Related Application The present case claims priority to, and the benefit of, GB 1613997.4 filed on 16 August 2016 (16/08/2016), the contents of which are hereby incorporated by reference in their entirety.

Field of the Invention

The present invention provides particles, such as microparticles, having self-assembled nanocrystals in a chiral nematic phase, and the use of the particles as dyes, for example to dye articles such as paper, card and clothes. Also provided are methods for preparing the particles by fluidic methods.

Background

Over the last two decades cellulose has attracted a growing interest due to its abundance and versatility when processed on the nanoscale in the form of cellulose nanocrystals. By strong acid hydrolysis, cellulose nanocrystals can be extracted from a variety of natural sources, producing stable aqueous suspensions that exhibit cholesteric liquid crystalline behaviour at higher concentrations.

The evaporation of a cellulose nanocrystal suspension on a flat substrate results in the formation of a solid film with a periodic chiral structure that can reflect visible light. For example, Revol et al. (J. Pulp Paper Science 1998, 24, 146) have described the preparation of reflective films prepared from a suspension of cellulose nanocrystals that was deposited on a Teflon surface. The films were shown to have structural colour in the infra-red, ultraviolet and visible ranges.

Park et al. (Chem. Phys. Chem. 2014, 15, 1477) have studied the formation of films prepared from suspensions of cellulose nanocrystals deposited onto large (25 mm) glass slides. The authors also looked at the effects of shear flow on the formation of helical structures within the film, and the film formation is studied whilst the suspension is dried under orbital shaking. Here, it is said that the shear flow ensures a uniformly aligned vertical helix restricted to the central part of the sample. However, the shear flow does not prevent variations in the pitch of the helix.

WO 95/21901 describes a solidified liquid crystal cellulose film prepared from a dispersion containing a chiral nematic phase of the cellulose. However, these is no description of a particle with the chiral nematic phase in a radial alignment. Although WO 95/21901 describes particulates, these are apparently prepared by disruption of the solid film, for example by milling.

The self-assembly of colloidal liquid crystals systems has been studied almost exclusively in planar geometries as solid films. The films may be used in a wide variety of applications including pressure or temperature sensors, amongst others.

However, more recently there has been increasing study into the effects of topological constraints. Of particular interest is the spherical geometry, where the curvature imposed for example by the interface of an emulsified droplet, leads to frustrated liquid-crystalline self- organization. This spherical topology, typically providing a nematic phase within a thin shell of material, has been shown to give rise to peculiar phenomena, with potential application in e.g. actuators or lasers. Jativa et al. (Soft Matter 2015, 11, 5374) report the self-assembly of cellulose nanocrystals in a shrinking droplet. Here, a very large droplet, around 1 to 10 μΙ_ (approx. 1 mm radius), containing a cellulose nanocrystal suspension in water, was permitted to shrink within a toluene-ethanol continuous phase. SEM and TEM images of the shrinking droplet were recorded. The authors note that they did not observe any pitch lines or reflection colours in the assembled droplets, where such are expected in the corresponding cellulose nanocrystal films. Indeed, there is no demonstration of an ordered internal nanostructure. The authors also report the formation of shells rather than particles in situations where the droplet is permitted to shrink at a relatively quick rate. Geng et al. (Scientific Reports 2016, 6, 26840) have described the formation of fully fluid liquid crystal shells. Here a shell of material is created by flowing a mixture of the molecular liquid crystal RO-TN 615 with chiral dopant CB14 between water-glycerol mixtures to form inner and outer phases, in a nested system. The authors show that polymerization of the shell components maintains optically useful properties in the material, whilst also improving robustness.

Here, the authors specifically emphasis that shell structures are preferred over simple droplets, which have a continuous distribution of material within the inner space. The authors argue that shells provide a more robust uniform alignment of the cholesteric phase. The authors conclude that cholesteric shells may be assembled without any degradation of order, and with a fully retained optical quality. Particles are said to contain a considerable number of defects, which manifest after the polymerization of the assembled shell, to give a product having a reduced reflection. The authors point to the use of very thin shells, having a thickness of 10 helical pitches (5 μηι in the system reported), as the most useful, as this allows for complete Bragg reflections of the wavelength and polarisation that match the cholesteric helix. The authors say there is no benefit in making the shell any thicker than this. Furthermore, the work of Geng et al. seeks to optimize the photonic cross communication between crystal shells for the purpose of secure authentication in a sample, and the optimal shell structures are those that give rise to unique optical patterns resulting from that cross communication. Here, the authors point to the use of a randomly generated spatial distribution of those shells to achieve a non-reproducible cross-communication pattern leading to the desired unclonable optical pattern.

Musevic (Phil. Trans. R. Soc. A 2013, 371, 20120266) reviews earlier work on the formation of a microlaser where a chiral nematic liquid crystal is dispersed in an isotropic insoluble medium. Chiral nematic microdroplets are formed, with a diameter of around 40 μηι. The presence of a surfactant anchors the liquid crystals to the phase boundary. The internal structure of the droplet is such that there is helical modulation of the chiral nematic liquid crystal, extending from the centre of the droplet to the surface. The product gives rise to a birefringent onion Bragg microresonator. This droplet is said to emit light in all directions, and is therefore said to be useful source of monochromatic light. The systems described here make use of an assembly generated from molecular liquid crystals.

Wang et al. (Angew. Chemie Int. Ed. 2016, 55, 12460) describe hydrogel microparticles formed by emulsion photopolymerization of cellulose nanocrystals. These microparticles are then used to form silica microspheres in a double-matrix templating method. The authors record a periodic spacing of the cellulose nanocrystals in the microparticles of about 5 μηι. The authors report radially oriented helical axes. Given the recorded pitch, the reflections from the microparticles are expected to be far from the visible range. Wang et al. was published after the priority date of the present case.

Li et al. (Nat. Commun. 2016, 7, 12520) describe the confinement of cholesteric suspensions of cellulose nanocrystals into droplets using microfluidic flow-focusing methods. The authors observe the reorganization of the liquid crystal phase into a monodomain Frank Pryce structure in some circumstances, recognizable between crossed polarizers as a concentric fingerprint pattern superimposed with a Maltese cross. The recorded pitch values were typically about 6 μηι. Li et al. was published after the priority date of the present case.

Cho et al. (Angew. Chemie Int. Ed. 2016, 55, 14014) report the preparation of droplets and microgels from cellulose nanocrystals using microfluidic flow-focusing methods. The recorded pitch values were in the range 5-6 μηι (for 126 μηι droplets) and 6-9 μηι (for 20 μηι droplets). Cho et al. was also published after the priority date of the present case.

Li et al. (Proc. Natl. Acad. Sci. 2017, 114, 2137) describe the assembly of nanoparticle arrays in the defects and disinclinations found within cholesteric cellulose nanocrystal droplets. The methods of preparation and analysis are the same as those described in Li et al. {Nat. Commun. 2016, 7, 12520). Li et al. was published after the priority date of the present case.

Cipparrone et al. (Adv. Mater. 2011 , 23, 5773) describe the formation of polymeric microparticles via photopolymerization of a micro-emulsion of molecular liquid crystal droplets. The microspheres were shown to contain a variety of internal configurations depending on the preparation conditions, including a Frank Pryce structure. The pitch was controlled by the addition of a chiral dopant allowing the use of the polymeric microparticles as, for example, a microlaser, when a dye was also incorporated. Cipparrone et al. later describe that the pitch of such a system can be tuned to reflect visible colours (Liquid Crystals Reviews. 2016, 4, 59).

The present case provides alternative structures that are based on the hierarchical self-assembly of nanocrystals (colloidal liquid crystals) in a confined geometry such as a droplet.

Summary of the Invention

In a general aspect the present invention provides a particle having self-assembled nanocrystals. The nanocrystals form a chiral nematic phase, otherwise known as a cholesteric phase, and this phase has a radial alignment in the particle. The chiral nematic phase is a helical nematic phase.

The particles have structural colour, of which the colour may be visible colour, infrared colour or ultraviolet colour. The structural colour derives from the presence of the chiral nematic phase, which gives rise to strong reflections at specific wavelengths. The periodicity of the cholesteric structure influences the structural colour as the cholesteric pitch may be varied by changes to the method of preparation. In a first aspect of the invention there is provided a particle having a self-assembly of a nanocrystal, where the particle has a chiral nematic phase, and the chiral nematic phase has a radial alignment within the particle.

The chiral nematic phase may have radial alignment within the particle. The radial alignment of a particle may be determined from the polarization micrograph for example when the particle is viewed through cross-polarizers, optionally with a first-order tint plate. The chiral nematic phase extends throughout the particle, and is not confined to a shell of material. The chiral nematic phase is also preserved during shrinkage, for example where water is at least partially removed from the system. The particles may have a very low size distribution, which is derived from their method of preparation, which allows for the uniform preparation of precursor droplets containing the nanocrystals. The inventors have found that it is possible to reliably measure the cholesteric pitch in particles at concentrations that are usually inaccessible from traditional pitch diagrams owing to the problem of kinetic arrest. Consequently, it is relatively straightforward to study the effects of changes to the methods of preparation on cholesteric pitch, thereby allowing for optimization of the pitch characteristics.

The methods of the invention also permit the formation of self-assemblies overall several hours rather than several days or weeks, thereby allowing desirable architectures to be prepared rapidly. This relatively quick synthesis is important in many systems, such as those based on cellulose nanocrystals, where structural errors induced by desulfation of the cellulose nanocrystals becomes prominent over longer assembly routes.

Across a range of nanocrystal concentrations, the cholesteric pitch in the particles of the invention also matches well with the cholesteric pitch found in films prepared from similar nanocrystal staring materials. Thus, the particles may be used in place of films in various applications.

At other nanocrystal concentrations, particularly higher concentrations, the cholesteric pitch in the particles of the invention departs from the cholesteric pitch found in films prepared from similar nanocrystal staring materials. Thus, the particles provide alternative structures to those films, and therefore the particles also provide alternative structural colour.

In a second aspect of the invention there is provided a dye composition comprising a particle of the first aspect of the invention. The dye composition may further comprise agents, such as solvents, for dyeing an article, such as food and beverage products, paper or card, or a clothing item, or for coating the composition onto a surface. Thus, the dye composition may be for dyeing an article, or the dye composition may be an ink or a paint, for providing decorative colour to a surface.

In a third aspect of the invention there is provided a method of preparing a particle having a self-assembly of a nanocrystal, where the particle has a chiral nematic phase, such as a particle of the first aspect of the invention, the method comprising the steps of:

(i) contacting a flow of a first phase and a flow of a second phase in a channel, thereby to generate in the channel a dispersion of discrete regions, preferably droplets, of the second phase in the first phase, wherein the second phase comprises nanocrystals, wherein the first and second phases are immiscible; and (ii) concentrating the discrete regions of the second phase comprising the

nanocrystals, thereby to generate the particle having a self-assembly of a nanocrystal, where the particle has a chiral nematic phase. The method may be a microfluidic method.

The methods of the invention allow for the preparation of particles at high volume and with a very small size distribution. The methods of the invention allow the formation of the nanocrystal self-assembly to be followed using standard optical instrumentation, thereby permitting a localized, quantitative investigation of the complex dynamic interaction of nanocrystals in suspension. The methods provide a practical route to obtaining highly hierarchical structures in a confined geometry from the nanometre to the macroscopic scale, using readily available nanocrystals. In a fourth aspect of the invention there is provided a method of preparing a particle having a self-assembly of a nanocrystal, where the particle has a chiral nematic phase, such as a particle of the first aspect of the invention, the method comprising the steps of:

(i) dispersing a second phase in a first phase, thereby to generate a dispersion of discrete regions, such as droplets, of the second phase in the first phase, wherein the second phase comprises nanocrystals, wherein the first and second phases are immiscible; and

(ii) concentrating the discrete regions of the second phase comprising the

nanocrystals, thereby to generate the particle having a self-assembly of a nanocrystal, where the particle has a chiral nematic phase.

Methods for the preparation of the particles of the invention may be performed using simple bulk dispersion methods.

Particles prepared by the methods of the invention may be subsequently processed, for example to stabilise the cholesteric phase. Such methods are well known in the art for the processing of films having self-assembled nanocrystals with a chiral nematic phase.

In a fifth aspect of the invention there is provided the use of a particle of the first aspect of the invention as a dye. In related aspects there is provided a method of dyeing an article using the particle, such as within a dye composition, and also provided is an article containing the particle of the invention, which article may be referred to as a dyed article.

The particles of the invention are particularly suitable for use as dyes given their structural colour, which includes visible and infrared colours. The particles are not subject to bleaching and do not suffer from angular dependent colour shifts or spatial dispersity of colour wavelength or intensity. These and other aspects and embodiments of the invention are described in further detail below.

