1: \documentclass[prl,twocolumn]{revtex4}

2: \usepackage{array,amssymb,graphicx}

3: 

4: %

5: %       Definition of wide equation command

6: %

7: % Usage \wideeqn{arg}{Text}

8: %     arg = t  - used if the wide equation is at the top of the page.

9: %     arg = b  - used if the wide equation is at the bottom of the page.

10: %     arg = m  - otherwise.

11: %

12: \def\wideeqn#1#2{

13: \begin{widetext}

14: \noindent

15: \if#1t

16: \else

17:     \raisebox{9pt}[0in][0.0in]

18:     {$\rule{3.4in}{0.4pt}\rule{0.4pt}{6pt}$\hspace{3.6in}}

19: \fi

20: #2

21: \if#1b

22: \else

23:     \raisebox{-9pt}[0in][0.0in]

24:     {\hspace{3.55in}$\rule{0.4pt}{6pt}\rule[6pt]{3.5in}{0.4pt}$}

25: \fi

26: \end{widetext}

27: \noindent

28: }

29: 

30: \begin{document}

31: 

32: \title{Effective theory of incompressible quantum Hall liquid crystals}

33: 

34: \author{Michael M. Fogler}

35: 

36: \affiliation{Department of Physics, Massachusetts Institute of

37: Technology, 77 Massachusetts Avenue, Cambridge, Massachusetts 02139}

38: 

39: %\date{\today}

40: 

41: \begin{abstract}

42: 

43: I propose an effective theory of zero-temperature phases of the quantum

44: Hall stipes: a smectic phase where the stripes are static and a novel

45: quantum nematic phase where the positional order is destroyed by

46: quantum fluctuations. The nematic is viewed as a Bose condensate

47: of dislocations whose interactions are mediated by a $U(1)$ gauge field.

48: Collective mode spectrum and the dynamical structure factor in the two

49: phases are calculated.

50: 

51: \end{abstract}

52: \pacs{PACS numbers: 73.40.Hm}

53: 

54: \maketitle

55: 

56: Stripe phases are extremely common in nature. Once studied mainly in the

57: context of pattern formation and soft condensed-matter systems

58: (convection rolls, ferrofluids, diblock co-polymers, {\it

59: etc.\/}~\cite{Seul_95}), they are now recognized to be important in

60: venues ranging from neutron stars~\cite{Neutron_star} to neural networks

61: of the brain~\cite{Chklovskii}. The ``hard'' condensed-matter community

62: was alerted to the relevance of stripes after their discovery in

63: transitional metal oxides, especially, high-$T_c$

64: cuprates~\cite{High_Tc_stripes}. The subject of this Letter is the

65: stripe phase in another remarkable correlated system: a two-dimensional

66: (2D) electron liquid in a transverse magnetic field. This particular

67: phase is formed when the Landau level occupation factor $\nu$ is close

68: to a half-integer and is larger than some critical value

69: $\nu_c$~\cite{Fogler_96}. Unlike the still debated case of high-$T_c$,

70: the theory of the quantum Hall stripes rests on a much more solid

71: foundation~\cite{Fogler_96,Moessner_96}. The quantity $1 / \nu$ plays

72: the role of a small parameter controlling the strength of quantum

73: fluctuations beyond mean-field. Thus, the mean-field stripe solution is

74: stable and adequately captures the main properties of the ground state

75: at large $\nu$. Numerical calculations~\cite{Critical_nu,Scarola_01}

76: indicate that $\nu_c \sim 4$ for the physically relevant case of Coulomb

77: interaction. This provides a natural explanation of the magnetotransport

78: anisotropy observed in GaAs heterostructures~\cite{Experiments} at $\nu

79: \geq \frac92$.

80: 

81: The states at $\nu \sim \nu_c$ are much less understood because the

82: fluctuations about the mean-field Hartree-Fock solution are large and

83: several almost degenerate states are in competition. Small changes in

84: microscopic parameters may therefore radically change the

85: nature of the ground state and bring to life novel quantum phases.

86: 

87: %

88: % FIG. 1

89: %

90: \begin{figure}

91: \includegraphics[width=2.1in,bb=118 308 557 518]{qn_1.ps}

92: %\vspace{0.1in}

93: \setlength{\columnwidth}{3.2in}

94: \caption{

95: (a) Smectic (b) Nematic. One dislocation is encircled.

96: \label{Fig_real_space}

97: }

98: \end{figure}

99: %

100: 

101: Particularly intriguing are proposed quantum {\it smectic\/} and quantum

102: {\it nematic\/} phases~\cite{Balents_96,Fradkin_99}. In the smectic

103: phase the system lacks translational symmetry (is periodically

104: modulated) along one of the spatial directions, say $\hat{\bf x}$, but

105: remains liquid-like along $\hat{\bf y}$. An example of the smectic phase

106: is a unidirectional charge-density-wave Hartree-Fock

107: state~\cite{Fogler_96}; however, the true quantum smectic must possess

108: certain amount of fluctuations beyond Hartree-Fock (shown pictorially in

109: Fig.~\ref{Fig_real_space}a) to prevent another periodic modulation along

110: $\hat{\bf

111: y}$~\cite{Fradkin_99,Yi_01,MacDonald_00,Fogler_in_preparation}. In

112: contrast, the nematic state is fully translationally invariant but lacks

113: the rotational invariance. It can be visualized as fluctuating stripes

114: preferentially aligned along the $\hat{\bf y}$-direction, see

115: Fig.~\ref{Fig_real_space}b. At present, concrete trial wavefunctions for

116: the quantum nematic are known only for specific filling

117: fractions~\cite{Musaelian_96}. This supports the idea that this phase is

118: physically realizable but forces us to seek approaches other than

119: ``wavefunction engineering'' to gain understanding. Below I construct

120: an effective long-wavelength low-frequency theory that applies to a

121: class of quantum Hall liquid crystals. It describes zero-temperature

122: properties and the quantum phase transition between these intriguing

123: states~\cite{Comment_on_1st_order}.

124: 

125: {\it Smectic phase\/}.--- The smectic emerges from a uniform state when

126: a pair of collective modes with wavevectors ${\bf q} = \pm q_0 \hat{\bf

127: x}$ goes soft and condenses~\cite{Comment_on_soft_modes}. Hydrodynamic

128: fluctuations of the electron density projected onto the topmost Landau

129: level become of the form $n({\bf r}, t) + {\rm Re}\,\{\Psi({\bf r}, t)

130: e^{-i q_0 x}\}$, where $n$ is a smooth density component (with $q \ll

131: q_0$), $\Psi = \Psi_0 e^{i q_0 u}$ is the smectic order parameter, and

132: $u$ is the deviation of the stripes (density maxima) from uniformity,

133: see Fig.~\ref{Fig_real_space}a. To build an effective theory I make a

134: crucial assumption that the Goldstone mode associated with $u$ is {\it

135: the only\/} low-energy degree of freedom of the system. In general,

136: other low-energy excitations, e.g., those inherited from the parent

137: uniform state may be present. Thus, our theory describes only a certain

138: class of possible smectic states, those that are superstructures on top

139: of incompressible quantum Hall liquids~\cite{Comment_on_compressible}.

140: I proceed by writing down the effective Hamiltonian for $u$ and $n$,

141: containing only the most relevant terms consistent with the

142: symmetry~\cite{DeGennes_book,Fogler_00}:

143: %

144: \begin{equation}

145: H = \frac{Y}{2} (\partial_x u)^2

146:   + \frac{K}{2} (\partial_y^2 u)^2

147:   + \frac12 \delta n {\bf U} \delta n + C \delta n \partial_x u.

148: \label{H}

149: \end{equation}

150: %

151: Here $Y$ and $K$ are phenomenological compression and bending moduli,

152: $\delta n = n - n_0$ is the deviation of local density from the

153: equilibrium value $n_0$, and ${\bf U}$ is the integral operator with

154: kernel $U({\bf r})$, the electron-electron interaction potential. $U$ is

155: Coulombic at large $r$ but is modified by many-body screening, exchange

156: and correlation effects at short distances (see more details

157: in~\cite{Fogler_96}). The last term in Eq.~(\ref{H}) accounts for the

158: dependence of the smectic period on $n_0$, with $C = Y \partial \ln q_0

159: / \partial n_0$. It vanishes at the half-filling due to electron-hole

160: symmetry but is nonzero otherwise.

161: 

162: One may wonder why include ``incompressible'' background $n$ in our

163: low-energy Hamiltonian~(\ref{H}). The reason is the important role $n$

164: plays in the dynamics of $u$. Indeed, $u$ determines the density

165: fluctuations near the star of the soft mode, e.g., $n_{{\bf q}_0 + {\bf

166: k}} = \frac12 (\Psi_0 e^{-i q_0 u})_{\bf k}$, while the density

167: operators $n_{\bf q}$ projected onto a single Landau level are

168: dynamically linked by the commutation relation~\cite{Girvin_86},

169: $[n_{\bf q}, n_{{\bf q}^\prime}] = 2 i \sin (\frac12 l^2 {\bf q} \wedge

170: {\bf q}^\prime) n_{{\bf q} + {\bf q}^\prime}$, where $l = \sqrt{\hbar /

171: m \omega_c}$ is the magnetic length and $\omega_c = e B / m c$ is the

172: cyclotron frequency for the external magnetic field $B$. The above

173: commutation relation is exact but difficult to deal with. I achieve a

174: simplification by replacing it with its expectation value expanded to

175: the lowest order in ${\bf q}$. This way I obtain the following

176: approximate commutation relation for use in our effective theory:

177: %

178: \begin{equation}

179:  [n_{\bf q}, u_{{\bf q}^\prime}] = (2 \pi)^2 l^2 q_y

180: \delta({\bf q} + {\bf q}^\prime)

181: \label{commutator}

182: \end{equation}

183: %

184: (all other commutators vanish in the $q \to 0$ limit). Introducing the

185: canonical momentum $p$ by $\delta n = -\partial_y p$, I unify

186: Eqs.~(\ref{H}) and (\ref{commutator}) into the effective

187: action~\cite{Comment_on_commutator},

188: %

189: \begin{eqnarray}

190:  A_{\rm sm} &=& \int\limits_0^\beta d \tau \int d^2 r \left\{

191: -\frac{i}{l^2} \hbar p \partial_\tau u + \frac{Y}{2} (\partial_x u)^2

192:   + \frac{K}{2} (\partial_y^2 u)^2

193:   \right.

194: \nonumber\\

195: \mbox{} &+& \left.\frac12 (\partial_y p) {\bf U} (\partial_y p)

196:   - C \partial_x u \partial_y p

197:   \right\}.

198: \label{A_sm}

199: \end{eqnarray}

200: %

201: This general form can be shown to pass two important tests. First,

202: after a change of variables it reproduces~\cite{Fogler_in_preparation}

203: an effective action derived for a specific model of the quantum Hall

204: smectic~\cite{MacDonald_00}. Second, the density structure factor,

205: easily calculated for the Gaussian theory~(\ref{A_sm}),

206: %

207: \begin{equation}

208: S({\bf q}, \omega) \simeq \frac{\hbar}{m}\, {\rm Im}\,

209: \frac{q_y^2 Q(q)}

210: {Q(q) \omega_p^2 (q_y / q)^2 - \omega^2 \omega_c^2 - i \omega \delta},

211: \label{Structure_smectic}

212: \end{equation}

213: %

214: coincides up to the Bose factor and dissipative terms with the finite

215: temperature result of Ref.~\cite{Fogler_00}. The notations used here are

216: $Q({\bf q}) = (\tilde{Y} q_x^2 + K q_y^4) / m n_0$, $\tilde{Y} = Y - C^2

217: / U$, and $\omega_p(q) = [n_0 U(q) q^2 / m]^{1/2}$. On the

218: $\omega > 0$ side, $S({\bf q}, \omega)$ consists of a single

219: $\delta$-function at the frequency of the magnetophonon mode,

220: %

221: \begin{equation}

222: \omega({\bf q}) = \frac{q_y}{q}\frac{\omega_p}{\omega_c}

223:                   \sqrt{Q({\bf q})}.

224: \label{omega_smectic}

225: \end{equation}

226: %

227: 

228: {\it Dislocations and duality\/}.--- 2D smectics can exist only at zero

229: temperature. At $T > 0$ thermal fluctuations of the stripes restore the

230: translational symmetry, so that the highest possible degree of ordering

231: is that of the nematic~\cite{DeGennes_book}. The actual crossover of the

232: Hamiltonian from the short-distance smectic~(\ref{H}) to the

233: long-distance nematic form is quite nontrivial. It is driven

234: by thermally excited dislocations, which have an ability to screen the

235: compressional stress~\cite{Toner_81}. It is natural to assume then that

236: the smectic-nematic {\it quantum\/} phase transition is also driven by

237: topological defects. Pictorially, the difference between the smectic and

238: nematic can be represented as follows. The dislocations are viewed as

239: lines in the (2 + 1)D space. In the smectic phase, they form small

240: closed loops (Fig.~\ref{Fig_worldlines}a) that depict virtual pair

241: creation-annihilation events; in the nematic phase arbitrarily long

242: dislocation worldlines exist and may entangle

243: (Fig.~\ref{Fig_worldlines}b), similar to worldlines of particles in a

244: Bose superfluid~\cite{Feynman_book}.

245: 

246: Now I will present a mathematical formalism supporting

247: these qualitative considerations. It is analogous to

248: the duality transformation introduced in Ref.~\cite{Fisher_89}.

249: 

250: %

251: % FIG. 2

252: %

253: \begin{figure}

254: \includegraphics[width=2.1in,bb=122 265 568 513]{qn_2.ps}

255: \vspace{0.1in}

256: \setlength{\columnwidth}{3.2in}

257: \caption{

258: Worldlines of dislocations in (a) smectic (b) nematic.

259: \label{Fig_worldlines}

260: }

261: \end{figure}

262: %

263: 

264: Our first step is to incorporate the dislocations into the effective

265: action~(\ref{A_sm}). This is accomplished by factorizing the smectic

266: order parameter, $\Psi = \Psi_0 e^{i q_0 u} \times \Psi_D$, where $u$ is

267: a regular single-valued function and $\Psi_D$ is the phase factor due to

268: the dislocations. The derivatives of $u$ in Eq.~(\ref{A_sm}) should now

269: be replaced by $\partial_\mu u - (i / q_0) \Psi_D^* \partial_\mu \Psi_D

270: \equiv (\partial_\mu u)_{\rm tot}$, $\mu = \tau, x, y$. Decoupling the

271: quadratic part of the action with a Hubbard-Stratanovich field

272: $\sigma_\mu$, I get

273: %

274: \begin{eqnarray}

275: A &=& A_D + \int\limits_0^\beta d \tau \int d^2 {\bf r}

276: [-i (\partial_\mu u)_{\rm tot} \sigma_\mu - H_a],

277: \label{A_Hubbard}

278: \\

279: H_a &=& -i \frac{C}{Y} l^2 \sigma_x \partial_y \sigma_\tau

280: + \frac{\sigma_x^2}{2 Y}

281: + \sigma_y (-2 K \partial_y^2)^{-1} \sigma_y

282: \nonumber\\

283: \mbox{} &+& \frac{l^4}{2} \partial_y \sigma_\tau

284: \left({\bf U} - \frac{C^2}{Y}\right)

285: \partial_y \sigma_\tau,

286: \label{H_a}

287: \end{eqnarray}

288: %

289: where $\hbar = 1$, $\sigma_\tau \equiv p / l^2$, and $A_D$ contains

290: terms describing dislocation cores (see below). Integration over $u$

291: gives the constraint $\partial_\mu \sigma_\mu = 0$, which I implement by

292: means of an auxillary $U(1)$ gauge field $a_\mu$,

293: %

294: \begin{equation}

295: \sigma_\mu = \epsilon_{\mu\nu\lambda}

296: \partial_\nu a_\lambda \equiv [\partial \times a]_\mu.

