1

CCATCCAT InstrumentationInstrumentation

Gordon Stacey representing the Gordon Stacey representing the efforts of many people involved in efforts of many people involved in

CCAT instrumentation studiesCCAT instrumentation studies

What is CCAT? A 25 m submm telescope that will operate at wavelengths A 25 m submm telescope that will operate at wavelengths

as short as 200 umas short as 200 um• Why 25 m?Why 25 m?

Matches ALMA continuum sensitivity in short submmMatches ALMA continuum sensitivity in short submm Significantly breaks confusion limit of smaller aperturesSignificantly breaks confusion limit of smaller apertures High altitude, smooth surface, large aperture High altitude, smooth surface, large aperture > 10 times > 10 times

more sensitive than current single dish facilities more sensitive than current single dish facilities Located in the Atacama desert in northern Chile at very Located in the Atacama desert in northern Chile at very

high elevation - 5600 mhigh elevation - 5600 m much of the time has PWV < 0.5 mmmuch of the time has PWV < 0.5 mm

Its location enables maximal synergy with ALMAIts location enables maximal synergy with ALMA• Locates sources for ALMA follow-upLocates sources for ALMA follow-up

Takes advantage of rapid growth in submm detector Takes advantage of rapid growth in submm detector technology to map large regions at high angular technology to map large regions at high angular resolutionresolution

22

What will we see?What will we see? Primary science Primary science

• Exploration of the Kuiper BeltExploration of the Kuiper Belt• Star and planetary system formation Star and planetary system formation • Sunyaev-Zeldovich EffectSunyaev-Zeldovich Effect• Surveys of star forming galaxies in the early Surveys of star forming galaxies in the early

UniverseUniverse These science topics emphasize wide-field These science topics emphasize wide-field

imaging – hence our first light instruments will imaging – hence our first light instruments will include camerasinclude cameras

Studies of primordial galaxies requires redshifts – Studies of primordial galaxies requires redshifts – we also include direct detection spectrometers we also include direct detection spectrometers

33

Resolving the Origins of the Resolving the Origins of the Cosmic Far-infrared BackgroundCosmic Far-infrared Background

44

Dole et al. (2006) and S. Dole et al. (2006) and S. OliverOliver1. 50% of the extragalactic background radiation is in the

FIR/submm2. Only a fraction the CFIRB has been accounted for with

galaxies3. The FIR/submm luminosity function must evolve strongly for

z > 0.

1. 50% of the extragalactic background radiation is in the FIR/submm

2. Only a fraction the CFIRB has been accounted for with galaxies

3. The FIR/submm luminosity function must evolve strongly for z > 0.

CCAT’s CCAT’s wavebandwaveband

CCAT, Hershel and ALMACCAT, Hershel and ALMA

URSI 2011 – CCAT Instrumentation55

Approximate FOV of first-light camera

ALMA primary beam(~7)

Simulated maps of the same patch of sky based on Simulated maps of the same patch of sky based on HerschelHerschel number number countscounts

At 450 At 450 m, CCAT and ALMA will have m, CCAT and ALMA will have approximately the same mapping approximately the same mapping

speed speed per beamper beam. .

With the 5With the 5 FOV first-light camera and FOV first-light camera and ~8,500 beams~8,500 beams, CCAT’s mapping , CCAT’s mapping speed will be ~8,500x higher.speed will be ~8,500x higher.

Identifying the Highest Redshift Identifying the Highest Redshift SourcesSources

66

>5>5 850 µm detection, 350 µm 850 µm detection, 350 µm nondetectionsnondetections

77

Baseline CCAT Instrumentation Baseline CCAT Instrumentation

Three Primary Science InstrumentsThree Primary Science Instruments• Submillimeter wave cameraSubmillimeter wave camera• Near millimeter wave cameraNear millimeter wave camera• Multi-object direct detection spectrometerMulti-object direct detection spectrometer

Z-spec Z-spec ZEUS/ZEUS-2ZEUS/ZEUS-2

• Transferred, and future instrumentationTransferred, and future instrumentation Full FoV camerasFull FoV cameras Heterodyne spectrometers/arraysHeterodyne spectrometers/arrays

88

Submm Camera: SummarySubmm Camera: Summary We envision a > 50,000 pixel submm camera at We envision a > 50,000 pixel submm camera at

first lightfirst light Primary band is 350 Primary band is 350 m ~ 40,000 pixels m ~ 40,000 pixels 5’ FoV 5’ FoV

• Filter wheel to access 450, 620, (200) Filter wheel to access 450, 620, (200) mm Dichroic splits off a long wavelength 850 Dichroic splits off a long wavelength 850 m bandm band

• Or perhaps more likely we will have an Or perhaps more likely we will have an (independent) mm wave camera for 740 (independent) mm wave camera for 740 m and m and longer wavelengthslonger wavelengths

• At least 10,000 pixels at longer wavelengthsAt least 10,000 pixels at longer wavelengths Advanced Technology Array Camera Advanced Technology Array Camera

ATACameraATACamera

99

Submm Camera Decision Tree – Submm Camera Decision Tree – Field of ViewField of View

The telescope delivers up to 1º FOV – why are we The telescope delivers up to 1º FOV – why are we designing to a 5’ FOV?designing to a 5’ FOV?• Science: Science: Initial science deliverable with 5’ FOV camerasInitial science deliverable with 5’ FOV cameras

• Image Scale: Image Scale: One can not couple the entire 1One can not couple the entire 1 FoV into a FoV into a background limited camera background limited camera smaller sub-systems smaller sub-systems

• Technology: Technology: Current technology suggests 40,000 pixels is Current technology suggests 40,000 pixels is a reasonable goal – this delivers Nyquist sampled images a reasonable goal – this delivers Nyquist sampled images over a 5’ over a 5’ 5’ FOV at 350 5’ FOV at 350 mm

tiling a 30’ FOV requires tiling a 30’ FOV requires one million pixelsone million pixels at 350 um, at 350 um, -- extremely expensive using today’s technologies-- extremely expensive using today’s technologies

Future developments will greatly reduce the Future developments will greatly reduce the costscosts – – therefore mega pixel cameras are postponed therefore mega pixel cameras are postponed

1010

Two Designs ConsideredTwo Designs Considered All reflective designAll reflective design

• Maximizes throughputMaximizes throughput• Minimizes emissivityMinimizes emissivity• Off-axis approach leads to BIG (3-4 m class) optics – Off-axis approach leads to BIG (3-4 m class) optics –

but 5’ FoV design not too bad…but 5’ FoV design not too bad… Transmissive design with field lensTransmissive design with field lens

• System is remarkably more compactSystem is remarkably more compact• Throughput and emissivity quite goodThroughput and emissivity quite good

Direct imagingDirect imaging• Would be fine at 200 um, over-sampled at longer Would be fine at 200 um, over-sampled at longer

wavelengthswavelengths• Problems with stray light…Problems with stray light…

1111

ATACameraATACamera Cornell – Caltech – Colorado collaborationCornell – Caltech – Colorado collaboration First light camera composed of sub-camerasFirst light camera composed of sub-cameras Dichroic 5’ field of view on CCATDichroic 5’ field of view on CCAT

• 40,000 pixel at 350 40,000 pixel at 350 m and 10,000 pixels at 850 m and 10,000 pixels at 850 mm FoV broken into 3 – 3’ “sub” fields (128×128): minimizing both FoV broken into 3 – 3’ “sub” fields (128×128): minimizing both

aberrations and window sizeaberrations and window size Final version could have several more sub-cameras Final version could have several more sub-cameras

1212

Ray-traceRay-trace Spot sizes quite good – circles are Airy diskSpot sizes quite good – circles are Airy disk Sub-cameras have Strehl ratios > 90% over nearly entire Sub-cameras have Strehl ratios > 90% over nearly entire

FoV (centered at angle of 0.07FoV (centered at angle of 0.07))

DetectorsDetectors Our preliminary design base-lined TES sensed SQUID Our preliminary design base-lined TES sensed SQUID

multiplexed arrays as in SCUBA-2multiplexed arrays as in SCUBA-2 Workable, within budgets for 40,000 pixel cameraWorkable, within budgets for 40,000 pixel camera Submm MKID devices are now the preferred optionSubmm MKID devices are now the preferred option

• Considerably less complex architecture that is more Considerably less complex architecture that is more readily scalable to large arraysreadily scalable to large arrays

• Considerably less complex read-out electronics as well.Considerably less complex read-out electronics as well.

1313

Considerably less cost

1414

MKID PrinciplesMKID Principles Photon detector is Photon detector is

incorporated into a incorporated into a superconducting resonator superconducting resonator circuitcircuit

Photon absorption causes Photon absorption causes the frequency and line-the frequency and line-width of the resonator to width of the resonator to changechange

Frequency domain Frequency domain multiplexing achieved by multiplexing achieved by designing resonators with designing resonators with slightly different resonant slightly different resonant frequencies and using a frequencies and using a broadband low noise broadband low noise microwave amplifier to microwave amplifier to read out the arrayread out the array

1515

Array Development at JPLArray Development at JPL

Lumped-element 350 µm direct absorption MKID pixel spiral inductor/absorber and an inter-digitated capacitor

16×16 array of TiN spiral lumped-element pixels 256 pixels coupled to one feedline visible at the top and bottom

1616

200 200 m MKID Devicem MKID Device

Demonstration of TiN far-IR MKID device at 200 m illustrating the inductive (frequency shift) and dissipative (resonance width) response to temperature (Peter Day et al. )

1717

Predicted SensitivityPredicted Sensitivity

Can detect Milky Way at z ~ 1 to 2!

1818

How Many SourcesHow Many Sources

4 hours/pixel, 2000 hour survey – 144 hours/pixel, 2000 hour survey – 14 survey in 2 survey in 2 years years

Approaches half a million sources/yearApproaches half a million sources/year

Transmillimeter Wave Camera Transmillimeter Wave Camera –Sunil Golwala–Sunil Golwala

Low wavelength Camera for CCATLow wavelength Camera for CCAT Antenna-coupled arrays of bolometersAntenna-coupled arrays of bolometers

• Can’t do 50,000 feed-hornsCan’t do 50,000 feed-horns• Single polarization antenna coupled design leads to Single polarization antenna coupled design leads to

a simple way to cover multiple bands with varying a simple way to cover multiple bands with varying pixel sizespixel sizes

• Nb slot antenna and microstrip limits shortest Nb slot antenna and microstrip limits shortest to > to > 740 um (405 GHz) 740 um (405 GHz)

Beam definition achieved with phased array Beam definition achieved with phased array antennaantenna

Signal detection with either MKIDS or TES Signal detection with either MKIDS or TES devicesdevices

1919

2020Sunil Golwala

Direct Detection SpectrometersDirect Detection Spectrometers For broad-band spectroscopy of broad, faint lines, direct For broad-band spectroscopy of broad, faint lines, direct

detection spectrometers are the instruments of choice.detection spectrometers are the instruments of choice.• Detectors are not subject to the quantum noise limit and are Detectors are not subject to the quantum noise limit and are

now sufficiently sensitive to ensure background limited now sufficiently sensitive to ensure background limited performance at high resolving powersperformance at high resolving powers

• Very large bandwidths Very large bandwidths ~ ~ are possible are possible Need to consider 3 types of direct detection spectrometersNeed to consider 3 types of direct detection spectrometers

• Fourier Transform spectrometersFourier Transform spectrometers: naturally broad band: naturally broad band

• Fabry-Perot interferometers: Fabry-Perot interferometers: high sensitivity, but must scanhigh sensitivity, but must scan

• Grating spectrometers: Grating spectrometers: spectral multiplexing monochrometerspectral multiplexing monochrometer Free space spectrometersFree space spectrometers Waveguide spectrometersWaveguide spectrometers

• Niche for all systems: here we focus on grating Niche for all systems: here we focus on grating spectrometers since we are interested in maximizing spectrometers since we are interested in maximizing point source sensitivitypoint source sensitivity 2121

Glenn2222

Glenn2323

Glenn2424

Z-Spec as a Redshift EngineZ-Spec as a Redshift Engine

Broad bandwidth is Broad bandwidth is very useful for very useful for determining determining redshifts of submm redshifts of submm galaxiesgalaxies

Observed Observed (redshifted) spacing (redshifted) spacing between CO between CO rotational lines rotational lines given by:given by:

= 115 GHz/(1+z)= 115 GHz/(1+z)

2525Lupu et al. 2010

Beyond COBeyond CO

Far-IR FS lines much Far-IR FS lines much more luminous than more luminous than COCO

Redshift engineRedshift engine Diagnostics of Diagnostics of

physical conditions physical conditions of gas and radiation of gas and radiation fieldsfields

Could cover submm Could cover submm – mm windows with – mm windows with 3 Z-spec like devices 3 Z-spec like devices

Simultaneous with Simultaneous with dichroic opticsdichroic optics

2626

2727

The Redshift (The Redshift (zz) ) andand E Earlyarly UUniverseniverse

SSpectrometer:pectrometer: ZEUSZEUS

S. Hailey-Dunsheath S. Hailey-Dunsheath Cornell PhD 2009Cornell PhD 2009

““Free-Space” submm (650 and 850 GHz) grating spectrometerFree-Space” submm (650 and 850 GHz) grating spectrometer

R R // ~ 1000 ~ 1000 BW ~ 20 GHz BW ~ 20 GHz T Trecrec(SSB) < 40 K (SSB) < 40 K ZEUS on CSO for several years – single beam on the sky ZEUS on CSO for several years – single beam on the sky Upgrade to ZEUS-2 a Upgrade to ZEUS-2 a

5 color (200, 230, 350, 450, 610 5 color (200, 230, 350, 450, 610 m bands); m bands); 40 GHz Bandwidth 40 GHz Bandwidth 10, 9, & 5 beam system10, 9, & 5 beam system

ZEUS-2

Stephen Parshley

Design ChoicesDesign Choices Choose R Choose R ~ 1000 ~ 1000

optimized for detection of optimized for detection of extragalactic lines extragalactic lines

Near diffraction limit:Near diffraction limit:• Maximizes sensitivity Maximizes sensitivity

to point sourcesto point sources• Minimizes grating size Minimizes grating size

for a given Rfor a given R Long slit in ZEUS-2 Long slit in ZEUS-2

• Spatial multiplexingSpatial multiplexing• Correlated noise Correlated noise

removal for point removal for point sourcessources

Choose to operate in n = Choose to operate in n = 2, 3, 4, 5, 9 orders which 2, 3, 4, 5, 9 orders which covers the 890, 610, covers the 890, 610, 450, 350 and 200 450, 350 and 200 m m windows respectivelywindows respectively

ZEUS Windows

2 3 4 5

ZEUS spectral coverage superposed on Mauna Kea windows on an excellent night

500 400600700800 300

Wavelength (m)

3030

ZEUS-2 Traces [CII] Cooling ZEUS-2 Traces [CII] Cooling LineLine

158 um [CII] line is dominant 158 um [CII] line is dominant coolant of neutral ISMcoolant of neutral ISM

ZEUS can detect [CII] at z ~ ZEUS can detect [CII] at z ~ 1 to 2 characterizing star 1 to 2 characterizing star formation in galaxies at the formation in galaxies at the historic peak of star historic peak of star formation in the Universeformation in the Universe

ZEUS provides a unique ZEUS provides a unique opportunity to explore this opportunity to explore this epoch through the [CII] lineepoch through the [CII] line

Approximately 40% of the Approximately 40% of the submm galaxy population submm galaxy population has redshifts such that the has redshifts such that the [CII] line falls in the 350 (z ~ [CII] line falls in the 350 (z ~ 1) or 450 (z~2) 1) or 450 (z~2) m windowsm windows

ZEUS [CII] Windows

Blain et al. 2002, Phys. Rep., 369, 111

With ZEUS-2 on CSO and APEX we can extend these studies from z >4 to 0.25 -- tracing the history of star formation from 12 Gyr ago, through its peak 10 Gyr ago to the present epoch

ZEUS-2ZEUS-1

ZEUS-2 Focal Plane Array: Natural Spatial ZEUS-2 Focal Plane Array: Natural Spatial MultiplexingMultiplexing

Upgrading to (3) NIST 2-d Upgrading to (3) NIST 2-d TES bolometer arrays TES bolometer arrays

Backshort tuned Backshort tuned 5 lines in 4 bands 5 lines in 4 bands

simultaneouslysimultaneously • 215 215 m (1.5 THz)m (1.5 THz)• 350 350 m (850 GHz)m (850 GHz)• 450 450 m (650 GHz)m (650 GHz)• 625 625 m (475 GHz)m (475 GHz)

Imaging capability (9-10 Imaging capability (9-10 beams)beams)

Simultaneous detection of Simultaneous detection of [CII] and [NII] in z ~ 1-2 [CII] and [NII] in z ~ 1-2 rangerange

First light in Januar 2012 First light in Januar 2012 on CSO with 400 um on CSO with 400 um array only array only

APEX later in 2012APEX later in 2012

Spectral Imaging CapabilitiesSpectral Imaging Capabilities

M51 - CO(1-0): BIMA Song (Helfer et al. 2003)

AstrophysicsAstrophysics• [CI] line ratio: [CI] line ratio: Strong Strong

constraints on T constraints on T

• 1313CO(6-5) lineCO(6-5) line: Strong : Strong constraints on CO opacity constraints on CO opacity

• [NII] line: [NII] line: Cooling of Cooling of ionized gas, and fraction of ionized gas, and fraction of [CII] from ionized media[CII] from ionized media

Mapping AdvantagesMapping Advantages• Spatial registration Spatial registration

“perfect”“perfect”

• Corrections for telluric Corrections for telluric transmission coupledtransmission coupled

• Expected SNR for the five Expected SNR for the five lines comparablelines comparable

12CO(7-6)

13CO(6-5)

[CI] 3P2 - 3P1

[NII] 3P1 - 3P0

[CI] 3P1 - 3P0

A Long Slit Free Space A Long Slit Free Space SpectrometerSpectrometer

ZEUS-2 is in 5ZEUS-2 is in 5thth order at 350 um - BW ~ 8% order at 350 um - BW ~ 8% RP~1000 RP~1000 20 cm collimated beam 20 cm collimated beam 0.6 m dewar 0.6 m dewar Could build a 1Could build a 1stst order free space grating order free space grating

spectrometer – BW 160% spectrometer – BW 160% RP~1000 RP~1000 60 cm collimated beam 60 cm collimated beam 1.5 to 1.8 m 1.5 to 1.8 m

dewardewar Advantages Advantages

• Flat focal planeFlat focal plane• Transmits both polarizationsTransmits both polarizations• Beams “dense-packed” but readily adapted to multi-Beams “dense-packed” but readily adapted to multi-

object spectroscopy object spectroscopy 3333

Multi-Object SpectrometersMulti-Object Spectrometers Free-space spectrometers like ZEUS-2 are trivially made Free-space spectrometers like ZEUS-2 are trivially made

into 1 (or 2) - d imaging systems, so it naturally into 1 (or 2) - d imaging systems, so it naturally becomes a multi-object spectrometer if we can “pipe” the becomes a multi-object spectrometer if we can “pipe” the light in. light in.

If configured in one band (say 350/450 If configured in one band (say 350/450 m), then the m), then the usable FoV of ZEUS-2 is > 20 beams usable FoV of ZEUS-2 is > 20 beams

To avoid source confusion, could configure with 10 feedsTo avoid source confusion, could configure with 10 feeds Z-Spec’s modularity also lends itself well to multi-beam Z-Spec’s modularity also lends itself well to multi-beam

configurations through stacking of the planar configurations through stacking of the planar waveguides.waveguides.

Light Pipes: Quasi-optical ApproachLight Pipes: Quasi-optical Approach

Periscope based Multi-Object SpectrometerPeriscope based Multi-Object Spectrometer Useful for observations of sources which have a low spatial density on the skyUseful for observations of sources which have a low spatial density on the sky Patrol regions over the focal plane assigned to each receiverPatrol regions over the focal plane assigned to each receiver Low transmission losses since only four reflections Low transmission losses since only four reflections

Goldsmith and

Seiffert

Confusion Confusion [CII] = FIR [CII] = FIR Continuum Detection LimitsContinuum Detection Limits

ZEUS Survey of 24 – z ~ 1 to 2 galaxies shows ZEUS Survey of 24 – z ~ 1 to 2 galaxies shows [CII]/FIR continuum ~ 0.2% [CII]/FIR continuum ~ 0.2%

Line/continuum ~ 10:1Line/continuum ~ 10:1 CCAT confusion limit: 1 mJy CCAT confusion limit: 1 mJy

10 mJy in line × 1.9 THz/1000/(1+z) 10 mJy in line × 1.9 THz/1000/(1+z)

or 1 × 10or 1 × 10-19-19 W/m W/m22 – easily detectable (10 – easily detectable (10/4hrs) with /4hrs) with ZEUS – like spectrometers on CCATZEUS – like spectrometers on CCAT

An image slicer grating (IFU) spectrometer might well An image slicer grating (IFU) spectrometer might well be quite useful – sources are crowded be quite useful – sources are crowded

3636

Is CCAT Spectroscopy Really Is CCAT Spectroscopy Really Necessary?Necessary?

CCAT will be the source finder for ALMACCAT will be the source finder for ALMA Detect sources with CCAT continuumDetect sources with CCAT continuum Detect sources and redshifts in spectral lines with CCATDetect sources and redshifts in spectral lines with CCAT Spatially (and spectrally) resolve lines with ALMASpatially (and spectrally) resolve lines with ALMA CCAT spectrometers are competitive for line searchesCCAT spectrometers are competitive for line searches

• Transparency and dish surface wins up to 2Transparency and dish surface wins up to 2

• System temperature wins a factor of 2 to 3System temperature wins a factor of 2 to 3

• Bandwidth: 10 settings vs. 1 setting per windowBandwidth: 10 settings vs. 1 setting per window

• 25 m dish vs. 12 m – wins a factor of 425 m dish vs. 12 m – wins a factor of 4 CCAT is equivalent to 2*2.5*4 = 20 antennas CCAT is equivalent to 2*2.5*4 = 20 antennas takes takes

(64/20)(64/20)22 = 10 times longer for CCAT, but CCAT covers = 10 times longer for CCAT, but CCAT covers entire band – entire band – so it comes out evenso it comes out even

But – CCAT will have multi-object spectrometers!But – CCAT will have multi-object spectrometers!3737

3838

SummarySummary CCAT’s facility first light instruments will consist of:CCAT’s facility first light instruments will consist of:

• Submm camera with > 50,000 pixels covering > 5’ Submm camera with > 50,000 pixels covering > 5’ FoVFoV

• Mm-wave camera with > 50,000 pixels covering ~ 20’ Mm-wave camera with > 50,000 pixels covering ~ 20’ FoVFoV

• Multi-object broad band direct detection Multi-object broad band direct detection spectrometersspectrometers

In addition we expect other “contributions” includingIn addition we expect other “contributions” including• Heterodyne receivers and arraysHeterodyne receivers and arrays• Specialize direct detection spectrometers (e.g. IFU, Specialize direct detection spectrometers (e.g. IFU,

FPI)FPI)• PolarimetersPolarimeters