326 IEEE TRANSACTIONS ON ELECTRON DEVICES, VOL. 63, NO. 1, JANUARY 2016

Application Relevant Evaluation of Trapping Effectsin AlGaN/GaN HEMTs With Fe-Doped BufferOlle Axelsson, Sebastian Gustafsson, Student Member, IEEE, Hans Hjelmgren, Senior Member, IEEE,

Niklas Rorsman, Member, IEEE, Hervé Blanck, Jörg Splettstoesser, Jim Thorpe,Thomas Roedle, and Mattias Thorsell, Member, IEEE

Abstract— This paper investigates the impact of differentiron (Fe) buffer doping profiles on trapping effects in microwaveAlGaN/gallium nitride (GaN) high electron mobility transis-tors (HEMTs). We characterize not only the current collapse dueto trapping in the buffer, but also the recovery process, whichis important in the analysis of suitable linearization schemes foramplitude modulated signals. It is shown that the simple pulseddc measurements of current transients can be used to investigatetransient effects in the RF power. Specifically, it is revealed thatthe design of the Fe-doping profile in the buffer greatly influencesthe recovery time, with the samples with lower Fe concentrationshowing slower recovery. In contrast, traditional indicators, suchas S-parameters and dc as well as pulsed I–V characteristics,show very small differences. An analysis of the recovery showsthat this effect is due to the presence of two different detrappingprocesses with the same activation energy (0.6 eV) but differenttime constants. For highly doped buffers, the faster processdominates, whereas the slower process is enhanced for less dopedbuffers.

Index Terms— Dispersion, gallium nitride (GaN), high electronmobility transistors (HEMTs), semiconductor device doping,trap levels.

I. INTRODUCTION

THE HIGH output power density and efficiency offered bypower amplifiers (PAs) based on gallium nitride (GaN)

high electron mobility transistors (HEMTs) make them

Manuscript received May 5, 2015; revised October 23, 2015; acceptedNovember 4, 2015. Date of publication November 23, 2015; date of cur-rent version December 24, 2015. This work was supported in part by theSwedish Governmental Agency of Innovation Systems, Chalmers Universityof Technology, Classic WBG Semiconductors AB, Comheat MicrowaveAB, Ericsson AB, Infineon Technologies Austria AG, Mitsubishi ElectricCorporation, Saab AB, SP Technical Research Institute of Sweden, and UnitedMonolithic Semiconductors, through the GigaHertz Centre in a Joint ResearchProject, and in part by the Research Project entitled Advanced III-Nitrides-Based Electronics for Future Microwave Communication and Sensing Systemsthrough the Swedish Foundation for Strategic Research. The review of thispaper was arranged by Editor K. J. Chen. (Corresponding author: OlleAxelsson.)

O. Axelsson, S. Gustafsson, H. Hjelmgren, N. Rorsman, and M. Thorsellare with the Department of Microtechnology and Nanoscience, ChalmersUniversity of Technology, Gothenburg SE-412 96, Sweden (e-mail:ollea@chalmers.se; sebgus@chalmers.se; hans.hjelmgren@chalmers.se;niklas.rorsman@chalmers.se; mattias.thorsell@chalmers.se).

H. Blanck and J. Splettstoesser are with United Monolithic Semiconduc-tors Gesellschaft mit beschränkter Haftung, Ulm 89081, Germany (e-mail:herve.blanck@ums-ulm.de; joerg.splettstoesser@ums-ulm.se).

J. Thorpe is with Irrierse/Experior Micro Technologies Ltd.,Derby DE1 2RJ, U.K. (e-mail: jim.thorpe@experiormicrotech.com).

T. Roedle is with NXP Semiconductors, Nijmegen 6534 AE,The Netherlands (e-mail: thomas.roedle@nxp.com).

Color versions of one or more of the figures in this paper are availableonline at http://ieeexplore.ieee.org.

Digital Object Identifier 10.1109/TED.2015.2499313

a strong candidate for high-frequency power applications.However, the dispersive behavior of these transistors inhibitsboth Continuous Wave performance and linearity, and hasdelayed their adoption in many microwave systems. Thedispersive effects are mainly caused by deep electron traps inthe buffer and at the surface. These dispersive effects are man-ifested as a decrease in maximum current (current slump) andON-state conductance (knee walkout) during large signaloperation, leading to lower maximum output power andefficiency [1], [2]. Also of great concern from a systemperspective but less studied on device level is the distortion ofamplitude modulated signals due to these memory effects. Lin-earization schemes using digital predistortion are impracticalfor long time constants, which require a large computationalcapacity and memory [3], [4].

Research on traps in GaN has primarily been focusedon understanding their physical mechanisms and minimizingthe current collapse. Earlier, much efforts were dedicated tominimize the effects of surface traps by utilizing differentpassivation schemes [1], [2], [5]. More recently, there has beena growing interest in the influence of the characteristics of theGaN-buffer on the large-signal performance of GaN HEMTs.The GaN-buffer is commonly doped with a deep acceptor suchas iron (Fe) to reduce leakage currents and increase breakdownvoltage [6], [7]. However, the design of the buffer, in termsof compensation doping concentration and profile is known tohave an effect on trapping phenomena. In particular, a trapwith activation energy ∼0.6 eV is thought to be localizedin the buffer, and it has been seen that its concentrationincreases with the Fe-dopant concentration [8]–[11]. In [11],it was predicted from simulations that Fe traps would causea small current collapse that is almost independent of theFe concentration above a certain threshold level. However,it has also been suggested that the 0.6 eV traps seen inmeasurements on the Fe-doped buffers are not due to theFe atoms but another trap indirectly affected by the doping [8].

This paper presents an application-focused investigation ofthe effect of different buffer Fe-doping profiles on GaN HEMTPA performance, focusing in particular on transient effectsdue to trapping after high-power operation. This is importantin the analysis of suitable linearization schemes but oftennot considered in other studies. It is shown that the simplepulsed dc measurements of current transients can be used toinvestigate transient effects in the RF power. While standarddevice characterization methods (e.g., dc and pulsed I–V ,S-parameters, and so on) reveal only small differences between

0018-9383 © 2015 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.

AXELSSON et al.: APPLICATION RELEVANT EVALUATION OF TRAPPING EFFECTS IN AlGaN/GaN HEMTs 327

Fig. 1. (a) Illustration of the three tested buffer doping profiles. (b) Pictureof a mounted transistor connected to the package with bond wires.

samples with different buffer Fe profiles, we show that therecovery process after a high-power pulse is significantlyaffected by the Fe doping. Furthermore, these effects areanalyzed by determining the time constants and activationenergies of the traps involved.

This paper is organized as follows. In Section II, the devicesunder test (DUT) and the differences in buffer doping areexplained. Section III compares the dc and small signal perfor-mance of the devices. Section IV describes the measurementsused to characterize trapping effects in the devices and presentsthe results of these measurements. Finally, the conclusions aredrawn in Section V.

II. TECHNOLOGY

The DUT is 0.5-μm gate length GaN HEMTs with6-μm × 400-μm gate width, fabricated by United MonolithicSemiconductors using Cree GaN wafers grown on SiC. Thegate-source and gate-drain distances are 1.5 and 3 μm, respec-tively. A gate-connected field plate extends 0.5 μm towardthe drain side and a source-connected field plate 2.3 μmfrom the source and above the gate. The lower part of theGaN buffer is Fe doped in order to obtain high resistivityand increase the breakdown voltage. Since Fe-doping in thechannel has been shown to cause severe current collapse [7],the Fe-doping profiles are designed to have an exponentialdecrease with a distance toward the AlGaN/GaN interface.This is inherently achieved during the metal–organic chemicalvapor deposition growth by memory effects after the closingof the Fe source. Fig. 1(a) shows the three tested buffer Feprofiles. The total buffer thickness is 1.8 μm for all devices.The standard doping profile has an approximately 1-μm-thick plateau with an Fe concentration in the 1018 cm−3

range, decreasing exponentially to around 1016 close to theAlGaN/GaN interface. Besides the standard buffer design,the Fe-doping close to the channel was varied by reducingthe thickness of the Fe plateau by 20% while keeping thetotal buffer thickness the same (reduced thickness) or byreducing the Fe concentration in the plateau by 50% (reducedconcentration). All measurements have been carried out ontwo devices of each type to assure the reproducibility ofobserved variations. However, only the data for one deviceof each type are presented, unless there are specific reasons tocomment on differences between nominally identical devices.The transistors were mounted in a package, and bond wireswere used to connect the gate, drain, and source terminals tothe package [Fig. 1(b)].

Fig. 2. DC characteristics of the three tested device types. (a) Drain current–voltage characteristics. Inset: measurement of the drain–source breakdownvoltage where the drain voltage is measured, while the gate voltage isswept and the drain current is kept at a constant 200 μA/mm. (b) Pinchoffcharacteristics at Vds = 0.5 V (--) and Vds = 50 V (-).

TABLE I

FIGURES OF MERIT FROM DC CHARACTERIZATION

III. DC, S-PARAMETER, AND TWO-TONE

CHARACTERIZATION

The devices were characterized by measuring thedc characteristics and the S-parameters up to 9 GHz. Thedrain current characteristics of the three device types aresummarized in Fig. 2 and Table I. Both the maximum current(at +1 V gate bias) and the maximum transconductancedecreased by ∼5%–10% in the devices with reduced Fe dop-ing. The breakdown voltage was measured using a currentinjection technique [12], where the drain current was keptconstant at 480 μA (200 μA/mm), while the gate voltagewas decreased to maintain the current level at higher Vds. Themeasured Vds is shown in Fig. 2(a) (inset). The more highlydoped buffers exhibit the highest breakdown voltage, but thevalues higher than 150 V were measured for all samples.

The pinchoff characteristics as a function of drain andgate voltage were characterized by the drain-induced barrierlowering (DIBL) and subthreshold slope (SS). DIBL is here

328 IEEE TRANSACTIONS ON ELECTRON DEVICES, VOL. 63, NO. 1, JANUARY 2016

Fig. 3. Comparison of (a) S21, (b) S22, and (c) Mason’s gain of the threedifferent types of devices for a bias point of Ids = 100 mA and Vds = 20 V.

TABLE II

FIGURES OF MERIT FROM SMALL SIGNAL CHARACTERIZATION

defined as the difference in gate threshold voltage in orderto obtain 1-mA current between 0.5 and 50 V drain bias,and SS is the slope of log10(Ids) with respect to the gatevoltage in the subthreshold region, measured in decade/V.As shown in Table I, DIBL is slightly higher in the deviceswith decreased Fe doping. Although small, the differencesare consistent across samples and expected, since a higherdoping level reduces the short-channel effects by improvingthe electron confinement in the channel [13]. The differencesin SS were small and inconsistent across different samples.The OFF-state drain current is ∼0.1 mA and dominated bygate leakage for all devices at Vds = 50 V.

The measured S-parameters show very small differencesbetween the different buffers. Fig. 3(a) and (b) showsS21 and S22, and the two S-parameters that can be expectedto be most affected by changes in the buffer. Fig. 3(c) showsMason’s gain and the extrapolation of fmax. Some figures ofmerit derived from the small signal characterization at this biasare summarized in Table II. The extrinsic transconductance,output conductance, and output capacitance were extractedfrom the Y-parameters at 100 MHz.

Furthermore, two-tone measurements have been performedon conjugate matched HEMTs at 2.65 GHz at several biaspoints and tone spacings and showed no significant variationsin the third-order intercept point (IP3) between the three devicetypes up to output powers of 29 dBm per tone. The HEMTsachieved an output IP3 of 43 dBm at 28 V and 200 mA bias.

In conclusion, the different Fe-doping profiles did notdemonstrate any significant effects on dc, small signal

Fig. 4. Ids–Vds characteristics under pulsed conditions for a gate voltageof +1 V, when the devices are pulsed for 1 μs from three different bias points.

performance and linearity at low power levels and none ofthe tested devices showed leakage currents or breakdownvoltages that are problematic for PA operation. This impliesthat the Fe buffer can be optimized to a large degree for othercharacteristics, such as reduced dispersive effects.

IV. CHARACTERIZATION OF TRAPPING EFFECTS

To characterize the different consequences of dispersiveeffects caused by electron trapping in the buffer, severalcharacterization methods are needed. The traditional methodof characterizing trapping effects is pulsed I–V measurements,which are carried out in Section IV-A. This gives informationon the current collapse and knee-walkout that affect the outputpower and efficiency. However, the pulsed I–V characteristicsdo not reveal transient effects on the device RF performanceafter a change in operating conditions, such as moving froma high RF power level to a lower power level.

Direct characterization of the transient behavior is per-formed in Section IV-B. We use a vector modulator to supply ahigh-power RF input signal to a matched HEMT at 2.65 GHzfor 300 μs before backing off the input power level by 15 dB.The output power is monitored in the time domain during andafter the pulse using a vector signal analyzer.

However, this method requires expensive instrumentationand is impractical on device level, since it requires engineeringof the output current and voltage waveforms (e.g., using tuners,matching networks, or active injection) in order to createrealistic operating conditions for the device. Therefore, a morepractical method to obtain information of these effects ondevice level is desired. In Section IV-C, we propose that theeffects of high-power microwave operation can be mimickedby dc-pulsing to a high drain voltage, and the transient effectscan be monitored by measuring the drain current after thepulse at an application relevant bias.

A. Pulsed I–V Characterization

To obtain information about current collapse due to gate anddrain lag, both the drain and gate voltages were pulsed for 1 μsfrom three different quiescent points: {Vgs (V), Vds (V)} ={0, 0}, {−4, 0}, {−4, 20}. Fig. 4 shows the measured draincurrent–voltage characteristics at Vgs = +1 V, for all quiescentpoints along with the dc data.

AXELSSON et al.: APPLICATION RELEVANT EVALUATION OF TRAPPING EFFECTS IN AlGaN/GaN HEMTs 329

Fig. 5. Transient measurement of (a) gain and (b) drain current after the three different transistors were subjected to a 300-μs pulse into 3-dB compression.The drain bias was 20 V, and the quiescent current was 100 mA. Black lines in (a): estimation based on drain current transients combines with the staticbias-dependent S-parameters. Black lines in (b): prediction of model (1) with two exponential terms. (c) Gain collapse 0.5 ms after the pulse, measured andestimated from Ids transients.

TABLE III

RELATIVE DROP IN KNEE CURRENT UNDER PULSED CONDITIONS

Table III summarizes the drop in current at the knee voltage(Vgs = +1 V, Vds = 4 V) for the quiescent points, relativeto the {0, 0} point. Pulsing the gate from below the pinchoffvoltage (−4 V) results in a small drop in current, by 2%–3%,whereas a large dc voltage (20 V) on the drain cause a largerdegradation in the maximum current, more than 10% for allbuffers. This indicates that traps excited by high fields couldcause degradation in the RF performance. The devices with thestandard buffer show slightly larger current slump comparedwith HEMTs with reduced doping.

B. Pulsed RF Transient Characterization

Fig. 5(a) and (b) shows the gain and dc drain current asa function of time after the RF-pulse for the three devicesat a quiescent drain voltage of 20 V and a drain currentof 100 mA. The output power is 35 dBm, leading to 3-dBgain compression. Immediately after the pulse, the gain dropsby ∼0.5 dB, accompanied by a drop in the drain current downto 60 mA. The gain drop is larger for quiescent bias pointscloser to the pinchoff.

Whereas all the different devices experience a similar dropin gain and current, the devices with the more highly dopedstandard buffer recovers very close to the quiescent levelswithin ∼10 ms, whereas the initial recovery is slower in thetwo variations with a lower Fe content.

The correlation between the drop in gain and drain currentmakes it possible to roughly evaluate transient effects mea-suring only the latter, allowing a much simpler setup. This isshown in Fig. 5(a), where the gain transient is estimated usingthe measured Ids(t) and the static S-parameter measurementsat different values of Ids. It is assumed that the matching

is only slightly affected by the current transient, so that theamplifier gain varies with drain current as |S21|2. Furthermore,it is also assumed that the dynamic |S21| during the transientis equal to the static |S21| measured at the same drain currentand voltage. Fig. 5(c) shows the measured and predicted gaincollapse versus output power 0.5 ms after the pulse, comparedwith the quiescent level, for the three doping profiles. Whilethis method does not exactly predict the transient, it gives arough indication of the gain collapse as well as the recoverytime.

Analysis of time constants can give insights on themechanisms behind the current collapse and subsequentrecovery [14], and on the reasons to the differences betweendifferent buffers. By fitting the measured data to a sum ofexponentials

Ids = Idsq +∑

an exp

[− t

τn

](1)

we can extract the different time constants τn (in the followingnumbered in order of magnitude with τ1 being the largestone) and corresponding amplitudes an using optimization.A good fit could be obtained for different conditions withtwo exponential terms in (1). Fig. 5(b) shows the fit of themodel to the measured transients. The time constants τn andthe corresponding amplitudes an at different output powerlevels are shown in Fig. 6. Two distinct time constants in themillisecond range, separated by 1–1.5 orders of magnitude,can be seen in all devices. The time constants are very similaracross the different devices, but their magnitudes differ. Thehighly doped standard buffer has a larger magnitude of thefaster time constant but a smaller magnitude of the slowerone, explaining the faster recovery in these devices. The timeconstants increase with power level, as well as the amplitudes.

C. Pulsed DC Transient Characterization

In order to mimic PA operation, we have pulsed the drainvoltage from a quiescent point of 20 V to higher values for1 μs and measured the current response after the pulse. Thegate voltage was kept constant at a value giving 100 mAquiescent current. The current was chosen as low as possible

330 IEEE TRANSACTIONS ON ELECTRON DEVICES, VOL. 63, NO. 1, JANUARY 2016

Fig. 6. (a) Time constants and (b) corresponding amplitudes of the Ids transient after the RF pulse, as the functions of peak power. The colors and markersin (b) represent the different doping profiles as in (a).

Fig. 7. Drain current transients after a 1 μs pulse to Vds = 40 V from aquiescent point of Vds = 20 V, for a quiescent current, Idsq = 100 mA. Theend of the pulse is at t = 0. Black lines: fit to model (1) including threeexponential terms.

to minimize effects of self-heating but high enough to keepthe transconductance in a fairly linear region during theexperiment. Fig. 7 shows the drain current transients afterpulses to Vds = 40 V from 20 V for a quiescent currentof 100 mA, compared with model (1) with three time constantsextracted using optimization. The time constants τn (withτ1 being the largest one) and the corresponding amplitudes an

are plotted in Fig. 8 versus the magnitude of the voltage pulse(�VDS = Vpulse − Vquiecscent).

The drain current shows a similar drop in current andsubsequent recovery as during the pulsed RF measurement(Figs. 5 and 6), indicating that pulsing the drain voltage canindeed be used to characterize transient effects in these transis-tors and find their associated time constants. The initial currentdrop is a few milliamperes larger in the standard devices,which is consistent with the results of the pulsed I–V charac-terization in Section IV-A, where the more highly doped stan-dard devices showed slightly stronger current collapse. On theother hand, the recovery is faster for the more highly dopedbuffer, similar to what the RF transient measurements showed.

The two slowest time constants τ1 and τ2, around300 and 10 ms, respectively, after a 40 V pulse, have thelargest amplitudes and, thus, dominate the current collapse

Fig. 8. Extracted parameters of model (1) using three terms, as the functionsof voltage pulse magnitude. Dashed line: 1 dB/dB slope, showing that the twoslower time constants are close to proportional to the voltage step.

and subsequent recovery. Similar to the case of RF pulsing,the difference between the samples with varied buffer dopingprofile is that reduced doping seems to lead to increased |a1|but reduced |a2|, thus making the recovery slower by shift-ing it to the slower time constant. The time constants alsodecrease with smaller voltage pulses, which is consistent withRF pulsing leading to shorter time constants at lower powerlevels.

Drain current transient measurements were also carriedout at elevated ambient temperatures in order to extract theactivation energies of the traps behind the different timeconstants. Fig. 9(a) shows the Arrhenius plots and the extractedenergies. The two larger time constants have similar activationenergies in the range of 0.5–0.6 eV for all samples, whereasthe activation energy of the fastest time constant is difficultto extract without faster measurement systems. The activationenergies were extracted using temperatures measured on thefixture, without considering any temperature gradient insidethe device. Assuming a 25 K/W · mm thermal resistancebetween the fixture and the buffer would result in a 21 K

AXELSSON et al.: APPLICATION RELEVANT EVALUATION OF TRAPPING EFFECTS IN AlGaN/GaN HEMTs 331

Fig. 9. Temperature dependence of the extracted parameters from model (1).(a) Arrhenius plot gives activation energies ∼0.6 eV for the two slowest timeconstants. (b) Amplitudes associated with the two time constants as a functionof temperature. Colors and markers: different doping profiles as in (a).

temperature rise at the quiescient bias and an 80 meV higheractivation energy. The elevated temperature also affected theamplitude of the different time constants [Fig. 9(b)]. Theamplitude of the slowest time constant |a1| increased forhigher temperatures, whereas |a2| decreased, resulting in thetotal current collapse staying constant with temperature.

V. CONCLUSION AND DISCUSSION

In this paper, the trapping effects in differently Fe-dopedGaN buffers in high-power microwave AlGaN/GaN HEMTswere evaluated. Traditional performance indicators, such asdc and pulsed I–V characteristics, S-parameters, and IP3, arevery similar between the different buffers and the small dif-ferences cannot conclusively be attributed to the buffer dopingrather than to unintended variation in epitaxy or processing.

However, a large impact of the buffer doping level is seenin the recovery after a large dc or RF pulse. We detecttwo detrapping processes with the same activation energybut different time constants. In the highly doped buffers,the initial current collapse is slightly larger, but a largerproportion of the trapped electrons is emitted through thefaster process, resulting in a faster recovery. In contrast, lowerFe concentration in the buffer and channel leads to a slowerrecovery, since a larger proportion of the electrons is emittedthrough the slower process.

The activation energy of the two time constants (0.6 eV)is in the same range as what have been found in the

previous studies of GaN buffers with as well as withoutFe doping [8]–[10], [15]–[18]. The faster time constantincreased in significance with increasing Fe doping similarto what was reported in [8]–[10]. However, the second timeconstant associated with the same energy with the oppositeFe concentration dependence was not found in other studies.In [8], a slower time constant was found at 0.82 eV, butit differed from our observations not only in the activationenergy, but also in the fact that it did not show notabledependence on Fe doping.

The similarity of the drain current transients after pulsingthe drain voltage from its operating point to the transientafter a high-power RF pulse shows that the former can beused as a simple method to characterize memory effects inthe RF domain and can be used as a complement to pulsedI–V characterization.

The almost identical dc and S-parameter performances ofthe different Fe-doping profiles imply that the buffer dopingprofile largely can be optimized for minimizing the negativeeffects of trapping in the buffer. However, it should be notedthat the buffer doping differences between the tested sam-ples are relatively small. Reducing the compensation dopingfurther would eventually lead to short-channel effects anddecreased breakdown voltage [13], as well as larger effects ontrapping [11].

REFERENCES

[1] S. C. Binari, P. B. Klein, and T. E. Kazior, “Trapping effects in GaNand SiC microwave FETs,” Proc. IEEE, vol. 90, no. 6, pp. 1048–1058,Jun. 2002.

[2] U. K. Mishra, L. Shen, T. E. Kazior, and Y.-F. Wu, “GaN-based RFpower devices and amplifiers,” Proc. IEEE, vol. 96, no. 2, pp. 287–305,Feb. 2008.

[3] B. Vassilakis and A. Cova, “Comparative analysis of GaAs/LDMOS/GaN high power transistors in a digital predistortion amplifier system,”in Proc. Asia-Pacific Microw. Conf. (APMC), Dec. 2005.

[4] F. M. Ghannouchi and O. Hammi, “Behavioral modeling and predistor-tion,” IEEE Microw. Mag., vol. 10, no. 7, pp. 52–64, Dec. 2009.

[5] Y.-F. Wu et al., “30-W/mm GaN HEMTs by field plate optimization,”IEEE Electron Device Lett., vol. 25, no. 3, pp. 117–119, Mar. 2004.

[6] S. Heikman, S. Keller, S. P. DenBaars, and U. K. Mishra, “Growth of Fedoped semi-insulating GaN by metalorganic chemical vapor deposition,”Appl. Phys. Lett., vol. 81, no. 3, pp. 439–441, 2002.

[7] V. Desmaris et al., “Comparison of the DC and microwave performanceof AlGaN/GaN HEMTs grown on SiC by MOCVD with Fe-doped orunintentionally doped GaN buffer layers,” IEEE Trans. Electron Devices,vol. 53, no. 9, pp. 2413–2417, Sep. 2006.

[8] M. Meneghini et al., “Buffer traps in Fe-doped AlGaN/GaN HEMTs:Investigation of the physical properties based on pulsed and tran-sient measurements,” IEEE Trans. Electron Devices, vol. 61, no. 12,pp. 4070–4077, Dec. 2014.

[9] M. Silvestri, M. J. Uren, and M. Kuball, “Iron-induced deep-levelacceptor center in GaN/AlGaN high electron mobility transistors:Energy level and cross section,” Appl. Phys. Lett., vol. 102, no. 7,pp. 073501-1–073501-4, Feb. 2013.

[10] D. W. Cardwell et al., “Spatially-resolved spectroscopic measurementsof Ec −0.57 eV traps in AlGaN/GaN high electron mobility transistors,”Appl. Phys. Lett., vol. 102, no. 19, pp. 193509-1–193509-4, May 2013.

[11] M. J. Uren, J. Möreke, and M. Kuball, “Buffer design to minimizecurrent collapse in GaN/AlGaN HFETs,” IEEE Trans. Electron Devices,vol. 59, no. 12, pp. 3327–3333, Dec. 2012.

[12] S. R. Bahl and J. A. del Alamo, “A new drain-current injectiontechnique for the measurement of off-state breakdown voltage inFETs,” IEEE Trans. Electron Devices, vol. 40, no. 8, pp. 1558–1560,Aug. 1993.

[13] M. J. Uren et al., “Punch-through in short-channel AlGaN/GaNHFETs,” IEEE Trans. Electron Devices, vol. 53, no. 2, pp. 395–398,Feb. 2006.

332 IEEE TRANSACTIONS ON ELECTRON DEVICES, VOL. 63, NO. 1, JANUARY 2016

[14] J. Joh and J. A. del Alamo, “A current-transient methodology fortrap analysis for GaN high electron mobility transistors,” IEEE Trans.Electron Devices, vol. 58, no. 1, pp. 132–140, Jan. 2011.

[15] P. Hacke, T. Detchprohm, K. Hiramatsu, N. Sawaki, K. Tadatomo, andK. Miyake, “Analysis of deep levels in n-type GaN by transientcapacitance methods,” J. Appl. Phys., vol. 76, no. 1, pp. 304–309, 1994.

[16] A. Armstrong et al., “Impact of carbon on trap states in n-type GaNgrown by metalorganic chemical vapor deposition,” Appl. Phys. Lett.,vol. 84, no. 3, pp. 374–376, 2004.

[17] H. K. Cho, K. S. Kim, C.-H. Hong, and H. J. Lee, “Electron traps andgrowth rate of buffer layers in unintentionally doped GaN,” J. Cryst.Growth, vol. 223, nos. 1–2, pp. 38–42, 2001.

[18] S. Gustafsson et al., “Dispersive effects in microwave AlGaN/AlN/GaNHEMTs with carbon-doped buffer,” IEEE Trans. Electron Devices,vol. 62, no. 7, pp. 2162–2169, Jun. 2015.

Olle Axelsson received the M.Sc. degree in engi-neering physics from the Chalmers University ofTechnology, Gothenburg, Sweden, in 2010, wherehe is currently pursuing the Ph.D. degree with theMicrowave Electronics Laboratory.

His current research interests include the character-ization of robustness, linearity, and trapping effectsin GaN HEMTs.

Sebastian Gustafsson (S’13) received the M.Sc.degree in electrical engineering from the ChalmersUniversity of Technology, Gothenburg, Sweden,in 2013, where he is currently pursuing the Ph.D.degree with the Microwave Electronics LaboratoryGroup.

His current research interests include RF metrol-ogy and GaN HEMT characterization.

Hans Hjelmgren (M’91–SM’11) received the Ph.D.degree in electrical engineering from the ChalmersUniversity of Technology, Gothenburg, Sweden,in 1991. His thesis work dealt with the numericalsimulation of hot electrons in GaAs devices.

He is currently an Associate Professor withthe Chalmers University of Technology. His cur-rent research interests include the different aspectsof TCAD.

Niklas Rorsman (M’10) received the M.Sc. degreein engineering physics and the Ph.D. degree in elec-trical engineering from the Chalmers University ofTechnology, Gothenburg, Sweden, in 1988 and 1995,respectively.

He returned to the Chalmers University of Tech-nology in 1998, where he is currently leadingthe microwave wide bandgap technology activitiesand investigating the application of graphene inmicrowave electronics.

Hervé Blanck received the M.S. and Ph.D. degreesfrom the University of Strasbourg, Strasbourg,France, in 1986 and 1989, respectively.

He then joined the Central Research Labora-tory, Thomson-CSF, Orsay, France, where he wasinvolved in the GaInP HBT development from1990 to 1996. In 1996, he joined United MonolithicSemiconductors Gesellschaft mit beschränkter Haf-tung, Ulm, Germany, where he has been the Managerof the Technology Development Department since2003.

Jörg Splettstoesser, photograph and biography not available at the time ofpublication.

Jim Thorpe, photograph and biography not available at the time ofpublication.

Thomas Roedle, photograph and biography not available at the time ofpublication.

Mattias Thorsell (S’08–M’11) received the M.Sc.and Ph.D. degrees in electrical engineering fromthe Chalmers University of Technology, Gothenburg,Sweden, in 2007 and 2011, respectively.

He is currently an Assistant Professor with theChalmers University of Technology. His currentresearch interests include the characterization andmodeling of nonlinear microwave semiconductordevices.