Summary of the Figures

Figure 1 shows (a) the phase behavior of cellulose nanocrystal suspensions of increasing concentration, as imaged under cross-polarizers, where a clear transition from pure isotropic to anisotropic phase is observed from left to right; (b) the calculated ratio of anisotropic phase present for each concentration investigated in (a) to compile the phase diagram (crosses). The specific concentrations investigated within microdroplets are indicated by the colored circles; and (c) a polarization micrograph of the generation of microfluidic water-in-oil droplets from a 14.5 wt % suspension of cellulose nanocrystals, as imaged under cross- polarizers (right) and with a first-order tint plate (left), illustrating the initial radial assembly. Figure 2 shows a comparison between (a) theoretical and (b) experimental images obtained from the confinement of cholesteric suspension of CNC within a spherical geometry, when viewed through cross-polarizers (top row) and upon addition of a first-order tint plate (bottom row). Upon loss of water the Maltese cross is retained, until the onset of buckling upon final drying. The CNC was used at an initial concentration of 7.3 wt %.

Figure 3 shows (a) a scheme of an evolution pitch diagram. The cholesteric pitch measured in the droplets (blue circles) is compared against a macroscale capillary measured by laser diffraction (red circles) and microscopy (red triangles). The pitch below 2 μηι was not measured due to the optical resolution limit. Upon increasing cellulose nanocrystal concentration, the measured cholesteric pitch in the droplets is initially consistent with the pitch measured in capillary, however for the case of confined suspensions a transition at c _g = 12% v/v (approx. 19 wt %) is observed. Extrapolation of the σ ¹ and σ ^1/3 trend lines correlates with the pitch measured by SEM for a dry film and microparticle respectively (diamonds). The capillary error bars correspond to the gradient in pitch observed as a function of position within the anisotropic phase; and (b) Schematics illustrating the effect on the helicoidal cholesteric structure upon three-dimensional contraction when confined within a sphere, as occurs after c _g (p∞ ^{1 3} , top), compared to unidirectional contraction in a planar geometry (p oc c ^"1 , bottom). The cellulose nanocrystal was used at an initial concentration of 7.3 wt % in water.

Figure 4 shows (a) an image of a dried cellulose nanocrystal microparticle, as imaged in transmission (left) and under cross polarizers with a first-order tint plate (right); and (b-d) SEM images of a dry, buckled cellulose nanocrystal microparticle, showing: (c) the clear ordering of cellulose nanocrystals on the surface and (d) the helicoidal assembly of cellulose nanocrystals with a defined pitch, p, within the particle. Figure 5 shows the polarization optical micrographs of particles developed from droplets containing 14.5 wt % suspension of cellulose nanocrystals in water (left), where the droplet and the particles are viewed through cross-polarizers (top row) and upon addition of a first- order tint plate (bottom row). The initial interference colour pattern is retained upon evaporation to form dry microparticles (right), with no evidence of radial ordering observed during this process.

Figure 6 shows the polarization optical micrographs of (a) developing droplets containing 10.9 wt % suspension of cellulose nanocrystals in water (top), where the droplet is viewed through cross-polarizers (right) and upon addition of a first-order tint plate (left); and (b) partially concentrated particles developed from the droplets, where the particle is viewed through cross-polarizers (bottom) and upon addition of a first-order tint plate (top). An initial Maltese cross pattern is observed in (a), which then disappears as the microdroplets exhibit an uncontrolled complex arrangement of the interference colour, which does not improve its order upon concentration of the suspension in the concentration step.

Figure 7 shows the polarization optical micrographs of (a) developing droplets containing 7.3 wt % of suspension of cellulose nanocrystals in water, as imaged under cross-polarizers (right) and upon addition of a first-order tint plate (left) showing the initial formation, and subsequent rapid loss, of a Maltese cross pattern; and (b) the resultant microdroplets contain factoids within a predominantly isotropic phase, that upon slow concentration can reorganize into (c) a radially ordered cholesteric phase, as observed under a first order tint plate. Figure 8 shows the polarization optical micrographs of three different partially concentrated particles, all prepared from developing droplets having an initial diameter of 140 μηι containing 7.3 wt % of suspension of cellulose nanocrystals in water, as imaged under cross-polarizers (bottom) and upon addition of a first-order tint plate (top), where there is (a) radial order throughout the entire diameter of the droplet, (b) significant radial ordering but with an isotropic region, and (c) a chiral nematic shell containing discrete factoids.

Figure 9 shows shows the polarization optical micrographs of (a) particles developed from droplets containing 5.8 wt % suspension of cellulose nanocrystals in water (left), where the droplet and the particles are viewed through cross-polarizers (bottom row) and upon addition of a first-order tint plate (top row). Here, a chiral nematic shell is formed without formation of intermediate factoids; and (b) particles having undergone further concentration as observed through cross-polarizers (right) and upon addition of a first-order tint plate (left). The formation of the chiral nematic shell without formation of intermediate factoids leads to an improved yield in radially-assembled particles.

Figure 10 shows the change in pitch (μηι) with the change in cellulose nanocrystal concentration (v/v) as exemplified by the evaporation of water from droplets with an initial diameter of 140 μηι (circles) and 50 μηι (diamonds) with an initial cellulose nanocrystal concentration of 5.8 wt % (ca. 4% v/v). The data point at [CNC] = 100% corresponds to the pitch measured by SEM, as illustrated in Fig. 4. The diameter of the droplet does not affect the trend in measured cholesteric pitch.

Figure 11 shows example SEM of the cross-section of (a) a cast film and (b) dry

microparticles of the same suspension of cellulose nanocrystals, as used for the pitch measurements in Figure 3. Figure 12 shows the change in droplet diameter (μηι) (and consequentially CNC

concentration as shown by relative amount) as a function of time (h) during the concentration process, for the droplet shown in Figure 2. The droplet diameter decreases linearly at a rate of 12 μηι.Ιτ ¹ . Figure 13 is a pair of SEM images of a microparticle prepared from a suspension having an increased ionic strength. The images shows a particle in cross-section (left image) with an expansion of the observed helical pitch (right image).

Figure 14 is a microscope image of the particles of Figure 13 that have been subsequently washed with ethanol and anisotropically collapsed.

Detailed Description of the Invention

The present invention provides particles having self-assembled nanocrystals. The nanocrystals are provided in a chiral nematic phase.

The particles of the invention may be reliably and reproducibly prepared by fluidic methods. The fluidic methods are easily adapted and allow for the preparation of particles having defined cholesteric pitches, thereby allowing for the formation of particles having alternative structural colours.

The inventors have studied the earlier work of Jativa et al. (Soft Matter 2015, 11, 5364), which is said to describe the confined self-assembly of cellulose nanocrystals in a shrinking droplet.

The work in the present case differs in significant aspects from the work of Jativa et al. The particles of the present case are distinguishable from the particles of Jativa et al. The fluidic methods of the present case are also distinguishable from the particles of Jativa et al. Jativa et al. study a shrinking droplet containing a suspension of cellulose nanocrystals in water. The droplets prepared by Jativa et al. are considerably larger than the droplets used in the exemplified methods of the present case. As noted previously, Jativa et al. say that they did not observe any pitch lines or reflection colours in the assembled droplets, where such are expected in the corresponding cellulose nanocrystal films. The article explains that "[t]he absence of black extinction lines in the polarized images of the factoids suggests that the carboxylated cellulose nanocrystals used in this work do not develop a helical arrangement under the dynamic conditions of this study."

The authors argue that the SEM images show rods assembling with a nematic order with local helicoidal structure. This is not convincing as the SEM image is too cropped to see a clear representation of any significant ordering. Furthermore, the inventors have observed that at the surface of a particle nanocrystal rods often follow the local topography, and this appears to be the case here. Without an SEM image of the particle cross-section, the authors' comments on the presence of assemblies with chiral nematic order cannot be verified.

In contrast, the TEM images do show a particle cross section. Here the authors describe regions of ordering either parallel or perpendicular to the image plane. However, this at best shows the presence of multiple domains with a nematic-like structure, and it does not show evidence of the helicoidal structure that is required for structural colour from this material.

The particles of the present case are therefore clearly distinguishable over the structures described by Jativa et al. Bardet et al. describe the formation of particles that are prepared from cellulose nanocrystal films. Here, a film having a chiral nematic phase and visible light structural colour is prepared, and the film is subsequently ball-milled to give a highly heterogeneous collection of particles. The particles are observed to have structural colour, which is attributable to the retention of the chiral nematic order within the particles. However, the particles of the present case are distinguishable from those of Bardet et al. in that the present particles have a chiral nematic order with a radial alignment within the particle. Such an ordering cannot be expected from the dry grinding of a nanostructure film.

Particles

The particles of the invention may be obtained or are obtainable by the methods of preparation described herein. The particles have a chiral nematic phase, and more specifically the particles have a cholesteric order. A particle of the invention has structural colour. Thus, the chiral nematic phase permits

Bragg reflections, and may do in the visible, infrared and ultraviolet regions of the spectrum. The colour may be ultraviolet colour, visible colour or infrared colour, and it is preferably visible colour. Visible colour refers to a colour with a wavelength in the range 400 to 700 nm. Infrared colour refers to a colour with a wavelength in the range 700 to 1 mm, most preferably 700 to 5 μηι. Ultraviolet colour refers to a colour with a wavelength in the range 100 to 400 nm, most preferably 200 to 400, such as 300 to 400 nm.

A reference to structural colour is a reference to the wavelength of light having a maximum reflectivity when normal incident light is directed onto the particle. A particle is distinguishable from a film. The particle has a spherical or substantially spherical structure, including a collapsed spherical structure, or the particle has an elliptical or substantially elliptical structure, including a collapsed elliptical structure. The film has a planar morphology. The structures of the particle and a film are dictated by their methods of manufacture.

In the present case, the particles are formed within a typically spherical droplet, and the particle takes the form of the droplet in which it is prepared, which may then be modified by the subsequent concentration (shrinking) of the droplet. Films are prepared by deposition of a suspension of nanocrystals onto a planar surface, such as described by Revol et al., where a nanocrystal suspension is deposited on a Teflon surface, Park et al., where a nanocrystal suspension is drop-cast onto a glass slide, and Dumanli et al., where the film is generated within a Petri dish.

A particle is typically a nanoparticle or a microparticle. Thus, the largest dimension of the particle may be in the nanometer or the micrometer range.

Generally, the particle is a microparticle.

The particle may have an average diameter of at most 50, 100, 150, 200, 250 or 500 μηι. The particle may have an average diameter of at least 1 , 5, 10 or 20 μηι.

The average diameter may be in a range selected from the upper and lower limits given above. For example, the average diameter may be within the range 10 to 200 μηι.

Diameter may refer to the largest cross-section of the particle.

The diameter may be determined from a transmission micrograph or SEM images. The diameter of larger particles, such as those with an average diameter of 1 μηι or more, may be determined from simple optical micrographs.

The particles of the invention are generally formed in droplets, which are allowed to shrink during a concentration process. Control of the droplet size therefore provides control of the particle size. A particle refers to a droplet containing the nanocrystals that has been permitted to shrink, thereby forming a self-assembly having a chiral nematic phase. Thus, the particle is a product where the nanocrystals are not entirely in suspension. The self- assembly process is associated with the solidification of the nanocrystals.

A particle of the invention may be substantially free of water. However, the particle may also contain some water. In the present case, the inventors have shown that the concentration of the particle, achieved by removal of a dispersed phase, such as water, alters the pitch of the chiral nematic phase. Thus, concentrating the nanocrystals within the particle has the effect of decreasing the pitch, and thereby shifting the structural colour to shorter wavelengths. The water content of the particle may be 98% v/v or less, 95% v/v or less, such as 90% v/v or less, 80% v/v or less, 70% v/v or less, 60% v/v or less, 50% v/v or less, 40% v/v or less, 30% v/v or less, 20% v/v or less, 10% v/v or less, 5% v/v or less, or 1 % v/v or less.

As noted above, some water may be present. The water content may be 10% v/v at most, 20% v/v at most, or 30% v/v at most.

Alternatively, the content of the particle may be expressed as the nanocrystal content, where high nanocrystal contents are preferred. The nanocrystal content may be expressed as a volume percentage of the total volume of the particle.

The nanocrystal content of the particle may be at least 2% v/v, at least 5% v/v, at least 6% v/v, at least 7% v/v, at least 8% v/v, at least 9% v/v, at least 10% v/v, at least 20% v/v, at least 30% v/v, at least 40% v/v, at least 50% v/v, at least 60% v/v, at least 70% v/v, at least

80% v/v, at least 90% v/v, at least 95% v/v or at least 99% v/v.

The nanocrystal content of the particle may be at least 3 wt %, at least 5 wt %, at least

7 wt %, at least 9 wt %, at least 10 wt %, at least 12 wt %, at least 14 wt %, at least 15 wt %, at least 30 wt %, at least 45 wt %, at least 60 wt %, at least 75 wt %, at least 90 wt %, at least 95 wt %, at least 97 wt %, at least 99 wt %, or at least 99 wt %.

The nanocrystal may make up substantially all of the particle (100% v/v or 100% wt %).

The particles of the present invention have a chiral nematic phase throughout the internal structure. The particle does not possess a hollow interior. Thus, the particle is not a capsule having only a shell of nanocrystals in a chiral nematic phase (a "hollow particle").

The distribution of the chiral nematic phase throughout the particle may be seen from the polarized optical microscopy images and the SEM images of the internal contents of the particle. The presence of chiral nematic organization throughout the entire diameter of the particle is seen from the appearance of concentric dark and bright rings within the polarized optical microscopy images. The arrangement of the helical axis is also seen from the characteristic Maltese-cross pattern in the polarized images, which is indicative of the radial alignment of the helical axes.

The inventors have found that the formation of the particles of the invention is favoured where the nanocrystal is provided in a suspension having a low level of anisotropy, for example at 20% or below, and most preferably where the level of anisotropy is very low, such as about 0%. Here, the formation of particles having an assembly of nanocrystals in a chiral nematic phase is regularly observed at the preferred nanocrystal concentrations. The particles of the invention may be formed from mixtures where the level of anisotropy is greater than 0%, as the work in the present case shows. Sometimes the higher anisotropic levels are associated with products having a shell of self-assembled material and an internal space that is occupied by tactoid structures. Thus, it is preferable to use mixtures having lower level of anisotropy.

The particles of the present case may be contrasted with the shells described by Geng et al. Here, fluidic techniques are used to develop a shell of molecular liquid crystals, and there is no crystal material provided in the internal space. The particle has a cholesteric structure, which is also referred to as a chiral nematic phase. Thus, the particles have a self-assembly that is not a non-helical chiral nematic.

The nanocrystals within the particle are in a helicoidal assembly. The helicoidal assembly may have a defined pitch within the particle.

The particle of the invention is formed from the self-assembly of nanocrystals within a confined space, such as a discrete region, which may be a droplet. The subsequent concentration of the particle may involve a change in the pitch of the chiral nematic phase. A gelation may be seen, where there is a transition from a liquid-like behaviour to a solid-like behaviour, which is seen from the change in pitch rate of change during the concentration step.

The cholesteric pitch, p, may be at most 3.0, at most 4.0, at most 5.0, at most 6.0, at most 7.0, at most 8.0, at most 9.0, or at most 10 μηι. The cholesteric pitch, p, may be at most 2.0 μηι.

The cholesteric pitch, p, may be at least 0.1 , at least 0.2, at least 0.4, at least 0.5, at least 0.6, at least 0.7, at least 0.8, at least 0.9, at least 1.0 or at least 2.0 μηι.

The cholesteric pitch, p, may be in a range selected from the upper and lower limits given above. For example, the cholesteric pitch, p, may be within the range 0.1 to 5 μηι, such as 0.4 to 4.0 μηι, such as 0.4 to 2.0 μηι.

A particle having a lower cholesteric pitch, p, may be associated with a particle having structural colour in the visible range. Thus, alternatively, the cholesteric pitch, p, may be within the range 0.2 to 0.7 μηι (200 to 700 nm), such as 0.3 to 0.5 μηι and such as 0.4 to 0.7 μηι.

The cholesteric pitch may be controlled by selection of appropriate conditions in the methods of preparation. As an example, the pitch may be altered by changes in the nanocrystal concentration within the particle in the second (dispersed) phase. Thus, an increase in the nanocrystal concentration is associated with a decrease in the pitch. The increase in concentration may be affected by at least partially removing the dispersed phase, typically water, from the particle, which consequently shrinks the droplet and effectively concentrates the nanocrystals, resulting in a decrease in the pitch.

The cholesteric pitch in a shrunken particle is dictated by the cholesteric pitch in the large particle, which is in turn influenced by the nanocrystal mixture used in the preparation methods. As the inventors show in the present case, the change in cholesteric pitch in a particle during the concentration step follows a relationship of ρχσ ¹ at lower nanocrystal concentrations within the particle, and then ρ∞σ ^ν3 at higher nanocrystal concentrations within the particle. Figure 3 in the present case shows the change in pitch with the change in concentration of the nanocrystal in the particle. The pitch change follows the trend p∞c ¹ at lower concentration before transitioning to p∞c ^1/3 at higher nanocrystal concentrations. The trend line shows a decrease in pitch with an increase in the nanocrystal concentration (as measured by volume percentage).

Therefore, to obtain a particle having a reduced pitch, such as would give rise to structural colour in the visible range, it may be appropriate to prepare a particle having a lower nanocrystal concentration and with a larger pitch, such as within the low infrared region, such as in the range 0.7 to 2.0 μηι, such as at a nanocrystal concentration in the range 1 to 10 % v/v. Concentrating the particle having a lower nanocrystal concentration will provide a particle having the described reduced pitch. In this particular embodiment, the intent is to reduce the trend line for the pitch decrease such that the pitch at a particular level of concentration is at the level needed to provide the desired structural colour.

Additionally, or alternatively, a change in the pitch of an at least partially concentrated particle may be modified by altering the concentration at which the change in pitch transitions from ρχσ ¹ to ρ∞σ ^1/3 . Where this transition is moved to higher nanocrystals concentration within the particle, the pitch will be reduced, and the wavelength of structural colour will also be reduced (for the reason that the more rapid drop in pitch following p∞c ¹ is sustained until greater nanocrystal concentrations, and the slower drop in pitch following ρ∞σ ^1/3 is delayed). The cholesteric pitch may be measured, in a dried particle for example, from SEM images of the particle, where the helicoidal assembly of the nanocrystals is visible, and the spacing between layers can be measured. An example SEM image is provided in Figure 4(d).

Polarized optical microscopy may also be used to determine pitch values, as described herein, and is suitable for use with those particles having a water content, and that have not yet been subject to collapse. Nanocrystals

The particles of the invention are formed from the self-assembly of nanocrystals within the confined geometry of a droplet. This self-assembly is observed during the shrinking of the droplet in the methods of preparation.

In general, any nanocrystalline material that is known to form cholesteric structures may be used in the present case. The nanocrystal is a nanocrystal of a chiral compound, such as cellulose.

A nanocrystal is typically rod-shaped. Thus, the crystal may be elongate with a length dimension considerably greater than the width dimension.

A nanocrystal for use in the present case, such as a cellulose nanocrystal, may be have a length that is at most 200, at most 500, at most 1 ,000, or at most 1 ,500 nm.

The nanocrystal for use in the present case, such as a cellulose nanocrystal, may have a length that is at least 50, at least 70, or at least 100 nm.

A nanocrystal for use in the present case, such as a cellulose nanocrystal, may be have a width that is at most 20, at most 30, at most 50 nm.

The nanocrystal for use in the present case, such as a cellulose nanocrystal, may have a width that is at least 1 , at least 3, at least 5, or at least 10 nm.

The aspect ratio for the nanocrystal may be at least 5, 10, 15 or 20.

The aspect ratio for the nanocrystal may be at most 40, 50, 100, 150 or 200.

The present inventors have found that nanocrystals having a low aspect ratio give rise to particles having a buckled, rather that wholly spherical, structure. Such structures have been found to have a reduced pitch compared to unbuckled particles, and the inventors have established that red-coloured particles may be prepared from cellulose nanocrystals when the aspect ratio is low.

However, the presence of nanocrystals having a low aspect ratio is not an essential feature for obtaining a particle having structural colour, and structural colour may be obtained in other ways, such as adaptations to the method of particle synthesis and work-up, as described herein, for example through the use of an additional washing and dry step.

Different nanocrystals for use in the present case may have different liquid crystal volume fractions for a set nanocrystal concentration in, for example, water. Below certain concentrations of the nanocrystal in, for example, water, there may be no visible liquid crystal volume. That is, the anisotropic phase may be absent at certain low concentrations of the nanocrystal. The nanocrystals are provided as a mixture, such as a suspension, for example in a second phase, which may be water. Droplets of the mixture are generated in a continuous phase, and these droplets are then concentrated which permits the formation of particles having a chiral nematic phase with a radial alignment. As noted, a nanocrystal for use in the present invention may be selected from any nanocrystal that is known to form a chiral nematic phase. Many nanocrystals are known in the art, and have been shown to form chiral nematic phases within, for example, films.

Examples of nanocrystals for use in the present case include cellulose nanocrystals, chitin nanocrystals, amyloid fibres, fd-viruses and organic colloidal rods, including phytosterol particles, amongst others.

A nanocrystal for use in the present case may be a nanocrystal that is known to form a chiral nematic phase, for example within a film, that gives rise to structural colour, and preferably structural colour that is visible or ultraviolet colour, such as ultraviolet colour. These nanocrystals may then be used to develop particles having a visible structural colour. As discussed in further detail below, the inventors have found that particles generally exhibit a reflectance maximum that is at a greater wavelength than the reflective maximum of the corresponding film (that is a film made from the same nanocrystal). Thus, selecting a film having ultraviolet structural colour, may allow formation of a particle having visible structural colour. Selecting a film having visible structural colour, may allow formation of a particle having infrared structural colour.

The nanocrystal may be a nanocrystal of a polysaccharide, such as glucose-containing polysaccharides, such as cellulose or a cellulose derivative, or chitin or a chitin derivative. The polysaccharide may be cellulose or a cellulose derivative.

In the present case nanocrystals of cellulose are used. Methods for the preparation of cellulose nanocrystals (CNCs) are well known in the art. Many types of cellulose

nanocrystals are known, and examples includes those cellulose nanocrystals obtained from different biological sources as well as those nanocrystals prepared in different ways from the same source.

A cellulose nanocrystal may be prepared from bacterial cellulose, animal cellulose, cotton cellulose, paper, such as filter-paper cellulose, or wood pulp cellulose. Cellulose for processing into nanocrystals may also be obtained from cotton, tunicate and valonia, for example. Typically, the cellulose material is hydrolysed, such as with acid, in the

preparation process, or the cellulose material is oxidized, such as with TEMPO-oxidized cellulose material. A preparation of a cellulose nanocrystal may include sonication of the cellulose sample.

Sonication, such as tip-sonication, of a nanocrystal suspension is typically used to improve the stability of that suspension, and the dispersion of the individual nanocrystals. It is also known that such techniques may be used to alter, such as red-shift, the pitch of the chiral nematic phase, although the mechanism for this is not known. In a worked example, the present inventors prepared a particle without an initial sonication of the nanocrystal suspension. Although an initial reduction in pitch was observed, of around 50% compared with the pitch recorded for the particles in the worked examples of the present case, the radial alignment of the chiral nematic phase in the particle was not seen.

Thus, the second phase may be sonicated, such as tip-sonicated, prior to use in the methods of the invention. The sonication may be varied in terms of the input energy and the duration of sonication, as will be understood to those of skill in the art.

The step of sonicating the nanocrystal suspension is not essential, and it is understood by the inventors that changes to the nanocrystal suspension, and changes to the particle drying steps may be made in order to avoid the need for a sonication step. Thus, particles having a radial alignment of the chiral nematic phase may be obtained without sonication. Thus a sonication step may be performed when needed, and may be performed where it is desirable to red-shift the pitch of the chiral nematic phase.

The processing procedures typically involve purification of the biological source material followed by the separation of the resulting cellulose material into its nanocrystalline components. Known methods for producing cellulose nanocrystals are described by Lagerwall et al. (NPG Asia Materials 2014, 6, e80), the contents of which are hereby incorporate by reference. In the worked examples of the present case, the cellulose nanocrystal is obtained by sulfuric acid hydrolysis of filter-paper. Thus, the cellulose may have sulfate groups, such as sulfate esters, on the surface of the cellulose backbone. Modifications to the cellulose surface by covalent or non-covalent functionalization may be used to modify the self-assembly process. For example, grafting polymers to the nanocrystal may improve colloidal stability without loss of cholesteric liquid crystal behavior, as recently demonstrated by Azzam et al.

Other components may be present with the nanocrystal which may be used to alter the physical and chemical properties of the product particle. Additives may also be added to the nanocrystal, such as acid, base, buffer or salt, to influence the self-assembly of the nanocrystals. These components are discussed in further detail below with respect to the second phase holding the nanocrystals during the methods of preparation. Adaptations to nanocrystals, and adaptations to the compounds for use in forming such nanocrystals, are described in the art. Such adaptations are made with a view to

maintaining the ability of the nanocrystal to form a chiral nematic phase. Methods of Preparation

The particles of the invention may be prepared from droplets containing nanocrystals.

The particles may be prepared using fluidic methods. Such methods allow for the preparation of a large number of particles having a very small size distribution. However, the particles may be formed in other ways, and the methods of preparation are not limited to fluidic methods.

In one aspect of the invention there is provided a method for the preparation of a particle, wherein the method comprises the step of:

(i) contacting a flow of a first phase and a flow of a second phase in a channel, thereby to generate in the channel a dispersion of discrete regions, preferably droplets, of the second phase in the first phase, wherein the second phase comprises nanocrystals, wherein the first and second phases are immiscible; and

(ii) concentrating the discrete regions of the second phase comprising the

nanocrystals, thereby to generate the particle.

The particle has a self-assembly of the nanocrystal in a chiral nematic phase. In particular the nematic phase is a helical nematic phase. The nanocrystals may be provided as a suspension in an aqueous second phase. However, other non-aqueous phases may be used, and the nanocrystals may be provided in an organic solvent second phase, for example in an apolar organic solvent second phase.

Step (ii) may be the at least partial contortion of the droplet contents, thereby to form a particle having a chiral nematic phase, where the chiral nematic phase has radial ordering within the particle.

In step (ii) concentrating may refer to a drying step. Step (ii) may include evaporation of the second phase into the atmosphere from the discrete region, such as from the droplet. This is less preferred as a concentrating method, as it is difficult to maintain close control of the concentrating rate.

