MOIRCS grisms

Available grisms and sensitivities

MOIRCS has been operating MOIRCS spectroscopic mode with two low-resolution grisms and a medium-resolution grism. From S08B we will add two high-resolution VPH grisms for users, under shared-risk policy. Users who are considering the use of the VPH grisms, please visit the webpage.

Two low-resolution "zJ500" and "HK500" grisms are characterized by its wide wavelength coveage and high throughput. The zJ500 grism is optimized for the observation of 0.9-1.78 micron regions, while the HK500 grism is optimized for 1.4-2.5 microns. Their resolution is designed to roughly R~500.

The medium resolution "R1300" grism is for a better observability of target lines between the atomospheric OH-line forest, especially in H-band wavelength region. We can use the grism for each JHK window by the use of JHK imaging filters for order-sorting purpose. The orders used for observation are 4th (partly 3rd) for J, 3rd for H, and 2nd for K. The grism effeciency become worse for shorter wavelengths (higher order). The actual spectral resolution is measured to be roughly 1500 under 0".5 width slit.

Estimated sensitivities listed are for a 5 sigma detection per pixel in 1 hour of on-source background-limited spectroscopy. They are calculated using the measured MOIRCS effeciency shown in Figure 5 considering the telescope effeciency and the atomospheric loss. The average sky brightness at each wavebands are used for calculation, and so these values should be treated as a crude estimate.

If the single line detection is the main aim on the observation, higher spectral resolution can work better in some cases as the sky OH lines can be separated more than R500 grism. And also, if the intrinsic line width is narrower than the resolution elements in R1300, higher resolution will work better. This is because the fraction of background emission relative to the target line flux per resolution element can be reduced for R1300, thus higher contrast of lines can be expected. The sensitivity for R1300 is actually be comparable compared with the case for R500 grating in H, and even better for K band. The actual performance may be updated occasionally.

Grism name	Operating range [um]	Resolution (0.5'' slit)	Dispersion [A/pixel]	Sensitivity (Vega magnitude) [mag/arcsec²]
zJ500	0.9-1.78 (*3)	700 @ J	5.57	J=19.2
HK500	1.3-2.5 (*4)	640 @ H 820 @ Ks	7.72	H=17.8 K=17.6
R1300 (*1)	1-2.5	1500 @ J(*2) 1600 @ H 1500 @ K	1.91 @ J 2.61 @ H 3.88 @ K	J=17.8 H=16.7 K=17.1

(*1) All values for R1300 grism are the converted value to 0".5 slit width from the measurements by 0".8 slit. Background-limited performance has been assumed here, although it is actually not true.
(*2) Extrapolated from H-band data. Actual measurement value is not available.
(*3) Without the order-sorting filter it will be 0.75-1.5 um, though the sensitivity below 0.8 um is very low and the 0th-order thermal light contaminates the spectra significantly.
(*4) See Note (3) below.

Note
(1) The straight-through wavelengths are 1.12 um for zJ500 and 1.93 um for HK500. The value for R1300 grism is 1.135 um for J, 1.502 um for H, and 2.240 um K, respectively (calculated values). Note that these values will shift slightly with the position of slits across the dispersion direction.

(2) The spectra will be shifted by +22 pixels to the spatial direction (+y direction in raw image) from the slit image position for R1300 grism (channel-1 side only). In wmdp_moircs software it is currently not displayed. For the channel-2-side R1300 grism there is no shift. Also, no move of the spectra across the spatial direction is seen for zJ500 and HK500 (both channel).

(3) Previously the zJ500 grism was supposed to use without filter, but that introduced the contamination of the second-order spectra at >1.5 um (namely, the detector has the sensitivity from ~0.73 um). In order to get a "clean" spectra to H-band window we installed a new order-sorting filter in October 2006, which limits the spectral range to the value we put on the table. If you need the information shorter than 0.9um and does not need the spectra longer than 1.5um, you can use the zJ500 grism without filter.

(4) We replaced the old OC_1.3 order-sorting filter to new one in 2007, because it suffered from strong second-order light at longer than 2.35um. The wavelength coverage for HK500 grism with new OC_1.3 is 1.3-2.5um without any second-order light. The existing order-sorting filter with thermal cut (OC_HK: 1.3--2.3um) is also available. The users who want to cut very strong thermal light component longer than 2.4um should use OC_HK. The overall background level of OC_HK is well below that of the OC_1.3 filter, which is the advantage of the use of OC_HK filter.

(5) You can take the H-band spectral data by both zJ500 and HK500 gratings. Sensitivity is 10-40% better for HK500 grating especially for red side, though the spectral resolution is a bit better for zJ500 grating.

(6) You can use the JHKKs imaging filters for spectroscopy of the zJ500/HK500 grisms for more muitiplicity.

(7)Y-band filter is not available for spectroscopy, because it is currently installed in the grism turret. We may re-install Y filter to different turret before S08B if possible.

Figure 4: (Left) The efficiencies of the zJ500 and HK500 grisms (from datasheet by the company). (Right) The actual measured values of the resolution for grisms at various wavelengths (Tokoku et al. 2007 Ph.D. thesis). OH night sky lines are used for measurement.

Figure 5a(left) & 5b(right): Total MOIRCS spectroscopic efficiency for zJ500 / HK500 grisms. Channel-2 data with 2"-width slit is used for evaluation. Note that the effeciency curves around the wavelength ranges with high atomospheric absorption clearly suffer from the residual of the absorption correction. Also note that the turn-over feature seen the longest edge of the spectra of the zJ500 grism is due to the artificial pattern seen in the detector. Figure 5b(right): Total MOIRCS spectroscopic efficiency for the R1300 grism with the JHKKs filters. Channel-2 data with 2"-width slit is used for evaluation. The turn-over feature seen the longest edge of the spectra of the H+R1300 grism is due to the artificial pattern seen in the detector.

Figure 5c: The order-sorting filters for zJ500 / HK500 grisms (installed mid October 2006) as well as the JHK filters for R1300 grism. For HK500 we also have the filter without thermal cut.

Images by courtesy of Dr.Tokoku.
Please note that all data on these pages are subject to change as the evaluation of the performance of MOIRCS progresses.

Created by Yuka Katsuno Uchimoto (Mar 2006)
Updated by Ichi Tanaka (2008-02-05)