Development of cartridge type 1.5THz HEB mixer receivers H. H. Chang 1, Y. P. Chang 1, Y. Y. Chiang 1, L. H. Chang 1, T. J. Chen 1, C. A. Tseng 1, C. P. Chiu 1, M. J. Wang 1 W. Zhang 2, W. Miao 2, S. C. Shi 2 D. Hayton 3, J. R. Guo 3 Institute of Astronomy and Astrophysics, Academia Sinica Purple Mountain Observatory, Nanjing, China Netherlands Institute for Space Research Delft University of Technology
Outline Introduction NbTiN and NbN HEB mixers. Backside LO injection Preliminary results of Cartridge development Summary and future plan
Greenland telescope (GLT) Observation: --VLBI (submm) -- Single dish observation: submm/thz=> multipixel THz receivers is required 2011/Oct-2012/Mar.
Receiver noise temperature of HEB mixers DSB niose temeprature(k) 2000 1800 1600 1400 1200 1000 800 600 400 200 10h /K B quasioptical, phonon cooled waveguide, phonon cooled quasioptical, diffusion cooled waveguide, diffusion cooled 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 Frequency (THz) Ref: A.D. Semenov et al. Supercond. Sci. Technol., 2002 (review paper); W. Zhang et al. APL, 2010
HEB mixer @ASIAA Started at 2011 Collaborator: Dr. Sheng-Cai Shi @Purple Mountain Observatory, Nanjing, China Superconducting bridge: NbTiN or NbN Bridge size: 1.5mm(w)x200nm(l) Au slot antenna Au contact pad NbTiN substrate Cross section of Bridge SiO 2 15 nm NbTiN Native SiO 2 SiO 2 TEM BF (450 kx)
NbTiN and NbN films by DC sputtering Nb/Ti target: 70/30 at. % working pressure: 7.5 mtorr Growth temperature: R.T. Deposition rate: ~10 A 0 /sec Ar/N2 flow rate: 12/3.5 sccm Nb target working pressure: 7.7 mtorr Growth temperature: R.T. Deposition rate: ~13 A 0 /sec Ar/N2 flow rate: 12/3.5 sccm Square resistance ( ) 200 150 100 50 NbTiN film d~15nm Onset Tc~10.2K DTc~0.4K Square resistance ( ) 250 200 150 100 50 NbN film d~15nm Onset Tc~11.1K DTc~0.6K 0 8 9 10 11 12 13 Temperature (K) 0 8 9 10 11 12 13 14 15 Temperature (K)
HEB-1.4T, fabrication process 1. Deposition of NbTiN or NbN by dc reactive magnetron sputtering 2. Patterning of alignment marks by photolithography substrate substrate substrate NbTiN or NbN deposition Alignment marks Contact pads 13nm~15nm,on HR Si substrate photolithography, Ti/Au:5/40nm EBL, RF clean: 5mins, in-situ Ti/Au:5/40nm (by E-beam evaporator) 3. Fabrication of contact pads by EBL substrate antenna EBL, Cr/Au:5/150nm and followed by photolithography, Ti/Au:10/300nm 4. Patterning of antenna with EBL and photolithography substrate Bridge width EBL, SiOx/Al: 20/7nm, RIE: 20+10+20+20 sec, Al removed 5.Foramiton of etching mask with EBL substrate 6. RIE etching substrate
Measurement system @ASIAA Cold load Beam splitter LNA Circulator HEB block LO Parabolic mirrors Window: 12mm Mylar IR filter: Zitex Beam splitter: 12mm Mylar mixer block: designed by PMO IF output DC bias Si lens chips
NbTiN HEB Mixers NbN HEB Mixers Trec (K) 6000 5000 4000 3000 2000 1000 @1.5THz Trx 0-2.5-2.0-1.5-1.0-0.5 0.0 0.5 1.0 1.5 2.0 2.5 Voltage (mv) P hot P cold 3 2 1 IF Power (mw) Trec (K) 14000 12000 10000 8000 6000 4000 2000 0 Trx @1.5THz -2-1 0 1 2 Voltage (mv) P hot P cold 5 4 3 2 1 IF Power (mw) Trec (K) 4000 3000 2000 NbTiN, 1.4THz twin-slot (by spectrum analyser) ponits, by tuable bandpass filter @ 1.4THz (Data by SRON) Trec (K) 20000 15000 10000 @1mV @2mV @ 0.85THz (Data by PMO) 5 GHz 1000 5000 0.5 1 1.5 2 2.5 3 IF frequency (GHz) 0 0.1 1 10 IF Freqency (GHz) 3.5 GHz
Native SiOx layer cleaning Device I h Device II Electron, Te :L:180nm, W:1.6mm t ep t pe L:220nm, W:1.6mm No cleaning. Phonon, Tp The SiOx layer on surface of substrate is cleaned. Au t es NbN SiOx Si Au Substrate, Tb Au NbN Si Au
Cross section image of Device II: contact pads No cleaning Cleaning Au Au Ti Ti NbN NbN SiO x Si SiO x Si Sample with cleaning show thinner SiOx layer.
Cross section image of Device II: bridge No cleaning Cleaning SiO 2 SiO 2 NbN NbN SiO x Si SiO x Si Sample with cleaning show thinner SiOx layer.
IF bandwidth @ 0.85 THz No cleaning IF bandwidth~3 GHz@1mV IF bandwidth~4 GHz@2mV Cleaning IF bandwidth~3.5 GHz@1mV IF bandwidth~5 GHz@2mV 15000 4 GHz 20000 5 GHz 10000 1mV 2mV 15000 Trec (K) 5000 Trec (K) 10000 5000 0 0.1 1 10 IF Freq. (GHz) 3 GHz 3.5 GHz 0 0.1 1 10 IF Freq. (GHz)
LO LO coupling Gas laser, P LO =~1mW Solid state P LO =~30uW Beamsplitter 3.5um Mylar 12um Mylar Trx 1200K 1400K Trx, corrected loss of beamsplitter 1130K 1150K Proposed LO injection: back side injection Si lens Si lens LO coming beam signal coming beam chip Trx will not be limited by beamsplitter for required LO pumping
Backside LO injection mixer block Front-side View Backside View IF output DC bias signal lens LO lens Front-side Piece chips on signal lens Backside Piece LO lens
Backside LO pumping 0.7 0.6 unpump pumping IV @1.5THz 0.5 Current (ma) 0.4 0.3 0.2 0.1 0.0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 Voltage (mv) HEB-1.4T-1-13_BOE-C --absorbed pumping power, P~72 nw (estimated with the isothermal method)
Pumping power estimation 2 nd reflection: loss:30% 1 st reflection: loss:30% Si lens Si lens LO coming beam signal coming beam Truncation of beam: loss:75% chip Coupling from bacside~9% (simulation results: radiation to the backside, air side, is 9%), Ref: D. F. Filipovic etal. IEEE Trans. on Microwave Theory Tech, 1993 The reflection and truncation of beam can be improved but theoretically, the Gaussian beam coupling from backside would be less than 9%. We are looking for other approach.
THz cartridge LO injected from the outside of cryostat through the vacuum window Beam splitter Window: 2mm HDPE IR filter: Zitex Beamsplitter: 12um Mylar LO Parabolic mirrors Cold load
THz cartridge Trx (K) 7000 6000 5000 4000 3000 2000 1000 The corrected Trx~ 1400 K, with removing the optic loss of uncoated Si lens. HEB mixer cartridge, with 12mm Mylar beamsplitter NbN HEB Uncoated Si lens 1.45THz 1.48THz 1.50THz 1.53THz 0.5 1.0 1.5 2.0 2.5 3.0 Voltage (mv)
Trx correction, cartridge receivers Uncorrected Trx=~2000K@1.5THz Beamsplitter (12mm Mylar) Vacuum window (2mm HDPE) IR filter (Zitex) Air loss Loss (db) -0.45-0.5-0.45-0.25-1.5 Gain 0.90 0.89 0.9 0.95 0.7 Uncoated Si lens@2.8k Trx, corrected (K) 1771 1544 1360 1278 894 *Air loss: 50cm, 50%humidity (absorption coefficient=~0.095m -1 @1.5THz) --cold load is higher than 77K due to higher condensed moisture near liquid nitrogen surface -- mixer noise is high, thinner NbN film is required
Toward multipixel HEB mixer receivers Item Note Mixer Chip LNA LO source Cartridge body Cold Optics Cryostat WCA Back end Electronics Based on single-pixel (PMO) design 0.5-4 GHz, provider: Caltech >15uW (goal:30uw), 1.45-1.55 THz, provider: VDI ALMA type (modified if necessary) Start from band10 optics design Quasioptcial coupling with using a Si lens array For single cartridge LO warm driver, provider: VDI Mixer bias, help from PMO Fist phase: use 8x 2GHz sub-bands Monitor and control, provider: help form PMO
Optics: schematic diagram M1 Beam splitter M3 M2 Lens array Mixer chip LO M4 Start from ALMA band10 optics. Quasioptical coupling with a Si lens array. LO injection: quasioptcial method, LO horn and M4 at 110K stage
Proposed LO configuration: Feed in at Q-band to reduce power loss. Drawback: X2x2 stage will generate 3W heat power. Q-band Feedthrough x 2 x 2 PA X 3 X 3 1550GHz bias X 3 110K 172GHz 300K DC feedthrough 43GHz cryostat synthesizer 14.3GHz
LO power estimation for multipixel Schematic diagram Beamsplitt er (12mm Mylar) IR filter (Zitex) Twin slot antenna integrate d with Si Lens Gain 0.1 0.9 0.9 0.5 Horn-tolens coupling r1 w *horn-to-lens coupling is a rough estimation, suggested by A. Gonzalez 170 160 Pixel-to-pixel distance:5mm; pixel diamter:4.5mm Available 1.5THz LO source (solid state source): P=30uW. Power density~ 0.47P 0 Recieving power (nw) 150 140 130 120 110 100 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 Beam waist (mm)
HEB mixer chips: --The Trx~ 1000 K, 1400 K and IF bandwidth of 1.5 GHz, 3.5 GHz, for the NbTiN and NbN HEB mixer, respectively. HEB mixer cartridge: --The corrected Trx~ 1400 K at 1.5 THz, with removing the optic loss of uncoated Si lens. Backside LO injection: --Theoretically, the Gaussian beam coupling from backside would be less than 9%. Future plan: Summary and future plan --We will develop 4-pixel HEB mixer cartridge because of the limitation of LO pumping power. The pixel number can be extended later if the LO power can be pushed up by vendor or using other technology.
Thank You for Your Attention