Chalenges of Silicon Materials Research硅材料的研究的挑战

1、manufacture of high resistivity, low oxygen czochralski manufacture of high resistivity, low oxygen czochralski siliconsiliconjune 2005olli anttilaokmetic fellow,okmetic oyjmanufacture of high resistivity, manufacture of high resistivity, low oxygen czochralski siliconlow oxygen czochralski silicono

2、utlineoutline backgroundbackground crystal growth and wafer manufacturingcrystal growth and wafer manufacturing wafer characteristicswafer characteristics high resistivity czhigh resistivity cz metals and lifetimemetals and lifetime dopantsdopants oxygenoxygen magnetic fieldmagnetic field silicon me

3、lt behavior silicon melt behavior conclusionsconclusionsokmetic worldwide presenceokmetic worldwide presencemarket overviewmarket overviewelectronic equipmentelectronic equipment$ 1000 billion / + 8% $ 1000 billion / + 8% * *) )semiconductorssemiconductors$ 250 billion /+15 % $ 250 billion /+15 % *

4、*) )silicon waferssilicon wafers$ 8$ 8 billion billion /+13 % /+13 % * *) ) polysilicon polysilicon $ 0,7 billion $ 0,7 billion +14 % +14 % * *) ) * *) ) 1970-2003 1970-2003 cagr cagr nokiamegatrends.ppt / 16.11.1998 / eku page: 6silicon wafers: key process stepssilicon wafers: key process stepswafe

5、r slicewafer slice final thickness typically 500-600 mfinal thickness typically 500-600 mcrystal growthcrystal growth bulk and some surface material propertiesbulk and some surface material propertieswafer etchwafer etch remove damageremove damage final backsidefinal backsidesilicon wafers: key proc

6、ess stepssilicon wafers: key process stepspolishpolish to obtain excellent flatness and surface propertiesto obtain excellent flatness and surface propertiesfinal clean and packagefinal clean and package remove particle, metallic, ionic, and organic remove particle, metallic, ionic, and organic cont

7、aminationcontamination ensure dependable storage behaviorensure dependable storage behaviorokmetic productsokmetic productspolished cz-silicon wafers for microelectronicspolished cz-silicon wafers for microelectronicsmems silicon wafers for micromechanicsmems silicon wafers for micromechanicsepitaxi

8、al wafersepitaxial waferssoi-wafers soi-wafers diameters: 100.200 mmdiameters: 100.200 mmdopants: boron, phosphorus, arsenic and dopants: boron, phosphorus, arsenic and antimony antimony resistivity: 0.002 2 kohm-cmresistivity: 0.002 2 kohm-cmorientations: 1-0-0, 1-1-1 and 1-1-0orientations: 1-0-0,

9、1-1-1 and 1-1-0oxygen: 618 ppmaoxygen: 1400 dipped to 1400 c melt.c melt.4. 4. start of body, after start of body, after completion of pletion of shoulder.5. 5. conical tail growth after conical tail growth after completion of pletion of body.czochralski (cz) crystal growthczochralski (cz) crystal g

10、rowthatmosphere: low pressure, high purity argon.atmosphere: low pressure, high purity argon.hot zone: isostatic high purity graphite and hot zone: isostatic high purity graphite and graphite felt.graphite felt.manufacture of high resistivity, manufacture of high resistivity, low oxygen czochralski

11、siliconlow oxygen czochralski siliconoutlineoutline backgroundbackground crystal growth and wafer manufacturingcrystal growth and wafer manufacturing wafer characteristics: lifetime and metalswafer characteristics: lifetime and metals high resistivity czhigh resistivity cz metals and lifetimemetals

12、and lifetime dopantsdopants oxygenoxygen magnetic fieldmagnetic field silicon melt behavior silicon melt behavior conclusionsconclusions0.010.11101010 at /cm2 fenicuzncrd.l.process capability: metalsprocess capability: metalsvpd-txrf, vantaa plant, q4/2000 - q1/2005, average valuesvpd-txrf, vantaa p

13、lant, q4/2000 - q1/2005, average valuesmetal levels have been stable at or below detection limit of 1e9 at /cmmetal levels have been stable at or below detection limit of 1e9 at /cm2 2. values include contamination from . values include contamination from handling during sampling and measurement.han

14、dling during sampling and measurement.q4/2000q1/2001q2/2001q3/2001q4/2001q1/2002q2/2002q3/2002q4/2002q1/2003q2/2003q3/2003q4/2003q1/2004q2/2004q3/2004q4/2004q1/20050.010.11101010 at /cm2 fenicuzncrd.l.process capability: metalsprocess capability: metalsvpd-txrf, vantaa plant, q4/2000 - q1/2005, aver

15、age valuesvpd-txrf, vantaa plant, q4/2000 - q1/2005, average valuesimagine that a coin is made to slide on a wafer, leaving a tail of 10imagine that a coin is made to slide on a wafer, leaving a tail of 109 9 /cm /cm2 2 metal atoms. this equals the typical contamination level of a metal atoms. this

16、equals the typical contamination level of a state-of-the-art clean. then on another wafer, and another, over and over again. after a while, the coin wears and it gets thinner. how state-of-the-art clean. then on another wafer, and another, over and over again. after a while, the coin wears and it ge

17、ts thinner. how far should we make the coin go, before it is all used up?far should we make the coin go, before it is all used up?lifetime as-oxidized, during past 24 monthslifetime as-oxidized, during past 24 monthsspecification 500 specification 500 w wcmcmboth n- and p-typeboth n- and p-typeno sp

18、ecific emphasis on surface passivationno specific emphasis on surface passivation0.01000.02000.03000.04000.05000.06000.07000.08000.09000.010000.014710 13 16 19 22 25 28 31 34 37 40 43 46 49lifetime ( s)average80 % of areacz vs. fz materialcz vs. fz materialfor high frequency and detector application

19、sfor high frequency and detector applicationsfz comes naturally oxygen leanfz comes naturally oxygen leanhigh resistivity readily available with fzhigh resistivity readily available with fzso, why look into czso, why look into czcz available in larger diameterscz available in larger diameterslower w

20、afer costlower wafer costbetter compatibility with advanced cmos processesbetter compatibility with advanced cmos processesoxygen brings significant improvement in thermal slip resistanceoxygen brings significant improvement in thermal slip resistanceoxygen gives significant radiation hardness advan

21、tage? oxygen gives significant radiation hardness advantage? cz vs. fz materialcz vs. fz materialfor high frequency and detector applicationsfor high frequency and detector applicationsdoes cz fulfill the requirements?does cz fulfill the requirements?dopant controldopant control? ?oxygen related don

22、orsoxygen related donors? ?metallic contaminationmetallic contamination? ? !yes!yes!recombination lifetime / diffusion length? recombination lifetime / diffusion length? !yes!yes!copscops? ? metals in siliconmetals in siliconimagine that a coin is made to slide on a wafer, leaving a tail of 10imagin

23、e that a coin is made to slide on a wafer, leaving a tail of 109 9 /cm /cm2 2 metal atoms. this equals the typical contamination level of a state-of-the-art metal atoms. this equals the typical contamination level of a state-of-the-art clean. then on another wafer, and another, over and over again.

24、after a clean. then on another wafer, and another, over and over again. after a while, the coin wears and it gets thinner. how far should we make the coin while, the coin wears and it gets thinner. how far should we make the coin go, before it is all used up? go, before it is all used up? from the e

25、arth to the sun, 150 million km! from the earth to the sun, 150 million km! another quiz:another quiz: common metallic contaminants in grown crystal appear common metallic contaminants in grown crystal appear typically at 10 ppqa (parts per quadrillion atomic) level. that is typically at 10 ppqa (pa

26、rts per quadrillion atomic) level. that is 0. and how many zeros?0. and how many zeros?cz vs. fz materialcz vs. fz materialfor high frequency and detector applicationsfor high frequency and detector applicationsdoes cz fulfill the requirements?does cz fulfill the requirements?dopant control?dopant c

27、ontrol?oxygen related donorsoxygen related donors? ?metallic contaminationmetallic contamination? ? !yes!yes!recombination lifetime / diffusion length? recombination lifetime / diffusion length? !yes!yes!copscops? ? high resistivity material: 6”p100high resistivity material: 6”p100normalnormal cruci

28、ble quality6 p100: 106990200300400500600700800050010001500distance from seed/mmresistivity/ohm cm intentional doping at about 1600 ohmcm level at seed end.intentional doping at about 1600 ohmcm level at seed end. typical seed-to-tail resistivity ratio is about 1.5; here much larger.typical seed-to-t

29、ail resistivity ratio is about 1.5; here much larger. high crucible boron (& al) content results in 1 kohmcm material.high crucible boron (& al) content results in 600 donors annihilated 600 c c back-end-of-the-line (beol) temperatures between 400 and 500 back-end-of-the-line (beol) temperat

30、ures between 400 and 500 c to be avoidedc to be avoided e.g., at 420 e.g., at 420 c, only 10-30 min allowed, depending on oi and resistivityc, only 10-30 min allowed, depending on oi and resistivity typical oi 5-8 ppma for high res. materialtypical oi 5-8 ppma for high res. materialoxygen control pa

31、rametersoxygen control parametersgas flowgas flowpressurepressurepurge tubes (gas flow patternpurge tubes (gas flow patterncrucible rotationcrucible rotationcrystal rotationcrystal rotationtemperature distributiontemperature distributionmagnetic fieldmagnetic fieldmcz puller vacuum system operating

32、at above 1400 vacuum system operating at above 1400 c c offers electrical power control, mechanical offers electrical power control, mechanical movements and gas flows. argon purge movements and gas flows. argon purge pressure is typically 20-30 mbar.pressure is typically 20-30 mbar. the i.d. of the

33、 magnet 1 m, max. power the i.d. of the magnet 1 m, max. power range 100 kw.range 100 kw.mcz stabilizes the melt and has some influence on oxygen levelmcz stabilizes the melt and has some influence on oxygen level magnet used by okmetic results to somewhat magnet used by okmetic results to somewhat

34、higherhigher oxygen level than regular czoxygen level than regular cz better melt stability better melt stability widens oxygen range;widens oxygen range; more aggressive control parameters can be used more aggressive control parameters can be usedresults from early crystals 10 results from early cr

35、ystals 10 yrs. agoyrs. ago0246812141820020406080100distance from seed (cm)12014016018020010oxygen concentration (ppma)16regularczmagnetic field changes oxygen behaviormagnetic field changes oxygen behaviorthicker laminar layer next to crucible thicker laminar layer next to crucible wallwall = slower

36、 oxygen dissolution = slower oxygen dissolutionthicker laminar layer at gas interface = thicker laminar layer at gas interface = tendency to tendency to increase increase oxygen in the meltoxygen in the meltbalance between these two effects balance between these two effects defines oxygen leveldefin

37、es oxygen levelslow crucible erosion = long crucible slow crucible erosion = long crucible lifetime, low dopant emission ratelifetime, low dopant emission rateprice to pay: more difficult control of price to pay: more difficult control of radial variations! both dopants and radial variations! both d

38、opants and oxygenoxygencz vs. fz materialcz vs. fz materialfor high frequency and detector applicationsfor high frequency and detector applicationsdoes cz fulfill the requirements?does cz fulfill the requirements?dopant control? dopant control? !yes?!yes?oxygen related donors?oxygen related donors?

39、!yes?!yes?metallic contaminationmetallic contamination? ? !yes!yes!recombination lifetime / diffusion length? recombination lifetime / diffusion length? !yes!yes!copscops? ? remains unresolved at this point, and should be addressed later, if needed; typical density remains unresolved at this point,

40、and should be addressed later, if needed; typical density 2 kohmcm p-type material manufacturable 2 kohmcm p-type material manufacturable100 mm material allows somewhat higher values100 mm material allows somewhat higher valuesn-type resistivity about factor 2 lowern-type resistivity about factor 2

41、loweroxygen 8 ppmaoxygen growth effectively from smaller, less unstable = growth effectively from smaller, less unstable meltmelt under crystal, minor impactunder crystal, minor impactdopant, oxygen distribution is not uniform even dopant, oxygen distribution is not uniform even with mczwith mczmelt

42、 flow is buoyancy driven turbulentmelt flow is buoyancy driven turbulentthe melt behavior is not quite so smooththe melt behavior is not quite so smoothrecombination lifetime after oxygen precipitationrecombination lifetime after oxygen precipitation125 mm n(100), 850 125 mm n(100), 850 c / 4h + 100

43、0 c / 4h + 1000 c /12 h oc /12 h o2 2shorter lifetime where more precipitatesshorter lifetime where more precipitatesring diameters give time scale, together with ring diameters give time scale, together with freezing interface curvaturefreezing interface curvaturequasiperiodic behavior, time scale

44、1 minquasiperiodic behavior, time scale 1 min open hot zone, early 90s open hot zone, early 90sbehavior is behavior is notnot external pull rate related external pull rate relatedlps measurement on 150 mm n(100) 5 ohmcm crystallps measurement on 150 mm n(100) 5 ohmcm crystal area shown 100 x 50 mmar

45、ea shown 100 x 50 mm sample as-grown and donor kill annealedsample as-grown and donor kill annealed freezing interface shape nicely visiblefreezing interface shape nicely visible time dependent freezing velocity can be time dependent freezing velocity can be measured if spectral response of measured

46、 if spectral response of measurement is knownmeasurement is knownsample cut sample cut growth axis growth axis silicon melt modeling:silicon melt modeling:temperature distributiontemperature distributionand velocity fieldand velocity field crucible rotation 5 rpm crucible rotation 5 rpm time depende

47、nt, chaotic behaviortime dependent, chaotic behavior heat to the crystal largely from the bottomheat to the crystal largely from the bottomit was at a meeting of the british association in london in 1932 that i remember hoarce lamb remarked: “i am an old man now, and when i die and go to heaven there are two matters on which i hope for enlightenment. one is quantum