297: \label{a}

298: \end{equation}

299: %

300: This leads to the (dual) action of the form

301: %

302: \begin{equation}

303: A_{\rm dual} = A_D + \int\limits_0^\beta d \tau \int d^2 {\bf r}

304:  (-i \Lambda a_\mu j_\mu + H_a),

305: \label{A_dual_first}

306: \end{equation}

307: %

308: where $\Lambda = 2 \pi / q_0$ and $j_\mu = (2 \pi i)^{-1}

309: \epsilon_{\mu\nu\lambda} \partial_\nu (\Psi_D^* \partial_\lambda

310: \Psi_D)$ is the dislocation 3-current. This current can be expressed in

311: terms of the second-quantized dislocation field $\Phi$,

312: %

313: \begin{equation}

314: j_\mu = t_\mu \Phi^* (-i \partial_\mu +  \Lambda a_\mu) \Phi,

315: \label{j_mu}

316: \end{equation}

317: %

318: which is assumed to be bosonic. Other types of quantum statistics

319: for dislocations are exotic alternatives, which are

320: not investigated here.

321: 

322: Making standard  assumptions about $A_D$, I arrive at

323: %

324: \begin{eqnarray}

325: A_{\rm dual} &=& \int\limits_0^\beta d \tau \int d^2 {\bf r}

326:  \biggl\{\frac{t_\mu}{2}|(-i \partial_\mu - \Lambda a_\mu

327:  - e_D a_\mu^{\rm ext})

328: \Phi|^2\biggr.

329: \nonumber\\

330: \mbox{} &+& \biggl.V(\Phi) + H_a[a]\biggr\}.

331: \label{A_dual}

332: \end{eqnarray}

333: %

334: The obtained dual theory, Eqs.~(\ref{H_a}), (\ref{a}), and

335: (\ref{A_dual}) describes $\Phi$ bosons interacting with each other and

336: with the $U(1)$ gauge field $a_\mu$. The phenomenological parameters

337: introduced above are as follows. Parameter $t_\tau \sim \hbar^2 / E_c$,

338: where $E_c$ is the dislocation core energy. It was estimated within the

339: Hartree-Fock approximation in Refs.~\cite{Fogler_96} and

340: \cite{Wexler_xxx}. Near $\nu \sim \nu_c$ where the interstripe

341: separation $\Lambda$ approaches the quantum fluctuation scale $l$, $E_c$

342: is expected to be much smaller than the Hartree-Fock value, but any

343: precise estimate is challenging. Parameter $t_x$ of dimension of ${\rm

344: energy} \times {\rm (length)}^2$ is the hopping matrix element for

345: dislocation motion in the $\hat{\bf x}$-direction, i.e., dislocation

346: {\it glide\/}. Such a glide requires quantum tunneling and is

347: exponentially small unless $\Lambda \lesssim l$. Parameter $t_y$

348: describes the dislocation climb, which also originates from the dynamics

349: on the microscopic length scales. One may recall that the climb requires

350: mobile point defects. Although such defects are not among fundamental

351: low-energy excitations of the theory, composite point defects, in the

352: form of short stripe segments or dislocation pairs, do exist. They are

353: most likely short lived because dislocations are not expected to form

354: bound pairs in the nematic phase. Yet during their limited life span

355: such composite defects can perfectly well assist the climb by

356: micromotions of constituent electrons. In general, I am not aware of any

357: fundamental principle that would protect $t_y = 0$ value, although it

358: seems reasonable that $t_y \ll t_x$ in our case.

359: 

360: Yet another phenomenological variable is the potential $V(\Phi) = m_\Phi

361: |\Phi|^2 + r_\Phi |\Phi|^4 + \ldots$, which accounts for a self-energy

362: and a short-range interaction between the dislocations; the scales of

363: $m_\Phi$ and $r_\Phi$ are set by $E_c$ and $E_c \Lambda^2$,

364: respectively. Finally, $e_D$ is electric charge of the dislocation that

365: couples to the external vector potential $a_\tau^{\rm ext} = a_x^{\rm

366: ext} = 0$, $a_y^{\rm ext} = B x$. This coupling is introduced only for

367: the sake of generality. Since I study electron liquid crystal phases

368: derived from incompressible liquids, I expect dislocations to be

369: electrically neutral, i.e., $e_D = 0$.

370: 

371: Another few comments are in order. The derived theory is meant to

372: capture only the dynamics of neutral (dipolar) excitations of the

373: system. The underlying incompressible state has its own dynamics

374: characterized, most importantly, by the quantized Hall conductivity.

375: 

376: There is a very strong similarity between our theory and the statistical

377: mechanics model of the 3D smectic-nematic transition studied

378: by Toner and others~\cite{Toner_82}.

379: 

380: Integrating out the gauge field from $A_{\rm dual}$ leads to the model

381: of dislocation lines interacting via an effective Biot-Savart potential

382: (which in our case is short-range for Coulombic

383: $U$~\cite{Fogler_in_preparation}). One may argue~\cite{Helfrich_78} that

384: the unlimited expansion of the dislocation loops occurs when the energy

385: cost for creating a large loop of length $L$ is compensated by the

386: ``entropic gain'' $L / l_0$, where the persistence length $l_0$ is

387: determined by short-range physics. Despite the physical appeal of this

388: argument, the gauge theory~(\ref{A_dual}) is presumably better suited

389: for a quantitative analysis (perhaps, along the lines of

390: Ref.~\cite{Toner_82}).

391: 

392: {\it Phases and their collective modes\/}.--- Let us now see how the

393: smectic and nematic states are reproduced in the dual theory. As

394: discussed above, the smectic phase corresponds to $\langle \Phi \rangle

395: = 0$. In this case $A_{\rm dual}$ reduces to $H_a$, which is quadratic.

396: The low-energy dynamics is that of a gas of noninteracting Goldstone

397: bosons, which are the aforementioned magnetophons. It is a simple

398: exercise to verify that their dispersion relation is given by

399: Eq.~(\ref{omega_smectic}).

400: 

401: In the nematic phase dislocations have condensed, $\langle \Phi \rangle

402: = \Phi_0 \neq 0$. In conventional local $U(1)$ gauge theories,

403: the appearance of such an order parameter triggers the Anderson-Higgs

404: mechanism, eliminating the gapless Goldstone modes. This can not be the

405: case here because the nematic state does break the continuous symmetry

406: with respect to spatial rotations. The seeming paradox is resolved due

407: to the peculiar feature of the present gauge theory: the nonlocality of

408: the gauge-field strength term $H_a$, see Eq.~(\ref{H_a}). By virtue of

409: that, the condensation of $\Phi$ merely stiffens the collective mode,

410: leaving it gapless. Neglecting terms proportional to $C$, I find that

411: %

412: \begin{equation}

413: \omega_1({\bf q}) =

414:  \left(\frac{m_x}{m_\tau} q_x^2 + m_x K q_y^2\right)^{1/2},

415: \:\: m_\mu \equiv t_\mu \Lambda^2 |\Phi_0|^2.

416: \label{omega_nematic_1}

417: \end{equation}

418: %

419: As for the magnetophonon mode of the smectic~(\ref{omega_smectic}), it

420: indeed acquires a small gap $\sqrt{m_y Y}$ at $q = 0$. It anti-crosses

421: with the acoustic branch~(\ref{omega_nematic_1}) near the point

422: $\omega_1^2({\bf q}) \sim m_y Y$, and at larger $q$ becomes the lowest

423: frequency collective mode with the dispersion relation

424: %

425: \begin{equation}

426: \omega_2({\bf q}) = \left[\frac{q^2_x q^2_y}{m^2 \omega_c^2} Y U(q)

427:                   + m_y Y\right]^{1/2}

428: \label{omega_nematic_2}

429: \end{equation}

430: %

431: only slightly different from~(\ref{omega_smectic}). At such $q$ the

432: structure factor of the nematic has two sets of $\delta$-functional

433: peaks,

434: %

435: %\begin{equation}

436: \[

437: S(\omega, {\bf q}) = \frac{\pi \hbar q_y^2}{m \omega_c^2}

438: \left[\frac{K q_y^4}{m n_0} \delta(\omega^2 - \omega_1^2) +

439: \frac{Y q_x^2}{m n_0} \delta(\omega^2 - \omega_2^2)\right],

440: \]

441: %\label{Structure_nematic}

442: %\end{equation}

443: %

444: which split between themselves the spectral weight of the single

445: collective mode of the smectic. The presence of the two modes can be

446: explained by the existence of two order parameters: a unit vector (more

447: precisely, director) ${\bf N}$ normal to the local stripe orientation

448: and the complex wavefunction $\Phi_0$ of the dislocation condensate.

449: Classical 2D nematics have two (overdamped) modes virtually for the same

450: reason~\cite{Toner_81}.

451: 

452: In conclusion, let us address measurable properties of the novel quantum

453: Hall states considered here. At low temperature both the parent

454: incompressible state and its liquid crystal descendants will be

455: insulating. If $T$ is not too small, formation of stripe superstructures

456: can be deduced from the anisotropic

457: magnetoresistance~\cite{Experiments}. On the other hand, the microwave

458: absorption will be anisotropic even at low $T$ and would enable one to

459: further distinguish between the smectic and nematic phases: the nematic

460: will show two dispersing collective modes while the smectic will produce

461: a single one. To circumvent disorder pinning effects, such measurements

462: should be done at high enough $q$.

463: 

464: {\it Acknowledgements}---This work is supported by MIT Pappalardo

465: Program in Physics. I wish to thank X.-G. Wen for useful discussions,

466: and also A.~Dorsey, L.~Radzihovsky, and C.~Wexler for valuable comments

467: on the manuscript.

468: 

469: \begin{thebibliography}{29}

470: 

471: \bibitem{Seul_95} M.~Seul and D.~Andelman,

472: Science, {\bf 267}, 476 (1995).

473: 

474: \bibitem{Neutron_star} S.~Reddy, G.~Bertsch, and M.~Prakash,

475: Phys.\ Lett.\ B {\bf 475}, 1 (2000).

476: 

477: \bibitem{Chklovskii} D.~B.~Chklovskii and A.~A.~Koulakov,

478: Physica A {\bf 284}, 318 (2000).

479: 

480: \bibitem{High_Tc_stripes}

481: %J.~M.~Tranquada {\it et. al.\/},

482: %\prl {\bf 73}, 1003 (1994); 

483: J.~M.~Tranquada {\it et. al.\/},

484: Nature {\bf 375}, 561 (1995);

485: S.~Mori, C.~H.~Chen, and S.-W.~Cheong,

486: Nature {\bf 392}, 473 (1998).

487: 

488: \bibitem{Fogler_96} A.~A.~Koulakov, M.~M.~Fogler, and

489: B.~I.~Shklovskii,

490: \prl {\bf 76}, 499 (1996);

491: M.~M.~Fogler, A.~A.~Koulakov, and

492: B.~I.~Shklovskii,

493: \prb {\bf 54}, 1853 (1996).

494: 

495: \bibitem{Moessner_96} R.~Moessner and J.~T.~Chalker,

496: \prb {\bf 54}, 5006 (1996).

497: 

498: \bibitem{Critical_nu} M.~M.~Fogler and A.~A.~Koulakov,

499: \prb {\bf 55}, 9326 (1997);

500: E.~H.~Rezayi, F.~D.~M.~Haldane, and K.~Yang,

501: \prl {\bf 83}, 1219 (1999)

502: N.~Shibata and D.~Yoshioka,

503: cond-mat/0101401.

504: 

505: \bibitem{Scarola_01} V.~W.~Scarola, K.~Park, J.~K.~Jain,

506: \prb {\bf 62}, R16259 (2001).

507: 

508: \bibitem{Experiments} M.~P.~Lilly, K.~B.~Cooper, J.~P.~Eisenstein,

509: L.~N.~Pfeiffer, and K.~W.~West,

510: \prl {\bf 82}, 394 (1999);

511: R.~R.~Du, D.~C.~Tsui, H.~L.~St\"ormer, L.~N.~Pfeiffer, and K.~W.~West,

512: Solid\ State\ Commun. {\bf 109}, 389 (1999).

513: 

514: \bibitem{Balents_96} L.~Balents,

515: Europhys.\ Lett. {\bf 33}, 291 (1996).

516: 

517: \bibitem{Fradkin_99} E.~Fradkin and S.~A.~Kivelson,

518: \prb {\bf 59}, 8065 (1999).

519: 

520: \bibitem{Yi_01} H.~Yi, H.~A.~Fertig, and R.~C\^ot\'e,

521: \prl (2001).

522: 

523: \bibitem{MacDonald_00} A.~H.~MacDonald and M.~P.~A.~Fisher,

524: \prb {\bf 61}, 5724 (2000).

525: 

526: \bibitem{Fogler_in_preparation} M.~M.~Fogler, in preparation.

527: 

528: \bibitem{Musaelian_96} K.~Musaelian and R.~Joynt,

529: J.\ Phys.\ Cond.\ Mat. {\bf 8}, L105 (1996).

530:  

531: \bibitem{Comment_on_1st_order} A finite-size study by Rezayi {\it et

532: al.\/}~\cite{Critical_nu} suggests that the transition from the stripe

533: phase to a uniform state as a function of the interaction parameters can

534: also occur via a first-order transition without the intermediate nematic

535: phase.

536: 

537: \bibitem{Comment_on_soft_modes} I am interested in a regime

538: with very strong quantum fluctuations where the smectic order parameter

539: is small and only the main harmonic of the soft mode is

540: important.

541: 

542: \bibitem{Comment_on_compressible} If the parent state is compressible,

543: the additional low-energy degrees of freedom would generate an effective

544: viscosity of Landau damping (or Caldeira-Leggett) type entering

545: Eq.~(\ref{H}). Its consequences are left for future study.

546:  

547: \bibitem{Comment_on_commutator} In principle, $\delta n = -\partial_y p

548: + f(u)$ with some function $f$. In the absence of dislocations the most

549: relevant term is $f = A \partial_x u$, which leads again to

550: Eq.~(\ref{A_sm}). If dislocations are allowed, the situation is more

551: complicated, see later in the main text.

552: 

553: \bibitem{DeGennes_book} P.~G.~de~Gennes and J.~Prost,

554: {\it The Physics of Liquid Crystals\/} (Oxford University Press, New York,

555: 1995).

556: 

557: \bibitem{Fogler_00} M.~M.~Fogler and V.~M.~Vinokur,

558: \prl {\bf 84}, 5828 (2000).

559: 

560: \bibitem{Girvin_86} S.~M.~Girvin, A.~H.~MacDonald, and P.~M.~Platzman,

561: \prb {\bf 33}, 2481 (1986).

562: 

563: \bibitem{Toner_81} J.~Toner and D.~R.~Nelson,

564: \prb {\bf 23}, 316 (1981). For application of this

565: theory to the quantum Hall stripes, see Refs.~\cite{Fradkin_99}

566: and \cite{Wexler_xxx}.

567: 

568: \bibitem{Wexler_xxx} C.~Wexler and A.~T.~Dorsey,

569: cond-mat/0009096.

570: 

571: \bibitem{Feynman_book} R.~P.~Feynman and A.~R.~Hibbs,

572: {\it Quantum Mechanics and Path Integrals\/} (McGraw-Hill, New York,

573: 1965).

574: 

575: \bibitem{Fisher_89} M.~P.~A.~Fisher and D.~H.~Lee,

576: \prb {\bf 39}, 2756 (1989).

577: 

578: \bibitem{Toner_82} J.~Toner,

579: \prb {\bf 26}, 462 (1982);

580: A.~R.~Day, T.~C.~Lubensky, and A.~J.~McKane,

581: \pra {\bf 27}, 1461 (1983).

582: 

583: \bibitem{Helfrich_78} W.~Helfrich,

584: J.\ Phys.\ (Paris) {\bf 39}, 1199 (1978).

585: 

586: \end{thebibliography}

587: \end{document}

588: 

589: ------------------------

590: 

591: An example of the quantum nematic is the state

592: with the wavefunction~\cite{Musaelian_96}

593: %

594: \begin{equation}

595: \Psi = \prod\limits_{j < k} z_{j k} (z_{j k}^2 - a^2 l^2)

596: \exp \left(-\sum_j |z_j|^2 / 4 l_B^2\right),

597: \label{nematic_trial_wavefunction}

598: \end{equation}

599: %

600: where $z_{j k} = z_j - z_k$, $z_j$ is the complex coordinate of $j$th

601: electron, $z_j = x_j + i y_j$, $l_B$ is the magnetic length, and $a$ is

602: a complex parameter that determines the degree of orientational order

603: and the direction of the stripes (the rotational invariance is broken if

604: $|a|$ exceeds some critical value). This particular wavefunction

605: corresponds to $\nu = \frac13$ but its generalization to higher Landau

606: levels is straightforward.

607: 

608: