ARTICLE pubs.acs.org/Langmuir
Role of Electric Field on Surface Wetting of Polystyrene Surface Bharat Bhushan*,† and Yunlu Pan†,‡ †
Nanoprobe Laboratory for Bio- & Nanotechnology and Biomimetics (NLB2), The Ohio State University, 201 W. 19th Avenue, Columbus, Ohio 43210-1142, United States ‡ Mechanical Engineering, Harbin Institute of Technology, Harbin, 150001, People's Republic of China ABSTRACT: The role of surface charge in fluid flow in micro/ nanofluidics systems as well as the role of electric field to create switchable hydrophobic surfaces is of interest. In this work, the contact angle (CA) and contact angle hysteresis (CAH) of a droplet of deionized (DI) water were measured with applied direct current (DC) and alternating current (AC) electric fields. The droplet was deposited on a polystyrene (PS) surface, commonly used in various nanotechnology applications, coated on a doped silicon (Si) wafer. With the DC field, CA decreased with an increase in voltage. Because of the presence of a silicon oxide layer and a space charge layer, the change of the CA was found to be lower than with a metal substrate. The CAH had no obvious change with a DC field. An AC field with a positive value was applied to the droplet to study its effect on CA and CAH. At low frequency (lower than 10 Hz), the droplet was visibly oscillating. The CA was found to increase when the frequency of the applied AC field increased from 1 Hz to 10 kHz. On the other hand, the CA decreased with an increasing peakpeak voltage at or lower than 10 kHz. The CAH in the AC field was found to be lower than in the DC field and had a similar trend to static CA with increasing frequency. A model is presented to explain the data.
1. INTRODUCTION Hydrophobicity plays an important role in fluid flow in various micro/nanofluidics based biomedical applications.13 Polymers, such as polydimethylsiloxane (PDMS) and PS, are widely used in micro/nanofluidics systems.4 During fluid flow, the channel surface can develop a surface charge which affects drag in fluid flow.5 Therefore, one needs to understand the effect of an electric field on drag and hydrophobicity. The electric field can be used for switchable hydrophobicity, which means reversibly switching between hydrophilicity and hydrophobicity,58 and even superhydrophobicity.9 It may find applications in droplet manipulation and fluid flow in various fields, such as drug delivery systems,10,11 biosensors, actuator systems, and bioseparation.12 In addition, an electric field can be used to manipulate microobjects, for example, on a PS surface with DI water8 and on polytetrafluoroethylene (PTFE) surface with both DI water13 and saline.14,15 When an electric field is applied to a droplet deposited on a dielectric layer, the liquid surface tension changes, which results in the change of CA, and is called electrowetting.58 Experiments to study the effects of DC and AC fields on the CA have been carried out. As an example, with an increase in DC field, the CA of a droplet decreased when deposited on PTFE surfaces in saline6,7 and on PS surfaces in DI water.5,8 In the presence of an AC electric field, the effects on the CA have some interesting features. In this case, the change of the CA was found to be larger than in the DC case with ionic liquids deposited on a PTFE surface.16 The CA was found to increase with an increasing frequency of AC voltage on PTFE surface with DI water.17 Oscillation of a droplet was reported on a PTFE surface with saline with an AC field.18 The CAH, which is the difference between advancing and receding CA, decreased when the amplitude of AC voltage increased19 and remained constant with an increasing r 2011 American Chemical Society
DC voltage. These effects may be of interest in drag reduction in fluid flow. As mentioned earlier, PS is widely used in micro/nanofluidic systems. In the present research, a systematic study of the effect of DC and AC fields on the CA and CAH of a droplet deposited on PS surface is carried out. To study the effect of AC voltage on Si substrate coated with PS, the AC field with positive and negative values cannot be used because the presence of the surface charge on Si substrate will generate a current which will destroy the coating when the direction of the electric field is changing. Therefore, an AC field with positive values has been used in this work. A model has been presented to explain the trend in the DC field data.
