Form Approved REPORT DOCUMENTATION PAGE OMB No. 0704-0188 Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing this collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden to Department of Defense, Washington Headquarters Services, Directorate for Information Operations and Reports (0704-0188), 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202- 4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to any penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number. PLEASE DO NOT RETURN YOUR FORM TO THE ABOVE ADDRESS. 1. REPORT DATE (DD-MM-YYYY) 2. REPORT TYPE 3. DATES COVERED (From - To) 12-04-2006 Journal Article POSTPRINT 2006 4. TITLE AND SUBTITLE 5a. CONTRACT NUMBER X-ray irradiation effects in top contact, pentacene based field 5b. GRANT NUMBER effect transistors for space related applications 5c. PROGRAM ELEMENT NUMBER 6. AUTHOR(S) 5d. PROJECT NUMBER R.A.B. Devine; Mang-Mang Ling,* Abhijit Basu Mallik,* Mark Roberts,* Zhenan Bao* 5e. TASK NUMBER 5f. WORK UNIT NUMBER 7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION REPORT NUMBER Air Force Research Laboratory *Dept of Chemical Engineering AFRL-VS-PS-JA-2006-1015 Space Vehicles Directorate Stanford University 3550 Aberdeen Ave SE 381 North South Mall Kirtland AFB, NM 87117-5776 Stanford, CA 94305 9. SPONSORING / MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR’S ACRONYM(S) AFRL/VSSE 11. SPONSOR/MONITOR’S REPORT NUMBER(S) 12. DISTRIBUTION / AVAILABILITY STATEMENT Approved for public release; distribution is unlimited. (Clearance #VS06-0059) 13. SUPPLEMENTARY NOTES Published in Applied Physics Letters 88, 151907 (2006) Government Purpose Rights 14. ABSTRACT Preliminary studies of the effect of x-ray irradiation, typically used to simulate radiation effects in space, on top contract, pentacene based field effect transistors have been carried out. Threshold voltage shifts in irradiated devices are consistent with positive charge trapping in the gate dielectric and a rebound effect is observed, independent of the sign of applied electric field during irradiation. Carrier mobility variations in positive electric field biased/irradiated devices are interpreted in terms of the effects of interface-state- like defects. 15. SUBJECT TERMS x- ray irradiation, radiation, radiation effects, pentacene 16. SECURITY CLASSIFICATION OF: 17. LIMITATION 18. NUMBER 19a. NAME OF RESPONSIBLE PERSON OF ABSTRACT OF PAGES Clay Mayberry a. REPORT b. ABSTRACT c. THIS PAGE Unlimited 19b. TELEPHONE NUMBER (include area Unclassified Unclassified Unclassified 4 code) 505-846-4049 Standard Form 298 (Rev. 8-98) Prescribed by ANSI Std. 239.18 APPLIEDPHYSICSLETTERS88,151907(cid:1)2006(cid:2) X-ray irradiation effects in top contact, pentacene based field effect transistors for space related applications R. A. B. Devinea(cid:1) AFRL/VSSE,3500AberdeenAvenue,KirtlandAFB,NewMexico87117 Mang-Mang Ling, Abhijit Basu Mallik, Mark Roberts, and Zhenan Bao DepartmentofChemicalEngineering,StanfordUniversity,381NorthSouthMall,Stanford, California94305 (cid:1)Received 17 January 2006; accepted 1 March 2006; published online 12April 2006(cid:2) Preliminary studies of the effect of x-ray irradiation, typically used to simulate radiation effects in space, on top contact, pentacene based field effect transistors have been carried out. Threshold voltageshiftsinirradiateddevicesareconsistentwithpositivechargetrappinginthegatedielectric andareboundeffectisobserved,independentofthesignofappliedelectricfieldduringirradiation. Carriermobilityvariationsinpositiveelectricfieldbiased/irradiateddevicesareinterpretedinterms of the effects of interface-state-like defects. © 2006 American Institute of Physics. (cid:3)DOI:10.1063/1.2194894(cid:4) Organic semiconductor field effect devices based upon perform irradiation and electrical characterization of these the metal-oxide-semiconductor field effect transistor devices an Aracor 4100 system was used together with a (cid:1)MOSFET(cid:2) structure remain a tantalizing prospect for future Hewlett-Packard 4142 instrument. Devices were probed us- applications in the area of flexible, lightweight, and confor- ing W tips in situ in the irradiator. Such x-ray irradiation is mal electronics. Although the carrier mobilities are signifi- typically used to simulate the effects of very energetic elec- cantly lower than those observed in classical, crystalline, or tron or proton radiation effects in space. For the W target in polycrystalline semiconductors, thus limiting the likely fre- the irradiator a 50 kV accelerating voltage and a current of quency response of organic based devices, other advantages 5 mA gave an exposure rate of 130 rad (cid:1)SiO (cid:2) s−1 as mea- 2 may outweigh this in specific applications. One area in sured using a Si diode. Exposures up to 500 krad (cid:1)SiO (cid:2) 2 which organic electronics may hold significant promise is in were accumulated and during these electric fields of very high altitude or space technology where conformality ±1 MVcm−1 were applied across the MOSFET gate oxide maybeusefuland,inparticular,whereweightconsiderations (cid:1)with the source and drain contacts shorted to ground poten- are imperative. In order for organic devices to be considered tial(cid:2). Prior to and following exposure, source/drain current for these applications, assessment of their reliability in the (cid:1)I (cid:2)wasmeasuredasafunctionofappliedgatevoltage(cid:1)V (cid:2) ds gs presence of radiation is essential and thus far little has been for a fixed source/drain voltage of −100 V. A typical mea- reportedonthissubject.1Furthermore,itisimportanttonote surement time was (cid:5)10 s including data transfer.All of the that radiation levels in space and at high altitude are such experimental curves were thus measured in the saturation that the primary mechanism of degradation is through accu- regime.5 A sequence of electrical measurements was also mulated defect generation as opposed to macroscopic mate- made in which MOSFET devices were stressed electrically rial destruction.Astudy of radiation effects therefore brings in the same way and for the same times as they were during us directly into the realm of defects in organic semiconduc- irradiationbutwithoutsubjectingthemtoanyradiation.This tors which also appears to be an area of growing interest.2–4 process was undergone in order to ascertain mobility and In order to begin to determine the consequences of irra- threshold voltage variations resulting from electrical stress diating organic based MOSFETs, we have taken the arche- alone. typicalmaterialpentacenewhichhasbeenwellcharacterized The freshly prepared devices had carrier mobilities and has one of the largest known carrier mobilities. The (cid:1)(cid:1)(cid:2)(cid:5)1 cm2V−1s−1 but by the time measurements were results of these preliminary studies are reported in the actually carried out in the irradiator (cid:1) had decreased following. to (cid:5)0.75 cm2V−1s−1. These values were determined in the Dry thermal oxides were grown to a thickness saturation regime for which5 of (cid:5)300 nm on degenerate n++ type Si wafers and the oxide I =(cid:1)Z/L(cid:2)(cid:1)C (cid:1)V −V (cid:2)2, (cid:1)1(cid:2) surfaces were then treated in a vapor of octadecyl triethox- ds ox gs th ysilane (cid:1)OTS(cid:2) for 5 h at 110 °C. Following this treatment where C is the gate oxide capacitance per cm2 and V is ox th pentacene was thermally evaporated to a thickness of the device threshold voltage. In the first instance, plots of 100 nm onto the oxide surface held at 80 °C. Finally, Au (cid:1)I (cid:2)1/2 as a function of V were made with V swept from ds gs gs electrodes (cid:1)45 nm(cid:2) were evaporated onto the pentacene sur- 0 to −100 V in −5 V steps. The slope of this plot then facethroughashadowmask.Thedrainandsourceelectrodes yielded (cid:3)(cid:1)Z/L(cid:2)(cid:1)C (cid:4)1/2 and the intercept V . This measure- ox th were 4000 (cid:1)m in width with a spacing of 200 (cid:1)m, giving a menthasseveraldirectadvantagesoverthealternativelinear devicechannelwidth(cid:1)Z(cid:2)tolength(cid:1)L(cid:2)ratioof20.Inorderto regime measurement5 of I as a function of V with V ds ds gs constant. Firstly, the saturation current is independent of V ds a(cid:2)Author to whom correspondence should be addressed; electronic mail: sothatcontactresistance6effectsareirrelevantandsecondly, [email protected] distortions in the initial behavior of Ids as a function of Vds 0003-6951/2006/88(cid:2)15(cid:1)/151907/3/$23.00 88,151907-1 ©2006AmericanInstituteofPhysics Downloaded 25 Aug 2006 to 129.238.237.96. Redistribution subject to AIP license or copyright, see http://apl.aip.org/apl/copyright.jsp 151907-2 Devineetal. Appl.Phys.Lett.88,151907(cid:2)2006(cid:1) FIG.1. Measuredvariationofthethresholdvoltageshift(cid:1)(cid:2)Vth(cid:2)indevices FIG.2. Variationofthecarriermobilitynormalizedtotheprestress,preir- subjecttoelectricalbiasstressingandelectricalbiasstress/x-rayirradiation. radiation value for different devices shown in Fig. 1. (cid:1)(cid:1)(cid:2) Bias stress of (cid:1)(cid:2)(cid:2)Biasstressof1MVcm−1,noradiation,(cid:1)(cid:2)(cid:2)biasstressof−1MVcm−1, 1MVcm−1,noradiation,(cid:1)(cid:2)(cid:2)biasstressof−1MVcm−1,noradiation(cid:1)the no radiation (cid:1)the equivalent time in seconds is given by the accumulated equivalent time in seconds is given by the accumulated dose/130(cid:2), (cid:1)(cid:3)(cid:2) dose/130(cid:2),(cid:1)(cid:3)(cid:2)1MVcm−1andirradiation,and(cid:1)(cid:4)(cid:2)−1MVcm−1andirra- 1MVcm−1 and irradiation, and (cid:1)(cid:4)(cid:2) −1MVcm−1 and irradiation (cid:1)lower x diation(cid:1)lowerxaxisforaccumulateddoseapplies. axisforaccumulateddoseapplies. resulting from electric field dependent Poole-Frenkel as we concluded, from threshold voltage measurements, that conduction7 can be ignored. bias stressing resulted in charge trapping (cid:1)positive or nega- The electrical bias stressing data for the threshold volt- tive(cid:2) in the semiconductor close to the semiconductor/ age shift (cid:1)(cid:2)V (cid:2) are shown in Fig. 1, (cid:1)(cid:1)(cid:2) for the case of a dielectric interface. th 1 MVcm−1fieldand(cid:1)(cid:2)(cid:2)forthecaseofa−1 MVcm−1field. In Fig. 1 we also present the data (cid:2)V obtained for th Since we are primarily concerned with radiation effects in devicessubjectedtoelectricalstressandsimultaneouslyirra- our measurements the x axis is shown in accumulated radia- diated with x rays. For both positively (cid:1)(cid:3)(cid:2) and negatively tion dose in krad (cid:1)SiO (cid:2). For the electric field stress data (cid:1)(cid:4)(cid:2) biased devices the effect of irradiation appears to be to 2 wherenoradiationwasusedtheappropriatestressingtimeis induce strong negative shifts of the threshold voltage shift found by dividing the accumulated dose by the dose rate of for accumulated doses up to (cid:5)75 krad (cid:1)SiO2(cid:2) followed by 130 rad (cid:1)SiO (cid:2) s−1. The initial threshold voltages prior to what appears to be a rebound effect. In order to assess what 2 stressing or irradiation were typically in the range of might be termed the “radiation” induced (cid:2)Vth we have sub- −13 to −17 V.Thebehaviorofthethresholdvoltageshift,at tracted the (cid:2)Vth values measured for bias stress alone from least for both positive and negative electric fields, is similar the values determined for devices subject to simultaneous to that reported by other authors.8 A positive variation in bias stress and irradiation.The resultant curves are shown in (cid:2)V with time is suggestive of trapping of negative charge Fig.3.Itwouldappearthatthereisstillasignificantnegative th either in the gate oxide dielectric or in the semiconductor close to the semiconductor/dielectric interface. Similarly, negative shifts in (cid:2)V are consistent with positive charge th trapping. Given the magnitude of the electric fields applied in both positive and negative cases (cid:1)small with respect to those required for Fowler-Nordheim injection, for example(cid:2) it is unlikely that the observed effects result from charges in thedielectricsincethepotentialbarriertochargeinjectionis large.Themostlikelytrappingsiteisthereforeintheorganic semiconductorandthedatainFig.1wouldthenevidencethe existenceofbothpositiveandnegativechargetrappingsites. It must be remembered by that the bias induced charge trap- ping is in addition to the substantial positive charge already presentintheunstressedmaterialorintheas-grownoxideas evidenced by the large, negative, as-made V values. th In Fig. 2, using a similar representation as for (cid:2)V , we th show the data obtained for the mobility variation as a func- tion of applied electric field and stress time. For simplicity themobilityvaluesarenormalizedtotheirunstressedvalues. Withintherangeofexperimentalerroronecanconcludethat FIG.3. Estimatedvariationofthethresholdvoltageduesolelytoeffectsof electrical stressing does not result in significant mobility radiation deduced by subtraction: (cid:2)Vth (cid:1)irr(cid:2)=(cid:1)(cid:2)Vth(cid:2)(cid:1)irradiation+biasstress(cid:2) variation at least up to (cid:5)3850 s, irrespective of the sign of r−a(cid:1)d(cid:2)iaVtitho(cid:2)n(cid:1)b,iasasntrdesso(cid:1)n(cid:4)ly(cid:2)(cid:2)(cid:1)(cid:3)−1(cid:2) 1MMVVcmcm−1−1dduurirningg ssttrreessssiinnggananddstrsetsrseisnsginagndanird- the applied electric field. This result is interesting inasmuch irradiation. Downloaded 25 Aug 2006 to 129.238.237.96. Redistribution subject to AIP license or copyright, see http://apl.aip.org/apl/copyright.jsp 151907-3 Devineetal. Appl.Phys.Lett.88,151907(cid:2)2006(cid:1) voltageshiftuptodosesintherangeof75–100 krad(cid:1)SiO (cid:2) negative charge trapping in oxides which compensates the 2 andthemagnitudesseemtobeindependentofthesignofthe effect of trapping positive charge. It must be underlined that applied bias electric field during irradiation. this effect can vary significantly with the processing under- In Fig. 2 we present the variation of the mobility as a gone by the thermally grown oxide so that no general state- function of accumulated x-ray dose. Irradiation in the pres- ments about orders of magnitude, etc., can be made. ence of negative bias results in little observable variation of On the basis of the preliminary measurements reported the mobility while irradiation in the presence of a positive here we are able to confirm previous measurements of posi- bias induces a clearly visible reduction in the carrier mobil- tiveandnegativechargebuildupsintheorganicsemiconduc- ity. In a classical model of reduction of mobility due to the tor, resulting from electrical stressing without irradiation. presence of interfacial states,9the threshold voltage shift can The most likely locality of this trapped charge is within the be written as10 semiconductor. When radiation is also present, charge trap- (cid:2)V =(cid:3)q/(cid:1)(cid:3)C (cid:2)(cid:4)(cid:1)(cid:2)(cid:1)/(cid:1)(cid:2)/(cid:1)1−(cid:2)(cid:1)/(cid:1)(cid:2), (cid:1)2(cid:2) ping in the gate oxide dominates the behavior of the thresh- th ox 0 0 old voltage but a reduction of the mobility under positive where q is the electronic charge, (cid:3)is a processing related bias and irradiation evidences the fact that there appears to constant,(cid:1) istheundegradedmobility,and(cid:2)(cid:1)isthevaria- be a buildup of interface-state-like defects. However, for an 0 tion in (cid:1) resulting from degradation. From Fig. 2 for the accumulateddoseof500 krad(cid:1)SiO (cid:2)weobservedamobility 0 2 positive bias/irradiated case we have (cid:2)(cid:1)/(cid:1)(cid:5)0.14 for an reduction of only 14% which strongly suggests that the or- 0 x-ray dose of 500 krad (cid:1)SiO (cid:2) so that (cid:2)V (cid:5)−2.6 ganicdevicesstudiedhereareintrinsicallyradiationhardand 2 th (cid:4)10−12/(cid:3)cm2V. A typical value11 of (cid:3) for Si is therefore potentially useful in the space environment. 2(cid:4)10−12 cm2 which suggests (cid:2)V (cid:5)−1.3 V, resulting from th “interfacelike” states. For the negative biased/irradiated case This work was supported under AFOSR Contract No. we have (cid:2)(cid:1)/(cid:1)(cid:5)0 so there would be no associated thresh- LRIR 06VS01COR. 0 old voltage shift. We note from Fig. 3 that a value of (cid:2)V (cid:5)−1.3 V would be difficult to discern in the scatter of 1L. F. Drummy, Y. J. Yang, and D. C. Martin, Ultramicroscopy 99, 247 th (cid:1)2004(cid:2). the data. We can therefore conclude that though significant 2R. A. Street, A. Salleo, and M. L. Chabinyc, Phys. Rev. B 68, 085316 differences occur in the behavior of the mobility when irra- (cid:1)2003(cid:2). diated under positive or negative bias, the associated thresh- 3D.V.Lang,X.Chi,T.Siegrist,A.M.Sergent,andA.P.Ramirez,Phys. old voltage shifts are too small to be clearly visible in the Rev.Lett. 93,076601(cid:1)2004(cid:2). behavior of (cid:2)V as a function of bias and irradiation. What 4J.E.NorthrupandM.L.Chabinyc,Phys.Rev.B 68,041202(cid:1)2003(cid:2). th 5S. M. Sze, Physics of Semiconductor Devices (cid:1)Wiley, NewYork, 1981(cid:2), is perhaps important is that irradiation under positive bias Chap.8. appears to result in the presence of interface-state-like de- 6P.V.Necliudov,M.S.Shur,D.J.Grundlach,andT.N.Jackson,J.Appl. fectswhereasallotherconditionsofbiasstressorbiasstress Phys. 88,6594(cid:1)2000(cid:2). and irradiation do not. 7P.Stallinga,H.L.Gomes,F.Biscarini,M.Murgia,andD.M.deLeeuw, The data in Fig. 3 indicate the presence of substantial J.Appl.Phys. 96,5277(cid:1)2004(cid:2). 8D. Knipp, R. A. Street, A. Völkel, and J. Ho, J. Appl. Phys. 93, 347 thresholdvoltageshiftduetoradiationinducedcharges.The (cid:1)2003(cid:2). sign and magnitudes of the shifts, at least up to accumulated 9P.J.McWhorterandP.S.Winokur,Appl.Phys.Lett. 48,133(cid:1)1986(cid:2). doses of (cid:5)100 krad (cid:1)SiO (cid:2), are consistent with observations 10R.A.B.Devine,J.-L.Autran,W.L.Warren,andK.L.Vanheusden,Appl. 2 of charge buildup in irradiated oxides.12 They are usually Phys.Lett. 70,2999(cid:1)1997(cid:2). 11S.C.SunandJ.D.Plummer,IEEETrans.ElectronDevices ED-27,1497 attributed to positive charge trapping. In other oxides, re- (cid:1)1980(cid:2). bound effects similar to those observed in Fig. 3 for doses 12P.Paillet,D.Hervé,J.-L.Leray,andR.A.B.Devine,Appl.Phys.Lett. (cid:5)100 krad (cid:1)SiO2(cid:2) have been attributed to the presence of 63,2088(cid:1)1993(cid:2). Downloaded 25 Aug 2006 to 129.238.237.96. Redistribution subject to AIP license or copyright, see http://apl.aip.org/apl/copyright.jsp