The JWST NIRCam grisms obtain R ~ 1,600 slitless spectroscopy at 2.4–5.0 µm in 2 science observing modes: wide field and single-object time series. The grisms may also help align JWST's mirrors.

On this page

NIRCam has grisms available for R = λ/Δλ ~ 1,600 slitless spectroscopy at long wavelengths (2.4–5.0 µm). These may be used during commissioning to help align JWST's mirrors (Shi et al. 2008) and afterward for science observations. Table 1 lists the properties of the grisms.

Each NIRCam module has 2 grisms installed on the long wavelength (LW) channel, which disperse spectra in orthogonal directions, along detector rows (grism R) and columns (grism C). For each observation, a grism in the LW pupil wheel is used in combination with a filter in the LW filter wheel. Broader filters yield wider spectral range and longer spectra (in detector pixels) at the expense of higher background and source confusion.

Table 1. NIRCam grism properties

Parameter	Value
Dispersion	1 nm/pixel
Resolving power, R = λ/Δλ	~1,600 at 4 µm
Undeviated wavelength (measured)	3.95 µm (3.92–3.98)

Observing modes

Two observing modes listed in Table 2 make use of the grisms: wide field slitless spectroscopy and grism time series. In most cases, only first order spectra are visible. Fainter second order spectra of some sources may be visible when F250M, F300M, F277W, or F322W2 filters are used in combination with the grism, though only the wide filters can be used for the grism time-series mode.

Table 2. NIRCam grism observing modes

Mode

Available filters

Usage

Wide field slitless spectroscopy

Medium: F250M, F300M, F335M, F360M, F410M, F430M, F460M, F480M

Wide: F277W, F322W2, F356W, F444W

For a wide field with a multitude of (possibly overlapping) spectra

Grism time series

Wide: F277W, F322W2, F356W, F444W

To monitor bright, isolated, time-varying objects

Resolving power

The grism spectral resolving power, R = λ/Δλ, is ~1,600 at 3.95 µm (the undeviated, zero-order wavelength), and decreases to R ~ 1,150 at 2.5 µm. It can be approximated by the function (shown in Figure 1) given in Section 3.2 of Greene et al. (2017):

R ~ 3.35(λ/µm)⁴ − 41.9(λ/µm)³ + 95.5(λ/µm)² + 536(λ/µm) - 240,

which is the result of a spline curve fit to the optical model prediction of the grism FWHM spectral resolving power R versus wavelength for point sources.

The FWHM spectral resolving power R = λ/Δλ ~ 1,600 at 3.95 μm, limited somewhat by the circular beam factor and diffraction. At shorter wavelengths, the resolving power is limited further by pixel sampling of the PSF, but this may be improved by dithering multiple observations. The resolving power can be approximated by R≃3.35 (λ∕μm)⁴-41.9(λ∕μm)³+95.5 (λ∕μm)² +536(λ∕μm)−240 to within a few percent. © Greene et al. (2017).

Dispersion

In each NIRCam module, the 2 grisms, grism R and grism C, disperse spectra along detector rows and columns, respectively. This dispersion does not change measurably with wavelength, and changes are small with field position (2% or less). Dispersion directions (increasing wavelength) are given for NIRCam ideal coordinates (X_ideal, Y_ideal) in Table 3 and Figure 2. Table 4 shows the dispersion lengths measured in pixels for the available filters, with the dispersions spanning roughly 1,000 pixels per micron of wavelength (Figure 3). Dispersion lengths in pixels are determined between the 5% power limits at wavelengths λ₁ and λ₂ and converted to pixels (1 nm/pixel).

Table 3. NIRCam grism dispersion directions

Element	Dispersion within:	Dispersion direction in module A	Dispersion direction in module B	Measured dispersion (nm/pixel)
Grism R	Rows	+X_ideal →	-X_ideal ←	1.004
Grism C	Columns	+Y_ideal ↑	+Y_ideal ↑	1.006

Dispersion directions (increasing wavelength) for the row (R) and column (C) grisms along the ideal coordinate axes (X_ideal, Y_ideal) in each module. Adapted from Figure 4 in Greene et al. (2017).

Table 4. NIRCam grism dispersion lengths in filters for the 2 modes

Filter	λ₁	λ₂	Dispersion Length (pixels)	Observing mode
F277W	2.395	3.179	784	Wide Field + Grism Time Series
F322W2	2.413	4.083	1670	Wide Field + Grism Time Series
F356W	3.100	4.041	941	Wide Field + Grism Time Series
F444W	3.835	5.084	1249	Wide Field + Grism Time Series
F250M	2.401	2.609	208	Wide Field
F300M	2.800	3.205	404	Wide Field
F335M	3.142	3.606	464	Wide Field
F360M	3.386	3.869	483	Wide Field
F410M	3.810	4.366	556	Wide Field
F430M	4.140	4.426	286	Wide Field
F460M	4.486	4.788	302	Wide Field
F480M	4.614	5.048	434	Wide Field

Table note:

Example dispersion lengths of grism spectra in various filters, measured in pixels between the 5% power limits at wavelengths λ1 and λ2. This numbers vary slightly, at a few percent levels over the field of view. The filters available for the grism time-series and wide field slitless spectroscopy modes are shown in Table 2.