2. EXPERIMENTAL SECTION The electric field experiments were performed as illustrated in Figure 1. A DI water droplet was placed on a PS surface spin coated on a p-type boron doped silicon wafer (120 Ω/m, Silicon Quest International) with a 300-nm thick silicon oxide coating grown thermally on the wafer. The silicon wafer was first cleaned in a sonication bath of acetone and then isopropyl alcohol. After that, a PS film obtained using a spin coating method was applied on the cleaned silicon wafer at a speed of 2000 rpm. The PS solution of concentration of 1% (w/v) by dissolving PS pellets (molecular weight 35000, Sigma Aldrich) in toluene (Mallinckrodt Chemical). To remove any remaining solvent, the PS sample was annealed in an oven at 53 ( 2 °C for 4 h. The thickness of the PS film was measured to be about 66 nm with an ellipsometer (model L116C, Gaertner Scientific Corporation). The CA of DI water Received: May 3, 2011 Revised: June 16, 2011 Published: June 17, 2011 9425
dx.doi.org/10.1021/la201636g | Langmuir 2011, 27, 9425–9429
Langmuir
ARTICLE
Figure 1. Schematic of the experimental setup used for measurement of the CA as a function of applied voltage. was about 94°. The sample was then glued to a metal sample plate which was used as an electrode with conductive silver paint to make sure the doped silicon substrate and the metal plate was electrically connected. A stainless steel wire was inserted into the droplet as another electrode. An electrical signal was generated by a function generator (33120A, HP) and was amplified by a power amplifier (model 50, Ling-Altec Electronics Inc.). A DC voltage and an AC voltage with a square wave were applied. The frequency of the AC voltage was varied from 0.5 to 104 Hz. The droplet shape was observed by using a high-speed camera (FS100, Canon) with 1152 864 pixels, and the image was recorded. CA was measured five times for each picture by using Scion Image on a PC. For CAH measurement, while inflating the droplet volume by a syringe, the maximum value of the CA before the contact line changes was recorded as advancing CA, and while sucking the droplet, the minimum value of the CA before the contact line changes was recorded as receding CA. The difference between advancing and receding CA was calculated as the CAH. In order to minimize the effect of evaporation, the whole process was finished in less than 10 min. When a dielectric surface, such as PS, is used between electrodes, the surface will have a tendency to electrostatically charge. When the PS coating is deposited on a silicon substrate which is a semiconductor, there will be another layer of charge on the Si substrate. When the PS coating is combined with another dielectric coating such as a silicon oxide layer, there will be additional charge in the silicon oxide and the two interfaces between silicon oxide and PS and Si. On a surface with that charge present, when the electric field has opposite polarity, the charge is additive which leads to large discharge current.20 This may damage the brittle PS coating. Considering that, an AC voltage with a minimum value of 0 V in square wave was applied to the Si substrate with PS coating.
3. MODEL FOR THE DC ELECTROWETTING A model was developed to obtain an expression which provides the relationship between CA and voltage. The surface tension (liquidsolid) γsl decreases when a voltage V is applied between the droplet and the substrate because of an electric double layer generated by the electric field.21 The surface tension at a voltage V can be given as follows: 1 γsl ðV Þ ¼ γsl ð0Þ CV 2 2
ð1Þ
Young’s equation combined with eq 1 describe the changed CA, θ, of the droplet with an electric field, which is the so-called Young-Lippmann equation,6,22 cos θ ¼ cos θ0 +
C 2 V 2γlv
ð2Þ
where θ0 is the CA in the absence of an electric field, C is the capacitance of the dielectric layer, and γlv is the liquidvapor interfacial tension.
Figure 2. DC voltage dependence of the shape and CA of the droplet. The voltage is increasing from 0 to 30 V, and then back to 0 V (the right image).
In this work, the dielectric layer has two parts: silicon oxide film and PS film (Figure 1). The capacitance of the dielectric layer can be given as follows: 1 1 1 d1 d2 ε1 d2 + ε2 d1 ¼ + ¼ + ¼ C C1 C2 ε0 ε1 ε0 ε2 ε0 ε1 ε2
ð3Þ
where C1, C2 are the capacitance of silicon oxide film and PS film, ε1, ε2 are the dielectric constants of the silicon oxide and PS, d1, d2 are the thickness of silicon oxide film and PS film, respectively, and ε0 is the dielectric constants of the vacuum. By combining eq 3 with eq 2, ε0 ε1 ε2 V 2 ¼ cos θ0 + kV 2 ð4Þ cos θ ¼ cos θ0 + 2ðε1 d2 + ε2 d1 Þγlv where k¼
ε0 ε1 ε2 2ðε1 d2 + ε2 d1 Þγlv
ð5Þ
The value of k can be calculated by the known value of ε0, ε1, ε2, d1, d2, and γlv, and also can be given by eq 4 with the known value of V and the measured values of θ and θ0. When Si, which is a semiconductor, contacts the metal, because of the different work functions, there is always a space charge layer.23 With different types of Si or different kinds of metal, the surface charge would be different. In this work, the p-type boron doped Si substrate contacted the silver paint, so there is a space charge layer which will reduce the effective voltage. That means when eq 4 is used as the model, the calculated value of k based on eq 5 will be larger than the fitted value based on eq 4.
4. RESULTS AND DISCUSSION Both DC and AC voltage were applied to the droplet. In the DC case, the CA and CAH as functions of different voltages were recorded. With AC voltage, CA as a function of different frequencies and peakpeak voltage was recorded. The oscillation of the droplet was visible, so the maximum and minimum of the CA were recorded as well. To avoid the influence of oscillation, CAH with an AC field was recorded as a function of different frequencies higher than 10 Hz. 4.1. DC Voltage Effects on Si Substrate with PS Coating. Figure 2 shows images of the droplet with different DC voltages. The CA was measured with an applied voltage from 0 to 30 V. The CA of the droplet decreases from 94° to 74° with an increase of the voltage. When the voltage is removed, the CA increases to 83° but does not increase enough to the original CA of 94° without any voltage. We believe the reason of this observation is that after the electric field was removed, there was residual the surface charge which provides an electric field to the system. The CA as a function of applied DC voltage is shown in Figure 3. Also shown are cosine values in order to fit the data to eq 4 to 9426
dx.doi.org/10.1021/la201636g |Langmuir 2011, 27, 9425–9429
Langmuir
ARTICLE
Figure 4. Schematic of the balance of horizontal forces at the contact line between surface and water. γsl(V) is the electric field dependent surface tension between solid and liquid, γsv and γlv are the solidvapor rec and liquidvapor interfacial tension, θ is the CA, Fadv max and Fmax are the pinning force when advancing and receding, respectively.
However, even with the effect of silicon oxide film, the calculated value based on eq 5 is still higher than the fitted value of k, which is in agreement with previous discussion. This observation suggests that, with Si substrate, the change of CA with electric field is smaller than with metal substrate because of the presence of silicon oxide layer and space charge layer. On the basis of the eqs 4 and 5 and proceeding analysis, it can be understood that, with the electric field, not only the applied voltage, but also the thickness of PS film or silicon oxide layer, can be used to control the effect of voltage. Different types of Si will induce different values of space charge, which also should affect the effect of voltage. Figure 3 also shows the CAH as a function of applied voltage. The change of CAH with voltage is not obvious. To understand this observation, the analysis of the balance of horizontal forces at the unit length of the contact line is carried out. The pinning rec force for advancing Fadv max and receding Fmax are assumed to have a same value but opposite directions. As shown in Figure 4, the balance can be given by the following:
Figure 3. (a) The CA of the droplet and its cosine value as functions of applied DC voltage. The modified YoungLippmann equation was fit to the cosine value of the CA. (b) The CAH of the droplet as a function of applied DC voltage. Error bars represent ( σ.
Table 1. Comparison of the Fitted Valuea and the Calculated Values of k = ((ε0ε1ε2)/(2(ε1d2 + ε2d1)γlv)), where ε0 = 8.854 1012 F/m (= N/V2), ε1(SiO2) = 4.42, ε2(PS) = 2.55 and γlv = 72.75 mN/m, respectively24 calculated value based on eq 5, (V2)
fitted value, (V2)
k (d1 = 0 nm,
k (d1 = 300 nm,
d2= 66 nm)
d2= 66 nm)
k
23.5 104
6.49 104
3.26 104
a
For the test sample: d1 = 300 nm, d2= 66 nm, and for the sample without the silicon oxide layer: d1 = 0 nm, d2= 66 nm. Fitted value of k was given by fitting the cosine value of measured CAs with eq 4.
calculate k. The fitted value of k was obtained as 3.26 104 V2 with an R2 value of 0.99. For comparison, the k value is also calculated using eq 5, and the data are presented in Table 1. There are two calculated value of k based on eq 5, one is for the model of the test sample, and the other is for a sample without regard to the silicon oxide layermetal substrate. It is obvious that the presence of the silicon oxide layer reduces the value of k.
adv for advancing : γlv cos θadv ¼ γsv γsl ðV Þ Fmax
ð6Þ
rec for receding : γlv cos θrec ¼ γsv γsl ðV Þ + Fmax
ð7Þ
where θadv and θrec are advancing and receding CA, respectively. As mentioned before, when a voltage is applied, γsl(0) decreases to γsl(V) while γlv, γsv,Fmax remain constant. So the change of the cosine value of advancing and receding CA will be: Δcos θrec ¼ Δcos θadv ¼
γsl ð0Þ γsl ðV Þ γlv
ð8Þ
That means with applied voltage, both cosθadv and cosθrec decrease in the same way. In this case, both the difference between the advancing and receding CA and the change of surface tension are small. So with a similar change in the cosine value, the change of CA is also similar, which results in the change of CAH being unobvious. 4.2. AC Voltage Effects on Si Substrate with PS Coating. Figure 5 shows the images of the droplet with different frequencies of an AC voltage with peakpeak value of 30 V. In low frequency (e10 Hz), the oscillation of the droplet is visible, and can be described by two images for each value of frequency; one is the droplet with the maximum CA labeled high, the other is the droplet with the minimum CA labeled low. Figure 6 shows the CA of the droplet as a function of the frequency of an AC voltage. It is observed that first the high CA and the low CA remain constant, followed by a steady increase above a frequency of about 1 Hz. The increase in CA with a frequency above 1 Hz is 9427
dx.doi.org/10.1021/la201636g |Langmuir 2011, 27, 9425–9429
Langmuir
ARTICLE
Figure 5. Frequency dependence of the shape and CA of the droplet. The peakpeak value of the AC voltage is 30 V, and the minimum value is 0 V. At lower frequency (lower than 10 Hz), the droplet oscillates, so two different CA for one frequency were recorded; one is the droplet with maximum CA labeled high and the other is the droplet with minimum CA labeled low.
similar to the published data on PTFE coated metal (Hong et al., 2008). The CAH as a function of the frequency is also shown in Figure 6. The CAH in an AC field was lower than in a DC field and increased as static CA with an increasing frequency of applied AC voltage. In the AC case, the voltage was switching between 0 and 30 V. The advancing CA will have a similar trend with static CA in an AC field. That means it will increase with an increasing frequency. However, for the receding CA, based on eq 7, the value of receding CA should be always obtained when the voltage is equal to 0 V. That means receding CA will remain constant with an AC field. Thus, with lower advancing CA and same receding CA, CAH in AC field should be lower than without voltage and should increase with frequency which is in agreement with the experimental data. Figure 6 also shows the CAs with different peakpeak values at different frequencies. The CA remains constant at high frequency, whereas it decreases at lower frequencies at an increasing rate at higher voltage. In this case, when the peakpeak value is lower than 16 V or the frequency is higher than 10 kHz, the effect of AC voltage on CA is not obvious. That means to get AC voltage effects on PS coated Si substrate with DI water, an AC voltage with higher peakpeak value and lower frequency is needed.
5. CONCLUSIONS The effects of DC and AC fields on CA and CAH for Si substrate with PS coating were studied. In the case of the application of a DC field, the CA of the droplet decreased with an increasing voltage. When the voltage was removed, the CA did not increase to the original value without any voltage. A model was developed to explain the experimental data. On the basis of the model, it was found that with an Si substrate, the change of CA with electric field is smaller than with a metal substrate. Moreover, for the case of the application of a DC field, one can control the change of the CA of a droplet deposited on an Si
Figure 6. (a) The CA and CAH of the droplet as function of the frequency of an AC voltage with a peakpeak value of 30 V. The frequency is in log scale. The “high” and “low” plots mean the maximum and minimum value of the CA, respectively. (b) The CA of the droplet as function of the peakpeak value of applied AC voltage at different frequencies. The minimum value of the AC voltage is 0 V. For 1 and 10 Hz, there are two plots labeled “high” and ”low” for the maximum and minimum value of CA, respectively, whereas at 10 kHz they overlap. Error bars represent ( σ.
substrate with PS coating not only by changing the applied voltage but also by changing the thickness of PS film or silicon oxide film. The type of Si can also affect the change of the CA. The CAH did not have an obvious change with DC field, because the advancing and receding CA decreased in the same way. In the case of an AC field, oscillation of the droplet was visible. Both minimum and maximum CAs either increased or remained constant at low frequency followed by an increase with frequency. CA remained constant with voltage at high frequency, whereas it decreased at lower frequencies at an increasing rate at higher voltage. On the basis of this observation, an AC voltage with higher peakpeak value and lower frequency should be used to realize variations in CA. The CAH in AC field was found to be lower than in DC field, and increased as static CA with increasing frequency due to the advancing CA decreased with AC filed and had a similar trend to static CA with increasing frequency while the receding CA remained constant. 9428
dx.doi.org/10.1021/la201636g |Langmuir 2011, 27, 9425–9429
Langmuir
ARTICLE
’ AUTHOR INFORMATION Corresponding Author
*E-mail:
[email protected].
’ REFERENCES (1) Nosonovsky, M.; Bhushan, B. Multiscale Dissipative Mechanisms and Hierarchical Surfaces: Friction, Superhydrophobicity, and Biomimetics; Springer-Verlag: Heidelberg, Germany, 2008. (2) Bhushan, B. Springer Handbook of Nanotechnology, 3rd ed.; Springer: Heidelberg, Germany, 2010. (3) Jung, Y. C.; Bhushan, B. J. Phys.: Condens. Matter 2010, 22, 035104. (4) Whitesides, G. M. Nature 2006, 442, 368–373. (5) Wang, Y.; Bhushan, B. Langmuir 2010, 26, 4013–4017. (6) Quilliet, C.; Berge, B. Curr. Opin. Colloid Interface Sci. 2001, 6, 34–39. (7) Mugele, F.; Baret, J. C. J. Phys.: Condens. Matter 2005, 17, R705–R774. (8) Bhushan, B.; Ling, X. J. Phys.: Condens. Matter 2008, 20, 485009. (9) Krupenkin, T.; Taylor, J. A.; Kolodner, P.; Hodes, M. Bell Labs Tech. J. 2005, 10, 161–170. (10) Sawahata, K.; Hara, M.; Yasunaga, H.; Osada, Y. J. Controlled Release 1990, 14, 253–262. (11) Qiu, Y.; Park, K. Adv. Drug Delivery Rev. 2001, 53, 321–339. (12) Roy, I.; Rao, M. V. S.; Gupta, M. N. Biotechnol. Appl. Biochem. 2003, 37, 9–11. (13) Cho, S. K.; Moon, H.; Kim, C. J. J. Microelectromech. Syst. 2003, 1, 70–80. (14) Lee, J.; Moon, H.; Fowler, J.; Schoellhammer, T.; Kim, C. J. Sens. Actuators 2002, 95, 259–268. (15) Pollack, M. G.; Fair, R. B. Appl. Phys. Lett. 2000, 9, 1725–1726. (16) Nanayakkara, Y. S.; Perera, S.; Bindiganavale, S.; Wanigasekara, E.; Moon, H.; Armstrong, D. W. Anal. Chem. 2010, 82, 3146–3154. (17) Hong, J. S.; Ko, S. H.; Kang, K. H.; Kang, I. S. Microfluid. Nanofluid. 2008, 5, 263–271. (18) Jung, M. O.; Sung, H. K.; Kwan, H. K. Langmuir 2008, 24, 8379–8386. (19) Li, F.; Mugele, F. Phys. Rev. Lett. 2008, 92, 244108. (20) Zaghloul, U.; Papaioannou, G.; Coccetti, F.; Pons, P.; Plana, R. Microelec. Reliability 2009, 49, 1309–1314. (21) Lippmann, G. Ann. Chim. Phys. 1875, 5, 494. (22) Israelachvili, J. Intermolecular and Surface Forces, 2nd ed.; Academic Press: New York, 1991. (23) Rhoderick, E. H. IEE Proc. 1982, 129, 1–14. (24) Haynes, M. W. Handbook of Chemistry and Physics, 91st ed.; CRC Press Inc.: Boca Raton, FL, 2010.
9429
dx.doi.org/10.1021/la201636g |Langmuir 2011, 27, 9425–9429
