High-Frequency Limits of Graphene Field-Effect Transistors with Velocity Saturation
High-Frequency Limits of Graphene Field-Effect Transistors with Velocity Saturation
Wilmart, Quentin;Boukhicha, Mohamed;Graef, Holger;Mele, David;Palomo, Jose;Rosticher, Michael;Taniguchi, Takashi;Watanabe, Kenji;Bouchiat, Vincent;Baudin, Emmanuel;Berroir, Jean-Marc;Bocquillon, Erwann;Fève, Gwendal;Pallecchi, Emiliano;Plaçais, Bernard
applied sciences Article High-Frequency Limits of Graphene Field-Effect Transistors with Velocity Saturation 1 1,2 1 1 1 Quentin Wilmart , Mohamed Boukhicha , Holger Graef , David Mele , Jose Palomo , 1 3 3 4 Michael Rosticher , Takashi Taniguchi , Kenji Watanabe , Vincent Bouchiat , 1 1 1 1 Emmanuel Baudin , Jean-Marc Berroir , Erwann Bocquillon , Gwendal Fève , 5 1, Emiliano Pallecchi and Bernard Plaçais * Laboratoire de Physique de l’Ecole Normale Supérieure, ENS, Université PSL, CNRS, Sorbonne Université, Université Paris-Diderot, Sorbonne Paris Cité, 75005 Paris, France; firstname.lastname@example.org (Q.W.); email@example.com (M.B.); firstname.lastname@example.org (H.G.); email@example.com (D.M.); firstname.lastname@example.org (J.P.); email@example.com (M.R.); firstname.lastname@example.org (E.B.); email@example.com (J.-M.B.); Erwann.Bocquillon@phys.ens.fr (E.B.); firstname.lastname@example.org (G.F.) Sustainable Energy Technologies Department, Brookhaven National Laboratory, Upton, NY 11973, USA Advanced Materials Laboratory, National Institute for Materials Science, Tsukuba, Ibaraki 305-0047, Japan; email@example.com (T.T.); WATANABE.Kenji.firstname.lastname@example.org (K.W.) Institut Néel, CNRS and Université Grenoble-Alpes, 38042 Grenoble, France; Vincent.Bouchiat@grenoble.cnrs.fr Institut d’Électronique, de Microélectronique et de Nanotechnologie, CNRS and University of Lille, 59491 Villeneuve d’Asq, France; Emiliano.Pallecchi@iemn.univ-lille1.fr * Correspondence: email@example.com or firstname.lastname@example.org Received: 29 November 2019; Accepted: 31 December 2019; Published: 8 January 2020 Abstract: The current understanding of physical principles governing electronic transport in graphene ﬁeld effect transistors (GFETs) has reached a level where we can model quite accurately device operation and predict intrinsic frequency limits of performance. In this work, we use this knowledge to analyze DC and RF transport properties of bottom-gated graphene on boron nitride ﬁeld effect transistors exhibiting pronounced velocity saturation by substrate hyperbolic phonon polariton scattering, including Dirac pinch-off effect. We predict and demonstrate a maximum oscillation frequency exceeding 20 GHz. We discuss the intrinsic 0.1 THz limit of GFETs and envision plasma resonance transistors as an alternative for sub-THz narrow-band detection. Keywords: Graphene transistor; velocity saturation; high-frequency; Dirac pinch-off; contact gating 1. Introduction Graphene is considered as a high-potential material for electronics, especially radio-frequency (RF) ﬁeld effect transistors (GFETs) , owing to the high intrinsic mobility of charge carriers [2–5]. The main challenge in this respect is to achieve current saturation at high bias to obtain signiﬁcant voltage and power gain. Two routes are considered. The ﬁrst relies on band-gap engineering using, e.g., lateral conﬁnement in graphene nanoribbon devices , which is often obtained at the cost of a reduced mobility. As in MOSFETs , the saturation results from a potential barrier induced at the drain side. The second route is velocity saturation, which promises high performances [8–10]. 2 1 1 It requires high mobility transistors (m & 1 m V s ) and results in large saturation currents (I 20 Acm ) . Limited by the Fermi energy at low doping, the saturation velocity is sat determined by optical phonon (OPs) scattering at large doping when E exceeds the optical phonon energy h ¯W [11,12]. In oxide-supported graphene, relevant OPs are substrate polar phonons (SPPs) OP [13–15] that have a moderate energy h ¯W . 50 meV. The saturation current being proportional SPP Appl. Sci. 2020, 10, 446; doi:10.3390/app10020446 www.mdpi.com/journal/applsci Appl. Sci. 2020, 10, 446 2 of 12 to the OP energy, power applications favor non-oxide dielectrics such as SiC (h ¯W 116 meV), SiC or hexagonal boron nitride (h-BN), which has two Reststrahlen bands, h ¯W 95–100 meV and h ¯W 170–200 meV, associated respectively with out-of-plane and in-plane optical phonons. Uniaxial II dielectrics, such as h-BN, actually sustain hyperbolic phonon polariton modes (HPhPs) that differ from SPPs (and intrinsic OPs), insofar as they propagate deep in the dielectric bulk, providing a unique radiative cooling pathway [11,16]. This paper discusses velocity saturation GFETs, their optimal design and operating conditions, their intrinsic frequency limit, and envisions the possibility to overcome this limit using plasma resonance devices [17,18]. It relies on experimental data taken on two series of h-BN-supported GFETs based on CVD graphene, which is appropriate for wafer scale applications. 2. Methods Samples used in this work are made of CVD graphene on a thin h-BN ﬂake deposited on top of a metallic bottom gate. The substrate consists in a 280 nm thermal SiO insulating barrier on top of high-resistivity Si suitable for RF measurements. Two types of gate electrodes are used: a thin tungsten electrode deposited on SiO (type A samples) for low-frequency devices and DC characterization, and a thick gold dual gate buried in SiO (type B samples) for RF devices. Several samples of both types have been measured presenting similar characteristics, but for simplicity, we focus on one illustrative sample of each type. Sample-A (Figure 1a) is made out of a 20-nm-thin tungsten ﬁlm that is etched to deﬁne the gate whose dimensions are L = 0.5 m for the length and W = 1.5 m for the width. In addition, we have included local contact gates (also tungsten) controlling the carrier density of the contacted graphene to reduce the contact resistance (see ). An h-BN ﬂake, of thickness t = 16 nm, is deposited BN on top of the gate by a dry transfer technique . The h-BN surface is cleaned by furnace annealing at 300 C under an Ar/H ﬂow. The CVD graphene is grown on copper following the pulsed CVD method at Institut Néel  and transferred with a wet technique on top of h-BN. Again, the polymer residues are removed by furnace annealing. Finally, 50 nm of Pd is evaporated for the electrical contacts, and devices are embedded in a 50 W-matched coplanar waveguide (CPW) used for DC and RF characterization. A schematic of a type-A device can be seen in Figure 1b. In very-high mobility devices such as the Zener-Klein transistor of  radiative cooling by hyperbolic phonon polaritons in the can be achieved, which is optimized using thicker (t & 100 nm) h-BN substrates. Note also that BN alternative WSe2 substrates can be used that provide even higher mobility at room temperature . The fabrication steps of sample-B (see Figure 2a below) are identical, except for the gate electrode: First a trench of SiO is realized with e-beam lithography and dry etching with a carefully controlled depth of 200 nm. Another e-beam lithography step deﬁnes a gate electrode within the SiO trench, and gold is evaporated up to a nearly perfect alignment with the SiO surface (an excess of a few nm of gold is preferable to avoid any air gap between the gate and h-BN). The gate dimensions are L = 0.8 m and W = 22 m, and h-BN thickness is 18 nm. The low contact resistance of sample-A (R 125 Wm) is minimized by contact gating implying that the total resistance is dominated by the channel. Fully ambipolar contact gating was demonstrated in Ref. ; in this work the channel length was small so that the transistor was controlled by the contact junction resistance. Here we work with long channels, longer than the optical phonon scattering length, to realize velocity saturation transistors with large voltage gain; contact gating is used instead to minimize the contact resistance. This geometry is well suited for the identiﬁcation of such mechanisms and their modeling. The sample-B geometry is best suited for maximizing the oscillation frequency f , as dual gates provide a better impedance match of the device and the thick max gold allows a reduction of the gate electrode resistance, which is a limiting factor for f . Our devices max 2 1 1 fulﬁll the ﬁled-effect mobility requirement m & 1 m V s at low temperature. Experiments are carried out in a Janis cryogenic probe station at 8 K in the band 0–40 GHz. A vector network analyzer (VNA) is used to measure the S-parameters of our transistors. Appl. Sci. 2020, 10, 446 3 of 12 ds a) b) drain BN graphene source drain gate hBN gate 1μm contact gate source 2 2 c) d) V =−3.5 → 0 V V =0 → 3.5 V g g 1.5 1.5 p-doping n-doping 1 1 0.5 0.5 0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8 V (V) V (V) ds ds 2500 2500 ds f) e) 2000 2000 0.1 0.2 V =0 → 0.8 V ds V =0 → 0.8 V 0.3 ds 1500 0.4 1500 0.5 0.6 0.7 1000 1000 0.8 −2 0 2 −2 0 2 V (V) V (V) Figure 1. Low frequency ﬁeld effect transistors (GFETs). (a) SEM image of sample-A. Note the absence of ungated area. (b) Electrical sketch of samples used in this study. For sample-A we have L = 500 nm, W = 1.5 m and t = 16 nm. (c) I–V characteristic of sample A in the p-doping regime. (d) I–V BN characteristic of sample-A in the n-doping regime. (e) DC resistance of sample-A. (f) RF small signal resistance of sample-A. The global increase of resistance with V is due to HPhP scattering only ds for negative gate voltages and both HPhP scattering and Dirac pinch-off for positive gate voltages. The resistance peak of dV /dI is the best working point for an RF transistor. ds V/I (Ω) I (mA) dV/dI (Ω) I (mA) Appl. Sci. 2020, 10, 446 4 of 12 a) c) 300 2 b) gate drain −1 g V m g r C 100 gd C C gs ds 50 −2 −5 0 5 −5 0 5 source V (V) V (V) e) d) 100 25 25 f exp. f exp. max f model f model max 20 20 15 15 10 10 5 5 0 0 0.1 −5 0 5 −5 0 5 1 10 V (V) V (V) g g f (GHz) Figure 2. Radio-frequency GFETs. (a) Optical image of sample-B before graphene and contact deposition. Dimensions: L = 800 nm, W = 22 m and t = 18 nm. (b) Small signal lumped BN elements circuit of an RF transistor. (c) RF small signal resistance r and transconductance g of sample-B measured at 2 GHz for biases in the 0–2 V range. (d) Logarithmic plot of the current H(f) and Mason U(f) gains measured between 1 GHz and 40 GHz after de-embedding. (e) V dependence of the corresponding measured cut-off frequencies (black circles) together with the model prediction (red lines). 3. Characterization of Low-Frequency GFETs Figure 1c,d shows the I–V characteristic of sample-A in the p-doping and n-doping regime respectively. Both regimes show a trend toward saturation, with larger currents on the p-side due 12 2 to a lower contact resistance (the contacted graphene is p-doped at ~0.7 10 cm by the contact gate kept at a constant voltage here). Figure 1e shows sample-A’s DC resistance for increasing bias voltages V . In the low doping regime, there is a broadening of the resistance peak as V is increased, ds ds which we attribute to a thermal effect. More interestingly, in the highly doped regime, we observe an increase of the resistance which is generally attributed to optical phonon scattering. However the relevant quantity is rather the small signal resistance dV /dI, shown in Figure 1f, which is directly ds obtained from the S-parameters and averaged over a 10 MHz to 4 GHz bandwidth. Indeed, on one hand, the small signal resistance is more sensitive to current saturation and on the other hand, it is used to describe the operation of an RF transistor . In addition to the increase of resistance in the doped regime, we observe a pronounced shift of the resistance peak towards positive gate voltages (the resistance peak shifts towards negative V when V is negative). This peculiar shape g ds is characteristic of locally gated devices  and not observed when using a remote Si gate. This effect, already introduced in Refs. [8,22], is a consequence of ﬁnite quantum capacitance effects and geometry which lead to a non-uniform carrier density in the channel at large V and eventually to a ds Dirac pinch-off. The next section is devoted to the development of a comprehensive model taking into account both mechanisms. Gain r (Ω) f (GHz) g (mS) f (GHz) max Appl. Sci. 2020, 10, 446 5 of 12 4. Simple Model of GFETs 4.1. Substrate Phonon Scattering Model The scattering by optical phonons has been identiﬁed as an important phenomenon of high-bias transport in high mobility materials like carbon nanotubes . However, the inﬂuence of the substrate cannot be neglected, and the impact of surface polar phonons (SPP) on transport properties has been invoked in several studies to explain the current saturation [8,12,14,15,24]. Those studies have been conducted with graphene on oxide samples, and to our knowledge, the effect of HPhPs has only recently been demonstrated using high-mobility exfoliated graphene on h-BN transistors . It was found that velocity saturation is provided by the lower W -HPhP band and hot electron cooling by the upper W -HPhP band. The out-of-plane optical phonons of the W band lift the sublattice degeneracy II I of graphene leading to an (unscreened) electron–phonon coupling reminiscent of the mass disorder mechanism reported in Ref. . In Refs. [11,16] exfoliated graphene was used and the mobility 2 1 1 was high enough (m > 3 m V s ) to access interband transport at very high bias, the so-called Zener–Klein regime, which is not reached in the present study due to limited mobility. The heuristic model used to describe HPhP scattering is inspired by the optical phonon-electron scattering model used by Yao et al. for carbon nanotubes . One considers that electrons are accelerated by an electric ﬁeld E = V /L up to the HPhP energy (here h ¯W ' 0.1 eV) to emit a ds phonon. In this notation, V is the voltage across the channel without potential drop at the contacts. ds h ¯W This interaction can be expressed as a scattering length for the electron–HPhP interaction l = , HPhP eE which is density-independent as represented by the dashed lines in Figure 3a for different electric ﬁelds. Our samples are not totally ballistic, and we need to account for a ﬁnite impurity scattering length l = mh ¯ pn/e, shown as a black solid line (the effective mobility is m = 1 m /V/s). The resulting imp 1 1 1 1 scattering length is given by the Matthiessen rule l = l + l , which is shown atE = 1 V m scatt imp HPhP in Figure 3a as a green solid line. As seen in the ﬁgure, at high bias, the transport is indeed dominated k W l 0 4e F scatt by HPhP scattering. The total scattering length l is translated into conductance G = . scatt h p L+l scatt It can be more conveniently expressed as a resistance: eV hp L hp ds R = + + (1) 2 2 4e k W Wnem 4e k W h ¯W F F I The ﬁrst term of Equation (1) is the ballistic resistance, the second is the well-known diffusive term, and the third corresponds to the HPhP scattering contribution; it is responsible for saturation and proportional to V . Note that the linear V dependence of the DC resistance (negative quadratic ds ds deviation in the current) depends only on the carrier density and the HPhP energy. That is how we can identify the h ¯W ' 0.1 0.01 eV energy of h-BN HPhPs as the main source of scattering and rule out the scattering by intrinsic graphene optical phonons or the W -HPhP band, as h ¯W h ¯W 0.2 eV. II OP II Note that the role of intrinsic OPs or W -HPhPs in energy relaxation can be identiﬁed using noise II thermometry respectively in suspended  and h-BN supported graphene [11,16]. This illustrates the difference between momentum relaxation (the resistance) and energy relaxation (the temperature or electronic distribution) mechanisms (Another example of such a difference, is acoustic phonon scattering which gives rise to a linear temperature dependence for resistivity but a cubic one for relaxation, reﬂecting single phonon and inelastic supercollision scattering respectively ). In the full saturation regime all terms are negligible compared to the HPhP term and one obtains the saturation current: 4e k W h ¯W F I I = (2) sat h p e Appl. Sci. 2020, 10, 446 6 of 12 Physically, the current saturation is due to the carrier velocity saturation v : sat I 2 h ¯W sat v = v (3) sat F neW p E Equations (1)–(3) strictly hold at large doping when E & h ¯W ; the crossover to Fermi energy F I 2 2 1/2 limited current at low doping can be taken into account by substituting (E + h ¯W ) for h ¯W  F I (The 2/p pre-factor, which stems from the angular projection of electron momentum along the current axis, is sometimes omitted in the literature ). In Ref.  the saturation current was deduced by 1 1 0 2 ﬁtting the differential conductance by the law R (V ) = R (0)V /[1 + (V /R(0)I )] , which can ds ds sat ds be deduced from Equation (1). The saturation is not complete in our samples due to limited mobility and we cannot extract I and v directly. In Figure 3b, we show the calculated drift velocity v as a sat sat d function of E (solid lines) for different densities from the current saturation model with h ¯W = 0.1 eV. 12 2 Further, we show the measured v at large density (n = 5 10 cm ), where channel density is mostly homogeneous. The very good agreement between theory and experiment, in a regime where HPhP scattering dominates, conﬁrms the relevance of our model. As a conclusion of this section, we note that the effect of high mobility is mainly to decrease the threshold voltage of saturation and thus to clarify the transition toward the current saturation regime. This is why full current saturation is only observed in high-mobility samples. a) b) l (0.5 V/μm) HPhP l (0.75 V/μm) imp HPhP l (1 V/μm) HPhP 12 −2 n=1 → 5 × 10 cm l (1 V/μm) scatt 0 0.2 0.4 0.6 0.8 1 −2 0 2 12 −2 𝓔 (V/µm) n (10 cm ) Figure 3. (a) Scattering length associated with the impurity scattering (black solid line) responsible for 4 2 the low ﬁeld mobility; here m = 10 cm /V/s, and with the HPhP scattering (dashed lines) for different in plane electric ﬁelds E = V /L. The green thick line is the total scattering length at the largest bias ds obtained by the Matthiessen’s law. (b) Carrier drift velocity simulated using v = V /(RneW) with R ds 12 2 from Equation (1), and the measured drift velocity at large doping (n = 5 10 cm ) in black circles. 4.2. Dirac Pinch-Off Model In a remote gate transistor, for example using a global silicon gate, the gate uniformly controls the charge of the semiconducting channel. However, considering a local gate transistor, where local 2 2 2 means that the quantum capacitance C = 2e E /(ph ¯ v ) and the geometrical capacitance per unit Q F area C = e e /t are on the same order of magnitude, the charge in the channel is not set by the g 0 r BN gate voltage only and it can be inﬂuenced by the electrostatic potential in the channel. Applying a voltage on the drain electrode results in a linear electrochemical potential along the channel if we assume diffusive conduction, which is the case at large bias due to HPhP scattering. We write the electrochemical balance at the position x in the channel: DE (x) eV (x) = eV (4) F gr ds scattering length (nm) v (10 m/s) d Appl. Sci. 2020, 10, 446 7 of 12 where V (x) is the electrostatic potential of graphene. This equation is combined with the electrostatic gr equilibrium n(x)e = C (V (x) V ), which allows calculating DE (x): g gr g 0 1 je j @ A DE = sign(e ) e 1 + 2 1 , (5) F W c 2 2 ph ¯ v 0 x with e = C and e (x) = eV eV . It results in an inhomogeneous DE within the channel. c g g 2 W F 2e ds This effect is sketched in Figure 4a, and the calculated DE (x) is shown in Figure 4b for increasing V . ds The corresponding carrier density in the channel as a function of V and x is shown in the color plot of Figure 4c. Starting from an n-doped channel and increasing V leads to a decrease of the carrier density ds on the drain side. Figure 4b shows that the neutrality can be reached and that for larger V there ds is even a localized neutral point whose location is shifted toward the source electrode. At the onset of neutrality on the drain side, increasing V induces an increase of the resistance, which ampliﬁes ds the sublinear behavior of the I–V characteristic and results in enhanced current saturation. This is the Dirac pinch-off. On the contrary, starting from a p-doped channel, an increase of V leads to an ds increase of the carrier density on the drain side which contributes to decreasing the channel resistance. This explains the observed asymmetric shape of the high bias resistance curve in Figure 1f. a) b) source drain 100 graphene μ* eV gr 𝓔 = 0 → 1 V/μm −50 μ* eV gr eV ds −100 0 100 200 300 400 500 x (nm) c) d) 12 −2 −1 n (10 cm ) r (Ωnm ) ds 3.5 2.5 3.5 𝓔 = 1 V/μm 𝓔 = 1 V/μm n doping 2.5 2 0 0 1.5 p doping −2.5 1 −5 −3.5 0.5 −3.5 0 100 200 300 400 500 0 100 200 300 400 500 x (nm) x (nm) Figure 4. (a) Sketch of the doping proﬁle across the graphene channel when V is increased. Starting ds from n-doping, DE (x) can be decreased down to neutrality at the drain side. This is the Dirac pinch-off. (b) Simulated E (x) along the channel for increasing in-plane electric ﬁeld. (c) Color plot of the local carrier density in the channel as a function of the x-position along the channel and the gate voltage V (d) Color plot of the local resistivity r (V , x) at V = 0.5 V. The total resistance of the device is the ds ds integration over x of this resistivity. 4.3. Hot Electron Effects To accurately describe the behavior of our transistor, we have to take into account the thermal broadening that arises at large V . Indeed, for high biased samples, the electronic temperature can ds V (V) E (meV) V (V) g F Appl. Sci. 2020, 10, 446 8 of 12 rise up to 800 K , which signiﬁcantly modiﬁes the energy distribution. We take into account this thermal broadening by writing the carrier density with a residual term n that depends on V : 0 ds u ! DE (x) t F n(x) = + n (V ) (6) 2 ds ph ¯ v 0 11 2 where n (V ) is the sum of an impurity term (n = 2 10 cm at zero bias) and a hot 0 imp ds D(k T ) 0 B electron term: n (V ) . Here, T is governed by the interaction between electrons and th e 2 2 ds ph ¯ v acoustic phonons of graphene. It has been shown in previous studies  that, at low doping and moderate impurity scattering, electron cooling is controlled by electron–phonon-impurity 2 3 supercollisions and that for a given Joule power input one has V /(RLW) = AT (In very high ds mobility h-BN supported/encapsulated samples, supercollisions are suppressed and the leading relaxation mechanism becomes black-body and luminescent W -HPhP emission). In this equation II we have A µ n and R µ 1/n so that AR is constant. Finally, the hot electron residual density reads as follows, 0 4/3 n = a(V ) , (7) th ds 3 2/3 2 4/3 where a 9 10 /(WL) cm V is a constant that depends on sample dimensions. This is used to calculate the local resistance r (V , x), which is plotted in Figure 4d. for V = 0.5 V. As expected, ds g ds the maximum resistance line follows the minimum carrier density line, with a width related to V . ds 4.4. Complete Model and Discussion Let us now calculate the total resistance of the graphene channel taking into account HPhP scattering, the non-uniform carrier density in the channel and the hot electron effect. We compute the 1 1 1 total scattering length l (x) = l (V , x) + l (x) and neglect the ballistic conductance to scatt g imp HPhP obtain the local resistance ph dx r (x)dx = p (8) ds l (x) 4e W n(x)p scatt The total channel resistance is then obtained by integrating over the channel length: R = r (x)dx (9) ds ds Finally, in short devices with high-mobility graphene, the contact resistance R is not negligible. Most of the contact resistance comes from the junction between contacted graphene and the channel. This issue has been overcome in a previous study with the transistor geometry of sample-A . It allows us to calculate the V dependent contact resistance. The total transistor resistance is then R(V , V ) = R (V , V ) + 2R (V ). For sample-A, we have R 125 Wm on the p-doping side g g c g c ds ds ds (where there are p-p junctions at the two contact edges) and R 275 Wm on the n-doping side (where there are p-n junctions at the two contact edges). The relevant quantity for an RF transistor is the small signal resistance dV /dI that we calculate ds ds from the current through the transistor I = V /R and the applied bias voltage V across the device. ds ds ds 4 2 1 1 The simulated dV /dI is plotted in Figure 5a where the mobility m = 10 cm V s and R are ds ds c determined from the zero bias transfer curve and the thermal broadening coefﬁcient a (from ) is in principle universal. The very good agreement between measurement and simulation shows that we are able to accurately predict the behavior of a local gate transistor driven at high bias knowing the graphene mobility, the contact resistance and the substrate phonon energy. It validates our transistor model and conﬁrms the role of the h-BN HPhPs at h ¯W = 0.1 eV in velocity saturation. To emphasize the effect of the local gate on the saturation we plot in Figure 5b, a simulation for a substrate thickness of t = 300 nm where all parameters are conserved (apart from V which is adjusted to conserve the BN Appl. Sci. 2020, 10, 446 9 of 12 zero bias density). There is a signiﬁcant difference between local and remote gate on the shape of the transfer curve. In Figure 5b, the slight asymmetry between p and n doping comes from the contact resistance difference. 2.5 2.5 local gate remote gate t = 15 nm t = 300 nm BN BN 2 2 V =0 → 0.8 V ds V =0 → 0.8 V ds 1.5 1.5 1 1 0.5 0.5 0 0 −2 0 2 −5 0 0 50 V (V) V (V) g g (a) (b) Figure 5. (a) Small signal resistance of the transistor obtained from the model (solid lines) and from the measured scattering parameters of sample-A at 2 GHz (circles). (b) Small signal resistance simulated for a remote gate transistor (h-BN dielectric thickness of 300 nm). The saturation effect is enhanced for the local gate device in the n-doping regime. 5. Characterization of RF-GFETs The performance of Sample-A as an RF transistor is assessed at high frequency. Measuring the S-parameters gives a direct access to the channel small signal resistance r and the transconductance g = dI /dV shown in the schematic of Figure 2b. We also measure the current gain H and the m g ds Mason gain U (equivalent to the power gain). Those gains are sensitive to extrinsic parameters such as stray capacitances and, for U only, the resistance of the gate electrode. After removing the effect of stray capacitances by a de-embedding procedure, we obtain the current gain cut-off frequency f = 9 GHz and the Mason’s gain cut-off frequency f = 7 GHz. The fact that f f is a distinctive feature of max max T velocity saturation GFETs, as f f in mobility-limited GFETs . The value of f is limited max T max here by the large resistance R 10 Ohms of the tungsten gate electrode. This is why we built a thick gold gate device in sample-B (shown in Figure 2a). In sample-B, the channel width is also further increased to W = 22 m, to increase the total current in the device. In terms of f , however, a compromise has to be found between a large max current which means a large transconductance and a larger gate resistance which strongly limits f . max The small signal characteristics of sample-B, i.e., the resistance r and transconductance g , are plotted in Figure 2c. One can notice the strong saturation effect on the n-side where the high bias r is larger than the zero bias resistance peak. In this sample, the n-side is also the low contact resistance side which results in a larger g on the n-side. Both effects contribute to a larger voltage gain for positive gate voltages. This is conﬁrmed by measuring H and U at the maximum bias and V = 2 V plotted in Figure 2d after de-embedding. In Figure 2e, we plot the V -dependence of f and f at V = 2.4 V. g max T ds Our model allows estimating r and g which is useful to determine the transistor ’s ﬁgures of merit from f = g /(2pC ) and m g f = q (10) max 2 (R + R )/r + 2p f C R G c T g G The predicted f and f in Figure 2e compare fairly well with our measurement. Both ﬁgures T max of merit present two peaks for p- and n-doping. In the case of f , the peak asymmetry comes from the dV/dI (kΩ) dV/dI (kΩ) Appl. Sci. 2020, 10, 446 10 of 12 contact resistance asymmetry. For f the peak asymmetry is stronger because it is also sensitive to a max stronger saturation on the n-doping side. The availability of a reliable model allows ﬁne tuning parameters to optimize f . It appears max that in our sample the large gate resistance is the main limitation. Engineering the transistor extrinsic elements can improve f by a large factor, for example, by using a T-shaped gate [9,30]. max Another important limiting factor is the contact resistance. 6. Perspectives on High-Frequency Graphene Devices In an attempt to assess the intrinsic frequency limits of GFETs, we include here an analysis of Zener–Klein transistors where contact resistance is low enough, mobility is large enough and channel length long enough (L = 4 m), to access a fully developed inter-band Zener–Klein tunneling regime . A quasi perfect current saturation (I & 20 Acm ) is achieved thanks sat to the combined intraband velocity saturation, interband Zener–Klein tunneling and Dirac pinch-off (Figure 1 of Ref. ). The small contact resistance R ' 150 Wm , large DC transconductance g & 2.5 Scm , and saturation resistance R lead to a large voltage gain G ' 8. From these m sat numbers, we estimate f f G/4g R ' 7 f (taking R R in Equation (10)) translating to max T m c T G c f & 100 GHz (taking f = g /2pC ' 16 GHz with L = 2 m and t = 20 nm). Considering max m g T BN the optimization of saturation velocity, contact resistance and channel length, these ﬁgures constitute the intrinsic frequency limit of GFETs. It still remains to be demonstrated with edge-contacted, h-BN encapsulated, GFETs at room temperature. This intrinsic limitation of GFETs can be circumvented relying on collective modes such as plasmons which beneﬁt from an enhanced velocity v v appropriate for sub-THz operation. p F This corresponds to plasma resonance transistors (PRTs) , which are multi-mode narrow-band detectors. Intrinsic Dirac-fermion plasma modes can be studied in plasma resonance capacitors (PRCs), which behave as quarter-mode resonators (fundamental at f = v /4L and harmonics f = (2k + 1) f ). In a recent study , an h-BN encapsulated PRC working at f = 40 GHz k 0 0 (v ' 4v ) was demonstrated, showing the relevance of long-wavelength plasma resonance devices in fully characterizing high-mobility graphene, offering a full analysis of the temperature and density dependence of the quantum capacitance C (T, n), kinetic inductance L (T, n) and damping or Q K mean-free path l (T, n). scat 7. Conclusions We have fabricated, modeled, and measured high-performance velocity-saturation GFETs showing promising perspectives for microwave detection and ampliﬁcation. Our modeling involves physical insight into microscopic mechanisms of velocity saturation and a proper account of Dirac pinch-off effects. Our work allows assessing the intrinsic frequency limits of GFETs using velocity saturation associated with Zener–Klein tunneling and opens perspectives “beyond-GFETs”, using plasma resonance techniques for sub-THz detection. The ﬁgure of merit is the f / f ratio which max T increases from f / f . 0.2 in the mobility-limited regime, over f / f 1 in the velocity-saturation max T max T regime, up to f / f & 5 in the Zener–Klein regime. The latter offers promising perspectives in max T RADAR and GSM applications. Author Contributions: Q.W., M.B., E.P. and B.P. conceived the experiment. T.T. and K.W. provided the hBN crystals. V.B. provided the CVD graphene. Q.W., M.B., H.G., J.P. and M.R. did the nanofabrication. Q.W., M.B., H.G. and B.P. did the electrical measurements. Q.W., M.B., H.G., D.M., E.B. (Emmanuel Baudin), J.-M.B., G.F., E.B. (Erwann Bocquillon) and B.P. participated to the data analysis and theoretical modelling. Q.W. and B.P. wrote the manuscript with contributions from all coauthors. All authors have read and agreed to the published version of the manuscript. Funding: The research leading to these results have received partial funding from the European Union Horizon 2020 research and innovation programme under grant No. 785219 “Graphene Core 2”, and from the ANR-14-CE08-018-05 “GoBN”. The work of Q.W. was supported by a DGA-MRIS scholarship. Acknowledgments: We thank thank L. Becerra for providing the tungsten ﬁlms. Appl. Sci. 2020, 10, 446 11 of 12 Conﬂicts of Interest: The authors declare no conﬂicts of interest. References 1. Wu, Y.; Jenkins, K.A.; Valdes-Garcia, A.; Farmer, D.B.; Zhu, Y.; Bol, A.A.; Dimitrakopoulos, C.; Zhu, W.; Xia, F.; Avouris, P.; et al. State-of-the-Art Graphene High-Frequency Electronics. Nano Lett. 2012, 12, 3062–3067. [CrossRef] 2. Wang, L.; Meric, I.; Huang, P.Y.; Gao, Q.; Gao, Y.; Tran, H.; Taniguchi, T.; Watanabe, K.; Campos, L.M.; Muller, D.A.; et al. One-Dimensional Electrical Contact to a Two-Dimensional Material. Science 2013, 342. [CrossRef] [PubMed] 3. Han, Z.; Kimouche, A.; Kalita, D.; Allain, A.; Arjmandi-Tash, H.; Reserbat-Plantey, A.; Marty, L.; Pairis, S.; Reita, V.; Bendiab, N.; et al. Homogeneous Optical and Electronic Properties of Graphene Due to the Suppression of Multilayer Patches During CVD on Copper Foils. Adv. Funct. Mater. 2014, 24, 964–970. [CrossRef] 4. Banszerus, L.; Schmitz, M.; Engels, S.; Dauber, J.; Oellers, M.; Haupt, F.; Watanabe, K.; Taniguchi, T.; Beschoten, B.; Stampfer, C. Ultrahigh-mobility graphene devices from chemical vapor deposition on reusable copper. Sci. Adv. 2015, 1, 2375–2548. [CrossRef] [PubMed] 5. Banszerus, L.; Schmitz, M.; Engels, S.; Goldsche, M.; Watanabe, K.; Taniguchi, T.; Beschoten, B.; Stampfer, C. Ballistic Transport Exceeding 28 m in CVD Grown Graphene. Nano Lett. 2016, 16, 1387. [CrossRef] 6. Meng, N.; Fernandez, J.F.; Vignaud, D.; Dambrine, G.; Happy, H. Fabrication and Characterization of an Epitaxial Graphene Nanoribbon-Based Field-Effect Transistor. IEEE Trans. Electron Dev. 2016, 58, 1594. [CrossRef] 7. Sze, S.M.; Ng, K. Physics of Semiconductor Devices, 3rd ed.; John Wiley & Sons: Hoboken, NJ, USA, 2007; p. 293. 8. Meric, I.; Han, M.Y.; Young, A.F.; Ozyilmaz, B.; Kim, P.; Shepard, K.L. Current saturation in zero-bandgap, top-gated graphene ﬁeld-effect transistors. Nat. Nanotechnol. 2008, 3, 654–659. [CrossRef] 9. Guo, Z.; Dong, R.; Chakraborty, P.S.; Lourenco, N.; Palmer, J.; Hu, Y.; Ruan, M.; Hankinson, J.; Kunc, J.; Cressler, J.D.; et al. Record Maximum Oscillation Frequency in C-Face Epitaxial Graphene Transistors. Nano Lett. 2013, 13, 942–947. [CrossRef] 10. Feng, Z.H.; Yu, C.; Li, J.; Liu, Q.B.; He, Z.Z.; Song, X.B.; Wang, J.J.; Cai, S.J. An ultra clean self-aligned process for high maximum oscillation frequency graphene transistors. Carbon 2014, 75, 249–254. [CrossRef] 11. Yang, W.; Berthou, S.; Lu, X.; Wilmart, Q.; Denis, A.; Rosticher, M.; Taniguchi, T.; Watanabe, K.; Fève, G.; Berroir, J.M.; et al. A graphene Zener–Klein transistor cooled by a hyperbolic substrate. Nat. Nanotechnol. 2018, 13, 47. [CrossRef] 12. Barreiro, A.; Lazzeri, M.; Moser, J.; Mauri, F.; Bachtold, A. Transport Properties of Graphene in the High-Current Limit. Phys. Rev. B 2009, 103, 076601. [CrossRef] [PubMed] 13. Fratini, S.; Guinea, F. Substrate-limited electron dynamics in graphene. Phys. Rev. B 2008, 77, 195415. [CrossRef] 14. Freitag, M.; Steiner, M.; Martin, Y.; Perebeinos, V.; Chen, Z.; Tsang, J.C.; Avouris, P. Energy Dissipation in Graphene Field-Effect Transistors. Nano Lett. 2009, 9, 1883. [CrossRef] [PubMed] 15. Perebeinos, V.; Avouris, P. Inelastic scattering and current saturation in graphene. Phys. Rev. B 2010, 81, 5442. [CrossRef] 16. Baudin, E.; Voisin, C.; Plaçais, B. Hyperbolic Phonon Polariton Electroluminescence as an Electronic Cooling Pathway. Adv. Funct. Mater. 2019. [CrossRef] 17. Graef, H.; Mele, D.; Rosticher, M.; Stampfer, C.; Taniguchi, T.; Watanabe, K.; Bocquillon, E.; Fève, G.; Berroir, J.M.; Teo, E.T.H.; et al. Ultra-long wavelength Dirac plasmons in graphene capacitors. emph. J. Phys. Mater. 2018, 1, 01LT02. [CrossRef] 18. Bandurin, D.A.; Svintsov, D.; Gayduchenko, I.; Xu, S.G.; Principi, A.; Moskotin, M.; Tretyakov, I.; Yagodkin, D.; Zhukov, S.; Taniguchi, T.; et al. Resonant terahertz detection using graphene plasmons. Nat. Commun. 2018, 9, 5392. [CrossRef] 19. Wilmart, Q.; Inhofer, A.; Boukhicha, M.; Yang, W.; Rosticher, M.; Morﬁn, P.; Garroum, N.; Fève, G.; Berroir, J.-M.; Placais, B. Contact gating at GHz frequency in graphene. Sci. Rep. 2016, 6, 21085. [CrossRef] Appl. Sci. 2020, 10, 446 12 of 12 20. Banszerus, L.; Sohier, T.; Epping, A.; Winkler, F.; Libisch, F.; Haupt, F.; Watanabe, K.; Taniguchi, T.; Müller-Caspary, K.; Marzari, N.; et al. Extraordinary high room-temperature carrier mobility in graphene-WSe2 heterostructures. arXiv 2019, arXiv:1909.09523v1. 21. Schwierz, F. Graphene transistors. Nat. Nanotechnol. 2010, 5, 487. [CrossRef] 22. Bai, J.; Liao, L.; Zhou, H.; Cheng, R.; Liu, L.; Huang, Y.; Duan, X. Top-Gated Chemical Vapor Deposition Grown Graphene Transistors with Current Saturation. Nano Lett. 2011, 11, 2555–2559. [CrossRef] [PubMed] 23. Yao, Z.; Kane, C.L.; Dekker, C. High-Field Electrical Transport in Single-Wall Carbon Nanotubes. Phys. Rev. Lett. 2000, 84, 2941–2944. [CrossRef] [PubMed] 24. Dorgan, V.E.; Bae, M.-H.; Pop, E. Mobility and saturation velocity in graphene on SiO . Appl. Phys. Lett. 2010, 97, 082112. [CrossRef] 25. Pallecchi, E.; Betz, A.C.; Chaste, J.; Fève, G.; Huard, B.; Kontos, T.; Berroir, J.-M.; Plaçais, B. Transport scattering time probed through rf admittance of a graphene capacitor. Phys. Rev. B 2011, 83, 125408. [CrossRef] 26. Laitinen, A.; Kumar, M.; Oksanen, M.; Plaçais, B.; Virtanen, P.; Hakonen, P. Coupling between electrons and optical phonons in suspended bilayer graphene. Phys. Rev. B 2015, 91, 121414. [CrossRef] 27. Betz, A.C.; Jhang, S.H.; Pallecchi, E.; Ferreira, R.; Fève, G.; Berroir, J.-M.; Plaçais, B. Supercollision cooling in undoped graphene. Nat. Phys. 2013, 9, 109–112. [CrossRef] 28. Betz, A.C.; Vialla, F.; Brunel, D.; Voisin, C.; Picher, M.; Cavanna, A.; Madouri, A.; Fève, G.; Berroir, J.-M.; Plaçais, B.; et al. Hot electron cooling by acoustic phonons in graphene. Phys. Rev. Lett. 2012, 109, 056805. [CrossRef] 29. Pallecchi, E.; Benz, C.; Betz, A.C.; Löhneysen, H.v.; Plaçais, B.; Danneau, R. Graphene microwave transistors on sapphire substrates. Appl. Phys. Lett. 2011, 99, 113502. [CrossRef] 30. Mele, D.; Mehdhbi, S.; Fadil, D.; Wei, W.; Ouerghi, A.; Lepilliet, S.; Happy, H.; Pallecchi, E. Graphene FETs Based on High Resolution Nanoribbons for HF Low Power Applications. Electron. Mater. Lett. 2018, 14, 133. [CrossRef] c 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.pngApplied SciencesMultidisciplinary Digital Publishing Institutehttp://www.deepdyve.com/lp/multidisciplinary-digital-publishing-institute/high-frequency-limits-of-graphene-field-effect-transistors-with-Bw0QfiOv5k
High-Frequency Limits of Graphene Field-Effect Transistors with Velocity Saturation