Transforming patterned defects into dynamic poly-regional topographies in liquid crystal oligomers

We create high-aspect-ratio dynamic poly-regional surface topographies in a coating of a main-chain liquid crystal oligomer network (LCON). The topographies form at the topological defects in the director pattern organized in an array which are controlled by photopatterning of the alignment layer. The defect regions are activated by heat and/or light irradiation to form reversible topographic structures. Intrinsically, the LCON is rubbery and sensitive to temperature changes, resulting in shape transformations. We further advanced such system to make it light-responsive by incorporating azobenzene moieties. Actuation reduces the molecular order of the LCON coating that remains firmly adhered to the substrate which gives directional shear stresses around the topological defects. The stresses relax by deforming the surfaces by forming elevations or indents, depending on the type of defects. The formed topographies exhibit various features, including two types of protrusions, ridges and valleys. These poly-regional structures exhibit a large modulation amplitude of close to 60%, which is 6 times larger than the ones formed in liquid crystal networks (LCNs). After cooling or by blue light irradiation, the topographies are erased to the initial flat surface. A finite element method (FEM) model is adopted to simulate structures of surface topographies. These dynamic surface topographies with multilevel textures and large amplitude expand the application range, from haptics, controlled cell growth, to intelligent surfaces with adjustable adhesion and tribology.

We characterized the degree of oligomerization for main-chain liquid crystal oligomers with azobenzene (Azo) concentrations of 1 wt.%, 5 wt.%, 10 wt.%, and 15 wt.%. Figure S1 shows the 1 H NMR spectrum of partial protons of those oligomers.The signals between 5.75 ppm and 6.45 ppm correspond to the six protons in the diacrylate end groups in the main-chain liquid crystal oligomers.The signals around 7.88 ppm match with the four aromatic protons of azobenzene.The signals around 8.15 ppm correspond to the four aromatic protons in the aromatic groups of RM82.In this study, we define the average degree of oligomerization (DO) as the number of mesogenic units per oligomer.Assuming the chain-extension reaction is complete, the expected DO of the oligomer is calculated using Equation (1).S Ac and S Ar are defined as follows: S Ac is the integration value of all protons in the diacrylate end groups divided by three, while S Ar is the integration value of all the aromatic protons in the mesogenic units.
In each oligomer chain, there are three pairs of protons with identical chemical environments in the acrylate end groups, and the backbone contains 4*DO aromatic protons.When normalizing S Ac to 2, S Ar will adjust proportionally, simplifying Equation (1) to Equation ( 2). Figure S1. 1 H NMR spectrum of partial protons in main-chain Oligomers with azobenzene concentrations of 1 wt.%, 5 wt.%, 10 wt.%, and 15 wt.%.
We studied the phase behavior of the main-chain oligomers before crosslinking.As shown in  4 We studied the glass transition temperature (T g ) of the liquid crystal oligomer networks.As shown in Figure S3, their glass transition temperatures occurred around -14 ºC, -15 ºC, -17 ºC, and -19 ºC for azobenzene concentrations of 1 wt.%, 5 wt.%, 10 wt.%, and 15 wt.%, respectively.These oligomer networks are in a rubbery state at room temperature.

Figure S3
. DSC analysis of liquid crystal oligomer networks with azobenzene concentrations of 1 wt.%, 5 wt.%, 10 wt.%, and 15 wt.%, respectively.The measurement was conducted at a ramping rate of 5 ºC min -1 . 5 We selected an LCON coating with a thickness of 4.9 μm and an azobenzene concentration of 15 wt.% to further study the permanent surface topographies.We observed that while continuously heating above 90 ºC, the surface started to undergo plastic deformation. 1pically, at 120 ºC, the deformation reached up to 3.89 μm (Figure S4c), with an amplitude of 80%.Upon cooling, the coating did not fully return to the initial flat state, retaining 28% of the deformation.6 We varied the coating thickness of an LCON coating and we observed that the thicker coating deforms greater.This can be interpreted as increased thickness providing a larger reservoir of materials available for migration.On the other hand, in the context of the surface of the coating, the materials below the surface can be considered as the soft sublayer.When the thickness of the coating increases, it is equivalent to increasing the thickness of the soft sublayer, which contributes to enhancing the deformation amplitude of the surface topographies.The sample measured contains 15 wt.% azobenzene.
To reveal the relationships between the deformation and coating thickness, we kept the UV light on and investigated the deformation of coatings with varying thicknesses at room temperature or 53 ºC, respectively.At room temperature, the increase in deformation with increasing thickness was not significant, while at 53 ºC, the deformations showed a notable increase, indicating a more significant synergistic effect.Typically, a coating with a thickness of 5 µm exhibits an amplitude of close to 60%.To study the contribution of spatial uniformity during actuation, we modify our FEM code by adding a spatial gradient to control the change in the scalar order parameter along the  thickness.To do so, we choose three functions: one linear and two exponential functions in the form of , where is the thickness of the coating, is the length scale related to ⅇ of oligomerization of the main-chain oligomers were calculated as follows: DO (1 wt.% Azo) = (33.91+ 0.47) / 4 = 8.60; DO (5 wt.% Azo) = (32.04+ 2.30) / 4 = 8.59; DO (10 wt.% Azo) = (30.27+ 4.99) / 4 = 8.81; DO (15 wt.% Azo) = (26.54+ 7.73) / 4 = 8.57.

Figure S5 .
Figure S5.Relationship between thermal response amplitude and thickness of the coating.

Figure S6 .
Figure S6.Influence of thickness on the deformation amplitude induced by light illumination at room temperature or 53 ºC, respectively.The sample contains 15 wt.% azobenzene.

Figure S7 .
Figure S7.Director field design in the FEM elastodynamic simulation.

Figure S8 .
Figure S8.Simulation of surface topography for a coating with a 3×3 array of +1 and -1 topological defects.The onset shows the side view of the regions.

(
-)  ×  ()  =    the penetration of light absorption during actuation, and is the value of at the top layer  ()  =   at a given time (FigureS9a).This way, the top layer of the sample undergoes full actuation, ; while the bottom layer is either partially actuated or remains unactuated, i.e., Δ =-1 Δ = 0 (in the case of linear gradient), see FigureS9b.We find out that, when there is a gradient, the amplitude of deformation on the surface of the sample is lower than the case of uniform actuation (FigureS9c).Consequently, we propose this as a possible reason for observing a lower amplitude of deformation by UV light actuation in our experiment.Although heat increases the temperature almost uniformly through the thickness of the coating, absorption by UV irradiation is significantly different on the top and bottom of the coating.

Figure
Figure S9.a) Proposed mathematical equations used to investigate the spatial dependency of the change in .b) Color plots of for each element across the thickness.The triangles show