1: \documentstyle[prl,aps,epsfig,multicol]{revtex}

2: %\textwidth=16truecm

3: %\textheight= 20truecm

4: %\hoffset=-1truecm

5: %\voffset=-2truecm

6: \begin{document}

7: \def\be{\begin{equation}}

8: \def\ee{\end{equation}}

9: \def\bearr{\begin{eqnarray}}

10: \def\eearr{\end{eqnarray}}

11: \def\tc{$T_c~$}

12: \def\tcl{$T_c^{1*}~$}

13: \def\c2{ CuO$_2~$}

14: \def\ruo{ RuO$_2~$}

15: \def\lsco{LSCO~}

16: \def\bi{bI-2201~}

17: \def\tl{Tl-2201~}

18: \def\hg{Hg-1201~}

19: \def\sro{$Sr_2 Ru O_4$~}

20: \def\rc{$RuSr_2Gd Cu_2 O_8$~}

21: \def\mgb{$MgB_2$~}

22: \def\pz{$p_z$~}

23: \def\ppi{$p\pi$~}

24: \def\sqo{$S(q,\omega)$~}

25: \def\tperp{$t_{\perp}$~}

26: \def\he4{${\rm {}^4He}$~}

27: \def\tss{${T_{ss}}$~}

28: 

29: \title{And quiet flows the supersolid \he4}

30: 

31: \author{ G. Baskaran \\

32: Institute of Mathematical Sciences\\

33: C.I.T. Campus,

34: Chennai 600 113, India }

35: 

36: \maketitle

37: \begin{abstract}

38: A superfluid having atomic scale superflow of a hexagonal lattice

39: of vortex and antivortex filaments, described by a single macroscopic

40: wave function is presented as a supersolid. As superfluid \he4

41: is pressurized, at a first order transition, rotons 

42: (atomic scale current circulation, a vortex loop) not only condense 

43: but also {\em expand and fuse into hexagonal or other 

44: complex superflow patterns}. The vortex 

45: core contains an excess density of non-condensate atoms. 

46: Further, a Kelvin (m = 0,

47: necklace) mode condenses in the vortex filaments. It results in a 3D

48: atom density wave of hcp symmetry. In our theory, superfluid

49: phase stiffness, rather than atom localization, imitates a solid

50: like rigidity.

51: \end{abstract}

52: 

53: \begin{multicols}{2}[]

54: Supersolid\cite{anlif,chester} is a quantum crystal that exhibits 

55: non classical moment of inertia\cite{leggett1}, 

56: a type of superfluid response.

57: Decades of efforts\cite{ExpReview}, following early theoretical 

58: suggestions, have 

59: culminated in a recent striking observation of non classical moment 

60: of inertia, by Kim and Chan\cite{kc1kc2}. This is 

61: yet another jewel in the crown of condensed \he4, an elusively

62: simple one component bose system; the deeper one searches and digs, 

63: the more surprises are in store.

64: This work has excited a renewed interest\cite{theory,pwaSS}

65: in the quantum many body 

66: theory of supersolid. It is likely to open new 

67: directions and find new phenomena in experiments; it also offers a good 

68: opportunity to bring the field of cold atom BEC, boson Mott insulators 

69: etc., closer to supersolid \he4. A correct theoretical understanding

70: of the supersolid phenomena will throw new light into old experimental 

71: quantum anomalies in the superfluid-solid helium 

72: interface\cite{RussianBook}, such as 

73: crystallization waves, Kapitza resistance etc.     

74: 

75: The supersolid predicted by Andreev, Lifshitz\cite{anlif} 

76: and Chester\cite{chester} has a 

77: reference crystal of localized atoms; large amplitude quantum

78: fluctuations create ground state vacancies or defects which 

79: undrgo condensation. In our mechanism superfluid is the

80: reference system and solid like rigidity 

81: emerges from the superfluid stiffness in a fundamental way.

82: A spatially periodic,

83: atom scale superfluid flow develops spontaneously in the ground 

84: state leading to a solid like response, in addition to 

85: superfluidity. We call it a roton fusion(ROFU) mechanism, as 

86: the atom scale flow pattern arises spontaneously

87: from a condensation and a complex fusion of real rotons (vortex

88: loop of atomic dimensions),

89: as we pressurize superfluid \he4. ROFU mechanism is one collective way of 

90: releasing the kinetic energy frustration, that is present in a solid with

91: localized atoms. It is a kind of spontaneous generation of 

92: staggered Abrikosov vortex lattice, created by 

93: hydrostatic pressure. Hydrostatic pressure, a scalar, mimics an 

94: internal staggered magnetic field. 

95: 

96: 

97: Roton minimum of superfluid \he4 has been viewed\cite{enz}, 

98: for quite some  

99: time as a soft mode that drives a superfluid-solid phase 

100: transition, as rotons have wave vectors close to the 

101: reciprocal lattice vectors of the hcp \he4 crystal.

102: A new and crucial ingredient in our theory is to use the 

103: non-trivial inner structure of roton, an atomic size quantized

104: vortex loop. In our theory rotons do more than condensation; they 

105: expand and fuse into ordered vortex and antivortex filaments

106: (atom scale thickness)  and 

107: loose their identity. 

108: \begin{figure}[h]

109: \epsfxsize 8cm

110: \centerline{\epsfbox{rofufig1.eps}}

111: \caption{ Figure 1. a) Region of roton minima in the $(k_x,k_y,0)$

112: plane. ${\bf Q}_1$, ${\bf Q}_2$ and ${\bf Q}_3$ are

113: Fourier component vectors of the macroscopic 

114: condensate wave function $\langle \Psi({\bf r})\rangle$.

115: b) Unit cell of the hexagonal vortex-antivortex array (tubes); 

116: it has a 2D atom density wave of hexagonal

117: symmetry c) Kelvin (azimuthal quantum number
 m = 0, necklace) 

118: mode with a finite $\pm Q_z$ condenses in the vortex in a 

119: staggered fashion. It results in an atom density wave of hcp 

120: lattice. Lattice sites (not position of localized atoms) are 

121: denoted by shaded circle.} 

122: \end{figure}

123: 

124: In superfluid \he4, nearly 90 \% of the atoms are out side

125: the zero momentum condensate. These large fraction of atoms

126: with finite momenta appear as spatially uniform quantum 

127: fluctuation of the superfluid vacuum. They are part of the 

128: vacuum and the ground state is a 100 \% superfluid.

129: When vortices and antivortices are spontaneously created in

130: the ground state, a part of the non-condensate fraction gets 

131: piled up in the vortex core region, creating a net atom 
 density 

132: wave of hcp symmetry.  The self consistent potential 
 that maintains 

133: this redistribution arises from the underlying
 superfluid stiffness. 

134: In this sense superfluid stiffness
 gives a solid like rigidity to 

135: the 3D atom density wave.
 A fragile 2\% superfluid condensate(as 

136: measured by Kim and 
 Chan), by a complex maneuver of its own flow, 

137: orchestrates 
 the 3D atom density wave and a rich `lattice dynamics'.

138: Periodic superflow leads to a periodic modulation of quantum 

139: fluctuation and non condensate component.

140: 

141: In what follows we develop some heuristic pictures, followed by 

142: a Bogoliubov theory and then discuss some consequences. 

143: 

144: Kim and Chan have discovered non-classical moment of 

145: inertia in pressurized solid \he4; they observe strong 

146: thermal hysteresis, small variation of supersolid transition 

147: temperature \tss for a pressures range, 25 to 60 bars, and an 

148: anomalous increase of \tss with addition of ${\rm {}^3He}$ 

149: impurity. More importantly, the authors indicate that the 

150: superfluid response may not be due to zero point vacancies or 

151: defects or interfaces. That is, it is likely one is dealing 

152: with a solid with integer number of atoms per unit cell and 

153: yet exhibiting a superfluid property. 

154: 

155: While building our theory we remember that 

156: at the superfluid-solid first order transition at T = 0,

157: inter \he4 atom distance decreases only by a small, $\sim 3 \%$,

158: (a density decrease $\sim 10 \%$). This means that locally

159: the quantum solid is no more crowded than the quantum liquid is. 

160: Thus any generalized rigidity that emerges on both sides is 

161: likely to have the same quantum character, namely some kind of

162: phase stiffness (ODLRO) arising from local coherent number 

163: fluctuations. Further, the superfluid

164: solid coexistence line in the P-T plane has nearly zero slope 

165: for $T < 0.2$ K indicating 

166: that the two phases have the same entropy per mole, according

167: to Clausius-Clapeyron equation $\frac{dP_m}{dT} =

168: \frac{S_L - S_S}{V_L - V_S}$. This means that density of 

169: states of low energy bosonic quasi particles is nearly 

170: the same for superfluid and supersolid at the coexistence 

171: line, indicating a possible deep connection between the

172: stiffness of a superfluid and rigidity of a supersolid.

173: 

174: 

175: Now we present a Bogoliubov theory to illustrate our ROFU mechanism. 

176: Bogoliubov theory is a mean field theory that works well for weakly 

177: interacting bosons. However, for a proper choice two body pseudo

178: potential, it captures qualitative features and some

179: quantitative features, including roton spectrum. The model 

180: Hamiltonian is:

181: \be

182: H = \sum 

183: (\epsilon_k -\mu)

184: b^\dagger_{\bf k}

185: b^{}_{\bf k} +

186: \frac{1}{\Omega}\sum V({\bf q}) b^\dagger_{\bf k}

187: b^{}_{\bf k-q} b^\dagger_{\bf k'}

188: b^{}_{\bf k' + q}

189: \ee

190: Here $\epsilon_k = \frac{\hbar^2}{2M}$ is the kinetic energy

191: of \he4 atoms with mass M, chemical potential $\mu$, created

192: by operators $b^{\dagger}$'s.  $V({\bf q})$ is the two body, 

193: effective or pseudo potential and $\Omega$ is the volume of the 

194: system.

195: 

196: In the uniform superfluid phase, Bogoliubov

197: theory starts by replacing the zero momentum

198: operator $b^{}_0 \rightarrow n_0^{\frac{1}{2}}e^{i\phi}

199: N^{\frac{1}{2}}$, a 

200: classical expectation value, with $n_0$ as a condensate 

201: fraction.

202: The Bogoliubov quasi particle spectrum of the uniform 

203: superfluid state has the well known form 

204: \be

205: \hbar\omega_k = \sqrt{\left(\frac{\hbar^2k^2}{2M}\right)^2 + 

206: \left( \frac{\hbar^2 k^2}{2M}\right)  2 n_0 V(k)}

207: \ee

208: where $n_0$ is the zero momentum condensate fraction. When 

209: the potential $V(k)$ is negative for a range of k, the spectrum

210: has a roton minimum; an example is Bruckner

211: pseudo potential, $V(k) \approx V_0\frac{\sin ka_0}{ka_0}$. 

212: with $a_0~ \approx ~ 2.1 \mathring{\rm A}$ and 

213: $V_0 \approx 15~K$.

214: 

215: 

216: Landau suggested roton as the first excitation that has associated 

217: with it a rotational velocity flow (and hence the name roton). 

218: Feynman viewed it as a single atom motion, but dressed by a 

219: back flow, the net effect being a quantized vortex loop of atomic

220: dimensions.

221: Recent variational study and wave packet analysis\cite{rotonReatto} 

222: confirms this and shows roton wave packet as a ball of

223: current disturbance, and little density variation. 

224: 

225: Within the Bogoliubov theory, the roton minimum (figure 1a)

226: reaches zero energy (become completely soft), when

227: $V(k) = - \frac{1}{2n_0}\frac{\hbar^2 k^2}{2M}$.

228: This indicates an instability of 

229: the uniform superfluid solution. We expect a roton condensation 

230: and reorganization of the ground state. 

231: A complex current flow associated with a roton, rather than a 

232: simple density variation, made us wonder what will be the 

233: consequence of a roton condensation. It became clear that 

234: {\em a primary consequence of roton condensation will be formation

235: of a current density wave, not atom density wave}.

236: This resulted in our roton fusion hypothesis. 

237: According to this hypothesis the 

238: condensed rotons expand and fuse into a hexagonal array of

239: vortices and antivortices. Our hypothesis is naturally 

240: influenced by the known hcp structure of solid \he4.

241: 

242: To test our hypothesis we have compared the energy of uniform

243: superfluid state with an off diagonal long range order (ODLRO) 

244: ansatz that contains a hexagonal array of vortices and 

245: antivortices. After some struggle we found a simple and elegant

246: ansatz - in phase superposition of three plane waves generates 

247: the desired vortex structure:

248: \bearr

249: \langle \Psi({\bf r})\rangle  \equiv \psi_0({\bf r})\sim

250: n_0^{\frac{1}{2}}e^{i\phi}(

251: e^{i{\bf Q}_1 \cdot {\bf r}}+

252: e^{i{\bf Q}_2 \cdot {\bf r}}

253: + e^{i{\bf Q}_3 \cdot {\bf r}})

254: \eearr

255: It also gives a lower energy 

256: than the uniform superfluid state, for a choice of $V(k)$.

257: Here ${\bf Q}_1= \frac{2\pi}{a}(\frac{2}{\sqrt 3},0)$, 

258: ${\bf Q}_{2,3} = \frac{2\pi}{a}(\frac{-1}{\sqrt 3},\pm 1)$,

259: ${\bf Q}_1 + {\bf Q}_2 + {\bf Q}_3 = 0$, `a' is the lattice parameter

260: of the hexagonal lattice. Here, $n_0$ and `a'

261: are variational parameters. The above ansatz leads to variation

262: of atom density only in the xy-plane, a 2D atom density wave.

263: This results in a supersolid behaviour along the x and y

264: directions and superfluid in z-direction - it is an anisotropic 

265: supersolid, a type of quantum liquid crystal.   Later we 

266: will improve it to get a 3D atom density wave of hcp 

267: symmetry. We can find more complex flow pattern (braided 

268: and knotted vortices) leading to an

269: atom density wave of hcp symmetry using more ${\bf Q}$'s etc.

270: 

271: One of the triangular sub lattices of the hexagonal lattice contains

272: vortices and the other antivortices (figure 1b). It is easy to show 

273: that equation (3) has asymptotic form 

274: \be

275: \langle \Psi ({\bf r})\rangle

276:  \sim x+iy~~~({\rm vortex}),~x-iy~~~({\rm antivortex})

277: \ee

278: respectively around the vortex and antivortex filaments of the two

279: sub lattices. 

280: 

281: In addition to global U(1) symmetry (overall phase of  

282: $\langle \Psi({\bf r})\rangle $) and translational symmetry 

283: there is a discrete symmetry breaking arising from P\&T violation. 

284: That is, another degenerate solution,

285: not connected by a global phase rotation of equation (5) is

286: obtained by the replacement ${\bf Q}_i \rightarrow - {\bf Q}_i$

287: in equation (3); equivalently, by an interchange

288: of vortices and antivortices. 

289: 

290: According to equation (4) the condensate fraction vanishes quadratically 

291: as we approach the core of line vortices. Further, the fraction

292: of the total particles condensed in our macroscopic wave function 

293: is $ < 1 $. Where are the rest of non-condensed particles ?

294: In the Bogoliubov theory of uniform superfluid, these are the 

295: particles that carry non zero momenta and they appear as a spatially 

296: uniform quantum fluctuations of the zero momentum condensate. 

297: They influence, for example the superfluid stiffness, through 

298: the spectrum; otherwise these finite momentum components are

299: not visible in the low energy dynamics of the irrotational 

300: superfluid.

301: 

302: In our ROFU, a spatial density

303: variation of the non-condensate fraction is induced.

304: In particular they appear as extra atom density in the normal

305: core region of our ordered vortices, where the condensate density

306: vanishes. This is somewhat counter intuitive, as one expects a 

307: depletion of fluid density at the vortex core.

308: It is known from an early work of Fetter\cite{fetter}

309: that within Bogoliubov 

310: theory, in the vortex core region the local atom density increases 

311: compared to the uniform background atom density. This is a 

312: remarkable non-local quantum effect, appearing within the 

313: Bogoliubov theory, as explained by Fetter.  

314: 

315: We elaborate this point further, by comparing a reference

316: Bose Einstein condensed (BEC) state

317: \be 

318: \Psi_{\rm BEC} \sim \prod_i 

319: (e^{i{\bf Q}_1 \cdot {\bf r}}+

320: e^{i{\bf Q}_2 \cdot {\bf r}}

321: + e^{i{\bf Q}_3 \cdot {\bf r}})

322: \ee

323: with the N-particle projected wave function of 

324: Bogoliubov theory for our ROFU solution:

325: \be

326: \Psi_{\rm N}[{\bf r}_i]

327: = \sum_{P} 

328: \chi ({\bf r}_{P1},{\bf  r}_{P2} ) \chi ({\bf r}_{P3}, {\bf r}_{P4}) .....

329: \chi ({\bf r}_{PN-1}, {\bf r}_{PN})

330: \ee

331: Summation over permutation P symmetrises the wave function.

332: The pair function $\chi({\bf r}_1, {\bf r}_2) \equiv

333: \eta ({\bf r}_1,{\bf r}_2) 

334: \psi_0 \left[ \frac{{\bf r}_1 + {\bf r}_2}{2} \right] $.

335: When the pair function $\eta({\bf r}_1, {\bf r}_2) = 1$, equation (6)

336: becomes the same as BEC (equation 5). In the BEC state, atom density, 

337: by construction, vanishes as we approach the core of the 

338: line vortices. In the Bogoliubov wave function, equation (6), the

339: pair function $\eta$ makes an important difference and creates a 

340: a pile up of atom density through pushing some of the non-condensate 

341: fraction to core regions.  The pair function $\eta$ represents, a 

342: repulsive correlation induced between any two particles in the

343: Bogoliubov theory. It shows how pairs of particle are pushed in 

344: and out of the condensate by the Bogoliubov process 

345: $ \langle b_{{\bf Q}_1}\rangle \langle b_{{\bf Q}_2}\rangle 

346: b^{\dagger}_{\bf k}b^{\dagger}_{{- \bf k} + {\bf Q}_1 +

347: {\bf Q}_2}$ etc., in a space dependent fashion. 

348: 

349: We want to make certain remarks about the nature of ODLRO

350: in our supersolid phase. Our primary order

351: is a long range order in momentum space, an ODLRO. Bosons 

352: condense in a single macroscopic wave function 

353: $\psi_0({\bf r})$ given by 

354: equation (5).It  has three

355: Fourier coefficients which are not independent: 

356: $\langle b_{\bf Q_1}\rangle $= 

357: $\langle b_{\bf Q_2}\rangle $ =  

358: $\langle b_{\bf Q_3}\rangle$ = $n_0^{\frac{1}{2}}e^{i\phi}$:

359: they have same amplitude and phase.

360: 

361: The spatial variation of atom density implied by our ODLRO

362: should not be thought of as a diagonal long range order (DLRO).

363: A spatial periodic ordering of a small fraction of the 

364: non-condensate fraction is induced self consistently in our theory.

365: For example in our Bogoliubov factorization we get an anomalous

366: term such as 

367: $ \langle 

368: b^{\dagger}_{\bf Q_1}\rangle

369: \langle b^{}_{\bf Q_2}\rangle

370: b^{\dagger}_{\bf k} 

371: b^{}_{\bf k - Q_1 + Q_2 }$.

372: A density wave of wave vector $({\bf Q_1 - Q_2})$ induced by

373: the above term has its origin in single particle condensation.

374: In principle we could generate such an anomalous term through

375: non-vanishing averages such as, $\sum_k\langle

376: b^{\dagger}_{\bf k} b^{}_{\bf k + Q_1 - Q_2}\rangle$.

377: This will be an independent diagonal long range order (DLRO)

378: parameter. 

379: 

380: The solution we have discussed so far, equation (3), has no

381: density variation along z-axis. In real \he4

382: we expect maximum amount of atom density to be concentrated

383: in the normal core region. Thus it has its natural tendency for 

384: local spatial order, arising from short range repulsions. 

385: We view this 

386: as a Kelvin (azimuthal quantum number m = 0, necklace or

387: sausage ) mode condenses (figure 1c) at wave vectors

388: $Q_z = \pm \frac{2\pi}{c}$ in the vortex filaments. We modify 

389: equation (3) and generate a periodic 

390: modulation of the vortex core size, along the vortex filament

391: in a staggered fashion,

392: \bearr

393: \langle \Psi ({\bf r})\rangle  &\sim& (x + iy) (1 + \epsilon \cos Q_zz)

394:   ~~{\rm and}~~ \nonumber \\ 

395: &\sim& (x-iy)(1 - \epsilon \cos Q_zz) 

396: \eearr

397: close to core of the vortices of the two sub lattices. In view of

398: the short range interaction of the model Hamiltonian,

399: the total energy is reduced further for small $\epsilon$, another 

400: variational parameter.  This vortex core modulation leads to an 

401: atom density wave of hcp symmetry and a 3D solid like rigidity. 

402: 

403: We can choose our model parameters of Hamiltonian (equation 1)

404: such that our hexagonal lattice contains an average of 2 atoms

405: per unit cell or close to it.

406: 

407: From our solution it follows that the superfluid density is

408: mostly concentrated outside the vortex core, which is

409: the interstitial region of the hcp atom density wave (figure 1b and

410: 1c).

411: 

412: So far we have sketched a Bogoliubov theory, a ROFU solution and 

413: some key features of the ground state. Our theory is far from

414: rigorous and complete. However, the physical argument for a 

415: spontaneous generation of microscopic circulating ground state

416: current is compelling. Bogoliubov (mean field) theory, in view of 

417: its non perturbative character is capable of finding possible 

418: new phases in dense liquid \he4. 

419: 

420: 

421: In what follows we show how our

422: theory qualitatively explains salient features of Kim-Chan's

423: results. Further, we briefly sketch interesting consequences 

424: that also follow; some of them are very unique to our theory. 

425: 

426: In our ROFU solution we 

427: discussed a 2D solid and a 3D solid. It is likely that as 

428: a function of pressure there is a small region in the 

429: pressure-temperature diagram where the 2D solid intervenes, 

430: as shown in figure 2.

431: \begin{figure}[h]

432: \epsfxsize 6.5cm

433: \centerline {\epsfbox{rofufig2.eps}}

434: \caption{ Schematic phase diagram for \he4 in the pressure

435: temperature plane. An intermediate region of anisotropic

436: supersolid (supersolid in x, y direction and superfluid

437: in z-direction) is suggested.}

438: \end{figure}

439: 

440: We can qualitatively discuss some features of finite 

441: temperature phase transitions that are unique to ROFU solution.

442: Thermally produced vortices in the ROFU condensate will 

443: proliferate, depin and melt the underlying vortex lattice.

444: Disappearance of ODLRO will result in a solid of localized atoms. 

445: As the process is a melting of a vortex lattice, 

446: i) phase transition will be a first order one and ii) small 

447: traces of 
 ${}^3{\rm He}$ atoms could help pin the vortices 

448: and thereby increase the supersolid-solid transition temperature. 

449: Both are consistent with Kim-Chan's observations. The anomalously

450: low critical velocity, for destruction of the non classical 

451: moment of inertia (super solidity), observed by Kim and Chan

452: is likely to arise from a collective depinning of the ground

453: state vortices from their weak self consistent potential. 

454: 

455: A key prediction of our theory is the presence of atom scale 

456: circulation. One important consequence of this is on lattice 

457: dynamics. The P \& T violation in the ground state and the first 

458: order dynamics of vortices makes the lattice dynamics different 

459: from the classical hcp solid. We get splitting of degeneracies etc.; 

460: some of the anomalous modes can be viewed as coupled Kelvin modes. 

461: We will discuss them in a future publication. 

462: 

463: In summary, our supersolid is a superfluid in disguise with ineresting

464: consequences.  

465: 

466: It is a pleasure to thank N. Kumar for an early (1971) and 

467: insightful introduction to Leggett's paper\cite{leggett1} 

468: and a recent discussion; Dung-Hai Lee for alerting me of Kim and 

469: Chan's work\cite{kc1kc2}; Moses Chan for a brief discussion 

470: at Hong Kong, implying `no room for vacancy' superfluidity

471: in their results.

472: 

473: \begin{references}

474: 

475: \bibitem{anlif}A. F. Andreev, I.M. Lifshitz,

476: Sov. Phys. JETP, {\bf 29} 1107 (1969)

477: \bibitem{chester}G.V. Chester, Phys. Rev. {\bf A 2} 256 (1970)

478: \bibitem{leggett1}A. Leggett, Phys. Rev. Lett., {\bf 25} 1543 (1970)

479: \bibitem{ExpReview} M. W. Meisel, Physica {\bf B 178} 121 (1992);

480: J. M. Goodkind, Phys. Rev. Lett., {\bf 89} 095301 (2002)

481: \bibitem{kc1kc2} E. Kim and M.H.W. Chan, Nature {\bf 427}

482: 225 (2004); Science, {\bf 305} 1941 (2004)

483: \bibitem{theory}D.M. Ceperley and B. Bernu,

484: Phys. Rev. Lett., {\bf 95}~155303  (2005); D. Weiss,

485: cond-mat/0407043; 

486: W. M. Saslow, cond-mat/0407166;

487: S. Koh, cond-mat/0411713;

488: N. Prokof'ev and B. Svistunov, cond-mat/0409472;

489: E. Burovski et al., cond-mat/0412644;

490: M. Tiwari and A. Datta, cond-mat/0406124

491: H. Zhai and Y.S. Wu, cond-mat/0501309;

492: X. Dai, M. Ma and F.C. Zhang, cond-mat/0501373

493: \bibitem{pwaSS}P.W. Anderson,

494: `Basic Notions of Condensed Matter Physics',

495: Ch 4, pp143, (Benjamin, 1984);

496: P. W. Anderson, cond-mat/0504731

497: \bibitem{RussianBook} A. Ya. Parshin, in Low temperature

498: Physics, Ed. A.S. Borovik-Romanov (MIR Publishers, Moscow, 

499: 1985) pp.15; K. N. Zinov'ea, ibid. pp 78

500: \bibitem{enz}T. Schneider and C.P. Enz, Phys. Rev. Lett.,

501: {\bf 27} 1186 (1971)

502: \bibitem{rotonReatto}D.E. Galli, E. Cechetti and L. Reattto,

503: Phys. Rev. Lett., {\bf 77} 5401 (1996)

504: \bibitem{fetter}A. L. Fetter., Annals of Physics, {\bf 70} 67 (1972)

505: \end{references}

506: \end{multicols}

507: \end{document}

508: 

509: 

510: The vortex core region contains the `atomic sites' of the 

511: hcp lattice. In in the vicinity of `atomic sites', superfluid 

512: fraction is 

513: absent but there is an excess of non-condensate fraction.

514: Thus in the ROFU state, superfluid is mostly an interstitial 

515: phenomena, particularly when

516: the number of atom per unit cell of the hcp lattice is 2. 

517: The corresponding vacancies have become part of the incoherent

518: non-condensate fraction. It means that vacancies do not bose 

519: condense. That is, even if we think of our ROFU state as

520: a solid with two atoms per unit cell, the interstials are

521: bose condensed, where as the vacancies are not. Thus {\em there 

522: is a fundamental difference between vacancies and interstitials}

523: in our theroy. This is consistent with the difference that

524: Anderson points in his early and recent discussions of 

525: supersolid \he4.

526: