Sensors and Actuators A 165 (2011) 27

Contents lists available at ScienceDirect

Sensors and Actuators A: Physical
j our nal homepage: www. el sevi er . com/ l ocat e/ sna
Capacitive tactile sensor array for touch screen application
Hong-Ki Kim, Seunggun Lee, Kwang-Seok Yun

Department of Electronic Engineering, Sogang University, 1 Shinsu-dong, Mapo-gu, Seoul 121-742, Republic of Korea
a r t i c l e i n f o
Article history:
Available online 13 January 2010
Keywords:
Tactile sensor
Flexible device
Touch screen
Multi-touch
Capacitive sensor
a b s t r a c t
In this paper, we propose and demonstrate a transparent and exible capacitive tactile sensor which is
designed for multi-touch screen application with force sensing. A sensor module is composed of 2Darray
tactile cells with a spatial resolution of 2mmto measure the touch force at multiple positions. The device
is fabricated by using transparent materials on a transparent plastic substrate. The optical transmittance
of the fabricated tactile sensor is approximately 86% in the visible wavelength region, and the maximum
bending radius is approximately 30mm. The cell size is 1mm1mm, and the initial capacitance of each
cell is approximately 900fF. The tactile response of a cell is measured with a commercial force gauge
having a resolution of 1mN. The sensitivity of a cell is 4%/mN within the full scale range of 0.3N.
2010 Elsevier B.V. All rights reserved.
1. Introduction
Atouchscreenis a display that candetect the presence andloca-
tion of a touch on a display area. Currently, touch screens, because
they provide very intuitive user interfaces, are widely used not
only in computer systems in the industry but also in hand-held
devices such as mobile phones, PDAs, and car navigation systems.
The important characteristics of a touch screen that is used as a
displayincludetransmittance, resolution, resistancetosurfacecon-
tamination, durability (lifetime), multi-touch recognition, display
size, and force sensing. Among these characteristics, multi-touch
recognition, which has recently been incorporated in several mod-
els of mobile phones and portable electronic devices, enables a
user to interact with a system by simultaneously using multiple
ngers. As will be discussed briey in this paper, it has been dif-
cult to apply multi-touchrecognitionto most classical touchscreen
technologies.
Various sensingtechnologies havebeendevelopedusingdiverse
approaches, and they are widely used in commercial products
using touch screens. Resistive [1], capacitive [2], optical using
infrared (IR) [3], and acoustic using surface acoustic wave (SAW)
[4] detection methods have been used in most conventional touch
screens. However, these types of touch screens recognize only
a single touch point. There are several technologies for multi-
touch recognition. The patterned capacitive-type touch screen
consists of transparent rowandcolumnelectrode arrays embedded
within some insulating material [5,6]. This arrangement moni-

Corresponding author. Tel.: +82 2 705 8915; fax: +82 2 705 8915.
E-mail address: ksyun@sogang.ac.kr (K.-S. Yun).
tors the change in capacitance that occurs at the point on the
screen where a nger is placed. Han reported multi-touch sens-
ing on rear-projected interactive screens based on the frustrated
total internal reection technique, which required a video cam-
era to monitor the nger locations [7]. The above-mentioned touch
screen technologies are well-adopted to a at panel display. How-
ever, nowadays, many studies have reported on exible displays,
because the at panel display using a glass substrate is fragile
and difcult to carry [8]. To be utilized in a exible display, the
tactile sensor for a touch screen should also exhibit exibility.
Therefore, in this work, a transparent and exible tactile sensor
has been designed for a multi-touch screen application. In addi-
tion, we are aiming at developing a touch sensor capable of force
sensing in order to discriminate among different levels of touch
strength.
