LETTERS
Photonic-crystal full-colour displays ANDRE´ C. ARSENAULT1,2*, DANIEL P. PUZZO1,2, IAN MANNERS3* AND GEOFFREY A. OZIN1* 1
Department of Chemistry, University of Toronto, 80 St George Street, Toronto M5S 3H6, Canada Opalux Incorporated, 80 St George Street, Toronto M5S 3H6, Canada 3 School of Chemistry, University of Bristol, Cantock’s Close, Bristol BS8 1TS, UK *e-mail:
[email protected];
[email protected];
[email protected] 2
Published online: 1 August 2007; doi:10.1038/nphoton.2007.140
In our information-rich world, it is becoming increasingly important to develop technologies capable of displaying dynamic and changeable data, for reasons ranging from valueadded advertising to environmental sustainability. There is an intense drive at the moment towards paper-like displays, devices having a high reflectivity and contrast to provide viewability in a variety of environments, particularly in sunlight where emissive or backlit devices perform very poorly. The list of possible technologies is extensive, including electrophoretic, cholesteric liquid crystalline, electrochromic, electrodewetting, interferometric and more. Despite tremendous advances, the key drawback of all these existing display options relates to colour. As soon as an RGB (red, green and blue) colour filter or spatially modulated colour scheme is implemented, substantial light losses are inevitable even if the intrinsic reflectivity of the material is very good. We describe a reflective flat-panel display technology based on the electrical actuation of photonic crystals. These materials display non-bleachable structural colour, reflecting narrow bands of wavelengths tuned throughout the entire visible spectrum by expansion and contraction of the photonic-crystal lattice. The material is inherently bright in high-light environments, has electrical bistability, low operational voltage, can be integrated onto flexible substrates, and is unique among all display technologies in that a continuous range of colours can be accessed without the need for colour filters or optical elements. Photonic crystals (PCs)1,2, materials with a periodic modulation in refractive index, can be sources of exceptionally bright and brilliant reflected colours arising from coherent Bragg optical diffraction3. Exemplified by gemstone opals, threedimensional PCs are readily available by means of colloidal selfassembly, making them a fertile test-bed for investigating concepts based on tunable structural colour. Synthetic preparations for silica or polymer microspheres with low dispersity (,2%) are well-developed, and their self-assembly into a close-packed ordered structure leaves a void volume of 26% available for further material infiltration or modification4,5. Tuning of colloidal PC optical properties has been effected by the infilling of metals, insulators, semiconductors and polymers of all types, and by the inversion of such constructs through removal of the template spheres. Given their bright colours with potential tunability, it seems obvious at first glance to use PCs as the active materials in refreshable full-colour digital displays. The fact that this has not yet been achieved highlights the difficult set of requirements that must be met by any potential candidate, including electrical
tunability, access to thin and homogeneous oriented films, large tuning range, relatively rapid response time, mechanical and cycling stability, low voltage/current requirements, and obviously the ability to implement the material into a feasible, practical, scalable and manufacturable device. In this report, we describe a material, process and device architecture meeting all such requirements, based on the electrical actuation of a silica – metallopolymer composite PC in a thin-film electrochemical cell. The design of our tunable PC display is based on a twocomponent responsive composite, which we have previously dubbed Photonic Ink (P-Ink)6. Our initial report described the solvent swelling and response to chemical redox agents of earlygeneration P-Ink composites, but in this report we demonstrate the first example of a dynamic electrical tunability, a working display device architecture, as well as a scalable and environmentally tolerant fabrication procedure. The first component in our P-Ink system is an inactive structural scaffold made of an ordered array of silica microspheres, and the second comprises a crosslinked