Step (ii) may include concentrating the discrete region to dryness. Thus the resulting particle is substantially free of the fluid used as the second phase.

Step (ii) may include initially collecting the outflow from the channel, thereby to collect the discrete regions, preferably droplets, prior to concentrating. In the method of the invention a dispersion of the second phase is created within the continuous first phase. In one embodiment, the second phase is an aqueous phase and the other phase is a water immiscible phase.

Preferably the first phase is substantially free of an alcohol solvent, such as an alkyl alcohol, such as substantially free of methanol and/or ethanol.

Preferably the second phase is substantially free of an alcohol solvent, such as alkyl alcohol, such as substantially free of methanol and/or ethanol.

The methods described by Jativa et al. describe the development of aqueous macrodroplets that are dispersed in a continuous phase of toluene and ethanol. The authors are able to show that droplet size can be decrease rapidly at high concentrations of ethanol in toluene. However, this rapid decrease in droplet size is also associated with an initial expansion of the droplet size as ethanol enters into the droplet.

In one embodiment, the nanocrystal is provided in a flow that is an aqueous phase. The aqueous phase is typically the second phase. In one embodiment, the second phase is an aqueous phase. The first phase is a water immiscible phase, for example an organic solvent or an oil phase.

In one embodiment, the flow of the second phase is brought into contact with the flow of the first phase substantially perpendicular to the first phase. In this embodiment, the channel structure may be a T-junction geometry. The path of the channel may follow the path of the flow of the first phase, in which case the second flow will be substantially perpendicular to the resulting combined flow in the channel. Alternatively, the path of the channel may follow the path of the flow of the second phase, in which case the first phase flow will be

substantially perpendicular to the resulting combined flow in the channel.

Methods utilising a T-junction geometry provide discrete regions, typically droplets, of the second phase in the first phase as a result of induced shear forces within the two phase system. As described herein, the shearing interactions developed in the flow system appear to disrupt radial ordering of nanocrystals at the interface of the two phases, as the discrete region travels along the channel.

In one embodiment, an additional flow of the first phase is provided. The first phase flows are brought into contact with each side of the second phase flow in a channel, and the flow of phases is then passed through a region of the channel of reduced cross-section (an orifice) thereby to generate a discrete region, preferably a droplet, of the second phase in the channel. Such methods, which have an inner second phase flow and two outer first phase flows, are referred to as flow-focussing configurations. Methods using flow-focussing techniques provide discrete regions, typically droplets, of the second phase in the first phase as a result of the outer first phase applying pressure and viscous stresses to the inner second phase, thereby generating a narrow flow of that phase. This narrowed flow then separates into discrete regions, typically droplets, at the orifice or soon after the combined flow has passed through the orifice.

In one embodiment, the discrete region is a droplet.

In one embodiment, the discrete region is a slug.

The droplet is substantially spherical.

In one embodiment, the average size of the droplet is at least 0.1 , 0.5, 0.6, 1 , 5, 10, 20, 30, or 40 μηι in diameter.

In one embodiment, the average size of the droplet at most 1 ,000, 600, 400, 200, 150, or 100 μηι in diameter.

In one embodiment, the average size of the droplet is in a range where the minimum and maximum rates are selected from the embodiments above. For example, the average size is in the range 40 to 400 μηι. The methods of the invention may be used to form droplets having a desired volume. In particular, the methods of the invention are suitable for developing droplets having a very small volume.

In one embodiment, the average volume of the droplet is at least 1 , at least 5, at least 10, at least 50, at least 100, or at least 500 fl_.

In one embodiment, the average volume of the droplet is at most 1 , at most 5, at most 10, at most 50, at most 100, at most 500 or at most 1 ,000 pL.

It is noted that the average volume of droplets described by Jativa et al. is in the range 1 to 10 μΙ_. After the discrete region is formed in the channel, the discrete region may be passed along the channel to a collection area. The residence time of the discrete region in the channel is not particularly limited. In one embodiment, the residency time is sufficient to allow the formation of self-assemblies within the droplet. Discrete regions of second phase are generated in the channel as the immiscible first phase shears off the second phase. The frequency of shearing is dependent on the flow rate ratio of the two phases.

In one embodiment, the flow rate is selected so as to provide a set number of discrete regions per unit time (discrete regions, such as droplets, per second).

The discrete regions may be prepared at a rate of at most 10,000, at most, 5,000, at most 1 ,000 or at most 500 Hz. The discrete regions may be prepared at a rate of at least 1 , at least 10, at least 50, at least 100, or at least 200 Hz.

In one embodiment, the discrete regions may be prepared at a rate that is in a range where the minimum and maximum rates are selected from the embodiments above. For example, the rate is in range 100 to 500 Hz.

The method comprises the step of (ii) concentrating the discrete region. This step at least partially removes solvent (which may be water) from the discrete region and may be referred to as desolvation.

There are no particular limitations placed on the method for concentrating the particles. In one embodiment, the droplets obtained may simply be allowed to stand at ambient conditions, and the solvent (such as water) permitted to evaporate. The droplets may also be heated to evaporate the solvent. These methods will generally result in relatively rapid and uncontrolled shrinking of the particle, which may cause disruption to particle. Greater control of the concentration process may be achieved by the methods described below.

In one embodiment, the second phase solvent is permitted to diffuse into the first phase, thereby resulting in the effective concentration of the nanocrystals in the droplet. Here, the first phase has some capacity to absorb and hold the solvent of the second phase.

The method of concentrating may also be referred to as the shrinkage of the droplet.

The change in droplet size, which is a reduction in the droplet size, may be expressed as a change in the droplet diameter over time.

The diameter of the droplet may be reduced at a rate of at least 1 , 2, 5, 10 μηι/h.

The diameter of the droplet may be reduced at a rate of at most 25, 30, 40, 50 or 100 μηι/h.

Typically, the observed decrease in the droplet diameter for large droplets will be less than the observed decrease in the droplet diameter for smaller droplets. Thus, in the worked examples of the present case a droplet having an initial diameter of around 139 μηι shrank at a rate of around 10 to 20 μηι/h, whilst a droplet having an initial diameter of around 50 μηι shrank at a rate of around 25 μηι/h. Alternatively, the diameter of the droplet may be reduced at a rate of at most 1 , 2, 5, 10 μηι/h. The droplet may be reduced at a very slow rate until the nanocrystal has self-assembled.

The inventors have found that a very slow rate of shrinkage, where the solvent loss from the droplet over time is very low, allows for the use of a nanocrystal suspension at a higher concentration. Under such conditions, the homogeneity of the internal particles structure may be improved. A change in the pitch of the produced microparticle was not observed with change in the initial concentration of the nanocrystal, coupled with a change in the drying rate.

The rate of solvent loss, typically water loss, from the droplet over time may be altered by changing the concentration of a surfactant contained in the continuous phase. Thus, where the surfactant concertation is reduced, the water loss from the droplet over time is reduced. It is believed that the surfactant aids transport of water through the oil. If water cannot be removed it locally saturates the continuous phase and further losses from the aqueous droplet are inhibited.

The inventors have used this approach to prepare droplets from an almost entirely anisotropic suspension, which can self-assemble over several hours (<24 h) with minimal loss of volume. In their original work, the inventors noted that the use of higher concentrations of nanocrystal in the nanocrystal suspension (such as the worked example where the CNC concentration was 10.9 wt %) is accompanied by gelation before significant self-assembly could occur. Altering the drying rate, for example by reducing the surfactant concentration in the continuous phase, permits self-assembly for higher concentrations of the nanocrystal.

Accordingly, with suitable control of the drying rate, the methods of the invention are not restricted to methods where the formed droplets are predominantly isotropic in initial composition. The methods of the invention typically simply need droplets where the contained suspension has not yet reached kinetic arrest (gelation). It is appreciated that the higher the concentration of the nanocrystals in the suspension, the higher the viscosity of that suspension, and the time necessary for the self-assembly of the nanocrystals also increases.

The inventors have found that the drying step may be separated into two phases, where the first phase is a slow, preferably very slow, drying step where the nanocrystal is permitted to self-assemble at near constant concertation. Following this self-assembly in the first phase, the drying rate may be change in a second phase, to allow for the evaporation of the second phase, typically water, from the forming particle. It is believed that the rate of evaporation in this second phase may be less important, as the self-assembly step has already occurred.

In one embodiment, the method optionally comprises a washing step, whereby the particles obtained are washed with a solvent. The purpose of the washing step may be to remove surfactant (where used) or any other component used in the particle-forming step. In one embodiment, a reference to a size of a droplet is also a direct reference to a particle that forms in the droplet, prior to the concentration of the nanocrystal. However, during the concentration process the particle size decreases and the size of the particle ultimately differs from, and is smaller than, the size of the droplet that is generated.

The droplet is a droplet formed in a channel of a fluidic device or a droplet that is collected from the channel of such a device.

A droplet formed directly after preparation is substantially spherical, and a particle that is formed at this stage is also substantially spherical. Desolvation may result in the collapse of the structure as the spherical edge becomes distorted.

The flow rate of the first phase and/or the second phase may be varied to allow preparation of droplets, and therefore particles, of a desired size. As the flow rate of the first phase is increased relative to the second phase, the average size of the droplet decreases, and the formed particle size decreases also.

It is noted that the dimensions and the geometry of the fluidic device may also be used to control the size of the droplets that are generated. Such considerations are familiar to those of skill in the art. Typically, the flow rate of the first phase is at least 1.5, 2, 3, 4, 5 or 10 times greater than that of the second phase.

The flow rates are typically selected to provide droplets of the second phase in the first phase. The flow rates of the first and the second phases may be selected so as to provide droplets having a desired average diameter.

The average particle size may be determined from measurements of a sample of droplets collected from the flow channel using simple microscopy techniques. In one embodiment, each droplet is a nanodroplet. That is a droplet have a diameter measureable in nanometres.

In one embodiment, each droplet is a microdroplet. That is a droplet have a diameter measureable in micrometres. The droplet formed from the fluidic preparation has a narrow size distribution. This may be gauged empirically by observation of the packing of collected droplets. A hexagonal close packing arrangement of collected droplets is indicative of a low monodispersity value, for example. In one embodiment, the droplets have a relative standard deviation (RSD) of at most 1.5%, at most 2%, at most 4%, at most 5%, at most 7%, or at most 10%. The worked examples in the present case show that the droplets have a coefficient of variation of less than 2%. It follows that the particles produced from the droplets may also have the same relative standard deviation if the droplets are concentrated in a uniform manner in step (ii).

The relative standard deviation may be a measurement of the largest dimension of the droplets or the resulting particles.

Fluidic techniques, and more specifically microfluidic techniques, are often considered not amenable to the large-scale preparation of products, such as particles. However, recent development in fluidics, particularly microfluidics, now allow for the large-scale preparation of dispersions using fluidic techniques. For example, a massive parallel flow-focussing device may contain 100 or more, such as 500 or more, flow focussing nozzles, each for the preparation of individual streams of a dispersed second phase in a continuous first phase. Such parallel devices are described by, for example, Amstad et al. (Lab chip 2016, 16, 4163) and may be referred to as millipede devices. As an alternative to the fluidic methods described above, particles may be prepared by the bulk dispersion of a second phase containing the nanocrystals in a continuous first phase. For example, shaking or vortexing a mixture of the second phase and the first phase, will generate a dispersion of discrete regions of the second phase in the first phase. The nanocrystals are permitted to self-assemble in the discrete regions thereby to form particles having a chiral nematic phase. The discrete regions may then be concentrated.

Accordingly, there is provided a method of preparing a particle having a self-assembly of a nanocrystal, where the particle has a chiral nematic phase, the method comprising the steps of:

(i) dispersing a second phase in a first phase, thereby to generate a dispersion of discrete regions, such as droplets, of the second phase in the first phase, wherein the second phase comprises nanocrystals, wherein the first and second phases are immiscible; and

(ii) concentrating the discrete regions of the second phase comprising the

nanocrystals, thereby to generate the particle having a self-assembly of a nanocrystal, where the particle has a chiral nematic phase.

The methods of the invention encompass methods for the generation of emulsions, such as microemulsions, using standard methods of emulsification, such as those methods that are based on simple dispersion and homogenisation, and membrane emulsification. In these methods, as with the fluidic methods, the suspension of the nanocrystal in the second phase is selected for formation of a particle having a chiral nematic phase. The continuous first phase may be selected for the most appropriate concentration rate. As an example, mechanical stirring of a bulk mixture, typically moderate or rapid stirring, can develop a dispersion of discrete regions of a second phase, such as a dispersion of droplets, at an equilibrium size range in a continuous first phase. This size range is defined by the stirring rate as well as the solvents used in the continuous and dispersed phases, together with any surfactants that are present in those phases. The use of bulk mechanical stirring methods to generate a suitable dispersion of cellulose nanocrystals is described by

Wang et al. {Angew. Chem. Int. Ed. 2016, 55, 12460).

As noted above, membrane emulsification methods may also be used as a bulk method to generate a dispersion of discrete regions in a continuous phase. As with other bulk methods, membrane emulsification allows for a large-scale preparation of a dispersion, but does so with some decrease in the dispersed phase homogeneity. Here, the second phase is passed through a porous, such as a microporous, membrane into a flow of the first phase, which passages across the surface of the membrane. The size and distribution of the dispersed discrete regions of the second phase in the continuous first phase depends on a number of factors, including the nature and size of the membrane pores and the degree of coalescence at the membrane surface and within the bulk composition.

