Article
Chiral Nematic Structure of Cellulose Nanocrystal Suspensions and Films; Polarized Light and Atomic Force Microscopy Derek G. Gray * and Xiaoyue Mu Received: 5 October 2015 ; Accepted: 10 November 2015 ; Published: 20 November 2015 Academic Editors: Aldo R. Boccaccini, Jung Kwon (John) Oh and Juergen Stampfl Department of Chemistry, McGill University, Pulp and Paper Building, 3420 University Street, Montreal, QC H3A-2A7, Canada;
[email protected] * Correspondence:
[email protected]; Tel.: +1-514-398-6182
Abstract: Cellulosic liquid crystalline solutions and suspensions form chiral nematic phases that show a rich variety of optical textures in the liquid crystalline state. These ordered structures may be preserved in solid films prepared by evaporation of solvent or suspending medium. Film formation from aqueous suspensions of cellulose nanocrystals (CNC) was investigated by polarized light microscopy, optical profilometry and atomic force microscopy (AFM). An attempt is made to interpret qualitatively the observed textures in terms of the orientation of the cellulose nanocrystals in the suspensions and films, and the changes in orientation caused by the evaporative process. Mass transfer within the evaporating droplet resulted in the formation of raised rings whose magnitude depended on the degree of pinning of the receding contact line. AFM of dry films at short length scales showed a radial orientation of the CNC at the free surface of the film, along with a radial height variation with a period of approximately P/2, ascribed to the anisotropic shrinkage of the chiral nematic structure. Keywords: cellulose nanocrystals; chiral nematic order; polarized light microscopy; fingerprint texture; reflection colours; coffee-stain effect; sessile droplets; contact line pinning
1. Introduction Cellulosic liquid crystalline materials, both cellulose derivatives [1,2] and cellulose nanocrystals [3,4] form chiral nematic phases that show a rich variety of optical textures in the liquid crystalline state. The ordered structures in the liquid crystalline state may be preserved with modification in solid films prepared by simple evaporation of solvent or suspending medium. Polarized light microscopy, supplemented by scanning probe microscopy and electron microscopy provides an accessible tool to observe the structure of these ordered suspensions and films. Here, we present some observations on the chiral nematic self-organization of cellulose-based systems, with emphasis on the aqueous suspensions of cellulose nanocrystals (CNC) and of films formed from CNC by evaporation. The factors that govern phase separation, chiral nematic pitch, mechanical properties and applications of CNC and other cellulosic nanocrystals have been reviewed elsewhere, [5–7] and will not be discussed in detail here.
Materials 2015, 8, 7873–7888; doi:10.3390/ma8115427
www.mdpi.com/journal/materials
Materials 2015, 8, 7873–7888
2. Results and Discussion Materials 2015, 8, page–page
2.1. Fluid Cellulosic Chiral Nematic Textures 2. Results and Discussion Materials 2015, 8, page–page
2.1.1. Fluid Phases, Fingerprint Texture
2. Results and Discussion 2.1. Fluid Cellulosic Chiral Nematic Textures
Above some critical concentration, suspensions of cellulose nanocrystals (CNC) form chiral 2.1. Fluid Cellulosic Chiral Nematic Textures 2.1.1. Fluid Phases, Fingerprint Texture nematic phases [3]. The liquid crystalline textures observed by polarized light microscopy depend Above some critical concentration, suspensions cellulose nanocrystals (CNC) chiral on the orientation of the optical axis of the helicoidalof assembly relative to that of form the microscope. 2.1.1. Fluid Phases, Fingerprint Texture nematic phases [3]. The liquid crystalline textures observed by polarized light microscopy depend Figure 1 shows a typical image of an aqueous CNC suspension with regions (a) of planar texture and Above some critical concentration, suspensions of cellulose nanocrystals (CNC) form chiral on the orientation of the optical axis of the helicoidal assembly relative to that of the microscope. Figure 1 (b) of fingerprint texture, where the chiral nematic axis is parallel to and orthogonal to the plane of nematic phases [3]. The liquid crystalline textures observed by polarized light microscopy depend shows a typical image of an aqueous CNC suspension with regions (a) of planar texture and (b) of the sample respectively. on the orientation of the optical axis of the helicoidal assembly relative to that of the microscope. Figure 1 fingerprint texture, where the chiral nematic axis is parallel to and orthogonal to the plane of the shows a typical image of an aqueous CNC suspension with regions (a) of planar texture and (b) of Asample respectively. variety of disclinations [8] where fingerprint lines intersect are also visible. The distance where the chiral nematic axis is parallel to and to of the plane of the betweenfingerprint the linestexture, in fingerprint texture is P/2, where P isorthogonal the also pitch the chiral nematic A variety of the disclinations [8] where fingerprint lines intersect are visible. The distance sample respectively. between the lines in the fingerprint texture is P/2, where P is the pitch of the chiral nematic structure. structure. The orientation of the CNC rods is shown in the idealized sketch in Figure 2. Over time, A variety of disclinations [8] where fingerprint lines intersect are also visible. The distance The orientation of the CNC rods is shown in the idealized sketch in Figure 2. Over time, the CNC the CNC tend to orient parallel to the walls of the microslide, so that that the chiral nematic axis between the lines in the fingerprint texture is P/2, where P is the pitch of the chiral nematic structure. tend to orient parallel the walls the microslide, that that the chiral nematic axis becomes becomes orthogonal to the to plane of theof sample, and theso fingerprint texture disappears. However, by The orientation of the CNC rods is shown in the idealized sketch in Figure 2. Over time, the CNC orthogonal to the plane of the sample, and the fingerprint texture disappears. However, by appropriate application of ato magnetic field the CNC unidirectional textures tend to orient parallel the walls of the to microslide, so samples, that that the chiral nematic fingerprint axis becomes appropriate application of a magnetic field to the CNC samples, unidirectional fingerprint textures orthogonal to may the plane of the sample, and the fingerprint texture disappears. However, by over extended areas be generated [9]. over extended areas may be generated [9]. appropriate application of a magnetic field to the CNC samples, unidirectional fingerprint textures over extended areas may be generated [9].
Figure 1. Image of 8% aqueous suspension (Batch SB‐8%, see experimental section) of cellulose Figure 1. Image of 8% aqueous suspension (Batch SB-8%, see experimental section) of cellulose nanocrystals in glass microslide. Regions (a) show planar texture and (b) show fingerprint texture Figure 1. Image of 8% aqueous suspension (Batch SB‐8%, see experimental section) of cellulose nanocrystals in glass microslide. Regions (a) show planar texture and (b) show fingerprint texture (see Figure 2). nanocrystals in glass microslide. Regions (a) show planar texture and (b) show fingerprint texture (see Figure 2). (see Figure 2).
Figure 2. Simplified sketch of organization of chiral nematic textures shown in Figure 1. Regions (a) Figure 2. Simplified sketch of organization of chiral nematic textures shown in Figure 1. Regions (a) and (b) correspond to of planar and fingerprint textures, respectively. The optical axis of Figure 2. Simplified sketch organization of chiral nematic textures shown in Figure 1. the Regions and (b) correspond to planar and fingerprint textures, respectively. The optical axis of the microscope is from top to bottom of the sketch, with the middle of the sample in focus. The light and (a) and (b) correspond to planar and fingerprint textures, respectively. The optical axis microscope is from top to bottom of the sketch, with the middle of the sample in focus. The light and of the dark lines under crossed polars originate from variations in birefringence due to the orientation of microscope is from top to bottom of the sketch, with the middle of the sample in focus. The light dark lines under crossed polars originate from variations in birefringence due to the orientation of the cellulose nanocrystals (CNCs) along and across the microscope optical axis [8]. the cellulose nanocrystals (CNCs) along and across the microscope optical axis [8]. and dark lines under crossed polars originate from variations in birefringence due to the orientation
2 of the cellulose nanocrystals (CNCs) along and across 2 the microscope optical axis [8].
7874
Materials 2015, 8, 7873–7888 Materials 2015, 8, page–page
2.1.2.Materials 2015, 8, page–page Flow Distortion of Fingerprint Texture 2.1.2. Flow Distortion of Fingerprint Texture As the concentration of CNC is increased, the chiral nematic phase is first observed in a 2.1.2. Flow Distortion of Fingerprint Texture As the concentration of CNC is increased, the chiral nematic phase is first observed in a biphasic biphasic region, where the isotropic and chiral nematic phases co-exist [3]. On standing, a sharp 2.1.2. Flow Distortion of Fingerprint Texture As the concentration of CNC is increased, the chiral nematic phase is first observed in a biphasic region, where the isotropic and chiral nematic phases co‐exist [3]. On standing, a sharp boundary boundary forms between the upper isotropic phase and the lower liquid crystalline phase [10,11]. region, where the isotropic and chiral nematic phases co‐exist [3]. On standing, a sharp boundary As the concentration of CNC is increased, the chiral nematic phase is first observed in a biphasic forms between the upper isotropic phase and the lower liquid crystalline phase [10,11]. However, However, fluid samples in sealed microslides are normally rotated from the vertical position to a forms between the upper isotropic phase and the lower liquid crystalline phase [10,11]. However, region, where the isotropic and chiral nematic phases co‐exist [3]. On standing, a sharp boundary fluid samples in sealed microslides are normally rotated from the vertical position to a horizontal horizontal position for viewing with a polarized light microscope, and the resultant sample flow fluid samples in sealed microslides are normally rotated from the vertical position to a horizontal forms between the upper isotropic phase and the lower liquid crystalline phase [10,11]. However, position for viewing with a polarized light microscope, and the resultant sample flow causes causes distortion of the observed texture, light as sketched in Figure 3 and illustrated Figure 4. The position for viewing with a polarized microscope, and the resultant sample inflow causes fluid samples in sealed microslides are normally rotated from the vertical position to a horizontal distortion of the observed texture, as sketched in Figure 3 and illustrated in Figure 4. The effect of the effectdistortion of the observed texture, as sketched in Figure 3 and illustrated in Figure 4. The effect of the of the gravity-driven flow at the interface leads to an apparent increase in the spacing of the position for viewing with a polarized light microscope, and the resultant sample flow causes gravity‐driven flow at the interface leads to an apparent increase in the spacing of the fingerprint fingerprint lines, due to the oblique viewing angle. A similar distortion is also observed in the gravity‐driven flow at the interface leads to an apparent increase in the spacing of the fingerprint distortion of the observed texture, as sketched in Figure 3 and illustrated in Figure 4. The effect of the lines, due to the oblique viewing angle. A similar distortion is also observed in the bulk of the liquid bulk gravity‐driven flow at the interface leads to an apparent increase in the spacing of the fingerprint of thelines, due to the oblique viewing angle. A similar distortion is also observed in the bulk of the liquid liquid crystalline phase (Figure 5). crystalline phase (Figure 5). Materials 2015, 8, page–page
crystalline phase (Figure 5). lines, due to the oblique viewing angle. A similar distortion is also observed in the bulk of the liquid crystalline phase (Figure 5).
Figure 3. Distortion of isotropic‐liquid crystalline phase boundary caused by flow of the sample fluid Figure 3. Distortion of isotropic-liquid crystalline phase boundary caused by flow of the sample fluid Figure 3. Distortion of isotropic‐liquid crystalline phase boundary caused by flow of the sample fluid when placed on the microscope stage. (a) Biphasic sample in glass microslide; (b) Cross‐section of
when placed on the microscope stage. (a) Biphasic sample in glass microslide; (b) Cross-section of when placed on the microscope stage. (a) Biphasic sample in glass microslide; (b) Cross‐section of Figure 3. Distortion of isotropic‐liquid crystalline phase boundary caused by flow of the sample fluid phase boundary. Viewing direction indicated by arrows. phase boundary. Viewing direction indicated by arrows. phase boundary. Viewing direction indicated by arrows. when placed on the microscope stage. (a) Biphasic sample in glass microslide; (b) Cross‐section of phase boundary. Viewing direction indicated by arrows.
Figure 4. Polarized light image of the isotropic‐chiral nematic phase boundary for an aqueous 6% Figure 4. Polarized light image of the isotropic‐chiral nematic phase boundary for an aqueous 6% CNC suspension (SB‐6%). The isotropic phase (dark under crossed polars) is on the left, the chiral Figure 4. Polarized light image of the isotropic-chiral nematic phase boundary for an aqueous 6% CNC suspension (SB‐6%). The isotropic phase (dark under crossed polars) is on the left, the chiral Figure 4. Polarized light image of the isotropic‐chiral nematic phase boundary for an aqueous 6% nematic phase is on the right. The sample is oriented as sketched in Figure 3. CNC suspension (SB-6%). The isotropic phase (dark under crossed polars) is on the left, the chiral nematic phase is on the right. The sample is oriented as sketched in Figure 3. CNC suspension (SB‐6%). The isotropic phase (dark under crossed polars) is on the left, the chiral
nematic phase is on the right. The sample is oriented as sketched in Figure 3. nematic phase is on the right. The sample is oriented as sketched in Figure 3.
Figure 5. Polarized light microscope image (crossed polars, wave plate) of 6% aqueous CNC of 6% aqueous CNC Figure 5. Polarized light microscope image (crossed polars, wave plate) suspension (SB‐6%), viewed between crossed polars with 530 nm red wave plate. The colors are due suspension (SB‐6%), viewed between crossed polars with 530 nm red wave plate. The colors are due Figure 5. Polarized light microscope image (crossed polars, wave plate) of 6% aqueous CNC
to the variations in orientation of the birefringent suspension. Figure 5. Polarized light microscope image (crossed polars, wave plate) of 6% aqueous CNC to the variations in orientation of the birefringent suspension. suspension (SB‐6%), viewed between crossed polars with 530 nm red wave plate. The colors are due suspension (SB-6%), viewed between crossed polars 3 with 530 nm red wave plate. The colors are to the variations in orientation of the birefringent suspension. 3 suspension. due to the variations in orientation of the birefringent 3
7875
Materials 2015, 8, 7873–7888
As the concentration of CNC in the suspensions increases, the chiral nematic pitch should in principle decrease to values of the order of visible light, where iridescent colours should be visible. Materials 2015, 8, page–page However, before such concentrations are reached, freezing-in of the pitch due to gelation and glass formation may occur [12]. Materials 2015, 8, page–page
As the concentration of CNC in the suspensions increases, the chiral nematic pitch should in principle decrease to values of the order of visible light, where iridescent colours should be visible. 2.1.3. Cellulosic Liquid Crystalline Solutions that Reflect Visible Light As the concentration of CNC in the suspensions increases, the chiral nematic pitch should in However, before such concentrations are reached, freezing‐in of the pitch due to gelation and glass principle decrease to values of the order of visible light, where iridescent colours should be visible. formation may occur [12]. While CNC suspensions don’t usually achieve pitch values small enough to reflect visible However, before such concentrations are reached, freezing‐in of the pitch due to gelation and glass light, molecularly dispersed solutions of cellulose derivatives often show iridescent colours due formation may occur [12]. 2.1.3. Cellulosic Liquid Crystalline Solutions that Reflect Visible Light
to reflection of circularly polarized light of visible wavelengths, although the colours take a long While CNC suspensions don’t usually achieve pitch values small enough to reflect visible light, time to 2.1.3. Cellulosic Liquid Crystalline Solutions that Reflect Visible Light form, presumably because the polymer chains take a long time to disentangle and adopt molecularly dispersed solutions of cellulose derivatives often show iridescent colours due to While CNC suspensions don’t usually achieve pitch values small enough to reflect visible light, chiral nematic order. Thus, for the (hydroxypropyl)cellulose (HPC)-water system [1], the effect of reflection of circularly polarized light of visible wavelengths, although the colours take a long time molecularly dispersed solutions of cellulose derivatives often show iridescent due to polymerto form, presumably because the polymer chains take a long time to disentangle and adopt chiral concentration on chiral nematic pitch is shown in Figure 6. These imagescolours were photographed reflection of circularly polarized light of visible wavelengths, although the colours take a long time by transmitted light between crossed polars. Thus, isotropic regions should be black, birefringent nematic order. Thus, for the (hydroxypropyl)cellulose (HPC)‐water system [1], the effect of polymer to form, presumably because the polymer chains take a long time to disentangle and adopt chiral on chiral shown in regions Figure 6. images were photographed by colour regions concentration should be bright andnematic planar pitch chiralis nematic inThese Figure 6a,c,d should display the nematic order. Thus, for the (hydroxypropyl)cellulose (HPC)‐water system [1], the effect of polymer transmitted light between crossed polars. Thus, isotropic regions should be black, birefringent of circularly polarized light nematic transmitted byshown the sample. other hand of photographed circularly polarized is concentration on chiral pitch is in Figure (The 6. These images were by reflectedregions should be bright and planar chiral nematic regions in Figure 6a,c,d should display the colour from the sample.) The crossed samplespolars. appear red isotropic (Figure 6a), green (Figure 6b) and (Figure 6c) and transmitted light between Thus, regions should be black, birefringent of circularly polarized light transmitted by the sample. (The other hand of circularly polarized is regions should be bright and planar chiral nematic regions in Figure 6a,c,d should display the colour blue (Figure 6d) to the naked eye, irrespective of the texture. The textures are characteristic of typical reflected from the sample.) The samples appear red (Figure 6a), green (Figure 6b) and (Figure 6c) of circularly polarized light liquid transmitted by the sample. (The other circularly and polarized low molar mass chiral nematic crystals, but the samples arehand veryof viscous, someis textures and blue (Figure 6d) to the naked eye, irrespective of the texture. The textures are characteristic of reflected from the sample.) The samples appear red (Figure 6a), green (Figure 6b) and (Figure 6c) require typical low molar mass chiral nematic liquid crystals, but the samples are very viscous, and some time (weeks) to form. and blue (Figure 6d) to the naked eye, irrespective of the texture. The textures are characteristic of textures require time (weeks) to form. Antypical low molar mass chiral nematic liquid crystals, but the samples are very viscous, and some example of the reflected light from a blue HPC sample illuminated by white light against a An example of the reflected light from a blue HPC sample illuminated by white light against a black background is shown in Figure 7. In view of the discussion below, it should be noted that the textures require time (weeks) to form. black background is shown in Figure 7. In view of the discussion below, it should be noted that the wavelengthAn example of the reflected light from a blue HPC sample illuminated by white light against a of reflected light, and hence the chiral nematic pitch, at the edge of the sample is smaller wavelength of reflected light, and hence the chiral nematic pitch, at the edge of the sample is smaller black background is shown in Figure 7. In view of the discussion below, it should be noted that the than that at the centre of the sample. In fact, the edge of the HPC sample shown in Figure 7 is clear, than that at the centre of the sample. In fact, the edge of the HPC sample shown in Figure 7 is clear, wavelength of reflected light, and hence the chiral nematic pitch, at the edge of the sample is smaller suggesting that the reflection wavelength has moved into the ultraviolet region as the sample dries. suggesting that the reflection wavelength has moved into the ultraviolet region as the sample dries. than that at the centre of the sample. In fact, the edge of the HPC sample shown in Figure 7 is clear, suggesting that the reflection wavelength has moved into the ultraviolet region as the sample dries.
Figure 6. (Hydroxypropyl)cellulose (HPC)/water polarized light microscope, crossed polars, polars, Figure 6. (Hydroxypropyl)cellulose (HPC)/water polarized light microscope, crossed transmission images (a) ~45% HPC, planar and focal conic textures; (b) 55% HPC, focal conic texture; Figure images 6. (Hydroxypropyl)cellulose (HPC)/water polarized light (b) microscope, crossed polars, texture; transmission (a) ~45% HPC, planar and focal conic textures; 55% HPC, focal conic (c) 55% HPC, oily streak texture; (d) ~65% HPC, planar and focal conic textures. The colours of the transmission images (a) ~45% HPC, planar and focal conic textures; (b) 55% HPC, focal conic texture; (c) 55% planar regions between crossed polars correspond to the reflection colours of the samples. HPC, oily streak texture; (d) ~65% HPC, planar and focal conic textures. The colours of the (c) 55% HPC, oily streak texture; (d) ~65% HPC, planar and focal conic textures. The colours of the planar regions between crossed polars correspond to the reflection colours of the samples. planar regions between crossed polars correspond to the reflection colours of the samples.
Figure 7. Reflection color from a thin 65 wt % solution of HPC in water, between a microscope slide and a 22 mm square cover glass. The sample is illuminated by white light against a black Figure 7. Reflection color from a thin 65 wt % solution of HPC in water, between a microscope slide Figure background. Evaporation of water leads to a higher concentration and shorter reflection wavelength at 7. Reflection color from a thin 65 wt % solution of HPC in water, between a microscope and a 22 mm square cover glass. The sample is illuminated by white light against a black edge of sample. slide and a 22 mm square cover glass. The sample is illuminated by white light against a black background. Evaporation of water leads to a higher concentration and shorter reflection wavelength at 4 concentration and shorter reflection wavelength background. Evaporation of water leads to a higher edge of sample.
at edge of sample.
4
7876
Materials 2015, 8, 7873–7888 Materials 2015, 8, page–page
2.2. Solid Cellulosic Films with Chiral Nematic Properties 2.2. Solid Cellulosic Films with Chiral Nematic Properties Cellulose and cellulose derivatives have long been exploited for their film‐forming properties. Cellulose and cellulose derivatives have long been exploited for their film-forming properties. Examples include cellophane, the the first first transparent transparent food wrap, and and cellulose cellulose acetate, whose use as Examples include cellophane, food wrap, acetate, whose use as base for photographic films has been replaced by its use as protective and passive optical elements in base for photographic films has been replaced by its use as protective and passive optical elements liquid combined with with their their in liquidcrystal crystalflat‐screen flat-screendisplays. displays.If Ifthese thesefilm‐forming film-formingproperties properties could could be be combined tendency to form chiral nematic solutions and suspensions, the unique optical properties of chiral tendency to form chiral nematic solutions and suspensions, the unique optical properties of chiral nematic liquid crystals would would become become available in a cheap easily handled solid film, film, based based on on an an nematic liquid crystals available in a cheap easily handled solid essentially renewable resource. To To date, the cellulose cellulose derivatives derivatives have proved difficult, in general essentially renewable resource. date, the have proved difficult, in general forming viscous viscous solutions solutions and and solids solids that that are are slow slow toto develop develop chiral chiral nematic nematic organization. organization. CNC CNC forming suspensions in water are less viscous, and chiral nematic suspensions and films are more readily suspensions in water are less viscous, and chiral nematic suspensions and films are more readily prepared. However, films remains challenging. In this prepared. However, control controlof ofthe theorientation orientationand andthe theCNC CNCin indry dry films remains challenging. In work, we focus on CNC films produced by simple evaporation of water from a sessile droplet of this work, we focus on CNC films produced by simple evaporation of water from a sessile droplet of CNC suspension on a flat surface. Hoeger et al. [13] showed by means of a convective‐shear CNC suspension on a flat surface. Hoeger et al. [13] showed by means of a convective-shear assembly assembly that capillary shear and capillary forces alignment generated alignment parallel and normal to the withdrawal that shear and forces generated parallel and normal to the withdrawal direction direction of a deposition plate, respectively. Both shear and capillary forces might be expected to act of a deposition plate, respectively. Both shear and capillary forces might be expected to act on the on the orientation of CNC deposited at the edge of an evaporating droplet. It is of interest to test orientation of CNC deposited at the edge of an evaporating droplet. It is of interest to test which which effect predominates. effect predominates. 2.2.1. Evaporation of Water from Edge of Covered Chiral Nematic CNC Suspensions 2.2.1. Evaporation of Water from Edge of Covered Chiral Nematic CNC Suspensions For samples sandwiched between microscope slide and cover glass, evaporation can only occur For samples sandwiched between microscope slide and cover glass, evaporation can only occur from the edge of the sample, resulting in a concentration gradient decreasing from the edge to the from the edge of the sample, resulting in a concentration gradient decreasing from the edge to the centre the sample. This gradient willwill influence the orientation of theof nanocrystals at the point where centre ofof the sample. This gradient influence the orientation the nanocrystals at the point they are frozen into the solid. Other factors will also affect the orientation and texture of the film where they are frozen into the solid. Other factors will also affect the orientation and texture of the produced on drying (Figure(Figure 8). When coverthe glass on top of aon droplet microscope film produced on drying 8). placing When the placing cover glass top of ona the droplet on the slide (Figureslide 8a), (Figure the droplet is forced to is spread forms as a thin layera of uniform thickness microscope 8a), the droplet forced asto itspread it forms thin layer of uniform (Figure 8b).(Figure This process generates both radial and transverse shear forces on shear the anisotropic thickness 8b). This process generates both radial and transverse forces on and the viscoelastic liquid crystalline fluid. Low viscosity samples will be drawn to the edge of the cover anisotropic and viscoelastic liquid crystalline fluid. Low viscosity samples will be drawn to the edge of the glass by capillary forces. If the contact and the slide and glass bycover capillary forces. If the contact lines between thelines fluidbetween and the the slidefluid and coverglass are not coverglass are not pinned, then the droplet radius may shrink during evaporation (Figure 8c). pinned, then the droplet radius may shrink during evaporation (Figure 8c).
Figure 8. Sketch of sample preparation between microscope slide and cover glass. (a) Sessile droplet Figure 8. Sketch of sample preparation between microscope slide and cover glass. (a) Sessile droplet with finite contact angle on microscope slide; (b) Droplet spreads under cover glass; (c) Droplet with finite contact angle on microscope slide; (b) Droplet spreads under cover glass; (c) Droplet shrinks due to edge evaporation. shrinks due to edge evaporation.
Complex textures are observed for dry films of 8% CNC suspensions allowed to dry between Complex textures are observed for dry films of 8% CNC suspensions allowed to dry between microscope slide and cover glass (Figure 9). For this sample and CNC concentration, the diameter of microscope slide and glass (Figure 9). initial For thisdroplet samplebetween and CNC concentration, diameter of the dry film was the cover same as that of the slide and cover the glass. Thus the the dry film was the same as that of the initial droplet between slide and cover glass. Thus the contact contact lines between the drying droplet and the glass surfaces must have been pinned, and the evaporation of water from the edge of the sample resulted in a decrease in thickness only; the situation sketched in Figure 8c did not apply here. The graininess of the texture was smallest close to 7877 5
Materials 2015, 8, 7873–7888
lines between the drying droplet and the glass surfaces must have been pinned, and the evaporation of water from the edge of the sample resulted in a decrease in thickness only; the situation sketched in Materials 2015, 8, page–page Figure 8c did not apply here. The graininess of the texture was smallest close to the edge of the sample the edge of the sample (bottom right of the image), increasing towards the centre of the cover glass. (bottom right of the image), increasing towards the centre of the cover glass. What appear to be Materials 2015, 8, page–page What appear to be residual areas of distorted fingerprint texture are evident away from the sample residual areas of distorted fingerprint texture are evident away from the sample centre. This suggests centre. This suggests that the relatively rapid drying at the edge of the sample generates a the edge of the sample (bottom right of the image), increasing towards the centre of the cover glass. that the relatively rapid drying at the edge of the sample generates a polydomain structure, but as the polydomain structure, but as the process proceeds, the domains enlarge until, at the centre, the What appear to be residual areas of distorted fingerprint texture are evident away from the sample process proceeds, the domains enlarge until, at the centre, the sample assumes an essentially planar sample assumes an essentially planar texture, oriented with the chiral nematic axis orthogonal to the centre. This suggests that the relatively rapid drying at the edge of the sample generates a texture, oriented with the chiral nematic axis orthogonal to the microscope slide and cover glass. microscope slide and cover glass. polydomain structure, but as the process proceeds, the domains enlarge until, at the centre, the sample assumes an essentially planar texture, oriented with the chiral nematic axis orthogonal to the microscope slide and cover glass.
Figure 9. Texture (polarized light microscope, crossed polars, 530 nm red wave plate) of 8 wt %
Figure 9. Texture (polarized light microscope, crossed polars, 530 nm red wave plate) of 8 wt % (Batch SB‐8%) CNC suspension on drying between microscope slide and cover glass. (Batch SB-8%) CNC suspension on drying between microscope slide and cover glass. Figure 9. Texture (polarized light microscope, crossed polars, 530 nm red wave plate) of 8 wt % (Batch SB‐8%) CNC suspension on drying between microscope slide and cover glass.
2.2.2. Evaporation of Water from Uncovered Droplets of Chiral Nematic CNC Suspensions
2.2.2. Evaporation of Water from Uncovered Droplets of Chiral Nematic CNC Suspensions
The edge‐wise evaporation of water from a sample of CNC suspension between glass slide and
2.2.2. Evaporation of Water from Uncovered Droplets of Chiral Nematic CNC Suspensions cover glass is slow and impractical on a larger scale. What happens if the droplet is left uncovered? The edge-wise evaporation of water from a sample of CNC suspension between glass slide and Aqueous droplets containing low concentrations of rod‐like CNC have been reported to display a cover glass is slow and impractical on a larger scale. What happens if the droplet is left uncovered? The edge‐wise evaporation of water from a sample of CNC suspension between glass slide and nematic‐like arrangement at the edge of the ring [14]. The orientation of CNC in the residue deposited cover glass is slow and impractical on a larger scale. What happens if the droplet is left uncovered? Aqueous droplets containing low concentrations of rod-like CNC have been reported to display a on a glass slide by the evaporation of a dilute suspension is readily observed by polarized light Aqueous droplets containing low concentrations of rod‐like CNC have been reported to display a nematic-like arrangement at the edge of the ring [14]. The orientation of CNC in the residue deposited microscopy with 530 nm red plate (Figure 10). The 530 nm red plate generates a red colour for nematic‐like arrangement at the edge of the ring [14]. The orientation of CNC in the residue deposited on a glass slide by the evaporation of a dilute suspension is readily observed by polarized light isotropic material and blue and yellow colours for birefringent material oriented as indicated in the on a glass slide by nm the red evaporation of a dilute suspension is readily observed by polarized light for microscopy with 530 platebecause (Figure The 530 nm redleft plate generates a red colour sketches. The colours are red faint, of 10). the suspension a generates thin film on the glass. The microscopy with 530 nm plate (Figure 10). dilute The 530 nm red plate a red colour for isotropic material and blue and yellow colours for birefringent material oriented as indicated quadrants occupied by the blue and yellow colours indicate that the orientation direction of the in isotropic material and blue and yellow colours for birefringent material oriented as indicated in the the sketches. The colours are faint, because of the dilute suspension left a thin film on the glass. nanocrystals deposited by the evaporating droplet is radial (parallel to the edge of the droplet). sketches. The colours are faint, because of the dilute suspension left a thin film on the glass. The The quadrants occupiedby bythe theblue blueand andyellow yellowcolours colours indicate orientation direction of the quadrants occupied indicate that that the the orientation direction of the nanocrystals deposited by the evaporating droplet is radial (parallel to the edge of the droplet). nanocrystals deposited by the evaporating droplet is radial (parallel to the edge of the droplet).
Figure 10. Polarized light image (crossed polars, 530 nm red wave plate) of dry film left on glass microscope slide by evaporating droplet of dilute (initially 0.52 wt %) aqueous CNC suspension. Two segments of the disc‐shaped residue are shown. The blue and yellow colours observed with the Figure 10. Polarized light image (crossed polars, 530 nm red wave plate) of dry film left on glass Figure 10. Polarized light image (crossed polars, 530 nm red wave plate) of dry film left on glass 530 nm red wave plate [15] indicate that the birefringent CNC are oriented radially, as indicated in microscope slide by evaporating droplet of dilute (initially 0.52 wt %) aqueous CNC suspension. microscope slide by evaporating droplet of dilute (initially 0.52 wt %) aqueous CNC suspension. the lower sketches. Two segments of the disc‐shaped residue are shown. The blue and yellow colours observed with the Two segments of the disc-shaped residue are shown. The blue and yellow colours observed with 530 nm red wave plate [15] indicate that the birefringent CNC are oriented radially, as indicated in Evaporation of suspensions above the critical concentration for liquid crystal formation showed the 530 nm red wave plate [15] indicate that the birefringent CNC are oriented radially, as indicated the lower sketches.
a more dramatic effect. Preliminary observations of the evaporation of droplets starting from CNC in the lower sketches. concentrations high enough to form chiral nematic suspensions showed the formation of films with Evaporation of suspensions above the critical concentration for liquid crystal formation showed 6 a more dramatic effect. Preliminary observations of the evaporation of droplets starting from CNC concentrations high enough to form chiral nematic suspensions showed the formation of films with
7878 6
Materials 2015, 8, page–page
a remarkable 3‐Dimensional iridescent pattern [16]. The contact line at the edge of the droplet was Materials 2015, 8, 7873–7888 pinned, and the CNC in colloidal suspension were transported during evaporation to the outer edge of the droplet by the so‐called “coffee‐stain” effect. Here we for examine the formation evaporation process in Evaporation of suspensions above the critical concentration liquid crystal showed a more dramatic effect. Preliminary observations of the evaporation of droplets starting from CNC more detail, focusing on the textures observed by polarized light microscopy and on the topology of concentrations high enough to form chiral nematic suspensions showed the formation of films with the films observed by profilometry and atomic force microscopy. a remarkable 3-Dimensional iridescent pattern [16]. The contact line at the edge of the droplet was The drying process initially causes a reduction in chiral nematic pitch, if for no other reason that pinned, and the CNC in colloidal suspension were transported during evaporation to the outer edge the concentration of chiral centres increases with increasing concentration. Thus, there are no reports of the droplet by the so-called “coffee-stain” effect. Here we examine the evaporation process in more of the reflection of circularly polarized light from fluid CNC suspensions, but iridescence detail, focusing on the textures observed by polarized light microscopy and on the topology of the is often films observed by profilometry and atomic force microscopy. observed for films made from such suspensions [17–21]. The drying process initially causes a reduction in chiral nematic if for other reason An example of the texture of an iridescent film is shown in pitch, Figure 11. noThe reflection band that the concentration of chiral centres increases with increasing concentration. Thus, there are no colours range from blue to red, as indicated on the insert, although the colours towards the edge of reports of the reflection of circularly polarized light from fluid CNC suspensions, but iridescence is the droplet are less intense due to the thickening and misalignment of the sample due to the coffee‐stain often observed for films made from such suspensions [17–21]. effect [16]. The dark lines are disclination lines at the edge of regions of essentially planar orientation. An example of the texture of an iridescent film is shown in Figure 11. The reflection band colours No fingerprint lines present, because chiral although nematic pitch for towards the sample is in the visible range from blueare to red, as indicated on the the insert, the colours the edge of the droplet are less intense due to the thickening and misalignment of the sample due to the coffee-stain region. Note that a wave plate was not used to generate the main image in Figure 11; the colours effect [16]. The dark lines are disclination lines at the edge of regions of essentially planar orientation. observed in transmission match those of the reflection band, as explained for the much more intense No fingerprint lines are present, because the chiral nematic pitch for the sample is in the visible region. colours shown by the (hydroxypropyl)cellulose/water system (Figure 6). Note that a wave plate was not used to generate the main image in Figure 11; the colours observed The in freely dried droplets another difference in that the transmission match those ofshow the reflection band, as explainedfrom for theedge much evaporation, more intense colours progression of colours from edge to centre is reversed. For covered droplets, evaporation from the shown by the (hydroxypropyl)cellulose/water system (Figure 6). The freely dried droplets show another difference from edge evaporation, in that the progression edge leads to a higher concentration and hence a shorter pitch at the edge, as shown for example in of colours from edge to centre is reversed. For covered droplets, evaporation from the edge leads to Figure 7. For freely dried droplets, as mentioned above, the longer wavelength reflections are near a higher concentration and hence a shorter pitch at the edge, as shown for example in Figure 7. the edge. A similar trend has been observed by Lagerwall et al. [4] and Dumali et al. [22]. However, for For freely dried droplets, as mentioned above, the longer wavelength reflections are near the edge. vacuum‐dried CNC films, the colours were blue at edge and red at centre [20]; evidently the rate of A similar trend has been observed by Lagerwall et al. [4] and Dumali et al. [22]. However, for drying influences the colour sequence. vacuum-dried CNC films, the colours were blue at edge and red at centre [20]; evidently the rate of drying influences the colour sequence. Even where iridescent reflection is not observed, the pitch values, indicated by fingerprint line Even where iridescent reflection is not observed, the pitch values, indicated by fingerprint line spacings (P/2) are much reduced on drying (Figure 12). In solid films, the line spacing is only visible spacings (P/2) are much reduced on drying (Figure 12). In solid films, the line spacing is only visible over small areas of the film, and the spacing often varies with location because the chiral nematic over small areas of the film, and the spacing often varies with location because the chiral nematic axis axis is not in inthe of the the film so that the apparent line isspacing is P/2 larger due to the is not theplane plane of film so that the apparent line spacing larger than due than to the P/2 oblique oblique viewing angle. viewing angle.
Figure 11. Polarized light image (transmission, between crossed polars) of film cast from 5.2 wt % Figure 11. Polarized light image (transmission, between crossed polars) of film cast from 5.2 wt % (SA-5.2%) aqueous CNC suspension on glass. The insert shows a reflection image of a similar droplet (SA‐5.2%) aqueous CNC suspension on glass. The insert shows a reflection image of a similar droplet dried on a black acrylic surface. The observed iridescent region is indicated by the arrow on the insert. dried on a black acrylic surface. The observed iridescent region is indicated by the arrow on the insert.
7879
Figure 11. Polarized light image (transmission, between crossed polars) of film cast from 5.2 wt % (SA‐5.2%) aqueous CNC suspension on glass. The insert shows a reflection image of a similar droplet Materials 2015, 8, 7873–7888 dried on a black acrylic surface. The observed iridescent region is indicated by the arrow on the insert.
Figure 12. 12. Fingerprint Fingerprint textures, polarized light microscope, crossed polars. (a) wt 5.2 % aqueous Figure textures, polarized light microscope, crossed polars. (a) 5.2 % wt aqueous CNC CNC suspension (Batch SA‐5.2%); (b) Film cast from same suspension. suspension (Batch SA-5.2%); (b) Film cast from same suspension.
7 2.2.3.Materials 2015, 8, page–page Evaporation of Water from Droplets of Chiral Nematic Suspensions: Stick-Slip Textures Materials 2015, 8, page–page
With a new batch of CNC, some novel textures were observed with the polarized light 2.2.3. Evaporation of Water from Droplets of Chiral Nematic Suspensions: Stick‐Slip Textures microscope. A distinct banded structure was observed near the edge of the film (Figure 13). At 2.2.3. Evaporation of Water from Droplets of Chiral Nematic Suspensions: Stick‐Slip Textures With a new batch of CNC, some novel textures were observed with the polarized light the very edge of the film (bottom right, next to the red area indicating the isotropic glass support) With a new batch of CNC, some novel textures were observed with the polarized light microscope. A distinct banded structure was observed near the edge of the film (Figure 13). At the a thinvery edge of the film (bottom right, next to the red area indicating the isotropic glass support) a thin blue band shows that the nanocrystals in this region are oriented radially. Moving towards microscope. A distinct banded structure was observed near the edge of the film (Figure 13). At the the centre of the sample distinct radial bands with more complex textures. Atthe higher very edge of the film (bottom right, next to the red area indicating the isotropic glass support) a thin blue band shows that are the several nanocrystals in this region are oriented radially. Moving towards magnification, these radial bands, of the order of 100 µm wide, are themselves seen to be made blue band shows that are the several nanocrystals in radial this region are oriented towards the up centre of the sample distinct bands with more radially. complex Moving textures. At higher centre of the sample are several complex At higher magnification, these radial bands, of the order of 100 μm wide, are themselves seen to be made up of of more or less radially aligned wavydistinct featuresradial aboutbands 2 µmwith widemore extending fortextures. up to several hundred magnification, these radial bands, of the order of 100 μm wide, are themselves seen to be made up of more or less radially aligned wavy features about 2 μm wide extending for up to several hundred micron (Figure 14). Their appearance suggests a distorted fingerprint pattern. Towards the centre micron (Figure 14). Their appearance suggests a distorted fingerprint pattern. Towards the centre of of themore or less radially aligned wavy features about 2 μm wide extending for up to several hundred samples, the texture becomes more uniform, the orientation of the features becomes more micron (Figure 14). Their appearance suggests a distorted fingerprint pattern. Towards the centre of the samples, the texture becomes more uniform, the orientation of becomes the features more random, and the spacing becomes larger. The radial banded structure morebecomes apparent at some the samples, texture becomes larger. more uniform, the orientation of the features becomes more random, and the spacing The radial banded structure becomes more apparent at locations on the film, where intense dark lines appear between the rings (Figure 15). The complex random, and the spacing becomes radial apparent at some locations on the film, where larger. intense The dark lines banded appear structure between becomes the rings more (Figure 15). The nature of these images the generated by the stick-slip evaporation process the shows that the orientation of some locations film, where intense dark lines appear evaporation between rings (Figure complex nature on of these images generated by the stick‐slip process shows 15). that The the the nanocrystals is far from a simple chiral nematic planar or fingerprint texture. complex nature of these images generated by the stick‐slip evaporation process shows that the orientation of the nanocrystals is far from a simple chiral nematic planar or fingerprint texture. orientation of the nanocrystals is far from a simple chiral nematic planar or fingerprint texture.
Figure 13. Polarized light image (crossed polars, 530 nm red wave plate) of film cast from 6 wt %
Figure 13. Polarized light image (crossed polars, 530 nm red wave plate) of film cast from 6 wt % CNC Figure 13. Polarized light image (crossed polars, 530 nm red wave plate) of film cast from 6 wt % CNC suspension (SB‐6%) on glass. suspension (SB-6%) on glass. CNC suspension (SB‐6%) on glass.
Figure 14. Polarized light image (crossed polars, 530 nm red wave plate) of film cast from 6 wt % Figure 14. Polarized light image (crossed polars, 530 nm red wave plate) of film cast from 6 wt % CNC suspension (SB‐6%) on glass showing banded texture at higher magnification. Figure 14. Polarized light image (crossed polars, 530 nm red wave plate) of film cast from 6 wt % CNC CNC suspension (SB‐6%) on glass showing banded texture at higher magnification.
suspension (SB-6%) on glass showing banded texture at higher magnification.
7880
Figure Materials 2015, 14. Polarized light image (crossed polars, 530 nm red wave plate) of film cast from 6 wt % 8, 7873–7888 CNC suspension (SB‐6%) on glass showing banded texture at higher magnification.
Figure 15. Polarized light image (crossed polars, 530 nm red wave plate) of film cast from 6 wt % Figure 15. Polarized light image (crossed polars, 530 nm red wave plate) of film cast from 6 wt % CNC CNC suspension (SB‐6%) on glass showing region with cracks between bands. suspension (SB-6%) on glass showing region with cracks between bands. Materials 2015, 8, page–page
8
2.2.4. Profilometry of Film Topography 2.2.4. Profilometry of Film Topography To interpret these images, profilometer scansscans and atomic force microscopy (AFM) images To try try and and interpret these images, profilometer and atomic force microscopy (AFM) of the surface of the dried droplets were obtained. As a first observation by means of optical images of the surface of the dried droplets were obtained. As a first observation by means of optical profilometry, there appear to be concentric rings of material, with raised edges. However, the rings profilometry, there appear to be concentric rings of material, with raised edges. However, the rings appeared to be discontinuous, and the sample signal was very noisy. It seemed that the oriented and appeared to be discontinuous, and the sample signal was very noisy. It seemed that the oriented and birefringent nature of the CNC film was interfering with the optical interferometric measurements birefringent nature of the CNC film was interfering with the optical interferometric measurements used instrument. ToTo avoid this this problem, samples were coated with gold to provide optically used by by the the instrument. avoid problem, samples were coated with gold to an provide an homogeneous surface, and the samples were scanned with the optical profilometer (Figure 16). In this optically homogeneous surface, and the samples were scanned with the optical profilometer (Figure 16). case a clearer topography emerges, with rings of increasing thickness towards the centre of the film. In this case a clearer topography emerges, with rings of increasing thickness towards the centre of The edges of the rings are often raised, and clearly the films contain radial cracks (coloured blue) that the film. The edges of the rings are often raised, and clearly the films contain radial cracks (coloured show the level of the underlying glass surface. blue) that show the level of the underlying glass surface
Figure 16. Optical profilometry of gold‐coated CNC film. Perspective view showing increasing step Figure 16. Optical profilometry of gold-coated CNC film. Perspective view showing increasing step heights towards the centre of the film. Blue regions indicate cracks in the film. heights towards the centre of the film. Blue regions indicate cracks in the film.
The heights of steps and cracks are shown in the X and Y profiles across the film (Figure 17) are The heights of steps and cracks are shown in the X and Y profiles across the film (Figure 17) are of the order of 2 μm, which is much smaller than the >50 μm height difference previously observed of the order of 2 µm, which is much smaller than the >50 µm height difference previously observed at at the edge of droplets with pinned contact lines [16]. However, the profilometry shows that the the edge of droplets with pinned contact lines [16]. However, the profilometry shows that the outer outer edge of each step is also raised, suggesting that a minor “coffee‐stain effect” is also occurring edge of each step is also raised, suggesting that a minor “coffee-stain effect” is also occurring during during evaporation. Thus a multiple “stick‐slip” situation is occurring in this case, as sketched in evaporation. Thus a multiple “stick-slip” situation is occurring in this case, as sketched in Figure 18. Figure 18. The radial nature of the cracks observed in Figures 15–17 indicates that there exists an The radial nature of the cracks observed in Figures 15–17 indicates that there exists an anisotropy in anisotropy in mechanical properties of the dry films that must reflect the local orientation of the mechanical properties of the dry films that must reflect the local orientation of the CNC. As with any CNC. As with any fibrous solid, the strong direction of films would be expected to be aligned along fibrous solid, the strong direction of films would be expected to be aligned along the direction of the the direction of the rod‐like CNC, with the weakest direction normal to the CNC orientation. The rod-like CNC, with the weakest direction normal to the CNC orientation. The radial alignment of the radial alignment of the CNC thus favors radial cracking during drying. CNC thus favors radial cracking during drying.
7881
during evaporation. Thus a multiple “stick‐slip” situation is occurring in this case, as sketched in Figure 18. The radial nature of the cracks observed in Figures 15–17 indicates that there exists an anisotropy in mechanical properties of the dry films that must reflect the local orientation of the CNC. As with any fibrous solid, the strong direction of films would be expected to be aligned along the direction of the rod‐like CNC, with the weakest direction normal to the CNC orientation. The Materials 2015, 8, 7873–7888 radial alignment of the CNC thus favors radial cracking during drying.
Figure 17. Optical profilometry of gold‐coated CNC film. (a) Topographic image, showing ridges Figure 17. Optical profilometry of gold-coated CNC film. (a) Topographic image, showing ridges (yellow) and cracks (black); (b) Corresponding X profile; (c) Y profile across image. (yellow) and cracks (black); (b) Corresponding X profile; (c) Y profile across image. Materials 2015, 8, page–page
9
Figure 18. Sketch of droplet evaporation under (a) pinned and (b) stick‐slip conditions; (c) Multiple Figure 18. Sketch of droplet evaporation under (a) pinned and (b) stick-slip conditions; (c) Multiple coffee‐stain deposition stages are shown, with rings of increasing thickness and raised edges. coffee-stain deposition stages are shown, with rings of increasing thickness and raised edges.
The contrast between the relatively high ring of CNC deposited at the edge of the evaporating The contrast between the relatively high ring of CNC deposited at the edge of the evaporating droplet for a pinned contact line [16] and the much more diffuse effect noted above suggests that the droplet for a pinned contact line [16] and the much more diffuse effect noted above suggests that tendency for contact line pinning is controlling the texture. Accordingly, we tried to decrease the the tendency for contact line pinning is controlling the texture. Accordingly, we tried to decrease pinning tendency by modifying the suspension with a suitable additive. Small amounts of a the pinning tendency by modifying the suspension with a suitable additive. Small amounts of a water‐soluble compound, propane‐1,2‐diol (propylene glycol, PG), were added to the evaporating water-soluble compound, propane-1,2-diol (propylene glycol, PG), were added to the evaporating droplet. PG has a lower surface tension and vapour pressure than water, and so should become droplet. PG has a lower surface tension and vapour pressure than water, and so should become enriched near the contact line during evaporation. The presence of the PG rich fluid at the contact enriched near the contact line during evaporation. The presence of the PG rich fluid at the contact line line inhibits contact line pinning during evaporation. Figure 19 shows a polarized light image of the inhibits contact line pinning during evaporation. Figure 19 shows a polarized light image of the film film prepared with PG; it shows none of the banding shown by PG‐free suspensions (Figures 16–18), prepared with PG; it shows none of the banding shown by PG-free suspensions (Figures 16–18), and and while dark lines attributed to cracking do occur, they are much closer together than in the case while dark lines attributed to cracking do occur, they are much closer together than in the case films films prepared from PG‐free suspensions. Thus the presence of the PG, presumably as a layer on the prepared from PG-free suspensions. Thus the presence of the PG, presumably as a layer on the surface surface of the glass, decreases the tendency for contact line pinning. Recently, dramatic effects on of the glass, decreases the tendency for contact line pinning. Recently, dramatic effects on contact contact line pinning and droplet mobility have been observed for water‐PG droplets on high‐energy line pinning and droplet mobility have been observed for water-PG droplets on high-energy glass glass surfaces [23]. While the glass surfaces that we used had not been subjected to the rigorous surfaces [23]. While the glass surfaces that we used had not been subjected to the rigorous procedure procedure used by Cira et al., nevertheless addition of low volatility low surface tension components used by Cira et al., nevertheless addition of low volatility low surface tension components such as such as PG to CNC suspensions would be expected to alter contact line pinning and hence the PG to CNC suspensions would be expected to alter contact line pinning and hence the orientational orientational order of the resulting film. order of the resulting film.
7882
surface of the glass, decreases the tendency for contact line pinning. Recently, dramatic effects on contact line pinning and droplet mobility have been observed for water‐PG droplets on high‐energy glass surfaces [23]. While the glass surfaces that we used had not been subjected to the rigorous procedure used by Cira et al., nevertheless addition of low volatility low surface tension components such as 2015, PG 8,to CNC suspensions would be expected to alter contact line pinning and hence the Materials 7873–7888 orientational order of the resulting film.
Figure 19. Polarized light microscope (crossed polars, 530 nm wave plate) of film cast from 8 wt % Figure 19. Polarized light microscope (crossed polars, 530 nm wave plate) of film cast from 8 wt % CNC suspension (SB‐8%) containing 2 wt % added propane‐1,2‐glycol (PG). CNC suspension (SB-8%) containing 2 wt % added propane-1,2-glycol (PG).
2.3. Atomic Force Microscopy of CNC Orientation and Topography in Dry Films 2.3. Atomic Force Microscopy of CNC Orientation and Topography in Dry Films The profilometric measurements reported above demonstrate the large scale deviations from The profilometric measurements reported above demonstrate the large scale deviations from planarity generated by colloidal forces during evaporation from anisotropic suspensions of CNC. planarity generated by colloidal forces during evaporation from anisotropic suspensions of CNC. At much higher magnifications, the orientations of individual nanocrystals in solid films may be At much higher magnifications, the orientations of individual nanocrystals in solid films may be observed. For films cast from dilute CNC suspensions, below the critical concentration for liquid observed. For films cast from dilute CNC suspensions, below the critical concentration for liquid crystal formation, the orientation of the nanocrystals is essential random (Figure 20), despite the fact crystal formation, the orientation of the nanocrystals is essential random (Figure 20), despite the that the suspension must pass through the concentration range where chiral nematic phase is fact that the suspension must pass through the concentration range where chiral nematic phase is expected. The ordering phenomenon is presumably slower than evaporation, or alternatively expected. The ordering phenomenon is presumably slower than evaporation, or alternatively gelation gelation occurs below the critical concentration for liquid crystal formation for this sample. occurs below the critical concentration for liquid crystal formation for this sample. Materials 2015, 8, page–page 10
Figure 20. 20. Atomic of Figure Atomic force force microscope microscope (AFM) (AFM) height height image image of of surface surface of of film film cast cast from from droplet droplet of isotropic (0.0032 wt %) CNC suspension. isotropic (0.0032 wt %) CNC suspension.
The surface orientation of CNC in films cast from suspensions above the critical concentration The surface orientation of CNC in films cast from suspensions above the critical concentration for for chiral nematic phase formation is quite different. Atomic force microscope (AFM) images of a chiral nematic phase formation is quite different. Atomic force microscope (AFM) images of a typical typical area of a film cast from an 8 wt % CNC suspension on a mica substrate are shown in area of a film cast from an 8 wt % CNC suspension on a mica substrate are shown in Figure 21. The Figure 21.The amplitude mode image (a) shows that the CNC are fairly uniformly oriented parallel amplitude mode image (a) shows that the CNC are fairly uniformly oriented parallel to the edge to the edge of the sample. The height mode image (b) shows that the surface is uneven, with ridges of the sample. The height mode image (b) shows that the surface is uneven, with ridges of the of the order of 10 nm high and around 2–3 μm apart, also oriented parallel to the edge of the sample. order of 10 nm high and around 2–3 µm apart, also oriented parallel to the edge of the sample. A A similar periodic relief at the surface of a chiral nematic oligomer has been observed by AFM, and similar periodic relief at the surface of a chiral nematic oligomer has been observed by AFM, and interpreted as relating to the half‐helical pitch [24]. Out‐of‐plane AFM deformations that correspond interpreted as relating to the half-helical pitch [24]. Out-of-plane AFM deformations that correspond to optical observations on dry CNC films have been observed for parabolic focal conic and to optical observations on dry CNC films have been observed for parabolic focal conic and fingerprint fingerprint textures [25]. The spacing of the ridges observed by AFM is of the same order of textures [25]. The spacing of the ridges observed by AFM is of the same order of magnitude as the magnitude as the line spacings of the fingerprint pattern observed by optical microscopy. It is line spacings of the fingerprint pattern observed by optical microscopy. It is reasonable to assume reasonable to assume that the anisotropic orientation of the CNC in the film is the cause. that the anisotropic orientation of the CNC in the film is the cause. A possible explanation is sketched in Figure 22. The orientation of CNC in successive adjacent A possible explanation is sketched in Figure 22. The orientation of CNC in successive adjacent nematic‐like layers of a chiral nematic structure close to an open surface is shown in the upper series nematic-like layers of a chiral nematic structure close to an open surface is shown in the upper series of cross‐sections. The chiral nematic axis is taken to be perpendicular to the page, and the sketches cover one half of the chiral nematic pitch. Assume that the concentration of CNC is sufficient to form a gel. The gel properties will depend on the orientation of the CNC; in particular, gels will shrink 7883 more in directions perpendicular to the CNC and less in directions parallel to the CNC. Thus the reduction in volume will be anisotropic, resulting in a periodicity of P/2 in height along the helicoidal axis, where P is the pitch in the final dry film. This type of anisotropic swelling and
interpreted as relating to the half‐helical pitch [24]. Out‐of‐plane AFM deformations that correspond to optical observations on dry CNC films have been observed for parabolic focal conic and fingerprint textures [25]. The spacing of the ridges observed by AFM is of the same order of magnitude as the line spacings of the fingerprint pattern observed by optical microscopy. It is reasonable to assume that the anisotropic orientation of the CNC in the film is the cause. Materials 2015, 8, 7873–7888 A possible explanation is sketched in Figure 22. The orientation of CNC in successive adjacent nematic‐like layers of a chiral nematic structure close to an open surface is shown in the upper series of cross-sections. The chiral nematic axis is taken to be perpendicular to the page, and the sketches of cross‐sections. The chiral nematic axis is taken to be perpendicular to the page, and the sketches cover one half of the chiral nematic pitch. Assume that the concentration of CNC is sufficient to form cover one half of the chiral nematic pitch. Assume that the concentration of CNC is sufficient to form aa gel. The gel properties will depend on the orientation of the CNC; in particular, gels will shrink gel. The gel properties will depend on the orientation of the CNC; in particular, gels will shrink more in directions perpendicular less in more in directions perpendicular to to the the CNC CNC and and less in directions directions parallel parallel to to the the CNC. CNC. Thus Thus the the reduction in volume will be anisotropic, resulting in a periodicity of P/2 in height along the helicoidal reduction in volume will be anisotropic, resulting in a periodicity of P/2 in height along the axis, whereaxis, P is where the pitch inthe thepitch final in drythe film. This ofThis anisotropic andswelling shrinkage is helicoidal P is final dry type film. type of swelling anisotropic and well-known on a larger length scale; the orientation of the cellulose microfibrils along the secondary shrinkage is well‐known on a larger length scale; the orientation of the cellulose microfibrils along wall of wood pulp fibres causes them to shrink on drying much more laterally than longitudinally, the secondary wall of wood pulp fibres causes them to shrink on drying much more laterally than with important consequences for the bonding and properties of paper sheet [26]. longitudinally, with important consequences for the bonding and properties of paper sheet [26].
Figure 21. AFM (a) amplitude and (b) height images of an area of film cast from droplet of 8 wt % Figure 21. AFM (a) amplitude and (b) height images of an area of film cast from droplet of 8 wt % aqueous CNC suspension (SB‐8%) on a mica substrate. The white arrow (a) indicates the aqueous CNC suspension (SB-8%) on a mica substrate. The white arrow (a) indicates the approximate approximate orientation of the nanocrystals; the black arrows (b) indicate the approximate orientation of the nanocrystals; the black arrows (b) indicate the approximate orientation and spacing orientation and spacing of ridged areas of the sample. The height colour scale in (b) is from −19 to +19 nm. of ridged areas of the sample. The height colour scale in (b) is from ´19 to +19 nm. Materials 2015, 8, page–page
11
Figure 22. 22. (a) The sketches show the orientation orientation of of CNC CNC in sequential sections sections of of aa chiral chiral nematic nematic Figure (a) The sketches show the in sequential helicoid with axis normal to the plane of the page. (In other words, the orientations shown side by helicoid with axis normal to the plane of the page. (In other words, the orientations shown side side should be envisaged as being one behind the other, with the sketch on the left being at the front.) by side should be envisaged as being one behind the other, with the sketch on the left being at the The CNC concentration is assumed high enough for gel formation, and the free surface is at the top front.) The CNC concentration is assumed high enough for gel formation, and the free surface is at of the sketches. On evaporation (b) the tendency of the CNC gel to shrink to a greater extent parallel the top of the sketches. On evaporation (b) the tendency of the CNC gel to shrink to a greater extent to the CNC orientation may lead to a surface ripple with a periodicity of P/2 along the chiral nematic axis. parallel to the CNC orientation may lead to a surface ripple with a periodicity of P/2 along the chiral nematic axis.
This model is clearly oversimplified; the drying process will generate in‐plane as well as out‐of‐plane shrinkage and the chiral nematic axis may well be at an angle to the evaporating surface. This model is clearly oversimplified; the drying process will generate in-plane as well as However, where clear fingerprint patterns are detected optically near a free surface, a corresponding out-of-plane shrinkage and the chiral nematic axis may well be at an angle to the evaporating surface. surface topology may be expected. The effect resembles the theoretical predictions of nanoscale However, where clear fingerprint patterns are detected optically near a free surface, a corresponding undulations and disclination lines for appropriately anchored chiral nematic liquid crystals, driven surface topology may be expected. The effect resembles the theoretical predictions of nanoscale by director surface gradients and elastic constants [27]. However, the surface distortion for CNC undulations and disclination lines for appropriately anchored chiral nematic liquid crystals, driven films is driven by the anisotropic shrinkage of a chiral nematic gel. by director surface gradients and elastic constants [27]. However, the surface distortion for CNC films is driven by the anisotropic shrinkage of a chiral nematic gel. 3. Experimental Section 3.1. CNC Preparation and Purification All the CNC suspensions were prepared 7884 as described previously [28]. Briefly, 40 g of dried Whatman cotton powder was reacted with a preheated sulfuric acid solution (64% w/w, 700 mL) at 45 °C for 45 min, after which the reaction was quenched by 5 L of deionized water (18.2 MΩ∙cm, Millipore Milli‐Q Purification System, EMD Millipore, Billerica, MA, USA) and left to settle for 2 h.
Materials 2015, 8, 7873–7888
3. Experimental Section 3.1. CNC Preparation and Purification All the CNC suspensions were prepared as described previously [28]. Briefly, 40 g of dried Whatman cotton powder was reacted with a preheated sulfuric acid solution (64% w/w, 700 mL) at 45 ˝ C for 45 min, after which the reaction was quenched by 5 L of deionized water (18.2 MΩ¨ cm, Millipore Milli-Q Purification System, EMD Millipore, Billerica, MA, USA) and left to settle for 2 h. The suspension was repeatedly concentrated and washed by centrifugation (6000 rpm, 10 min cycles) until the supernatant appeared turbid, the suspension was then extensively dialysized against deionized water until the effluent remained at neutral pH (Spectra/Por 4, Molecular weight cut-off 12,000–14,000 Daltons). Following the removal of excess acid, the suspensions were dispersed by sonication using a Vibracel sonicator (five 7-min cycles at 60% energy output, Vibracel Sonics and Materials, Newtown, CT, USA). An ice bath was used to prevent overheating. Two batches of CNC, labelled SA and SB were prepared. The main difference was in the ion exchange treatment used to remove excess ion content from the suspensions. Batch SA was stirred over Dowex Marathon MR-3 (Dow Chemical Co., Midland, MI, USA) for 24 h. Batch SB was also stirred over Dowex Marathon MR-3 for 24 h, but then the mixed bed resin was filtered off and replaced by Dowex Marathon C, a strong acid cation exchange resin that had previously been rinsed repeated with ethanol to remove colour. The second resin treatment was to ensure that the resin was in acid form. The suspensions were next filtered through glass microfibre filters (0.45 µm pore size) to remove any particulate impurities introduced by the sonicater tip, and concentrated by means of a Labconco RapidVap vacuum evaporation system. The concentration of the suspensions was determined gravimetrically by drying small volumes in an oven at 105 ˝ C. Sample SA was used as a 5.2 wt % suspension, and sample SB was concentrated to give 6 wt % (SB-6%) and 8 wt % (SB-8%) suspensions. The CNC samples were stored at ´4 ˝ C for at least 2 weeks until the suspensions had separated into isotropic and anisotropic phases. Small amounts of each sample were sealed in glass micro-slides (VitroCom, Mountain Lakes, NJ, USA) and kept standing vertically for 2–3 days before imaging by polarized light microscopy. 3.2. Polarized Light Microscopy Photographs of the liquid crystalline suspensions and the dried films prepared from CNC were taken with a Nikon Eclipse LV100POL polarized light microscope and DS-Fi1 camera (Nikon, Tokyo, Japan). A 530 nm red wave plate was inserted between crossed polars to indicate CNC orientation [15]. Some images of chiral nematic cellulose derivatives were prepared as described in the cited references. 3.3. Fabrication of CNC Films CNC suspensions were drop cast onto glass (for transmitted light microscopy and optical profilometry), black acrylic (for reflected light imaging) or mica surfaces (for atomic force microscopy). The suspensions were allowed to evaporate undisturbed at ambient conditions (room temperature 20–25 ˝ C, relative humidity 20%–60%) for 12 h to give disc-shaped solid films on the glass, acrylic or mica surfaces. 3.4. Optical Profilometry The topography of mm-scale areas of the dry CNC films was investigated by optical profilometry. CNC films deposited on glass substrates were scanned with an optical profilometer (Wyko NT8000, Veeco Instruments Inc., Plainview, NY, USA). The initial results were noisy, but the quality of the topographic images was much improved by coating the surface of the CNC film with
7885
Materials 2015, 8, 7873–7888
a 20 nm thick layer of gold by means of a NexDep Ebeam Evaporator (Angstrom Engineering Inc., Kitchener, ON, Canada). 3.5. Atomic Force Microscopy The topography of the CNC film surfaces at much smaller length scales was investigated by AFM. To prepare very dilute CNC samples on mica for AFM imaging of individual nanocrystals, a drop of poly-L-lysine (0.01 wt % in water) was placed onto a freshly cleaved mica surface (Electron Microscope Supplies, Hatfield, PA, USA), rinsed with water, followed by a drop of dilute CNC suspension (~0.005 wt %). The samples were left to dry in Petri dishes under ambient conditions. For images of CNC orientation in films cast from chiral nematic suspensions, a droplet of 8 wt % suspension was allowed to evaporate on the mica surface. A Multi-Mode-3 AFM (Digital Instruments, Bruker, Billerica, MA, USA) was used in tapping mode to image the samples, with probes obtained from NanoScience (Si, N-type, 0.01–0.025 Ω/cm, 200–400 kHz resonance frequency, with nominal tip radius of 10 nm). 4. Conclusions Simple evaporation of a droplet of aqueous CNC suspension leads to a surprising range of self-organizing phenomena: 1. Starting with a dilute suspension of CNC, well below the critical concentration for liquid crystal formation, the CNC tend to align radially at the edge of the droplet. 2. For more concentrated suspensions that are initially above the critical concentration, some radial initial orientation of CNCs at the edge of the film is observed. This is in line with recent observations of the orientation of CNCs in shear flow, where the acid-form CNCs self-oriented across the flow direction [29]. 3. Drying droplets of concentrated CNC display a topography governed by evaporation-driven mass transfer (the “coffee-stain effect”). 4. The magnitude of the coffee-stain effect is greatest for strongly pinned droplets, where the height near the edge of the dry film is several times the height in the middle of the film [16]. 5. When pinning of the contact line is weaker, a “stick-slip” situation is observed, with layers of increasing thickness are laid down by the receding contact line. The layers give a complex optical texture; radial cracking of the drying film was sometimes observed. 6. The surface topography of dry films at short length scales showed a radial orientation of the CNC at the free surface of the film, along with a radial height variation with a period of approximately P/2, ascribed to the anisotropic shrinkage of the chiral nematic structure. 7. The rich liquid crystalline textures displayed by the evaporation of water from CNC suspensions depend on the interplay of concentration gradients that influence chiral nematic pitch, director orientation and gelation with the effects of surface orientation and the elastic constants normally associated with liquid crystalline phases. The qualitative results here are very preliminary, but hopefully indicate some of the variables involved in the formation of chiral nematic solids from cellulose nanocrystal suspensions. Acknowledgments: Funding from the Natural Sciences and Engineering Research Council Canada, and support from Fonds Québecois de la Recherche sur la Nature et les Technologies through the Center for Self-Assembled Chemical Structures are gratefully acknowledged. We thank M. Ramkaran for assistance with the AFM imaging. Author Contributions: X.M. prepared the CNC suspensions and films, took the photomicrographs and contributed to the manuscript. D.G.G. supervised the work and wrote the manuscript. Conflicts of Interest: The authors declare no conflict of interest.
7886
Materials 2015, 8, 7873–7888
References 1. 2. 3. 4.
5. 6. 7. 8. 9.
10. 11. 12. 13.
14. 15. 16. 17. 18. 19. 20. 21. 22.
23.
Werbowyj, R.S.; Gray, D.G. Liquid crystalline structure in aqueous hydroxypropyl cellulose solutions. Mol. Cryst. Liq. Cryst. Lett. 1976, 34, 97–103. [CrossRef] Nishio, Y.; Sato, J.; Sugimura, K. Liquid crystals of cellulosics: Fascinating ordered structures for the design of functional material systems. Adv. Polym. Sci. 2015. [CrossRef] Revol, J.F.; Bradford, H.; Giasson, J.; Marchessault, R.H.; Gray, D.G. Helicoidal self-ordering of cellulose microfibrils in aqueous suspension. Int. J. Biol. Macromol. 1992, 14, 170–172. [CrossRef] Lagerwall, J.P.F.; Schütz, C.; Salajkova, M.; Noh, J.H.; Park, J.H.; Scalia, G.; Bergström, L. Cellulose nanocrystal-based materials: From Liquid Crystal Self-Assembly and Glass Formation to Multifunctional Thin Films. Available online: http://www.nature.com/am/journal/v6/n1/abs/am201369a.html (accessed on 10 January 2014). Habibi, Y.; Lucia, L.A.; Rojas, O.J. Cellulose nanocrystals: Chemistry, self-assembly and applications. Chem. Rev. 2010, 110, 3479–3500. [CrossRef] [PubMed] Klemm, D.; Kramer, F.; Moritz, S.; Lindström, T.; Ankerfors, M.; Gray, D.; Dorris, A. Nanocelluloses: A new family of nature-based materials. Angew. Chem. Int. Ed. 2011, 50, 5438–5466. [CrossRef] [PubMed] Moon, R.J.; Martini, A.; Nairn, J.; Simonsen, J.; Youngblood, J. Cellulose nanomaterials review: Structure, properties and nanocomposites. Chem. Soc. Rev. 2011, 40, 3941–3994. [CrossRef] [PubMed] Chandrasekhar, S. Liquid Crystals, 2nd ed.; Cambridge University Press: Cambridge, UK, 1992; pp. 213–299. Revol, J.F.; Godbout, L.; Dong, X.M.; Gray, D.G.; Chanzy, H.; Maret, G. Chiral nematic suspensions of cellulose crystallites; phase separation and magnetic field orientation. Liq. Cryst. 1994, 16, 127–134. [CrossRef] Dong, X.M.; Kimura, T.; Revol, J.F.; Gray, D.G. Effects of ionic strength on the phase separation of suspensions of cellulose crystallites. Langmuir 1996, 12, 2076–2082. [CrossRef] Chen, W.; Gray, D.G. Interfacial tension between isotropic and anisotropic phases of a suspension of rod-like particles. Langmuir 2002, 18, 633–637. [CrossRef] Mu, X.; Gray, D.G. Formation of chiral nematic films from cellulose nanocrystal suspensions is a two-stage process. Langmuir 2014, 30, 9256–9260. [CrossRef] [PubMed] Hoeger, I.; Rojas, O.J.; Efimenko, K.; Velev, O.D.; Kelley, S.S. Ultrathin film coatings of aligned cellulose nanocrystals from a convective-shear assembly system and their surface mechanical properties. Soft Matter 2011, 7, 1957–1967. [CrossRef] Uetani, K.; Yano, H. Self-organizing capacity of nanocelluloses via droplet evaporation. Soft Matter 2013, 9, 3396–3401. [CrossRef] Introduction to Polarized Light Microscopy. Available online: https://www.microscopyu.com/articles/ polarized/polarizedintro.html (accessed on 2 November 2015). Mu, X.; Gray, D.G. Droplets of cellulose nanocrystal suspensions on drying give iridescent 3-D “coffee-stain” rings. Cellulose 2015, 22, 1103–1107. [CrossRef] Revol, J.-F.; Godbout, L.; Gray, D.G. Solid films of cellulose with chiral nematic order and optically variable properties. J. Pulp Paper Sci. 1998, 24, 146–149. Beck, S.; Bouchard, J.; Berry, R. Controlling the reflection wavelength of iridescent solid films of nanocrystalline cellulose. Biomacromolecules 2011, 12, 167–172. [CrossRef] [PubMed] Beck, S.; Bouchard, J.; Chauve, G.; Berry, R. Controlled production of patterns in iridescent solid films. Cellulose 2013, 20, 1401–1411. [CrossRef] Tang, H.; Guo, B.; Jiang, H.; Xue, L.; Li, B.; Cao, X.; Zhang, Q.; Li, P. Fabrication and characterization of nanocrystalline cellulose films prepared under vacuum conditions. Cellulose 2013, 20, 2667–2674. [CrossRef] Dumanli, A.G.; van der Kooij, H.M.; Kamita, G.; Reisner, E.; Baumberg, J.J.; Steiner, U.; Vignolini, S. Digital color in cellulose nanocrystal films. ACS Appl. Mater. Interfaces 2014, 6, 12302–12306. [CrossRef] [PubMed] Dumanli, A.G.; Kamita, G.; Landman, J.; van der Kooij, H.M.; Glover, B.J.; Baumberg, J.J.; Steiner, U.; Vignolini, S. Controlled, bio-inspired self-assembly of cellulose-based chiral reflectors. Adv. Opt. Mater. 2014, 2, 646–650. [CrossRef] [PubMed] Cira, N.J.; Benusiglio, A.; Prakash, M. Vapour-mediated sensing and motility in two-component droplets. Nature 2015, 519, 446–450. [CrossRef] [PubMed]
7887
Materials 2015, 8, 7873–7888
24. 25. 26. 27. 28. 29.
Meister, R.; Dumoulin, H.; Halle, M.A.; Pieranski, P. The anchoring of a cholesteric liquid crystal at the free surface. J. Phys. II 1996, 6, 827–844. [CrossRef] Roman, M.; Gray, D.G. Parabolic focal conics in self-assembled solid films of cellulose nanocrystals. Langmuir 2005, 21, 5555–5561. [CrossRef] [PubMed] Page, D.H.; Tydeman, P.A. Consolidation of the Paper Web; Bolam, F.M., Ed.; British Paper and Board Makers Association: London, UK, 1965; pp. 371–396. Rofouie, P.; Pasini, D.; Rey, A.D. Nano-scale surface wrinkling in chiral liquid crystals and plant-based plywoods. Soft Matter 2015, 11, 1127–1139. [CrossRef] [PubMed] Kloser, E.; Gray, D.G. Surface PEG-grafting of cellulose nanocrystals in aqueous media. Langmuir 2010, 26, 13450–13456. [CrossRef] [PubMed] Tatsumi, M.; Teramoto, Y.; Nishio, Y. Different orientation patterns of cellulose nanocrystal films prepared from aqueous suspensions by shearing under evaporation. Cellulose 2015, 22, 2983–2992. [CrossRef] © 2015 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons by Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).
7888