Electron Microscopy of Polymer Cholesteric Liquid Crystal Flakes







Cathy Fromen

University of Rochester, Rochester, NY 14627
Department of Chemical Engineering


 Final Project, Optics 307

Spring 2009 


1.       Abstract

2.                   Liquid Crystals

3.                   PCLC Flakes

Experimental Procedure

1.                  Methods and Materials


Results and Discussion

1.                     Neat Undoped PCLCs

2.                    Two Layer PCLC+PEDOT

3.                    Uniformly Doped PCLC


1.                  Conclusion

2.                 Acknowledgements

3.                 References



1. Abstract

Polymer Cholesteric Liquid Crystals (PCLCs) exhibit advantageous optical and electrical properties, making them useful for display technologies.  PCLCs reflect wavelengths proportional to the cholesteric pitch length and also exhibit motion when suspended in a host fluid: translation due to electrophoresis and reorientation due to Maxwell-Wagner polarization.  Recent studies have shown that internal flake doping or flake layering can reduced the electric field required for such motion.1-2  Uniformly doped PCLC flakes were created by adding 10% carbon black (wt%) to PCLCs prior to alignment and layered PCLC flakes were created by spin-coating a thin film of PCLC with a conductive layer of Poly(3,4-ethylenedioxythiophene), also known as PEDOT.  In this project, PCLC flake doping and layering was investigated using electron microscopy.  Electron micrographs of undoped PCLC flakes were obtained and a distinct fingerprint texture was found on the surface.  Spin-coated layers of PEDOT on PCLC flakes were found to have a similar fingerprint structure, and the PEDOT layers were found to be approximately 1um thick.  Uniformly doped carbon black flakes did not exhibit the fingerprint structure, but a less organized wrinkled sheet texture was found.  This was determined to be due to the distribution of carbon black conglomerates, which vary in size from a few nanometers to as large as 500nms. 

2. Liquid Crystals

The term “liquid crystal” refers to phase of matter, also called a mesophase, with structural properties between those of crystals and liquids.  Liquid crystals (LCs) have long-range orientation order, as in a crystalline structure, but the molecules themselves are often anisotropic, resulting in a degree of order found between the two phases.3  There are two different classes of liquid crystals: lyotropic and thermotropic.  Lyotropic materials are found in solution and enter the mesogen phase at a particular concentration.  Thermotropic materials enter the liquid crystal phase between a certain temperature range, beginning at the crystalline melting point.  Thermotropic liquid crystals are then divided into different categories based on their internal order.  These categories include the nematic, cholesteric and smectic phases.  The nematic is the simplest and possesses long-range orientational order defined by a director n, but is completely anisotropic in other directions.  The smectic phase is more complex, where the structure is not only defined by long-range orientational order, but the presence of internal layers results in weak translational order as well.  The particular liquid crystals discussed in this project fall into the chiral or cholesteric phase.  In this phase, the director n, which is the unit vector describing the average direction of orientation along the long axis of the structure, rotates 360˚ throughout the material to form a helix.

This helix can be described by a pitch length, P, which is the length required for a complete 360˚ rotation of n.  As a result of the internal alignment of the liquid crystals, the material exhibits a selective reflection effect.  The PCLCs in this project align in a left-handed helix; due to the selective reflection effect left-handed light will be completely transmitted through the material, while right-handed light at a specific wavelength of λ0 will be reflected.  λ0 is the wavelength of selective reflection and is equal to the refractive index of the material,  nav­, multiplied by P, the pitch length.4


This project expands on previous work done with polymer cholesteric liquid crystals (PCLCs).  As the name implies, PCLCs are thermotropic materials that can exist in the cholesteric phase.  PCLCs differ from typical cholesteric liquid crystals due to the large size of the molecules themselves, which are in fact polymers.  As a result, the order obtained in the LC phase can be maintained below the melting temperature because the large polymer molecules are “frozen” into place.  Thus, PCLCs can exist with optical and electrical liquid crystal properties at a temperature below the melting point.  The specific PCLCs in this project are noncrosslinkable cyclic polysiloxanes substituted with mesogenic groups which are connected to the backbone by aliphatic spacers.  The colors are supplied by the manufacturer and they differ in ratios of chiral to nonchiral side chains, thereby changing the pitch length.  As a result, PCLC technology has the ability to result in selective reflection across the visible light spectrum.


3. PCLC Flakes

In order to utilize these optical properties on a micro scale, PCLC “flakes” were made.  Shaped PCLC flakes were made by casting a thin film of PCLC material into a mold.  Using soft lithography, a mold was made of polydimethylsiloxane (PDMS) with 60µm x 20µm wells.  Solid PCLC material was heated into the LC phase, occurring above 50˚C, and was then aligned into a thin film onto using a shearing force.  This thin film of PCLC was spread across the PDMS mold, producing aligned PCLCs in the individual wells.  The material was then cooled below its melting point, where the LC alignment was maintained.  The PCLC flakes are then removed from the mold by laminating the thin film onto a glass substrate, from which the flakes can be easily removed and suspended in a host fluid.5


The specific electrical properties of these PCLC flakes have been studied in detail in previous projects.1  The PCLC flakes are known to exhibit various types of motion, including reorientation and  translation.  In reorientation, the PCLC flake rotates 90˚ along their long axis to align with the applied electric field.  As a result of the PCLC flake’s dimensions, reorientation changes the PCLC flake from a position where it reflects light back to the viewer, to a position where the reflected light cannot be seen by the viewer.  Reorientation occurs from Maxwell-Wagner polarization in an AC field, which describes the induced dipole resulting from an electric field’s effects on the interface of two materials with dissimilar dielectric properties.  The second type of motion exhibited by PCLC flakes is translation, occurring in a DC field.  Translation is driven by electrophoresis; due to interfacial charging, the electric field creates a double layer around the now charged PCLC flake and this fluid motion causes the flake to move.  Both types of motion, translation and reorientation, have possible applications for display technology.


PCLC flake technology has the potential to make a huge impact in the reflective particle display industry.  This industry is expected to make huge strides in the near future, with products such as the Amazon Kindle already a commercial success.  However, there are many obstacles still facing current reflective particle displays.  In particular, current particle display technologies lack the ability for a full color display.  E-Ink, used in the Kindle, only has potential for a two-color display without the application of color filters.  Due to the selective reflection effect, PCLC flake technology can offer brilliant colors over the full visible spectrum.  However, PCLC flake technology also has major obstacles to overcome: higher power requirements, 50% reflected light loss and microencapsulation challenges. Recent advances have occurred in microencapsulation of PCLC flakes, offering potential to apply the technology to a flexible application, as well as advances in flake doping to further lower the power requirements for flake motion. 


In order to increase the conductivity of individual flakes, methods of flake doping has been explored.  Various types of carbon black was dispersed in the PCLC material before aligning in the LC phase, and it was found that uniform doping can increase the conductivity of the PCLC flakes. 1  This allows a dramatic drop in translation and reorientation times.  Flake layering has also resulted in increased flake conductivities.  Recently procedures have been developed for creating two layer PCLC flakes, using combinations of different colored PCLCs, with a conductive polymer, PEDOT.  These novel doping methods can increase the viability of the technology by decreasing the required energy input.


The purpose of this project is to utilize electron microscopy to explore further PCLC technology and the effects of flake doping and layering.  The self-assembly of the cholesteric liquid crystals results in interesting flake surface features, which were able to be resolved by both the SEM and the TEM.  The effects on this liquid crystalline structure due to the addition of dopant or a conductive layer were also studied.  In addition, the distribution of the carbon black dopant in doped PCLC flakes was observed, as well as resulting thickness of PEDOT layers on the flakes.



Experimental Procedure


1. Methods and Materials

Three types of flake samples were observed in this project: neat undoped PCLC flakes, 10% carbon black doped flakes and two layered PCLCs with a PEDOT layer.  Shaped flakes were prepared in a PDMS mold and were laminated to a microscope slide.  They were then removed from the slide and attached to an SEM pin using carbon tape, or applied directly to a lacey carbon TEM grid.  Due to the insulating properties of the PCLC material, the SEM pins were then coated with gold.  Coating is a crucial step in observing the PCLC flakes; it provides a conductive path for the electrons to follow to ground, which eliminates effects of charging, it increases the emission of secondary electrons, the prime focus in this project, and it increases the stability of the polymer sample, which might otherwise deteriorate in the beam.  A gold sputter-coater found in room 216 in Wilmont Hall was used for 30 seconds at 15mamps, resulting in a coating of approximately 3nm.  Also viewed in this project was a thin film of carbon black.  Carbon black was dissolved in methylene chloride, the same solved used to dissolve the solid PCLC material prior to alignment, and was then deposited directly to an SEM pin.  The instruments used in this project were the Zeiss-Supra 40VP SEM and the FEI Tecnai F20 TEM found in room 216 in Wilmont Hall at the University of Rochester.


Results and Discussion


1.     Neat Undoped PCLC Flakes

PCLC flakes have traditionally been viewed using light microscopy.  The selective reflection effect results in PCLCs appearing a brilliant color.  Below are two images of neat undoped PCLC flakes of various colors and shapes.  Rectangular flakes are 20x60μms. 


PCLC flakes can be easily viewed using traditional light microscopy: image on the left shows PCLC flakes in a PDMS mold, while image on the left shows different varieties of shaped flakes covering a range of colors, with the background image showing an empty PDMS mold. 


While light microscopy can accurately capture the selective reflection, electron microscopy is required to resolve much smaller features of the liquid crystalline structure.  In this project, the secondary electron detectors were used to resolve surface features of the PCLC flakes.  Both the SE2 detector and the InLens detector were able to resolve surface features of the PCLC flake.  These micrographs show a Fingerprint Texture, indicated by dark lines outlining a spiral structure. 4  These “wormy” lines are spaced apart approximately half  the pitch length of the material.  These particular flakes exhibit selective reflection in the green, which would result in a theoretical pitch length around 360nm. When measured using the ImageJ software6, these bands are approximately 200nm apart, as compared to the expected theoretical value of 180nm.  These surface features are best resolved using the InLens detector at short working distances, which maximizes the secondary electrons returned back towards the beam.

SE2 Micrographs:


SE2 micrographs of neat undoped PCLC flakes of increasing magnification

InLens Micrographs: 



InLens micrographs of neat undoped PCLC flakes of increasing magnification.  Surface fingerprint texture of PCLCs found yielding “wormy” lines separated by approximately 200nm


2. Two Layered PCLC Flakes with PEDOT Layer

Two layered flakes have also been viewed using light microscopy.  However, PEDOT exhibits high transmission in the visible light range, resulting in a completely clear polymer coating.  Light microscopy is unable to resolve any specific features of the PEDOT layer, and images look identical to neat undoped PCLC flakes.

Light Microscopy image of PEDOT coated PCLC flakes

The secondary electron detectors were used to examine the effect of layering PCLC flakes with PEDOT.  Again the surface features of the flakes were resolved using primarily the InLens detector.  These micrographs show the same Fingerprint texture as seen in the undoped PCLC flakes, even through the PEDOT layer.  However, unlike in the undoped PCLC flakes, the spiral structures are raised by a few nanometers and appear only close to the sides of the flakes.  Towards the center of the flakes, the flow pattern of the PEDOT becomes very apparent, resulting in a much less coherent spiral structure, until the spiral structure is unrecognizable at all. 



InLens micrographs of the surface features of two layer PEDOT coated flakes.  Micrograph on the left shows PEDOT coating with surface features resolved.  These surface features are magnified in the following two images, yielding the traditional fingerprint texture, as well as the raised fingerprint and flow patters on the far right.


Micrographs of the PEDOT coated PCLC flakes were also used to characterize the PEDOT layer, which ranges from 400nm to 1.5μm in thickness.6


InLens micrographs of the side of two PEDOT coated PCLC flakes: PEDOT layer found on top side of both flakes, ranging from 400-700nm in thickness


3. Uniformly Doped PCLC Flakes with 10% Carbon Black

Uniformly doped PCLC flakes with 10% carbon black were also viewed first using light microscopy.  While similar in appearance to undoped PCLC flakes, the addition of the carbon black yields regions within the flake that absorb light and appear black to the viewer at a high magnification.  Recent studies and tests on flake conductivity have brought into question the uniformity of the carbon black doping, and one of the main goals of this project was to study the distribution of carbon black in the flakes.  From the following image, it is apparent that light microscopy could not accomplish this goal alone.

10%cb solution 

Uniformly doped PCLC flakes containing 10% Carbon black (wt%).  Flakes are 20x60μm in dimension.

In order to study the distribution of the carbon black within the PCLC flakes, many different techniques were employed.  First, it was hoped that due to the increase in flake conductivity, carbon black doped flakes could be viewed without the gold surface coating.  However, charging occurred quite rapidly.  Uncoated flakes were then viewed using the variable pressure option in the SEM.  Leaking air into the tank and using the VPSE detector dramatically improves the quality of the images obtained, but lowers the ability for high resolution, as apparent in the following images.

Variable Pressure Micrographs


Electron Micrographs of 10% carbon black doped PCLC flakes viewed using variable pressure of 30Pa


As a result of the poor resolution, all future samples were coated with gold to avoid charging. 


To assist in the search for carbon black within the PCLC flakes, a film of carbon black was deposited directly to a SEM pin, allowing for characterization of carbon black structures outside of the PCLC film.  This film yielded interesting results; the micrographs reveal that carbon black tends to conglomerate into a few distinct dimensions.  The smallest particle size able to be resolved yielded carbon black conglomerates on the order of 10-50nm.  However, these small conglomerates were then easily found in larger structures ranging from 150nm to a few microns.  These micrographs were important in identifying carbon black conglomerations in the following flake samples.


Image of Carbon Black Film



InLens micrographs of carbon black film.  Top two micrographs show large carbon black conglomerations, while bottom two micrographs show organization into 100-200nm spherical structures.

Following analysis of the carbon black film, the surface features of the doped carbon black flakes were then viewed using the SEM.  Immediately, a noticeably different surface structure as compared to the neat undoped PCLC flakes becomes apparently.  In the neat undoped PCLC flakes, the surface yielded the distinct fingerprint pattern.  With the 10% doped PCLC flakes, there is no defined spiral structure, but rather a wavy lined pattern similar to a wrinkled sheet.  It appears as though the fingerprint pattern was disrupted by the addition of the carbon black, forming less organized surface structures, and presumably less internal order as well.  The wrinkled sheet pattern still contains the “wormy” lines mentioned before, with spacing still approximately 200nms.  However, the loss of the tight fingerprint spiral is indicative of a change in organization.  Also of note, the wrinkled sheet seems to exits on both the top and bottom surfaces of the flakes and on the sides as well.  Surfaces with damage from the sheering process yield the wrinkled sheet as frequently as surfaces lacking the sheer damage, and no surfaces were viewed without the wrinkled sheet pattern.

Secondary Electron Micrographs



InLens electron micrographs of 10% carbon black doped PCLC flakes.  The wrinkled sheet texture is apparent in all six samples, showing a wide variety of flake angles.


While these secondary electron images only reveal information about surface topology, much can be inferred about the internal structure as well.  As mentioned, one of the main goals of this project was to explore the uniformity of the carbon black doping within the PCLC flake.  Recent conductivity tests had shown discrepancies between aligned films of PCLCs doped with carbon black versus an unaligned mixture of the two components.  It was thought that the sheering processes separated the PCLC and the carbon black, isolating the carbon black from the surface of the flake and altering the conductivity of the flake.  However, the electron micrographs obtained for this project suggest this is not the case.  Taking the top right micrograph from the previous section, large conglomerations of carbon black can be located directly on the surface of the flake.  These conglomerations are circled below, shown alongside the original image.  Unfortunately, from only the secondary electron micrographs it is impossible to determine completely if these conglomerations are carbon black or merely PCLC impurities.  However, while impurities in the PCLC structure were seen in the neat undoped PCLC flakes, found to be typically less than 100nms (see above), the impurities here are generally larger, typically around 500nms.  Also, their strategic placement seems to correspond to alterations in the wrinkled sheet pattern and are most likely responsible for the absence of the fingerprint texture.  Both the x-ray detector and the backscattered electron detector were employed to distinguish any difference between these impurities and the remaining PCLC material.  It was hoped that the polysiloxanes making up the PCLCs would result in different x-ray or backscattered electron signals compared to the carbon black.  However, the PCLCs are also comprised of mesogen liquid crystal groups in which carbon is the main constituent.  As a result, neither detector was able to distinguish any additional information regarding these impurities.


InLens electron micrographs of wrinkled sheet texture in 10% carbon black doped PCLC flakes.  On left is unaltered micrograph, on right impurities determined to be carbon black have been circled


Finally, TEM micrographs of the uniformly doped PCLC flakes were obtained.  Flakes were placed onto a TEM carbon grid, and the side regions of the flakes were imaged.  This was done because the center regions of the flakes were too thick to allow for transmission, but variations in the side morphology allowed for imaging of thinner regions of the flakes.  These TEM images support conclusions drawn previously by the SEM analysis.  Carbon black conglomerates of various sizes were found distributed throughout the samples, varying from as large as 200μm, to a few hundred nanometers, to tens of nanometers and possibly smaller.

TEM Images

Created by Digital Micrograph, Gatan Inc.             Created by Digital Micrograph, Gatan Inc.             Created by Digital Micrograph, Gatan Inc.

TEM images of 10% carbon black doped PCLC flakes.  Carbon black conglomerates of decreasing size shown from left to right.



Conclusions and Acknowledgements


1. Conclusion

In this project, the surface features and liquid crystalline order was explored using electron microscopy.  Neat PCLC flakes were characterized and a distinct fingerprint texture was revealed.  A slight variation of this fingerprint texture was found in the layered PEDOT coated PCLC flakes, but was completely absent in the uniformly doped 10% carbon black flakes.  Looking into this structural various, it was concluded that the distribution of carbon black conglomerates throughout the sample was responsible for the alteration.  Further work with higher concentrations of carbon black doping might yield more definite results.


2. Acknowledgements

Thanks to Jerry Cox, Ken Marshall, Dr. Stephen Jacobs and the Laboratory for Laser Energetics for introducing me to PCLC, the loves of my academic life.


Special thanks to Brian McIntyre for all of his patience, assistance, excellent ideas, support and teaching prowess; without his help none of this project would ever have been accomplished.


And thanks to MF, the HTML Master.


3. References

1) Kosc, T. Z., Motion of Polymer Cholesteric Liquid Crystal Flakes in an Electric Field. PhD Thesis, University of Rochester, Rochester, New York, 2003.


2) G. Cox and C. Fromen, PEDOT flake layering, internal report, 8-2008.


3) Hamley, I.W. Introduction to Soft Matter. West Sussex: Wiley & Sons, 2007.


4) S. Jacobs, Optics and Liquid Crystals for Chemical Engineers, class notes CHE 447/MSC 434, Spring 2009.


5)  G. Cox, Microencapsulation effect – standard cell type, internal report, 6-3-2008.


6) ImageJ software, version 1.41 for Windows, downloaded from http://rsbweb.nih.gov/ij/download.html on March 23, 2009.


Cathy Fromen, April 2009



Please enter any comments, criticisms, questions, etc. below.

Your name:

Email address: