**5. Integrated optics interferometry for astrophysics**

#### **5.1 Stellar interferometry with integrated optics**

Optical long baseline interferometry is a technique that is undoubtedly providing high angular resolution observations in optical astrophysics. Fizeau in 1868 [25] was the first to attempt using interferometry for astronomical observations, without reaching the wanted result, eventually proposed and revised by Stéphane in 1874 [26]. Nevertheless, only in 1921, Michelson and Pease [27] first succeeded in measuring stellar diameters with a single telescope equipped with a pupil mask. The schematics and characteristics of their apparatus are reported in **Figure 14**. Unfortunately, their interferometer was not that sensitive to allow further investigation. In fact, a 1.0 milliarcsecond diameter on the sky translates to 0.5 μm in optical path difference (OPD), on a *B* = 100 m baseline (see **Figure 14**).

In practice, modern direct interferometry only started in 1975 with Labeyrie [28] who produced stellar interference with two separated telescopes.

Modern long baseline interferometry requires the combination of several stellar beams collected from different apertures (telescopes). The first interferometers started working with only two apertures, such as GI2T [29], SUSI [30], PTI [31], IOTA [32], COAST [33] and NPOI [34, 35]. The increase in the number of apertures was one of the major features of new generation interferometers. Today, the principal operational interferometric observatories, which use this type of instrumentation, include VLTI [35–45], and CHARA [37].

Current projects are using interferometers to search for extrasolar planets, either with nulling techniques, by astrometric measurements of the reciprocal motion of the star or through direct imaging.

In **Figure 15**, the geometry of the ideal interferometer is reported. Let us specify the incident source flux power F in units of energy incident per unit time per unit crosssectional area, and the collecting area of the apertures A1 and A2 as A. Then, apart from some efficiency factors, the detected power *P* expresses as per the following:

$$P = 2AF\left(1 + \cos\ k(s\ \ B + d\_1\ \ d\_2)\right),\tag{7}$$

#### **Figure 14.**

*(a) Scheme of the two slits mask experiment from Michelson and Pease. A star, with an α angular diameter, is imaged after its light passes through a double slit mask, with B as the slit distance. An interferogram appears as a function of the optical path difference (OPD), with the first minimum at OPD = λ/2, where λ is the wavelength of the impinging light. The fringes disappear when the OPD overcomes the source coherence length, e.g. (b) and (c) show different coherence length interferograms.*

#### **Figure 15.** *Ideal stellar interferometer schematics.*

where *k* = *2π/λ*, *s* is the unit vector normal to wavefront propagation and *B* is the baseline vector between the two apertures.

In the space of relative delay *D* = *s\* B* + *d1 \* d2*, *P* varies harmonically between zero and 2*AF* with period *λ*. It is important to recall that for conventional imaging, the limiting angular resolution *α* follows the well-known relationship *α λ/D*, (where *D* is

### *Integrated Optics and Photonics for Optical Interferometric Sensing DOI: http://dx.doi.org/10.5772/intechopen.103770*

the telescope pupil diameter and *λ* is the wavelength), therefore, in comparison, an interferometric system provides the measure of interference fringes between two beams at higher angular frequencies, in the order of *α λ/B*.

The complex visibility of these fringes is proportional to the Fourier transform of the object intensity distribution (Van-Cittert Zernike theorem), hence allowing to resolve particulars, very narrow from the angular point of view. Following these principles, stellar interferometry is offering to present days astronomers the ability to study celestial objects in unprecedented detail. It is possible to see details on the surfaces of stars and even to study celestial bodies close to a black hole [46] (**Figure 16**).

Stellar interferometry has become even more effective due to the advent of high sensitivity detectors and of large aperture telescopes. Nevertheless, to implement it, a complex system of mirrors has normally to be set up to bring the light from the different telescopes, constituting the synthetic aperture, to the instruments, where it is combined and processed (see scheme in **Figure 17**). This is technically demanding,

**Figure 16.**

*(a) Conventional star image, (b) same with two telescopes stellar interferometry having baseline* B *= 10*D*, and, (c) corresponding interference pattern profile.*

#### **Figure 17.**

*Schematics of IO apparatus required for two telescopes stellar interferometry. Light from the same source collected by telescopes 1 and 2 is injected in 2x2 IO beam splitters/combiners 1 and 2, respectively. One output port of each combiner is used to perform photometric adjustment whereas the second ports are combined in a third IO 2x2 beam splitter/combiner, at which output ports the interferometric signal is collected and processed.*

**Figure 18.**

*(a) Fibre pig-tailed IO y-branch mounted on mechanical support, (b) SEM picture of the IO Y-junction, allowing beam splitting as well as fibre signal combining.*

as the light paths have to be set equal to within 1.0 μm over distances of a few hundred metres, in order to avoid an OPD offset out of the coherence length.

For ground-based interferometers, the source phase is corrupted by atmospheric turbulence. This prevents an absolute measurement of the source phase. However, it is possible to measure the difference in the source phase between two wavelengths. In practice, stellar interferometry requires star tracking techniques to compensate for astronomic seeing due to atmospheric turbulence.

In recent years, integrated optics and photonics technology, inherited from the telecom field and micro-sensor applications, was proposed for astrophysical interferometry. Results obtained with components coming from micro-sensor application were first presented by Berger et al. in a seminal series of dedicated works [38–44]. These authors demonstrated the validity and feasibility of the integrated optics technology for astronomical interferometry, by using telecom fibre coupler/combiners. Following a complete laboratory characterisation of the optical properties of the applied IO components, a first set-up was tested at the Infrared Optical Telescope Array (IOTA) observatory, in Arizona.

The above-mentioned studies demonstrated that beam combiners are very stable and lead to precise measurements. Moreover, IO components are versatile and easy to handle. In particular, the number of optical alignment adjustments strongly simplifies, which dramatically reduces the complexity of multiple-beam combinations for aperture synthesis imaging (**Figure 18**).

Other examples of IO based stellar interferometers are present by the VLTI, where the VINCI apparatus, based on IO beam combiners and fibre optics components, has allowed astronomers to reach the unprecedented resolution of 4.0 milliarcseconds in sky observations [46].

#### **5.2 IO Mach-Zehnder micro-interferometers for earth and space remote sensing**

Absorption or emission spectroscopy is largely adopted for remote sensing in both Earth and Space exploration, on board dedicated satellite platforms.

In this case, all general resources (weight, encumbrance, energy consumption, etc.) are particularly limited, furthermore, the onboard instrumentation is exposed to harsh environmental conditions (vibrations, ultra-high vacuum, radiation, temperature gradients, etc.). For these reasons, IO devices can represent a very important solution, particularly when based on monolithic structures (**Figure 19**).

Integrated scanning micro-interferometers with Mach–Zehnder geometry, have been designed and produced by using MEOS (Micro Electro Optical Systems) technologies.

*Integrated Optics and Photonics for Optical Interferometric Sensing DOI: http://dx.doi.org/10.5772/intechopen.103770*

**Figure 19.**

*(a) Carbon fibre telescope integrating three IO micro-interferometers. (b) Detail of an integrated MZ microinterferometer equipped with front-end optics, readout electronics and packaging, ready to use (the overall package length is 12 cm).*

The obtained micro-devices are based on integrated optical waveguides on LiNbO3 (LN) crystals, electrically driven, without moving parts, by exploiting the electrooptical properties of the material. These IO devices are Fourier Spectrometers in that they operate the Fourier Transform of the input radiation spectral distribution, which is eventually recovered starting from the output signal by means of Fast Fourier Transform (FFT) techniques.

Such micro-interferometers weigh a few grams, require a power consumption of a few mW and, in principle, can operate in the whole LN transmittance range (0.36 μm–4.5 μm).

In the literature several works have been reported [12, 47, 48] describing the development of a whole series of micro-interferometric apparatuses, demonstrating in principle the applicability of IO MEOS technology for Space exploration and Earth remote sensing. The micro-interferometers were produced on x-cut LiNbO3 singlecrystal substrates, by applying non-conventional micromachining techniques, based on high-energy particle beams processing.

Performances were particularly tested in the 0.4 μm–2.5 μm spectral window, with some extension also in the 2.5 μm–4.5 μm range. In the Visible region 0.4 μm–0.7 μm this microsystem demonstrated a spectral resolution suitable for detecting the

**Figure 20.** *(a) Solar radiation interferogram, (b) corresponding FFT (solid), reference (dot).*

**Figure 21.**

*(a) Raw interferogram as obtained from an integrated scanning MZI, and (b) the absorption analysis of the NO2 analyte, (lower curve), introduced in a wide band light (upper curve).*

characteristic lines of the solar spectrum together with the absorption bands of common gases present in Earth's atmosphere (see **Figures 20** and **21**).
