Appl. Phys.

B 60, 405 407 (1995)

Applied P csB

and Optics

Springer-Verlag 1995

Rapid communication

Wavelength dependence of the phase retardation of a quarter-wave plate

Abstract. We describe an optical heterodyne polarization interferometer that can be efficiently used for the precision measurement of the change in the state of polarization of a light wave induced by polarizing optical devices. This technique is used to measure the change with wavelength of the phase retardation of a quarter-wave plate. A theoretical derivation is presented to permit computation of the wavelength dependence of the phase shift induced by a quarter-wave plate. PACS: 42.25. - p; 42.79.Ci

Heterodyne interferometry can be efficiently used to measure the phase with a high degree of experimental accuracy [1]. This is accomplished by transferring the phase from the optical frequency co to a much lower frequency [2-5]. The phase measurement ultimately relies on the measurement of the phase difference between the output signals of two photodetectors which can be measured electronically using a vector voltmeter and/or oscilloscope, thus overcoming the limitations of traditional optical fringe interferometry. This technique can be used to study the overall change in the state of polarization of a light wave upon passing through a transmission-type device. In the present paper, we describe a heterodyne-type Mach-Zehnder polarization interferometer that permits the measurement of the State of Polarization (SOP) parameters of polarizing optical devices. This technique permits us the measurement of the phase retardation of a quarter-wave plate at various wavelengths. A theoretical analysis is presented that permits the determination of the wavelength dependence of the SOP parameters of a quarter-wave plate.

(2 = 632.8 nm) and a linearly polarized argon-ion laser as the light sources. The argon laser produces three wavelengths, 2 = 488 nm, 2 = 496.5 nm and 2 = 514.5 nm. The Acousto-Optic Modulator (AOM) provides two beams. One of these beams, the signal beam at frequency co, is normally incident on a Glan-Thompson polarizer (GT) which polarizes the transmitted beam at an azimuth of 45 to the horizontal x-axis. It then passes through a Quarter-Wave Plate (QWP) (design wavelength 2 = 514.5 nm), mounted on a Motorized Rotary Stage (MRS). The second beam is the reference at a frequency co + Am, (Am = 40 MHz) which is also. linearly polarized at 45 to the horizontal. Both beams combine at the Beam Splitter BS to obtain the heterodyne optical beam. A broad-band Polarizing Beam Splitter (PBS) divides the incident light into p and s waves, which are detected by two silicon photodiode/amplifiers, A and B, respectively. The two photodetectors are connected to a two-channel vector voltmeter to measure the beat photocurrents and the phase difference between the two signals. The accuracy of the phase measurement we thus obtain is of the order of 0.1 .

2 Theory To discuss the principle of the measurement, let us assume that the p-wave is polarized along the X-axis and the s-wave along the Y-axis. The interference at the BS is given by:

Ix = eRr + Isx + 2~R~IsxCOs(Acot + 0s()- 0R(]),

(la) (lb)

1 Experimental setup
The optical heterodyne system is schematically illustrated in Fig. 1. We have used a linearly polarized HeNe laser

In (la) and (lb), Isx and Isx are the intensities of the p-waves of the reference and signal beams, respectively, IR, and Is~ are the corresponding intensities of the s-waves (o) (o) (o) and 0s() - 0Rx and 0s, 0Ry are the phase differences for the p- and s-waves. From (la) and (lb), the ratio SB/Aof the beat intensities at the beat frequency Am and the phase difference 0 of channel B with respect to channel A, as measured by the
-

406
100

~ = 488nrn = 496.5n'm

X= 514.5n m - ~ 3.=632.8nmy

GT MRS ^ LJ" QWP

-100

,50

100

150

0
Fig. 1. Experimental arrangement

[Degreel

Fig. 2. Phase lag as function of the rotation angle of the quarterwave plate under test. The experimental and theoretical computed values are represented by dots and ,full curves, respectively

vector voltmeter, are given by

Ly

&/A(6) = / ~

N/ IR~ Is~

= Iz~)11XsC> l. (6)

(2a)

1.6
l& 1,2

= arg [-;~{s (3)] - arg (ZCR)). )

(2b)
O.

We have defined Z(R = E ()/E()R, and Zs = E()/E,sx as the ) / Rx ~) ()s / ratio of the complex phasor c o m p o n e n t s of the reference and signal b e a m s respectively, .(o) ~UR~- ~(o) = a r g " (o), ~URx LZR ), ~o) _ ~s~O)= arg(z~o))and 6 is the wavelength-dependent phase retardance of the Q W P under test. Equations (2a) and (2b) express the simple fact that the S O P p a r a m e t e r s of the signal b e a m are completely determined by the relative amplitude and the relative phase of one oscillating c o m p o n e n t with respect to the other orthogonal c o m p o n e n t . The relationship between the input Z{s and the output i) X{s ellipses of polarization on the signal a r m of the inter) ferometer is described by the Polarization Transfer Function (PTF) of the polarizing optical device [-4]:

0.6

75

100

125

150

t7g

0
1.8 b 1.6 1.4

[Degree]

.--"~*------- ~.= 496.5nm

1 o,8

Z(s o)
-

T22X(si) q- T21 T12Z{si) -.}- T11 '

0.6

(4)

~5

50

7S

100

125

150

175

[Degree]

where T u (i,j = 1, 2) are the Jones-matrix elements of the optical system to be measured. E q u a t i o n (4) is valid in the absence of nonlinearity and other frequency-changing processes. In order to examine the effect of the wavelength changes on the measured relative phase ~, let us consider that in our experimental conditions )~(R =Z~Sl)= 1, ) a r g 0 ~ )) = 0. F u r t h e r m o r e , the Cartesian Jones matrix of a linear retarder whose fast axis is inclined at an azimuth 0 to the Y-axis is given by [-4]

Fig. 3a, b. Ratio of beat intensities as a function of the rotation angle of the quarter-wave plate under test. The experimental and theoretical computed values are represented by dots and full curves, respectively; (a): 2 = 488nm, 2 = 514.5nm; (b): 2 = 496.5 nm, 2 = 632.8 nm

By using (4) and (5) together with (2a) and (2b), we determine S~/A and ~ as a function of the wave-plate azimuth angle 0 and phase retardance 6 sin 6 cos 20 6 2sin 2 ~ cos 2 20

T = ( cOs2Oei6/2 q- sin20e-ia/2

isin20sin 6/2

isin 20sin6/2 "] sin2 0e ia/2 + cos2Oe-ia/2]"

~(0, 6) = arctan
1 -

(6)

(5)

407
105

9o

fitted values of ~(0, 6) and SB/A(O, 6) calculated from (6) and (7), respectively, are represented by the full curves in Figs. 2 and 3. Figure 4 shows the values of the plate retardation as a function of the wavelength. The full line is the result of the linear fit of the plate retardation 6 in the considered wavelength range. The rate of change of retardation d6/d2 with wavelength for our wave plate is found to be -0.18/nm.
4 Conclusion
475 525 575 625

[~]
Fig. 4. The change in wavelength of the quarter-wave plate retardation 6. Measured values of retardation (data points); linear fit of

(full curve)
(1 - sin 2 40 sin 4 6/2) 1/2 1 + sin40 sin 2 6/2

SB/A(O, 6) =

(7)

We have described the principle of an optical heterodynetype Mach-Zehnder detection interferometer based on polarization to measure the SOP parameters of polarizing optical devices. We have used this technique for the measurement of the wavelength dependence of the quarter-wave-plate retardation. We point out that this optical heterodyne technique can be used to measure with a high degree of experimental accuracy the phase retardation induced by a cascade of N polarizing optical devices which process the same optical wavelength, or the phase retardation of a wave plate over a variety of wavelengths.
References

3 Experimental results

The Q W P was rotated at different angles 0 and the corresponding SOP parameters were measured. Figure 2 shows the measured values of the phase at each wavelength. Figures 3a, b show the experimental data of the ratio of the beat intensities as a function of the wave-plate rotation angle. The experimental data of the phase ~(0, 6) and SB/A(O, g~) were fit by (6) and (7) and the values of the plate retardation 6 were computed at each wavelength. The

1. 2. 3. 4. 5. 6.

R. Calvani, R. Caponi, F. Cisternino: Opt. Commun. 54, 63 (1985) N.A. Massie, R.D. Nelson, S. Holly: Appl. Opt. 18, 1797 (1979) R.N. Shagam, J.C. Wyant: Appl. Opt. 17, 3034 (1978) L. Chin-hwa, C. Chien, C. Keh-su: Appl. Opt. 29, 5159 (1990) Y. Lin, Z. Zhou, R. Wang: Opt. Lett. 13, 553 (1988) R.M.A. Azzam, N.M. Bashara: Ellipsometry and Polarized Light (North-Holland, Amsterdam 1977) pp. 28-30