Measurement of change in refractive index in polymeric flexible substrates using wide field interferometry and digital fringe analysis

Gyanendra Singh and Dalip Singh Mehta*

Laser Applications and Holography Laboratory, Instrument Design Development Centre, Indian Institute of Technology Delhi, Hauz Khas, New Delhi 110016, India *Corresponding author: dsmehta@iddc.iitd.ac.in Received 17 July 2012; revised 15 October 2012; accepted 2 November 2012; posted 8 November 2012 (Doc. ID 172804); published 7 December 2012

Indium tin oxide coated polyethylene terephthalate (PET) polymeric films are widely used as substrates for future optoelectronic devices, such as organic LEDs, organic thin film transistors, and organic solar cells. These PET substrates are thin, flexible, and rugged. But residual stresses are trapped in polymeric substrates due to their manufacturing process, and this leads to the birefringence in flexible displays. In this paper we report the measurement of the change in refractive index of PET substrates using Mach Zehnder interferometry and the Fourier transform fringe analysis technique. Change in refractive index was observed by means of bending the PET substrate. This change in birefringence varies the optical path difference between the two arms of the interferometer, leading to the fringe shift. From the fringe shift the phase change was extracted as a function of bending, and the change in the refractive index was determined experimentally for two wavelengths, i.e., red and green color lasers. We found that the value of change in the refractive index of these substrates increases on bending of the substrates. 2012 Optical Society of America OCIS codes: 120.2650, 120.3180, 120.5050.

1. Introduction

Displays are now an integral part of our daily life, as they are used in our homes, workplaces, vehicles, etc. [1]. Over the last few years display devices have been fabricated using liquid crystal technologies that include twisted nematic LCDs [2], cholesteric LCDs [3], and polymer-dispersed liquid crystal displays [4]. But all these displays are fabricated on glass substrates, which are breakable and bulky. Recently there has been great progress in fabricating display devices and many other optoelectronic devices onto flexible substrates. Flexible electronics and photonics is the future concept for display applications.
1559-128X/12/358413-10$15.00/0 2012 Optical Society of America

Flexible substrates are being widely used for emissive displays including organic light-emitting diodes (OLEDs) [5], electroluminescent displays [6], fieldemitting diodes [7], organic solar cells [8], and even flexible plasma displays [9]. One of the most fundamental requirements for flexible electronics is the availability of a good substrate for device fabrication that doesnt change its properties while bending. The most important physical property of these polymeric substrates is their flexibility in the complete operating device. Other important properties are their low volume, light weight, easy scalability in size, robustness, and low cost [10]. These polymeric substrates are made up of thin plastics substrates, like polyethylene terephthalate (PET), polycarbonate, and polyethylene sulfone polyester films [11,12]. One of the major advantages of using polymer sheets as
10 December 2012 / Vol. 51, No. 35 / APPLIED OPTICS 8413

the substrate is that we can easily attain the desired thickness and shape of the substrate by the stretching process and mechanical stress [10]. But there is a problem of birefringence related to polymeric substrates, which is the difference in the refractive index in two orthogonal directions [13], which might change their optical properties during the operation of the devices. The birefringence in PET substrates originates during the fabrication process, in which the mechanical stress causes the molecular reorientation of polymeric materials leading to structural anisotropy and to birefringent optical properties i.e., the refractive indices in different directions [14]. Optical birefringence is a desired property in many applications, like polymers of different types and materials being used for the fabrication of linear and circular polarizers, pseudodepolarizers, and phase plates with tailored characteristics for working at different spectral ranges [15,16]. Tuning of the polymer film can be done during the production process, but the unwanted residual stress that remains inside the substrates degrades the optical quality of the substrates [17]. Birefringence is always present for plastic substrates because of the residual stress, while the bending of these substrates leads to an additional controllable birefringence due to the external applied stress. Further, during the operation of optoelectronic devices fabricated on PET substrates, the temperature may change the birefringence. Therefore, it is important to study the change in the refractive index by means of bending the substrates over the large area. Birefringence studies of polymer substrates have been done using various techniques, e.g., Stchakovsky et al. presented polarimetric characterization of optically anisotropic substrates by using phase modulated spectroscopic ellipsometry as well as liquid crystal Mueller matrix polarimetry [10]. Van Horn and Winter [18] determined the birefringence and orientation in biaxially stretched polymer films and sheets using the conoscopic measurement technique. Medhat et al. [19] obtained fringes of equal tangential inclination by curve-induced birefringence in a Fortypan photographic plate. El-Dessouki et al. [20] used double-exposure speckle photography for measuring the birefringence of a curved sheet, and Russo [14] gave an alternative method for the determination of birefringence in stretched polymeric substrates by determining the phase difference using three different laser wavelengths and a tilting compensator. Shabana and Jaleel [21] studied the optical properties of highly birefringent polymer films using different interferometric techniques. Among all the above-mentioned techniques, optical interferometry is one of the best methods for the study of birefringence, as it can most precisely calculate the optical path difference over the entire visible spectrum for which optical parameters can be determined. In this paper we report the study of the optical birefringence of a commercially available indium tin
8414 APPLIED OPTICS / Vol. 51, No. 35 / 10 December 2012

oxide (ITO)-coated PET (ITO/PET) substrate used in flexible electronics (photonics) by MachZehnder interferometry (MZI). The change in refractive index was observed by means of bending of the PET substrate for two different monochromatic wavelengths (red and green lasers). Fourier transform fringe (FTF) analysis was used to reconstruct the phase maps of the interferograms recorded at different amounts of bending of the substrates. From this phase change, the change in refractive index was calculated. In the following section the theoretical background is given, followed by experimental details and results analysis.
2. Theoretical Background

The two-dimensional (2D) FTF analysis technique has been regarded as one of the best techniques, being fast and reliable for extracting the required phase information. Fourier fringe analysis has undergone several modifications over time for improving the quality of result interpretation. The interference fringes recorded by CCD array are displayed on the computer screen. The equation for the intensity of fringes recorded by detector array is in a digitized form that can be expressed as a 2D sinusoidal intensity distribution I x; y by the following expression [22,23]: I x; y I R x; y I O x; y p 2 I R x; y I O x; y cos 0 x; y;

(1)

where I R x; y and I O x; y are the intensities of light coming from the reference and object arms of the MZI, respectively, and 0 x; y is the 2D phase distribution of the interferogram. In general, for the FTF analysis a 2D fringe pattern can be expressed as hx; y ax; y bx; y cos2 f x; y x; y; (2) where ax; y is the variation of background illumination (irradiance variations arising from nonuniform light reflection or transmission), bx; y is the amplitude modulation of the fringes, f x; y is the spatial carrier frequency, and x; y is the phase of the interference fringes. The input fringe pattern given in Eq. (2) can be rewritten in the following form for the convenience of explanation [22]: gx; y ax; y cx; y exp2 if 0 cx; y exp 2 if 0 ; with 1 cx; y bx; y expix; y; 2 (4) (3)

where  denotes the complex conjugate. Taking the Fourier transform of Eq. (3) with respect to x, the expression for the Fourier spectrum is shown below:

H f x ; y Af x ; y Cf x f 0 ; y C f x f 0 ; y; (5) where the capital letters denote the Fourier spectra, f 0 is the initial spatial carrier frequency, and f x is the spatial carrier frequency with a variable x. The Fourier spectra in Eq. (5) are separated by the carrier frequency f 0. We make use of either of the two spectra on the carrier, say Cy; f x f 0 , and translate it by f 0 on the frequency axis toward the origin to obtain Cf 0 ; y. Note that the unwanted background variation ax; y has been filtered out in this stage. By computing the inverse Fourier transform of Cf ; y with respect to f , we obtain Cx; y using fast Fourier transform (FFT) again. The 2D phase map of the interferogram can be computed by using following expression:  x; y tan
1

Fig. 1. Schematic diagram of the bending of the ITO/PET substrate.

 ImCx; y : ReCx; y

R (6)

4F 2 C2 : 8F

(9)

From the phase maps corresponding to interferograms recorded at different amounts of bending of the ITO/PET substrate, we can calculate change in phase, which is related to the corresponding change in refractive index of the ITO/PET substrate. The change in phase of interferograms with bending can be expressed by the following expression: x; y 2 2 OPD nx; ydx; y; i i (7)

where n x; y is the change in refractive index of the ITO/PET substrate, which is a function of the amount of bending, i (i 1, 2 corresponding to red and green colors, respectively) is the wavelength of laser light, and dx; y is the geometrical thickness of the PET substrate, which is constant. In Eq. (7), i and dx; y are known constants, and hence by measuring the phase difference with different amounts of bending using Eq. (7), one can determine the change in the refractive index of the ITO/PET substrate using the following expression:   i x; y nx; y : 2 dx; y (8)

Here R, F , and C show the radius of curvature, out-of-plane displacement, and chord length, respectively. The calculated radius of curvature for different amounts of bending of the PET substrate was 5, 3, and 2.5 cm [24]. In order to determine the value of radius of curvature R the distances C and F have been calculated. An MZI was used for recording the fringes using the ITO/PET substrate as a test plate. Two spatially separated equivalent paths are present in the MZI, which is used for the measurement of the optical phase shift. The optical path length is changed by means of introducing the test plate in one arm, called the object arm, of the interferometer. For constructive interference, the phase difference is an even multiple of , and for destructive interference, it is an odd multiple of . The phase difference is given by Eq. (7). Through bending, stress is applied to the ITO/PET substrate, which results an additional controlled birefringence along with the residual birefringence (due to the manufacturing process) of the substrate. The change in curvature due to bending of the ITO/PET substrate is shown in Fig. 2, which depicts the presence of two types of stresses, i.e., tension and compression stress toward the convex and concave sides

The change in the refractive index is valuable for determining the properties of ITO/PET substrates. As the refractive index is the fundamental property of the material, any change in it leads to a change in the final devices [OLEDs, organic thin film transistors (OTFTs), and organic solar cells] fabricated using these substrates. Figure 1 shows the schematic diagram of the variation of the radius of curvature of an ITO/PET substrate on increasing the bending. The radius of curvature has been calculated by using simple trigonometry (Pythagorean theorem), and the relation is expressed by

Fig. 2. (Color online) Schematic diagram of the ITO/PET substrate in bending form, illuminated by the collimated monochromatic (red or green) light. 10 December 2012 / Vol. 51, No. 35 / APPLIED OPTICS 8415

of the bending, respectively. The magnitude of these stresses varies according to the thickness of the film. Uniform additional birefringence is present on the substrate due to the presence of these stresses [19]. The optic axis of the substrate changes according to the bending curvature, and it always makes the tangent to the circle, as shown in Fig. 2. Therefore, the optic axis is direction dependent rather than varying in a straight line. As the stress applied to the substrate increases due to bending, that leads to a decrease in radius of curvature r. Without bending, the radius of curvature can be assumed to be r . For a very thin substrate, the radius of curvature for both faces of the bended substrate can be considered nearly equal. Under these circumstances the interference fringes are symmetrical around the axis OX. Then the distance of the fringes from both side of the symmetric axis is the same and can be called the diameter of the fringe 2Z. From Fig. 2, the angle of incidence can be given by [19] r d cos p ; r d2 Z2 (10)

q dr d r d2 Z2 m n:

(12)

Since Z varies with angle , which is the angle between the direction of the incidence beam and the radial direction, the relation between the order of the fringe and square of the radius, i.e., Z2 , can be given by   2r d2 m: (13) Z2 dn More details about the derivation of Eq. (13) can be found in [19]. Equation (13) gives the linear relation between square of radius Z and order of the fringe, while the value of the change in refractive index (n) varies directly with the curvature (1r). Equation (13) is used for calculating the birefringence of the ITO/PET substrate using fringes of equal tangential inclination by Medhat et al. [19] and El-Dessouki et al. [20]. We have used the full field phase measurement method to determine the change in n using Eq. (7). We compared our results with the theoretical model proposed by Medhat et al. and El-Dessouki et al., and we found close agreement with the theory.
3. Experimental Details

where Z is the distance between symmetric axis OX and the fringe, which is the radius of the fringe (PQ), and d is the thickness of the substrate. From RSQ, cos dl, where l is the thickness along the path of incident light. Putting the value of cos in Eq. (10), we can get l dr dr d2 d2 12 . The change in interference pattern occurs due to the variation of the value of l. When the phase difference is an even multiple of 2, the constructive interference takes place for which [19] nl m ; (11)

where m is the integer order and is the fractional order of the interference fringes. Substituting the value of l in Eq. (11), we get [19]

A birefringence study of the ITO/PET (surface resistivity is 35 sq, D is 5 mil, and transmittance is 550 nm, >86%, product no. 639311, Sigma Aldrich, USA) substrate was done using the optical interferometric technique, for which an MZI was used, which is shown in Fig. 3. A green diode pumped solid state laser with a wavelength of 1 532 nm and a red HeNe laser with a wavelength of 2 632.8 nm, each with 10 mW of power, were mixed at a beam splitter (BS). Both beams were passed through a beam expander (BE) and a spatial filter (P) unit simultaneously. A collimating lens (CL) of focal length 24 cm, diameter 30 mm, was used to collimate the beams. The collimated beams were then made

Fig. 3. (Color online) Schematic diagram of a MachZehnder interferometer used for the study of the ITO/PET substrate. 8416 APPLIED OPTICS / Vol. 51, No. 35 / 10 December 2012

incident on a beam splitter (BS1 ), and were allowed to pass through the aperture (A). The collimated light was divided into two beams by BS1 (5050), and the two resulting beams (the sample beam and the reference beam) were reflected by mirrors M1 and M2 , respectively. Two beams then passed through the second beam splitter BS2 (5050), and interference fringes were observed and recorded by a color CCD camera interfaced with computer. The ITO/PET substrate used for the interferometric study was placed between one of the arms of the MZI, and interferograms were recorded for the different amounts of bending of the substrate. The thickness of the ITO coated substrate is 0.129 mm, which was measured by screw gauge and also was obtained from the manufacturer. A pictorial representation of the bending given to the PET film substrate is shown in Fig. 2. The experiment was carried out by the following procedure. The ITO/PET substrate was placed in one arm of the MZI, and interferograms were recorded for different amounts of bending of the PET substrate for dual wavelengths (1 532 nm and 2 632.8 nm). First a green laser was switched on, and the ITO/PET without any bending was placed in one of the arms of the MZI (as shown in Fig. 1) and the interference fringe was recorded. The green laser was blocked and a red laser was switched on, and the interferogram in the presence of the ITO/PET substrate was recorded under identical conditions as the sample. Bending was applied to the ITO/PET substrate, and the interferograms for the green and red lasers were recorded sequentially by means of switching on one laser and blocking the other laser and vice versa. The bending of the ITO/PET substrate was further increased and the experiment was repeated, and interferograms for the red and green lasers were recorded.
4. Results and Discussion

Figures 4(a)4(d) show the interferograms recorded when the green laser was switched on with

increasing bending of the ITO/PET substrate. Figures 4(e)4(h) show similar results with the red laser switched on with increasing bending of ITO/ PET substrate. The interferograms were analyzed by the FTF analysis method, which we have discussed in the theory section. First the 2D FFT of each interferogram was computed, and the first-order spectrum of each was filtered out and inverse fast Fourier transform (IFFT) was computed. A 2D wrapped phase map of each interferogram was obtained, and the phase map was unwrapped using a MATLAB program. The pictorial representation for the analysis of the interferograms is shown in Figs. 5(a)5(h), using the interferograms shown in Figs. 4(d) and 4(a), respectively, as examples. A similar procedure was used for the phase map of rest of the interferograms. For the analysis of the interferograms shown in Fig. 3, we have used the area, which is subject to the maximum change in the fringe shape, having the dimensions 648 376 pixels. A similar procedure was followed for computing the phase map of a reference interferogram. The unwrapped phase map of the object [Fig. 5(d)] is subtracted from the reference phase map [Fig. 5(h)]. Figure 6(a) shows the subtracted phase map. From this phase map the change in the refractive index profile was calculated using Eq. (8), and the result is shown in Fig. 6(b). From this figure, it can be seen that the change in the refractive index is at its maximum at the center of the ITO/PET substrate. Analysis for all the interferograms shown in Fig. 4 has been done by using the Fourier transform method for two wavelengths. Figure 6 shows the analysis for the 1 532 nm, where gray level images of the recorded interferograms are used. In this study we have used Fig. 7(a) as the reference interferogram recorded without bending of the ITO/PET substrate. The interferograms shown in Figs. 7(b)7(d) are recorded with bending of the ITO/PET substrate. The wrapped phase maps for the interferograms in Figs. 7(a)7(d) are shown in Figs. 7(e)7(h). The unwrapped phase

Fig. 4. (Color online) Optical interferograms recorded for different amounts of bending of the ITO/PET substrate using the setup shown in Fig. 3 for dual wavelengths. 10 December 2012 / Vol. 51, No. 35 / APPLIED OPTICS 8417

Fig. 5. (Color online) (a) Object interferograms; (e) reference interferograms; (b), (f) Fourier spectra; (c), (g) wrapped phase maps; (d), (h) unwrapped phase maps.

maps are calculated from the wrapped phase maps, shown in Figs. 7(e)7(h), using the MATLAB code. The reference unwrapped phase map is subtracted from the object phase maps one by one, and the calculated phase difference is shown in Figs. 7(i)7(k). Figures 7(i)7(k) represent the phase difference for the subtraction of the unwrapped phase map corresponding to the reference interferogram given in Fig. 7(a) and the object interferograms shown in Figs. 7(b)7(d). Changes in the refractive index corresponding to the phase differences represented by Figs. 7(i)7(k) are shown below in Figs. 7(l)7(n), which are calculated by using Eq. (8). The value for the change in refractive index is given in Table 1. The same analysis was done for the interferograms recorded using the red laser having wavelength 2 632.8 nm, which is pictorially indicated in Fig. 8.

The change in the refractive index for the given bending of the ITO/PET substrate is given in Table 1 for both red and green wavelengths. We have noticed that on increasing the bending of the substrate, the change in refractive index also increasing. The change in refractive index is very close for the same bending given for the two wavelengths. We know that plastics are made up of long chains of molecules (hydrocarbons) known as polymers. We have bended the plastic substrate for our experiment, which is pictorially represented in Fig. 1. The bend shown in Fig. 1 depicted that the ITO/PET substrate from inside the curved surface is compressed and stretched from the outside. The cross section of the ITO/PET substrate remains unchanged during bending, because when it compresses from the inside, the material expands sideways [25]. When the red

Fig. 6. (Color online) (a) Phase difference of unwrapped phase maps shown in Figs. 5(d) and 5(h) and (b) 2D change in refractive index calculated from phase difference. 8418 APPLIED OPTICS / Vol. 51, No. 35 / 10 December 2012

Fig. 7. (Color online) Interferograms for wavelength 1 532 nm (a) without bending and (b)(d) with gradually increasing bending; (e)(h) the corresponding wrapped phase maps for the recorded interferograms (a)(d); (i)(k) the phase differences for the unwrapped phase maps with respect to the reference interferograms; (l)(n) the changes in refractive index corresponding to the phase differences shown in (i)(k).

and green laser light passes through the ITO/PET substrates, we observe the phase change due to the bending, which leads to the change in refractive index. The phase change caused by bending the substrate increases the optical path difference, and this additional path difference is added to the phase
Table 1. Change in Refractive Index by Applied Bending to ITO/ PET Substrate for Dual Wavelength

S. No. 1 2 3

Change in Refractive Index (1 532 nm) G1 -O 0.1531 103 G2 -O 0.4387 103 G3 -O 1.047 103

Change in Refractive Index (2 632.8 nm) R1 -O 0.1334 103 R2 -O 0.3485 103 R3 -O 0.876 103

change. Stress induced alignment (anisotropy) of the molecules that compose the PET substrate is another reason behind this change in the refractive index. Stressing in an arbitrary direction leads to a random orientation of the polymer molecule in the PET substrate, and the cumulative effect of the ITO thin film due to the bending is also a reason behind the change in refractive index. The change in the refractive index is due to the strained polymer configuration, which causes the change in the birefringence. However, the phase change is also wavelength dependent. Birefringence is related to dispersion, so the change in the refractive index of the PET substrate is related to the wavelength of the light passing through the sample.
10 December 2012 / Vol. 51, No. 35 / APPLIED OPTICS 8419

Fig. 8. (Color online) Interferograms for wavelength 1 632.8 nm: (a) without bending and (b)(d) with gradually increasing bending; (e)(h) the corresponding wrapped phase maps for the recorded interferograms (a)(h); (i)(k) the phase differences for the unwrapped phase maps with respect to the reference interferograms; (l)(n) the changes in refractive index corresponding to the phase differences shown in (i)(k).

As ITO is an optically transparent film throughout the visible spectrum, the refractive index of these films is modeled by the first-order Sellmeier dispersion relation, where a, b, and 0 are constants [26]: n2 1 a b2 2 2 0 : (14)

absolute values of n0 and ne are not required for the calculation of n [26]: n ne no n C2 : (16)

Under standard assumptions, the Sellmeier formula reduces to the Cauchy relation, where A and B are the constants for the given PET substrate [26]: n A B2 : (15)

The dispersion of the birefringent film is given by the following equation, where n is the birefringence at a long wavelength and C is a constant, while
8420 APPLIED OPTICS / Vol. 51, No. 35 / 10 December 2012

According to normal dispersion, by increasing the wavelength, the refractive index decreases. With the above explanations, we can say that the change in the refractive index is the phenomenon of both the bending of the ITO/PET substrate and the wavelength of the laser light used for recording the interferograms, which leads to a change in the phase difference and refractive index. We have observed that the changes in the refractive index for two different wavelengths increase with an increase in the bending of the ITO/PET substrate. The order of change in the

refractive index is nearly same for both wavelengths. This change in refractive index (birefringence) on bending the plastic substrates used for organic electronics and photonics may lead to further changes in various electro-optical properties of the devices, such as conductivity, optical transmission, and the spectrum of light. The present study reported in this paper may give some insight into these properties. The graph of changes in the refractive index for the different bending curvatures of the PET substrate is plotted against the reciprocal of the radius of curvature, which is shown in Fig. 9. From the graph it is clear that the change in the refractive index is inversely proportional to the bending curvature applied to the PET substrate. The change in the refractive index versus reciprocal of radius of curvature for the green and red wavelengths is shown by the green (with squares) and red (with circles) curves, respectively. These results confirm the theoretical model given in Section 2. Using the present method we can determine the n at each pixel. Our method is based on the measurement of the phase change due to the change in the refractive index. Therefore, the accuracy of the determination of change in the refractive index depends on the phase measurement. The error analysis is done by using Eq. (8), and the maximum error for calculating the value of n is given by n d ; n d (17)

ogram. We have taken the average of and n for all the bending curvatures to calculate the maximum error. For green and red wavelengths, the calculated maximum error is 4.29 106 and 3.589 106 , respectively.
5. Conclusions

In conclusion, a commercially available ITO/PET substrate has been investigated for different amounts of bending using a MZI and the Fourier fringe analysis technique. In this study we have found that the refractive index of the PET substrate increases with greater bending. This change of refractive index may lead to changes in other optical properties of the flexible substrate, such as the transmission, the spectrum profile of the light passing through it, and the conductivity of the devices that are fabricated on this substrate. This study helps us to signify the degree of bending that can be applied to flexible substrate based devices so that their spectrum profile may not change drastically, as the emission spectrum is a very important parameter of display devices such as OLEDs and OTFTs. G. Singh is thankful to UGC [F.No. 10-2(5)/2007(i)E.U.II], New Delhi, for financial assistance. The authors are also thankful to CSIR for financial grant no. 03(1147)/09/EMR-II.
References
1. D. R. Cairns and G. P. Crawford, Electromechanical properties of transparent conducting substrates for flexible electronic displays, Proc. IEEE 93, 14511458 (2005). 2. T. J. Gardner and R. P. Wenz, STN alignment on rib-spaced plastic LCD substrate, in Society for Information Display 1995 Digest of Technical Papers (Society for Information Display, 1995), Vol. 16, pp. 695698. 3. J. L. West, V. H. Bodnar, Y. Kim, and H. Wonderly, Multi-color, cholesteric displays using plastic substrates, in Proceedings of the Sixth International Display Workshops (Society for Information Display, 1999), pp. 335338. 4. C. D. Sheraw, L. Zhou, J. R. Huang, D. J. Gundlack, T. N. Jackson, M. G. Kane, I. G. Hill, M. S. Hammond, J. Campi, B. K. Greening, J. Francl, and J. W. West, Organic thin-film transistor-driven polymer dispersed liquid crystals, Appl. Phys. Lett. 80, 10881090 (2002). 5. A. N. Krasnov, High-contrast organic light-emitting diodes on flexible substrates, Appl. Phys. Lett. 80, 38533855 (2002). 6. A. B. Chwang, M. A. Rothman, S. Y. Mao, R. H. Hewitt, M. S. Weaver, J. A. Silvernail, K. Rajan, M. Hack, J. J. Brown, X. Chu, L. Moro, T. Krajewski, and N. Rutherford, Thin film encapsulated flexible organic electroluminescent displays, Appl. Phys. Lett. 83, 413415 (2003). 7. O. J. Lee and K.-H. Lee, Fabrication of flexible field emitter arrays of carbon nanotubes using self-assembly monolayers, Appl. Phys. Lett. 82, 37703772 (2003). 8. T. Aernouts, P. Vanlaeke, W. Geens, J. Poortmans, P. Heremans, S. Borghs, R. Mertens, R. Andriessen, and L. Leenders, Printable anodes for flexible organic solar cell modules, Thin Solid Films 451, 2225 (2004). 9. C. A. Wedding, J. W. Guy, O. M. Strbik, D. K. Wedding, R. P. Wenzlaff, and W. W. Olsen, Flexible AC plasma displays using plasma spheres, in Society for Information Display Digest of Technical Papers Vol. 35 (Society for Information Display, 2004), pp. 815818. 10. M. Stchakovsky, C. Caillaud, M. Foldyna, R. Ossikovski, and E. G. Caurel, Polarimetric characterization of optically anisotropic flexible substrates, Thin Solid Films 516, 14141418 (2008). 10 December 2012 / Vol. 51, No. 35 / APPLIED OPTICS 8421

where , , and d are the error in the calculation for wavelength of the light source, phase change by bending the substrate, and thickness of the PET film, respectively. The error in finding the value of 102 nm, 1 102 rad, and d 1 103 mm. Error analysis is done for both green and red wavelengths used for recording the interfer-

Fig. 9. (Color online) Graph of the change in induced birefringence by the bending curvature versus the reciprocal of the radius of curvature for the red and green wavelengths.

11. K. Noda, H. Sato, H. Itaya, and M. Yamada, Characterization of Sn-doped In2 O3 film on roll-to-roll flexible plastic substrate prepared by DC magnetron sputtering, Jpn. J. Appl. Phys. 42, 217222 (2003). 12. H. M. Kim, S. K. Jeung, J. S. Ahn, Y. J. Kang, and C. K. Je, Electrical and optical properties of In2O3ZnO films deposited on polyethylene terephthalate substrates by radio frequency magnetron sputtering, Jpn. J. Appl. Phys. 42, 223227 (2003). 13. M. F. Weber, C. A. Stover, L. R. Gilbert, T. J. Nevitt, and A. J. Ouderkirk, Giant birefringent optics in multilayer polymer mirrors, Science 287, 24512456 (2000). 14. R. Russo, An alternative method for the determination of the birefringence in stretched polymeric films, Polym. Testing 20, 283286 (2001). 15. A. M. Title, Improvement of birefringent filters. 2: achromatic waveplates, Appl. Opt. 14, 229237 (1975). 16. H. Fu, Z. Yan, and M. Mansuripur, Measurement of the wavelength dependence of birefringence for optical disk substrates, Appl. Opt. 33, 74067411 (1994). 17. H. Daly, K. Nguyen, B. Sanschagrin, and K. Cole, The buildup and measurement of molecular orientation, crystalline morphology, and residual stresses in injection molded parts: a review, J. Injection Molding Technol. 2, 5985 (1998). 18. B. L. Van Horn and H. H. Winter, Conoscopic measurement of birefringence and orientation in biaxially stretched polymer films and sheets, Macromolecules 36, 85138521 (2003).

19. M. Medhat, N. I. Hendawy, and A. A. Zaki, Fringes of equal tangential inclination by curvature-induced birefringence, Opt. Laser Technol. 35, 3135 (2003). 20. T. A. El-Dessouki, N. I. Hendawy, and A. A. Zaki, Measuring birefringence of curved sheet and single crystals by doubleexposure speckle photography, Opt. Lasers Eng. 47, 622628 (2009). 21. H. M. Shabana and A. A. Jaleel, Optical properties of highly birefringent polymer films using simple techniques, Int. J. Polym. Mater. 54, 10091018 (2005). 22. M. Takeda, Spatial-carrier fringe-pattern analysis and its applications to precision interferometry and profilometry: an overview, Ind. Metrol. 1, 7999 (1990). 23. M. Suematsu and M. Takeda, Wavelength-shift interferometry for distance measurements using the Fourier transform technique for fringe analysis, Appl. Opt. 30, 40464055 (1991). 24. H.-C. Chen, K.-T. Huang, and Y.-M. Lo, Measurement of residual stress for ITO/PET substrates by the double beam shadow moir interferometer, Appl. Opt. 51, 15661571 (2012). 25. G. P. Crawford, Flexible Flat Panel Displays (Wiley, 2005). 26. M. J. Escuti, D. R. Cairns, and G. P. Crawford, Opticalstrain characteristics of anisotropic polymer films fabricated from a liquid crystal diacrylate, J. Appl. Phys. 95, 23862390 (2004).

8422

APPLIED OPTICS / Vol. 51, No. 35 / 10 December 2012