Electrically controlled localized charge trapping at amorphous fluoropolymer-electrolyte interfaces

Charge trapping is a long-standing problem in electrowetting-on-dielectric (EWOD), causing reliability reduction and restricting its practical applications. Although this phenomenon has been investigated macroscopically, the microscopic investigations are still lacking. In this work, the trapped charges are proven to be localized at three-phase contact line region by using three detecting methods -- local contact angle measurements, electrowetting (EW) probe, and Kelvin Probe Force Microscopy (KPFM). Moreover, we demonstrate that this EW-induced charge trapping phenomenon can be utilized as a simple and low-cost method to deposit charges on fluoropolymer surfaces. Charge density near the three-phase contact line up to 0.46 mC/m2 and the line width with deposited charges ranging from 20 to 300 micrometer are achieved by the proposed method. Particularly, negative charge densities do not degrade even after harsh testing with a water droplet on top of the sample surfaces for 12 hours, as well as after being treated by water vapor for 3 hours. These findings provide an approach for applications which desire stable and controllable surface charges.

ABSTRACT: Charge trapping is a long-standing problem in electrowetting-on-dielectric (EWOD), causing reliability reduction and restricting its practical applications. Although this phenomenon has been investigated macroscopically, the microscopic investigations are still lacking. In this work, the trapped charges are proven to be localized at three-phase contact line region by using three detecting methods --local contact angle measurements, electrowetting (EW) probe, and Kelvin Probe Force Microscopy (KPFM). Moreover, we demonstrate that this EWinduced charge trapping phenomenon can be utilized as a simple and low-cost method to deposit charges on fluoropolymer surfaces. Charge density near the three-phase contact line up to 0.46 mC/m 2 and the line width with deposited charges ranging from 20 to 300 µm are achieved by the proposed method. Particularly, negative charge densities do not degrade even after "harsh" testing with a water droplet on top of the sample surfaces for 12 hours, as well as after being treated by water vapor for 3 hours. These findings provide an approach for applications which desire stable and controllable surface charges.

INTRODUCTION
Amorphous fluoropolymers (AFPs) such as Teflon AF and Cytop are popular materials for various applications 1-7 because of the unique combination of favorable material properties, such as chemical inertness, mechanical strength, water repellency, dielectric strength, optical transparency, and easy solution processability [7][8] . For these reasons, AFPs are also predominantly used as insulating and hydrophobic layer in electrowetting (EW) devices [8][9][10][11][12][13][14] . EW, which is often also denoted as 'electrowetting-on-dielectric' (EWOD) to emphasize the relevance of the dielectric layer, relies on the fact that ionic charge carriers have in general a rather low affinity to the weakly polarizable AFPs. However, at the same time, fluoropolymers have been used for decades as charge storage (electret) materials with applications in electro-mechanical transductions, such as microphones, micro-electro-mechanical systems (MEMS) and electric generators [15][16][17][18][19] . These applications rely on the fact that charges, once deposited on or within AFPs, remain stable due to the wide electronic band gap and deep energetic traps. The purpose of the present work is to shed light on these two antagonistic aspects of charge repellence and charge storage in AFPs that jointly control the injection of charge carriers into AFPs in EWOD at high voltage. Such EW-induced charge injection, if done in a controllable way, will offer interesting opportunities for generating permanent charge patterns on AFPs.
The reliability of any EW applications in microfluidics [20][21] , optofluidics [22][23] , display technology [24][25] , and energy harvesting 26 relies on the reproducibility, performance, and durability of the dielectric layer; thus the stability of AFPs is particularly important [27][28][29][30] . Charge trapping induces the degradation of the electrical response of AFP films, leading to contact angle saturation and failures in EWOD devices 26,[31][32] . Early experiments with composite dielectrics displayed a reversible response and symmetric saturation for positive and negative bias voltage, suggesting substantial mobility of both types of charge carriers upon injection into the AFP films 33 . More recent studies displayed a strongly asymmetric and sometimes irreversible response 34,35 . Other investigations identified a relationship of macroscopic charge injection and/or contact angle saturation with molecular scale properties of the system, demonstrating better EW stability for bulkier salt ions 29 (e.g. surfactants, ionic liquids) and for bulk fluid molecules 36 (e.g. glycols).
More recently, it has been also reported that Teflon AF materials got permanent negative surface charges upon extended (several hours) contact with water in the absence of any applied voltage 37 .
Since most studies on EW-induced charging phenomena mainly focus on the response of the macroscopic contact angles and the total electrical currents, there is a significant lack of quantitative description of the underlying microscopic charge trapping phenomenon. Moreover, a clear understanding of the correlation between the charge trapping and the macroscopic wetting characteristics has remained elusive. It has been recognized that diverging electric fields in the vicinity of the three phase contact line (TPCL) cause various types of non-linear response of the materials during EW, which may limit the minimum contact angle [38][39] . Nevertheless, it is still not clear whether the heterogeneity of the electric fields leads to the charge trapping and induces permanent change in the local surfaces of AFPs. It was assumed that the charge injection process essentially follow the distribution of the electric field with its well-established divergence near the TPCL [40][41] . However, several recent studies have revisited the topic of charge injection in EWOD using local surface potential measurements with non-contact electrostatic probes 32,[42][43] .
Surprisingly, the measured surface potential distributions were reported to be rather broad with a maximum in the center of the droplet, thereby challenging the classical view based on the local field divergence 32,[42][43] .
In this work, we analyze the charge distribution generated on AFP surfaces by EW at high voltage with unprecedented lateral resolution, and explore the usage of the EW-induced charge injection for localized charge storage at AFP surfaces. Three complementary techniques have been used to reveal the local charge distribution on single layer AFP surfaces. We demonstrate that EWinduced charge injection is highly localized. Based on this, a simple and low-cost approach is proposed and validated to generate stable charge patterns with controllable length scale and density. As a result, we can tune the surface properties of AFP surfaces at microscale level without complex microfabrication processes and the related instruments. The excellent stability of the negative trapping charges, in particular in a "harsh" environment under water or high humidity, suggests its potential for a wide range of applications requiring stable surface charges 44-48 .