The 4 wide filters available for both observing modes are shown, along with all the medium filters for WFSS. Broader filters produce longer spectra and only first order spectra are visible in most cases. Fainter second order spectra of some sources (shown in grey) may be visible when using F250M, F300M, F277W, or F322W2. The red boxes show the full detector width (2,040 photo-sensitive pixels) along the x-axis, while the y-axis extent of the box and the y-placement of each spectrum is arbitrary (note that WFSS is offered only with SUBARRAY = FULL). In this example, the source is placed at the center of the x-axis (x=1024, marked by an X symbol and the dotted line). This corresponds to the location of the undeflected wavelength of 3.95 µm, or alternatively would correspond to the direct image position if the grism was not in the beam. The thick portions of the spectra footprints correspond to wavelengths for which the total throughput is larger than 50% of the maximum throughput for that setting. The thinner part corresponds to wavelengths for which the total throughput is larger than 1% of the maximum throughput for that setting. In this figure, it is possible to appreciate the displacement of the NIRCam grism spectra with respect to the direct image, as well as the relative position of spectra obtained using different filters. Note that sources that are not at the center of the field of view will produce spectra that are translated with respect to those shown in the figure. It is recommended to review the NIRCam WFSS field-of-view page to learn how to optimize the source placement within the field of view according to the science needs.

Sensitivity

Figures 4 and 5 show NIRCam module A first and second order sensitivities that include contributions from the JWST and NIRCam optics, filters, detector quantum efficiency (QE), and grism. The module B grism is less sensitive by ~30%, and is not shown.

NIRCam module A, grism R, first order sensitivities including JWST and NIRCam optics, filters, detector quantum efficiency (QE), and grism. The module B grism is less sensitive by ~30% (not shown).

NIRCam module A second order sensitivities including JWST and NIRCam optics, filters, detector quantum efficiency (QE), and grism. The module B grism is less sensitive by ~30% (not shown).

Throughput

The module A grisms have higher throughput efficiency due to anti-reflection coating on both sides. Module B grisms are only coated on the smooth (non-grooved) side, resulting in 30% lower transmission as well as ghosts (faint reflections) of some bright spectra.

Second order throughputs of the module A and B grism and available filters (module A filters shown, module B is similar), including all JWST and NIRCam optics and detector quantum efficiencies. The grism throughput must be multiplied with that of a selected filter.

Grism throughput tables for both modules and for both 1st and 2nd orders can be downloaded here. Note that these are grism-only throughputs. Filter-only throughputs, can be dowloaded here. To obtain the full system throughputs, the grism-only throughputs need to be interpolated at the same wavelengths as the filter-only throughputs and multiplied by the filter-only throughputs.

Hardware

Fabrication of the NIRCam silicon grisms is described by Jaffe et al. (2008) and Gully-Santiago et al. (2010). The grism diameter is 48 mm, with 42 mm optically usable. The spacing between grooves is 15.36 µm. The blaze angle is offset slightly from the prism angle, as it was chosen to maximize efficiency near the mean wavelength of the LW modules (3.7 µm), which is offset from the undeviated wavelength (3.95 µm). Table 5 lists the basic manufacturing specifications of the grisms.

Table 5. NIRCam grism hardware properties

Parameter	Value
Prism angle	6.16°
Blaze angle	5.75°
Blaze wavelength	3.7 µm
Groove frequency	65 grooves/mm
Maximum thickness	8.0 mm
Diameter	48.0 mm, circular with side flats
Clear aperture	42.0 mm, circular
AR coating	2.4 to 5.0 µm
Peak transmission	70.5% at 3.7 µm (module A) 52% at 3.7 µm (module B)

NIRCam grism photograph — © Greene et al. (2010).

References

Greene, T., Beichman, C., Gully-Santiago, M. et al. 2010, SPIE 7731
NIRCam: development and testing of the JWST near-infrared camera

Greene, T. et al. 2017, JATIS, 035001
λ = 2.4 to 5 μm spectroscopy with the James Webb Space Telescope Near-Infrared Camera

Gully-Santiago, M., Wang, W., Deen, C. et al. 2010, SPIE 77393S
High-performance silicon grisms for 1.2-8.0 μm: detailed results from the JWST-NIRCam devices

Jaffe, D. T., Wang, W., Marsh, J. P. et al. 2008, SPIE 70103L
Fabrication and test of silicon grisms for JWST-NIRCam

Shi, F., King, B. M., Sigrist, N. et al. 2008, SPIE 70102E
NIRCam Long Wavelength Channel grisms as the Dispersed Fringe Sensor for JWST segment mirror coarse phasing

Notable updates	08 Jul 2024 Pre-launch figures removed and updated 25 Sep 2017 Edited information to match recent paper (Greene, T. et al. 2017, JATIS, 035001)
Originally published	21 Dec 2016

JWST User Documentation

NIRCam Grisms