An increase in the concentration of the nanocrystals in the second phase may be associated with an increase in anisotropy in the mixture. Conversely, a decrease in the concentration of the nanocrystals in the second phase may be associated with a decrease in anisotropy in the mixture. For example, the cellulose nanocrystals used in the worked example show complete anisotropy in water at around 15 wt % and above. Complete loss of anisotropy is seen at around 6 wt % and below.

The concentration of the nanocrystals in the second phase is chosen to provide a mixture where the second phase has low anisotropy (high isotropy). The present inventors have found that the use of nanocrystal mixtures, such as aqueous suspensions, where there is high anisotropy (low isotropy) do not provide particles having a substantial cholesteric phase. Rather, the particles have a disordered anisotropic structure without any particular re-ordering. In the present case, the use of a completely anisotropic suspension of cellulose nanocrystals in water was found to give a particle lacking order, and the particle remained in a disordered anisotropic state. Similarly, where the suspension of cellulose nanocrystals in water has some isotropic character, with a predominant anisotropy, the particle that results still has a disordered anisotropic state.

Through analysis of the polarization optical micrographs of the fluidic preparation methods, the inventors have seen that the initial stages of the droplet formation involve an immediate radial ordering of the nanocrystal. However, this arrangement is rapidly disrupted by chaotic advection caused by the shearing interactions in the flow methods. The shear forces are believed to induce multiple topological defects into the forming self-assembly, thereby preventing the formation of a radial geometry. However, when the level of isotropy is increased further (and therefore the level of anisotropy is decreased further) there is a change in the behaviour of the nanocrystals within the droplet. At higher levels of isotropy, the formation of radial order throughout the droplet is observed, and this radial order is carried through into the product particle.

At very low concentrations, where there is complete isotropy, the nanocrystals do not form particles where there is radial order throughout the entire diameter. Instead, under very low concentrations shell formation is observed, and these shells have a transient cholesteric order. The shell is disrupted during the concentration process.

Changes in anisotropy can be readily achieved by, for example, concentrating or diluting the nanocrystals.

The amount of nanocrystal present, to achieve the preferred level of anisotropy, will depend upon on the nanocrystal in question and the liquid that it is suspended in. For example, Lagerwall et al. has shown that the anisotropy levels for a range of cellulose nanocrystals differs for a particular set nanocrystal concentration in water. Thus, the nanocrystal concentration is selected appropriately to the nanocrystal for use. In the present case, cellulose nanocrystals in water are preferably used in a concentration range of from 4 to 10 wt %. Where the cellulose nanocrystals are used in an organic apolar solvent, the cellulose nanocrystal may be used at a higher concentration whilst still providing a low level of anisotropy. Thus, the cellulose nanocrystal concentration may be in the range 15 to 40 wt %.

The nanocrystal may be present in the mixture at an amount of at most 7, at most 8, at most 10, most 15, at most 20, at most 25, at most 30, at most 35, at most 40 wt %.

The nanocrystal may be present in the mixture at an amount of at least 1 , at least 2, at least 3, at least 4, at least 5, or at least 6 wt %.

The nanocrystal may be present in the mixture in an amount selected from a range with the upper and lower limits selected from the values given above. For example, the nanocrystal may be present in the mixture in an amount selected from 4 to 10 wt %.

Of course, the wt % values that are chosen will depend upon the level of anisotropy that results from the use of a particular nanocrystal. For the cellulose nanocrystal used in the present work, a concentration from 4 to 10 wt % is suitable of forming particles having radial order throughout the particle.

Thus, additionally or alternatively, the amount of nanocrystal used may be expressed in terms of the level of anisotropy within the nanocrystal suspension. The nanocrystal may be present in a mixture where the level of anisotropy is at most 10, 15, 20, 25, 30, 35, 40, 45, 50, 55 or 60 %. Additionally, or alternatively, the nanocrystal may be present in a mixture where the level of anisotropy is at most 65, 70, 75, 80, 85, 90 or 95%. The nanocrystal may be present where the mixture is substantially all anisotropic

(substantially 100% anisotropy).

The nanocrystal may be present in a mixture where the level of anisotropy is at least 0%, at least 0.1 , at least 0.5, at least 1 , at least 2, at least 5 %.

The nanocrystal may be present in the mixture where the level of anisotropy selected from a range with the upper and lower limits selected from the values given above. For example, nanocrystal may be present in the mixture where the level of anisotropy is selected from 5 to 60 %.

The nanocrystal may be present as a suspension where the isotropic phase dominates. The level of isotropy required for a particular nanocrystal mixture may also be gauged experimentally by preparing droplets using the methods described herein, and the appropriate concentration and levels of isotropy may be determined from the droplets and particles that result having a radial order throughout the entire diameter. The level of anisotropy may be gauged by visualisation of a nanocrystal mixture under cross- polarizers. The relative amounts of isotropic and anisotropic phases may be taken as the relative volumes of each phase in the nanocrystal mixture. See, for example Figure 1 (a) in the present case, where visual images of samples with different levels of anisotropy are shown.

The inventors have found that particles having a very high level of anisotropy do not give rise to particles having a chiral nematic phase extending through the particle. Where the level of anisotropy is very high, the fluidic methods yield a particle having a disordered anisotropic structure throughout the particle. As noted above, a particle of the invention has a chiral nematic phase extending through the particle, as can be seen the polarization micrograph of the particle product.

Where the level of anisotropy is very low, the fluidic methods sometimes yield a capsule having a shell of material only. Here, the internal space is not occupied by a self-assembly having a chiral nematic phase. However, this occurs only when the concentration of the nanocrystal is very low, and therefore the effect is associated not with the level of anisotropy as such.

The second phase is substantially homogenous. Prior to use the second phase may be homogenised, for example, by vortexing a suspension of the nanocrystal in the second phase. Alternatively, the sample suspension may be sonicated. In one embodiment, the discrete region is formed at ambient temperature.

In one embodiment, the discrete region is formed at about 5, 10, 15, 20, 25, or greater than

25°C. After the particle is formed it may be further processed in a further step, for example to stabilise the chiral nematic phase. Such steps are known in the art for stabilising films having self-assembled nanocrystals with a chiral nematic phase.

The preferred nanocrystals for use in the present case are cellulose nanocrystals. Particles prepared from the self-assembly of these nanocrystals may be processed by vacuum treatment or treatment with base, such as potassium hydroxide, in order to remove the sulfate functionality from the cellulose. This stabilises the assembly in the cholesteric particle, and will also prevent dissolution of the particle when it is placed in water. Suitable processing methods for films, which may be adapted for the particles of the present case, are described by Bardet et al. and Giese et al.

A particle prepared by the methods of the invention may be washed, for example with an organic solvent, and subsequently dried. It has been found that washing with certain solvents may be used as a method to reduce pitch length in a particle of the invention. A substantially spherical particle may be swelled in a non-dissolving solvent, such as ethanol, and then subsequently dried. The resulting dried particle has a slightly flattened appearance and the recorded pitch is less than that recorded for the spherical particle, and this reduction in pitch may be in the region of 100 to 200 nm. For example, the inventors have observed that a cellulose nanocrystal particle having a minimum pitch of about 0.7 μηι may be washed with ethanol, and then dried, to yield a particle having observable red-coloured regions, which are indicative of a reduction in the pitch length into the visible region.

Where a buckled particle is produced in the methods of the invention, for example where a nanocrystal having a low aspect ratio is used, the pitch may be further reduced by washing the particle in a method of solvent-induced compression, as described above.

More generally, buckling of the particle, induced by any means, such as those methods described herein, may be used as a mechanism to blue-shift the pitch. Apparatus

The methods of the present invention call for a flow of a second phase and a flow of a first phase, which is immiscible with the second phase, to be brought together in a channel, thereby to generate a dispersion of the second phase in the first phase. Methods for the generation of a flow of a first phase and a second phase are well known in the art. In one embodiment, each flow may be generated from a syringe under the control a programmable syringe pump. Each syringe is loaded with an appropriate second phase or water-immiscible phase.

In the method of the invention, droplets may be collected only when the flows are at the required flow rate.

The channel in which the second phase and first phase flows are contacted is not particularly limited.

In one embodiment, the channel is a microfluidic channel.

In one embodiment, the channel has a largest cross-section of at most 1 ,000, at most 500, at most 200, at most 100 or at most 50 μηι.

In one embodiment, the channel has a largest cross-section of at least 0.1 , at least 1 , at least 10 or at least 20 μηι. The channel may be provided in an appropriate substrate. The substrate is one that will not react with the components used in the formation of the particle, such as the nanocrystals, and such as the first and second phases.

The substrate may be a PDMS-based substrate.

The channels may be prepared or treated such that they are not wettable by the second phase. For examples, where the second phase is water, the channels may be hydrophobic, including hydrofluoric.

The preparation of substrates for use in fluidic flow techniques are well known to those with skill in the art. The channel may have a depth of at least 0.1 , at least 1 , at least 10 or at least 20 μΐη.

The channel may have a depth of at most 50, at most 80, at most 100 or at most 200 μΐη.

Second Phase The second phase is immiscible with the first phase. The second phase may be referred to as a dispersed phase, particularly once it has contacted the first phase and is separated into discrete regions, such as droplets.

In one embodiment, the second phase is an aqueous phase. Therefore, the first phase is water immiscible. Typically the nanocrystal is provided in an aqueous phase, such as a suspension in the aqueous phase. This is the second phase.

In another embodiment, the second phase is a non-aqueous phase. The first phase is immiscible with this non-aqueous phase. The self-assembly of nanocrystals, such as cellulose nanocrystals, in non-aqueous liquid is known. For example, organic, such as apolar organic, liquids may be used, and the nanocrystal may be provided as a suspension in these fluids. Elazzouzi-Hafraoui et al. describe the self-assembly and chiral nematic properties of cellulose nanocrystals from suspensions in cyclohexane. Cheung et al.

describe the formation of chiral nematic structures in cast films from suspension of cellulose nanocrystals in organic solvents, such as dimethylformamide (DMF) and toluene. A second phase may be a combination of liquids, such as water and a second liquid. In some embodiments the second phase may be a combination of water and an alcohol, such as butanol, methanol or ethanol, although this is less preferred. The combination of water and alcohol is typically used where the first phase is an organic oil. In one embodiment, the flow rate of the second phase is at most 1 ,000, at most 500, at most 250, at most 200, at most 150 or at most 100 μ ι.

In one embodiment, the flow rate of the second phase is at least 0.05, at least 0.1 , at least 0.5, at least 1 , at least 5, at least 10, or at least 50 μ ι.

In one embodiment, the flow rate of the second phase is in a range where the minimum and maximum rates are selected from the embodiments above. For example, the flow rate of the second phase in the range 50 to 150 μ ι.

The flow rate of the second phase refers to the flow rate of that phase before the phase is contacted with the first phase. In the worked examples, a flow of the second phase at 80 μ-Jh is used.

The second phase may be provided with additional components to modify the formation of the particle during the concentration step and/or to modify the physical or chemical properties of the particle product. A component that is present in the second phase may be become present, such held, within the particle product.

For example, the second phase may additionally hold a label, such as a dye, such as a fluorescent dye, to give a labelled particle. Other components, such as carbon black, such as particles, such as nanoparticles, of carbon black, may be used to enhance or supplement the structural colour of the particle.

As well as particles of carbon black, other colloidal particles, such as nanoparticles, may be added to the second phase for incorporation into the particle. Examples include graphene, gold, silver and quantum dot particles. Such particles may be used where they do not disrupt the chiral nematic phase.

In a further example, the second phase may additionally hold polymers or non-volatile solvents, such as glycerol, whose presence is capable of influencing the behaviour of the particle during the concentration process, for example to reduce the bucking of the particle. These components, may also be used to modify, such as increase, the cholesteric pitch in the particle product. Other components, such as surfactants may be provided in the second phase, for example on the surface of the nanocrystals, to improve mechanical properties.

The additional components may be added into the second phase by simple admixing of the component with the nanocrystals and the second phase fluid.

It is known that components that are added to the second phase may preferentially locate to the core of a particle. Li et al. (Nat. Commun. 2016, 7, 12520) note that particles of, for example, metal or carbon may partition to the core of a cholesteric droplet. Similarly, the droplets and particles described herein may include an additional components, such as microparticles, that are predominantly located in the core of the droplet or particle.

First Phase

The first phase comprises a component that is immiscible with the second phase. The first phase may be referred to as a continuous or carrier phase.

In one embodiment, the flow rate of the first phase is at most 10,000, at most 5,000, at most 1 ,000, at most 500, or at most 250 μ τ

In one embodiment, the flow rate of the first phase is at least 10, at least 50, or at least 100 μ ι.

In one embodiment, the flow rate of the first phase is in a range where the minimum and maximum rates are selected from the embodiments above. For example, the flow rate of the first phase in the range 100 to 250 μΙ_/ϊι.

The flow rate of the first phase refers to the flow rate of that phase before the phase is contacted with the second phase.

