We propose a new family of achromatic phase shifters for infrared nulling interferometry. These key optical components can be seen as optimized Fresnel rhombs, using the total internal reflection phenomenon, modulated or not. The total internal reflection indeed comes with a phase shift between the polarization components of the incident light. We propose a solution to implement this vectorial phase shift between interferometer arms to provide the destructive interference process needed to disentangle highly contrasted objects from one another. We also show that, modulating the index transition at the total internal reflection interface allows compensating for the intrinsic material dispersion in order to make the subsequent phase shift achromatic over especially broad bands. The modulation can be induced by a thin film of a well-chosen material or a subwavelength grating whose structural parameters are thoroughly optimized. We present results from theoretical simulations together with preliminary fabrication outcomes and measurements for a prototype in Zinc Selenide.
© 2007 Optical Society of America
Despite the tremendous progress in the indirect detection techniques and the increasing number of detected extrasolar planetary objects, exoplanet hunters are still struggling with the technical challenges hampering the conception of an observatory that will allow the direct imaging and characterization of Earth-like planets. Indeed, even our best telescopes can’t see them directly because of the large flux ratio between the planet and its parent star. For example, an Earthlike exoplanet is typically ∼ 109 times fainter than its host star in the visible spectrum, ∼ 106 in the thermal infrared. Among all raised ideas to reach the needed very high dynamic range, such as visible coronagraphy, infrared nulling interferometry proposed by R. Bracewell in 1978  appears to be a most promising technique. The nulling interferometry consists in adjusting the phases (with a π phase shift in the two-telescope case) of the beams coming from various telescopes (two or more) to produce a fully destructive interference on the optical axis, nulling the starlight while letting the planetary signal pass through.
One of the most critical units of a nulling interferometer is the phase shifter. The difficulty is to conceive it to perform well enough over the whole operational bandpass. Remotely characterizing exoplanet atmospheres, which is our ultimate goal, requires spectroscopic analysis over large spectral bands. For example, the Darwin Infrared Space Interferometer  considered by the European Space Agency (ESA), or the Terrestrial Planet Finder-Interferometer (TPF-I) considered by NASA, will operate in the infrared wavelength range spanning from 6 to 18 (or 20) μm. Several studies have been initiated by both agencies, with the purpose of selecting the best achromatic phase shifter (APS). Exhaustive listing or detailed trade-off between different existing APS is beyond the scope of this paper. However, let us cite the four most promising concepts that are currently being examined:
- - the dispersive plate APS , directly inspired by the techniques used by optical designers to minimize lens chromatic aberrations, uses a given number of glass or dielectric plates, whose materials and thicknesses are optimized together with the free-air optical path difference (OPD), mutually neutralizing the various dispersion gradients to produce a given phase shift over the desired bandpass;
- - the focus-crossing APS , where an achromatic phase shift of π is produced by making the beam cross a focus along one of the interferometer arm;
- - the field-reversal APS , which is based on the achromatic reversal of the electric-field vector on one of the two interfering waves, by means of a rotational shearing interferometer with a fixed shear of 180 degrees;
- - the vectorial APS, subject of this paper, which consists in spatially distributing a phase retardance that is primarily affecting the orthogonal polarization components.
According to , for an interferometer with stellar leakages similar to those of a Bracewell two-telescope interferometer, a sensible value for the mean instrumental leakage contribution to the null depth would be
In addition to the APS imperfections there are many systematic sources of deterioration of the null depth: e.g., telescope pointing errors, photometric unbalance, OPD errors, etc. Consequently, the requirement for the limitation on the null depth due to the APS alone, must be significantly lower than expressed in Eq. 1, i.e., N = 10-6 or better. In practice, we will always apply a reasonable security by requiring a few 10-7.
As we will show, the vectorial APS family is a promising alternative to the three first mentioned APS. Indeed, as compared to the difficulty of finding the dispersive plate APS working point(s), the vectorial APS implementation is straightforward. Moreover, unlike the focus-crossing and field-reversal APS, vectorial APS do not come with a subsequent pupil flip. The pupil inversion, creating a pair of images for one single planet, induces several limitative consequences. First, the spatial coherence γ of the interfering beams must be of the order of the null specification on the APS, i.e. γ ≤ 10-6. Second, Hénault  recently stressed out that the pupil-flip APS could be less favorable because the pupil inversion seems to substantially impact extinction maps, implying S/N losses and weakening the detection capacity of the instrument . These problems can only be mitigated with monomode spatial filtering. Third, contrary to the mirror APS which only provides a π phase shift (some interferometer configurations require other phase shift values), the vectorial APS is theoretically able to satisfy any requirements in this respect. For instance, it could provide the 90°, 120° and 240° phase shifts required by the three telescope nuller  (TTN), either in the planar or non-planar configuration.
We recently proposed the idea of using subwavelength gratings in total internal reflection (TIR) incidence as achromatic phase shifters . Theoretical calculations using the Rigorous Coupled Wave Analysis  (RCWA) pointed to interesting results. The purpose of the present paper is fourfold:
- - put the subwavelength grating in TIR incidence APS in the more general context of modulated Fresnel rhomb APS for infrared nulling interferometry;
- - presenting a general tolerance study of Fresnel rhomb-based vectorial APS;
- - introducing some manufacturing and design considerations for the actual prototype under fabrication at CSL (“Centre Spatial de Liège”) to be tested at room (298 K) and cryogenic (100 K) temperatures at IAS (“Institut d’Astrophysique Spatiale”, Orsay, France);
- - presenting some preliminary retardance measurement results obtained at 632.8 nm.
The TIR phenomenon comes with a phase shift between the vectorial s and p polarization components. This vectorial phase shift takes the following general form 
where θ is the angle of incidence, greater or equal to θc, the critical angle defined as sin θc = nti = nt/ni and where ni and nt are the refractive indices of the incident and emergent media, respectively. This property is exploited in the well-known Fresnel rhombs, which are retarders to be used alone as polarization optics.
2.1. Double rhomb configuration
Single Fresnel rhombs are known to be very sensitive to incidence angle variations (at the sub-arcmin level). This drawback would be penalizing in our application since the thermal infrared interferometric beam is likely to slightly diverge because of diffraction while somewhat wandering around because of vibrations. For this reason, we chose the double rhomb configuration for its known insensitivity to incidence variations . Indeed, the pairs of reflections in the two rhombs are complementary (see Fig. 1 and Fig. 2); an increase in the first two TIR angles due to departure from nominal incidence leads to a decrease in the last two angles. Thus if the phase shift varies linearly with the TIR angle, the retardance changes are cancelled. Therefore a wider range of incidence angle variations can be tolerated: up to several degrees.
2.2. Modulated total internal reflection achromatic phase shifter
Classical Fresnel rhombs are limited by the intrinsic index dispersion of the rhomb bulk material . Engraving a subwavelength grating on the TIR interface or simply depositing a single thin layer of a well chosen material leads to a significant improvement. Indeed, the electromagnetic field evanescent interaction with the optimized micro-structure or thin film allows us to tune the index ratio nti of Eq. 2. It is to be noted that in the double rhomb configuration, the number of subwavelength gratings to be imprinted (resp. thin film to be deposited) on the TIR interface may be limited to only two out of four. Indeed, the dispersion compensation artificially introduced by the modulated interface can already and substantially improve the global behavior of the component in terms of phase shift achromaticity.
2.2.1. Total internal reflection grating
The principle of the total internal reflection grating achromatic phase shifter (TIRG APS) is to use a subwavelength grating in the TIR incidence condition . One dimensional sub-wavelength gratings turn out to be artificially anisotropic . It means that the structure can be associated with two synthetic effective indices, one for each polarization component: TE (transverse electric, vibrating parallel to the grating grooves, or s) and TM (transverse magnetic, vibrating perpendicular to the grooves, or p). These effective indices, nTE and nTM are totally dependent on the grating and incidence geometries (see Fig. 5), which can be wisely engineered to compensate for the natural material dispersion.
2.2.2. Total internal reflection thin film
Instead of engraving a subwavelength grating onto the TIR interface, one can deposit a layer of a well-chosen foreign material. The principle of the so-called total internal reflection thin film achromatic phase shifter (TIRTF APS) is to make use of such a thin film coated with an appropriate thickness on the TIR interface. This principle has been known for quite a long time in the optical where MgF 2 thin films are commonly deposited on BK7 or Silica rhombs to improve the angular and chromatic behavior of commercial Fresnel rhombs .
Implementing two strictly identical π retarders into the two interferometer arms and rotating them by ninety degrees with respect to each other around the interferometer optical axis, is a practical solution to spatially distribute the initial vectorial phase shift (see Fig. 3); “vectorial” meaning between the polarization components. The ninety degree rotation between the retarders around the optical axis must be respected at ±2 arcmin to reach the 10-7 secure null depth level.
3. Theoretical analysis
This section is devoted to the theoretical analysis of the TIRG and TIRTF APS. The purpose of this analysis was to optimize (with the “simplex-search” optimization method , highly robust when the parameter space is highly discontinuous) the APS component in order to “achromatize” the Darwin/TPF-I working wavelength range: 6–18 μm. However, due to practical constraints like wavefront filtering or dichroic limitations, the latter is expected to be divided into two or three sub-bands. In the two-band case, the first one ranges from 6 to 11 microns while the second one ranges from 11 to 18 microns. Before going into the details of the theoretical analysis, one had to choose the materials according to the bandwidth specifications mentioned here above. Regarding this matter, we immediately had to discard common infrared materials like Silicon (Si) and Gallium Arsenide (GaAs) for their strong multi-phonon absorption features beginning between 8 and 12 microns . For the same reason, we also had to discard Zinc Sulphide (ZnS) and Germanium (Ge), other widespread infrared materials.
The choice for the bulk material constituting the rhomb in fact revealed to be very difficult and severely limited since the material has to be perfectly transparent up to 18 microns (20 if possible) and of course available in large ingots of very good optical quality (good homogeneity, low impurity, etc.). For this reason, we focussed on common long-wave infrared materials such as Zinc Selenide (ZnSe), Cadmium Telluride (CdTe), and KRS-5. Although ZnSe is in general used in the 11–18 μm range, CdTe and KRS-5 are the only ones perfectly clear up to 18 microns. Note that ZnSe, ZnS or even CVD (Chemical Vapor Deposition) Diamond can be used as the TIRTF APS layer materials in their phonon absorption ranges since their working thicknesses are very thin and the subsequent absorptions therefore negligible. These selected materials are common in infrared applications, they cover a large refractive index spectrum (n = 2 - 4) and their processing (polishing, etching, sputtering, etc.) is in general well known for most of them. The refractive index dispersions will be taken from  for Diamond, from  for ZnSe, from  for KRS-5 and from  for ZnS, CdTe.
To simulate grating responses in the subwavelength domain, we used an algorithm based on the RCWA, which is also called the Fourier modal method because it is based on Fourier decompositions of the physical characteristics of the grating and fields so that the resolution takes place in the frequency space.
3.1. ZnSe Rhomb
In its polycrystalline form, ZnSe is available in large quantities and volumes at a reasonable cost. It is also relatively convenient to polish with very good surface qualities. Moreover, its thermal properties are very attractive, with a low thermal expansion coefficient (7.1× 10-6/K) and a rather good thermal conductivity (0.18 W/cm/K). Unfortunately, the chosen double-rhomb configuration inevitably lengthens the optical path in the material. For instance, in the present case, a working angle of about 65.03° would lead to a physical path of about 17 cm inside the material for an entrance beam diameter of 15 mm. In the second Darwin sub-band (11 – 18 μm), such a long path is penalizing since ZnSe begins its phonon absorption around 14 microns. Indeed, the absorption coefficient  k is equal to 4.24 × 10-6 at 14 μm and 298 K, leading to absorption of about 50%. This value reduces to 2.52 × 10-6 at 100 K, giving an absorption of 30%, which remains acceptable for demonstration purposes. For this reason, a ZnSe TIRG APS in the double-rhomb configuration will be practically limited to the 6 – 14μm wavelength range. One of the main outcomes of the theoretical results presented in Table 1 is that the ZnSe bare Fresnel rhomb (without any modulation of the TIR interfaces) is not performing well enough for the Darwin baseline 10-6 specification (see Fig. 4, dotted line). However, depositing a foreign material at the TIR interface or imprinting a subwavelength grating onto it allows overcoming this limitation in the 6 – 11μm wavelength range. In the larger 6 – 14μm range, only the subwavelength grating solution is able to meet the specification (Fig. 4, continuous line) while the thin film one just misses it by a small factor (Fig. 4, dashed line).
3.2. CdTe Rhomb
Availability of CdTe ingots is more limited than ZnSe ones. Polishing of CdTe is quite delicate but currently under evaluation for improvement by several manufacturers. CdTe, which also possesses rather good thermal characteristics (expansion coefficient of 5.9 × 10-6/K and conductivity of 0.062 W/cm/K), nevertheless appears to be a viable potential solution. Indeed, as already stated, CdTe is one of the selected rhomb materials perfectly clear up to 18 microns. Moreover, theoretical RCWA results are excellent, showing better performance than ZnSe (see Table 1). First of all, the CdTe bare Fresnel rhomb solution is worth considering since it is theoretically performing well enough, at least for the Darwin first sub-band but unfortunately not for the second one. Depositing a thin film of a foreign material (e.g., Diamond, ZnSe or ZnS) overcomes this limitation. As far as the subwavelength grating solution is concerned, results are comfortably in the specifications for both bands. Let us emphasize that there exists a most interesting theoretical solution which would “achromatize” the full Darwin wavelength range from 6 to 18 microns with a single APS consisting of an optimized double CdTe rhomb modulated with a subwavelength grating (Table 1).
3.3. KRS-5 Rhomb
KRS-5 infrared dispersion is extremely low , making it an ideal material for a Fresnel rhomb. Indeed, results displayed in Table 1 confirm its very good potential. Unfortunately, KRS-5 is a very difficult material to handle. It is toxic, brittle and is not easily polished (to our knowledge its polishing was never demonstrated at qualities better than λ/4 rms, λ = 632.8 nm). It is to be emphasized that KRS-5 is common in infrared applications despite its bad thermal conductivity (5.4 × 10-3 W/cm/K) and expansion coefficient (5.8 × 10-5/K). KRS-5 was also considered since it is one of the rare materials perfectly transparent above 15 microns. Finally, KRS-5 cannot be micro-structured nor easily treated with foreign materials because of its chemical reactivity.
4. Design and tolerancing of a modulated Fresnel rhomb prototype
In this section, we will discuss the design of a modulated Fresnel rhomb prototype intended at being one of the APS selected for study, fabrication and test in the framework of one of ESA’s Darwin R&D preparatory activities: Nulltimate , which is a project of infrared cryogenic (100 K) nulling testbed. This section will be devoted to the tolerancing of the subwavelength grating and the macroscopic rhomb. The choice of the material for the prototype has been made thanks to a trade-off matrix. Even if CdTe shows the best theoretical characteristics in terms of transparency and performances, ZnSe was the retained material to make the prototype in the 6 – 14μm Nulltimate wavelength range for the following reasons:
- - in its polycrystalline state, ZnSe is a priori easy to source, and a convenient material to handle and polish (surface qualities of λ/30 rms with λ = 632.8 nm are routinely obtained);
- - it is compatible with the thickness-adjustable beam splitters of the Nulltimate test bench which are planned to be in ZnSe;
- - it is transparent in the optical, letting the HeNe 632.8-nm interferometric metrology beam pass through.
4.1. TIRG tolerancing
The fabrication of the TIRG APS will be based on micro-electronic technologies. The first classical step consists in imprinting a photomask of the grating in a photoresist coated on the chosen substrate material. The precision of this step is critical because it defines once and for all the lateral dimensions of the grating (Fig. 5): its period Λ and the so-called feature line, i.e., the period multiplied by the filling factor f. This pattern will then uniformly be transferred into the substrate by an appropriate reactive plasma-beam etching down to the desired depth (h). The fabrication has to be interactive to properly compensate for process errors. In situ monitoring is a possible solution  but not the only one as we will discuss.
4.1.1. Manufacturing scenario
Let us now consider the ZnSe TIRG APS designed for the 6 – 14 μm wavelength range. A rough optimization for this range leads to a 900-nm period. Then, assuming this fixed period, best solutions are searched using the “simplex” optimization method  coupled to the RCWA algorithm with the filling factor f, the grating thickness h and the incidence angle θ left as free parameters. Results of this optimization are displayed in Fig. 6 (top left), where the optimal null depth is plotted versus the feature line, i.e., the product of the 900-nm fixed period and the varying filling factor. Continuous variations are now imposed to the feature line while letting the optimization algorithm find the corresponding adjustment of the thickness that minimizes the null depth (see 6, top right). The best null depth, in this case the minimum one, is about 1.5 × 10-7. This optimal value is obtained for the 220-nm feature line. The corresponding adjusted thickness is 1.43 μm. If we calculate the grating aspect ratio, we find 6.5 which is demanding given the nature of the material to be subsequently etched. To relax this difficulty, we chose the conservative width of 250 nm for the feature line which corresponds to an adjusted thickness of 1.23 μm and a subsequently reduced aspect ratio of about 5. Let us now fix the feature line to 250 nm, keep the period at 900 nm and vary the thickness artificially, letting the optimization algorithm find the corresponding incidence polar angle θ adjustment that minimizes the null depth. Results of this analysis are shown in Fig. 6, middle left and right. Provided that the polar angle can be adjusted with a sub-arcmin precision, the N ≈ 2.5 × 10-7 tolerance on the thickness definition is 1.23 ± 25 nm, i.e., about 2%, which is feasible.
The conclusion of this tolerance analysis is that, provided that there are interactions between measurements and manufacturing at each key step, the tolerances on the parameter definitions are rather comfortable, up to a few tens of nanometers for the feature line and the thickness. A posteriori correction is reported on the incidence angle which is very convenient since the double-rhomb geometry still ensures the insensitivity to the entrance beam incidence configuration (input angle and beam divergence, for instance) at the macroscopic level.
4.1.2. Grating slope angle
Departure from the nominal assumed grating profile, i.e. a perfectly rectangular one, is likely to naturally emerge from the plasma-beam etching process (Fig. 7). The calculated tolerance on the grating slope angle α is 5°, which is very tight and necessitates very directional and anisotropic plasma etching processes.
4.2. TIRTF tolerancing
As far as an eventual TIRTF APS component is concerned, tolerancing can be envisaged in another way. Since the parameter space is limited to two variables, i.e. the thickness of the layer and the angle of incidence, the working points can be traced in two-dimensional maps. From Fig. 6 bottom, we can conclude that the tolerance on the thickness is of ∼ 10%. It is to be noted that coatings are routinely deposited with a precision of ∼ 1% on the thicknesses.
4.3. Surface roughness and material homogeneity
We have conducted a complete tolerancing study of roughness and homogeneity effects using a Monte-Carlo method coupled with rigorous diffraction analysis. The conclusion is that the surface roughness and grating thickness variability should not exceed 10 nm rms. Concerning linewidth roughness (feature line uniformity), it should not depart from 25 nm rms. As far as material refractive index homogeneity (δn) is concerned, the conclusion is that the current ZnSe synthesizing method (CVD) provides an excellent optical quality (δn ≈ 3 × 10-6) that should not affect the nulling performance of the interferometer . Small defects should moreover be filtered out by the planned spatial filters.
4.4. Rhomb geometry and ghosts
The geometry of the rhomb (Fig. 8) is well constrained by the TIR optimal angle (θ 1 ≈ 65°) and the dimension of the incoming interferometric beam, i.e. B= 15 mm in diameter (the input face size E will be 20 by 20 mm). Each of the four rhombs must be strictly identical with arcsec precision for the angles and micrometric precision for the dimensions, which can be ensured to a certain degree by polishing them together in a same batch. Slight departure from nominal dimension will impact OPD, but can be compensated thanks to the thickness-adjustable beam splitters of the testbed which are also in ZnSe. In order to dump stray light originating from spurious reflections at the ZnSe/air interfaces, we chose to apply anti-reflective treatments (R ≈ 1.5%). Unfortunately, given the stringent constraints of nulling interferometry, we also had to apply well-chosen wedges between rhomb interfaces. In order to further mitigate ghosts, we have used ray-tracing analysis which led to an optimal wedge value of w = 1°. Both rhombs of the double-rhomb configuration possess the same but opposed wedge angle so that deviation and dispersion are annihilated and therefore do not impact the nulling process (Fig. 9). Let us also mention that the rhomb spacing D must be identical at the 0.1 mm-level between both interferometer arms in order to avoid differential dispersion of the OPD.
In this section, we will present some manufacturing details for the ZnSe Fresnel rhomb APS.
5.1. ZnSe Rhomb cutting and polishing
We have conducted cutting and polishing trials of ZnSe rhombs with the appropriate global geometry and dimensions (Fig. 10). Results of ZYGO interferometric measurements in terms of surface quality comfortably fall in the specifications, and beyond our expectations with surface figures of λ/100 rms (λ = 632.8 nm). Concerning the roughness, WYKO profilometer measurements give a rather good 3 nm rms.
5.2. TIR Grating manufacturing
We chose to manufacture the micro-pattern using holographic lithography and dry etching. The first one is necessary for masking the parts of the substrates to be protected during the etching step (Fig. 11, right). Direct laser writing or classical mask exposure techniques are not applicable onto the rhomb facets since they are designed to accommodate thin and flat substrates such as wafers for micro electronics. The holographic recording allows us to overcome this limitation by imprinting the photoresist mask thanks to a two-wave sinusoidal interference fringe pattern (Lloyd mirror mounting, Fig. 11, left). A good understanding and control of the recording and developing parameters (as energy dose, development duration, resist thickness, resist optical contrast, etc.) have produced reproducible rectangular binary profiles with a controlled filling factor (at the 10-nm level) instead of the natural sine profile. After its processing, the remaining photoresist pattern serves as a lithographic mask for the subsequent reactive plasma beam etching (RPBE) process, which makes use of both the ballistic effect and chemical reactivity of a beam of reactive ions to transfer structures into a substrate. The various parameters (gas melanges, beam energy, beam incidence, etc.) characterizing the etching process are optimized for the transfer into various materials. The interest of such a technique comes particularly from its high selectivity, the potential to efficiently etch one material and not another coexisting one, used as a mask. The infrared material (ZnSe or CdTe) is engraved using a selective chemistry (CH 4/H 2  or chlorine-based  compounds), depending on the crystal phase of the material which leaves intact the photoresist mask but etches the infrared substrate. Validation trials are currently under progress. They aim at identifying the best process conditions ensuring the most efficient transfer of the diffractive structure from the mask into the material. Highest selectivity, homogeneity and reproducibility are the key factors under optimization.
5.3. Mechanical mount
The two mechanical mounts (one per interferometer arm) of the two ZnSe double rhombs must be thoroughly optimized to account for thermo-mechanical effects. Indeed, the component shall be measured at room (298 K) and cryogenic (100 K) temperatures. It will thus be cooled down and the mount must allow for an efficient heat transfer while compensating differential contractions. The alignment of the two rhombs must be controlled around 3 axes with an arcmin precision. The double rhomb spacing difference between both arms, and thus between the two corresponding mechanical mounts must be below 0.1 mm. The mounts must also account for the ZnSe sensitivity to stress birefringence. In order to quantify the latter to help the mount design, a finite element analysis was conducted and resulted in the conclusion that the rhombs must lay freely at 45° in their mounts (Fig. 12). Indeed, given the stress-optics coefficient of ZnSe, C λ=10.6μm ≈ -12 brewsters , even its own weight constraint σ can induce a substantial birefringence Δn = C λ=10.6μmσ which is just within the specifications.
6. Measurement of a ZnSe Fresnel rhomb retardance at 632.8 nm
We conducted preliminary measurements of the performance of a ZnSe Fresnel rhomb at 632.8 nm with a simple polarimeter arrangement (Fig. 13). A governing equation relates the ratio of the intensities (corresponding to ) and I 0(corresponding to P20) of the two orthogonal polarization states to the orientations of the input/output polarizer with respect to the retarder (α), and the retardance (Γ)
In addition to the retardance measurement, the experiment allowed us to disentangle several parasitic effects that could prevent the rhomb to reach its theoretical phase shifting potential. The rhomb TIR angle of 65°±0.1° should theoretically provide a 91°±0.5°phase shift at 632.8 nm. The primary source of perturbation was identified as the surface and bulk scattering (∼ 1%). Fortunately, we have demonstrated that it can be mitigated at least at the 10-5 level (accuracy of our optical setup) by inserting a spatial filter at the rhomb output, as planned for Nulltimate too. The secondary identified source was the spurious parasitic reflections. Tilting the rhomb by ∼ 1° was necessary to get rid of the ghost, in agreement with our ray-tracing analysis that concluded to the necessity of wedging the final rhombs (Sect. 4.4). We have then measured a phase shift of 88.8°±1.5° [Fig.14](uncertainty of the optical setup) which is consistent with the tilt-corrected theoretical value of 90.2°±0.5°(uncertainty of the measured rhomb angle). Given the different systematics in our experiment at 632.8 nm, we can extrapolate an upper boundary for the expectable null depth at 6 μm: N < 10-4. Moreover, this experiment allowed us to identify the way to mitigate every potential sources of disturbance imputable to the sole rhomb APS so that its final performance should comply to the Darwin specifications.
To summarize, we have presented a new family of APS for thermal infrared radiations relying on the TIR phenomenon, modulated or not. The modulation can be induced either by an integrated subwavelength grating or a deposited thin film. Theoretical results show remarkable improvements over the classical Fresnel rhomb technique, which is always limited by the intrinsic dispersion of the bulk material used. We have also presented some design key points and encouraging preliminary measurements for the ZnSe prototype under manufacturing. In the framework of R&D activities for the Darwin mission, this prototype will be tested on the Nulltimate test bench at IAS. Results will be the subject of a forthcoming paper.
The authors acknowledge the financial support of ESA.
References and links
01. R.N. Bracewell, “Detecting Non Solar Planets by Spinning Infrared Interferometer,” Nature 274, 780–781 (1978). [CrossRef]
02. A. Léger, J.M. Mariotti, B. Mennesson, M. Ollivier, J.L. Puget, D. Rouan, and J. Schneider, “Could We Search for Primitive Life on Extrasolar Planets in the Near Future,” Icarus 123, 249–255 (1996). [CrossRef]
03. A.L. Mieremet, J.J. Braat, H. Bokhove, and K. Ravel, “Achromatic phase shifting using adjustable dispersive elements,” Proc. SPIE 4006, 1035–1041 (2000). [CrossRef]
04. J. Gay and Y. Rabbia, “Principe d’un coronographe interférentiel,” C. R. Acad. Sci. Paris 322, 265–271 (1996).
05. E. Serabyn and M.M. Colavita, “Fully Symmetric Nulling Beam Combiners,” Appl. Opt. 40, 1668–1671 (2001). [CrossRef]
06. B. Chazelas, F. Brachet, P. Bordé, B. Mennesson, M. Ollivier, O. Absil, A. Labèque, C. Valette, and A. LégerA., “Instrumental stability requirements for exoplanet detection with a nulling interferometer: variability noise as a central issue,” Appl. Opt. 45, 984–992 (2006). [CrossRef] [PubMed]
08. C. Hanot et al., Proc. SPIE 6693-62, in press (2007).
09. A.L. Karlsson, O. Wallner, J.M. Perdigues Armengol, and O. Absil, “Three telescope nuller based on multibeam injection into single-mode waveguide,” Proc. SPIE 5491, 831–841 (2004). [CrossRef]
10. D. Mawet, C. Lenaerts, P. Riaud, J. Surdej, S. Habraken, and D. Vandormael, “Use of subwavelength gratings in TIR incidence as achromatic phase shifters,” Opt. Express 13, 8686–8691 (2005). [CrossRef] [PubMed]
11. M.G. Moharam and T.K Gaylord, “Rigourous Coupled-Wave Analysis of planar grating diffraction,” J. Opt. Soc. Am. 71, 811–818 (1981). [CrossRef]
12. M. Born and E. Wolf, “Reflection and refraction of a plane wave,” in Principles of Optics eds. (Cambridge University press, seventh edition, 1997), 49–53.
13. K. B. Rochford, A. H. Rose, P. A. Williams, C. M. Wang, I. G. Clarke, P. D. Hale, and G. W. Day, “Design and performance of a stable linear retarder,” Appl. Opt. 36, 6458–6465 (1997). [CrossRef]
16. J.C. Lagarias, J. A. Reeds, M. H. Wright, and P. E. Wright, “Convergence Properties of the Nelder-Mead Simplex Method in Low Dimensions,” SIAM Journal of Optimization 9(1), 112–147 (1998). [CrossRef]
17. Gary J. Hawkins, Spectral Characterisation of Infrared Optical Materials and Filters (PhD Thesis - The University of Reading UK, 1998).
18. F. Peter, Z Phys 15, 358–368 (1923). [CrossRef]
19. W. J. Tropf, “Temperature-dependent refractive index models for BaF2, CaF2, MgF2, SrF2, LiF, NaF, KCl, ZnS and ZnSe,” Opt. Eng. 34, 1369–1373 (1995). [CrossRef]
20. W. J. Tropf, M.E. Thomas, and M.J. Linevsky, “Infrared refractive indices and thermo-optic coefficients for several materials,” Proc. SPIE 3425, 160–171 (1998). [CrossRef]
21. A. Labèque, B. Chazelas, F. Brachet, C. Commeaux, P. Blache, A. Léger, M. Ollivier, T. Lepine, and C. Valette, “The Nulltimate project: building and testing, at low temperature, achromatic phase shifters to prepare the Darwin mission,” Proc. SPIE 5491, 999–1010 (2004). [CrossRef]
22. P. Lalanne, P. Pichon, P. Chavel, E. Cambril, and H. Launois, “Interferometric Characterization of Subwavelength Lamellar Gratings,” Appl. Opt. 38, 4980–4984 (1999). [CrossRef]
23. W. R. Chen, S. J. Chang, and Y. K. Su et al., “Refractive Ion Etching of ZnSe, ZnSSe, ZnCdSe and ZnMgSSe by H2/Ar and CH4/H2/AR,” Jpn. J. Appl. Phys. 39, 3308–3313 (2000). [CrossRef]
24. K. Kurisu, T. Hirai, and K. Fuseet al., “Development of a Diffractive Optical Element for Laser Processing,” SEI Technical Review 53, 86–91 (2002).
26. C. Yang and P. Yeh, “Artificial uniaxial and biaxial dielectrics with the use of photoinduced gratings,” J. Appl. Phys. 81, 23–29 (1997). [CrossRef]