9 Luglio, 2003 LTD 10 – Genova, Italia 1

Parameter Comparison for Low-NoiseMo/Au TES Bolometers

Dominic BenfordHarvey MoseleyJohannes StaguhnChristine Allen

NASA / Goddard Space Flight Center

Jay ChervenakThomas StevensonWen-Ting Hsieh

LTD 10 – Genova, Italia 29 Luglio, 2003

Context

! Developing bolometers for far-IR astronomy! Building arrays of SQUID-multiplexed bolometers for

SOFIA/SAFIRE, ASTRO/SPIFI, GBT/3mmCam! Related presentations:

" Staguhn et al. Detector noise analysis talk R05" Chervenak et al. Detector fabrication description Y18" Allen et al. Detector fabrication description Y24" Dicker et al. GBT/3mmCam instrument talk W03" Moseley et al. 2D (semiconducting) array talk Y06

LTD 10 – Genova, Italia 39 Luglio, 2003

Bolometer Requirements

! Parameters:" Tbath ~ 300mK (±20%)" Power ~ 1pW (±3x)" NEP ~ 10-17 W/√Hz" τ ≤ 10ms" Absorbs at ~100µm-1mm

" TC ~ 450mK (±10%)" RN ~ 400mΩ (±20%)" G(TC) ~ few x 10-11 W/K (±3x)

TC

Tbath

G

C

(Absorber)

τ = C/G

Optical Power

LTD 10 – Genova, Italia 49 Luglio, 2003

Mechanical Implementation

! 1x8 linear arrays of Pop-Up Bolometers

! Folding allows arrays to be close-packed

! Thermal isolation high, since flexure hinges are thin

! SHARC-II (12x32=384 pixel)semiconducting Pop-Up array (Moseley et al.)

LTD 10 – Genova, Italia 59 Luglio, 2003

History of our TES bilayers

! Both Mo/Au and Mo/Cu fabricated

! Good performance demonstrated, but not sufficiently repeatable

! Always some excess noise, often out-of-band, sometimes 5x-10x thermal noise limit

LTD 10 – Genova, Italia 69 Luglio, 2003

Approach: Boundary Conditions

! Current flowing in TES can take many possible paths

! Initial trials indicated that normal metal bars on sides could help constrain possible paths, leading to more orderly current flow

! Other constraints may be more likely to prevent moving “chunks” of superconducting regions

Lindeman et al. model

Superconducting

Normal

LTD 10 – Genova, Italia 79 Luglio, 2003

Trial #1: Longitudinal Bars

! In order to better understand the effect of boundary condition geometry on detector performance, we produced several nearly identicalbolometers with different normal metal bars.

! Several designs of longitudinal bars produced

! Some designs are wedged to remove isotropy

#0

#1

#2

#3

#4

LTD 10 – Genova, Italia 89 Luglio, 2003

Trial #2: Transverse Bars

! Several designs of transverse bars produced

! Some designs are interdigitated; others are continuous

#0

#1

#2

#3

LTD 10 – Genova, Italia 99 Luglio, 2003

Transition Measurements

! 4-wire measurements of resistance vs. temperature, with several bias currents to estimate self-heating and critical currents.

! Addition of transverse bars does cause sharpness to degrade, butstill achieve α≈100

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.45

0.50

0.500 0.501 0.502 0.503 0.504 0.505 0.506 0.507 0.508 0.509 0.510Temperature (K)

Ch 0

Ch 1

Ch 2

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.425 0.430 0.435 0.440 0.445 0.450 0.455Temperature (K)

Ch2 cut

Ch3 cut

Longitudinal bars Transverse bars

LTD 10 – Genova, Italia 109 Luglio, 2003

Behavior on Bias (I)

! I(V) curves taken on all devices; calibrated to determine resistance and power as a function of base temperature.

! Some calibration uncertainties remain.

0

5

10

15

20

25

30

35

40

0 1 2 3 4 5 6 7 8 9 10TES Bias Voltage (µV)

C0R4

C0R0

C0R2C0R1

C0R3

0

2

4

6

8

10

12

14

16

18

20

0 1 2 3 4 5 6TES Bias Voltage (µV)

C1R2C1R3

C1R0

C1R1

Longitudinal bars Transverse bars

LTD 10 – Genova, Italia 119 Luglio, 2003

Behavior on Bias (II)

! Resistance vs. bias yields close to expected values.! Some calibration uncertainties remain.

Longitudinal bars Transverse bars

0

100

200

300

400

500

600

0 1 2 3 4 5 6 7 8 9 10TES Bias Voltage (µV)

C0R4C0R0

C0R2

C0R1C0R3

0

50

100

150

200

250

300

350

400

450

500

0 1 2 3 4 5 6TES Bias Voltage (µV)

C1R2

C1R3

C1R0

C1R1

LTD 10 – Genova, Italia 129 Luglio, 2003

Behavior on Bias (III)

! Power vs. bias yields similar value for all pixels (as should be the case).! Power on transition quite constant (as should be the case).! Some calibration uncertainties remain.

Longitudinal bars Transverse bars

0

20

40

60

80

100

120

140

160

180

200

0 1 2 3 4 5 6 7 8 9 10TES Bias Voltage (µV)

C0R4

C0R0

C0R2

C0R1

C0R3

0

10

20

30

40

50

60

70

80

90

100

0 1 2 3 4 5 6TES Bias Voltage (µV)

C1R2C1R3

C1R0C1R1

LTD 10 – Genova, Italia 139 Luglio, 2003

Pixel Thermal Conductance

! Thermal conductance measured by average power on transition for many bath temperatures

! Power law fit:" TC=443mK" G(TC) = 6·10-10 W/K (higher than desired)" Index G=G0TN, N~2 (though not tightly

constrained)" Psat = 84pW (with Tbath=0)." Phonon noise: 6·10-17 W/√Hz

! Next device mask will reduce leg width, to reduce Psat and G(TC) by ~10x.

0

5

10

15

20

25

30

35

40

45

350 360 370 380 390 400 410 420 430 440 450Temperature (mK)

TC

= 443 mKG(T

C) = 0.58 pW/mK

Index ~ 2P

sat = 84 pW

LTD 10 – Genova, Italia 149 Luglio, 2003

Dynamic Measurements

! Measured complex impedance Z(ω) for bolometers, as perLindeman et al.

! Results prove tricky to interpret; devices too fast for reliable high frequency impedance measurements

! Time constants of 1/τe~5kHz.

-90

-80

-70

-60

-50

-40

-30

-20

-10

0

10

-80 -60 -40 -20 0 20 40 60

Imag

inar

y Im

peda

nce

(mž)

Real Impedance (mž)

SuperconductingZ=24.0mžZ=32.0mžZ=40.0mžZ=52.5mž

10 kHz

~100 Hz

5.5 kHz

4.5 kHz

4.0 kHz

4.0 kHz

X

X

X

X

X

XX

XXX

XXXXXXXXXXXXXXX

XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

-500

-400

-300

-200

-100

0

100

200

300

-400 -300 -200 -100 0 100 200 300 400

0mž

34mž

44mž

55mž

68mž

81mž

96mž

112mž

129mž

147mž

166mž

187mž

208mž

231mž

255mž

281mž

307mž

335mž

363mž

393mž

X 400mž

410mž

421mž

421mž

LTD 10 – Genova, Italia 159 Luglio, 2003

Noise Measurements

! The subject of Johannes Staguhn’s talk; in order not to spoil the surprise:

The geometry of the bars does affect the noise.

0

50

100

150

200

250

300

0 5000 10000 15000 20000 25000Frequency (Hz)

0

50

100

150

200

250

300

0 5000 10000 15000 20000 25000

Cur

rent

Noi

se (p

A/¦H

z)

Frequency (Hz)

LTD 10 – Genova, Italia 169 Luglio, 2003

Optical Measurements

! Bismuth absorbing film deposited on bolometers to impedance math to free space.

! Dark tests after deposition show same result as tests prior to deposition.

! Optically functioning bolometers are demonstrated in the dark; prior tests have verified optical performance.

Optical measurement of a blackbody calibrator from an earlier, similar bolometer.

LTD 10 – Genova, Italia 179 Luglio, 2003

Summary

! Several nearly identical bolometers produced, with different normal metal bars. Bolometers are optically active detectors.

! Tests of quasi-DC parameters (TC, RN, α, Psat, etc.) show good consistency.

! Noise performance can be improved by changing bar geometry.

! Measurements & analysis ongoing to quantify improvements and differences. Time constant (impedance) analysis to be refined.