next up previous contents
Next: HETG Spectra
Up: AXAF News
Previous: ACIS Pileup

LETG Order Separation

Extracting a first order spectrum from LETG data is a challenge. The HRC-S detector (the prime read-out detector for the LETG) provides no energy resolution with which to separate overlapping orders. There are several approaches to separating the contribution from high orders from the first order spectrum, including both instrumental filters and modeling. The possibilities are summarized in the LETG chapter in the Proposer's Guide and the instrumental filters are discussed in the Observatory Guide. However, because the filters work in rather complicated ways, we are expanding on the discussions here. In the first two sections we present a graphic overview of the HESF ((High Energy Suppression Filter) and the LESF (Low Energy Suppression Filter). In the final section of this article, we summarize the contribution from high orders to LETG spectra.

HESF (High Energy Suppression Filter)

The HESF is summarized in Fig. 19.

The geometry is summarized in the ``High Energy Suppression Filter" article in this newsletter. Full details of the design are provided in the HRC chapter of the Observatory Guide. The figure and brief description in this article are to provide the user with a quick overview of how the filter works.

Figure 19 shows the effective area from the combination of the HRMA, the LETG grating, and the HRC-S detector. This is shown in the center of the figure for orders 1 and 3 as a function linear distance on the detector. (Order 2 is more heavily suppressed.) Only one side of the spectrum is shown; the other is similar but slightly shifted. For convenience, 4 scales have been provided at the top of the figure to show the wavelength scale for first and third order (in Å) and the corresponding energy scales. The central region (left side of the figure in the graph) has a gap between the two sections of the filter. Progressing outwards (to the right in the single-sided figure), the two parts of the filter are coated in Cr and amorphous C in turn. The physical locations of the two coatings of the reflection flat are shown immediately below the effective areas. Figs. 6.20 and 6.21 in the Observatory Guide and also Figs. 11 and 12 in this newsletter show the filter geometry.

   figure230
Figure 19: The High Energy Suppression Filter (HESF). Effective area for orders 1 and 3 is shown in the two central curves. The scale on the bottom axis is the distance in the dispersion direction in mm. The top four scales show wavelength and energy scales for both first and third orders. The horizontal line immediately under the effective area curves indicates the spatial location of the Cr and C portions of the HESF. The four horizontal lines at the bottom of the figure indicate the wavelength regions which are absorbed by each portion of the filter for each order. For any part of the filter, its effect can be determined by comparing the wavelengths it intercepts spatially with the wavelengths it absorbs. See text for further discussion.

The filter makes use of the different absorbing properties of the two coatings as a function of energy at the appropriate reflection angle. (See Figures 6.24 and 6.25 in the Observatory Guide.) The vertical lines in Fig. 19 relate the location of the parts of the filter to the wavelengths of the spectrum on the detector. Four lines at the bottom of Fig. 19 show the energies absorbed by each coating region. The top two lines show the wavelength region absorbed by Cr in first order (top) and in third order (below). Below that similar lines are shown for the C filter region.

To step through the details of the order separation, start with the vertical lines which show that in the Cr region of the filter the first order spectrum (top axis) covers approximately 33 to 56 Å. This is an energy range which is not absorbed by the Cr reflection (top dashed horizontal line below). On the other hand, the vertical lines show that in third order the spectrum in the Cr region is from about 12 to 19 Å (second axis from the top), a range which is absorbed by the Cr filter (lower horizontal dashed line), hence third order light is suppressed in this wavelength region. (At shorter wavelengths the spectrum passes through the hole in the filter and is unchanged.) Similar comparisons for the region of the filter coated in C show that first order light is virtually unaffected, while third order light is suppressed because of the respective energy ranges of the spectrum that fall on the filter.

LESF (Low Energy Suppression Filter)

   figure240
Figure 20: The Low Energy Suppression Filter (LESF). Effective area curves are shown for orders 1 and 3 in the two central curves. The scale on the bottom axis is the distance along the dispersion direction in mm. The top four scales show wavelength and energy scales for both first and third orders. The two horizontal lines immediately beneath the effective area curves indicate the spatial locations of the various regions of both the normal and the suppessed filters (labeled a to d) which differ in the thickness of aluminum and polyimide. The four horizontal lines at the bottom of the figure indicate wavelengths absorbed by each filter region. The effect of any region can be determined by comparing the wavelengths it intercepts with the wavelengths it suppresses. See text for further discussion.

LESF is summarized in Fig. 20. Full details of the geometry of the filter are given in the Observatory Guide Fig. 6.1. It is also summarized in the article ``The HRC UV/Ion Shield" in this newsletter, including Fig. 8 showing the filter geometry. Again, the example shown in Fig. 20 is the HRMA-LETG-HRC-S effective area curves with the wavelength and energy scales at the top for first and third order.

Under the effective area curves in the center of the graph, the locations of the regions of both the normal UVIS and the suppression `strip' are shown labelled a, b, c, and d as in Fig. 6.15 in the Observatory Guide HRC chapter. They are schematic, since the boundaries are not sharp in the converging beam. Under these are lines showing the wavelength region which is suppressed in each segment in first order. The limiting wavelength used is that at which the the transmission drops below 10% in Fig. 6.15.

In comparing the location of segment a in the normal filter (vertical lines show the wavelength region in first order) with the horizontal line showing the wavelength region suppressed, it is clear that segment a does not suppress the wavelengths that fall on it. Segments b and c similarly have little effect. In the d region in the LESF suppression strip, however, the location of the segment is such that wavelengths that fall on it in first order are those that are absorbed, removing light at wavelengths longer than about 55 Å. A similar comparison for the LESF shows that third order light is largely unchanged in any of the segments. This means that an observer can observe a source through the LESF to obtain a measure of the high order flux. This can be used to estimate/model the high order contribution to the spectrum through the `normal' region of the filter.

Summed High Order Contribution

   figure250
Figure 21: The partial contribution of higher orders in the LETG. The effective area as a function of the distance in the dispersion direction for first order is shown as the solid line. The dashed line is the sum of the effective areas from orders 2 through 9. Orders higher than 9 also make a significant contribution. (Figure prepared by Belinda Wilkes)

Figure 21 shows the contribution of high orders 2 through 9 to the effective area compared with the effective area in first order as a function of distance along the detector. The contribution of the high orders is substantial. Figs. 8.6 and 8.7 in the Proposers' Guide show that the effective area in individual orders only drops slowly as order number increases, so orders >9 also make a significant contribution. On the other hand, the sharp drop in effective area at high energy means that only orders up to about 25 need to be considered.

Nancy Remage Evans


next up previous contents
Next: HETG Spectra
Up: AXAF News
Previous: ACIS Pileup

Roy Kilgard
Mon Dec 22 10:05:00 EST 1997