In fact, touch sensors with force sensing have been researched
for the last few years as tactile sensors mainly for articial
skin for robot applications [9,10], minimally invasive surgery
[11,12], wearable computers [13], and mobile or desktop hap-
tic devices [14]. Four popular pressure-sensing mechanisms
for tactile sensors have been reported: resistive, piezoresistive,
piezoelectric, and capacitive-sensing mechanisms. In resistive sen-
sors, a resistance change induced from the resistive material
squeezed between electrodes is measured [15]. A piezoresis-
tive sensing mechanism uses a strain gauge to measure the
deformation of a tactile cell [16]. A piezoelectric mechanism
measures the accumulation of charges and the resulting volt-
age buildup as a membrane is forced. However, a piezoelectric
sensor cannot detect static force [17]. A capacitive-sensing
mechanism measures the capacitance change induced by the
change in the gap between the electrodes [9]. However, most
0924-4247/$ see front matter 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.sna.2009.12.031
H.-K. Kim et al. / Sensors and Actuators A 165 (2011) 27 3
of these devices are not suitable for touch screen display sys-
tems because of the non-transparency of the materials they are
made of. In order to meet the requirement for tactile sensors
for multi-touch screens for exible display applications, we have
introduced a capacitive tactile sensor array constructed with
polycarbonate (PC) lms and indiumzinc-oxide (IZO) electrodes
for exibility and transparency. In this paper, we present the
concept, fabrication, and experimental results of our sensor in
detail.
2. Design
Fig. 1 shows the cross-sectional view and the dimension of a
unit cell of the proposed tactile sensor array. The upper and bottom
substrates are transparent PC lms with a thickness of 120m. A
thin transparent IZOlayer was used as the electrodes and the signal
lines. The twoelectrodes formeda capacitor separatedbya distance
of 13m by SU-8 spacers. The cell size and electrode size were
2mm2mmand 1mm1mm, respectively. The capacitance of a
cell can be expressed as
C =
1
(t
a
/
0
A) +(t
d
/
d

0
A)
, (1)
where
0
is the permittivity in free space,
d
is the relative per-
mittivity of the SU-8 insulation layer, t
a
is the air-gap distance,
t
d
is the thickness of the SU-8 insulator layer, and A is the elec-
trode area. The initial capacitance of a cell was estimated to be
Fig. 1. Cross-sectional view of a tactile cell and its dimensions.
926fF using Eq. (1) assuming that the relative permittivity of SU-
8 was 3.2. When a touch pressure was applied on the surface of
the upper plate, the gap between the two plates decreased and the
capacitance increased until the gap was closed. By measuring the
capacitance for all the capacitive array cells, we could determine
the touch position and the applied force on multiple locations.
The membrane deection and resultant capacitance change as
the touch force applied must be considered for a capacitive cell
design. These factors were examined by the nite element method
(FEM) simulation for a capacitive cell with the dimensions given
Fig. 2. Center deection (solid line: calculated, dashed line: simulation) and capacitance (dashdot line) for various applied forces.
4 H.-K. Kim et al. / Sensors and Actuators A 165 (2011) 27
Fig. 3. Fabrication processes of the proposed tactile sensor: (a) electrode layer for-
mation, (b) spacer formation on top plate, (c) electrode layer formation on bottom
plate, (d) insulation layer coating, and (e) the completed device after bonding pro-
cess.
Fig. 4. Fabricated tactile sensor: (a) exibility and (b) magnied view of touch sen-
sor.
in Fig. 1 using a COMSOL multiphysics simulator (COMSOL Inc.).
Fig. 2(a) shows two examples of simulation results: the initial sta-
tus with zero touch force (left) and with the touch force of 68mN
at a point where the upper plate just begins to touch the bottom
plate (right). Fig. 2(b) shows the center deection and the resulting
capacitance for various applied forces. The solid line is the center
deectionversus the appliedforce, andthe dashedline is the capac-
itance change. The initial capacitance was estimated to be 938fF,
which was close to the calculated value of 926fF. The upper plate
begantotouchthe bottomplate whenthe appliedforce was 68mN;
the capacitance at this point was approximately 3.4pF. Another
important factor that must be considered is mechanical response
time of the cell membrane because slowresponse time will result in
afterimage lag on display. The calculated and simulated resonance
frequency of the designed membrane is about 21.5kHz which is
fast enough comparing with 60Hz, a general refresh time of display
pixel.
3. Fabrication
In our design, we used transparent PC lms as structural mate-
rials, SU-8 (Microchem Co.) as spacers and an insulator, and an IZO
thin lm as electrodes. The fabrication process is shown in Fig. 3.
Each layer was processed separately and bonded together using
SU-8 as the bonding material, which was also used as an insula-
tor and a spacer. We used an IZO-coated PC lm. The thickness of
the PC lm and the IZO thin lm was 120m and 130nm, respec-
tively. For photolithography, the lms were mounted on a silicon
wafer. For the top plate, rst, the IZOlayer was patterned (Fig. 3(a))
using general photolithography and wet etching. The solution with
Fig. 5. (a) Optical transmittance of tactile sensor measuredwithspectrophotometer
and (b) tactile sensor on LCD display of mobile phone.
H.-K. Kim et al. / Sensors and Actuators A 165 (2011) 27 5
hydrochloric acid and nitric acid (HCl:HNO
3
=3:1) was used as an
IZO wet etchant, and the etch time was approximately 5s at room
temperature. Next, SU-8 2007 (MicrochemCo.) was spin-coated to
have a thickness of 13m and patterned to form spacers on the
top substrate (Fig. 3(b)). As the bottomplate, the IZOlayer was pat-
terned again for the bottom electrode (Fig. 3(c)). Then, a thin SU-8
2005 was spin-coated to have a thickness of 5m, forming an insu-
lation layer between the top and the bottom electrodes (Fig. 3(d)).
Next, the top substrate was aligned with the bottomsubstrate, and
pressure was appliedat roomtemperature (Fig. 3(e)). Then, the two
substrates that were bonded together were heated on a hot plate
at 95

C for 1min to cure the thin SU-8 layer. Finally, the SU-8 was
hardened after UV exposure and post-exposure bake at 95

C for
1min.
Fig. 4 shows the fabricated tactile sensor. The initial device
was designed to have 2020 capacitive cells, and the size of
the entire sensor module was 6cm6cm, including the intercon-
nection pads. The fabricated sensor exhibited good exibility, as
shown in Fig. 4(a). Fig. 4(b) shows the magnied view of the fab-
ricated tactile sensor. The overlap area of each capacitive cell was
1mm1mm, and the diameter of SU-8 spacer was 200m.
4. Experimental results
The transparency of the fabricated tactile sensor was measured
usinga UV/Visible spectrophotometer (SCINCO). The average trans-
mittance was approximately 86% in a visible light range from
380nm to 770nm, as shown in Fig. 5(a). We placed the tactile sen-
Fig. 6. (a) Measurement setup for single tactile cell characterization and (b)
schematic representation of readout circuits for the fabricated sensor module.
Fig. 7. Measured response (solid line) and simulation result (dashed line) of the
fabricated cell for various touch forces.
sor on top of the LCD display of a commercial mobile phone to test
the visibility, and as seen in Fig. 5(b), there was no interference or
decrease in visibility.
We set up custom-made equipment for touch force char-
acterization. Fig. 6(a) displays our setup for the contact force
measurement. A force gauge with a tip was used to precisely
Fig. 8. (a) Photograph of rubber stamps and their touch images captured by the
fabricated tactile sensor module and (b) multi-touch tactile images captured from
the fabricated sensor. Areas of two neighboring unit cells are designated with red
and blue dashed squares. (For interpretation of the references to color in this gure
legend, the reader is referred to the web version of the article.)
6 H.-K. Kim et al. / Sensors and Actuators A 165 (2011) 27
apply pressure on a specic capacitive cell. The end of the tip
was treated to have a at rectangular shape with the dimension
of 1mm1mm, and the force gauge from AIKOH Engineering
Co. had a force resolution of 1mN. Fig. 6(b) shows the custom-
designed readout circuitry. First, each tactile cell was selected by a
row decoder and reset. Then, it was charged to V
step
. When the
cell was selected by a column decoder, the stored charge was
transferred to the feedback capacitance (C
f
) and generated out-
put voltage, as given in the equation. The signal was processed
by a custom-designed eld programmable gate array (FPGA) chip,
and the nal image was displayed by LabVIEW (NI). In order to
remove the offsets from the circuit, we designed the circuit to
read a single cell twice with and without resetting the feedback
capacitance.
Fig. 7shows the measuredresponse of the fabricatedcell for var-
ious touch forces. Further, the FEM simulation result is depicted as
a dashed line in this gure for the sake of comparison. A single cell
was pressed by using a micro-force gauge with a tip, as shown in
Fig. 6(a). All experimental data are the averaged value of 10 mea-
surements on different cells and standard deviation is less than
6.7%. The initial capacitance of a cell was measured to be approx-
imately 900fF, which was close to the theoretical value of 926fF
obtained fromEq. (1) and the simulated value of 938fF. The capac-
itance increased linearly with the applied force before 0.1N and
saturated after that pressure, which implied that both the upper
and the bottomelectrodes were incontact withthe insulationlayer
between them. Moreover, we can see that the experimental results
Fig. 9. (a) Tactile sensor attached on cylindrical structure with radius of curvature
of 30mm and (b) movement of touch points on screen according to nger motion.
adequately followed the simulation result before saturation.
Multi-touch tactile images captured from the fabricated sen-
sor are shown in Fig. 8. Pressure was applied by using a rubber
stamp with the letter T on it, and the corresponding image
was clearly captured in Fig. 8(a). Further, several point images
according to various touch pressures are seen in Fig. 8(b) and
the areas of two neighboring units cells in post processing dis-
play programare designated with the red and blue dashed squares
in this gure. The program was designed to change the both of
darkness and size of color in a cell area. In this experiment, we
rst applied pressures on different locations at the same time
using several tips and capacitance values of each cell are mem-
orized. To nd the pressure values on each cell giving recorded
capacitance values, we applied pressure on each cell using force
gauge with sharp tip. We can clearly see that the brightness and
size of the point images increased in proportion to the touch
pressure.
Sliding experiments on a curved surface were also performed,
andtheir results are showninFig. 9. The tactile sensor was attached
on a cylindrical structure with the radius of curvature of 30mm, as
shown in Fig. 9(a). Two ngers touched two different points on
the tactile sensors and moved on the surface with slight pressure.
The sliding speed of the ngers was approximately 2cm/s. Fig. 9(b)
shows that the touch points on the screen satisfactorily follow the
sliding of the ngers on the curved tactile sensor.
5. Conclusions
In this study, a new exible and fully transparent tactile sen-
sor for touch screen applications was proposed and successfully
demonstrated. A sensor module consisted of a 2020 tactile cell
array with a spatial resolution of 2mm. The fabricated tactile sen-
sor module exhibited good exibility with a maximum radius of
curvature of 30mm and captured multi-touch images. The cell to
cell variation of capacitive response was measure as 6.7%. Even
though the proposed tactile sensor modules need more optimiza-
tion in the design and fabrication to increase the uniformity, they
can be a good candidate for touch screens for exible display in the
future with their exibility, transparency, and capability for force
sensing.
Acknowledgements
This work was supported by the IT R&D program of MKE
[2009-F-024-02, Development of Mobile Flexible IOP Platform], the
National Research Foundation of Korea (NRF) grant funded by the
Korea government (MEST) [2009-0076641] and a research grant
from Sogang University in 2008.
References
[1] R.S. Cok, R.R. Bourdelais, C.J. Kaminsky, Flexibleresistivetouchscreen, USPatent
2004/0212599 A1 (2004).
[2] P.W. Kalendra, W.J. Piazza, Automatic calibration of a capacitive touch screen
used with a xed element at screen display panel, US Patent 5283559 (1994).
[3] R.W. Doering, Infrared touch panel, US Patent 4868912 (1989).
[4] R. Adler, P.J. Desmares, An economical touch panel using SAWabsorption, IEEE
Trans. Ultrason. Ferroelectr. Freq. Control 34 (1987) 195201.
[5] H. Philipp, Capacitive sensor and array, US Patent 6452514 (2000).
[6] S. Hotelling, J.A. Strickon, B.Q. Huppi, Multipoint touchscreen, US Patent
2006/0097991 (2006).
[7] J.Y. Han, Low-cost multi-touch sensing through frustrated total internal reec-
tion, UIST05, October 2326, 2005, pp. 115118.
[8] P. Mach, S.J. Rodriguez, R. Nortrup, P. Wiltzius, J.A. Rogers, Monolithically
integrated, exible display of polymer-dispersed liquid crystal driven by
rubber-stamped organic thin-lm transistors, Appl. Phys. Lett. 78 (2001)
35923594.
[9] H.-K. Lee, S.-I. Chang, E. Yoon, A exible polymer tactile sensor: fabrication and
modular expandability for large area deployment, J. Microelectromech. Syst.
15 (2006) 16811686.
H.-K. Kim et al. / Sensors and Actuators A 165 (2011) 27 7
[10] J. Engel, J. Chen, C. Liu, Development of polyimide exible tactile sensor skin, J.
Micromech. Microeng. 13 (2003) 359366.
[11] M.E.H. Eltaib, J.R. Hewit, Tactile sensing technology for minimal access
surgerya review, Mechatronics 13 (2003) 11631177.
[12] M.H. Lee, H.R. Nicholls, Tactile sensing for mechatronics: a state of art survey,
Mechatronics 9 (1999) 131.
[13] T. Hoshi, H. Shinoda, A sensitive skin based on touch-area-evaluating tactile
elements, in: Proceedings of the Symposium on Haptic Interfaces for Virtual
Environment and Teleoperator Systems, Alexandria, 2006, pp. 8994.
[14] D.C. Ruspini, K. Kolarov, O. Khatib, The haptic display of complex graphical
environments, in: Proceedings of the 24th Annual Conference on Computer
Graphics and Interactive Techniques, 1997, pp. 345352.
[15] M. Shimojo, A. Namiki, M. Ishikawa, R. Makino, A tactile sensor sheet using
pressure conductive rubber with electrical-wires stitched method, IEEE Sens.
J. 4 (2004) 589595.
[16] B.J. Kane, M.R. Cutkosky, G.T.A. Kovacs, A traction stress sensor array for use
in high-resolution robotic tactile imaging, J. Microelectromech. Syst. 9 (2000)
425434.
[17] J. Dargahi, N.P. Rao, S. Sokhanvar, Design and microfabrication of a hybrid
piezoelectric-capacitive tactile sensor, Sens. Rev. 26 (2006) 186192.
Biographies
Hong-Ki Kimreceived his BS degree in Physics fromChungnamNational University
in 2007, and his MS degree in Electronic Engineering from Sogang University in
2009. He joined the LATTRON Co., Ltd, Korea in 2009. His research area includes
Bio-MEMS and Ceramic Device.
Seunggun Lee received his BS degrees in Electronic Engineering from Sogang Uni-
versity in 2009. He is currently pursuing his MS degree in Electronic Engineering
from Sogang University. His research area includes Touch Sensors.
Kwang-Seok Yun received his BS degree in Electronics Engineering from Kyung-
pook National University in 1996, MS and PhD degrees in Electrical Engineering
and Computer Science from Korea Advanced Institute of Science and Technology
(KAIST) in 1997 and 2002, respectively. He was a post-doctorial researcher at Uni-
versity of California, Los Angeles from 2005 to 2007. He joined the Department of
electronic Engineering at Sogang University, Korea in 2007, where he is now an
Assistant Professor. His current research area includes micro total analysis systems,
Lab-on-a-chip, MEMS, and micro sensors and actuators.