasing the frequency of the magnetic field rotation, fmf , until it reached ∼0.29Hz where the maximum possible magnetic torque, Tm (θ = 45°), was required to counteract the viscous drag (Movie S1: Part 2). A. Time dependence of the angular displacement, α, made by the director, n , during rotation of a planar LCP microdisk in a ∼165 mT magnetic field rotating at different frequencies. B. Example polar graphs illustrating the uniform disk rotation in the subcritical regime (fmf ≈ 0.07 Hz) and the periodically interrupted disk rotation in the supercritical regime (fmf ≈ 0.39 Hz). C. Schematics depicting the dynamics of the director and magnetic field vector in the sub– and supercritical regimes of the LCP disk rotation. Explanation of the symbols used is given in the left panel. D. Step‐out frequency (fso ) measurements: the average angular displacement, Δα, made by the director before reversing the rotation direction in the supercritical regime (see the inset in panel A), measured as a function of the magnetic field frequency (fmf ) for a range of field strengths; the hyperbolic fits of the graphs were used to determine the step‐out frequency as a function of the applied field strength; 2–5 disks were analysed for each field. E. Experimental and theoretical dependence of the step‐out frequency on the magnetic field strength. The error bars are small compared to symbols representing the experimental data and, therefore, are not visible in the plot. Upon further increase in fmf , the maximum magnetic torque could no longer ensure fd = fmf , and the disk transitioned into the supercritical regime (orange plots in Figure 3A). In this case, the director continued to gradually fall behind the field even after θ passed 45° (Figure 3C, right; Movie S1: Part 3). When θ exceeded 90°, the magnetic torque, Tm∼sin2θ, reversed its sign and caused the disk to reverse its spinning direction. The director and field now moved toward each other until they became antiparallel, after which Tm restored its original sign and the disk resumed rotation in the same direction as B ; the whole process repeated. The overall motion was not uniform as for fmf ≤ fso but became interrupted with periods of backward rotation (Figure 3B, right). The greater fmf , the more rapidly θ reached 90° and later k·90° (k = 1, 2, …) and, hence, the more frequently the rotation was interrupted, as can be seen in Figure 3A. While the supercritical regime can be useful for certain applications (e.g. when an oscillating motion is required), the fully controlled, uniform rotation of the subcritical regime is likely to have a broader applicability. The insight into the disk rotation in the supercritical regime is used in the next section to precisely measure the step‐out frequency as a function of the field strength. Step‐Out Frequency To identify fso , the average angular displacement, Δα, the director makes before it reverses the rotation direction (Figure 3A, inset panel) was measured as a function of the magnetic field frequency. As follows from Figure 3A, in the supercritical regime, Δα should gradually decrease from infinity at fmf = fso to 0 at f mf ≫ f so . The dependence is expected to take a form of a function with a vertical asymptote coinciding with the step‐out frequency. Such curves were measured for a range of magnetic field strengths, with 2–5 different disks analysed for each field value; all the data are presented in Figure 3D. f so was determined by fitting each experimental curve with a hyperbolic function y=A+Bx−fso, with one of the fitting curves shown for each field value. Finaly, the average of multiple f so measurements was taken and used to plot f so as a function of the field strength (Figure 3E). The step‐out frequency increased with increasing field strength and reached ∼0.6 Hz at the maximum applied field of ∼250 mT. Even though the curves measured for a particular field strength differed between the disks, the standard deviation of the corresponding fso in most cases did not exceed 5%, indicating good reproducibility. Notably, the experimental results show an overall agreement with the theoretical predictions based on Equation 4 and parameters of the system (Table 1): both curves follow a quadratic trend, and the theoretical and measured step‐out frequencies are in the same order of magnitude. However, the real disks systematically entered the supercritical regime at frequencies ∼30% lower than theoretically predicted. This discrepancy in part originated from the underestimation of the resistance coefficient, R, caused by the ellipsoid approximation of the disks. However, since an unlikely large (∼30%) error in R would be needed to explain the observed difference by the resistance coefficient alone, the alignment distortion induced by monomer diffusion during printing has likely contributed to it as well: the associated monomer flow interfered with the order imposed by alignment layers, leading to a non‐fully planar molecular configuration and, hence, a reduced particle magnetic moment resulting in a smaller fso . Knowing the characteristic step‐out frequencies of LCPs, potential applications of the magnetic torque‐driven manipulation of these materials in the context of microrobotics can be discussed. Two bio‐inspired mechanisms can be employed to convert the achieved rotational motion into propulsion: the corkscrew motion of flagellated bacteria and the whip‐like tail beating of sperm cells. While bacteria revolve helical flagella at frequencies of 30–100 Hz [61, 62], spermatozoa beat elastic tails at 5–20 Hz [63, 64, 65, 66], which, assuming a beating amplitude of 30°, translates into a rotational frequency of 1–3 Hz. Considering the maximum frequency of ∼0.6 Hz achieved in this work, the design of LCP sperm‐like microswimmers may be attempted, though the effect of the tail on the resistance coefficient needs to be considered. The corkscrew motion may also be considered if the magnetic setup is modified to increase the field strength to 3‐3.5 T, which, based on the quadratic trend of the experimental data, should assist in reaching frequencies close to ~100 Hz. Another type of propulsion that can be used to realize translation via rotation is surface‐rolling [67, 68]. To illustrate such motion, an LCP microblock (∼50 µm × 75 µm × 200 µm) with the director parallel to the block long side was prepared and positioned next to the pool wall in the presence of a ∼200 mT field rotating at ∼0.1 Hz. The interplay between the magnetic torque and surface friction acting on the particle resulted in rolling of the LCP microwalker along the wall (Movie S3). Importantly, the results obtained are relevant to biological scenarios since the viscosity of many bodily liquids like lymph (∼1.5‐2.2 mPa·s) [69, 70] and cerebrospinal fluid (∼0.7‐1.0 mPa·s) [71] are close to that of the SPT solution employed in this work (1.15 mPa·s) (see Table S2 for more examples). Programming the Magnetic Response of LCP Microparticles A unique advantage of LCPs is the possibility of high‐resolution spatial programming of the molecular alignment and, hence, magnetic moment of LCP samples (through its dependence on the internal molecular organization). The alignment programming can, therefore, be used to create microrobots with custom magnetic profiles as well as compositionally and geometrically identical devices with distinct magnetic responses, offering a way of achieving advanced multimodal motion of individual microrobots, differentiated control of multiple microrobots via the same magnetic signal (the multiagent mode), and complex swarm behavior [72, 73, 74]. Below, we consider two examples of the magnetic response programming. Defining the LCP Microparticle Orientation Relative to the External Magnetic Field In contrast to ferro‐, ferri‐, and superparamagnetic materials, the shape of an LCP sample does not affect the magnitude or direction of its magnetic moment [54]. As a consequence, the magnetic easy axis coincides with the director which, therefore, defines the particle orientation in the magnetic field: in a state aligned with a static magnetic field ( n ∥ B ), the angle between the chosen geometrical axis, G , characterizing the particle orientation, and n defines and equals to the angle between G and B (Figure 4A). To demonstrate this principle, three LCP microblocks (∼100 µm × 75 µm × 200 µm) were printed in a planar cell in such a way that their geometrical axes G (parallel to their long sides) formed 0°, 45°, and 90° with the molecular alignment direction (Figure 4B). Compelled to align LC moieties with B (∼200 mT), the particles adopted orientations in which G formed the same angle with B as with n (Figure 4C). Such orientation programming may be used to achieve multiagent control where geometrically identical microrobots, driven by the same magnetic stimulus, move in different directions due to a difference in their orientation relative to the applied field [75]. Programming the magnetic response of LCP microparticles via molecular alignment. (A, B, C). Controlling the particle orientation in the magnetic field: (A) the orientation of an LCP microobject geometrical axis, G , relative to the magnetic field, B , is defined by its orientation relative to the director, n ; (B) fabrication of LCP microblocks with n at 0°, 45°, and 90° to G and (C) their alignment in a ∼200 mT static magnetic field. (D, E, F, G). Adjusting the step‐out frequency: (D) schematic diagrams, polarized images, and (E) fabrication process of the 100Pl and 25Pl‐75H/25P‐75O LCP microblocks; comparison between (F) the 100Pl and 25Pl‐75H microblocks and (G) the 25Pl‐75H and 25P‐75O microblocks: (top) step‐out frequencies and (bottom) angular displacements measured as a function of time for different magnetic field frequencies. All experiments were performed in a ∼200 mT magnetic field. Adjusting the Step‐Out Frequency Endowing areas of an LCP particle with different alignments offers a means to alter the magnetic moment and thereby the step‐out frequency of the system. One way to achieve it is to make a specific area of the LCP object unresponsive to the magnetic field by inducing either the isotropic alignment (Tm∼Δχ=0 in this case) or a uniaxial alignment with n orthogonal to B (θ=90∘,Tm∼sin(2·θ)=0). To demonstrate the latter approach, two LCP microblocks (∼70 µm × 75 µm × 275 µm in size) were prepared: one with fully planar alignment (denoted by 100Pl) and the other with one quarter planar and three quarters homeotropic (i.e. vertical relative to the block plane) alignment (denoted by 25Pl‐75H; Figure 4D). The step‐out frequency of the 25Pl‐75H block was expected to be 1/4 that of the 100Pl sample. The photoalignment technique was used to prepare the 25Pl‐75H block with the microscale‐resolution alignment patterning. A cell with SD1 photoalignment layers was programmed by means of linearly polarized UV light generated by the DMD to have two regions with SD1 molecules pointing in orthogonal directions in the cell plane (Figure 4E, i). Guided by the photoalignment layers, LCP precursor molecules adopted the programmed molecular configuration once filled in the cell (Figure 4E, ii). Next, a microblock (∼70 µm × 75 µm × 275 µm in size) was printed at the border between the two regions in such a way that three quarters of the particle had the director orthogonal to the block geometrical axis (the O‐region) and the other quarter had n parallel to it (the P‐region). The resulting sample configuration was denoted as 25P‐75O and, by rotating by 90° along its axis, could be transformed into the desired 25Pl‐75H configuration. The 100Pl microblock was printed in the same cell in one of the two areas with uniform planar alignment. The quality of alignment within the prepared microblocks was examined by observing them through crossed polarizers (Figure 4D, bottom panels). The 100Pl sample as well as the P‐ and O‐ regions of the 25P‐75O block configuration exhibited strong birefringence indicative of good planar alignment. In contrast, the 25Pl‐75H state showed planar alignment in one quarter of the sample and signs of partially distorted homeotropic alignment in the remaining part, as evidenced by light transmission through the region in both polarizer orientations. Such distortion was expected considering the preceding experiments performed on the planar LCP microdisks and can be explained by the printing‐induced diffusion. Similar to the case of the LCP disks, considering the strong birefringence observed for the 100Pl and 25Pl–75O samples suggesting the overall good alignment, both 100 Pl and 25Pl‐75H samples were deemed suitable for testing. With the same approach used to determine fso of planar LCP disks, the step‐out frequencies of the two blocks were measured in the presence of a ∼200 mT rotating magnetic field. The 100Pl and 25Pl‐75H microblocks exhibited fso = 0.23 Hz and fso = 0.08 Hz, respectively (Figure 4F, top; Figure S9; Movie S4). These results confirm that the combinatorial alignment indeed enables effective tuning of fso . The experimentally observed ratio of ∼3 between the two step‐out frequencies, which is slightly lower than the expected value of 4, is likely attributable to the alignment distortion discussed earlier. In particular, the reduced planar molecular order decreased the experimental step‐out frequency of the 100Pl sample while distortion of the homeotropic alignment increased fso for the 25Pl‐75H block. Making microdevices with distinct step‐out frequencies can allow for their selective control [76, 77, 78]. For instance, in the simple case of the 100Pl and 25Pl‐75H blocks, the magnetic field frequency range comprises three zones (Figure 4F, bottom). In the first zone where fmf≤fso25Pl−75H, both particles spin in synchrony with the field, and their uniform motion can be utilized to perform a useful task. In the second zone where fso25Pl−75H