Subscriber access provided by University of Sussex Library
Article
Development of biphasic formulations for use in electrowettingbased liquid lenses with a high refractive index difference Matthias S. Ober, Daniel Dermody, Mathieu Maillard, Franck Amiot, Géraldine Malet, Benjamin Burger, Caroline Woelfle-Gupta, and Bruno Berge ACS Comb. Sci., Just Accepted Manuscript • DOI: 10.1021/acscombsci.8b00042 • Publication Date (Web): 16 Jul 2018 Downloaded from http://pubs.acs.org on July 24, 2018
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Combinatorial Science
Development of biphasic formulations for use in electrowettingbased liquid lenses with a high refractive index difference Matthias S. Ober,*†,I Daniel Dermody,† Mathieu Maillard,‡,II Franck Amiot,‡,III Géraldine Malet,‡ Benjamin Burger, ‡ Caroline Woelfle-Gupta† and Bruno Berge‡,IV †
Core R&D, The Dow Chemical Company, Midland, Michigan 48674, United States Corning Technology Center – Lyon; formerly Varioptic S.A., Bâtiment Tony Garnier, 24 rue Jean Baldassini, 69007 Lyon, France.
‡
Supporting Information Placeholder
ABSTRACT: Commercial electrowetting-based liquid lenses are optical devices containing two immiscible liquids as an optical medium. The first phase is a droplet of a high refractive index oil phase placed in a ring-shaped chassis. The second phase is electrically conductive and has a similar density over a wide temperature range. Droplet curvature and refractive index difference of two liquids determine the optical strength of the lens. Liquid lenses take advantage of the electrowetting effect, which induces a change of the interface’s curvature by applying a voltage, thereby providing a variable focal that is useful in autofocus applications. The first generation of lens modules were highly reliable, but the optical strength and application scope was limited by a low refractive index difference between the oil and conductive phase. Described herein is an effort to increase the refractive index difference between both phases, while maintaining other critical application characteristics of the liquids, including a low freezing point, viscosity, phase miscibility and turbidity after thermal shock. An important challenge was the requirement that both phases have to have matching densities and hence had to be optimized simultaneously. Using high throughput experimentation in conjunction with statistical design of experiments (DOE), we have developed a series of empirical models to predict multiple physicochemical properties of both phases and derived ideal locations within the formulation space. This approach enabled the development of reliable liquid lenses with a previously unavailable refractive index difference of ΔnD of ≥0.290, which enabled true optical zooming capability.
INTRODUCTION Electrowetting-based liquid lenses, initially developed by the startup Varioptic S.A., now availabe under the brand name Corning® Varioptic® Lenses,1-3 cover a wide range of applications in consumer and industrial optical systems. The liquid lenses provide access to high resolution optics with autofocus functionalities without the need for mechanical moving parts. They are used in many applications including barcode readers, medical imagery, or biometry. Liquid lenses are composed of two immiscible liquids encapsulated into a sealed cell.3 One of these liquids is a hydrophilic conductive phase and the other one is a hydrophobic oil. The interface between both liquids is spherical and forms a diopter due to refractive index difference. When applying voltage to the system, the liquid/liquid meniscus’ curvature changes due to electrowetting,4-8 and with it, the focal distance of the lens. The liquid lens technology offers many advantages over traditional technologies: low power consumption, high speed, shock and vibration resistance, resistance to wear, and high optical quality. Critical to the development of
reliable and functional lenses is the formulation development of liquid/liquid pairs that are suitable to meet application requirements.3 One of the key parameters of the liquid pair is its refractive index difference, as it directly determines the optical power range.9 Whereas low refractive index differences (0.080 < ΔnD < 0.120) are sufficient for autofocus applications, high ΔnD values are needed to implement zooming capabilities.3 Unfortunately, interface refractive index difference between two liquids can be modified over a much more limited range than the step between a solid and surrounding air. For instance, feasible lowest refractive index liquids are in the range n=1.3-1.4 whereas the upper limit is in the range of n = 1.7-1.8, but these high refractive liquids have significant issues regarding chemical stability and formulation performance. Hence pushing refractive index difference toward high values of above ΔnD ≥ 0.290 drastically reduces the size of the formulation space (Figure 1) that also meets other application requirements. A particularly important requirement is the need that both phases have a minimum density difference to prevent gravity
ACS Paragon Plus Environment
ACS Combinatorial Science
from deforming the liquid-liquid interface. In addition, low viscosity and interfacial tension are advantageous for high speed and short response times.10 Low wettability of the water phase on the substrate is necessary to reduce lens hysteresis.11 Cross miscibility, resistance to temperature shocks and low melting points of the two phases have to be minimized over a wide temperature range to ensure functionality at different environmental conditions. Many of these requirements are antagonistic. The main challenge is to find the best compromise between all these parameters while increasing the refractive index difference. We approached this problem in three stages. The goal of the first stage was to identify formulation candidate compounds and develop a high-resolution empirical model of the refractive index, melting point and density of the formulation space of the aqueous conductive phase. An important discovery here was the inclusion of salts of fluorinated acids, which allowed minimization of its refractive index. During the second stage, we evaluated new candidate compounds for the organic phase by evaluating their performance in binary biphasic systems in combination with common candidate compounds for the aqueous or hydrophilic phase. In the final stage we developed comprehensive statistical models of properties of oil phases and oil phase formulations paired with a corresponding ideal water phase as derived from the water phase model obtained during the first stage. These final stage models enabled us to identify optimal ranges within the formulation space and guided further commercial development. On a larger scenery, this example illustrates how high throughput experimentation in combination with statistical design of experiment can solve a complex multi-parameter industrial challenge. Refractive Index histogram
Number of candidates
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
100 80 60 40 20 0
Refractive index
Figure 1. Refractive index distribution taken from the Varioptic S.A. optical fluid database. A refractive index difference ΔnD ≥ 0.290 be obtained only with a limited number of compounds.
EXPERIMENTAL TECHNIQUES Materials Materials were purchased from Sigma-Aldrich (Sigma-Aldrich Corporation, St. Louis, Missouri, now owned by Merck KGaA) or Gelest (Gelest Inc., Morrisville, Pennsylvania) and used without further purification. Oligo(diphenylethers), in-
Page 2 of 14
cluding Santovac MCS-293 and Santolight SL-5267 were purchased from SantoVac Fluids (now SantoLubes LLC, Missouri, US).
General Methods Refractive index The refractive index measurements were performed on an Atago refractometer RX-7000α. The temperature of the sample was held at 20 °C and sample size was approximately 0.2 mL. For pure liquids, refractive indices were re-measured in-house and compared with supplier data for calibration purposes. Miscibility Miscibility of two-component biphasic liquid systems at variable temperatures was measured as follows. First, the refractive indices nD,pure of the pure hydrophobic and corresponding hydrophilic phases were determined at the temperature of interest as described. Next, the phases were combined and equilibrated by agitating the mixture overnight at the prescribed temperature. The refractive indices nD,eq of the top and bottom phase were remeasured at an identical temperature. The volume fraction of Phase 1 in Phase 2 as a function of temperature was estimated by using the equation Vphase 1 in phase 2 T 2, 2, , pure , eq Vphase 2, eq, total 2, 2, , pure , pure assuming an additive behavior of the refractive indices (AragoBiot approximation).12 Miscibility of multi-component biphasic liquid systems was estimated by treating each phase as a single component (having the refractive index of the multi-component phase). Density Densities were measured with an Anton Paar DMA-500 density meter at 25 °C by injecting 1 mL of sample into the measurement cell. Melting Point Melting points were determined by DSC measurements in a single measurement cycle. A 5 mg sample, encapsulated into a hermitic aluminum DSC pan, was equilibrated for 1 min at 25 °C. Next, the sample was cooled to −90 °C with a temperature gradient of −10 K/min, equilibrated at −90 °C for 1 min and finally heated to 45 °C with a temperature gradient of 10 K/min. Generally, the maximum of melting peak endotherms, mp, rather the onset of melting mo, was used as a melting point. If several phase transitions or melting points were observed, the one with the highest temperature was used as the data point for model fitting. Turbidity The turbidity of the oil phase / conductive phase system after thermal shock was determined by high-throughput nephelometry measurements. Turbidity originates from partial miscibility from liquids at high temperature, after a thermal cycling, condensation occurs in oversaturated liquids, forming light scattering dispersion of immiscible liquids.13 Turbidity is characterized by two main parameters, the maximum turbidity related to the amount and size of liquid dispersion, and the recovery time necessary to retrieve liquids within transparency specifications. First, 150 µL of the hydrophilic and 100 µL of the hydrophobic phase were pipetted into a 96-well optical glass plate that was sealed with a corresponding 96-well silicone lid. The plate was then heated to 85 °C (compound selection stage) or 70 °C
ACS Paragon Plus Environment
Page 3 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Combinatorial Science
(formulation DOE) for 20 min in a convection oven. After temperature equilibration, the plate was removed from the oven, the lid was removed and the hot plate was immediately placed into an Ascent Nepheloskan nephelometer (Thermo Labsystems). The turbidity of the oil/conductive phase binary systems of the plate was measured at 5 min intervals with an integration time of 1000 ms/well. Proprietary formulations with a known behavior, as well as water blanks were used as internal standards. At least two, usually three independent measurements of each sample were performed and the turbidity/time curves were averaged among these measurements. Viscosity Viscosities were measured at variable temperatures on a Brookfield CAP 2000+ instrument with the appropriate spindle for the expected viscosity range. High-throughput measurements of viscosities were performed by using the Dow TADM Viscosity experiment,14 utilizing a Hamilton Microlab® Star dispensing robot and measuring the pressure drop during aspiration. All high-throughput viscosity measurements were performed at room temperature. Solubility (semi-quantitative) Experiments were conducted combinatorically using a 96 well plate containing individual 1.5 mL vials. A small amount of any of the solid candidate compound series (~5 mg) was combined with a series of liquids of interest. The vials were sealed and the plate was shaken over night at room temperature. The next day, the vials were visually inspected for solubility of the solid. High Throughput Powder Dispensing Solid components of high throughput experiments were dispensed using an Autodose Powdernium (Symyx) solids handler. The robotic handler was used in a Many to Many configuration with on-deck weighing. 25 mL or 50 mL hoppers were used as sources and the powders were dispensed into 1 mL vials. Design of Experiments, surface modeling and data visualization Constrained D-optimal mixture designs were calculated with JMP,15 which was also used to perform multivariate regression and model development with the acquired data points. JMP was further utilized to search for optimal areas in the formulation space, sorting and clustering of data, generation of bivariate plots and phase diagrams. Numeric data, as received from turbidity or miscibility screens, was arranged into a matrix (ASCII format) and plotted utilizing GnuPlot.16 Library composition diagrams were generated with Freeslate Library Studio.17
RESULTS AND DISCUSSION Formulation development Biphasic formulations consist of a low refractive index conductive hydrophilic phase and a high refractive index, non-conductive hydrophobic phase. The formulation of these phases for liquid lens applications have to meet a variety of physicochemical requirements. The most important constraints are (1) low cross miscibility ( wNaTFA wwater > 3 wNaBr
Additional constraints
wwater + wEG + wTMG + wNaBr + wNaTFA = 1
All samples were measured by DSC, melting points were determined within a range of −90 ° C, and 25 °C (Figure s3, SI). As expected, not all of the parameters were statistically significant in the model, and we removed parameters with the highest P-Values until R2adj did not increase any further and was within proximity of R2. Data was fitted the following model equation optimized to exclude statistically insignificant coefficients:
1.4660
1.4310
Potassium acetatec)
1.4527
NaTFA
1,3-Propanediolb)
1.4393
Sodium triflatec)
1.3679
EG,water
1.3721
NaBr,NaTFA
Density model An empirical density model for the multi-component conductive phase had been previously developed by Varioptic SA, and
0.8
0.3
Ethylene glycol
index of distilled water available in-house against Cargille reference standard, b) Remeasured in-house, c) Determined by regression as described in text.
High boundary
Water
Sodium bromidec)
a) Refractive
Low boundary
Ethylene glycol
1.33273
Sodium trifluoroacetate
Weight fraction (wi)
Compound
b)
c)
Page 4 of 14
conductive phase
EG,water NaTFA
EG
NaTFA
TMG
EG
EG
TMG
water
water,NaTFA NaBr
EG water
NaTFA
water
EG
water
TMG,NaTFA water
NaBr
EG,NaTFA wEG
NaBr
NaTFA
NaTFA TMG
NaTFA
NaTFA,water
NaTFA
water
Eq. 4
Coefficients ai, bi,j and ci,j and a regression summary are listed in the SI (Table s8) Combined model By combining the density, refractive index and melting point models of the conductive phase, constraining the combined model for a maximal allowable freezing point (e.g. −25 °C) and iteratively solving for minimal achievable refractive index, it is possible to predict an ideal conductive phase formulation and
ACS Paragon Plus Environment
Page 5 of 14
its associated refractive index over a wide density range (Figure 2). Notably, it is not possible to obtain a refractive index of less than 1.37 at a melting point of below -25 °C for conductive phases that do not utilize NaTFA as one of the anti-freeze components. Formulations that contain any amount of NaBr are not ideal concerning the investigated properties, as they never have the lowest possible refractive index within the investigated space. Formulations containing a TMG component are only advantageous, if very low densities (between 1.03 and 1.15) are targeted. 1.4
index of a conductive phase that does not freeze at −25°C is approximately 1.35; the intersection point of the −25 °C melting point isoline intersects with the x-axis (0% EG, a solution of ~43% (w/w) NaTFA in water).
Lowest attainable RI for formulations with mp
