the quantum hydrodynamic smooth potentialdownloads.hindawi.com/journals/vlsi/1998/089074.pdf ·...
TRANSCRIPT
VLSIDESIGN1998, Vol. 6, Nos. (1-4), pp. 17-20Repnnts available directly from the publisherPhotocopying permitted by license only
(C) 1998 OPA (Overseas Publishers Association) N.V.Published by license under
the Gordon and Breach Science
Publishers imprint.Printed in India.
The Quantum Hydrodynamic Smooth Effective PotentialCARL L. GARDNER** and CHRISTIAN RINGHOFER
Department ofMathematics, Arizona State University, Tempe, AZ 85287-1804
An extension of the quantum hydrodynamic (QHD) model is discussed which is valid forclassical potentials with discontinuities. The effective stress tensor for the QHD equationscancels the leading singularity in the classical potential at a barrier and leaves a residualsmooth effective potential with a lower potential height in the barrier region. The smoothingmakes the barrier partially transparent to the particle flow and provides the mechanism forparticle tunneling in the QHD model.
Keywords: quantum hydrodynamic model, smooth effective potential
In his lectures on Statistical Mechanics, Feynmanderives an effective quantum potential by a Gaussian
smoothing of the classical potential. After demon-strating that the effective free energy based on theeffective potential is accurate for smooth classical
potentials like the anharmonic oscillator, he goes onto say that "it fails in its present form when the [clas-sical] potential has a very large derivative as in thecase of hard-sphere interatomic potential" or forpotential barriers in quantum semiconductor devices.In this note, we discuss an extension of Feynman’sideas to a smooth effective potential for the quantumhydrodynamic (QHD) model that is valid for the tech-
nologically important case of potentials with disconti-nuities.The QHD equations have the same form as the
classical hydrodynamic equations:
+ w--(nli) 0 (1)oxi
) )V mnuj(mnuj) + (mnuiuj- Pij) -n (2)
OW 0(uiW ujPij + qi) --nui
v
(3)
where repeated indices are summed over and where n
is the particle density, u is the velocity, rn is the parti-cle mass, Pij is the stress tensor, V is the classicalpotential energy, W is the energy density, q is the heatflux, and TO is the ambient temperature. Collisioneffects are modeled by the relaxation time approxima-tion, with momentum and energy relaxation times xpand xw. Quantum effects enter through the expressionfor the stress tensor (and for the energy densityderived from the stress tensor).
Originally the quantum correction to the stress ten-
sor in the QHD equations was given to O(h2) andinvolved second derivatives of the classical potential.In the spirit of Feynman, Ferry and Zhou derived a
* Research supported in part by the U.S. Army Research Office under grant DAAH04-95-1-0122.+ Corresponding author. E-mail [email protected]. Research supported in part by ARPA under grant F49620-93-1-0062.
18 CARL L. GARDNER and CHRISTIAN RINGHOFER
smoothed quantum potential for the QHD equationsby linearizing an equation for the equilibrium densitymatrix. The Feynman effective partition functioninvolves a smoothed potential of the form
Va2(X) ] dy { (x- y)2 ) V(y) (4)x/2aexp
2a2
where a2 oc fJh2/m and 1/T is the inverse tempera-ture. The Ferry-Zhou effective stress tensor involves
the difference between the smoothed and the localquantum potential-h272n/8mn + V. Their smoothingfunction is of the form exp{-(x-y)Z/2a2}/Ix-yl.Neither of these two approaches which involve justa spatial averaging has enough smoothing to handlediscontinuities in V, since second spatial derivativesof the smoothed potential appear in both Feynman’sfree energy and the QHD equations.
To derive the new effective stress tensor and energydensity, we construct an effective density matrix as an
O(V) solution to the Bloch equation. Then using theeffective density matrix, we take moments of the
quantum Liouville equation to derive the QHD equa-tions with the effective stress tensor and energy den-sity [2].The effective density matrix has the form
p(,X,y)
{ m (x-y)2exp -2--0.04
0.02
0
-0.02
-0.04
-0.06
Smooth------Bloch
where V is given in center-of-mass coordinates
R=. (x+y),s=x-yby
f ( 2m[5 )3/22--- dfJ’ d3X x
{ 2m[ X,2}exp(13 ,) ([3 + [3,)h2
P’s)] (6)
In constructing the effective density matrix, it is the
change 6V (max V rain V}) over a character-istic length scale that is assumed small rather than I3V.The change 6V over a characteristic length scale
may not be small at a potential step. In fact, for a 0.2eV barrier, 6V 8 at 300 K and 32 at 77 K. How-ever, numerical comparisons demonstrate excellentagreement between the first three moments of the
equilibrium full density matrix and the effective den-
sity matrix for the Bloch equation with a barrier
potential for [6V < 4 (see Fig. 1), and good qualita-tive agreement for 138V < 20 (see Fig. 2). Themoments of the O(h2) density matrix are in severe
quantitative and qualitative disagreement with themoments of the full density matrix.
Using the effective density matrix in the moment
expansion of the quantum Liouville equation, we
-150 -i00 -50 0 50 i00 150
X
FIGURE Quantum term in the energy in eV for electrons in a 0.1 eV double barrier in OaAs at T 300 K. The barriers are 25/i and thewell is 50 A wide. x is in A for all figures
THE QUANTUM HYDRODYNAMIC SMOOTH EFFECTIVE POTENTIAL 19
0.2
0 1
0
-0.1
-0.2
-0.3
-0.4
SmoothBloch
- .so - oo -so o so--x
FIGURE 2 Quantum term in the energy in eV for electrons in a 0.5 eV double barrier
obtain the QHD conservation laws as the first threemoments with
h2n 2VPij- -nTij-
4mT oa5:i3xj (7)
3 2 ----]J’2nV2vW -nT + -mnu +
8rnT(8)
where the "quantum potential" is
v(13,x)1 13 2 2m13 3/:2
fo fa3x,)h2) x
exp -(_ ,)(+,)h(X’-x) V(X’). (9)
e quantum coection to the classical stress tensorand energy density is valid to all orders of and to
first order in gV,d involves bo a smooing inte-
gration of the classical potential over space and an
averaging integration over temperature.We define the 1D smoo effective potential in the
momentum conseation equation (2) as the most sin-
gulp of V- P l"
h2 d2VU V 4
4mT dx2"(10)
e double integration (over both space and tempera-ture) provides sufficient smooing so that the Plte in the smooth effective potential actually ccels
the leading singularity in the classical potential at a
barrier (see Fig. 3), leaving a residual smooth effec-tive potential with a lower potential height in the bar-rier region. This cancellation and smoothing makesthe barriers partially transparent to the particle flowand provides the mechanism for particle tunneling inthe QHD model. Note that the effective barrier heightapproaches zero as T 0. This effect explains influid dynamical terms why particle tunneling is
enhanced at low temperatures. As T -+ o, the effec-tive potential approaches the classical double barrier
potential and quantum effects in the QHD model are
suppressed.
10.7 20.50.25 1
0
Log [T/300 K]-i00
io0
FIGURE 3 Smooth effective potential for electrons in GaAs for50/ wide unit potential double barriers and 50/, wide well as afunction ofx and logl0(T/300 K)
20 CARL L. GARDNER and CHRISTIAN RINGHOFER
ReferencesD. K. Ferry and J.-R. Zhou, "Form of the quantum potentialfor use in hydrodynamic equations for semiconductor devicemodeling," Physical Review, vol. B 48, pp. 7944-7950, 1993.
[2] C. L. Gardner and C. Ringhofer, "The smooth quantumpotential for the hydrodynamic model," Physical Review, vol.E 53, pp. 157-167, 1996.
BiographiesCarl L. Gardner is Professor of Mathematics at Ari-zona State University. His current research interestslie in classical and quantum semiconductor devicesimulation and computational fluid dynamics.
Christian Ringhofer is Professor of Mathematicsat Arizona State University. His current researchinterests include classical and quantum transportequations and moment models for semiconductordevice modeling.
International Journal of
AerospaceEngineeringHindawi Publishing Corporationhttp://www.hindawi.com Volume 2010
RoboticsJournal of
Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014
Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014
Active and Passive Electronic Components
Control Scienceand Engineering
Journal of
Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014
International Journal of
RotatingMachinery
Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014
Hindawi Publishing Corporation http://www.hindawi.com
Journal ofEngineeringVolume 2014
Submit your manuscripts athttp://www.hindawi.com
VLSI Design
Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014
Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014
Shock and Vibration
Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014
Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawi Publishing Corporation http://www.hindawi.com
Volume 2014
The Scientific World JournalHindawi Publishing Corporation http://www.hindawi.com Volume 2014
SensorsJournal of
Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014
Modelling & Simulation in EngineeringHindawi Publishing Corporation http://www.hindawi.com Volume 2014
Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014
Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014
Navigation and Observation
International Journal of
Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014
DistributedSensor Networks
International Journal of