JWST NIRSpec's Cosmic Ray Experience at L2

TL;DR

JWST NIRSpec dark exposures reveal how cosmic rays interact with detectors behind shielding at L2 and inform calibration strategies. The study details the detector chain (two Teledyne H2RG HgCdTe arrays) behind Mo shielding, the typical hit footprint (~7.1 pixels) and energy deposition (~$6~\mathrm{keV}$) corresponding to ≈$5200$ electrons, and a linear energy transfer of ~ $0.86~\mathrm{keV~\mu m^{-1}}$, with the energy-to-DN conversion involving $E_g=hc/\lambda_\mathrm{co}$ and detector gains $g_c$ and $g_\mathrm{pp}$; the IRS$^2$ readout is noted for improving CR detection. The observed shielded hit rate decreased from approximately $4.3$ to $2.3~\mathrm{ions~cm^{-2}}~s^{-1}$ over the first ~3 years, and solar-cycle expectations suggest rising rates to ~$4.3~\mathrm{ions~cm^{-2}}~s^{-1}$ by early 2027 and potentially ~$6~\mathrm{ions~cm^{-2}}~s^{-1}$ in the early 2030s, including rare snowball hits and secondary showers whose origins may include heavy ions or shielding-induced secondaries; implications for calibration, observing efficiency, and mission planning extend to future infrared facilities such as the Nancy Grace Roman Space Telescope.

Abstract

We characterize cosmic ray interactions in blanked-off \JWST NIRSpec ``dark'' exposures. In its Sun/Earth-Moon L2 halo orbit, \JWST encounters energetic ions that penetrate NIRSpec's radiation shielding. The shielded cosmic ray hit rate decreased from approximately $4.3$ to $2.3~\mathrm{ions~cm^{-2}}~s^{-1}$ during the first three years of operation. A typical hit affects about 7.1~pixels necessitating mitigation during calibration and deposits around $6~\mathrm{keV}$ in the $λ_\mathrm{co} = 5.4~μ$m HgCdTe detector material (equivalent to $\sim5200$ charges). The corresponding linear energy transfer is about $0.86~\mathrm{keV~μm^{-1}}$. As we are currently near solar maximum, galactic cosmic ray flux is expected to increase as solar activity declines, leading to an anticipated rise in the NIRSpec rate from $2.3$ to $4.3~\mathrm{ions~cm^{-2}}~s^{-1}$ by early 2027 and potentially reaching $\sim6~\mathrm{ions~cm^{-2}}~s^{-1}$ in the early 2030s. We investigate rare, large ``snowball'' hits and, less frequently, events with secondary showers that pose significant calibration challenges. We explore their possible origins as heavy ions, secondary particles from shielding, or inelastic scattering in the HgCdTe detector material. We discuss the implications of these findings for future missions including the Nancy Grace Roman Space Telescope.

Figures (3)

• Figure 1: JWST is in a "halo" orbit about the Sun/Earth-Moon L2 Lagrange point. L2 is about $1.5\times10^6~\mathrm{km}$ from the earth, outside the Moon's orbit, and on the opposite side of the Earth from the Sun. Credit: Based on a figure from Evans2003
• Figure 2: Noon-midnight cross section of the near-Earth environment. Magnetosphere plasma extends out to about the bow shock looking toward the sun and trails further downstream on the anti-Sun side. JWST's L2 halo orbit is about $235~\mathrm{R}_\oplus$ to the right of the earth. This is deep space, where the earth's magnetosphere provides essentially no shielding. The Advanced Composition Explorer ( ACE) is located at L1. This is also deep space. We would expect ACE to experience a similar GCR environment to JWST. The GOES satellites are in geostationary orbits. There, the Earth's magnetic shielding provides some protection. We therefore expect JWST and ACE to see similar GCRs. One might reasonably expect GOES to see somewhat fewer GCRs on account of the weak, but still present protection from the earth's magnetosphere.
• Figure 3: NIRSpec has two Teledyne H2RG HgCdTe near infrared detector arrays. Light enters from the top. Cosmic rays come in from all sides. Those that come through the bottom pass through about $\approx$7 mm of molybdenum, a thin balanced composite structure (BCS), thin layers of epoxy, a thin silicon readout integrated circuit (ROIC), and thin indium bumps before reaching the HgCdTe detector layer. The molybdenum (Mo) package is included in the shielding on the back side of the detector. The BCS is a proprietary component that Teledyne uses to minimize thermally induced strain. Everything except the molybdenum base is ignored in the shielding calculation.