• 沒有找到結果。

Fully quantitative characterization of CMOS–MEMS polysilicon/titanium thermopile infrared sensors

N/A
N/A
Protected

Academic year: 2021

Share "Fully quantitative characterization of CMOS–MEMS polysilicon/titanium thermopile infrared sensors"

Copied!
4
0
0

加載中.... (立即查看全文)

全文

(1)

SensorsandActuatorsB161 (2012) 892–900

ContentslistsavailableatSciVerseScienceDirect

Sensors

and

Actuators

B:

Chemical

jou rn a l h o m e pag e :w w w . e l s e v i e r . c o m / l o c a t e / s n b

Fully

quantitative

characterization

of

CMOS–MEMS

polysilicon/titanium

thermopile

infrared

sensors

Chung-Nan

Chen

InstituteofPhotonicsandCommunications,NationalKaohsiungUniversityofAppliedSciences,Kaohsiung,Taiwan,ROC

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Received29August2011

Receivedinrevisedform9November2011 Accepted22November2011

Available online 30 November 2011 Keywords: CMOS Infraredsensor MEMS Specificdetectivity Thermalproperty Thermopile

a

b

s

t

r

a

c

t

ThisstudydemonstratesafullyquantitativecharacterizationofahighlysensitiveCMOS–MEMS polysili-con/titaniumthermopileinfraredsensorbyusingthesimulationsofnon-sequentialraytracingandsolid conduction,andthemeasurementsofvoltageresponseandfrequencyresponseinatmosphereandin vacuum.

Thethermaltimeconstantsof17.0msinairand37.0msinvacuumforthepolysilicon/titanium ther-mopilewithagold-blackabsorberwereestimatedbythemeasurementsoffrequencyresponse.The solidconductance,gasconductance,radiationloss,andheatcapacitanceofthethermopilewere char-acterizedas112␮W/K,141␮W/K,5.88␮W/K,and4.40␮J/Kinatmospherebythesimulationofsolid conductionusingANSYSandthemeasurementsoffrequencyresponse.Thevoltageresponsivity, sen-sornoise,noiseequivalentpower,andspecificdetectivityofthegold-blackcoatedthermopileinair wereestimatedas63.1V/W,27.0nV/Hz1/2,0.43nW/Hz1/2,and1.87×108cmHz1/2/Wbythesimulationof receivedopticalpowerusingLightToolsraytracingsoftwareandthemeasurementsofvoltageresponse. ItshowsthatthesensorhasthehighestspecificdetectivitycomparedtothepublishedCMOS–MEMS thermopilesinatmosphereduetothedesignoflowsolidconductanceandhighemissivity. Eventu-ally,theSeebeckcoefficientofthepolysilicon/titaniumpairwasfirstevaluatedandhasamagnitudeof 170.2␮V/K.

© 2011 Elsevier B.V. All rights reserved.

1. Introduction

Thermopile infrared sensors were widely used as the sens-ingelementsofnondispersiveinfrared(NDIR)gasdetectors[1,2], thermal imagers [3,4] and non-contact infrared thermometers [5].Thermalimagersandinfraredthermometersarethedevices thattransforminfraredradiationintotemperaturereading.NDIR gasdetectorsconsistofband-passfilters, infraredemitters,and infrared sensors which could detect the concentration of spe-cificgasesbythesignaldifferenceresultedfromtheabsorption of infraredpower by thegases.In order to enhancethe selec-tivityofgassensing,thebandwidthsoftheband-passfiltersare usuallypreferredtobeasnarrowastheabsorptionbandwidths ofthespecificgaseswhichwouldresultin theweakandslight receivedinfraredpowerofthesensors.Itis particularly impor-tantandnecessarytodevelop ahighlysensitiveinfraredsensor forNDIRgassensing.Thermopilesarethermal-typeinfrared

sen-∗ Correspondenceaddress:InstituteofPhotonicsandCommunications,National KaohsiungUniversityofAppliedSciences,415ChienKungRoad,SanminDistrict, Kaohsiung80778,Taiwan,ROC.Tel.:+88673814526x3354;fax:+88673832771.

