1.1

UVI instrument requirements and scientific goal

The scientific motivation for SMILE-UVI is to image the consequences of the solar wind-magnetosphere interaction and provide the most complete/highest resolution global images of the aurora ever taken. Data from SMILE-UVI will usher in a new era of geospace research. For the first time, researchers will track the causal chain of space weather through geospace, from the solar wind driver to its ultimate consequences in the radiation belts and ionosphere. Being the only global auroral imager during the mission timeframe, its data will be used for a multitude of reasons by researchers around the world.

Figure 1:

The aurora is powered by the solar wind/magnetosphere interaction, produced by magnetospheric plasma processes, and affects the ionosphere and upper atmosphere.

As discussed extensively in Donovan et al. [1], after nearly four decades of auroral imaging from space the result has been images with spatial resolution typically around 100 km, with temporal resolution of 30 seconds to two minutes (or longer). The historically low SNR of UV imagers leaves low-light level emissions undetected, due in part to low suppression of out-of-band signal results in low daylight suppression. Although auroral imaging from space has been tremendously important in the evolution of geospace science, in a very real sense we have only just begun to capitalize on the richness of opportunities that auroral imaging from space has to offer (see Figure 2). We have only limited information about the auroral distribution of time and space scales and how that distribution evolves in time. Our understanding of how the spatial extent of aurora of different types (e.g., patchy, Alfvenic, etc.) evolves in response to changing geospace conditions is very limited and in most cases comes from the partial view afforded by ground-based instruments. It is mandatory to increase the knowledge on key parameters such as the open flux in the polar cap through long duration geomagnetic events such as Steady Magnetospheric Convection and magnetic storms, which is one of the main goal of the SMILE mission.

Figure 2:

After nearly four decades of auroral imaging from space the result has been images such as those shown here (left) provided by the Canadian UV imager on the Swedish Viking satellite [2] and (right) the Canadian built UV “WIC” imager on the NASA IMAGE satellite [3]. While UV imaging from space has been tremendously important to geospace science, improving the imaging quality especially in the dayside is very important.

For SMILE, the baseline requirements for the UV images of Earth’s aurora have been set at 150 km or better spatial resolution from locations on orbit greater than 19 RE geocentric. The further requirements on detection of the dayside cusp and the polar cap boundary (from apogees) drive the high level specifications listed below;

Table: 1

High level SMILE-UVI specifications

To reach the spectral requirements of SMILE-UVI specific coatings are being developed. These filters are one of the key elements of the mission, directly coupled to the science objectives. These filters will be implemented as 4 reflective optical coatings made of thin film multilayers deposited on each mirror of the UVI. The objective of these filters is to select the scientific waveband between 160 and 180 nm. Out-of-band, the combined four mirrors have to give a rejection ratio as high as possible to reject light from solar diffusion, dayglow and unwanted atomic lines. The expected rejection ratio, all components combined (detector included) has to be in the range of 10^-8 – 10^-9.

3.1

Thin film materials

This section presents the refractive index and extinction coefficient of each material involved in the coatings. They are depicted on Figure 4 to Figure 6. Due to a lack of information in the UV-visible range above 250 nm for LaF3, only Far-UV range is presented. However, a number of additional ellipsometry measurements in the UV-visible range (from 190 nm to 800 nm) are foreseen in the near future to be performed at CSL to evaluate the filter performances in the visible.

Figure 4:

Optical constants of MgF2 from [10] (dots) and interpolated data (continuous line).

Figure 5:

Optical constants of LaF₃ from [10] (dots) and interpolated data (continuous line).

Figure 6:

Optical constants of Al from [10] (dots) and interpolated data (continuous line).

3.2

Coating spectra optimized for normal incidence

Thin film multilayers have been optimized according to π-multilayer equation for different H/L ratio and for different incidence angles. Reflectance spectra in the FUV region corresponding to multilayers optimized for normal incidence and specific values of H/L ratio are depicted below for each type of coatings.

Figure 7:

Reflectance spectrum for a [MgF₂/Al]₂₅ multilayer with H/L=400 at normal incidence.

Figure 8:

Reflectance spectrum for a [LaF₃/MgF₂]₂₅ multilayer with H/L=0.3 at normal incidence.

3.3

Angular dependency for a given coating stack

One of the main concern of interference filters is their dependencies with the incidence angle. While speaking about wide field telescope, one can be worried by the wide range of ray angles hitting the optical surfaces. Coating angular sensitivity has then to be considered. Coatings are optimized for a specific angle, let say θ = 10° in the present simulations which corresponds to the average incidence angle on the first mirror. However, ray analysis shows that minimum and maximum incidence angles are quite different, ranging from θ = 5° to θ =15° while keeping the same coating configuration, obviously. It turns out that coating performances are robust when the angle of incidence θ varies. There is no significant decrease of central wavelength throughput. The main effect is a wavelength shift of the spectral curve. In both coating types, the increase of the AOI blueshifts the curve maximum as shown on Figure 9.

Figure 9:

Reflectance spectra for a multilayer with H/L=0.3 optimized for θ=10°, at several angles of incidence θ.

4.1

Methodology

Based on the optical design of the UVI instrument, a first analysis of the angles of incidence distribution is performed for every field. This analysis is done with Code V. Along the rays and for each ray generated, a set of four angles of incidence (AOI) is calculated. These AOI will then be used to generate the spectral properties of the reflected beam through the optical telescope. The calculation is based on the fact that a ray contains all the spectral content and no wavelength dispersion occurs in the instrument. The spectral properties due to reflections along the rays are combined by multiplying the spectral response, while these responses are averaged over all the rays defining the fields and distributed over the entrance pupil. Results for central field is represented on Figure 10. Other fields have been analyzed. While central field has a symmetric profile around average angle because of on-axis instrument, off-axis fields present non-symmetric angle profile as depicted on Figure 11 and Figure 12.

Figure 10:

Distribution of ray incidence angles reflected on the four UVI-mirrors. The represented field is the central one (0,0).

Figure 11:

Angle distribution on the four UVI mirror for field (0,5).

Figure 12:

Angle distribution on the four UVI mirror for field (3.5,3.5).

4.2

Throughput evaluation

Using the ray analysis preformed in the previous section and the coating performance evaluation tool presented in section 3, the throughput of each ray entering into the instrument can be evaluated. Ray throughput is averaged over the whole rays set entering through the instrument pupil. Each useful field is also analyzed. The following graphs (Figure 13 & Figure 14) present the spectral response of each individual reflection on the mirrors considering the specific angle of incidence associated to a specific ray. The dashed line corresponds to the equivalent filter response by all four mirrors and normalized to 1 mirror. Four specific rays are represented among a set of thousand rays.

Figure 13:

[LaF3/MgF2] Spectral property of the individual reflection through the optical system. Four specific rays are represented among thousand. M1 to M4 reflectivity are represented for these specific rays. Dashed line corresponds to a hypothetic filter including the effect of the four.

Figure 14:

[LaF3/MgF2] all rays represented after four reflections – Field 0,0

Figure 14 (left) represents the coatings response along all thousand rays after the four mirror reflections. The right-graph is the averaged responses on the (0,0)- field focal point. Results are presented in log-scale to highlight the depth of the coating reflectivity. On Figure 15 is represented the equivalent filter combining all the coating effects on the four mirrors. The same methodology has been applied on useful fields of view of the UVI-instrument. It is important to note that the filter response is quite similar for the three first mirror while the spectral reflectivity response of the fourth one is blueshifted. This effect is easily explained by the average angle of incidence which is approximately twice the angle of incidence of the three other mirrors. Figure 16 represents the spectral response for all fields.

Figure 15:

[LaF3/MgF2] Equivalent filter corresponding to the combination of the four mirrors contributions reduced to one mirror.

Figure 16:

[LaF3/MgF2] Spectral throughput for all fields.