Abstract
Recently, we have suggested a scaleinvariant model for a nanotransistor. In agreement with experiments a closetolinear threshold trace was found in the calculated I_{D } V_{D}traces separating the regimes of classically allowed transport and tunneling transport. In this conference contribution, the relevant physical quantities in our model and its range of applicability are discussed in more detail. Extending the temperature range of our studies it is shown that a closetolinear threshold trace results at room temperatures as well. In qualitative agreement with the experiments the I_{D } V_{G}traces for small drain voltages show thermally activated transport below the threshold gate voltage. In contrast, at large drain voltages the gatevoltage dependence is weaker. As can be expected in our relatively simple model, the theoretical drain current is larger than the experimental one by a little less than a decade.
Introduction
In the past years, channel lengths of fieldeffect transistors in integrated circuits were reduced to arrive at currently about 40 nm [1]. Smaller conventional transistors have been built [29] with gate lengths down to 10 nm and below. As wellknown with decreasing channel length the desired longchannel behavior of a transistor is degraded by shortchannel effects [1012]. One major source of these shortchannel effects is the multidimensional nature of the electrostatic field which causes a reduction of the gate voltage control over the electron channel. A second source is the advent of quantum transport. The most obvious quantum shortchannel effect is the formation of a sourcedrain tunneling regime below threshold gate voltage. Here, the I_{D } V_{D}traces show a positive bending as opposed to the negative bending resulting for classically allowed transport [13,14]. The sourcedrain tunneling and the classically allowed transport regime are separated by a closeto linear threshold trace (LTT). Such a behavior is found in numerous MOSFETs with channel lengths in the range of a few tens of nanometers (see, for example, [29]).
Starting from a threedimensional formulation of the transport problem it is possible to construct a onedimensional effective model [14] which allows to derive scaleinvariant expressions for the drain current [15,16]. Here, the quantity arises as a natural scaling length for quantum transport where ε_{F }is the Fermi energy in the source contact and m* is the effective mass of the charge carriers. The quantum shortchannel effects were studied as a function of the dimensionless characteristic length l = L/λ of the transistor channel, where L is its physical length.
In this conference contribution, we discuss the physics of the major quantities in our scaleinvariant model which are the chemical potential, the supply function, and the scaleinvariant current transmission. We specify its range of applicability: generally, for a channel length up to a few tens of nanometers a LTT is definable up to room temperature. For higher temperatures, a LTT can only be found below a channel length of 10 nm. An inspection of the I_{D } V_{G}traces yields in qualitative agreement with experiments that at low drain voltages transport becomes thermally activated below the threshold gate voltage while it does not for large drain voltages. Though our model reproduces interesting qualitative features of the experiments it fails to provide a quantitative description: the theoretical values are larger than the experimental ones by a little less than a decade. Such a finding is expected for our simple model.
Theory
TsuEsaki formula for the drain current
In Refs. [13,14], the transport problem in a nanoFET was reduced to a onedimensional effective problem invoking a "singlemode abrupt transition" approximation. Here, the electrons move along the transport direction in an effective potential given
(see Figure 1b). The energy zero in Equation 1 coincides with the position of the conduction band minimum in the highly ndoped source contact. As shown in [14]
Figure 1. Generic nchannel nanofield effect transistor. (a) Schematic representation. (b) Onedimensional effective potential V^{eff}.
where E_{k = 1 }is the bottom of the lowest twodimensional subband resulting in the zconfinement potential of the electron channel at zero drain voltage (see Figure 4b of Ref. [13]). The parameter W is the width of the transistor. Finally, V_{D }= eU_{D }is the drain potential at drain voltage U_{D }which is assumed to fall off linearly.
Experimentally, one measures in a wide transistor the current density J, which is the current per width of the transistor that we express as
Here is the number of equivalent conduction band minima ('valleys') in the electron channel and I_{0 }= 2eε_{F}/h. In Refs. [15,16] a scaleinvariant expression
was derived. Here, m = μ/ε_{F }is the normalized chemical potential in the source contact, v_{D }= V_{D}/ε_{F }is the normalized drain voltage, and v_{G }= V_{G}/ε_{F }is the normalized gate voltage. As illustrated in Figure 1(b) the gate voltage is defined as the energy difference μ  V_{0 }= V_{G}, i.e., for V_{G }> 0 the transistor operates in the ONstate regime of classically allowed transport and for V_{G }< 0 in the sourcedrain tunneling regime. The control variable V_{G }is used to eliminate the unknown variable V_{0}. For the chemical potential in the source contact one finds (see next section)
where u = k_{B}T/ε_{F }is the normalized thermal energy. Equation 4 has the form of a TsuEsaki formula with the normalized supply function
Here, F_{1/2 }is the FermiDirac integral of order 1/2 and is the inverse function of F_{1/2}. The effective current transmission depends on which is the normalized energy of the electron motion in the yzplane while is their energy in the xdirection. In the next sections, we will discuss the occurring quantities in detail.
Chemical potential in source and draincontact
For a wide enough transistor and a sufficient junction depth a (see Figure 1) the electrons in the contacts can be treated as a threedimensional noninteracting electron gas. Furthermore, we assume that all donor impurities of density N_{i }are ionized. From charge neutrality it is then obtained that the electron density n_{0 }is independent of the temperature and given by
Here m^{e }is the effective mass and N_{V }is the valleydegeneracy factor in the contacts, respectively. In the zero temperature limit a Sommerfeld expansion of the FermiDirac integral leads to
Equating 7 and 8 results in
which is identical with (5) and plotted in Figure 2. As wellknown, with increasing temperature the chemical potential falls off because the highenergy tail of the Fermidistribution reaches up to ever higher energies.
Figure 2. Normalized chemical potential vs. thermal energy according to Equation 9 in green solid line and parabolic approximation in red dashdotted line.
Supply function
As shown in Ref. [14] the supply function for a wide transistor can be written as
This expression can be interpreted as the partition function (loosely speaking the "number of occupied states") in the grand canonic ensemble of a noninteracting homogeneous threedimensional electron gas in the subsystem of electrons with a given lateral wave vector (k_{y}, k_{z}) yielding the energy in the yzdirection. Formally equivalent it can be interpreted as the full partition function in the grand canonic ensemble of a onedimensional electron gas at the chemical potential μ  ε. Performing the limit the Riemann sum in the variable can be replaced by the FermiDirac integral F_{1/2}. It results that
with the normalized transistor width w = W/λ. For the scaling of the supply function in Equation 11 we define (see Ref. [14])
where and we use the identity V_{0}= ε_{F }= m  v_{G}. For the source contact we write
leading to the first factor in the square bracket of the TsuEsaki equation 4. In the drain contact, the chemical potential is lower by the factor V_{D}. Replacing μ → μ  V_{D }yields
Below we will show that for transistor operation the low temperature limit is relevant (see Figure 2). Here, one may apply in leading order (resulting from a Sommerfeld expansion) and F_{1/2}(x → ∞) → exp (x). Since V_{0 }> 0 the factor v_{G } m is negative and we obtain from (12)
From Figure 3 it is seen that for ε below the chemical potential the supply function is well described by the squareroot dependence in the limit. If ε lies above the chemical chemical one obtains the limit which is a small exponential tail due to thermal activation.
Figure 3. Supply function in the source contact (see Equation 6) for u = 0.1 and v_{G }= 0 (black line), lowtemperature limit according to Equation 15 for α < 0 (red dashed line) and α > 0 (green dashed line). Because of the small temperature m(u) ~ 1 so that occurs at .
Current transmission
The effective current transmission in Equation 16 is given y
It is calculated from the scattering solutions of the scaled onedimensional Schrödinger equation
with β = 2m*V_{0}L^{2}/ħ^{2 }= l^{2}(m  v_{G}), and ŷ = y/L. The scaled effective potential is given by , , and ,where . As usual, the scattering functions emitted from the source contact obey the asymptotic conditions and
with and .
As can be seen from Figure 4, around the current transmission changes from around zero to around one. For weak barriers there is a relatively large current transmission below one leading to drain leakage currents. For strong barriers this remnant transmission vanishes and we can approximate the current transmission by an ideal one.
Figure 4. Scaled effective model. (a) Scaled effective potential. (b) Effective current transmission at u = 0.1, v_{D }= 0.5, and v_{G }= 0 ( = 0.504 and m = 0.992). The considered characteristic lengths are l = 4 (red, weak barrier, β = 15.87) and l = 25 (green, strong barrier, β = 619.8). The ideal limit (Equation 19) in blue line.
To a large extent the Fowler Nordheim oscillations in the numerical transmission average out performing the integration in Equation 4.
Parameters in experimental nanoFETs
Heavily doped contacts
In the heavily doped contacts the electrons can be approximated as a threedimensional noninteracting Fermi gas. Then from (8) the Fermi energy above the bottom of the conduction band is given by
For n^{++}doped Si contacts the valleydegeneracy is N_{V }= 6 and the effective mass is taken as . Here m_{1 }= 0.19m_{0 }and m_{2 }= 0.98m_{0 }are the effective masses corresponding to the principle axes of the constant energy ellipsoids. In our later numerical calculations we set ε_{F }= 0.35 eV assuming a level of sourcedoping as high as N_{i }= n_{0 }= 10^{21 }cm^{3}.
Electron channel
In the electron channel a strong lateral subband quantization exists As wellknown [17] at low temperatures only the two constant energy ellipsoids with the heavy mass m_{2 }perpendicular to the (100)interface are occupied leading to a valley degeneracy of g_{v }= 2. The inplane effective mass is therefore the light mass m* = m_{1 }entering the relation
Here ε_{F }= 0.35 eV was assumed. One then has in Equation 3 I_{0 }= ~ 27μA and with λ ~ 1 nm as well as = 2 one obtains J_{0 }= 5.4 × 10^{4 }μA/μm.
Results
Drain characteristics
Typical drain characteristics are plotted in Figure 5 for a low temperature (u = 0.01) and at room temperature (u = 0.1). It is seen that for both the temperatures a LTT can be identified. We define the LTT as the j  v_{D }trace which can be best fitted with a linear regression j = σ^{th}v_{D }in the given interval 0 ≤ v_{D }≤ 2. The best fit is determined by the minimum relative mean square deviation. The gate voltage associated with the LTT is denoted with . It turns out that at room temperature lies slightly above zero and at low temperatures slightly below (see Figure 5c). In general, the temperature dependence of the drain current is small. The most significant temperature effect is the enhancement of the resonant FowlerNordheim oscillations found at negative v_{G }at low temperatures. From Figure 5d, it can be taken that the slope of the LTT σ^{th }decreases with increasing l and increasing temperature. For "hot" transistors (u = 0.2) a LTT can only be defined up to l ~ 10.
Figure 5. Calculated drain characteristics for l = 10, v_{G }starting from 0.5 with decrements of 0.1 (solid lines) at the temperature (a) u = 0.1 and (b) u = 0.01. In green dashed lines the LTT. For u = 0.1 the LTT occurs at a gate voltage of = 0.05 and for u = 0.01 at = 0.05. (c) , and (d) σ^{th }versus characteristic length for u = 0.01 (black), u = 0.1 (red), and u = 0.2 (green).
Threshold characteristics
The threshold characteristics at room temperature are plotted in Figure 6 for a "small" drain voltage (v_{D }= 0.1) and a "large" drain voltage (v_{D }= 2.0). For the largest considered characteristic length l = 60 it is seen that below zero gate voltage the drain current is thermally activated for both considered drain voltages. A comparison with the results for l = 25 and l = 10 yields that for the small drain voltage the I_{D } V_{G }trace is only weakly effected by the change in the barrier strength. In contrast, at the high drain voltage the drain current below v_{G }= 0 grows strongly with decreasing barrier strength. The drain current does not reach the thermal activation regime any more, it falls of much smoother with increasing negative v_{G}. As can be gathered from Figure 8 this effect is seen in experiments as well. We attribute it to the weakening of the tunneling barrier with increasing v_{D}. To confirm this point the threshold characteristics for a still weaker barrier strength (l = 3) is considered. No thermal activation is found in this case even for the small drain voltage.
Figure 6. Calculated threshold characteristics at u = 0.1 (a) for l = 60 and (b) l = 25, and (c) l = 3. The dashed straight lines in blue are guides to the eye exhibiting a slope corresponding to thermal activation.
Discussion
We discuss our numerical results on the background of experimental characteristics for a 10 nm gate length transistor [4,5] reproduced in Figure 7. As demonstrated in Sect. "Parameters in experimental nanoFETs" one obtains from Equation 21 a characteristic length of λ ~ 1 nm under reasonable assumptions. For the experimental 10 nm gate length, we thus obtain l = L/λ = 10. Furthermore, Equation 20 yields the value of ε_{F }= 0.35 eV. The conversion of the experimental drain voltage V into the theoretical parameter v_{D }is given by
Figure 7. Drain characteristics in experiment and theory. (a) Experimental drain characteristics for a nanotransistor with L = 10 nm [4,5]. Our assumption for the LTTis marked with a green dashed line leading to a threshold gate voltage of = 0.15V. (b) Theoretical drain characteristics for l = 10 and u = 0.1 (see Fig. 5a) with the green dashed threshold characteristic at = 0.05.
The maximum experimental drain voltage of 0.75 V then sets the scale for v_{D }ranging from zero to v_{D }= 0.75 eV/0.35 eV ~ 2. For the conversion experimental gate voltage V_{G }to the theoretical parameter v_{G }we make linear ansatz as
where is the experimental threshold gate voltage (see Figure 8a). The constant β is chosen so that converts into . In our example, it is shown from Figure 8a = 0.15 V and from Figure 8b = 0.05, so that β = 0.2 eV. To match the experimental drain characteristic to the theoretical one we first convert the highest experimental value for V_{G }into the corresponding theoretical one. Inserting in (23) V_{G }= 0.75 V yields v_{G }~ 0.5. Second, we adjust the experimental and the theoretical drain currentscales so that in Figure 7 the curves for the experimental current at V_{G }= 0.7 and the theoretical curve at v_{G }= 0.5 agree. It then turns out that the other corresponding experimental and theoretical traces agree as well. This agreement carries over to the range of negative gate voltages with thermally activated transport. This can be gathered from the I_{D } V_{G }traces in Figure 8. We note that the constant of proportionality in Equation 23 given by 1 eV is more then ε_{F }which one would expect from the theoretical definition v_{G }= V_{G}/ε_{F}. Here, we emphasize that the experimental value of e V_{G }corresponds to the change of the potential at the transistor gate while the parameter v_{G }describes the position of the bottom of the lowest twodimensional subband in the electron channel. The linear ansatz in Equation 23 and especially the constant of proportionality 1 eV can thus only be justified in a selfconsistent calculation of the subband levels as has been provided, e.g., by Stern[18].
Figure 8. Threshold characteristics in experiment and theory. (a) Experimental threshold characteristics for the nanotransistor in Fig. 7a. (b) Theoretical threshold characteristics for l = 10 and u = 0.1 with the blue dashed lines corresponding to thermal activation.
The experimental and the theoretical drain characteristics in Figure 7 look structurally very similar. For a quantitative comparison we recall from Sect. "Parameters in experimental nanoFETs" the value of J_{0 }= 5.4 × 10^{4}μA/μm. Then the maximum value j = 0.15 in Figure 7b corresponds to a theoretical current per width of 8 × 10^{3}μA/μm. To compare with the experimental current per width we assume that in the yaxis labels in Figures 7a and 8a it should read μA/μm instead of A/μm. The former unit is the usual one in the literature on comparable nanotransistors (see Refs. [29]) and with this correction the order of magnitude of the drain current per width agrees with that of the comparable transistors. It is found that the theoretical results are larger than the experimental ones by about a factor of ten. Such a failure has to be expected given the simplicity of our model. First, for an improvement it is necessary to proceed from potentials resulting in a selfconsistent calculation. Second, our representation of the transistor by an effectively onedimensional system probably underestimates the backscattering caused by the relatively abrupt transition between contacts and electron channel. Third, the drain current in a real transistor is reduced by impurity interaction, in particular, by inelastic scattering. As a final remark we note that in transistors with a gate length in the micrometer scale shortchannel effects may occur which are structurally similar to the ones discussed in this article (see Sect. 8.4 of [10]). Therefore, a quantitatively more reliable quantum calculation would be desirable allowing to distinguish between the shortchannel effects on micrometer scale and quantum shortchannel effects.
Summary
After a detailed discussion of the physical quantities in our scaleinvariant model we show that a LTT is present not only in the low temperature limit but also at room temperatures. In qualitative agreement with the experiments the I_{D } V_{G}traces exhibit below the threshold voltage thermally activated transport at small drain voltages. At large drain voltages the gatevoltage dependence of the traces is much weaker. It is found that the theoretical drain current is larger than the experimental one by a little less than a decade. Such a finding is expected for our simple model.
Abbreviation
LTT: linear threshold trace.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
UW worked out the theroretical model, carried out numerical calculations and drafted the manuscript. MK carried out numerical calculations and drafted the manuscript. HR drafted the manuscript. All authors read and approved the final manuscript.
References

Auth C, Buehler H, Cappellani A, Choi Hh, Ding G, Han W, Joshi S, McIntyre B, Prince M, Ranade P, Sandford J, Thomas C: 45 nm Highk+Metal Gate StrainEnhanced Transistors.

Yu B, Wang H, Joshi A, Xiang Q, Ibok E, Lin MR: 15 nm Gate Length Planar CMOS Transistor.

Doris B, Ieong M, Kanarsky T, Zhang Y, Roy RA, Dokumaci O, Ren Z, Jamin FF, Shi L, Natzle W, Huang HJ, Mezzapelle J, Mocuta A, Womack S, Gribelyuk M, Jones EC, Miller RJ, Wong HSP, Haensch W: Extreme Scaling with UltraThin Si Channel MOSFETs.

Doyle B, Arghavani R, Barlage D, Datta S, Doczy M, Kavalieros J, Murthy A, Chau R: Transistor Elements for 30 nm Physical Gate Lengths.

Chau R, Doyle B, Doczy M, Datta S, Hareland S, Jin B, Kavalieros J, Metz M: Silicon NanoTransistors and Breaking the 10 nm Physical Gate Length Barrier.
61st Device Research Conference 2003; Salt Lake City, Utah (invited talk)

Tyagi S, Auth C, Bai P, Curello G, Deshpande H, Gannavaram S, Golonzka O, Heussner R, James R, Kenyon C, Lee SH, Lindert N, Miu M, Nagisetty R, Natarajan S, Parker C, Sebastian J, Sell B, Sivakumar S, St Amur A, Tone K: An advanced low power, high performance, strained channel 65 nm technology.

Natarajan S, Armstrong M, Bost M, Brain R, Brazier M, Chang CH, Chikarmane V, Childs M, Deshpande H, Dev K, Ding G, Ghani T, Golonzka O, Han W, He J, Heussner R, James R, Jin I, Kenyon C, Klopcic S, Lee SH, Liu M, Lodha S, McFadden B, Murthy A, Neiberg L, Neirynck J, Packan P, Pae S, Parker C, Pelto C, Pipes L, Sebastian J, Seiple J, Sell B, Sivakumar S, Song B, Tone K, Troeger T, Weber C, Yang M, Yeoh A, Zhang K: A 32 nm Logic Technology Featuring 2ndGeneration Highk + MetalGate Transistors, Enhanced Channel Strain and 0.171 μm^{2 }SRAM Cell Size in a 291 Mb Array.

Fukutome H, Hosaka K, Kawamura K, Ohta H, Uchino Y, Akiyama S, Aoyama T: Sub30nm FUSI CMOS Transistors Fabricated by Simple Method Without Additional CMP Process.

Bedell SW, Majumdar A, Ott JA, Arnold J, Fogel K, Koester SJ, Sadana DK: Mobility Scaling in ShortChannel Length Strained GeonInsulator PMOSFETs.

Sze SM: Physics of Semiconductor Devices. New York: Wiley; 1981.

Thompson S, Packan P, Bohr M: MOS Scaling: Transistor Challenges for the 21st Century.

Brennan KF: Introduction to Semiconductor Devices. Cambridge: Cambridge University Press; 2005.

Nemnes GA, Wulf U, Racec PN: Nanotransistors in the LandauerBüttiker formalism.
J Appl Phys 2004, 96:596. Publisher Full Text

Nemnes GA, Wulf U, Racec PN: Nonlinear IV characteristics of nanotransistors in the LandauerBüttiker formalism.
J Appl Phys 2005, 98:84308. Publisher Full Text

Wulf U, Richter H: Scaling in quantum transport in silicon nanotransistors.

Wulf U, Richter H: Scaleinvariant drain current in nanoFETs.

Ando T, Fowler AB, Stern F: Electronic properties of twodimensional systems.
Rev Mod Phys 1982, 54:437. Publisher Full Text

Stern F: SelfConsistent Results for nType Si Inversion Layers.
Phys Rev B 1972, 5:4891. Publisher Full Text