E-mailaddress:cn [email protected]

sorswhichtransferthermalradiationintothermoelectricvoltage. Therefore, theperformances of thermopiles,suchas responsiv-ity,specificdetectivityandresponsespeed,arequitedependent ontheirthermalproperties.Athermopilesensorcomprisesmany thermocouplesconnectedinseriesandtheconnectionsofthe ther-mocouplesdivideintohotjunctionsandcoldjunctions.Ingeneral, thehotjunctionsarethermallyisolatedfromtheirsubstrateand aninfraredabsorberiscoatedontothesurfaceofthethermopile forimprovingthesensitivityoftheinfraredsensor.Asathermopile absorbstheinfraredradiationemittedfromaninfraredsource,the temperaturedifferencebetweenhotjunctionsandcoldjunctionsis formedandthethermoelectricvoltagebetweenthetwoterminals ofthermocouplesinseriesiscreatedatthesametime.The magni-tudeoftheoutputvoltageisdirectlyproportionaltothenumberof thermocouples,theSeebeckcoefficientofthethermocouple mate-rialpair,andthetemperaturedifferencebetweenhotjunctionsand coldjunctions.Andthetemperaturedifferenceisindirectratioto theemissivityoftheinfraredabsorber,receivedinfraredpower, andthereciprocalofthermalconductance.Thevoltage responsiv-ityofaninfraredsensorisdefinedastheratioofoutputvoltageto receivedinfraredpowerwhichspecifiestheinput–outputgainofa sensorsystem.However,specificdetectivityisthemostimportant specificationofaninfraredsensortoindicatethesignal-to-noise 0925-4005/$–seefrontmatter © 2011 Elsevier B.V. All rights reserved.

(2)

C.-N.Chen/SensorsandActuatorsB161 (2012) 892–900 893

ratioofanopticalsensor.Themagnitudeofspecificdetectivityfor athermopileinfraredsensorisindirectproportiontotheemissivity oftheinfraredabsorber,theSeebeckcoefficient,andthe recipro-calofthermalconductance.Themainheatlossesforthermopile devicesaresolidconductionandgasconductionsincetheradiation lossisrelativelysmallandnegligible.Forthermalmicrosensors,the gasconductancemayevendominatethebehaviorofheattransfer andtheperformanceinatmosphere[6,7].Inaddition,theresponse timeconstantsofthermaldevices,whicharedefinedastheratio ofheatcapacitancetothermalconductance,arealsohighly depen-dentontheirthermalproperties.It showsthatthequantitative characterizationofthethermalpropertiesisveryusefuland valu-ablefortheanalysisofthesensorperformancesandthedesignof thermalsensors.

For thereasons ofmass production, IC integration,and low cost, a CMOS compatible process was frequently employed as the front-end process of a MEMS thermopile infrared sen-sor [8–11,3,4,12–15]. In past years, polysilicon and aluminum werethe most common materialsfor forming the thermocou-plepairs of CMOSthermopiles since thefabrication process of a polysilicon/metal(poly/metal) thermopile is relativelysimple and compatible. However,the thermal conductivityof a CMOS aluminum film is about 200W/mK [16] which is an order of magnitudegreater than those of polysilicon and titanium [17]. Thehighsolidconductance ofa polysilicon/aluminum(poly/Al) thermopilecouldresultinalowsensitivity.Besides,an interfer-encestructure was used toserve as an infrared absorber of a CMOS–MEMSthermopilewhichhaslowabsorptanceanda nar-rowabsorptionbandwidth[2,18].Instead,aporousgoldblackfilm withabsorptanceof>90%andwideabsorptionbandwasproposed forthermalinfrareddetectors[19–21].Inourwork,thepurpose ofthedesign of polysilicon/titanium(poly/Ti) thermopiles isto improvethespecificdetectivityofsensorsbyhighlyreducingthe solid conductance andincreasing theemissivityof theinfrared sensorwithoutmuch loweringtheSeebeckcoefficientand rais-ingJohnsonnoise.Furthermore,thepostprocessofthethermopile wasdesignedtobeasimplephotomaskless process.Inorderto extremelylowerthesolidconductanceofapoly/metalthermopile, a0.6-␮maluminumfilminthestandardfirstmetallizationprocess wasreplacedbya0.1-␮mtitaniumfilmwithalowthermal con-ductivity.TheSeebeckcoefficientsofapoly/Tipairandapoly/Al pairshouldbealmostthesamesincepolysiliconhasmuchlarger absoluteSeebeckcoefficientthanthoseoftitaniumandaluminum [18,22–24].Besides,theJohnsonnoiseofapoly/Tithermopileis almostequivalenttothatofapoly/Al thermopilewithan iden-ticallayoutsincepolysilicondominatestheelectricresistanceof thesensor.AftertheCMOSprocess,thesuspendedstructureofthe poly/Tithermopilewasformedbyfront-sideanisotropicetching ofsiliconin adual-dopedTMAHsolutionwithlowetchrateof exposedaluminumpads[25].Forfurtherlesseningsolid conduc-tance,acceleratingetchingprocessandachievingahighfillfactor, narrowandlongline-shapedetchingwindowsweredesignedand located between thethermocouples. Eventually,a porous gold blacklayerwasthermallyevaporatedontothesurfaceofthe sen-sorand patterned in situ by metal mask technique toachieve highabsorptionofinfraredpower.Thispaperis focusedonthe characterizationofalltheimportantthermalpropertiesand perfor-mancesofthepoly/Tithermopile,suchasthesolidconductance,gas conductance,radiationloss,heatcapacitance,thermaltime con-stant,emissivity,Seebeckcoefficient,responsivity,Johnsonnoise, shotnoise,temperaturefluctuationnoise,noiseequivalentpower (NEP), and specific detectivity, by adopting the LightTools ray tracingsimulation of receivedinfraredpower,theANSYS finite elementanalysis(FEA)ofsolidthermalconduction,andthe mea-surementsofvoltageresponse,frequencyresponseandvacuum response.

Fig.1.HeattransferofaMEMSthermopile.

2. Theory,designandsimulation

2.1. Theory

The thermal behavior of a packaged thermopile under an infraredradiationof˚isgovernedbytheheatflowequation[26] CdTh

dt +G(Th−Ta)=ε˚, (1)

whereCistheheatcapacitanceofthethermopilesensor,Thisthe temperatureofhotjunctions,tisthetime,Gisthethermal conduc-tancebetweenhotjunctionanditssurroundingwithatemperature ofTa,andεistheemissivitywhichrevealstheabsorptanceofthe infraredabsorber.ThesolutionofthetemperaturedifferenceT betweenthehotjunctions anditsenvironment inthetransient stateis[26]

T= ε˚G



1

1+ω22, (2)

where ω is the modulated angular frequency of the incident infraredpower,and =C/Gis thethermaltime constant ofthe sensor.

TheoutputthermoelectricvoltageVcausedbytherising tem-peratureofhotjunctionscanbeexpressedby

V=N˛AB(Th−Tc), (3)

inwhichNisthenumberofthermocouplesinseries,˛AB isthe Seebeckcoefficientbetweenthetwothermocouplematerials,Tcis thetemperatureofcoldjunctionswhichisalmostthesame mag-nitudewiththeambienttemperatureTa.Therefore,thevoltage responsivityRVofthethermopilecanbegivenby

RV≡ V ˚= N˛ABε G 1



1+ω22 = RV 0



1+ω22 , (4)

whereRV0=N˛ABε/Gistheflat-bandresponsivityinthethermal steadystate.ForCMOSpoly/metalthermopileswiththesame lay-out,thenumberofthermocouples isfixedandthevariancesof theSeebeckcoefficientsbetweenpolysiliconandCMOS–metalsare verysmallsincepolysilicondominatesthethermoelectriceffect ofthepoly/metalthermopiles.Therefore,theresponsivityofthe poly/metalthermopilesinthethermalsteadystateisalmostonly indirectratiototheiremissivityandthereciprocalofthermal con-ductance.

Asshownin Fig.1,there arethree typesofheat lossesthat ariseinaMEMSthermopile:solidconduction,gasconductionand radiation.Asthesuspendedstructureofathermopileabsorbsthe incidentinfraredpowerradiatedfromatarget,someheat dissi-patesthroughthesolidsuspendedstructuretosiliconsubstrate.

(3)

894 C.-N.Chen/SensorsandActuatorsB161 (2012) 892–900

ThesolidconductanceGsisdirectlyproportionaltothethermal conductivityksofthesuspendedstructureandisexpressedby Gs=kswd

l , (5)

wherew,dandlarethewidth,thicknessandlengthofthe sup-porting leads of the suspended membrane. And some heat is transported to thesubstrate by thecollisions of gasmolecules betweenthehotsuspendedmembraneandthesubstratewithinthe cavity.ThegasconductanceGgisindirectratiotothefree molec-ularconductivityofgaskgandtheareaoftheinfraredsensorAs. Besides,thegasconductanceisafunctionofpressureP[7,27].For athermopilewithone-sideheatsink,thegasconductanceisgiven by Gg=  2−kgAsP



P t P+Pt



, (6)

whereistheaccommodationcoefficientofgas,Ptisanempirical transitionparameterwhichisdirectlyproportionaltothe recip-rocalofgapdepthbetweenthedeviceandheatsink[27].Asthe pressurePisnegligiblecomparedtothetransitionpressure,thegas conductanceisdirectlyproportionaltothepressureofthecavity. Itmeansthatavacuumpackageisveryusefultodiminishthegas conductance.Radiationlossisthephenomenonofheatexchange betweenathermaldeviceanditsenvironment.Theradiation con-ductanceGrcouldbederivedfromtheStefan–Boltzmannlawand givenby

Gr=(ε+εb)As(Th2+Ta2)(Th+Ta)≈4(ε+εb)AsTa3, (7) whereεbistheemissivityofthebottomsurfaceofthesensor,is theStefan–Boltzmannconstant,andthetemperatureofhot junc-tionsisconsideredtobeclosetotheambienttemperaturesince thetemperaturerisingofhotjunctionsunderthermalradiationis tiny.Thetotalthermalconductanceofathermopilesensorisgiven by

G=Gs+Gg+Gr. (8)

Ingeneral,theradiationconductanceofathermopileis negli-giblesincethetemperatureofhotjunctionsisonlyslightlyhigher thantheambienttemperature.However,thegasconductanceofa MEMSthermaldeviceinatmosphereissignificantduetothe minia-turedimensionofcavitygap.Forreducingthegasconductance,a vacuumorafilling-gasenvironmentwasoftenusedbypublished papersandmanufacturers[18].

ThemainnoisesourceofathermopilesensorisJohnsonnoise duetonobiassupplyrequirementduringthesensor’soperation [28].Johnsonnoiseisexpressedas

vJ

=



4kBTsRsf, (9)

wherekBistheBoltzmannconstant,Tsistheabsolutetemperature ofthethermopile,Rsistheresistanceofthesensor,andfisthe noisebandwidth.Inaddition,shotnoiseisgivenby[26]

vs

=Rs·



2qIf, (10)

inwhichqistheelementarychargeofanelectron,and Iisthe current.Thetemperaturefluctuationnoiseresultedfromthe tem-peraturefluctuation



T2



1/2inbackgroundiswrittenby[26]

vTF

=N˛AB·



T2



1/2 = RV 0 ε



4kBT2Gft, (11)

whereft=1/4isthebandwidthoftemperaturefluctuationnoise. Thetotalsensornoiseofathermopilesensorisexpressedas

vn

=



v

2

J+

v

2s+

v

2TF. (12)

NoiseequivalentpowerNEPandspecificdetectivityD*arethe mostsignificantspecificationsofaninfraredsensortoqualifythe signal-to-noiseperformanceofthesensor.NEPisdefinedasthe receivedinfrared powerwhich hasa signal-to-noise ratio of1. TheNEPofathermopileinthesteadystatecanbeexpressedand approachedas NEP≡ R

vn

V ≈ G N˛ABε



4kBTsRsf. (13)

Forapoly/metalthermopile,thepolysiliconlinesdominatethe resistanceandthethermoelectriceffectofthesensor.Considera simplepolysiliconcantileverwhichhasadimensionoflpinlength, wpinwidth,dpinthickness,andaresistivityofp,theresistance rofthecantileveralongthelengthofpolysiliconlineisgivenby r=plp/wpdp.IfthepolysiliconcantileverisdividedintoNequal lineswiththesamewidthofwp/Nbyconsideringthattheintervals betweenthepolysiliconlinesarenegligible.Thentheresistanceof thepolysiliconlinesinseriescouldbewrittenas

Rs≈N·p lp (wp/N)dp =

N2r. (14)

AndtheNEPcanberewrittenas

NEP= G

˛ABε



4kBTsrf. (15)

ItshowsthatthedependenceofNEPonthethermocouple num-berNdisappearsduetotherapidlyincreasingofsensorresistance. D*isdefinedasthereciprocalofNEPandnormalizedtothesensor areaAsandthebandwidthf

D∗≡



Asf NEP = ˛ABε G



As 4kBTsr . (16)

ItimpliesthatthemagnitudeofD*isalmostonlydependent ontheemissivityandthethermalconductanceofthesensorfora poly/metalthermopilewiththesameprocessandsensorsize.

Inthiswork,theabsorptionspectrumoftheabsorberlayerwas measuredbyusingFouriertransforminfraredspectrometry(FTIR) in order toestimatethe emissivityε of thepoly/Ti thermopile infraredsensor[5].AccordingtothestatementofKirchhoff’slawof thermalradiation,theemissivityofabodyequalsitsabsorptivity atthermalequilibrium.Therefore,theemissivityoftheabsorber layercouldbeevaluatedbythecalculationofaverageabsorptivity ofinfraredradiationoveraspecifiedwavelengthrange.

The important performances of thermopiles wereevaluated bythesimulationofreceivedopticalpower,themeasurementof thermoelectricvoltage,andthetheoreticalcalculationofsensor’s noises.Inordertoestimatethereceivedpower˚ofaTO5-packaged thermopileatvarioustemperaturesofablackbodyinfraredsource andatambienttemperature,araytracingsimulationwascarried outbyadoptingLightToolssoftware.Theresponsivityofthe ther-mopileisdefinedastheratiooftheoutputthermoelectricvoltage Vtothereceivedopticalpower˚,inwhichtheoutputvoltagewas obtainedbythemeasurementofsensor’svoltageresponseunder infraredradiationandthecorrespondingreceivedopticalpower wassimulatedbyLightTools.Thesensornoise

vn

ofthethermopile iscalculatedbyusingEqs.(9)–(12)andtheNEPofthesensorwas evaluatedbycalculatingtheratioofsensornoisetoresponsivity. Subsequently,thespecificdetectivityD*couldbefurthergivenby utilizingitsdefinitioninEq.(16).

Thethermalpropertiesofthermopileswerefurtherfoundand analyzed bythe simulation of solid conductance, the measure-mentsoffrequencyresponsesinairandinvacuum.Toestimate thesolidconductanceofthermopilesensors,thetemperature dis-tributionsofhotjunctionsunderinfraredradiationwassimulated byadoptingANSYSFEAsoftware.Inthesimulation,anexact3D modelofapoly/metalthermopilechipwasconstructedandthe

(4)

900 C.-N.Chen/SensorsandActuatorsB161 (2012) 892–900

4. Conclusion

ACMOS–MEMSgold-blackcoatedpoly/Tithermopileinfrared

sensorwithahighspecificdetectivitywasfabricatedbyusinga

0.8-␮msix-maskCMOSprocessandapostprocesswithoutany

lithographystep.Thethermalpropertiesandperformancesofthe

poly/Tisensorhavebeenfullycharacterizedbysimulationsand

measurements in this study. The characterization resultsshow

that thepoly/Tithermopilehasexcellent responsivity and

spe-cificdetectivityof63.1V/Wand1.87×108cmHz1/2/Wduetothe

lowthermalconductivityoftitanium.Thespecificdetectivityof

the CMOSpoly/Ti thermopile withthe designs of low thermal

conductanceandhighemissivityis6.1timeshigherthanthatof

standardCMOSpoly/Al thermopileswithoutgoldblackcoating.

Thecontributionsof solidconduction, gasconduction and

radi-ationlossontheheatlossesofthethermopilewereabout43%,

55%and2%.Andthegasconductancecouldbefurtherneglected

and the specific detectivity of the sensor would be raised to

4.06×108cmHz1/2/W by operating under a pressure less than

0.02Torr.Moreover,theSeebeckcoefficientofn+-polysiliconand

titaniumwasfirstestimatedandhasamagnitudeof170.2␮V/K.It

showsthatpolysiliconmaterialdominatestheSeebeckcoefficient

ofCMOS–MEMSpoly/metalthermopiles.

Acknowledgements

TheauthorwouldliketothanktheNationalScienceCouncil

oftheRepublicofChina,Taiwan,forfinanciallysupportingthis

researchunderContractNo.NSC97-2221-E-151-006.Theauthor

wouldalsoliketothanktheNationalNanoDeviceLaboratoriesfor

thesupportingofprocessfabrication.

References

[1]S.Udina,M.Carmona,G.Carles,J.Santander,L.Fonseca,S.Marco,A micro-machinedthermoelectricsensorfornaturalgasanalysis:thermalmodeland experimentalresults,Sens.ActuatorsB134(2008)551–558.

[2] A.Roncaglia,F.Mancarella,G.C.Cardinali,CMOS-compatiblefabricationof thermopileswithhighsensitivityinthe3–5␮matmosphericwindow,Sens. ActuatorsB125(2007)214–223.

[3] A.D. Oliver, K.D. Wise, A 1024-element bulk-micromachined thermopile infraredimagingarray,Sens.ActuatorsA73(1999)222–231.

[4]A.Schaufelbühl,N.Schneeberger,U.Münch,M.Waelti,O.Paul,O.Brand,H. Baltes,C.Menolfi,Q.Huang,E.Doering,M.Loepfe,Uncooledlow-costthermal imagerbasedonmicromachinedCMOSintegratedsensorarray,J. Microelec-tromech.Syst.10(2001)503–510.

[5]C.N.Chen,Temperatureerroranalysisandparameterextractionofan8–14-␮m thermopilewithawavelength-independentabsorberfortympanic thermome-ter,IEEESens.J.11(2011)2310–2317.

[6]C.N.Chen,W.C.Huang,ACMOS–MEMSthermopilewithlowthermal conduc-tanceandanearperfectemissivityinthe8–14-␮mwavelengthrange,IEEE ElectronDeviceLett.32(2011)96–98.

[7]X.Wang,C.Liu,Z.Zhang,S.Liu,X.Luo,Amicro-machinedPiranigaugefor vacuummeasurementofultra-smallsizedvacuumpackaging,Sens.Actuators A161(2010)108–113.

[8]R.Lenggenhager,H.Baltes,J.Peer,M.Forster,Thermoelectricinfraredsensors byCMOStechnology,IEEEElectronDeviceLett.13(1992)454–456. [9]J.Schieferdecker,R.Quad,E.Holzenkämpfer,M.Schulze,Infraredthermopile

sensorswithhighsensitivityandverylowtemperaturecoefficient,Sens. Actu-atorsA47(1995)422–427.

[10]M.Müller,W.Budde,R.Gottfried-Gottfried,A.Hübel,R.Jähne,H.Kück,A ther-moelectricinfraredradiationsensorwithmonolithicallyintegratedamplifier stageandtemperaturesensor,Sens.ActuatorsA54(1996)601–605. [11]H.Baltes,O.Paul,O.Brand,MicromachinedthermallybasedCMOS

microsen-sors,Proc.IEEE86(1998)1660–1678.

[12]C.Calaza,N.Viarani,G.Pedretti,M.Gottardi,A.Simoni,V.Zanini,M.Zen,An uncooledinfraredfocalplanearrayforlow-costapplicationsfabricatedwith standardCMOStechnology,Sens.ActuatorsA132(2006)129–138. [13] S.J.Chen,C.H.Shen,Anewhigh-filling-factorCMOS-compatiblethermopile,

IEEETrans.Instrum.Meas.56(2007)1231–1238.

[14] Y.Li,H.Zhou,T.Li,Y.Wang,Y.Liu,Y.Wang,CMOS-compatible8× 2thermopile array,Sens.ActuatorsA161(2010)120–126.

[15] D.Xu,B.Xiong,Y.Wang,Self-alignedthermoelectricinfraredsensors,IEEE ElectronDeviceLett.31(2010)512–514.

[16] M.vonArx,O.Paul,H.Baltes,Process-dependentthin-filmthermal conductiv-itiesforthermalCMOSMEMS,J.Microelectromech.Syst.9(2000)136–145. [17]C.G.Mattsson,K.Bertilsson,G.Thungstrom,H.E.Nilsson,H.Martin,Thermal

simulationanddesignoptimizationofathermopileinfrareddetectorwithan SU-8membrane,J.Micromech.Microeng.19(2009)055016.

[18]A.Graf,M.Arndt,M.Sauer,G.Gerlach,Reviewofmicromachinedthermopiles forinfrareddetection,Meas.Sci.Technol.18(2007)R59–R75.

[19]D.J.Advena,V.T.Bly,J.T.Cox,Depositionandcharacterizationoffar-infrared absorbinggoldblackfilms,Appl.Opt.32(1993)1136–1144.

[20]W.Becker,R.Fettig,W.Ruppel,Opticalandelectricalpropertiesofblackgold layersinthefarinfrared,InfraredPhys.Technol.40(1999)431–445. [21]N.Nelms,J.Dowson,Goldblackcoatingforthermalinfrareddetectors,Sens.

ActuatorsA120(2005)403–407.

[22]A.W.VanHerwaarden,P.M.Sarro,ThermalsensorsbasedontheSeebeckeffect, Sens.Actuators10(1986)321–346.

[23]M.Boutchich,K.Ziouche,P.Godts,D.Leclercq,Characterizationofphosphorus andboronheavilydopedLPCVDpolysiliconfilmsinthetemperaturerange 293–373K,IEEEElectronDeviceLett.23(2002)139–141.

[24]J. Xie, C. Lee, M.F. Wang, Y. Liu, H. Feng, Characterization of heavily dopedpolysiliconfilmsforCMOS–MEMSthermoelectricpowergenerators,J. Micromech.Microeng.19(2009)125029.

[25]K.Biswas,S.Kal,EtchcharacteristicsofKOH,TMAHanddualdopedTMAHfor bulkmicromachiningofsilicon,Microelectron.J.37(2006)519–525. [26]P.W.Kruse,D.D.Skatrud,UncooledInfraredImagingArraysandSystems,

Aca-demicPress,1997.

[27] B.C.S.Chou,Y.M.Chen,M.Ou-Yang,J.S.Shie,AsensitivePiranivacuumsensor andtheelectrothermalSPICEmodeling,Sens.ActuatorsA53(1996)273–277. [28]M.C.Foote,E.W.Jones,T.Caillat,Uncooledthermopileinfrareddetector lin-eararrayswithdetectivitygreaterthan109cmHz1/2/W,IEEETrans.Electron

Devices45(1998)1896–1902. Biography

Chung-NanChenwasborninTaiwan.HereceivedtheB.S.degreeinphysicsfrom ChineseCultureUniversity,Taipei,Taiwan,in1988,theM.S.andPh.D.degreesin electro-opticalengineeringfromNationalChiaoTungUniversity,Hsinchu,Taiwan, in1993and2000,respectively.HisPh.D.dissertationisthetechnologydevelopment ofmicrobolometerIRFPAdetectors.HeservedfortwoyearsintheTaiwanArmy.He waswithOptoTechCorporation,Hsinchu,Taiwan;hewasinvolvedintheR&Dof MEMSproductsandbecameadeputymanageroftheDivisionofMicroelectronics from1996to2005.In2005,hebecameanAssistantProfessorwiththeInstitute ofPhotonicsandCommunications,NationalKaohsiungUniversityofApplied Sci-ences,Kaohsiung,Taiwan.Hisresearchinterestsincludethermalsensors,MEMS andsemiconductorprocess,andinfraredengineering.

數據

Fig. 1. Heat transfer of a MEMS thermopile.

參考文獻

相關文件

substance) is matter that has distinct properties and a composition that does not vary from sample

好了既然 Z[x] 中的 ideal 不一定是 principle ideal 那麼我們就不能學 Proposition 7.2.11 的方法得到 Z[x] 中的 irreducible element 就是 prime element 了..

Wang, Solving pseudomonotone variational inequalities and pseudocon- vex optimization problems using the projection neural network, IEEE Transactions on Neural Networks 17

Then, we tested the influence of θ for the rate of convergence of Algorithm 4.1, by using this algorithm with α = 15 and four different θ to solve a test ex- ample generated as

Particularly, combining the numerical results of the two papers, we may obtain such a conclusion that the merit function method based on ϕ p has a better a global convergence and

Then, it is easy to see that there are 9 problems for which the iterative numbers of the algorithm using ψ α,θ,p in the case of θ = 1 and p = 3 are less than the one of the

volume suppressed mass: (TeV) 2 /M P ∼ 10 −4 eV → mm range can be experimentally tested for any number of extra dimensions - Light U(1) gauge bosons: no derivative couplings. =>

For pedagogical purposes, let us start consideration from a simple one-dimensional (1D) system, where electrons are confined to a chain parallel to the x axis. As it is well known