Preparation of Teflon films
ITO/glass substrates were cleaned in a Liquid Crystal Display (LCD) cleaning line for G2.5 glass (400 mm × 500 mm). Subsequently, 800 nm thick AFP films were prepared by screen printing Teflon AF 1600 solution (The Chemours Company, USA), followed by baking on a hot plate at 95 °C for 1 min to remove residual solvent and additional baking in an oven at 185 °C for 30 min to anneal the film. All processes were carried out in a clean room. More details on the fabrication process can be found in Ref. 49 .

Surface charging
Teflon surfaces were charged by applying DC voltages Uc of up to ± 140 V for 2 to 15 min between the ITO electrodes on the substrate (kept at electrical ground potential) and a platinum (Pt) wire (0.1 mm diameter) immersed into a 5 µL drop of de-ionized water (MilliQ). Charging the surface was performed at room temperature (~25 °C) in a closed container filled with vapour-saturated air.
The generic setup is shown in Figure S1  in which depends only weakly on the applied voltage (see Figure S2). was limited to ensure that the simultaneously measured current on the substrate remained below 1 µA for all experiments.

Contact angle measurements
The wettability of the samples was measured using a commercial contact angle goniometer (OCA- with negligible hysteresis. The EW response was probed by applying a triangular waveform (±30 V) with a period of 60 s to the probe drop. The maximum voltage during the EW-surface characterization measurements was kept deliberately low to ensure that the system displays a parabolic response following the equation Here, is the contact angle under applied voltage of , and = 0 = 2. For more detailed aspects of EW measurements including their interpretation in the presence or absence of surface charges, see Ref. 8 .

Kelvin probe force microscopy (KPFM)
To characterize the electrostatic potential of the surface in more detail and with high lateral resolution, KPFM measurements were performed using a commercial atomic force microscopy Here ∂C(z)/ ∂z is the gradient of the capacitance between tip and sample surface and is the trapping voltage. Splitting the force according to their frequency ( ), we obtain the static (FDC) and dynamic (Fω and F2ω) contributions, as usual, The amplitude of is proportional to − . To obtain the in amplitude modulation KPFM, is adjusted such that becomes minimal. For a system with a perfectly homogeneous dielectric film and a bottom electrode layer, the surface potential is expected to be identical with the trapping voltage . More details on measuring with the KPFM can be found in S.I. and Figure S3.

Macroscopic surface wettability upon charging
The schematic and working principle of EWOD is shown in Figure 1a. When a voltage ( ) is applied on the dielectric layer via a electrolyte droplet and the bottom electrode, a pulling force ( ) from the applied electric field pulls the TPCL toward the outside direction of the droplet, and thus changes the contact angle. This pulling force = To investigate the charge trapping phenomenon in EWOD, we deposited a 5 µL drop of deionized water on the sample surface in air, as described above, and abruptly turned the charging voltage to as high as −120 . As a response, the drop spread within a few tens of ms from the initial contact angle of ~115° to ~ 70° (Figure 1b). Subsequently, a slow relaxation to ( ℎ ) ≈ 80° took place for approximately 1 min. Along with the increase in contact angle, the radius of footprint area of the drop decreased (Figure 1c). This macroscopic contact angle retreat phenomenon has also been observed in previous reports 28,34,42 . According to the humid environment, this relaxation was not caused by evaporation (see Figure S1). After 300 s, when the charging voltage was turned off, subsequent inspection of the samples by optical and by atomic force microscopy ( Figure 4c) did not display any appreciable variation of the surface topography.
As discussed above, the contact angle variation in EW is the joint effect of the materials surface (interface) tension, the applied voltage, the reversible and the irreversible trapping charges.
In the present experiment, the material surface (interface) tension and the applied voltage ( = −120 V) were kept constant, and the pulling force contributed by the applied voltage of was =

Local contact angles
To investigate whether the charge trapping was reversible and spread at the entire drop-substrate interface, we removed the charging drop (after turning off the charging voltage) and subsequently investigated the surface properties in several manners. First, the wettability of the surface were investigated at high lateral resolution using a contact angle measurement with a much smaller probe droplet (0.3 µL) (the setup shown in Figure S4). To minimize disturbing effects of contact angle hysteresis, these measurements were carried out in ambient oil.

Electrowetting response
The reduction of ( = 0) presented in Figure 2 alone does not clearly indicate whether the effect was caused by local chemical variation of the surface along the contact line, which would reduce Young's angle in eq. (1), or whether it is indeed caused by the expected injection of surface charge, which would give rise to a finite value of . To distinguish between these two scenarios, we performed EW measurements using smaller probe droplets (0.5 µL). Ramping the voltage applied to the droplet up and down in triangluar fashion, we found that the decrease in contact angle with increasing voltage was asymmetric, as expected in the presence of a static surface charge (Figures 3b and c). The asymmetry was found to be strongly position-dependent on te surface, being much more pronounce close to the contact line during charging, Region 2 in Figure   3, as compared to the central part of the charging drop (Region 1). At the same time, maximum contact angle, the apex of the curves, corresponding to the contact angle at zero charge was also slightly decreased as compared to the pristine case. Fitting the eq. (1) to the data shown in Figure   3c, we found that the trapping voltages were of (1) = −16 and

Kelvin Probe Force Microscopy (KPFM) measurement
Considering the fact that the probe droplets spread upon applying the voltage, and thus their footprint area increases quickly, one may wonder whether the probe drop remains within the narrow ring around the original contact line where the deposited charges are presumably trapped.
If the probe drop spreads beyond the charging area, the measured asymmetry of the EW response and the value of would in part reflect the finite lateral extent of the deposited charge pattern rather than its absolute value.
To overcome the resolution limitation of contact angle-base detecting method, we performed slightly smaller than that suggested by the contact angle and EW response measurements. The absolute value of the local surface potential in the KPFM measurements was around -10 V which was consistent with the EW-response measurements at the same charging conditions (shown in Figure S5 and 5a). Nano-sized KPFM probe in ambient air and macroscopic EW-probed drops in ambient oil thus experienced the same surface charge density , which could be obtained from the measured voltages using eq. (2) with = . Thus far, these results from three types of micro-and nano-scale measurements revealed and confirmed that the charges are indeed trapped at the AFP surfaces after EW process and accumulate at the TPCL regions.  To optimize the local surface charge density within the rim, we varied the applied voltage and duration for injecting charges, as well as the polarity of the voltage ( Figure 5 and Figure  In contrast, the maximum positive charge density that we could deposit using the opposite charging polarity was only half of the negative ones, and more importantly , was unstable ( Figure   5c). It relaxed within a few hours of continuous probing a water drop in ambient oil. Due to the unstable nature of the positive charges, studies involving these were not explored in more detail.

Controlling of charge trapping behaviors
The difference between positive and negative charges suggests a stronger affinity of negative charge carriers to the polymer, which is also indicated in previous results [59][60] . The charge density up to −0.35 / 2 could be achieved within charging time of 5 under a charging voltage of -120V, indicating that 10 times higher charge density than that of the spontaneous charges 37 was reached, and almost hundred times faster than that from spontaneous charge accumulation at Telfon AF surfaces in contact with water at the same pH value 37 . This demonstrates the power of electric fields in immobilizing charge carriers at Teflon-water interfaces. Figure 5d illustrates the charge generation in electrowetting governed by two types of processes: the reversible adsorption and the irreversible trapping. Charges adsorbed in shallow traps on the Teflon surfaces were considered to be reversibly adsorbed. Since Teflon is porous on the molecular scale, traps deeper within the films may also be energetically deeper and hold back ions in an irreversible manner leading to a finite trapped charge. The strong electric fields near the contact lines enable much faster trapping and higher charge densities. In addition, the traps for positive charges are shallower than those for negative ones.

Creating narrow charge distributions
According to the classical EW theory, 40-41 the high electric charge densities which should be responsible for the charge injection, are localized within a region of the order of the thickness of the dielectric layer. A solution of the electrostatic problem adapted to the parameters of the present experiments shows that the region, in which the local electric field exceeded the average field / under the charging drop by more than a factor of two, is less than 1 µm in width, as shown in Figure 6. Following this thought, the intrinsic charge generation mechanism should allow to generate much narrower charge distribution region than the measured width of 200 to 300 µm. We attribute this to the relaxation of the contact line position during the charging process that accompanies the contact angle relaxation (Figures 1b and c).  Figure S8.
The slowly receding contact line was believed to leave behind a trace of charges on the surface, which eventually formed the observed rim. In order to reduce the width of the deposited rim of charges, we suppressed the geometric relaxation of the drop during charging by confining it between two parallel plates at a distance of ℎ = 100 µm (Figure 7a and 7b). ℎ was simply achieved by putting a 100 µm thick glass space between the two plates. The lower surface was a Teflon-coated substrate as above and the upper one was an ITO coated glass that served as an electrode during the charging process. These two surfaces confined a drop during charging and reduce the displacement of the contact line ∆ for the same amount of contact angle relaxation ∆ as in Figure 1, with ∆ ∝ ℎ ∆(cos ). KPFM measurements after removing the top surface and the drop demonstrate that indeed a much narrower rim of charges was deposited with a width of about 20 µm, as shown in Figure 7. The average surface potential within the rim was -10 V, corresponding to the trapped charge density = −0.22 / 2 . From these results, we could also conclude that a much smaller but finite charge density was deposited at the solid-liquid interface away from the contact line, as seen in

CONCLUSION AND OUTLOOK
In this work, we reveal that,   where the ( ) is the capacitance per unit area of the air capacitor formed between the probe and the dielectric surface. ( ) depends on the distance between the probe and the dielectric layer (z), and can be calculated by ( ) = 0 ⁄ . is the capacitance per unit area of the dielectric layer.
, and are the potential at the probe, at the dielectric surface and the bottom electrode.
Given the total charge in the system is zero, we get where is the trapping charge density on the surface of dielectric layer.
From Eq. S1 -S3, we get: Here, we should notice that the surface potential of the dielectric layer is not constant during the measurement. The dielectric surface potential ( , ) does not only depend on the trapping charge density , but also on z and . We define the trapping voltage as = / . Only if the probe is very far away from the dielectric surface, the surface potential equals the trapping voltage, = = / . The electric energy of the system contain two parts: the energy from the capacitance and the energy from the source (battery) , can be written as: The energy of the capacitance contains the energy in the air capacitance between the probe and the surface ( ) and the energy in the dielectric layer , and can be calculated as: The energy from the source is: Where 0 is the initial energy stored in the source. is the overlapping area of the probe and the dielectric surface. According to Eq. S1, S2 and S3, the charge density on the probe is : Substitute Eq. S10 to Eq. S9, we get Substitute Eq. S8 and Eq. S11 to Eq. S7, the electric energy in the system is Given the trapping voltage = / and the total capacitance ( ) of the probe and the dielectric layer: the electrical energy is given by: The electric force on the AFM probe is given by the gradient of the energy: Because ∂C ∂z is negative, this force is attractive. Since the potential on the tip is the sum of an AC voltage ( sin ) and a DC voltage ( ), can be written as: = + sin ( 16) while the bottom electrode is grounded: = 0. Thus, the electric force is: