<?xml version="1.0" encoding="UTF-8"?><!DOCTYPE article  PUBLIC "-//NLM//DTD Journal Publishing DTD v3.0 20080202//EN" "http://dtd.nlm.nih.gov/publishing/3.0/journalpublishing3.dtd"><article xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" dtd-version="3.0" xml:lang="en" article-type="research article"><front><journal-meta><journal-id journal-id-type="publisher-id">JMP</journal-id><journal-title-group><journal-title>Journal of Modern Physics</journal-title></journal-title-group><issn pub-type="epub">2153-1196</issn><publisher><publisher-name>Scientific Research Publishing</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="doi">10.4236/jmp.2012.39143</article-id><article-id pub-id-type="publisher-id">JMP-22682</article-id><article-categories><subj-group subj-group-type="heading"><subject>Articles</subject></subj-group><subj-group subj-group-type="Discipline-v2"><subject>Physics&amp;Mathematics</subject></subj-group></article-categories><title-group><article-title>
 
 
  Nonabelian Dualization of Plane Wave Backgrounds
 
</article-title></title-group><contrib-group><contrib contrib-type="author" xlink:type="simple"><name name-style="western"><surname>adislav</surname><given-names>Hlavatý</given-names></name><xref ref-type="aff" rid="aff1"><sup>1</sup></xref><xref ref-type="corresp" rid="cor1"><sup>*</sup></xref></contrib><contrib contrib-type="author" xlink:type="simple"><name name-style="western"><surname>Miroslav</surname><given-names>Turek</given-names></name><xref ref-type="aff" rid="aff1"><sup>1</sup></xref><xref ref-type="corresp" rid="cor1"><sup>*</sup></xref></contrib></contrib-group><aff id="aff1"><addr-line>Czech Technical University in Prague, Faculty of Nuclear Sciences and Physical Engineering, B?ehová 7, 115 19 Prague 1, Czech Republic</addr-line></aff><author-notes><corresp id="cor1">* E-mail:<email>hlavaty@fjfi.cvut.cz(AH)</email>;<email>turekm@km1.fjfi.cvut.cz(MT)</email>;</corresp></author-notes><pub-date pub-type="epub"><day>24</day><month>09</month><year>2012</year></pub-date><volume>03</volume><issue>09</issue><fpage>1088</fpage><lpage>1095</lpage><history><date date-type="received"><day>June</day>	<month>30,</month>	<year>2012</year></date><date date-type="rev-recd"><day>July</day>	<month>31,</month>	<year>2012</year>	</date><date date-type="accepted"><day>August</day>	<month>10,</month>	<year>2012</year></date></history><permissions><copyright-statement>&#169; Copyright  2014 by authors and Scientific Research Publishing Inc. </copyright-statement><copyright-year>2014</copyright-year><license><license-p>This work is licensed under the Creative Commons Attribution International License (CC BY). http://creativecommons.org/licenses/by/4.0/</license-p></license></permissions><abstract><p>
 
 
  We investigate plane-parallel wave metrics from the point of view of their (Poisson-Lie) T-dualizability. For that purpose we reconstruct the metrics as backgrounds of nonlinear sigma models on Lie groups. For construction of dual backgrounds we use Drinfel’d doubles obtained from the isometry groups of the metrics. We find dilaton fields that enable to satisfy the vanishing beta equations for the duals of the homogenous plane-parallel wave metric. Torsion potentials or 
  B-fields, invariant w.r.t. the isometry group of Lobachevski plane waves are obtained by the Drinfel’d double construction. We show that a certain kind of plurality, different from the (atomic) Poisson-Lie T-plurality, may exist in case that metrics admit several isometry subgroups having the dimension of the Riemannian manifold. An example of that are two different backgrounds dual to the homogenous plane-parallel wave metric.
 
</p></abstract><kwd-group><kwd>Sigma Model; String Duality; pp-Wave Background</kwd></kwd-group></article-meta></front><body><sec id="s1"><title>1. Introduction</title><p>Sigma models can serve as models of string theory in curved and time-dependent backgrounds. Solution of sigma-models in such backgrounds is often very complicated, not to say impossible. On the other hand, there are many backgrounds whose properties were thoroughly investigated and it is therefore interesting to find if they can be transformed to some others. Important example of such transformation is so called Poisson Lie T-duality.</p><p>In their seminal work [<xref ref-type="bibr" rid="scirp.22682-ref1">1</xref>] Klimč&#237;k and Ševera set conditions for dualizability of backgrounds and gave formulas for their transformation. Since then several examples of dualizable sigma models were constructed, see e.g. [2- 4]. Unfortunately, most of the examples are not physically interesting. The purpose of this paper is to show that physical backgrounds that admit sufficiently large group of isometries are naturally dualizable and therefore equivalent in a sense to some others. In this paper we are going to investigate four-dimensional plane-parallel wave metrics [5-8] from this point of view.</p><p>The basic concept used for construction of dualizable sigma models is Drinfel’d double-Lie group with additional structure. The Drinfel’d double for a sigma model living in curved background can sometimes be found from the knowledge of symmetry group of the metric. More precisely, in the Drinfel’d double there are two equally dimensional subgroups whose Lie algebras are isotropic subspaces of the Lie algebra of the Drinfel’d double. In case that the metric has sufficient number of independent Killing vectors, the isometry group of the metric (or its subgroup) can be taken as one of the subgroups of the Drinfel’d double. The other one then must be chosen abelian in order to satisfy the conditions of dualizability. Short summary of the dualization procedure described e.g. in [<xref ref-type="bibr" rid="scirp.22682-ref9">9</xref>] is given in the next section.</p></sec><sec id="s2"><title>2. Elements of Poisson-Lie T-Dual Sigma-Models</title><p>Let G be a Lie group and <img src="28-7500792\ff80e53d-7b08-470a-89af-0f18a3c6e152.jpg" /> its Lie algebra. Sigma model on the group G is given by the classical action&#160;</p><disp-formula id="scirp.22682-formula72291"><label>(1)</label><graphic position="anchor" xlink:href="28-7500792\2a92914e-07d4-4e6a-923d-d41756db6779.jpg"  xlink:type="simple"/></disp-formula><p>where F is a second order tensor field on the Lie group G. The functions <img src="28-7500792\6492a250-3212-4d89-9667-a50b31bba622.jpg" /> are determined by the composition <img src="28-7500792\54c16dce-763a-4cff-bed2-bacfdef36b24.jpg" /> where <img src="28-7500792\1ef9c813-3903-4f7f-b7af-4181a42fc26d.jpg" /> and <img src="28-7500792\82f9f5ef-9a14-4e49-8e8c-e85f3f006540.jpg" /> are components of a coordinate map of neighborhood <img src="28-7500792\5c568908-10e7-48a3-b24d-c28a6a47688b.jpg" /> of element<img src="28-7500792\7e29442e-1493-454c-9800-8a1a02cd8806.jpg" />.</p><p>Equivalently the action can be expressed as&#160;</p><disp-formula id="scirp.22682-formula72292"><label>(2)</label><graphic position="anchor" xlink:href="28-7500792\7b4927be-a87f-4bb6-a29b-cc462dfa44c8.jpg"  xlink:type="simple"/></disp-formula><p>where <img src="28-7500792\2a7ee2b5-113f-4dad-9532-b337ca1b405b.jpg" /> are right-invariant fields</p><p><img src="28-7500792\715a9656-e371-4554-831c-8e26e8fa2061.jpg" />. The relationship between E and F is given by the formula</p><disp-formula id="scirp.22682-formula72293"><label>(3)</label><graphic position="anchor" xlink:href="28-7500792\95a306d8-3c38-4a86-a580-4ec17657f51f.jpg"  xlink:type="simple"/></disp-formula><p>where <img src="28-7500792\26e1ec03-7862-4db2-acea-af64ee3f7a66.jpg" /> are the components of right invariant forms<img src="28-7500792\93dbfbe4-7c7c-4594-b90a-9fb8ccf84a9d.jpg" />. The equations of motion derived from the action (1) have the following form&#160;</p><disp-formula id="scirp.22682-formula72294"><label>(4)</label><graphic position="anchor" xlink:href="28-7500792\684493b8-3169-4ed7-b99b-728aaaebd9ce.jpg"  xlink:type="simple"/></disp-formula><p>where <img src="28-7500792\5b810465-1b9e-4bee-b054-3eecfa1fe94a.jpg" /> are components of the Levi-Civita connection associated with the second order tensor field F This tensor field is a composition of the metric (a symmetric part) and the torsion potential (an antisymmetric part). The condition of dualizability of sigma-models on the level of the Lagrangian is given by the formula [<xref ref-type="bibr" rid="scirp.22682-ref1">1</xref>]&#160;</p><disp-formula id="scirp.22682-formula72295"><label>(5)</label><graphic position="anchor" xlink:href="28-7500792\2e7a9406-0aee-4afd-8c84-3512095a11d9.jpg"  xlink:type="simple"/></disp-formula><p>where <img src="28-7500792\b1f2fbea-21ac-4fbd-bc8b-5eac8a3e5726.jpg" /> are structure coefficients of the dual algebra <img src="28-7500792\06ad3779-c3d6-45f4-b475-5a6068a54ae5.jpg" /> and <img src="28-7500792\655e5b66-c9b8-4375-b65f-c0ee8a1cee8f.jpg" /> are left-invariant fields on the Lie group G. The algebras <img src="28-7500792\8e08271c-ad18-4823-aa76-ab707a4b5644.jpg" /> and <img src="28-7500792\a9efafe8-a7b2-4e6c-a993-f6f3e7ade4ad.jpg" /> then define the Drinfel’d double that enables to construct tensor F satisfying (5).</p><sec id="s2_1"><title>2.1. The Drinfel’d Double and Poisson-Lie T-Duality</title><p>As mentioned in the Introduction the Drinfel’d double D is defined as a connected Lie group whose Lie algebra <img src="28-7500792\9106ae5d-edc3-4279-93bc-743bceaee34e.jpg" /> can be decomposed into pair of subalgebras<img src="28-7500792\8bc9076c-1d81-4529-9d47-71ee0d2aca16.jpg" />, <img src="28-7500792\fe85dfd3-a4b6-4873-a85c-0d7ffe9946b8.jpg" />maximally isotropic with respect to a symmetric ad-invariant nondegenarate bilinear form <img src="28-7500792\f356a1c9-01cd-4531-8e27-a513205b231b.jpg" /> on<img src="28-7500792\cad73dd2-295a-4d78-aafc-935ba029e472.jpg" />.</p><p>Under the condition (5) the field Equations (4) for the <img src="28-7500792\c555441c-d57a-4033-8977-aa77eb23ceaa.jpg" />-model can be rewritten as equation for the mapping <img src="28-7500792\2c58141f-0340-499f-b546-697f5a43ca9d.jpg" /> from the world-sheet <img src="28-7500792\ebf3e3e6-e1a0-4a31-ab6b-5931c6b7548c.jpg" /> into the Drinfel’d double D&#160;</p><disp-formula id="scirp.22682-formula72296"><label>(6)</label><graphic position="anchor" xlink:href="28-7500792\9e1dc2cb-4980-4135-9dfc-bfb971822cb7.jpg"  xlink:type="simple"/></disp-formula><p>where subspaces<img src="28-7500792\cd3525fa-19fb-48d0-82ba-245676d76ff6.jpg" />,</p><p><img src="28-7500792\498ef36a-6178-47bb-8664-aaab7e1aea50.jpg" />are orthogonal w.r.t. <img src="28-7500792\dbd7eddf-f9de-4a9e-b9fb-281f28bc4295.jpg" />and span the whole Lie algebra<img src="28-7500792\17d08256-9692-4a2d-91e8-f7d66950021a.jpg" />.<img src="28-7500792\ebe92b33-bced-4f99-8506-50738f7919d1.jpg" />, <img src="28-7500792\fb4a7452-5722-442c-98c7-b88d362bd9d7.jpg" />are the bases of <img src="28-7500792\e935776f-16c1-41d0-9f2a-e9d6d4f58944.jpg" /> and<img src="28-7500792\21c4cbf9-9d08-46b2-b7f9-1b812b7c6df9.jpg" />.</p><p>Due to Drinfel’d, there exists unique decomposition (at least in the vicinity of the unit element of D) of an arbitrary element l of D as a product of elements from <img src="28-7500792\4e39983e-5ee6-4de7-9dd2-4b747729afbf.jpg" /> and<img src="28-7500792\416fa3be-9969-4cbf-bef9-4af5d555b2ec.jpg" />. The solutions of Equation (6) and solution <img src="28-7500792\dc39f55e-b1e4-4ee1-9686-24d1e7e98402.jpg" /> of the Equation (4) are related by&#160;</p><disp-formula id="scirp.22682-formula72297"><label>(7)</label><graphic position="anchor" xlink:href="28-7500792\18b12fb6-9aec-4d35-b67f-f3ee39bd5c1f.jpg"  xlink:type="simple"/></disp-formula><p>where<img src="28-7500792\98e71e54-dd70-4c78-9bf9-8b6a8b4ec3c2.jpg" />, <img src="28-7500792\7a367c16-882f-4b78-9737-9d0853724511.jpg" />fulfil the equations&#160;</p><disp-formula id="scirp.22682-formula72298"><label>(8)</label><graphic position="anchor" xlink:href="28-7500792\252c18d8-0610-4003-938b-7e61d9a96ec9.jpg"  xlink:type="simple"/></disp-formula><disp-formula id="scirp.22682-formula72299"><label>(9)</label><graphic position="anchor" xlink:href="28-7500792\5aaa7970-dbf3-4e15-bc1e-f4ac78e6cd72.jpg"  xlink:type="simple"/></disp-formula><p>The matrix <img src="28-7500792\3efa7eae-dc6b-468e-8b2b-c7c07e016129.jpg" /> of the dualizable <img src="28-7500792\961a6327-d4de-4c46-93b3-e47fca964407.jpg" />-model is of the form</p><disp-formula id="scirp.22682-formula72300"><label>(10)</label><graphic position="anchor" xlink:href="28-7500792\d84365a5-f82a-4bf8-ae09-72e2714f4f04.jpg"  xlink:type="simple"/></disp-formula><p>where <img src="28-7500792\71ceb72d-cae4-4513-bcfc-f173402230ef.jpg" /> is a constant matrix, <img src="28-7500792\782c82b8-9af1-4e96-aac4-6c24f673f456.jpg" />is given by the formula&#160;</p><disp-formula id="scirp.22682-formula72301"><label>(11)</label><graphic position="anchor" xlink:href="28-7500792\167d6583-c274-42e0-a44e-0af1331646d7.jpg"  xlink:type="simple"/></disp-formula><p>and matrices<img src="28-7500792\277e3856-8c8a-42bc-9e3f-4e9f68b84afe.jpg" />, <img src="28-7500792\8bb90557-1d60-4c15-9576-6901ba0420b9.jpg" />, <img src="28-7500792\b43163e2-6c11-4e70-8544-47097651553b.jpg" />are given by the adjoint representation of the Lie subgroup G on the Lie algebra of the Drinfel’d double in the basis <img src="28-7500792\a17ffe39-2ca8-42b2-a21a-6e7d8d104377.jpg" /><sup>1</sup></p><disp-formula id="scirp.22682-formula72302"><label>(12)</label><graphic position="anchor" xlink:href="28-7500792\5dcda895-444f-4c57-8b7f-bd0ed51cc6fd.jpg"  xlink:type="simple"/></disp-formula><p>Let us note that <img src="28-7500792\f910fd61-e52b-4f46-82bc-78dbac613391.jpg" /> is the value of <img src="28-7500792\be90f3a4-8045-4e16-9a21-49c9e453a68d.jpg" /> in the unit <img src="28-7500792\0d81184a-9df5-4f96-8d9e-5f50fb495a6f.jpg" /> of the group <img src="28-7500792\539c5e11-c9f4-40f9-8caa-0ebf9801db1c.jpg" /> because<img src="28-7500792\0d609c4b-610f-4582-bb04-a1080f7ce5ac.jpg" />.</p><p>The dual model can be obtain by the exchange&#160;</p><disp-formula id="scirp.22682-formula72303"><label>(13)</label><graphic position="anchor" xlink:href="28-7500792\20b0180b-d677-4928-b964-3af000b3be9f.jpg"  xlink:type="simple"/></disp-formula><p>Solutions of the equations of motion of dual models are mutually associated by the relation&#160;</p><disp-formula id="scirp.22682-formula72304"><label>(14)</label><graphic position="anchor" xlink:href="28-7500792\26a872d1-c3a8-42f8-a1aa-7c6d49e02631.jpg"  xlink:type="simple"/></disp-formula></sec><sec id="s2_2"><title>2.2. Poisson-Lie T-Plurality</title><p>Generally, more than two decompositions<sup>2</sup> (Manin triples) of Lie algebra <img src="28-7500792\7ef23575-cc2a-49f3-8534-72116a0ba4b1.jpg" /> of the Drinfel’d double can exist. This possibility leads to Poisson-Lie T-plurality. Let <img src="28-7500792\79bc0533-572c-486f-82b9-105e8ca33f53.jpg" /> is another decomposition of the Drinfel’d algebra <img src="28-7500792\af1df5df-3a49-4d7d-a8fc-23937656ff5b.jpg" /> into a pair of maximal isotropic subalgebras. Then the Poisson-Lie T-plural sigma model is given by the following formulas [<xref ref-type="bibr" rid="scirp.22682-ref10">10</xref>]&#160;</p><disp-formula id="scirp.22682-formula72305"><label>(15)</label><graphic position="anchor" xlink:href="28-7500792\67f30bff-a6f9-407f-9b0d-9db2061a3bc9.jpg"  xlink:type="simple"/></disp-formula><disp-formula id="scirp.22682-formula72306"><label>(16)</label><graphic position="anchor" xlink:href="28-7500792\10e03b8f-9c90-4025-b443-4f25cf38060b.jpg"  xlink:type="simple"/></disp-formula><p>where the matrices <img src="28-7500792\03ad8279-e37a-47db-8394-64758fa39f07.jpg" /> determine the relationship between the bases of the appropriate decompositions <img src="28-7500792\a6d66300-9d13-456c-99d1-93a7654661d1.jpg" /> and<img src="28-7500792\96c2a62e-9b8b-4e8e-b304-322f267d358d.jpg" />&#160;</p><disp-formula id="scirp.22682-formula72307"><label>(17)</label><graphic position="anchor" xlink:href="28-7500792\5f01ad23-f5c2-451e-857b-5aa1669ccc2f.jpg"  xlink:type="simple"/></disp-formula><p>The relationship between the classical solutions of the two Poisson-Lie T-plural sigma-models is given by a possibility of two decompositions of the element <img src="28-7500792\9d5619cd-1bbb-48a0-8827-748f8f9f7584.jpg" /> as&#160;</p><disp-formula id="scirp.22682-formula72308"><label>(18)</label><graphic position="anchor" xlink:href="28-7500792\f5e217a6-3f03-4b0c-b0d4-cba077e2d4a3.jpg"  xlink:type="simple"/></disp-formula><p>The Poisson-Lie T-duality is then a special case of Poisson-Lie T-plurality for<img src="28-7500792\1f2a4d2d-132e-45f7-8aff-6ddba9a5bd0e.jpg" />.</p></sec></sec><sec id="s3"><title>3. Homogenous Plane Wave Metrics</title><p>Homogenous plane wave is generally defined by the metric of the following form [5,6]</p><disp-formula id="scirp.22682-formula72309"><label>(19)</label><graphic position="anchor" xlink:href="28-7500792\7a51dc07-3d7c-4c1e-85a0-3d6dc68335c2.jpg"  xlink:type="simple"/></disp-formula><p>where <img src="28-7500792\c5836b55-10df-4617-a0a2-7a1e7b6bf30d.jpg" /> is the standard metrics on Euclidean space <img src="28-7500792\9dd7e89d-d926-4f6a-854d-1fbc705b6058.jpg" /> and <img src="28-7500792\e7806caf-0264-4a63-989c-a6ca094fd93a.jpg" /> The form of this metric seems to be simple, but explicit construction of sigma models can be very complicated. Therefore, we have focused on the special case of isotropic homogenous plane wave metric<img src="28-7500792\d46aedaa-f90c-4969-aac7-21a170c5f3dd.jpg" />&#160;</p><disp-formula id="scirp.22682-formula72310"><label>(20)</label><graphic position="anchor" xlink:href="28-7500792\1e5aa6ab-9965-449b-a9e7-e6af10c94613.jpg"  xlink:type="simple"/></disp-formula><p>Metric (20) has a number of symmetries important for the construction of the dualizable sigma models. It admits the following Killing vectors&#160;</p><disp-formula id="scirp.22682-formula72311"><label>(21)</label><graphic position="anchor" xlink:href="28-7500792\eb87c325-5c3c-4f05-842d-3923f47b0fb8.jpg"  xlink:type="simple"/></disp-formula><p>where <img src="28-7500792\f4c58f1e-1f68-4107-b2a4-85b88054ade4.jpg" /> satisfies&#160;</p><disp-formula id="scirp.22682-formula72312"><label>(22)</label><graphic position="anchor" xlink:href="28-7500792\94b38722-ae9b-48aa-bd2b-8fba95f8464a.jpg"  xlink:type="simple"/></disp-formula><p>The Killing vectors <img src="28-7500792\0320703c-7c5f-4a6f-b10f-10e8355396e6.jpg" /> are generators of orthogonal rotations in<img src="28-7500792\0f3ed8b6-f8cb-43a7-b1cc-444e5405492d.jpg" />. For special choice of&#160;</p><disp-formula id="scirp.22682-formula72313"><label>(23)</label><graphic position="anchor" xlink:href="28-7500792\e823cb90-921f-48dd-bb14-030235ae3def.jpg"  xlink:type="simple"/></disp-formula><p>there are further isometries related to the scaling of the light-cone coordinates&#160;</p><disp-formula id="scirp.22682-formula72314"><label>(24)</label><graphic position="anchor" xlink:href="28-7500792\5c8d21db-9a1d-487f-a8f0-5769b9494fc9.jpg"  xlink:type="simple"/></disp-formula><p>The specific form of <img src="28-7500792\407f7031-cedc-4c49-a3c4-34366aa6a9f8.jpg" /> enables us to calculate the function <img src="28-7500792\044c0541-d445-4529-9d9e-3ae5f15dc522.jpg" /> explicitly. The Killing vectors of the metric (20) for <img src="28-7500792\af367f8f-cabe-4183-bcfa-90ebeab742af.jpg" /> are&#160;</p><disp-formula id="scirp.22682-formula72315"><label>(25)</label><graphic position="anchor" xlink:href="28-7500792\5617776c-1e42-4706-82d4-286c328052fd.jpg"  xlink:type="simple"/></disp-formula><p>where D is the generator associated with the scaling symmetry and<img src="28-7500792\be05e2ba-6d33-4eb5-9da2-a118ef054bd3.jpg" />.</p><p>In the following we shall investigate the case<img src="28-7500792\93bab893-1682-46a0-a0f8-aab95772dfa7.jpg" />. It means that the metric tensor in coordinates <img src="28-7500792\bead1603-dcb4-4d31-916e-7bd5635821ca.jpg" /> reads&#160;</p><disp-formula id="scirp.22682-formula72316"><label>(26)</label><graphic position="anchor" xlink:href="28-7500792\154362ca-fff7-4ed5-99b8-45b9b81c2d22.jpg"  xlink:type="simple"/></disp-formula><p>This metric is not flat but its Gaussian curvature vanishes. Note that it has singularity in<img src="28-7500792\011b6ac2-0f51-48a3-b656-d8eed39cfb4a.jpg" />. It does not satisfy the Einstein equations but the conformal invariance conditions equations for vanishing of the <img src="28-7500792\d2b7ea06-d65f-45fb-bbff-59f90cd2dd76.jpg" />-function</p><disp-formula id="scirp.22682-formula72317"><label>(27)</label><graphic position="anchor" xlink:href="28-7500792\b0f423d4-27ea-4ca4-85bb-0afc2c61ccb6.jpg"  xlink:type="simple"/></disp-formula><disp-formula id="scirp.22682-formula72318"><label>(28)</label><graphic position="anchor" xlink:href="28-7500792\32e8978d-6fa9-4b43-a233-6861fe68f443.jpg"  xlink:type="simple"/></disp-formula><disp-formula id="scirp.22682-formula72319"><label>(29)</label><graphic position="anchor" xlink:href="28-7500792\002b6430-1c7d-4ea6-a37e-c5c0be92d35f.jpg"  xlink:type="simple"/></disp-formula><p>where the covariant derivatives<img src="28-7500792\9fb29201-4b77-454c-b0a0-c3f5ad4dea37.jpg" />, Ricci tensor <img src="28-7500792\d2ecfbd9-cbe3-4382-9688-b7073c7d2754.jpg" /> and Gauss curvature <img src="28-7500792\63bac1ca-9862-431c-85a1-cd02a3a66fb6.jpg" /> are calculated from the metric <img src="28-7500792\7ce2cd4f-c22d-45a2-aa98-a94cadff4cf2.jpg" /> that is also used for lowering and raising indices. Torsion <img src="28-7500792\636ffadc-99e2-4e1a-a5a4-40b5e7fd3002.jpg" /> in this case vanishes and dilaton field is [<xref ref-type="bibr" rid="scirp.22682-ref5">5</xref>]</p><disp-formula id="scirp.22682-formula72320"><label>(30)</label><graphic position="anchor" xlink:href="28-7500792\0616dbbe-e85e-42f8-b117-a8aceb2d9ef5.jpg"  xlink:type="simple"/></disp-formula><p>The metric (26) admits the following Killing vectors<sup>3</sup></p><disp-formula id="scirp.22682-formula72321"><label>(31)</label><graphic position="anchor" xlink:href="28-7500792\0578d236-5f62-4ecc-9ed8-c14a5ae89c5b.jpg"  xlink:type="simple"/></disp-formula><p>One can easily check that the Lie algebra spanned by these vectors is the semidirect sum <img src="28-7500792\2b8e5878-1078-4fce-8c3a-bc13ed9b0135.jpg" /> where <img src="28-7500792\197be0f4-57bb-4fba-81f4-07b463ff2bce.jpg" /> and ideal <img src="28-7500792\8cda187f-ecf9-4025-a80f-8ff6016a6c76.jpg" />. The algebra <img src="28-7500792\35f72064-2f0d-4cb3-ae62-0fd5295f4205.jpg" /> is abelian and its generators can be interpreted as dilation in <img src="28-7500792\db36cd61-5871-4d5a-83a2-cf16941908da.jpg" /> and rotation in<img src="28-7500792\5b4a4ae4-e9e7-478f-b4de-1b07fa9655f8.jpg" />. Generators of the algebra <img src="28-7500792\e690c35c-d753-4e87-a2a2-f07ef6c25d67.jpg" /> commute as two-dimensional Heisenberg algebra with the center<img src="28-7500792\655b8523-1d2f-48ee-bebe-d018cf5100fd.jpg" />.</p>Construction of Dual Metrics<p>As explained in Section 2, dualizable metric can be constructed by virtue of Drinfel’d double. For this goal the Lie algebra <img src="28-7500792\0e2608b1-9055-45c6-995e-27a55e3bf9d6.jpg" /> of the Drinfel’d double can be composed from the four-dimensional Lie subalgebra <img src="28-7500792\2d4780a6-db7e-478a-baaa-0e75e68d0de5.jpg" /> isomorphic to the four-dimensional subalgebra of Killing vectors and four-dimensional Abelian algebra<sup>4<img src="28-7500792\1950501e-8693-4db7-9970-b5e0eb1ee199.jpg" /></sup>. Moreover, the four-dimensional subgroup of isometries must act freely and transitively [<xref ref-type="bibr" rid="scirp.22682-ref1">1</xref>] on the Riemannian manifold M where the metric (20, 23) is defined so that<img src="28-7500792\20aaff29-0bc8-46c8-a9ab-984ed6242a1e.jpg" />.</p><p>Using the method described in [<xref ref-type="bibr" rid="scirp.22682-ref11">11</xref>] for semisimple algebras we find that up to the transformation<img src="28-7500792\016e8eeb-44c9-4c6d-84fa-2632d519bc27.jpg" />, i.e. <img src="28-7500792\4ad7f854-c2fb-4282-af32-a815c1120043.jpg" />there are six classes of four-dimensional subalgebras of the isometry algebra of the homogeneous plane wave metric isomorphic to&#160;</p><p>• <img src="28-7500792\f82a250d-d014-487e-afaa-8c297e900ecf.jpg" /></p><p>• <img src="28-7500792\88769170-42bc-4050-b2c0-b6276e4470a5.jpg" /></p><p>• <img src="28-7500792\a093d42d-6368-48af-a35d-003391e3ca97.jpg" /></p><p>• <img src="28-7500792\70105547-ff69-4453-b5ed-b5fae0c8e922.jpg" /></p><p>• <img src="28-7500792\e1a729d7-73df-45a4-94ea-614c3f3415d0.jpg" /></p><p><img src="28-7500792\69528f7d-3a59-457a-90c4-2198f10f254b.jpg" /></p><p>where <img src="28-7500792\8baca0f0-653e-4fcc-8072-0bc27779c7af.jpg" /> are arbitrary parameters.</p><p>Infinitesimal form of transitivity condition can be formulated as requirement that four independent Killing vectors can be taken as basis vectors of four-dimensional vector distribution in M. In other words, these Killing vectors must form a basis of tangent space in every point of M. It means that in every point of M there is an invertible matrix <img src="28-7500792\0eceb64e-4f0f-4238-8d98-45c6213d8e24.jpg" /> that solves the equation</p><disp-formula id="scirp.22682-formula72322"><label>(32)</label><graphic position="anchor" xlink:href="28-7500792\d63e4f07-e0f2-494c-af23-d9321cc8b0fc.jpg"  xlink:type="simple"/></disp-formula><p>where<img src="28-7500792\e87d978a-a6be-4ee7-a63b-92e84301f62f.jpg" />, <img src="28-7500792\381ee175-952b-42d5-a6eb-5529d857ef81.jpg" />and <img src="28-7500792\ae126531-a6bb-4cf8-88a5-de9452f04b55.jpg" /> form a basis of the subalgebra.</p><p>Infinitesimal form of requirement that the action of the isometry subgroup is free says that if in any point of M there is a vector of the corresponding Lie subalgebra such that its action on the point vanishes then it must be null vector.</p><p>By inspection we can find that the only four-dimensional subalgebras that generate transitive actions on M are isomorphic to <img src="28-7500792\72d1431c-a67f-490c-aadf-1f47af149122.jpg" /> or <img src="28-7500792\eb0c2530-6be4-4d68-b9bd-97b92c92007f.jpg" />. Their non-vanishing commutation relations are&#160;</p><disp-formula id="scirp.22682-formula72323"><label>(33)</label><graphic position="anchor" xlink:href="28-7500792\0ea24a10-0592-436c-81c6-0bb05648f515.jpg"  xlink:type="simple"/></disp-formula><p>and</p><disp-formula id="scirp.22682-formula72324"><label>(34)</label><graphic position="anchor" xlink:href="28-7500792\c5546018-8ed9-4b3c-80c5-3eae76428312.jpg"  xlink:type="simple"/></disp-formula><p>respectively where <img src="28-7500792\f9f26681-8993-4beb-976d-4b6e5e8200f0.jpg" /> and <img src="28-7500792\66d64df5-b115-48cb-b650-bb90f3903ecb.jpg" /> are real parameters. One can also check that the action of both corresponding groups of isometries is free. In the following we shall find metric dual to (26) that follows from its Drinfel’d double description where <img src="28-7500792\e5a8ddd0-acf1-4bc7-8420-2c1bfbc83b13.jpg" /> is isomorphic either to algebra spanned by <img src="28-7500792\6c4d908e-f87c-4c49-a5e0-3ea3995c11d9.jpg" /> or by <img src="28-7500792\938137b7-442c-41f6-9091-db20079ec56e.jpg" />.</p><p>Let us start with construction of the Drinfel’d double following from the algebra isomorphic to (33) and dual Abelian algebra. Assume that the Lie algebra <img src="28-7500792\05607a46-244c-4bc8-8929-d77705356086.jpg" /> is spanned by elements <img src="28-7500792\ae44df65-9049-4d77-b553-1d3b47b2843b.jpg" /> with commutation relations&#160;</p><disp-formula id="scirp.22682-formula72325"><label>(35)</label><graphic position="anchor" xlink:href="28-7500792\d40ee303-859b-4512-ac76-b033dd7c97cc.jpg"  xlink:type="simple"/></disp-formula><p>where <img src="28-7500792\9941bd63-88e9-47c3-a158-f779a3404f7c.jpg" /> and <img src="28-7500792\575ec055-1a04-4764-8095-b30093a84af9.jpg" /> are arbitrary real parameters. The basis of left-invariant vector fields of the group generated by <img src="28-7500792\1f3c8ab7-cecf-403d-8fe0-0dfa9979ceb8.jpg" /> is</p><disp-formula id="scirp.22682-formula72326"><label>(36)</label><graphic position="anchor" xlink:href="28-7500792\8a2a9a6f-65f7-4f30-863a-a02aa2e0dab2.jpg"  xlink:type="simple"/></disp-formula><p>where <img src="28-7500792\14fec579-10e5-470b-8143-5a6ae791faea.jpg" /> are group coordinates used in parametrization&#160;</p><disp-formula id="scirp.22682-formula72327"><label>(37)</label><graphic position="anchor" xlink:href="28-7500792\ae2b6570-a816-4027-bde0-aa28e044ff84.jpg"  xlink:type="simple"/></disp-formula><p>To be able to obtain the metric (26) by the Drinfel’d double construction first we have to transform it into the group coordinates. Transformation between group coordinates x<sub>1</sub>, x<sub>2</sub>, x<sub>3</sub>, x<sub>4</sub>, and geometrical coordinates u, <img src="28-7500792\4e7934b8-cbc8-41d4-9144-b2cbaeaf615d.jpg" />, x, y is</p><disp-formula id="scirp.22682-formula72328"><label>(38)</label><graphic position="anchor" xlink:href="28-7500792\dbd8f27d-f5a8-4eff-9bc5-c1cbbdf9d05a.jpg"  xlink:type="simple"/></disp-formula><p>It converts the Killing vectors <img src="28-7500792\7977a5d8-9a35-462a-83d6-53c9965c7854.jpg" /> into the left-invariant vector fields (36) and the metric (26) into the form</p><disp-formula id="scirp.22682-formula72329"><label>(39)</label><graphic position="anchor" xlink:href="28-7500792\5e5b993d-3327-46e5-8c6b-929f92692e78.jpg"  xlink:type="simple"/></disp-formula><p>that is obtainable by (3) and (10). To get the matrix <img src="28-7500792\8e5257ba-d5b1-4067-b8f4-e727efe8cc44.jpg" /> necessary for construction of the dual model we note that it is given by the value of <img src="28-7500792\18c3e351-3531-428d-8d73-d365c9b40103.jpg" /> in the unit of the group, i.e. by value of <img src="28-7500792\fbbd8032-e729-4a6d-9645-37d9f36b2c1d.jpg" /> for<img src="28-7500792\cf029bd8-3de0-488c-84b0-699764991f7f.jpg" />.&#160;</p><disp-formula id="scirp.22682-formula72330"><label>(40)</label><graphic position="anchor" xlink:href="28-7500792\650f00dc-ec75-4dba-b335-ea5a515dc3e0.jpg"  xlink:type="simple"/></disp-formula><p>The dual tensor on the Abelian group <img src="28-7500792\5a6a9ebb-afde-445f-9e64-4182690bd791.jpg" /> constructed by the procedure explained in the Section 2, namely by using (3), (10) and (13) is</p><disp-formula id="scirp.22682-formula72331"><label>(41)</label><graphic position="anchor" xlink:href="28-7500792\7bc6d5b7-7030-48eb-863a-e17d6e8aa579.jpg"  xlink:type="simple"/></disp-formula><p>One can see that the dual tensor has also antisymmetric part (<img src="28-7500792\0b86ff2a-7dac-4ada-a0f6-995257f2f01b.jpg" />-field or torsion potential)&#160;</p><disp-formula id="scirp.22682-formula72332"><label>(42)</label><graphic position="anchor" xlink:href="28-7500792\0e3d0f1e-74b0-4e31-96f0-0c1f732ad539.jpg"  xlink:type="simple"/></disp-formula><p>and its torsion <img src="28-7500792\d12fc08a-a4d2-4478-aa22-baa60e41b360.jpg" /> is</p><disp-formula id="scirp.22682-formula72333"><label>(43)</label><graphic position="anchor" xlink:href="28-7500792\5c029d9f-b469-4af0-9c43-a59c43525d9a.jpg"  xlink:type="simple"/></disp-formula><p>The Gauss curvature of its symmetric part vanishes but the Ricci tensor is nontrivial. Dual metric that is symmetric part of (41) does not solve the Einstein equations either but again we can satisfy conformal invariance conditions (27)-(29) by the dilaton field</p><disp-formula id="scirp.22682-formula72334"><label>(44)</label><graphic position="anchor" xlink:href="28-7500792\0e059651-12dc-4475-a39b-b9f30de65e9e.jpg"  xlink:type="simple"/></disp-formula><p>If we use the subalgebra of isometries spanned by <img src="28-7500792\97b5b53f-c0a6-4862-adaf-5887c38efaad.jpg" /> instead of that spanned by <img src="28-7500792\ae8beae2-4905-4f4b-a94f-e7b24c8555ee.jpg" /> then the transformation between group coordinates x<sub>1</sub>, x<sub>2</sub>, x<sub>3</sub>, x<sub>4</sub>, and geometrical coordinates u, <img src="28-7500792\ba776898-53d1-4b38-98bf-bfd083f0a372.jpg" />, x, y is&#160;</p><disp-formula id="scirp.22682-formula72335"><label>(45)</label><graphic position="anchor" xlink:href="28-7500792\8cb96e6e-28b6-43d7-b5bd-471b4e1cd2dc.jpg"  xlink:type="simple"/></disp-formula><p>the matrix <img src="28-7500792\c352de52-59e0-4a0a-af1c-d5083cfa1385.jpg" /> gets again the form (40) and we get another tensor dual to (26)</p><disp-formula id="scirp.22682-formula72336"><label>(46)</label><graphic position="anchor" xlink:href="28-7500792\58f344de-ce80-4b42-9de9-f786487a7339.jpg"  xlink:type="simple"/></disp-formula><p>Even though it is not symmetric its torsion is zero. It satisfies the conformal invariance conditions (27)-(29) with the dilaton field</p><disp-formula id="scirp.22682-formula72337"><label>(47)</label><graphic position="anchor" xlink:href="28-7500792\7a1a5076-662c-45c5-ad3e-7551c9b13ff7.jpg"  xlink:type="simple"/></disp-formula></sec><sec id="s4"><title>4. Lobachevsky Plane Waves</title><p>Another type of metrics that have rather large group of isometries are so called Lobachevsky plane waves [7,8]. They are of general form</p><disp-formula id="scirp.22682-formula72338"><label>(48)</label><graphic position="anchor" xlink:href="28-7500792\639f028f-9521-40e6-9ffa-09a4a0f63926.jpg"  xlink:type="simple"/></disp-formula><p>They satisfy Einstein equation with cosmological constant <img src="28-7500792\eff7790b-5459-42ed-8a99-64684a0828a2.jpg" /> iff</p><disp-formula id="scirp.22682-formula72339"><label>(49)</label><graphic position="anchor" xlink:href="28-7500792\1031268d-990a-4acb-9226-01cc8b4dd6af.jpg"  xlink:type="simple"/></disp-formula><p>The Gauss curvature of this metric is<img src="28-7500792\83d38098-e63a-4793-a6b2-4e078bce9a41.jpg" />. For special forms of function H the metric (48) admits various sets of Killing vectors. All of them are subalgebras of a vector space spanned by</p><disp-formula id="scirp.22682-formula72340"><label>(50)</label><graphic position="anchor" xlink:href="28-7500792\71924193-24ae-40ce-8dde-6c8ef040e40b.jpg"  xlink:type="simple"/></disp-formula><p>A bit surprisingly, all these seven independent vector fields found in [<xref ref-type="bibr" rid="scirp.22682-ref7">7</xref>] form a Lie algebra even though they are not Killing vectors of the same metrics (it depends on the form of<img src="28-7500792\c5be21da-dc3b-4623-af6a-b743d9135ae5.jpg" />). We are interested in metrics that admit at least four independent Killing vectors because they can be interpreted as dualizable backgrounds for sigma models in four dimensions.</p><p>As mentioned in the Section 3.1, for construction of dualizable metrics we need a four-dimensional subalgebra of Killing vectors that generates group of isometries that acts freely and transitively on the four-dimensional Riemannian manifolds. Here we shall investigate metrics of the form (48) where that<img src="28-7500792\db3a3a6c-1e0f-416c-b6d5-59dcf0a0af25.jpg" />, i.e.</p><disp-formula id="scirp.22682-formula72341"><label>(51)</label><graphic position="anchor" xlink:href="28-7500792\7133ea68-960d-4170-bb0f-9a8b5caef966.jpg"  xlink:type="simple"/></disp-formula><p>It solves the Einstein equation with the cosmological constant <img src="28-7500792\6ce2818f-aca0-4c41-8ab8-97f2ca583866.jpg" /> for <img src="28-7500792\58873d46-fc57-4679-9fe6-a8650f22b8ad.jpg" /> [<xref ref-type="bibr" rid="scirp.22682-ref12">12</xref>].</p><sec id="s4_1"><title>4.1. Construction of the Dual Metric</title><p>The metric (51) has five-dimensional Lie group of isometries generated by the Killing vectors K<sub>I</sub>, K<sub>II</sub>, K<sub>III</sub>, K<sub>IV</sub>, K<sub>V</sub>, K<sub>VI</sub>. Their nonzero commutators read</p><disp-formula id="scirp.22682-formula72342"><label>(52)</label><graphic position="anchor" xlink:href="28-7500792\70120cd3-68d5-4a90-b918-7d1758b6e1fe.jpg"  xlink:type="simple"/></disp-formula><p>Four-dimensional subalgebras of the Lie algebra (52) for generic <img src="28-7500792\4d41d2a3-98c8-4ea5-a651-8baa93084bc0.jpg" /> are isomorphic to one of the following algebras:&#160;</p><p>• <img src="28-7500792\6625b158-db08-49b0-8cbd-8a246011c71c.jpg" /></p><p>• <img src="28-7500792\17265728-c98c-4460-9841-cb79a8e77cad.jpg" /></p><p><img src="28-7500792\227aa905-3b53-4d32-9fe1-095ad184a79f.jpg" /></p><p>It is easy to check that the only subalgebra of these that satisfy the condition of transitivity (32) in every point of M is the first one. Its action is free on M as well so that we can use it for dualization of the metric (51).</p><p>In the following we shall consider the case <img src="28-7500792\5f00fa5f-01a1-490e-bdf2-6462eaf80d77.jpg" /> because <img src="28-7500792\3d113261-ccbd-4146-a09a-fba87cafee49.jpg" /> do not bring anything qualitatively different. It means that for dualization we shall use the algebra <img src="28-7500792\621a5b99-9fc2-4e44-90d0-26b4e3bd0b28.jpg" /> spanned by K<sub>I</sub>, K<sub>III</sub>, K<sub>IV</sub>, K<sub>VI</sub> with nonzero commutation relations</p><disp-formula id="scirp.22682-formula72343"><label>(53)</label><graphic position="anchor" xlink:href="28-7500792\7fdd7925-7ca3-4148-bbd6-8fad68e9f709.jpg"  xlink:type="simple"/></disp-formula><p>The corresponding Drinfel’d double is generated by the algebra <img src="28-7500792\be8844ae-79e9-41cf-b4e6-a0b7a76394b7.jpg" /> defined by the commutation relations (53) and four-dimensional Abelian algebra. The basis of leftinvariant vector fields of the group generated by <img src="28-7500792\a6fc8de2-1b27-4721-9d5c-d266c714227b.jpg" /> is</p><disp-formula id="scirp.22682-formula72344"><label>(54)</label><graphic position="anchor" xlink:href="28-7500792\ae7ec555-f2a8-4eff-af8c-7f0e3954a353.jpg"  xlink:type="simple"/></disp-formula><p>where x<sub>1</sub>, x<sub>2</sub>, x<sub>3</sub>, x<sub>4</sub> are group coordinates used in parametrization</p><p><img src="28-7500792\ac76f7b1-818b-4145-9f4b-98636ef62877.jpg" /></p><p>and X<sub>1</sub>, X<sub>2</sub>, X<sub>3</sub>, X<sub>4</sub> are generators of <img src="28-7500792\540775dd-9c4b-453d-bff0-240174c99db9.jpg" /> satisfying</p><disp-formula id="scirp.22682-formula72345"><label>(55)</label><graphic position="anchor" xlink:href="28-7500792\88a7b7c5-97bb-4ea1-892e-0048bccebb29.jpg"  xlink:type="simple"/></disp-formula><p>Transformation between group coordinates and coordinates u, <img src="28-7500792\0dcc781f-bf7a-4c08-b751-1c3f0280901b.jpg" />, x, y of the Lobachevsky manifold is</p><disp-formula id="scirp.22682-formula72346"><label>(56)</label><graphic position="anchor" xlink:href="28-7500792\76e87683-5d90-4014-af64-8bb09bd45569.jpg"  xlink:type="simple"/></disp-formula><p>This transformation converts the Killing vectors K<sub>I</sub>, K<sub>III</sub>, K<sub>IV</sub>, K<sub>VI</sub> into the left-invariant vector fields (54) and the metric (51) into</p><p><img src="28-7500792\3d973cd5-62d7-42e9-b09e-fe76e8a2a536.jpg" /></p><p>The value of this metric for<img src="28-7500792\26b9d048-0a6a-4c49-855d-4cda4936b341.jpg" />, i.e. in the unit of the group, gives the matrix</p><disp-formula id="scirp.22682-formula72347"><label>(57)</label><graphic position="anchor" xlink:href="28-7500792\fbc6db9b-e937-4e92-b934-ad05acb3a92d.jpg"  xlink:type="simple"/></disp-formula><p>Having this matrix we can construct the dual tensor. It is again obtained using (3), (10) and (13) and has the form</p><p><img src="28-7500792\b8d9a7d7-9095-46c6-a4c5-7d6b277cc3d5.jpg" /></p><p>This tensor has nonzero and nonconstant Gauss curvature and torsion.</p></sec><sec id="s4_2"><title>4.2. B-Field</title><p>The Drinfel’d double construction enables to add the Bfield (torsion potential) to the metric so that the resulting tensor <img src="28-7500792\db7df4bd-1646-44b8-8288-ef98c1bead17.jpg" /> is invariant with respect to the same isometry group as the metric itself. Namely, changing <img src="28-7500792\18c9943d-a67a-423b-806a-fe9b88675206.jpg" /> to<sup>5</sup></p><disp-formula id="scirp.22682-formula72348"><label>(58)</label><graphic position="anchor" xlink:href="28-7500792\51d5e88f-6d14-42fe-a7c2-a5a81bfefbbd.jpg"  xlink:type="simple"/></disp-formula><p>and applying the formula (3), (10), we get covariant tensor that after the transformation (56) acquires the form</p><disp-formula id="scirp.22682-formula72349"><label>(59)</label><graphic position="anchor" xlink:href="28-7500792\26a792ab-712c-4ba3-adcd-7ebca1737905.jpg"  xlink:type="simple"/></disp-formula><p>Its symmetric part is the metric (51). This tensor is again invariant with respect to the isometry group generated by K<sub>I</sub>, K<sub>III</sub>, K<sub>IV</sub>, K<sub>VI</sub>. For <img src="28-7500792\b8acffd7-9508-4b3d-a8fc-b2fe748a48d1.jpg" /> the invariant group can be extended by the generator K<sub>V</sub>.</p><p>Torsion <img src="28-7500792\860e78a9-1d56-4423-9079-a9e112556d28.jpg" /> obtained from the antisymmetric part of <img src="28-7500792\6c2ab84f-f455-4fae-8b91-1c890e7e09d0.jpg" /> is</p><disp-formula id="scirp.22682-formula72350"><label>(60)</label><graphic position="anchor" xlink:href="28-7500792\0d02b165-b4d7-4da3-b245-b215db7de21c.jpg"  xlink:type="simple"/></disp-formula><p>As the tensor (59) was obtained by the Drinfel’d double construction it is possible to dualize it but the result is too extensive to display.</p></sec></sec><sec id="s5"><title>5. Conclusions</title><p>Isometry groups of metrics can be used for construction of their (nonabelian) T-dual backgrounds. Sufficient condition for that is that the metric have an isometry subgroup whose dimension is equal to the dimension of the Riemannian manifold and its action on the manifold is transitive and free.</p><p>We have shown that for the plane wave metrics (26) and (51) such isometry subgroups exist and the metrics can be dualized by the Poisson-Lie T-duality transformation. We have determined the metrics and B-fields dual to the plane waves. For homogeneous plane waves (26) we have also found the dilaton field that guarantees conformal invariance of the dual metric.</p><p>Metrics that possess isometry group whose dimension is greater than the dimension of the Riemannian manifold may have several duals. More precisely, if the metric admits various isometry subgroups with above given properties then we can construct several backgrounds dual to the metric. This phenomenon is another kind of plurality of sigma models different from the Poisson-Lie T-plurality described in the Section 2.</p><p>An example of this type of plurality is provided by the plane wave metric (26) with isometry subgroups generated by Killing vectors <img src="28-7500792\ead2a3e9-139b-4b56-8374-1deab0f6a58f.jpg" /> or by <img src="28-7500792\376c9da0-c735-4017-827b-9ebd37a363d1.jpg" /> (see (31) producing two dual backgrounds (41) and (46)). To decide if this plurality is different from the Poisson-Lie T-plurality one has to check whether the eight-dimensional Drinfel’d double s generated by the four-dimensional abelian algebra and algebras spanned by <img src="28-7500792\a780a67a-5824-479e-84cb-ad19cfadec0f.jpg" /> or <img src="28-7500792\c8f95689-a588-44ca-bf09-79c927ede2d4.jpg" /> are isomorphic by a transformation that leave the constant matrix (40) invariant. This is, however, very difficult task that might be investigated in the future.</p></sec><sec id="s6"><title>6. Acknowledgements</title><p>This work was supported by the research plan LC527 of the Ministry of Education of the Czech Republic. Consultation with P. Winternitz and L. Šnobl on classification of subalgebras are gratefully acknowledged.</p></sec><sec id="s7"><title>REFERENCES</title></sec><sec id="s8"><title>NOTES</title></sec></body><back><ref-list><title>References</title><ref id="scirp.22682-ref1"><label>1</label><mixed-citation publication-type="other" xlink:type="simple">C. Klim?ík and P. ?evera, “Dual Non-Abelian Duality and the Drinfeld Double,” Physics Letters B, 1995, pp. 455-462. </mixed-citation></ref><ref id="scirp.22682-ref2"><label>2</label><mixed-citation publication-type="other" xlink:type="simple">M. A. Lledo and V. S. Varadarajan, “SU(2) Poisson-Lie T-Duality,” Letters in Mathematical Physics, Vol. 45, No. 3, 1998, pp. 247-257. doi:10.1023/A:1007498803198</mixed-citation></ref><ref id="scirp.22682-ref3"><label>3</label><mixed-citation publication-type="other" xlink:type="simple">K. Sfetsos, “Poisson-Lie T-Duality beyond the Classical Level and the Renormalization Group,” Physics Letters B, Vol. 432, No. 3-4, 1998, pp. 365-375. 
doi:10.1016/S0370-2693(98)00666-2</mixed-citation></ref><ref id="scirp.22682-ref4"><label>4</label><mixed-citation publication-type="other" xlink:type="simple">L. Hlavaty and L. ?nobl, “Poisson-Lie T-Plurality of Three- Dimensional Conformally Invariant Sigma Models II: Nondiagonal Metrics and Dilatonpuzzle,” Journal of High Energy Physics, No. 10, 2004. </mixed-citation></ref><ref id="scirp.22682-ref5"><label>5</label><mixed-citation publication-type="other" xlink:type="simple">G. Papadopoulos, J. G. Russo and A. A. Tseytlin, “Solvable Model of Strings in a Time-Dependent Plane-Wave Background,” Classical and Quantum Gravity, pp. 969- 1016, [hep-th/0211289]. </mixed-citation></ref><ref id="scirp.22682-ref6"><label>6</label><mixed-citation publication-type="other" xlink:type="simple">M. Blau and M. O’Loughlin, “Homogeneous Plane Waves,” Nuclear Physics B, Vol. 652, No. 1-2, 2003, pp. 135-176.  
doi:10.1016/S0550-3213(03)00055-5</mixed-citation></ref><ref id="scirp.22682-ref7"><label>7</label><mixed-citation publication-type="other" xlink:type="simple">S. T. C. Siklos, “Lobatchewski Plane Gravitational Waves in Galaxies, Axisymmetric Systems and Relativity,” M. A. H. MacCallum, Ed., Cambridge University Press, Cambridge, 1985, p. 247. </mixed-citation></ref><ref id="scirp.22682-ref8"><label>8</label><mixed-citation publication-type="other" xlink:type="simple">J. Podolsky, “Interpretation of the Siklos Solutions as Exact Gravitational Waves in the Anti-De Sitter Universe,” Classical and Quantum Gravity, Vol. 15, No. 3, 1998, pp. 719-733. doi:10.1088/0264-9381/15/3/019</mixed-citation></ref><ref id="scirp.22682-ref9"><label>9</label><mixed-citation publication-type="other" xlink:type="simple">C. Klim?k, “Poisson-Lie T-Duality,” Nuclear Physics A, 1996, pp. 116-121, [hepth9509095]. </mixed-citation></ref><ref id="scirp.22682-ref10"><label>10</label><mixed-citation publication-type="other" xlink:type="simple">R. von Unge, “Poisson-Lie T-Plurality,” Journal of High Energy Physics, 2002, [hepth0205245]. </mixed-citation></ref><ref id="scirp.22682-ref11"><label>11</label><mixed-citation publication-type="other" xlink:type="simple">J. Patera, P. Winternitz and H. Zassenhaus, “Continuous Subgroups of the Fundamental Groups of Physics. I. General Method and the Poincaré Group,” Journal of Mathematical Physics, Vol. 16, No. 8, 1975, pp. 1597-1614.</mixed-citation></ref><ref id="scirp.22682-ref12"><label>12</label><mixed-citation publication-type="other" xlink:type="simple">V. R. Kaigorodov, “Einstein Spaces of Maximum Mobility,” Soviet Physics Doklady, Vol. 7, 1963, p. 893.</mixed-citation></ref></ref-list></back></article>