130404.PostdocSeminar-Copyrightversion

High-power semiconductor lasers for in-situ sensing of atmospheric gasesCarl BorgentunMicrodevices laboratory (MDL), JPL/Caltech

Copyright 2013 California Institute of Technology.Government sponsorship acknowledged.

Postdoc seminar Carl Borgentun

OutlineBackground

Laser design and fabrication

Performance

Next target2013-04-04Postdoc seminar Carl Borgentun*


Background2013-04-04Postdoc seminar Carl Borgentun*


Gas sensing2013-04-04Postdoc seminar Carl Borgentun*EarthsciencePlanetary scienceSafetyCredit: NASA


Credit: Richard W. Pogge, Ohio state universityAbsorption spectroscopy basicsCredit: David Sayres, Harvard2013-04-04Postdoc seminar Carl Borgentun*http://www.astronomy.ohio-state.edu/~pogge/Ast161/Unit5/atmos.html


Laser spectroscopy basicsLaserGas1980sLiquid helium-cooled lasers (1000kg)1990sLiquid nitrogen-cooled lasers (70 kg)2000sThermoelectrically cooled lasers (0.1 kg)2013-04-04Postdoc seminar Carl Borgentun*


Example application Earth scienceCarnegie airborne observatoryCredit: Carnegie institute of science2013-04-04Postdoc seminar Carl Borgentun*Earthscience


Example application Planetary scienceMars polar landerMars science laboratoryCredit: NASA2013-04-04Postdoc seminar Carl Borgentun*Planetary science


Example application SafetyCarbon monoxide monitoring instrumentCredit: NASA2013-04-04Postdoc seminar Carl Borgentun*Safety


Ozone depletionChlorine compounds destroy ozone, but only at cold temperatures.Water vapor in the stratosphere increases the threshold temperature, at which ozone destruction can take place. Risk of thinner ozone layer not only at the poles, but also at lower latitudes where people, animals, and plants live.

2013-04-04Postdoc seminar Carl Borgentun*James G. Anderson et al., Science, 337 (6096), 835-839, 2012.Various explanations for high mixing ratios of water vapor, distinguishable by measuring water isotopologues.


Traditional absorption techniques are not sensitive enough for the simultaneous measurement of water and its less abundant isotopologues.Jim Andersons group at Harvard are developing a cavity-enhanced instrument.A high-power laser source at the right wavelength is needed.In this case: Right wavelength ~ 3777 cm-1 (2.65 m)In this case: High-power > 10 mWNeed for more sensitive instrumentCredit: NASA; Jim Anderson, Harvard2013-04-04Postdoc seminar Carl Borgentun*


Laser design and fabrication2013-04-04Postdoc seminar Carl Borgentun*


Two criteria:Optical gainCavity

Laser crash course2013-04-04Postdoc seminar Carl Borgentun*Active mediumCavityPumpingMirrorOutput beam


Molecular Beam EpitaxyCredit: JPL2013-04-04Postdoc seminar Carl Borgentun*66


Processing2013-04-04Postdoc seminar Carl Borgentun*Etch ridgeInsulate and contactEtch grating


Single-mode emission2013-04-04Postdoc seminar Carl Borgentun*


GratingDistributed feedback (DFB) laser using a Bragg grating => single-mode emission.Etched, i.e. non-metal => less loss.Laterally coupled => no epitaxial re-growth necessary.Credit: JPL2013-04-04Postdoc seminar Carl Borgentun*


SEM picturesCredit: Cliff Frez2013-04-04Postdoc seminar Carl Borgentun*


Coating, cleaving, and bondingCredit: Cliff Frez2013-04-04Postdoc seminar Carl Borgentun*Wavelength:2.0 3.5 m


Performance2013-04-04Postdoc seminar Carl Borgentun*


Power characteristicsOutput power at 35 mW @ 10C, 600 mA.Still more than 20 mW @ 30C, 600 mA.No sign of thermal roll-over.Dips in LI curves are due to absorption of ambient water (feature, not a bug!).

2013-04-04Postdoc seminar Carl Borgentun*


Spectral performanceWavelength is tunable by temperature and/or drive current.2013-04-04Postdoc seminar Carl Borgentun*


Reliability and life-timeReliability and life-time test: devices submitted to elevated drive currents and mount temperatures.Over 3000 hours of accumulated testing has been performed, without a single failure.Peak wavenumber stabilizes after 100-200 hours. The change in peak wavenumber is less than 2 cm-1. The output power is not reduced.2013-04-04Postdoc seminar Carl Borgentun*


Lasers integrated into TO-3 packages.No significant degradation in performance noticed after packaging.PackagingBeforeAfter2013-04-04Postdoc seminar Carl Borgentun*


Shape of output beamDivergent and highly elliptical, normal for edge-emitters.Divergence angles: ~ 110 and 40 degrees.2013-04-04Postdoc seminar Carl Borgentun*


Collimating the beam2013-04-04Postdoc seminar Carl Borgentun*LaserLaser


Package with internal lens2013-04-04Postdoc seminar Carl Borgentun*


Next target: CH42013-04-04Postdoc seminar Carl Borgentun*


Initial results2013-04-04Postdoc seminar Carl Borgentun*Hit target wavelength (3058 cm-1).Increase in output power (6 mW vs 1.5 mW)


SummarySuccessfully delivered 2 laser modules emitting more than 10 mW at 3777 cm-1 to Harvard for water detection.

Developing laser package complete with collimating optics.

Initial results show promising outlook for future methane detection missions like TLS.2013-04-04Postdoc seminar Carl Borgentun*


AcknowledgementsSupervisor:Siamak Forouhar

Colleagues:Cliff Frez, Ryan Briggs, Mahmood Bagheri2013-04-04Postdoc seminar Carl Borgentun*


Thanks for the attention!2013-04-04Postdoc seminar Carl Borgentun*


High-power semiconductor lasers for in-situ sensing of atmospheric gasesCarl Borgentun, Microdevices laboratory (MDL)


One of NASAs and JPLs large areas of interest have long been gas detection. It is very useful in earth science looking at greenhouse gases - as well as planetary science to understand the chemistry of atmospheres , and astronaut safety for early fire detection for instance.

Many times you would use a mass spectrometer. Wonderful instrument, lots of information. Needs vacuum, which means vacuum pumps and valves and everything that entails. Launching that into orbit or beyond is costly.

Laser spectrometer on the other hand. Small, convenient. These days, almost no NASA mission is launched without a laser somewhere.*Absorption spectroscopy. Fingerprints of many gases, especially in the infrared. Example of the Infrared Absorption of the Atmosphere from 1-28 microns.

Stress: not the same spectra! More than 20 mum, less than 1 nm. Full 100%, less than 1%.

If you have a narrow-band light source, you can separate different species. Fortunately, such a light source exists the laser!*A basic laser spectroscopy experiment might look something like this. If you want to do in-situ sensing, you can enclose the gas and have the laser light do multiple passes through the cavity. Of course, the more passes you allow, the more power you have to input since you lose a little light at every reflection.

A problem with lasers is that they require cooling that can be quite bulky. Nowadays, the size have shrunk significantly.Very long lifetimes, upwards of 20 years or more, no need for filling of cooling liquids!

TLS. 20 cm cavity, 80 passes, 16 m total path length.*ecosystem chemistry, structure, biomass, and biodiversity,*Polar: Water and carbon dioxideMSL: methane and carbon dioxide**Overtones => longer path length needed. This instrument uses a cavity with so high finesse i.e. high mirror reflectivities that the total path length is 2 km (as compared to TLS, 16 m).

Harvard has used commercially available lasers, but found that they were not sufficient for the required sensitivities of this mission.

*Two criteria:- Optical gain- Cavity

*Layered structure, like Dagwood sandwich.

QWs 66 Angstrom ~ atomic size.

MBE machine:Toxic gases, high vacuum, 400 degC => $million machine

One wafer = 100s of lasers.

*Ridge etch:- Smooth side walls (otherwise scattering, interfere with grating)Not too shallow (current spreading)Not too deep (mode loss)Narrow (single spatial mode)

Grating:Very close to the ridgeCannot punch through active region (lossy)

Insulation and contact:Define current pathHeatspreading

*Special sandwich: GaSb, less mature material system. Much of the fabrication and processing has an experimental touch to them, since we have to develop new methods. Advantage is the wavelength coverage: any wavelength between 2.0 and 3.5 mum.

Layer functionalities, both optical and electrical.*Grating is small, so tiny that we cannot use optical lithography. We need to use electrons! Electron beam lithography: $6 million

*Take a look at the scale!Human hair cross-section ~ 50 mum.

Grating pitch is less than a mum.*More critical moments: the best of lasers can be ruined by lousy management or poor solder.

Not every laser in bar is a good one. Varying grating pitch for fabrication tolerance.

Thickness is just 100 mum.

Coating material: cannot electrically short, good adhesion.*Describe tuning with temperature and drive current.*Lasers usually long-lived, several 10 000s hours.These lasers use new material system, need to show that our lasers will not die immediately after delivery, kinda like a warranty.

Only good lasers are tested.

Burn-in: weed out weak devices. Wavelength changes first 100-200 hours: not a problem, but something we need to watch. Investigating right now.

*Packaging for systematical purposes: easy to build and maintain.Also decouple the laser from the environment => longer life-time.*Now, with TLS, one of the major issues was with the beam divergence.

*TLS power 1.5 mW @ -20C.3058 cm-1 = 3.27 mum.*

130404.PostdocSeminar-Copyrightversion

Documents

Transcript of 130404.PostdocSeminar-Copyrightversion