1: \documentstyle[aps,prl,multicol,fancyheadings,epsfig,amsbsy,amssymb,amstex]{revtex}
2: \newcommand{\lrule}{ \noindent
3: \rule{0.5\textwidth}{0.1mm}\rule{0.1mm}{3pt}\newline }
4: \newcommand{\rrule}{ \noindent \parbox{\textwidth}{
5: \hfill\rule[-3pt]{0.1mm}{3pt}\rule{0.5\textwidth}{0.1mm}} }
6: \def\bbbone{{\mathchoice {\rm 1\mskip-4mu l} {\rm 1\mskip-4mu l}
7: {\rm 1\mskip-4.5mu l} {\rm 1\mskip-5mu l}}}
8: \def\bbbc{{\mathchoice {\setbox0=\hbox{$\displaystyle\rm C$}\hbox{\hbox
9: to0pt{\kern0.4\wd0\vrule height0.9\ht0\hss}\box0}}
10: {\setbox0=\hbox{$\textstyle\rm C$}\hbox{\hbox
11: to0pt{\kern0.4\wd0\vrule height0.9\ht0\hss}\box0}}
12: {\setbox0=\hbox{$\scriptstyle\rm C$}\hbox{\hbox
13: to0pt{\kern0.4\wd0\vrule height0.9\ht0\hss}\box0}}
14: {\setbox0=\hbox{$\scriptscriptstyle\rm C$}\hbox{\hbox
15: to0pt{\kern0.4\wd0\vrule height0.9\ht0\hss}\box0}}}}
16: \pagestyle{fancy} \pagestyle{fancyplain} \footrulewidth 0.4pt
17: \plainheadrulewidth 0.4pt \plainfootrulewidth 0.4pt
18: \lhead{\large LA-UR-01-XXXX}
19: \chead{ \today}
20: \rhead{\sl }
21: \cfoot{\sc\thepage}
22: \lfoot{}
23: \rfoot{}
24:
25: \textheight9.75in %Simulate the actual PRB textheight.
26: \topmargin-25mm %Make it fit on a letter size sheet.
27:
28:
29: \newcommand{\beq}{\begin{eqnarray}}
30: \newcommand{\eeq}{\end{eqnarray}}
31: \newcommand{\s}{\sigma}
32: \newcommand{\bs}{\bf \sigma}
33: \newcommand{\bj}{{\bf j}}
34: \newcommand{\bE}{{\bf E}}
35: \newcommand{\bH}{{\bf H}}
36: \newcommand{\bF}{{\bf F}}
37: \newcommand{\br}{{\bf r}}
38: \newcommand{\on}{{\overline{n}}}
39:
40:
41: \begin{document}
42:
43: \title{Spin-drift transport and its applications}
44: \author{Ivar Martin}
45: \address{Theoretical Division, Los Alamos National Laboratory,
46: Los Alamos, NM 87545}
47:
48: \date{January 25, 2002}
49:
50: \maketitle
51:
52: \begin{abstract}
53:
54: We study the generation of non-equilibrium spin currents in
55: systems with spatially-inhomogeneous magnetic potentials. For
56: sufficiently high current densities, the spin polarization can be
57: transported over distances significantly exceeding the intrinsic
58: spin-diffusion length. This enables applications that are
59: impossible within the conventional spin-diffusion regime.
60: Specifically, we propose dc measurement schemes for the carrier
61: spin relaxation times, $T_1$ and $T_2$, as well as demonstrate the
62: possibility of spin species separation by driving current through
63: a region with an inhomogeneous magnetic potential.
64:
65: \end{abstract}
66: \pacs{PACS Numbers: XXXXX}
67: \begin{multicols}{2}
68:
69:
70: The rapidly developing field of spin-sensitive electronics
71: promises to reinvigorate conventional semiconductor
72: electronics\cite{general}. Utilizing the carriers' spin degrees of
73: freedom in addition to their charge opens possibilities for
74: conceptually new {\em spintronic} devices. Already implemented
75: are GMR read-heads and non-volatile magnetic RAM (MRAM); there are
76: also many proposals for other devices \cite{zutic}. Much of the
77: promise of spintronic devices lies in the enhanced flexibility of
78: the spin degrees of freedom compared to electrical charge. While
79: creating local charge imbalance generally involves large
80: electrostatic Coulomb energies, non-equilibrium spin polarizations
81: can be created at a very low cost. As a consequence, spin
82: polarizations are inherently susceptible to manipulation, e.g. by
83: external magnetic fields or by current drive. Indeed, it has been
84: observed experimentally that non-equilibrium spin polarization can
85: be transported over distances exceeding 100 micrometers
86: \cite{kik}, with only moderate diffusive spread. This would be
87: impossible in the case of the non-equilibrium charge
88: inhomogeneity, since the Coulomb repulsion would blow such
89: inhomogeneity apart almost instantly.
90:
91: The immunity of the spin polarization with respect to electrical
92: screening makes it possible to satisfy {\em drift} transport
93: conditions, for which the local spin polarization can be carried
94: downstream by an electrical current over distances proportional to
95: the drift velocity. This situation can be obtained for
96: sufficiently large current densities, when the {\em spin-drift}
97: length becomes larger than the {\em spin-diffusion} length (both
98: defined below). In this Letter we explore this new regime, readily
99: achievable in spin transport, but almost never realized in
100: electron-hole transport. It is studied based on general
101: formulation for spin transport using spin drift-diffusion
102: equations.
103:
104: The spin-drift regime is generally favorable for spintronic
105: purposes as it can provide larger spin-injection depths compared
106: to the spin-diffusive regime. Here we propose two specific
107: applications of the spin-drift regime: (1) dc transport-based
108: measurement of longitudinal ($T_1$) and transverse ($T_2$) spin
109: relaxation times; (2) spin species separation by inhomogeneous
110: magnetic fields. To determine the spin relaxation times, we
111: propose to measure the spin polarization decay profile in the
112: presence of constant current; the decay length can then be related
113: either to $T_1$ or $T_2$, depending on the experimental setup.
114: Unlike conventional pump-probe techniques, there is no need for
115: ultrafast time-resolved detection, and the spin polarization
116: profile can be measured either by optical\cite{kik} or some other
117: techniques. To spatially separate spin-up from spin-down carriers,
118: we propose to drive current in a 2D geometry containing a region
119: of inhomogeneous magnetic field. This technique can be used as a
120: flexible alternative to the standard spin injection techniques.
121:
122: To study spin transport we use the drift-diffusion approximation
123: for spin currents in the presence of an inhomogeneous
124: spin-dependent potential $V_\s(\br) = - g \mu_B {\bs} \cdot \bf
125: H$. At this point we assume that in the plane of the electron
126: gas, the applied magnetic field $\bH$ can vary in value but not in
127: orientation. Then all carriers in the sample can be divided into
128: spin-up ($\sigma = 1/2$) and spin-down ($\sigma = -1/2$) species
129: that feel magnetic potentials equal in magnitude but opposite in
130: sign, $V_\uparrow(\br) = - V_\downarrow(\br)$. For the external
131: potential smaller than the thermal energy in the non-degenerate
132: case, $V_\s(\br) < k_B T$, or less than the Fermi energy in the
133: degenerate case, $V_\s(\br) < E_F$, the local carrier density will
134: remain essentially unchanged, both with and without current drive.
135: Hence, in what follows we neglect electrical screening effects
136: since they only modify the results in the order $(V_\s/max(k_B
137: T,E_F))^2$. In the drift-diffusion approximation, the
138: spin-polarized current is
139: \beq\label{eq:j_s}
140: \bj_\s = n_\s \mu (q \bE - \nabla V_\s) - q D \nabla n_\s,
141: \eeq
142: where $q = \pm e$ is the carrier charge, $n_\s$ the number density
143: of the spin species $\s$, $D$ the diffusion constant, $\mu$ the
144: mobility, $\bE$ the external electric field, and $\nabla$ denotes
145: the gradient operator. We assumed that the transport is unipolar
146: (e.g. n-doped semiconductor), and the diffusion constant is the
147: same for both spin species\cite{flatte}. We do not include the
148: Lorentz force, which also acts on the moving electrons, since it
149: affects both spin species identically and, moreover, is
150: compensated through the classical Hall effect for a confined
151: sample geometry. In practice, local variation of the magnetic
152: potential can be achieved either through the use of micromagnets
153: that can create strongly inhomogeneous external magnetic fields,
154: or by magnetically doping semiconductors, which can produce strong
155: position dependence in the carrier $g$-factor\cite{ohno}.
156:
157: The dynamically generated non-equilibrium spin polarization
158: relaxes towards its equilibrium value. Under constant drive, the
159: relaxing spins are replenished by the divergence of the spin
160: current. In the relaxation-time approximation,
161: \beq\label{eq:cont}
162: \nabla\cdot\bj_s &=& - \delta_\s/T_1,\\
163: \delta_\s &=& n_\s - \overline{n}_\s,\\
164: \overline{n}_\s &=& n_0 \exp(-V_\s/k_BT)/2.
165: \eeq
166: The expression for the equilibrium density, $\overline{n}_\s$,
167: corresponds to non-degenerate carriers. For degenerate carriers,
168: the Boltzman distribution should be replaced by the Fermi-Dirac
169: distribution. In the non-stationary case, there is an extra time
170: derivative, $\partial\delta_\s/\partial t$, on the left hand side
171: of the continuity equation. Combining Eqs.~(\ref{eq:j_s}) and
172: (\ref{eq:cont}), the dc equation for spin density under uniform
173: drive ($\nabla\cdot \bE = 0$) is
174: \beq\label{eq:main}
175: \mu \bE \cdot \nabla(\on_\s + \delta_\s) - \frac{\mu}{q} \nabla
176: (\delta_\s \nabla V_\s) - D \nabla^2 \delta_\s = - \frac{\delta_\s
177: }{T_1}
178: \eeq
179: The first term on the left hand side represents the drag of the
180: spin density by the electric field ($\mu \bE = v_d$ -- the drift
181: velocity), the second term, as we will see, is usually
182: unimportant, and the last term represents diffusion. Similar
183: descriptions have been previously considered by several authors
184: \cite{darryl}, however, with the focus on the {\em diffusive}
185: regime (neglecting $\bE \delta_\s$ terms). The drift terms have
186: only been invoked in the spatially localized regions, such as
187: depletion layers and Schottky barriers\cite{darryl2}. The
188: primary goal of the present work is to explore the {\em
189: spin-drift} regime that occurs for larger drives, that is when the
190: spin-drift length, $L_E = \mu E T_1$, exceeds the spin diffusion
191: length, $L_D = \sqrt{D T_1}$.
192:
193:
194: We first consider the one dimensional case with spin-polarized
195: negative half-space. The static equilibrium density is
196: \beq
197: \on_\s - n_0/2 = \s S\Theta(-x),
198: \eeq
199: where $\Theta$ is the step function equal to unity for positive
200: arguments and zero otherwise; $S$ is the static spin density in
201: the polarized region. The step in the density profile implies a
202: step in the magnetic potential, $V_\s$. The density modification
203: in the presence of a uniform current drive is described by the 1D
204: version of Eq.~(\ref{eq:main}). The strongest singularity is
205: caused by the jump in $V_\s$ at $x = 0$ (second term in
206: Eq.~(\ref{eq:main})), which implies discontinuity in the density,
207: $\delta_\s$,
208: $$\frac{\delta_\s(+0) - \delta_\s(-0)}{\delta_\s(+0) + \delta_\s(-0)} =
209: \frac{\mu}{2qD}(V_\s(-0) - V_\s(+0)).$$
210:
211: Within our assumptions about the magnitude of the magnetic
212: potential, $V_\s < max(k_BT, E_F)$, this correction is small and
213: hence can be neglected. Also, it causes charge imbalance, and
214: hence is screened out for distances exceeding the Coulomb
215: screening length\cite{stu}. The solution of the homogeneous
216: equation for $\delta_\s$ is
217: \beq
218: \delta_\s(x) &=& A\exp(\lambda_1 x) + B\exp(\lambda_2 x), \\
219: \lambda_{1(2)} &=& \frac{L_E}{2L_D^2}\left( 1 + (-) \sqrt{1 +
220: 4L_D^2/L_E^2}\right).