Where a flow focusing technique is used to develop discrete regions of a second phase, the flow rates of the two first phases may be the same. The values given above may refer to the combined flow rate of the two first phases, to the flow rate of each first phase.

The first phase may be less dense than the second phase. Thus, when the dispersed second phase is collected within the first phase, the second phase will sink within the first phase, and will subsequently not be exposed to the atmosphere. In this situation the rate of loss of the fluid of the second phase, such was water, may be reduced. The inventors have found that a slower concentration of the droplet contents is helpful in controlling the structure of the particle product.

The first phase may additionally comprise a surfactant. The surfactant is provided in the first phase in order to stabilise the micro-emulsion that is formed in the fluidic preparation methods. The step of forming the discrete region (such as a droplet) may require the presence of a surfactant. Furthermore, the presence of a surfactant is useful in limiting or preventing the coalescence of the droplets collected. Additionally, changes in the concentration of the surfactant may be used to alter the rate of water loss from the droplet and particle during the concentration process, and this may be used to alter the structure of the final product particle. The surfactant chosen is not particularly limited, and encompasses any surfactant that is capable of promoting and/or stabilising the formation of discrete regions, such as droplets, of the second phase in the first phase.

A surfactant for use in the present case is a non-ionic surfactant, such as the sorbitan oleate surfactant which is available as Span 80 (RTM) from Sigma-Aldrich. Such surfactants are especially useful for stabilising water-in-hydrocarbon oil emulsions, such as water-in hexadecane, for example.

Alternatively, the surfactants comprise an oligomeric perfluorinated polyether (PFPE) linked to a polyethyleneglycol. Such surfactants are especially useful for stabilising water-in- fluorocarbon oil emulsions, for example.

In the present case the choice of fluid and surfactant for the first phase is taken with a view to the preferred rate of liquid loss from the droplet and the particle during the concentration step. As mentioned previously the droplet and the particle may not be exposed to atmosphere during the concentration process, and any loss of the second phase fluid, such as water, from the droplet and the particle is into the first phase only. It is under these circumstances that there is closest control of the liquid loss from the droplet and particle. In other embodiments a droplet or particle may be exposed to the atmosphere, and the liquid loss may be to the atmospheres under controlled conditions, for example in a controlled humidity chamber. However, arrangements of this type are more complex, and are less preferred.

The surfactant is present at most 0.1 %, at most 0.2%, at most 0.5%, at most 0.75%, at most 1 %, at most 2%, at most 5% w/w to the total phase.

The surfactant is present at least 0.05% or at least 0.07% w/w to the total phase.

In one embodiment, second phase has a limited solubility in the first phase. As noted above, the concentration step may involve loss of liquid, such as water, from the droplet or the particle into the first phase.

In one embodiment, second phase has a solubility in the first phase of at most 50, at most 20, at most 10, or at most 5 ppmw.

In one embodiment, the first phase has a solubility in the second phase of at most 50, at most 20, at most 10, or at most 5 ppmw. Aqueous Phase

The present invention calls for the use of an aqueous phase as the dispersed phase in the methods of the invention. Methods for the preparation of suitable aqueous mixtures comprising nanocrystals will be apparent to those of skill in the art, and examples preparations are given in the worked examples of the present case.

In the present case the nanocrystals are provided as a suspension in water. The aqueous phase may have an ionic strength or a pH that is selected for a particular pitch value in the particle of the invention. Thus changes to the ionic strength or the pH can be used to alter, such as decrease the pitch.

For example, an increase in ionic strength may be used to decrease the pitch of the chiral nematic phase in the product particle. Without wishing to be bound by theory, the present inventors believe that an increase in ionic strength moves the gelation point to a higher concentration, allowing for a greater twisting of the cholesteric helix during the formation of the particle, for example during the concentrating procedure. The ionic strength may be altered, such as increased, by the addition of a salt and/or an acid and/or a base to the aqueous phase, by the increased concentration of a salt in the aqueous phase, or by the change in acidity of the aqueous phase. The salt is not particularly limited, and it may be an inorganic salt such as NaCI. The acid is not particularly limited, and it may be an inorganic acid such as HCI or H2SO4.

Very large increases in ionic strength are generally to be avoided, as the inventors have noted that found to result in particles having an increased disorder within the core, and also an increase in the pitch of the chiral nematic phase. In an example in the present case, the inventors have found that the pitch in a particle prepared as described herein may be reduced to around 0.7 μηι (as measured by SEM) by increasing the ionic strength of the aqueous phase using H2SO4.

Without wishing to be bound by theory, the inventors believe that the observed changes in the recorded pitch occur at the limits of the electrolyte concentration. At lower ionic strengths, which is the lower limit in the system (near zero), electrostatic repulsions are dominant and very long-range, promoting kinetic arrest. At low concentrations of the nanocrystal this is generally referred to as a repulsive Wgner glass. At higher ionic strengths, the upper limit in the system, the colloidal stability is disrupted and this leads to either a slow evolution of the particle into an attractive colloidal gel, or to flocculation and precipitation. The specific ionic strength needed to produce a desired result (a desired pitch) will depend upon the nanocrystal, and particular the surface charge of that nanocrystal, and the nanocrystal suspension itself, including the relative concertation of the suspension.

Typically, the amount of a monovalent salt (such as NaCI) to the mass of nanocrystal is kept below 200 mmol/kg, such as below 100 mmol/kg.

Water Immiscible Phase

The present invention calls for the use of a phase that is immiscible with water. That phase may be an oil-based phase (oil phase) or an organic solvent-based phase (organic phase), or a combination of the two. In one embodiment, the water immiscible phase is a liquid phase.

In one embodiment, the water immiscible phase has as a principal component an organic solvent. For example, the organic solvent is selected from chloroform, octane and hexadecane, such as hexadecane.

The oil phase has, as a principal component, an oil. The oil is a liquid at ambient temperature.

In one embodiment, the oil is an organic oil, such as a hydrocarbon-based oil. An example of a hydrocarbon-based oil for use in the invention is hexadecane.

In one embodiment, the oil is a mineral oil.

In one embodiment, the oil is a fluorinated hydrocarbon oil.

In one embodiment, the oil is a perfluorinated oil. An example of a perfluorinated for use in the invention is FC-40 (Fluorinert as available from 3M).

In one embodiment, the oil is a silicone oil.

The water immiscible phase is inert. That is, it does not react with the nanocrystals, or any other component used to form a particle of the invention.

The water immiscible phase may be selected for its limited ability to absorb water from the developing particle during the concentration step, for example to a maximum water content as discussed above.

Analysis of Particle

A particle of the invention may be analysed by simple bright field microscopy to determine the shape of the particle surface. The images obtained may also be used to determine the cross-section, typically the diameter, of the particle. The particle may also be analysed for shape and cross-section, amongst others, thickness using scanning electron microscopy and atomic force microscopy, such as described in the worked examples of the present case. Polarized optical microscopy may be used to analyse phases within the particles, such as the presence of isotropic and anisotropic phases of the nanocrystal. The worked examples describe the use of polarized optical microscopy with analyses performed in transmission mode with crossed polarizers. SEM images may be used to ascertain the presence of the chiral nematic phase, and may be used to measure the pitch of the helicoidal structures.

Use of Particles The particles of the invention are suitable for use as dyes, owing to their ability to provide structural colour, including visible, infrared and ultraviolet colour.

The spacing of the layers within the particle may be controlled in order to provide a particle having a desired structural colour.

The particles of the invention may be use as dyes or pigments in place of traditional dyes or pigments, for example, in food and beverage colouring, publishing and clothing manufacture.

Accordingly, the invention also provides a dye composition comprising the particle of the invention.

The composition optionally further comprises agents, such as solvents, for dyeing an article, such as paper or card, or a clothing item. Suitable agents for use in a dye composition are well known to those of skill in the art.

The composition may be an ink or a paint and the composition may further comprise agents suitable for use in coating the surface to which the ink or paint is to be applied.

The composition may a colouring additive for food or beverage, and the additive may comprise further ingestible ingredients for use in the preparation of the food or beverage.

The invention also provides an article, such as a foodstuff, beverage, paper or card, or a clothing item, comprising the particle of the invention.

The particles of the invention provide the advantage over traditional dyes in that they are not subject to bleaching. Furthermore, particles having structural colour, in contrast with films having structural colour, do not suffer from angular dependent colour shifts, such as those visible to the eye, or spatial dispersity of colour wavelength or intensity. Also provided by the present invention is a method of dyeing an article, such as paper or card, or a clothing item, the method comprising the step of contacting the article with a dye composition comprising the particle of the invention. The particles of the invention may also find use in applications that are unrelated to structural colour. For example, the particles, such as microparticles, may find use as beads, such as microbeads, in consumer and healthcare products, including cleaning compositions and cosmetic compositions. The particles of the invention may be prepared from

biopolymers, such as polysaccharides, and these may be used to generate beads that have biocompatible and biodegradable properties. This provides an alternative to current microbeads, which are prepared from synthetic polymers, and which suffer from the problem of bio-accumulation, most particularly in the marine environment.

Other Preferences

Each and every compatible combination of the embodiments described above is explicitly disclosed herein, as if each and every combination was individually and explicitly recited.

Various further aspects and embodiments of the present invention will be apparent to those skilled in the art in view of the present disclosure.

"and/or" where used herein is to be taken as specific disclosure of each of the two specified features or components with or without the other. For example "A and/or B" is to be taken as specific disclosure of each of (i) A, (ii) B and (iii) A and B, just as if each is set out individually herein.

Unless context dictates otherwise, the descriptions and definitions of the features set out above are not limited to any particular aspect or embodiment of the invention and apply equally to all aspects and embodiments which are described.

Certain aspects and embodiments of the invention will now be illustrated by way of example and with reference to the figures described above.

Experimental and Results

Materials

Hexadecane (99%) and Span 80 were purchase from Sigma Aldrich and Fluka respectively and were used without further purification. The initial suspension of 14.5 wt % cellulose nanocrystals was prepared from filter paper, as described below, with subsequent formulations diluted with deionized water (Millipore Milli-Q Gradient A10, resistivity

>18 MQ.crrr ¹ ). Instrumentation

Microdroplets were imaged in transmission using a Vision Research Phantom Miro ex4-M fast camera, attached to an Olympus IX-71 inverted microscope (10x-64x objectives).

Polarized optical microscopy was performed in transmission with crossed polarizers. In order to indicate CNC orientation a sensitive tint-plate (Olympus U-TP530) was additionally inserted between the crossed polarizers. Scanning electron microscopy (SEM) images were acquired using a Zeiss Leo Gemini 1530VP system, working at 90° with respect to the electron beam. SEM samples were mounted on aluminum stubs using conductive carbon tape and, to minimize surface charging, sputtered with a 5-10 nm layer of Au/Pd (Emitech K550; I = 55 mA for 10 s). The acceleration voltage used was 2.0 kV and the working distance was 1-2 mm. Atomic Force Microscopy (AFM) images were acquired with an Agilent 5500, collected in tapping mode (OTESPA-R3 tip) and at room temperature over a 25 μηι ² area. AFM samples were prepared by drop casting 10 μΙ_ of a diluted CNC

(cellulose nanocrystal) suspension on to poly(L-lysine) functionalized mica. After deposition, the samples were rinsed with deionized water and dried under nitrogen flow.

Cellulose Nanocrystal Suspension

Cellulose nanocrystals were obtained from the hydrolysis of 'Whatman No.1' cellulose filter paper (30 g) with sulfuric acid (64 wt %, 420 ml_) at 64 °C for 30 min, before quenching using Milli-Q ice and water. Soluble cellulose residues and acid were removed by

centrifugation (three steps at 20,000g for 20 min) and dialysis against deionized water (MWCO 12-14 kDa membrane) and a stable suspension of [CNC] = 2.2 wt % was obtained. Conductivity titration against sodium hydroxide indicated [S0 ^2_ ] = 205 mmol.kg ^-1 of CNC. The suspension was tip sonicated in an ice bath (Fisherbrand Ultrasonic disintegrator 500 W, amplitude 30% max. 8200 J.g- ¹ of CNC) and vacuum filtered (8.0 μηι then 0.8 μηι nitrocellulose, Sigma-Aldrich).

The suspension was concentrated by heating at 60 °C in a water bath for 12 h, resulting in a 14.5 wt% (approx. 9.4% v/v) suspension of CNC. The ionic content of this concentrated suspension was deduced from conductometric titration against sodium hydroxide, complemented by pH measurements before and after potassium chloride addition and corresponds to the following: 36% surface charge loss by CNC partial desulfation ([S0 ^2_ ] = 131 mmol.kg- ¹ of CNC) [CNC ^" ] = 19 mm, the release of [S0 ² -] _fre e = 10.7 mm, [H ⁺ ] _fre e = 18.7 mm and [H ⁺ ] _to tai = 40.4 mm. Planar films of CNC were prepared by casting 1.0 ml. of a 2.0 wt % CNC suspension in a 3.5 cm diameter polystyrene Petri Dish, before allowing to evaporate under ambient conditions. Phase Diagram and Pitch Measurement in Capillaries

The initial 14.5 wt % CNC suspension was diluted with deionized water using a high precision scale, vortexed and transferred to a flat capillary of sufficiently large inner dimensions to eliminate any confinement effects (1.00 x 10.00 x 50 mm) and sealed with glass plates and nail polish. The self-assembly was observed after 4 days and later after 95 days with no noticeable change. The cholesteric pitch in the glass capillaries was then determined using: (i) Polarized optical microscopy (TPIan Nikon, 20x (NA=0.30, WD=30mm) and 50x (NA=0.40, WD=22mm) objectives, images not shown). The pitch was measured as twice the period of the fingerprint pattern, taking either an average over 10 pitch distances or using a FFT of the image processed with ImageJ. (ii) Laser diffraction performed using a laser (λ = 531.8 nm) and observing the diffraction pattern in transmission (images not shown). The pitch was derived using Bragg's law, as adapted by Kahn to include Snell law correction. In order to account for possible sample inhomogeneity in the vertical dimension, the pitch was measured at regular intervals throughout the anisotropic phase.

Droplet-Based Microfluidics

Monodisperse water-in-oil microdroplets were generated within a hydrophobic flow-focusing microfluidic device. These were manufactured from PDMS via soft lithography, whereby: (i) the microchannel network was designed in silico (AutoCAD), (ii) printed as a negative photomask and (iii) transferred onto a silicon wafer spin-coated with SU-8 photoresist via

UV-photolithography to form a mold. PDMS and the cross linker (Sylgard 184 elastomer kit, Dow Corning) in a 10: 1 ratio were poured onto this mold and allowed to stand overnight at 70°C. The PDMS layer, imprinted with the microfluidic channel design, was removed and using a biopsy punch (1.0 mm) inlets and an outlet were formed. The imprinted PDMS and a glass substrate were exposed to oxygen plasma for 8 s and then pressed together to seal the microfluidic channels. To study the self-assembly of CNC within the microdroplet a 200 μηι wide flow-focusing junction (A) was employed with a channel depth of 80 μηι. For structural analysis of solid microparticles a smaller flow-focusing junction (43 x 43 μηι, B) was employed to generate the templating microdroplets. To render the channels fluorophilic they were immediately flushed with a 0.5% v/v solution of trichloro(1 H, 1 H,2H,2H-perfluorooctyl)silane in Fluorinert FC-40 (3M) and subsequently cured at 120°C overnight. This hydrophobic surface- modification was also applied to glass slides that were used in 'shrinking droplet' studies, ensuring low wetting of the surface by the aqueous microdroplets.

To generate microdroplets, the continuous oil phase and the discrete aqueous phase were injected into the microfluidic device via two syringe pumps (PHD 2000, Harvard Apparatus) with controlled flow rates of 200 μΙ_.Ιτ ¹ and 80 μΙ_.Ιτ ¹ , respectively. At the intersection, the shear forces caused the formation of aqueous droplets in oil (typically for a 7.3 wt% CNC suspension: 0A = 139.5 ± 2.5 μηι and B = 50.1 ± 0.6 μηι). The continuous phase comprised of the organic oil, hexadecane, with 2.0 wt% Span 80 surfactant. The dispersed phase consisted of an aqueous suspension of cellulose nanocrystals; this was diluted as appropriate with deionized water from an initial 14.5 wt % CNC suspension and vortexed to homogenize the sample prior to injection into the microfluidic device. Once the micro- emulsion was generated, it exited the microfluidic device through microbore polythene tubing (0i = 380 μηι, I ~ 10 cm) and was collected onto a microscope slide.

The aqueous phase was allowed to slowly diffuse into the oil at RTP, until solid

microparticles were formed. During droplet shrinkage a linear decrease in the droplet diameter was observed (Figure 12). This was typically in the range of 10 - 20 μηι.Ιτ ¹ for the large droplets (0A) and increasing to 25 μηι.ΐΎ ¹ for the smaller droplets (0B) reported in Figure 4 (see ESI). Microparticle Analysis

Residual surfactant was removed from dry microparticles by washing with n-hexane, prior to imaging by SEM. To image the interior of a droplet, it was fractured using the following protocol: the particles (on a substrate) were first placed in a nitrogen atmosphere, cooled in liquid nitrogen and finally mechanically crushed. The low temperature made the droplets more brittle, while the low humidity inhibited condensation of water.

Phase Diagram in the Capillaries The phase diagram (Figure 1 b) reports the proportion of anisotropic phase as a function of the total concentration of CNC in the suspension, calculated as the ratio of the volume of anisotropic phase to the total sample volume, and accounting for the uneven bottom and upper meniscus. From this diagram one can observe that the suspension is fully isotropic below ci = 5.97 wt % and fully anisotropic above CA = 14.59 wt % (values obtained from linear extrapolation).

The coexistence regime follows a higher slope at lower concentrations, with a transition around 9.7 wt % in the present case, and related to the change of pH and ionic strength as the sample concentration varies.

Pitch Measurements in the Capillaries

The cholesteric pitch in the glass capillaries was determined using both polarized optical microscopy and laser diffraction, as described below. Optical Microscopy

Long working distance objectives (TPIan Nikon, 20 x (NA = 0.30, WD = 30mm) and 50 x (NA = 0.40, WD = 22mm) objectives) were used in order to increase the depth of focus (approx. 61 μηι (20 x) and 19 μηι (50 x) as estimated from Berek's formula), which in turn helps with blurring out the fingerprint pattern of misaligned cholesteric domains. As a result, only the domains whose helices lay parallel to the observation plan contribute to visible fingerprints, with a tolerance angle δθ below 10 degrees allowing little uncertainty on the determined pitch values (error [1-1/cos(<50)] < 2%) for the considered pitch values).

Laser Diffraction

The formula given below:

Arj— n p sin arcsin

was used for the average optical index, n, calculated as the average of the effective optical indices given by Bruggeman modeling (described below). In this geometry, the diffracted peaks are observed in transmission, allowing higher diffraction orders to be observed for larger pitches and improve the pitch measurement. The diffracted light is mainly linearly polarized along the helix axis of the diffracting domains. This contrasts with the case of smaller pitches comparable with the wavelength of visible light. In the latter case, the first order diffraction is observed in reflection. At the limit of normal incidence, the diffracted light is bound in wavelengths (photonic band-gap) and is fully circularly polarized, with no existence of higher order diffraction peaks as long as the cholesteric helix profile remains sinusoidal.

Pitch Variation along the Vertical Dimension of a Capillary

The strong agreement between the two techniques confirms the validity of the observation conditions, in contrast to other pitch measurements available in the literature. Moreover, it allows detecting slight change of the pitch value from the bottom to the top of the capillaries, where measurements were performed at regular intervals. To our knowledge, this is the first clear observation of vertical pitch gradients in the anisotropic phase. Local size fractionation, salting-out gradients, or building-up of hydrostatic pressure in the lower levels of the anisotropic phase could explain this variation. Given this variation, we attributed to each prepared sample the average pitch of each series, as reported in Fig. 3, and we displayed as error bars the minimum and maximum of the locally measured values.

The Change of Pitch with Concentration in a Capillary

The pitch values in the capillaries initially follow a power law of p c ¹ as expected from Straley modeling of chiral non-flexible rods, though a slight change occurs around 9.7 wt % where the power law gets closer to approx. ^{5 3} as expected from Odijk modeling of semi- flexible chiral polymers. Such a small change in the power law is also visible in other CNC studies reported so far. Trends in the Evolution Pitch Diagram

Comparison of microdroplets of different size with a low starting concentration

Microdroplets of different size (-140 and -50 μηι) both with a low starting concentration of 7.3 wt % were prepared and their change in pitch upon concentration compared in

Figure 10. In both cases the cholesteric domains form a spherical shell and consequentially the trend in the pitch measurements overlaps reasonably well, following a power law from approx. c ^_1 to approx. ^{1 3} . This indicates there is negligible dependence on the size of the droplets on the pitch measurement.

Comparison Between Microdroplets at a High Starting Concentration with the Capillary.

Microdroplets of large size (-140 μηι) were prepared from a higher starting concentration ([CNC] = 10.9 wt %) and compared with the capillary series. The concentration dependence of the pitch in both cases overlap, with a change of power law (matching previous reports), initially closer to approx. σ ¹ and then to cr ^5/3 . Importantly, both these droplets and the capillaries display a polydomain structure, which reduces topological constraint on the pitch relaxation (i.e. the local nematic director on each end of a large cholesteric domain has to rotate fast to accommodate for the creation of more cholesteric bands). This polydomain structure can explain the small discrepancy noted in Figure 3, where geometrically-confined self-assembly into a large monodomain cholesteric shell (-140 μηι, [CNC] = 7.3 wt %) leads to long-range constraints on pitch relaxation. It is interesting to note that the trend in pitch in the measured droplets switches to approx. c ^1/3 at a threshold concentration at a value similar to c _g observed in Figure 3 of the manuscript.

The Rate of Water Loss from Shrinking Droplets

The aqueous microdroplets are submerged beneath a thin layer of hexadecane oil, and consequentially water loss from the droplets to the air is dependent upon diffusion through this barrier. The rate of water loss was found to be most dependent on two parameters that dominated over any variation in the droplets themselves: (i) the thickness of the oil barrier and (ii) the amount of surfactant present within the oil. The first factor is readily apparent when a large number of droplets are dispersed across a single large oil droplet, with those near the edge shrinking in a matter of minutes compared to hours for those nearer the center. The second factor is attributed to the surfactant acting as a micellar carrier for lost water; if the surfactant is diluted it was observed that the rate of water loss could be massively retarded. To overcome these factors, a constant surfactant concentration of 2.0 wt % span 80 was used throughout experiments and microdroplets near the edge of the macro-scale oil droplet (where complete loss of water in under an hour prevents radial ordering occurring prior to kinetic arrest) were discounted from all studies.

For the droplets reported in the manuscript, we observed a linear decrease in the droplet diameter, typically in the range of 10 - 20 μηι/h for the large droplets and increasing to 25 μηι/h for the small droplets used for SEM. The droplet shrinkage upon removal of water from the surface can be modeled by the following differential equation:

V(t+dt) = V(t) - Drate ^* S(t) ^* dt

where V and S represent the volume and the surface of the droplet respectively, and Drate is the diffusion rate through the interface (in m.s ^"1 , i.e. homogeneous to a diffusion coefficient multiplied by an interface thickness).

After explicating the radius R of the droplet, the equation simplifies to:

(dR / dt) = -Drate

leading to a linear radius dependency R(t) = R(0) - D _ra te ^* t, as shown on Figure 12, from which we extract D _ra te = 1.66 * 10 ^"3 m.s ^-1 . From this it is possible to evaluate the strain (έ) experienced by the assembly of the colloidal particles upon drying (ύεΙ dt) = d \n(R) I dt at (ds/dt) _t =o = -2.6 * 10 ^"5 S ^"1 and crossed c _g at (ds/dt) _cg = -3.3 * 10 ^"5 s- s-

Modeling of the optical behavior of cholesteric droplets In order to confirm the formation of a uniform radial chiral nematic phase inside the microdroplets, the polarized microscopy images of our droplets were compared with those obtained from numerical simulations, following a method well established in the literature. These were produced assuming a director field n of spherical components

(n _r , ΠΘ, ηψ) = (0, οοε( +2π r/p), 8ίη( +2π r/p)), where (r, θ, <f>) are the usual spherical coordinates, p is the helix pitch, as described in the Frank-Pryce model.

The liquid crystal phase is described as an effective birefringent medium with the optical axis parallel to the local nematic axis. The effective medium refractive indices and local birefringence are obtained from those of cellulose nanocrystals (A?H = 1.6180, n _± = 1.5436) and water (n _w = 1.33), with the effect of CNC concentration accounted by Bruggeman's theory applied to a dense assembly of aligned and non-conductive, intrinsically birefringent rods. The resulting birefringence is typically low and, as long as the diameter of the droplet is large compared to the wavelengths of visible light and the refractive index contrast between the droplet and the carrier fluid (hexadecane, ΠΗΘΧ = 1.43) is low, scattering and refraction of light at the surface of the droplet and multiple reflections in the bulk can be neglected. Light propagation is then only determined by the phase shift between orthogonally polarized waves and can be numerically implemented with the Mueller matrices method. Images were obtained assuming collimated monochromatic incident light and plotting the intensity of light transmitted through the second polarizer in different positions, using a color scale. To simulate the transmission through an additional sensitive tint plate, implemented in the experimental setup to infer the orientation of the rods slow axis, we determined the transmitted intensity of monochromatic light at three different wavelengths AR = 700 nm, AG = 530 nm and A _B = 470 nm) and produced colored images. The value of the pitch, p, introduced in the Frank-Pryce director field was the one measured from the fingerprint texture as twice the spacing of the bright lines. The simulations thus obtained reproduced to a high extent the features observed in the micrographs with and without the full wave plate.

Bruggeman's Theory

Bruggeman's theory is relevant for nanocomposites containing high volume fraction of small spherical or ellipsoidal inclusions that can be either isotropic (disoriented) or aligned. In the following expression we consider the specific case of perfectly aligned rod-like inclusions of dielectric material of negligible absorption (i.e. negligible imaginary permittivity) that is intrinsically birefringent (i.e. use of different dielectric values for the inclusion in parallel and perpendicular directions). l ^u . where

where φ is the CNCs volume fraction, the optical indices are related to the permittivities at optical frequencies by en = n/i ² , e± = n± ² , and e _w = n _w ² , and Nn = 0, Λ/±= 1/2 are the depolarization factors respectively in the parallel and the perpendicular directions with respect to the elongated CNC axis. The mass fraction [CNC] in wt% and its volume fraction ^ are related using the formula:

[CNC](wt%) = ρ φ Ι (ρΦ + ¾ (1 - φ))

using p = 1 ,630 kg.nr ³ for cellulose and po = 1 ,000 kg.nr ³ for water.

Discussion

An aqueous suspension of cellulose nanocrystals was prepared as described above. To characterize the lyotropic properties of this suspension, it was diluted to give a series of concentrations from 14.5 to 4.7 wt % CNC and the proportion of anisotropic phase was evaluated at each concentration (see Figure 1 (a)). This enabled the construction of a traditional phase diagram, as show in Figure 1 (b). This phase diagram allows for determination of the critical values of CNC concentration for this specific suspension at which the transition from isotropic to anisotropic phase occurs.

In order to understand the impact of geometrical confinement, it was necessary to study how the initial concentration of the CNC suspension affected self-assembly within a micron-scale droplet. Microdroplets were generated in a single step as an aqueous emulsion in hexadecane oil within a polydimethylsiloxane (PDMS) flow-focusing microfluidic device, as described in the Experimental Methods. At the flow-focusing junction the aqueous CNC suspension intersected perpendicularly with flows of hexadecane oil, resulting in

segmentation into monodisperse microdroplets (coefficient of variation < 2%), with a diameter defined by the geometry of the flow focus, and the relative flow rates and viscosities of the immiscible solutions. Once formed, the droplets were collected via microbore tubing onto a fluorophilic substrate for further study. It should be noted that the aqueous droplets are denser than the surrounding oil, and as such, settle onto the surface of the fluorophilic substrate rather than at the air-oil interface. The presence of this oil layer slows the loss of water from the droplets, allowing them to be studied over timescales from hours to days.

Microdroplets with a typical diameter of 140 μηι were prepared from a series of CNC concentrations across the phase transition, as indicated by the circles in Figure 1 (b). In all cases, the optical anisotropy of the suspension allowed for the ordering of CNC domains to be visualized during droplet formation by polarized optical microscopy, as described in the Experimental Methods. This is exemplified with a 14.5 wt % suspension of CNC, as denoted by the uppermost circle in the phase diagram (Figure 1 (b)). Here, the generation of an interface between the two fluids at the neck of the flow focus resulted in an immediate radial ordering of the CNC within the microdroplet, giving rise to a pronounced Maltese cross-like pattern when imaged under crossed-polarizers (Figure 1 (c)). However, this arrangement is rapidly disrupted by chaotic advection induced within the microdroplet by shearing interactions with its surroundings as it travels along the microfluidic channel. At such a high concentration, the suspension is highly viscous and is almost entirely in a liquid crystalline phase. For this reason, the shear experienced by the cholesteric phase inside the droplet induces many topological defects. This droplet consequentially remains trapped in this disordered anisotropic state without any particular re-ordering of the structure. Upon subsequent removal of water (by diffusion into the oil) and corresponding concentration of the confined suspension, this arrangement is preserved (Figure 5). A similar trend is observed for a lower concentration of 10.9 wt % (see circle in Figure 1 (b)). Here, despite microdroplets containing both isotropic and anisotropic phases, the relaxation towards a radial geometry was again not significantly observed after initial droplet formation (Figure 6). A markedly different assembly process is observed when the isotropic phase is dominant within the microdroplet, as is the case for 7.3 wt % suspension of CNC (see circle in

Figure 1 (b)). As before, radial ordering of the liquid crystalline structure is observed upon generation of the microfluidic droplets, however mixing within the droplet as it flows along the channel results in microdroplets in a predominantly isotropic phase, containing clearly defined factoids (Figure 7). Upon the loss of water from the droplet the factoids re-arrange, resulting in the formation of an ordered chiral nematic shell, growing inwards from the water- oil interface (Figure 2). Depending on the number and dimensions of the factoids (which is influenced by the individual composition of each microdroplet), either a chiral nematic shell containing free factoids, or a radial order throughout the entire diameter of the droplet is obtained (Figure 8,). In the latter case, such ordering can be preserved until the onset of buckling during the final stages of water loss from the droplet. The inventors attribute buckling to the interplay between increasing Laplace pressure at lower radius of curvature and the resistance to isotropic compression of the solidified CNC shell of the microdroplet, allow loss of the remaining water content without laterally compressing the rigidified surface. The presence of the chiral nematic organization throughout the entire diameter of the droplet is confirmed by the numerically simulated pattern shown in Figure 2(a). In particular, the observation of concentric dark and bright circles corresponds to the fingerprint pattern of the cholesteric structure. The superimposed Maltese-cross pattern is due to the isoclines of the radial cholesteric helix axis aligned with the axes of the crossed polarizers, in agreement with a planar anchoring of the CNC local director with the droplet interface. Low viscosity and a homogeneous composition are expected to increase the proportion of droplets retaining the radial chiral nematic order, with the in situ formation of a chiral nematic phase expected to reduce the generation of shear-induced topological defects. This was confirmed with droplets solely containing an isotropic suspension of CNC (5.8 wt %, circle in Figure 1 (b)), where the formation of independent factoids within the volume of the droplet was no longer observed upon droplet shrinkage, with the isotropic-anisotropic phase transition instead initiated exclusively at the water-oil interface (Figure 9).

Furthermore, lower concentrations have been tested. However, for values lower than 4 wt % the loss of water in ambient conditions leads only to a thin shell with transient cholesteric order. This shell is subsequently disrupted by buckling prior to the droplet core becoming sufficiently concentrated.

The observation of the microdroplet shrinkage process by polarized optical microscopy allowed for monitoring of the pitch evolution as a function of the water content. This particular configuration enables the construction of an 'evolution pitch diagram' which depicts the behavior of the cholesteric phase from low concentration at the equilibrium towards the final dry state. The pitch (p) is measured as twice the periodicity of the fingerprint pattern observed between crossed polarizers, while the suspension concentration is calculated from the droplet diameter. Variation in the pitch measured across a single droplet was within experimental error, indicating that the CNC concentration was uniform throughout the shrinking droplet. Similarly, droplets with comparable CNC concentration displayed the same cholesteric pitch. The values of the pitch measured during the loss of water from the microdroplets are reported in Figure 3 and compared to the values of the pitch independently obtained from measurement in a glass capillary (see Experimental Methods). It is worth noting here that for a given initial value of CNC concentration, the same trend in pitch is measured irrespective of the dimensions of the initial droplets (Figure 10).

The essence of the self-assembly process of CNC in suspension is illustrated by Figure 3. At a low concentration of CNC in suspension, the cholesteric pitch observed both in the droplets and in the capillaries overlap and appears to be inversely proportional to

concentration, p∞c ¹ , as expected from Straley modeling of chiral non-flexible rods. The small discrepancy between the two pitch measurements above 9.7 wt % is attributed to the cholesteric monodomain within the droplets. In contrast, droplets prepared from [CNC] =10.9 wt % remain trapped in a polydomain structure, and as such more closely follow the capillary data. Significantly, above a critical concentration denoted here c _g , a transition after which the pitch scales as p∞c ^1/3 is observed. This transition is attributed to the manifestation of the kinetic arrest, where the sample cannot relax. Indeed, such kinetic arrest is expected to take place at some point during shrinkage towards a fully-dried sphere of self-assembled CNCs. The value of c _g for this specific suspension (12% v/v ~ 19 wt % in Figure 3) was found to be comparable between individual radially-ordered cholesteric droplets, irrespective of the initial size, CNC concentration or rate of water removal. The concentration at which this transition occurs has been discussed and addressed in the literature as a key factor in the

understanding of the self-assembly of CNC, but remains challenging to assess. As the suspension gets kinetically trapped, the cholesteric structure cannot relax over time, but it can be still affected by the local shear experienced upon drying and therefore it is sensitive to any geometrical constraints. In this system, the spherical geometry leads to a three- dimensional contraction of the cholesteric structure, in agreement with the observed power law. This contrasts with the usual configuration where a film is cast onto a planar substrate, which leads to a unidirectional, vertical straining of the cholesteric structure and

consequentially to a similar power law of ρχσ ¹ . For this reason, the spherical geometry allows for a clear discrimination of the transition upon kinetic arrest.

Finally, Figure 4 shows the morphology of CNC micro particles after the complete loss of water. For this experiment, smaller microdroplets (50 μηι in diameter) are employed to minimize the effects of buckling on the particle structure. Similarly, to the case of planar CNC films, the chiral nematic nature can be maintained, evidenced by the clear helicoidal structure observed in the scanning electron microscopy (SEM) image reported in

Figure 4(d). Moreover, the value of the pitch measured directly from such images, closely matches the extrapolated confined pitch for a CNC concentration near 100%, as indicated by the dashed blue line in Figure 3. In contrast, casting a planar film via slow evaporation using the same suspension (Figure 11) results in a cholesteric pitch consistent with the standard power law behavior (dashed red line). Additional Experimental and Results

Further experiments in support of the invention are described in detail below, with

comparisons with the particles prepared according to the processes described above.

Change in Pitch with Change in Ionic Strength

In the original work described above, a particle was prepared having a measured pitch of around 1.3 μηι (see Figure 3). The preparation was repeated, but with the addition of further acid to the aqueous suspension. Thus, the initial suspension contained 7.25 wt% CNC with 8 mM H2SO4 (1 10 mmol/kg CNC) and the pitch of the dried particle prepared from the suspension was determined to be about 0.7 μηι. Figure 13 is an SEM cross-section of the microparticle product prepared from this suspension. The particle was embedded in a polymer matrix to allow for cross-sectioning.

This a significant reduction from the pitch value described in the work above, where only 5 mM H2SO4 (69 mmol/kg) was present in the initial suspension. Significantly increasing the acid concentration to 15 mM (205 mmol/kg), was found to give spherical particles (that is particles having less buckling) that did not show significant self-assembly during the drying process.

For simplicity, the added electrolyte was sulfuric acid, but salt (such as NaCI) was seen to act similarly. Thus, changes in ionic strength may be use to change the pitch in the particle product.

Change in Pitch with Washing and Drying

The particles prepared in the method above were swollen in ethanol and subsequently dried, thereby anisotropically collapsing the particle. This method can produce spots of visible colour in the particle. The observed red colour shows in some areas of the particles shows that the pitch has been compressed to around 0.4 μηι by this post-treatment.

Figure 14 is a microscope image of the particles produced in this method. Red colour spots are clearly visible in the imagery across all the particles produced.

Changes in Drying Rate

In the original work described above, a particle was prepared in a continuous phase having a surfactant at a relatively high concertation (2 wt % Span 80 in hexadecane). In a further experiment, the concentration of the surfactant in the continuous phase was reduced during the particle preparation. It was found that a droplet stored in a dish where the surrounding oil phase containing low surfactant concentration (0.5 wt % Span 80 in hexadecane) evaporated much more slowly than one stored in higher concentration (2 wt % Span 80 in hexadecane), with a linear contraction of the diameter by 1 μηι/h and 5 μηι/h respectively. This is sufficiently slow, such that the interval for self-assembly to occur before gelation is up to 24 hours, enabling the relaxation of droplets of higher initial CNC concentration than previously observed. Once self-assembly has occurred it is believed that the drying rate is no longer significant. References

All documents mentioned in this specification are incorporated herein by reference in their entirety. Amstad et al. Lab Chip 2016, 16, 4163

Azzam et al. Langmuir 2016, 32, 4305

Bardet et al. Carbohydrate Polymers 2015, 122, 367

Cheung et al. ACS Macro Lett. 2013, 2, 1016

Cho et al. Angew. Chem. Int. Ed. 2016, 55, 14014

Cipparrone et al. Adv. Mater. 2011 , 23, 5773

Cipparrone et al. Liquid Crystals Reviews. 2016, 4, 59

Detsi, E., Rijksuniversiteit Groningen, Zernike Institute, PhD Thesis, 2012

Dumanli et al. ACS Appl. Mater. Interfaces 2014, 6, 12302

Elazzouzi-Hafraoui et al. J. Phys. Chem. B 2009, 113, 1 1069

Geng et al. Scientific Reports 2016, 6, 26840

Jativa et al. Soft Matter 2015, 11, 5374

Klemm et al. Comprehensive Cellulose Chemistry, Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, FRG, 1998

Lagerwall et al. NPG Asia Materials 2014, 6, e80

Lakhtakia et al. J. Phys. D: Appl. Phys. 1997, 30, 230

Li et al. Nat. Commun. 2016, 7, 12520

Li et al. Proc. Natl. Acad. Sci. 2017, 114, 2137

Musevic Phil. Trans. R. Soc. A 371 , 20120266

Park et al. Chem. Phys. Chem. 2014, 15, 1477

Revol et al. J. Pulp Paper Science 1998, 24, 146

Robinson et al. Discuss. Faraday Soc. 1958, 25, 29-42

Wang et al. Angew. Chem. Int. Ed. 2016, 55, 12460

WO 95/21901.