PowerPoint Presentation

“Mind the Gap” session

243

AAS Meeting, 9 January 2024

UV CMOS Detectors for

CASTOR and Beyond

Chaz Shapiro

sponsorship acknowledged.

JPL Collaborators:

John Hennessey, Michael Hoenk, April Jewell,

Todd Jones, Shouleh Nikzad

Thanks to Pat Côté (NRCC, CASTOR science PI)

for slide contributions

CASTOR

The Cosmological Advanced Survey Telescope for Optical and uv Research

CASTOR Mission Overview

 Light-weighted Zerodur 1m primary mirror

 Three Mirror Anastigmat with 0.25 deg

FoV

 Active M2 for WFE compensation

 Fine steering mirror for image stabilization

 1063 kg spacecraft and 10 Gbps optical downlink

 MAC200-small-SAT bus

 800 km polar SSO for efficient surveys

 Dichroic separation of wavebands

 Optical coatings on all mirrors for red leak control

 Minimum 5 year mission (Goal:10 years)

 Combination of Legacy Surveys (64%), Guest Observer

programs (25%), Target-of-Opportunity programs (8%),

and calibration time (3%)

 14 candidate Legacy Surveys advanced to SRL4 during

Phase 0

 To be revisited once the partnership is finalized

“Our highest recommendation at the very large investments scale is for CASTOR,

an exciting mission with a broad and compelling science case, and which would be

Canada’s first marquee space astronomy mission.”

— Canadian Astronomy Long-Range Plan 2020

Focal planes

Acronym!

CASTOR

The Cosmological Advanced Survey Telescope for Optical and uv Research

Imaging and UV Spectroscopy

 Wide-field UV/blue-optical imaging

 3 x 310 Megapixel arrays sampled at 0.1”/pixel

 Simultaneous imaging in three channels

 Back-illuminated large-format CMOS detectors with ALD AR

coatings and 2D-doping

 Two spectroscopic modes

 Deployable grism (150-400nm, R~350) over full FOV

 UVMOS with DMD selector (150-300nm, R~1000-2000) over

207” x 117” (offset from FOV)

 Grism and UVMOS instruments make up roughly one third of the time

requested for legacy surveys

 Grism spectroscopy is used in 6/14 surveys

 UVMOS spectroscopy is used in 7/14 surveys

 Spectroscopic science drivers include the progenitors of kilonovae, the

evolution of the cosmic web, AGN reverberation mapping, flare rates in

M dwarfs, the surface chemistry of Trans Neptunian Objects, and more

(Spectroscopic)

SN Type II

SN Type Ia

CASTOR detector requirements

• Low dark current < 0.01 e-/px/s

• Low read noise < 6 e- rms

• Low power dissipation

• Pixel pitch <= 10 µm

• Ability to do 10 Hz fine guiding on main array

without disrupting science integrations

• Large, buttable arrays to reduce chip gaps and

number of detectors

• Radiation hard

 CMOS solution

• High UV QE  2D doping and AR-coating

Teledyne-e2v CIS301

9K x 8.6K pixels ; 10µm pitch

Teledyne-e2v CIS120

2K x 2K pixels ; 10µm pitch

~1 Gpx focal plane divided into 3

channels/bands

• Detector coatings chosen by

CASTOR to maximize in-band QE

for each channel

• Out-of-band light rejected by

dichroics and mirror coatings

Nominal bands with detector AR coatings

FUV

150-300

NUV

300-400

VIS

400-550

JPL Engagement in CASTOR

❑ CASTOR partners (NRCC, Honeywell, ABB) have

been working closely with JPL since 2017, with the

ultimate goal of securing a NASA partnership.

❑ Science JPL/IPAC scientists made key

contributions to the 2019 CASTOR Science

Maturation study, serving as co-leads on 7 of 8

science working groups for the Phase 0 study.

(POC: Jason Rhodes)

 Technology JPL’s Advanced Detector group and

Teledyne-e2v have delivered 2D doped and AR

coated CIS120 prototypes to CSA as part of their

Space Technologies Development Program (POC:

Chaz Shapiro)

CIS120 wafer inspection

FUV coating models

2D-Doping: Expanding design space for UV missions of all sizes

• Detectors are the heart of an instrument, driving major mission design

trades: mass, power, optical configuration, observing strategy, cost

• Solid state detectors continue to grow in scale, sensitivity, and capability.

Choose from: CCD (incl. Skipper, EMCCD), CMOS, APD, SPAD...

• …but they are not natively UV-sensitive

• JPL’s 2D-doping process converts Si-based devices into UV detectors,

allowing UV missions to take full advantage of latest technologies. 2D-

doping and subsequent coatings:

• provide high, stable QE in UV

• are agnostic to device architecture – process on backside doesn’t “see” frontside

• allow QE to be tailored to required wavelength bands

• are radiation tolerant (Hoenk et al. 2014)

jpl.nasa.gov

• Precision coatings via Atomic Layer Deposition

• Single and Multilayer coatings can tailor response in and out of the desired

band.

• Coatings deposited on detector eliminate standalone filters and reduce optical

complexity.

Antireflection (AR) Coatings and Metal Dielectric Filters

(MDF) optimize QE

Jewell, et al., Proc. SPIE 8820 (2013) 88200Z

Nikzad, et al., Applied Optics, 51, (2012) 365

Solar blind MDF

Narrow band AR

Broadband AR

See talks by

April Jewell

John Hennessey!

The Future: Explorers and

Flagships

https://asd.gsfc.nasa.gov/luvoir/design/

Instrument Baseline Detector

High Definition Imager (HDI) Delta-doped CMOS array;

200-1100nm

Ultraviolet Multi-Object

Spectrograph (LUMOS)

Delta-doped CMOS array;

200-400nm (MCP 100-200)

Pollux (European Contribution) Delta-doped CCDs;

97-390nm

ECLIPS (Coronograph) Delta-doped EMCCD; 200-

525nm

https://www.jpl.nasa.gov/habex/

Instrument Baseline Detector

Workhorse Camera (HWC) Delta-doped CCD/CMOS;

150-950nm

UV Spectrometer Delta-doped EMCCD

(backup); 115-300nm

Coronograph Delta-doped EMCCD;

450-670nm

Starshade Delta-doped EMCCD;

200-450nm

LUVOIR

• 2D-doped detectors are baselined

or backups for several instruments

• Working with industry to develop

devices that meet challenging

requirements

• E.g. LUVOIR needs large, buttable

arrays with <7µm pixels and fast,

low noise readout.

(Some wavelength

ranges shown span

multiple channels)

LUVOIR & HabEx

Habitable Worlds Observatory!

Backup

Notional Schedule

CASTOR

The Cosmological Advanced Survey Telescope for Optical and uv Research

TDAMM Science with CASTOR UVMOS

Applications of intermediate-resolution UVMOS spectroscopy for

TDAMM science

 Supernovae spectra: characterization of CSM and progenitor stars from

Fe II, Ni II and Mg II absorption depths, plus the height and slope of the

UV continuum

 Spectroscopic follow up of kilonovae and other transients out to

distances of 500 Mpc

 UV emission-line observations of tidal disruption events

 Characterization of super-luminous and infant supernovae from UV

continuum measurements and equivalent widths for lines of ionized Si,

C, Mg and Ti between 220 and 270 nm

CASTOR

The Cosmological Advanced Survey Telescope for Optical and uv Research

Galaxies in the Cosmic Web with CASTOR

EAGLE simulation: McAlpine et al. (2016)

 Grism and UVMOS slit-less grism spectra will capture

redshifts for 1000s of Lyman-α emitters (LAEs) and

Lyman Break Galaxies (LBGs) per pointing

 CASTOR grism redshifts will trace the cosmic web to

z~3

 This will allow us to study galaxy evolution in the full

range of cosmic environments to 11 Gyr before the

present

Partnerships and Collaborations

 NASA Desired contribution: CASTOR’s three WFI cameras (see below). Additional contributions

are possible and welcome

 UKSA Participation approved in Jan. 2023 through a new bilaterals program in space science

 Contributions under discussion with CSA, including detector qualification, testing, electronics; cryo-

cooler; filters; coatings; secondary; data flows

 Detector procurement discussions are ongoing, focusing on a separate UKSA funding stream

 India/ISRO In December 2021, ISRO had expressed strong interest in a partnership

 Expected contributions: launch, UVMOS, ground stations

 A timely agreement could not be reached. CASTOR is now seeking other partners to provide these

items

 Spain/France Strong interest in the UVMOS (formerly an Indian contribution).

 South Korea Discussions underway

jpl.nasa.gov

The Silicon UV Problem

Passivation by 2D-Doping shrinks the dead layer

Hoenk et al., APL, 61 (1992) 1084

Nikzad et al., Proc. SPIE 4139 (2000) 250

2D-doping provides the maximum

possible QE, limited only by photon

transmission into the silicon.

“delta” doping =

Delta function profile

Dead layer

shallow absorption

= poor QE

jpl.nasa.gov

2D-doping Process: Molecular Beam Epitaxy (MBE)

• Device wafer (up to 8”) bonded to handle wafer to protect

frontside structures and add stability

• Backside surface is thinned to epitaxial Si layer

• New Si deposited by e-beam evaporation

• “Delta” layer of dopant added: P-type doping (B) or N-type (Sb)

• Additional Si cap added

• OR repeat for “superlattice” doping for extra stability

• Deposited silicon layers typically 1-3 nm

• Wafer is diced, devices are completed with packaging

Veeco Gen 200

Wafer in MBE