221: \eeq
222: Matching the ($x > 0$) and ($x < 0$) solutions at $x = 0$ and
223: requiring vanishing $\delta_\s$ at infinity, the spin density,
224: $s=\delta_\uparrow - \delta_\downarrow$, for the positive drift
225: velocity ($qE > 0$) is
226: \beq\label{eq:1D_sol}
227: s = \frac{S}{\sqrt{1 + 4L_D^2/L_E^2}}
228: \times
229: \left\{\begin{array}{cc}
230: \exp(\lambda_1 x), &x < 0, \\
231: \exp(\lambda_2 x), &x > 0\end{array}\right. .
232: \eeq
233: The solution for the negative drift velocity can be obtained from
234: this solution as $ -s (-x)$.
235:
236: Two interesting special cases correspond to the {\em spin-drift}
237: ($L_E \gg L_D$) and the {\em spin-diffusion} ($L_E < L_D$)
238: regimes. In the spin-drift case,
239: \beq
240: s = \left\{\begin{array}{lc}
241: {S}\exp(-x/L_E), &x> 0, \\
242: 0, &x < 0\end{array}\right..
243: \eeq
244: And in the spin-diffusion case,
245: \beq
246: s = \frac{S L_E}{2 L_D}\exp(-|x|/L_D).
247: \eeq
248:
249: \begin{figure}[htbp]
250: \begin{center}
251: \includegraphics[width = 3.0 in]{Fig_1D.eps}
252: \vspace{0.5cm} \caption{Spin polarization drag induced by electric
253: current in 1D geometry, Eq.~(\ref{eq:1D_sol}). The dashed line
254: corresponds to the spin polarization profile in the absence of
255: current. The indicated strength of the electric field, $E$, is in
256: V/cm. Note that the stronger the applied electric field is, the
257: farther away the spin polarization can be dragged. The parameters
258: are characteristic of $10^{16}$~cm$^{-3}$ Si-doped GaAs: $\mu =
259: 3000$~cm$^2$/Vs, $T_1 = 100$~ns, $T = 1.6$~K. Carriers are
260: assumed to be non-degenerate.} \label{fig:1D}
261: \end{center}
262: \end{figure}
263:
264: The series of solutions for various strengths of the electric
265: field are shown in Figure~\ref{fig:1D}. The parameters are
266: characteristic of $10^{16}$~cm$^{-3}$ Si-doped GaAs\cite{kik}:
267: $\mu = 3000$~cm$^2$/Vs, $T_1 = 100$~ns, $T = 1.6$~K. The drift
268: condition for this system is realized for $E > 2\sqrt{k_BT/q\mu
269: T_1}
270: \sim 2$~V/cm. Experimentally, much larger electric fields
271: have been applied to this system\cite{kik}.
272:
273:
274:
275: Under spin-drift conditions, the characteristic length of the
276: exponential decay of spin polarization is proportional to the
277: longitudinal spin relaxation time, $L_E = v_d T_1$. In this
278: regime, there is almost one-to-one correspondence between the
279: distance from the interface and the time spent by electrons in the
280: region with zero field, $x = v_d t$, which is only weakly altered
281: by diffusion. Therefore, by measuring the decay length and knowing
282: the drift velocity, $v_d = \mu E$, one can directly determine
283: $T_1$. An advantage compared to the conventional pump-probe
284: approach\cite{kik} is that the measurement can be conducted in a
285: dc setup that doesn't require ultrafast optics and electronics.
286: The spin polarization detection with sub-micron spatial resolution
287: can be done by standard optical techniques based on Kerr/Faraday
288: rotation, or possibly using magnetic resonance force microscopy
289: (MRFM)\cite{mrfm} by modulating the current at the cantilever
290: frequency\cite{roukes}.
291:
292: In addition to the longitudinal relaxation time, $T_1$, which
293: defines how fast the longitudinal spin polarization disappears, an
294: almost identical approach can be applied to determine the
295: transverse spin relaxation time, $T_2$, which measures the
296: electron spin decoherence time. For that, in the setup described
297: above, a weak uniform magnetic field, $H_\perp$, should be applied
298: {\em perpendicular} to the direction of the strong field that
299: exists at $x < 0$. In the region with the strong field, this
300: additional component will only weakly alter the the direction of
301: the total field. However, in the region where originally there was
302: no magnetic field, this component will cause the precession of the
303: drifting spin polarization. The spatial period of the precession
304: \beq
305: L_P = 2\pi v_d/\omega_L,
306: \eeq
307: is related to the Larmor frequency, $\omega_L = g\mu_B
308: H_\perp/\hbar$, which is about 62 MHz at $H_\perp = 100$ G for $g
309: = -0.44 $. For the driving electric field $E = 10$ V/cm and the
310: above parameters, $L_P \sim 5\ \mu$m. Such spatial modulation can
311: be resolved both by optical means or with MRFM. To determine the
312: transverse spin relaxation time $T_2$ one needs to measure spatial
313: dependence of the spin polarization component perpendicular to
314: $H_\perp$ and then fit it to
315: \beq\label{eq:T_2}
316: S \propto \exp(-x/v_d T_2)\cos(2\pi x/L_P).
317: \eeq
318: Here we neglected the diffusive spread of the spin polarization,
319: which will case an analogue of NMR inhomogeneous broadening. The
320: diffusive spread will matter, however, only when it becomes
321: comparable to the precession length, $L_P$. This effectively
322: places an upper limit on the distance from the interface for
323: which the precession-induced oscillation can be resolved, $L <
324: |\mu E L_P^2/D|\sim 250\ \mu$m for the above parameters. The
325: upper limit can be modified by the choice of the transverse
326: magnetic field $B_\perp$ with the only constraint that $\omega_L
327: T_2 > 1$. Finally, the carrier $g$-factor itself can be determined
328: from the Larmor frequency, which can be extracted from the
329: precession length.
330:
331: \begin{figure}[htbp]
332: \begin{center}
333: \includegraphics[width = 3.0 in]{Fig_2D.eps}
334: \vspace{0.5cm} \caption{Spin species separation by inhomogeneous
335: magnetic field. An electric current flows in 2D (film) through a
336: region of localized magnetic field, $V_\s(r) = - 0.6 \s
337: \exp[-(r/5)^{10}]$, with $r$ in micrometers and energy $V_\s$ in
338: Kelvin. The spin polarization is measured in the units of the
339: total density. The strength of the electric field, $E$, is in
340: V/cm. Notice the region of {\it negative} (``minority'') spin
341: polarization that first forms were the current impinges on the
342: region of the magnetic field. For large enough drift velocities,
343: the ``majority'' and the ``minority'' carriers forms a three-tail
344: structure that extends over the drift length, $L_E$, downstream.
345: The material parameters are the same as in Fig.~\ref{fig:1D}.}
346: \label{fig:2D}
347: \end{center}
348: \end{figure}
349:
350: Another application that is only possible in the spin-drift regime
351: is the {\em spatial} separation of spin species by driving
352: electric current through a region with an inhomogeneous magnetic
353: field. To demonstrate this effect, we consider a two-dimensional
354: system with a compact region containing magnetic field. Unlike
355: the one-dimensional case considered above, the ``minority''
356: (opposite to magnetic field) spin carriers can avoid this region
357: by flowing around it. Indeed, this is what we observe by solving
358: numerically the 2D spin-relaxation drift-diffusion equations,
359: Fig.~\ref{fig:2D}. The calculation is performed on a $100 \times
360: 150$ lattice that spans $30\times45\ \mu$m region for parameters
361: relevant to Si-doped GaAs (specified above). The field is
362: localized in 10 $\mu$m disk. For large enough electric field,
363: there is formation of the negative spin polarization where the
364: current impinges on the boundary of the magnetic field
365: region\cite{dassarmaPN}. The minority spin polarization is then
366: dragged downstream by the current until it relaxes and diffuses
367: away. Notice that in the zero-current state there is $no$
368: negative spin polarization in the system. The ``majority'' spin
369: polarization is ripped off the local field region, and together
370: with the ``minority'' carriers forms a three-tail structure that
371: extends over the spin-drift length, $L_E$, downstream. Therefore,
372: driving current through this system leads to an effective
373: separation of the two spin species. A similar effect can be
374: expected also in the ballistic regime realized in the ultra-high
375: mobility 2DEG structures.
376:
377: The effect that we discussed above is analogous to the classic
378: Stern-Gerlach effect; however, unlike the original effect the spin
379: separation can be achieved even with electrons. In the original
380: Stern-Gerlach experiment, {\em neutral} spin-1/2 particles
381: passing through a region with a magnetic field gradient get
382: deflected by the force $F = g \mu_B (\s\cdot\nabla)\bH$, and
383: produce two spots on the screen that represent two possible
384: projections of spin $\s$. For charged particles, the situation is
385: complicated due to the Lorentz force that acts to scramble the
386: separation. It was first pointed out by Bohr that in the case of
387: electrons, the observation of Stern-Gerlach effect is impossible
388: because the deflection by the Lorentz force exceeds the
389: spin-related splitting\cite{SGrec}. The Bohr argument does not
390: apply, however, to constrained geometries, e.g. a thin film
391: sample, where the Lorentz force is compensated by the Hall
392: voltage. This makes the {\em solid-state Stern-Gerlach} effect
393: possible. It is important to stress, however, that despite the
394: appearance, the mechanism for spin separation here is distinctly
395: different from the the classic Stern-Gerlach effect. A {\em
396: time-dependent} Stern-Gerlach effect was recently considered by
397: Fabian and Das Sarma\cite{dassarmaSG}.
398:
399: In the above discussion we assumed a possibility of a magnetic
400: field profile that is inhomogeneous but always pointing along the
401: same line, e.g. perpendicular to the sample plane. This is an
402: important assumption because the fringe fields have ability to
403: destroy spin polarization through precession. For the $10^{16}$
404: Si:GaAs, a 100 Gauss fringe field at 10 V/cm drive could scramble
405: the polarization after 5 $\mu$m. However, since the fringe fields
406: decay rapidly in space, their detrimental effects can be reduced
407: by increasing the drive or by reducing the size of the
408: micromagnets. Moreover, for thin-film samples, the effects of the
409: fringe fields can be essentially eliminated by placing the sample
410: between two closely spaced magnet terminals (split ring) or
411: between two ``identical'' magnets. Alternatively, one could
412: selectively boost the $g$-factor, e.g. by locally doping II-VI
413: semiconductors with Mn. This way, when placed even in a relatively
414: weak uniform external magnetic field, different parts of the
415: sample will feel very different magnetic potentials. By combining
416: these approaches, one can engineer a variety of strong magnetic
417: potentials which are not affected by the fringe fields.
418:
419:
420: In summary, we explored the spin-drift regime, which is realized
421: for sufficiently large but realistic current densities in
422: semiconductors. In this regime, the spin density can be carried
423: by the current over distances controlled by the spin-drift length,
424: which can significantly exceed the spin-diffusion length. This
425: opens a possibility of designing spintronic devices with the spin
426: properties dynamically controllable by the applied bias. This is
427: in sharp contrast with the conventional particle-hole
428: semiconductor electronics where electrostatic screening
429: practically precludes large scale dynamic manipulation of the
430: relative particle-hole density. As specific applications, we
431: proposed measurement schemes for the carrier spin relaxation
432: times, $T_1$ and $T_2$, as well as demonstrated the possibility of
433: spin species separation by driving current through a region with
434: an inhomogeneous magnetic potential. The latter is analogous to
435: the classic Stern-Gerlach effect.
436:
437: Upon completion of this paper, author became aware of the related
438: one-dimensional study of the spin-drift regime by Yu and
439: Flatt\'{e}\cite{yu}.
440:
441: We thank S.A. Crooker, J. Albrecht, D.L. Smith, J. Fabian, and
442: S.A. Trugman for useful discussions. This work was supported by
443: DARPA SPINs program.
444:
445:
446: \begin{thebibliography}{99}
447: \bibitem{general} G. Prinz, Science {\bf 282}, 1660 (1998); S.A. Wolf {\em et
448: al.}, Science {\bf 294}, 1488 (2001).
449:
450: \bibitem{zutic} For an extensive list of references, see I. \v{Z}uti\'{c}, preprint
451: cond-mat/0112368 (2001).
452:
453: \bibitem{kik}
454: J. Kikkawa and D.D. Awschalom, Nature (London){\bf 397}, 139
455: (1999).
456:
457: \bibitem{flatte} For detailed analysis of spin diffusion in doped and intrinsic
458: semiconductors, see M.E. Flatte and J.M. Buyers, Phys. Rev. Lett.
459: {\bf 84}, 4220 (2000).
460:
461: \bibitem{ohno} Y. Ohno {\em et al.}, Nature (London) {\bf 402}, 790 (1999); M.
462: Oestreich {\em et al.}, Appl. Phys. Lett. {\bf 74}, 1251 (1999);
463: R. Freiderling {\em et al.}, Nature (London) {\bf 402}, 787
464: (1999); B.T. Jonker {\em et al.}, Phys. Rev. B {\bf 62}, 8180
465: (2000).
466:
467: \bibitem{darryl} Mark Johnson and R. H. Silsbee, Phys. Rev. B 35, 4959
468: (1987); D.L. Smith and R.N. Silver, Phys. Rev. B {\bf 64}, 5323
469: (2001).
470:
471: \bibitem{darryl2} \v{Z}uti\'{c}, J. Fabian, and S. Das Sarma, preprint
472: cond-mat/0106085 (2001); J. Albrecht and D.L. Smith, unpublished.
473:
474: \bibitem{stu} H.U. Baranger and J.W. Wilkins, Phys. Rev. B {\bf
475: 30}, 7349 (1984); S.A. Trugman and A.J. Taylor, Phys. Rev. B {\bf
476: 33}, 5575 (1986).
477:
478: \bibitem{mrfm}J. A. Sidles , Phys. Rev. Lett. {\bf 68}, 1124 (1992); D. Rugar
479: {\em et al.}, Science {\em 264}, 1560 (1994); Z. Zhang, P. C.
480: Hammel, and P. E. Wigen, Appl. Phys. Lett. {\bf 68}, 2005 (1996);
481:
482: \bibitem{roukes} M. Roukes and P.C. Hammel, private communication.
483:
484: \bibitem{dassarmaPN} In the 1D geometry, the dynamical generation
485: of the ``negative'' polarization has been observed by
486: \v{Z}uti\'{c}, J. Fabian, and S. Das Sarma, preprint
487: cond-mat/0106085 (2001).
488:
489: \bibitem{SGrec} For a recent discussion of the Bohr-Pauli-Mott argument and
490: possible exceptions, see B. M. Garraway and S. Stenholm, Phys.
491: Rev. A {\bf 60}, 63 (1999).
492:
493: \bibitem{dassarmaSG} J. Fabian and S. Das Sarma, preprint cond-mat/0104146
494: (2001).
495:
496: \bibitem{yu} Z. G. Yu and M. E. Flatt\'{e}, preprint
497: cond-mat/0201425 (2002).
498:
499: \end{thebibliography}
500: \end{multicols}
501: \end{document}
502: