To be published in Applied Optics: MeasurementofsurfaceprofileofAxiconlenswithapolarizationphase shiftingshearinginterferometer Authors: SanjibChatterjeeandY.Kumar Accepted: 13June2011 Posted: 17June2011 Doc.

ID: 148061 Title:

OSA

Published by

Measurement of surface profile of Axicon lens with a polarization phase shifting shearing interferometer
Sanjib Chatterjee* and Y. Pavan Kumar Raja Ramanna Centre for Advanced Technology Dept. of Atomic Energy, Indore-452013, India * Corresponding author: schat@rrcat.gov.in Abstract: A Twyman Green interferometer (TGI) based polarization phase shifting

OSA
presented. In this technique, the annular beam generated due to passing of an expanded collimated laser beam traveling along the axis of revolution of the transparent glass axicon element, is spilt up into reflected and transmitted components, having plane of polarizations in orthogonal planes, by the polarization beam splitter (PBS) cube of the TGI based optical setup. The spilt up components are made to travel unequal paths along the two arms of the TGI and are recombined by the PBS. Because of the difference in path lengths traveled by the annular conical beams, a linear shear is introduced along the radial direction between the interfering components. Thus the resulting interference pattern gives a map of optical path difference (OPD) between two successive close points along a radial direction on the conical surface of the axicon lens. The OPD map along radial directions and hence the slopes/profiles of the conical surface are obtained by applying polarization phase shifting interferometry. Results obtained for an axicon lens are presented. Optical Society of America OCIS codes: 120.0120, 120.3180, 120.5050, 120.6650 1

shearing interferometric technique for testing of conical surface of axicon lens is

Published by

1. Introduction J. H. McLeod defined axicon as a type of optical element which is a figure of revolution [1-3]. Axicon may be reflecting, transmitting or diffracting [4] and has the property that a point source on its axis is imaged to a range of points along its axis. This extended focal depth has been utilized in many applications, such as, high resolution optical coherence tomography [5], two photon excitation fluorescence microscopy [6], generation of laser produced plasmas [7], optical pumping of infrared laser systems [8] etc. Axicons have been applied for the production of radial and tangential polarization of light beams [9].

OSA
axicon mirror has been reported [10]. Axicons are very useful for generation of dark hollow beam [11]. Hollow laser beams generated using axicons has been applied for trapping of cold atoms [12]. Axicons have been widely used for the generation of quasi non-diffracting Bessel beams [13, 14]. The ring of coherent light formed by a combination of an axicon and a focusing lens has been utilized for trepanning [15]. It is important to characterize optical quality of axicon. A simple Twyman Green interferometer (TGI) based setup for testing cone angle, roundness and concentricity of axicons has been discussed in ref. [16]. Ref. [17] discussed a double-pass unequal path polarization TGI for the testing of wave front distortion produced by large aperture diameter non-linear reflaxicons. A Mach-Zehnder interferometer based testing setup has been used for testing surface form error of lens axicons as reported in [18]. In this method, the cone of light beam formed due to passing of expanded collimated laser beam through a test axicon lens is allowed to interfere with a plane reference wave. The distortion of the conical axicon surface is obtained from the asymmetry of the concentric 2

Laser acceleration of electrons using radially polarized CO2 laser beam produced by an

Published by

circular fringe system. Ref. [19] describes an interferometric fringe projection technique in which a fringe pattern produced by two interfering plane wave fronts is projected through the axicon. The distortion introduced by the conical surface of the axicon is obtained by applying phase shifting interferometry (PSI). Present paper describes a polarization phase shifting shearing interferometric technique for testing of conical surface of axicon lens. 2. Principle Optical schematic of the TGI based setup for the measurement of surface slope of a

OSA
polarized He-Ne (l=632.8nm) laser beam represented by a plane wave front ABCD is allowed to fall on the AX placed with its axis along the direction of propagation of the beam. The transmitted beam from the AX is split up into reflected and transmitted components with linear orthogonal polarizations, i. e., p and s components by the polarization beam splitter (PBS). Each component beam passes through a quarter wave plate (QWP) [QWP1/QWP2 with axes at 45 to the direction of polarizations], which transforms the state of polarization of the linearly polarized beam to circular polarization. Each beam falls on respective plane mirror [M1/M2]. On double passing through the QWP [QWP1/QWP2] each circularly polarized beam becomes linearly polarized in orthogonal direction and thus emerges out through the other port of the PBS. QWP3 transforms the state of polarizations of the component beams to opposite circular polarizations, of which the linear components transmitted by the linear polarizer/analyzer (P) interfere. Lens (L) transfers the interference pattern formed on a rotating diffuser screen (RD) to the plane of a two dimensional CCD detector array and the fringes are 3

transparent glass axicon (AX) is shown in Fig. 1. Expanded, collimated, circularly

Published by

grabbed by a frame grabber and personal computer (PC). Polarization phase shift can be introduced between the beam components reflected from M1 and M2 by varying the angular orientation of the pass direction of P [20]. When the separations of M1 and M2 from the PBS are equal, the component beams emerge out of the PBS would completely overlap. Unequal separations of M1 and M2 from the PBS would produce a linear shear along radial directions between the component beams. This is illustrated in Fig. 2, which shows the equivalent ray diagram, for an incident collimated beam represented by plane wave

OSA
virtual images of the AX formed by M1 and M2, respectively. The rays corresponding to M1 and M2 are represented by single and double arrows, respectively. The bottom part of the Fig. 2 shows the overlapping beams in a plane normal to the axis of the AX. Fan of rays from AX1 are shown by B1A1, D1C1 and that from AX2 by B2A2, D2C2, while the beam overlapping is denoted by B2A1, D1C2. Clearly the linear shear in the radial directions depends on the longitudinal separations between AX1 and AX2 which in turn depends on the difference between the individual separations of the M1 and M2 from the PBS. The optical path difference (OPD) variation, Wr, for linear shear (r) along a radial direction (r) normal to direction of propagation of the fan of rays can be written as [21]: Wr= W(r) - W(r-r) (1) where W(r) represents the optical path variation along the radial direction r. Wr is determined by using polarization phase shifting interferometry. A series of interferograms are captured as the phase difference between the interfering components is 4

front ABCD, in a plane passing through the axis of the AX. AX1 and AX2 represent the

Published by

changed, in steps, by a constant amount, by varying the angular orientation of the pass direction of P (Fig. 1). The intensity at a point (x, y) in the interference field can be written as [22] I (x,y) = I0 (x,y) { 1 + V( x,y) cos [ (x,y) + j] } (2)

where I0 (x, y), V(x, y), (x, y) are the mean intensity, fringe visibility and original phase difference between the two interfering beams at the point (x, y), respectively and j= (j-1) , with j= 1,2, 3and 0 < < ., represents phase shift between the adjacent frames. We have used an algorithm [23] which uses five different values of intensity, I1,

OSA
= arc tan [2(I4-I2)/ (I1+I5-2I3)] (3) For an error in phase shift, (so that =/2 + , where is a small quantity) the error in is given by [23] = (2/4) sin 2 (4) which shows the insensitivity of this algorithm to phase shift miss-calibration type system error. Taking absolute values of numerator and denominator of Eq. (3), is first calculated in modulo /2 and then comparing signs of sin and cos , [ref. 24, pp. 366] is transformed to modulo 2. The calculated phases are wrapped and unwrapping / phase integration is performed by comparing the phase difference between adjacent pixels [25]. Unwrapped phase values are least square fitted and deviation from a reference plane is obtained. Phase variations are converted to optical path variations by using the following relation 5

I2, I3, I4, I5 corresponding to j=1, 2, 3, 4, 5 and =/2 and is given by

Published by

= (/2) where and represent optical path length and wavelength of light, respectively.

(5)

We are actually interested in slope variation (WS/s) along a radial direction denoted by s on the surface of the AX. It is evident that W(r) depends on the surface form error W(s), on the surface of AX. Thus WS, corresponding to Wr and the corresponding linear shear s, can be obtained from known mathematical relationships to be discussed in Sec. (2.1) and hence the desired slope variation can be calculated.

OSA
Suppose a plane wave front, traveling along the axis of the AX, falls on the AX. To obtain WS, we need to consider the optical path difference (OPD) suffered by the traveling plane wave due to variation in surface form of the conical surface of the AX. The wave front distortion, in terms of OPD, suffered by a plane wave front traveling normally through a plane glass-air interface having a step of thickness t is given by (n-1) t, where n is the refractive index of the glass medium. The part of the plane wave front passing through the step suffers a phase/path delay with respect to the rest of the plane wave front which passes through the plane interface. For non normal incidence, as in case of AX (Fig. 1), the OPD can be written as [(n-1) (t / cos a)], where a (the angle between the plane and the conical surfaces of the AX) is the angle of incidence, since the glass path traversed by the portion of the wave front passing through the step becomes (t / cos a). Thus we get Wr = [(n-1) (WS / cos a)] and by rearranging, we obtain WS = [Wr cos a/(n-1) ] (6) 6

2.1. Calculation of WS

Published by

2.2. Calculation of s In Fig. 3, AX1 and AX2 represent virtual images of AX formed by M1 and M2 {Fig. 1], as discussed in Sec. 2. A ray of light traversing parallel to the axis of the AX falls on the surfaces of AX1 and AX2 at D1 and D2, respectively. The refracted rays D1O1 and D2P1 are shown by double arrowed lines. Angle of incidence is equal to the base angle a, the angle between the plane and the conical surfaces of the AX. Applying Snells law, the angle of refraction is given by, r = sin -1(n sin a) =na, for small values of a,

OSA
suffered by the incident ray is = (n-1) a. The normal separation between D1O1 and D2P1 is given by PO= D2F=r = D1D2 sin =x sin , where D1D2=x is the separation between AX1 and AX2, which is equal to twice of the difference in separations of M1 and M2 from BS. The normal drawn at F, P and O are shown by dotted lines. From Fig. 3, we can write, PO= r=D2O cos r= s cos r, where D2O= s. Combining both the relations, r =x sin , and r= s cos r, we get, x = (s cos r)/ sin. As the fringes are formed on RD, we need to find the relationship between the length scales on RD and the direction s on the surface of the AX. We have O1P1= r= P1O2 cos = rd cos , where P1O2 = rd. Thus we have, r= s cos r= rd cos and s = (rd cos )/ cos r. Also, we have Wr = Wd cos , where the suffix d refers to OPD variation along a radial direction on RD. Substituting Wr, in Eq. (6), we get WS = [Wd cos cos a/(n-1) ] 3. Results & discussions (7) 7

where n is the refractive index of the material of the AX and the angular deviation

Published by

Results obtained for a home made BK-7 glass AX of aperture diameter 75.0mm and base angle of a=5.5are presented. We have taken s = 1.0mm and obtained x 20.0mm [Sec. 2.2]. We have s = 1.009rd, Ws = 1.92 Wd, for our case, using relations derived in Sec. 2.1 & 2.2. M1 & M2, which are mounted in precision linear translation stages, are first positioned equidistant from the BS, by adjusting M1, M2 at the focal points of converging beams produced due to splitting of a converging beam of laser light, produced by a corrected telescope objective lens placed (temporarily) in place of the AX, by BS. The fluff off fringes due to interference of the emergent beams reflected by M1 and M2

OSA
setup. Next, either M1 or M2 is linearly shifted by appropriate distance (x/2 10.0mm) for obtaining the desired shear s = 1.0mm. Fig. 4 shows one of a series of interferograms which are grabbed by applying quarter wavelength geometrical phase shifts between the sheared interfering components. It is evident that Fig. 4 is an interference map corresponding to OPD variation, Wd between two close points separated by rd, along radial directions. By applying PSI, Wd at each point along a diameter of the ring shaped interference field can be found out. For simplicity and for some limitations of the software used by us, we have presented limited OPD data for the section of the AX corresponding to the radial portions lying along a row of pixels of the two dimensional CCD array. Fig. 5 & Fig. 6 show variations of Wd along the radial direction Xd and the fitted lines. Fig. 7 & Fig. 8 show the variations after tilt subtraction. The variations of slopes s = (Ws/ s) on the section of the AX are shown by Fig. 9 & Fig. 10, while the corresponding height variations are shown by Fig. 11 and Fig. 12, respectively. 8

ensures identical positioning of the mirrors (M1 & M2) in the arms of the TGI based

Published by

In the present technique, the conical fan of rays, produced by the AX, in the arms of the TGI based setup (Fig. 1), incident on QWP1 and QWP2 at a small angle of = 2.83. For zero order quartz retardation plates, this is well within the specified acceptance angle. Polymer based retarders are much less sensitive to variation in incident angle than quartz retarders and a polymer retarder changes by less than 1% over a 10 incidence angle. Thus for testing AX with relatively larger base angles, polymer based true zero order QWPs should be used. The TGI based optical setup discussed in Sec. 2 can also be used for testing of reflective

OSA
reflective AX with collimated laser beam and for allowing the reflected beam from the AX to enter the TWI based test setup. 4. Conclusions A new TGI based shearing interferometric technique for the measurement of surface profile/ slope errors of axicon lenses is presented. It is evident that the methods reported in ref. [18] and [19] are applicable for axicon lenses with smaller base angles. For AXs with relatively larger base angles the fringe density, in these methods, would be high and may go well beyond the resolution limit of the CCD array. Thus the methods are not useful for testing of AX with relatively larger base angles. The present technique has much higher upper limit for base angle, as discussed in Sec.3. Also, the present technique can be used for testing of reflective axicon by introducing minor modifications. Since the present method is based on TGI, a good vibration isolated table is needed. Acknowledgements: 9

AX by using a non-polarizing cube beam splitter in the input side for illuminating the

Published by

We wish to thank S. S. Negi of our Optical Workshop for his help in setting up the experimental arrangement and Rishipal of our Optical workshop for drawing of the figures.

OSA
10

Published by

References 1. J. H. McLeod The axicon: A new type of optical element. J. Opt. Soc. Am. 44 592-597 (1954) 2. J. H. McLeod Axicons and their uses J. Opt. Soc. Am. 50 166-169 (1960) 3. Z. Jaroszewicz, A Burvall, A. T. Friberg, Axicon the most important optical element, Opt. & Photonics News, April, 34-39 (2005) 4. T. Kololuoma, K. Kataja, S. Juuso, J. Aikio, Fabrication and characterization of hybrid glass based axicons, Opt. Eng. 41, 3136-3140 (2002)

OSA
coherence tomography, Opt. Letts. 27 243-245 (2002) 6. P. Dufour, M. Piche, Y. De Koninck and N. McCarthy, Two photon excitation fluorescence microscopy with a high depth of field using an axicon, Appl. Opt. 45 9246-9252 (2006) 7. M. Blanchard, Y. Dastous, G. Roy, and B. Tremblay, Axicon lenses to produce very long plasmas for lasers, J. Opt. Soc. Am. 68, 1389 (1978) 8. M. Couture and M. Piche, Optical pumping of infrared lasers using an axicon geometry, J. Opt. Soc. Am. A 2, P38 (1985) 9. F. P. Schafer, On some properties of Axicons, Appl. Phys. B39 1-8 (1986) 10. W. D. Kimura, G. H. Kim, R. D. Romea, L. C. Steinhauer, I. V. Pogorelsky, K. P. Kusche, R. C. Fernow, X. Wang and Y. Lin, Laser acceleration of relativistic electrons using the inverse Chernokov effect, Phys. Rev. Letts. 74 546-549 (1995) 11

5. Z. Ding, H. Ren, Y. Zhao, J. S. Nelson and Z. Chen High resolution optical

Published by

11. M. D. Wei, W. L. Shiao, Y. T. Lin, Adjustable generation of bottle and hollow beams using an axicon, Opt. Com. 248 7-14 (2005) 12. I. Manek, Y. B. Ovchinnikov, R. Grimm, Generation of a hollow laser beam for atom trapping using an axicon, Opt. Com. 147 67-70 (1998) 13. G. Scott and N. McArdle, Efficient generation of nearly diffraction free beams using axicon, Opt. Eng. 31 2640-2643 (1992) 14. J. Arlt, K. Dholakia,Generation of higher order Bessel beams by use of an axicon, Opt. Commun. 177 297-301 (2000)

OSA
thermal effects for spatial trepanning, Opt. Eng. 45 014301 (2006) 16. S. D. Fantone, Simple method for testing an axicon, Appl. Opt. 20, 3685-3686 (1981) 17. J. Hayes, K. L. Underwood, J. S. Loomis, R. E. Parks and J. C. Wyant, Testing of non-linear diamond turned reflaxicons, Appl. Opt. 20,235-239 (1981) 18. D. Kupka, P. Schlup, R. A. Bartels, Self referenced interferometry for the characterization of axicon lens quality, Appl. Opt. 47, 1200-1205 (2008) 19. M. de Angelis, S. D. Nicola, P. Ferraro, A. Finizio, G. Pierattini, Test of a conical lens using a two beam shearing interferometer, Opt. & Lasers in Eng. 39, 155-163 (2003) 20. S. Chatterjee, Y. P. Kumar, and B. Bhaduri Measurement of surface figure of plane optical surfaces with polarization phase shifting Fizeau interferometer, Opt. & Laser Tech. 39, 268-274 (2007) 12

15. Zeng, Latham and Kar, Shaping of annular laser intensity profiles and their

Published by

21 . Strojnik, G. Paez, and M. Mantravadi Lateral shear interferometers, in Optical Shop Testing, D. Malacara, Ed. p. 122-184, Wiley (2007) 22. P Hariharan, Phase shifting interferometery: minimization of systematic errors, 23. Opt. Eng. 39, 967-969, (2000)

P. Hariharan, B. F. Oreb, T. Eiju, Digital phase shifting interferometery: a simple error compensating phase calculation algorithm, Appl. Opt., 26,

2504-2505, (1987). 24. K. Creath, Phase measurement interferometry techniques, in Progress in Optics, xxviii. E. Wolf, Ed. p. 349-393, North-Holland Amsterdam: (1988)

OSA
25. D. Malacara, S. Malacara, and Z. Malacara, in Interferogram Analysis for Optical Testing, p. 248-55, Marcel Dekker, New York: (1998) 13

Published by

Figure captions: Fig. 1 Optical schematic of the TGI based optical setup for the measurement of surface slopes/profile of Axicon lens Fig. 2 Generation of linear shear in the radial directions between the overlapping conical beams due to the longitudinal separation between the virtual images AX1 and AX2 of AX in M1 and M2 [Fig. 1], respectively. Fig. 3 Geometrical construction for the computation of s Fig. 4 Interferogram captured with s=1.0mm

OSA
Fig. 5 Variation of OPD between two closely separated (rd) points along a radial direction [A1B2] (Fig. 2) denoted by xd Fig. 6 Variation of OPD between two closely separated (rd) points along a radial direction [D1C2] (Fig. 2) denoted by xd Fig. 7 Variation of OPD shown in Fig. 4 after tilt subtraction Fig. 8 Variation of OPD shown in Fig. 5 after tilt subtraction Fig. 9 Slope variation along the section [A2B2/ A1B1] of the AX in the plane of the figure (Fig. 2) Fig. 10 Slope variation along the section [C2D2/ C1D1] of the AX in the plane of the figure (Fig. 2) Fig. 11 Height variation along the section [A2B2/ A1B1] of the AX in the plane of the figure (Fig. 2) Fig. 12 Height variation along the section [C2D2/ C1D1] of the AX in the plane of the figure (Fig. 2) 14

Published by

OSA
Fig. 1 Optical schematic of the TGI based optical setup for the measurement of surface slopes/profile of Axicon lens 15

Published by

OSA
Fig. 2 Generation of linear shear in the radial directions between the overlapping conical beams due to the longitudinal separation between the virtual images AX1 and AX2 of AX in M1 and M2 [Fig. 1], respectively. 16

Published by

OSA
Fig. 3 Geometrical construction for the computation of s 17

Published by

OSA
Fig. 4 Interferogram captured with s=1.0mm 18

Published by

OSA
Fig. 5 Variation of OPD between two closely separated (rd) points along a radial direction [A1B2] (Fig. 2) denoted by xd 19

Published by

OSA
Fig. 6 Variation of OPD between two closely separated (rd) points along a radial direction [D1C2] (Fig. 2) denoted by xd 20

Published by

OSA
Fig. 7 Variation of OPD shown in Fig. 4 after tilt subtraction 21

Published by

OSA
Fig. 8 Variation of OPD shown in Fig. 5 after tilt subtraction 22

Published by

OSA
Fig. 9 Slope variation along the section [A2B2/ A1B1] of the AX in the plane of the figure (Fig. 2) 23

Published by

OSA
Fig. 10 Slope variation along the section [C2D2/ C1D1] of the AX in the plane of the figure (Fig. 2) 24

Published by

OSA
Fig. 11 Height variation along the section [A2B2/ A1B1] of the AX in the plane of the figure (Fig. 2) 25

Published by

OSA
Fig. 12 Height variation along the section [C2D2/ C1D1] of the AX in the plane of the figure (Fig. 2) 26

Published by