network of polyferrocenylsilane (PFS), an iron-based metallopolymer. The crosslinked PFS network component actively mediates the lattice spacing of the silica spheres through an electrochemically driven swelling and shrinking process. Thin films of monodisperse silica spheres made by sol –gel polymerization7 were deposited on indium tin oxide (ITO) coated glass plates by convective self-assembly8. The voids between the silica spheres were then infiltrated by a lowmolecular-weight PFS with pendant unsaturated C5 5C bonds, to which was added a small amount of a multifunctional thiol and radical photoinitiator. Ultraviolet irradiation generates chemical crosslinks between the polymer chains through a well-known thiol-ene reaction9, leading to an infinite metallopolymer network matrix. As would be expected from a polymer-network-based PC composite10–15, P-Ink films show colour changes based on solvent swelling. Solvent influx into the network causes a lattice increase, and is manifested as a redshift of the Bragg diffraction peak. The increase in lattice constant on swelling is shown schematically in Fig. 1a and can be directly visualized by the SEM images shown in Fig. 1b. The degree of swelling is influenced both by the swelling solvent as well as the degree of crosslinking, which determines the ultimate range of tunability16. The effect of crosslink density can be clearly seen when the composites are swollen with toluene (Fig. 1c). Electrical actuation of a P-Ink film was enabled by its incorporation into a sealed thin-layer electrochemical cell. A schematic of the cell is shown in Fig. 2a, and consists of the
468
nature photonics | VOL 1 | AUGUST 2007 | www.nature.com/naturephotonics
© 2007 Nature Publishing Group
LETTERS
Infiltration
Actuation
Decreasing crosslink density
Figure 1 Fabrication and mechanochromic behaviour of thin-film electroactive PCs. a, Fabrication scheme and schematic of tuning mechanism. b, SEM images of an as-infiltrated (left) and swollen (right) sample. c, Optical pictures showing areas of thin-film PFS – silica composites and their crosslinker-dependent solvent swelling and colour-tuning behaviour.
15
Glass superstrate
10
Electrolyte
PC composite
Current ( A)
ITO electrode
5 0 –5
ITO electrode
–10
Glass substrate
–15 0.0
0.4
0.8
1.2
Potential (V)
Figure 2 Constructing a PC electrochemical cell. a, Schematic of the two-electrode electrochemical cell with relevant components. The P-Ink film is approximately 3 – 5 mm in thickness, and the cell gap can range from tens to hundreds of micrometres. b, A cyclic voltammogram of the silica – PFS composite using a Ag/AgCl reference electrode, showing the characteristic two redox peaks of PFS. c, Image of an assembled multipixel electrochemical cell with attached electrode leads. d,e, The same cell after oxidizing the middle pixel with 1.5 V (d) and 2.0 V (e). The reflected colours are due to Bragg diffraction from the composite PC. nature photonics | VOL 1 | AUGUST 2007 | www.nature.com/naturephotonics
© 2007 Nature Publishing Group
469
LETTERS 60
50 620 600
50
580
max
(nm)
40
560
40 Reflectance (%)
Reflectance (%)
540
30
520 500 0
20
1,000 2,000 Applied potential (mV)
3,000
30
20 10
10
0
0 450
500
550 600 Wavelength (nm)
650
400
700
500 Wavelength (nm)
600
680 660 Bragg peak maximum (nm)
640 620 600 580 560 540 520 500 0
20
40
60
80
100
Number of cycle 230 660 220 640
210 Maximum Bragg shift (nm)
Bragg maximum (nm)
620 600 580 560 540
200 190 180 170 160
520
150
500
140 130
480 0
20
40 Time (min)
60
0.0
80
0.5
1.0 1.5 Crosslinker content (mol%)
2.0
2.5
Figure 3 Electromechanochromic behaviour of silica – PFS PC thin films. a, Reflection spectra taken at increasing applied voltages. The inset shows the dependence of the peak position on the applied voltage, clearly showing a continuum of accessible colours. b, Spectra of a sample showing reflection of all three primary colours upon incremental oxidation. c, The Bragg peak maximum versus number of cycles for a series of 100 oxidation – reduction cycles. The voltage was cycled between 2.0 V and – 2.0 V, holding the sample at a particular voltage for approximately 15 s to ensure equilibrium. d, Electrical bistability measurement. The sample was oxidized at 2.0 V, after which the leads were disconnected and the peak position monitored as a function of time. e, Dependence of the maximum tuning range on the crosslinker content of the composite.
P-Ink composite supported on ITO –glass as the working electrode, a hot-melt ionomer spacer (Dupont), and an ITO– glass counterelectrode. The cell is filled with an organic solvent-based liquid electrolyte by vacuum filling, and sealed with epoxy. It is
known that PFS polymers in solution as well as in supported films display reversible electrochemical oxidation and reduction, with the partial electronic delocalization along the polymer backbone leading to a continuously tunable degree of
470
nature photonics | VOL 1 | AUGUST 2007 | www.nature.com/naturephotonics
© 2007 Nature Publishing Group
LETTERS
50 µm
Figure 4 Images of P-Ink pixel arrays. a – c, Pictures of a P-Ink film coated onto an array of 0.3-mm ITO lines. Every other line was oxidized to various degrees to generate coloured patterns. a, Alternating dark green and red stripes made by applying a voltage to every other line in the array. b, The lines have been shifted into the near infrared by applying a higher voltage and are thus transparent, allowing the black background behind the sample to be visualized. c, By applying different voltages to individual lines, multicolour lines are also possible. d, Image of an oxidized single line, showing a vivid green on pale violet. e, The line shown in d, visualized in an optical microscope, demonstrating a 25-mm edge resolution adequate for display applications.
oxidation17. The cyclic voltammogram of this material (measured in a three-electrode configuration; Fig. 2b) shows a profile characteristic of such main-chain ferrocene-based polymers, with two wide oxidation waves in the forward scan. The first peak corresponds to oxidation of every other iron atom along the chain, the second to full oxidation, but in reality these transitions are broad and overlapping owing to partial charge delocalization along the polymer chain, as well as a potential buildup within the polymer network on increasing oxidation. Once the completed cell is connected to a potentiostat, it becomes an electrically actuated electrochemical cell, as shown in Fig. 2c. When an oxidative potential is applied to the P-Ink composite, electrons are drawn out of the iron atoms in the PFS backbone while anions from the electrolyte (surrounded by a solvent shell) are driven into the polymer to neutralize its positive charge buildup. The influx of both ions and solvent into the polymer causes it to swell and push apart the layers of spheres, redshifting the reflected optical diffraction peak (Fig. 2d, e). Applying a reducing potential runs the system in reverse, with electrons docking onto the PFS backbone and the anions being expelled into the electrolyte. As reported in a previous study18,
the expansion in the PC lattice occurs mainly perpendicular to the substrate owing to covalent anchoring of the composite. By virtue of their continuously tunable state of oxidation, P-Ink films display voltage-dependent continuous shifts in reflected colours. The graph in Fig. 3a illustrates how the Bragg peak gradually shifts through a given range when incrementally varying the applied voltage. Although the reflectivity of these samples is modest, its magnitude can be increased by increasing the thickness of the active P-Ink layer. The position and magnitude of the accessible tuning range depends both on the starting sphere size and on the degree of crosslinking (Fig. 3e), and to a lesser degree on the polymer molecular weight. By appropriately choosing the starting sphere size and crosslinker content, a single sample can be tuned to all three primary colours (Fig. 3b). Once a particular voltage has been applied, the peak shift is completed within 1–2 s, but data were collected after application of the voltage for 2 min to ensure that equilibrium had been reached. It is important to note that the reflected wavelength can also be switched into the near-infrared region, rendering the materials transparent and providing a black or white state depending on the background.
nature photonics | VOL 1 | AUGUST 2007 | www.nature.com/naturephotonics
© 2007 Nature Publishing Group
471
LETTERS The cycling stability of the materials was found to be good, with a series of 100 oxidation–reduction cycles, as shown in Fig. 3c. As can be clearly seen, the peak maximum shifts reversibly, with its position returning to within nanometres after each voltage application. The electrical bistability of this material was evaluated by applying an initial voltage pulse to redshift the diffraction peak, then disconnecting the electrical leads and measuring the drift of the peak in time. As shown in Fig. 3d, the P-Ink material has a high intrinsic bistability, with the peak position stable over an approximately 2-h period. To demonstrate the potential of electroactive PCs in full-colour displays, multi-pixel devices were built and tested. This was achieved by depositing composite PC films onto arrays of 0.3-mm-wide ITO strips on glass. In this way, each strip could be individually addressed by connecting it in a circuit to the counterelectrode. Optical images of devices in operation are shown in Fig. 4. Figure 4a shows a P-Ink sample with alternating dark green and red stripes, made by applying a voltage to every other line in a 0.3-mm ITO line array. When a higher voltage is applied to each of the activated lines, their colour could be shifted out of the visible wavelength range (Fig. 4b), making them transparent. This transparency allows the black backing to be seen, and provides a dark state. Multicolour lines are also possible (Fig. 4c), achieved by applying different voltages to individual lines. The colour tuning range can be shifted to lower wavelengths by using smaller spheres, with Fig. 4d showing a bright green oxidized line bracketed by pale violet lines on either side (shown also in an optical microscope image in Fig. 4e). It can be readily appreciated that, although the film on top of the electrode array is continuous, only the area directly over the contacted electrode is oxidized or reduced when a voltage is applied. This is extremely important for future manufacturing, because pixel arrays can be created without the need for patterning and segmenting the active layer. The inherent flexibility of our silica –polymeric composite PCs makes it possible to integrate them into flexible display components. In addition to rigid devices, we also constructed flexible devices by using ITO-coated PET sheets as working electrodes and counterelectrodes. Although these flexible devices are less developed, we were able to obtain similar colour shifts as with devices assembled from rigid ITO– glass substrates. Electrically tunable PCs, as described in this report, can clearly provide electronic displays with unique and attractive characteristics. In contrast to any other display technology, these materials can reflect colours having a wavelength that is voltage tunable throughout the visible range, without the need for expensive colour filters or sub-pixellation of red –green – blue elements, thereby wasting much less of the accessible light output. However, to achieve a full-colour display, a number of additional milestones must be met. Extensive stability testing is still needed, including extended cycling and a battery of environmental tests. Another important consideration will be how to properly deal with the angle dependence of a PC pixel. Our preliminary investigation has shown that this problem can probably be adequately corrected through a combination of optical layers (as used in LCD screens) and controlled internal disorder (as used in cholesteric LCDs). In addition, to achieve full colour, the hue, lightness and saturation of each pixel must be controlled, and it is still too early to identify the optimal way
of tuning these characteristics in a P-Ink display. Hue selection is apparently straightforward with the P-Ink system as it is voltage dependent. To manipulate the saturation and lightness, however, a colour-mixing technique such as lateral integration (used in Qualcomm’s iMoD displays) or vertical integration (used in cholesteric LCDs) may be necessary. These properties, coupled with an inherent bistability and low operating voltage, offer power requirements compatible with a number of applications, such as outdoor and indoor signage, portable electronics, architectural panels, active camouflage and full-colour electronic paper. As PC theory is scalable throughout the whole electromagnetic range, P-Ink coatings can also be readily integrated into devices modulating UV, visible or infrared light reflection and transmission. Received 24 November 2006; accepted 22 June 2007; published 1 August 2007. References 1. Yablonovitch, E. Inhibited spontaneous emission in solid-state physics and electronics. Phys. Rev. Lett. 58, 2059– 2062 (1987). 2. John, S. Strong localization of photons in certain disordered dielectric superlattices. Rev. Lett. 58, 2486–2489 (1987). 3. Schroden, R. C., Al-Daous, M., Blanford, C. F. & Stein, A. Optical properties of inverse opal photonic crystals. Chem. Mater. 14, 3305 –3315 (2002). 4. Lopez, C. Materials aspects of photonic crystals. Adv. Mater. 15, 1679– 1704 (2003). 5. Arsenault, A. et al. Towards the synthetic all-optical computer: Science fiction or reality? J. Mater. Chem. 14, 781 –794 (2004). 6. Arsenault, A. C., Miguez, H., Kitaev, V., Ozin, G. A. & Manners, I. A polychromic, fast response metallopolymer gel photonic crystal with solvent redox tunability: A step towards photonic ink (P-Ink). Adv. Mater. 15, 503 –507 (2003). 7. Stober, W., Fink, A. & Bohn, E. Controlled growth of monodisperse silica spheres in the micron size range. J. Colloid Interface Sci. 26, 62–69 (1968). 8. Jiang, P., Bertone, J. F., Hwang, K. S. & Colvin, V. L. Single-crystal colloidal multilayers of controlled thickness. Chem. Mater. 11, 2132–2140 (1999). 9. Hoyle, C. E., Lee, T. Y. & Roper, T. Thiol-enes: Chemistry of the past with promise for the future. J. Polym. Sci. Part A – Polymer Chem. 42, 5301–5338 (2004). 10. Asher, S. A., Holtz, J., Liu, L. & Wu Z. J. Self-assembly motif for creating submicron periodic materials. Polymerized crystalline colloidal arrays. J. Am. Chem. Soc. 116, 4997–4998 (1994). 11. Holtz, J. H., Holtz, J. S. W., Munro, C. H. & Asher, S. A. Intelligent polymerized crystalline colloidal arrays: Novel chemical sensor materials. Anal. Chem. 70, 780 –791 (1998). 12. Saito, H., Takeoka, Y. & Watanabe, M. Simple and precision design of porous gel as a visible indicator for ionic species and concentration. Chem. Commun. 2126– 2127 (2003). 13. Jethmalani, J. M. & Ford, W. T. Diffraction of visible light by ordered monodisperse silica – poly(methyl acrylate) composite films. Chem. Mater. 8, 2138–2146 (1996). 14. Fudouzi, H. & Xia, Y. N. Photonic papers and inks: Color writing with colorless materials. Adv. Mater. 15, 892– 896 (2003). 15. Jiang, P., Smith, D. W., Ballato, J. M. & Foulger, S. H. Multicolor pattern generation in photonic bandgap composites. Adv. Mater. 17, 179 –184 (2005). 16. Takeoka, Y. & Watanabe, M. Tuning structural color changes of porous thermosensitive gels through quantitative adjustment of the cross-linker in pre-gel solutions. Langmuir 19, 9104 –9106 (2003). 17. Rulkens, R. et al. Linear oligo(ferrocenyldimethylsilanes) with between two and nine ferrocene units: Electrochemical and structural models for poly(ferrocenylsilane) high polymers. J. Am. Chem. Soc. 118, 12683–12695 (1996). 18. Arsenault, A. C. et al. Vapor swellable colloidal photonic crystals with pressure tunability. J. Mater. Chem. 15, 133– 138 (2005).
Acknowledgements A.C.A. would like to thank the Natural Sciences and Engineering Research Council of Canada (NSERC) for a CGS scholarship. I.M. thanks the European Union for a Marie Curie Chair, the Royal Society for a Wolfson Research Merit Award (2005), and the Canadian Government for a Canada Research Chair. G.A.O. thanks the Government of Canada for a Research Chair in Materials Chemistry. The authors would like to thank G. Masson and D. Rider for help with polymer synthesis, and V. Kitaev for providing monodisperse microspheres. The authors would like to thank Materials and Manufacturing Ontario (MMO), NSERC, the Government of Canada and the University of Toronto for generous funding.
Author contributions A.C.A. and G.A.O. conceived the experiments; D.P.P. and A.C.A. performed the experiments; A.C.A., D.P.P., G.A.O. and I.M. analysed the experimental results; A.C.A. wrote the manuscript.
Competing financial interests The authors declare competing financial interests: details accompany the full-text HTML version of the paper at www.nature.com/naturephotonics. Reprints and permission information is available online at http://npg.nature.com/reprintsandpermissions/
472
nature photonics | VOL 1 | AUGUST 2007 | www.nature.com/naturephotonics
© 2007 Nature Publishing Group