<?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">OJM</journal-id><journal-title-group><journal-title>Open Journal of Microphysics</journal-title></journal-title-group><issn pub-type="epub">2162-2450</issn><publisher><publisher-name>Scientific Research Publishing</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="doi">10.4236/ojm.2013.33010</article-id><article-id pub-id-type="publisher-id">OJM-35702</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>
 
 
  Matrix Quasi-Exactly Solvable Jacobi Elliptic Hamiltonian
 
</article-title></title-group><contrib-group><contrib contrib-type="author" xlink:type="simple"><name name-style="western"><surname>ncilla</surname><given-names>Nininahazwe</given-names></name><xref ref-type="aff" rid="aff1"><sub>1</sub></xref><xref ref-type="corresp" rid="cor1"><sup>*</sup></xref></contrib></contrib-group><aff id="aff1"><label>1</label><addr-line>Institut de Pédagogie Appliquée, Université du Burundi, Bujumbura, Burundi</addr-line></aff><author-notes><corresp id="cor1">* E-mail:<email>nininaha@yahoo.fr</email></corresp></author-notes><pub-date pub-type="epub"><day>09</day><month>08</month><year>2013</year></pub-date><volume>03</volume><issue>03</issue><fpage>53</fpage><lpage>59</lpage><history><date date-type="received"><day>April</day>	<month>11,</month>	<year>2013</year></date><date date-type="rev-recd"><day>May</day>	<month>12,</month>	<year>2013</year>	</date><date date-type="accepted"><day>May</day>	<month>20,</month>	<year>2013</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 construct a new example of 2 &#215; 2-matrix quasi-exactly solvable (QES) Hamiltonian which is associated to a poten
  tial depending on the Jacobi elliptic functions. We establish three necessary and sufficient algebraic conditions for the previous operator to have an invariant vector space whose generic elements are polynomials. This operator is called quasi-exactly solvable.
 
</p></abstract><kwd-group><kwd>Technology; Preference for Quality; Volume of Trade; Vertical Intra-Industry Trade</kwd></kwd-group></article-meta></front><body><sec id="s1"><title>1. Introduction</title><p>In quantum physics, one of the main mathematical problems consists in constructing the spectrum of a linear operator defined on a suitable domain of Hilbert space. In most cases, this type of problem cannot be explicitly solved, in other words the eigenvalues of the Hamiltonian cannot be computed algebraically. However, in few cases, some of which turn out to be physically fundamental, the spectrum can indeed be found explicitly. The two major examples of this kind are the celebrated harmonic quantum oscillator and the hydrogen atom (i.e. 3- dimensional Schr&#246;dinger equation coupled to an external Coulomb potential). These examples are called exactly solvable in the sense that the full spectrum of the Hamiltonian is found explicitly.</p><p>In the last few years, a new class of operators which is intermediate to exactly solvable and non solvable operators has been discovered [1-4]: the quasi-exactly solvable (QES) operators, for which a finite part of the spectrum can computed algebraically.</p><p>Although scalar QES operators have been classified in one variable [<xref ref-type="bibr" rid="scirp.35702-ref5">5</xref>] and in several variables [<xref ref-type="bibr" rid="scirp.35702-ref6">6</xref>], a classification of matrix QES operators is still missing.</p><p>More recently, interesting tools for classification of 2 &#215; 2-matrix QES operators in one spatial dimensional [7- 9] and in creation and annihilation operators [<xref ref-type="bibr" rid="scirp.35702-ref10">10</xref>] have been constructed.</p><p>In the Ref. [<xref ref-type="bibr" rid="scirp.35702-ref9">9</xref>], PT-symmetric, QES 2 &#215; 2-matrix Hamiltonians are analyzed with the emphasis set on the reality properties of the eigenvalues. The authors considered both trigonometric and hyperbolic 2 &#215; 2-matrix Hamiltonians.</p><p>A set of necessary and sufficient conditions (i.e. QES conditions) for 2 &#215; 2-matrix operators to preserve a vector space of polynomials have been proposed. These QES conditions constitute the so-called QES analytic method.</p><p>This paper is organized as follows: In the Section 2, based on the Ref. [<xref ref-type="bibr" rid="scirp.35702-ref9">9</xref>], we briefly recall the QES analytic method used to investigate the quasi-exact solvability of 2 &#215; 2-matrix operators. In Section 3, along the same lines as in the Ref. [<xref ref-type="bibr" rid="scirp.35702-ref9">9</xref>], we apply the QES analytic method in order to construct a new 2 &#215; 2-matrix QES Hamiltonian depending on Jacobi elliptic functions. We will consider two values of the constant δ: the case δ = 1 and the case δ = 2. The interesting results will be found.</p></sec><sec id="s2"><title>2. QES Analytic Method</title><p>A general test to check whether a <img src="1-1220048\6685208f-8672-4be9-a8cf-49cb04b6503c.jpg" />-matrix differential operator H (in a variable<img src="1-1220048\c6152b9c-8491-47f7-86b9-571fba938688.jpg" />) preserves a vector space whose components are polynomials is proposed [<xref ref-type="bibr" rid="scirp.35702-ref9">9</xref>]. After a gauge transformation and a change of variable on the operator H lead to a new operator <img src="1-1220048\011e52a2-d52e-4483-a366-e30272f89d61.jpg" /> which can be decomposed as follows</p><disp-formula id="scirp.35702-formula1608"><label>, (1)</label><graphic position="anchor" xlink:href="1-1220048\6d7ad5ea-c145-49b5-bed6-b1cfc7346e32.jpg"  xlink:type="simple"/></disp-formula><p>with</p><p><img src="1-1220048\76a679a9-3624-4aab-9acf-dc155c876f25.jpg" />.</p><p>Here <img src="1-1220048\ca5c86cb-9741-4fec-bc4b-fcbb8f9ebbda.jpg" /> denote homogeneous differential operators, C<sub>s</sub>, D<sub>s</sub> are arbitrary constant and <img src="1-1220048\c7bb0cfd-56f4-4f23-b357-c70ef0d0c4be.jpg" /> are integers. More precisely, the diagonal components of <img src="1-1220048\28a01da0-6764-4593-af23-2444d331319f.jpg" /> are differential operators and the off-diagonal components</p><p><img src="1-1220048\da29602a-502c-424a-812a-86d2d822871f.jpg" />and <img src="1-1220048\ecee6c5c-35a0-41bc-88c8-350bce1b2537.jpg" />are respectively proportional to <img src="1-1220048\f8a164f8-062a-483f-94e6-449754cdccdf.jpg" /> and <img src="1-1220048\675cc68b-ece0-4124-9c3a-2691f502c3ff.jpg" /> with <img src="1-1220048\fb6ed1bb-de70-48b7-a85b-a4bd955a5200.jpg" /> and<img src="1-1220048\5b19eeea-45f7-430e-a140-f245e927c726.jpg" />. The operators <img src="1-1220048\e2bf6de5-a29a-45df-91ef-fe6224b8a74a.jpg" /> and <img src="1-1220048\733c60c4-3344-492b-8c4f-ea82b4c2ebdf.jpg" /> have lower degrees in all their components than the corresponding components in<img src="1-1220048\c1d3950c-2114-44a2-bc05-2eef8913a091.jpg" />.</p><p>In order to obtain QES conditions for<img src="1-1220048\01b8377f-eddc-40d2-8936-6e865d2f67d2.jpg" />, the generic vector of the vector space <img src="1-1220048\5d5f560d-9da3-4e57-899b-224e0c564f0f.jpg" /> is</p><disp-formula id="scirp.35702-formula1609"><label>, (2)</label><graphic position="anchor" xlink:href="1-1220048\55044fc1-e2f6-42a2-b535-bf727bce7e57.jpg"  xlink:type="simple"/></disp-formula><p>Where <img src="1-1220048\6cbf588d-408e-4a2a-a9d3-9c3d27c550d2.jpg" /><img src="1-1220048\11a7fc01-410c-4667-8841-1e6c7daab0e7.jpg" /> are complex parameters. As a consequence the 2 &#215; 2-matrices <img src="1-1220048\b1192612-7466-4a93-9636-60d15ab53d0d.jpg" /> are defined by</p><disp-formula id="scirp.35702-formula1610"><label>(3)</label><graphic position="anchor" xlink:href="1-1220048\421e6fb2-9a8a-43bf-89ce-9bb31e972102.jpg"  xlink:type="simple"/></disp-formula><p>The three QES conditions for <img src="1-1220048\a8b5cc6e-58ea-4af4-96f9-cb1904693718.jpg" /> to have an invariant vector space are as follows [<xref ref-type="bibr" rid="scirp.35702-ref9">9</xref>]</p><disp-formula id="scirp.35702-formula1611"><label>(4)</label><graphic position="anchor" xlink:href="1-1220048\b3df1e07-98f9-4b76-86c7-43d17aa0394b.jpg"  xlink:type="simple"/></disp-formula><p>In the next step, we will apply in a systematic way the previous QES analytic method in order to construct a 2 &#215; 2-matrix QES Hamiltonian associated to a potential depending on the Jacobi elliptic functions [11,12].</p></sec><sec id="s3"><title>3. QES Jacobi Hamiltonian</title><sec id="s3_1"><title>3.1. Case δ = 1</title><p>In this section we apply the QES analytic method established previously to check whether a particular 2 &#215; 2-matrix operator is QES. We consider Schr&#246;dinger N &#215; N-matrix operator with potential depending on the Jacobi elliptic functions of the form [<xref ref-type="bibr" rid="scirp.35702-ref11">11</xref>]:</p><disp-formula id="scirp.35702-formula1612"><label>(5)</label><graphic position="anchor" xlink:href="1-1220048\45dc8117-2c7d-4760-8643-31ce2d5f1bbf.jpg"  xlink:type="simple"/></disp-formula><p>with</p><disp-formula id="scirp.35702-formula1613"><label>(6)</label><graphic position="anchor" xlink:href="1-1220048\861786f0-f083-4cc2-96b0-57bac0aafb56.jpg"  xlink:type="simple"/></disp-formula><p>where a<sub>j</sub>, b<sub>j</sub> denote real constants (without loss of generality we assume<img src="1-1220048\c357e311-fca0-48ab-a318-64c763c79ca5.jpg" />) and V<sub>I</sub> is symmetric off-diagonal matrix of the form</p><p><img src="1-1220048\66d0cb3f-6aaa-4657-9add-2d87f03ad454.jpg" /></p><p>Note that the above Hamiltonian is to be considered on the Hilbert space of periodic functions on [0, 4 K(k)].</p><p>The properties of the Jacobi functions that are useful to make calculations are listed in the relations (12) and (13).</p><p>The case <img src="1-1220048\ce5d4e13-123e-4baa-8a20-df93693da910.jpg" /> corresponds to the Lam&#233; equation [<xref ref-type="bibr" rid="scirp.35702-ref11">11</xref>]. We will treat in details the case <img src="1-1220048\1c55e56e-71f4-4ab8-9954-98137adc5284.jpg" /> which corresponds to the following operator</p><disp-formula id="scirp.35702-formula1614"><label>(7)</label><graphic position="anchor" xlink:href="1-1220048\cd74ab55-7069-4b4d-9407-afc1b3bb5408.jpg"  xlink:type="simple"/></disp-formula><p>with</p><p><img src="1-1220048\f6a5dcc4-faca-4eff-bad9-18e851f91448.jpg" /></p><p><img src="1-1220048\78e7d8f6-3ea7-42ff-b314-2525f906c562.jpg" />is the matrix identity and<img src="1-1220048\f9cc0a15-c499-499c-93af-807ac8547213.jpg" /><img src="1-1220048\0f239b27-1973-4a16-8d44-875e7600e7a1.jpg" /> denote real constants. Note that the sum <img src="1-1220048\f2c2c849-11eb-4740-8653-6d874e16be0d.jpg" /> is the potential associated to the Hamiltonian H(z).</p><p>Using the following change of function (i.e. the gauge transformation), the gauge Hamiltonian is written as follows</p><disp-formula id="scirp.35702-formula1615"><label>(8)</label><graphic position="anchor" xlink:href="1-1220048\daefada3-98ba-49eb-b2a2-30aa78446ba9.jpg"  xlink:type="simple"/></disp-formula><p>where</p><disp-formula id="scirp.35702-formula1616"><label>(9)</label><graphic position="anchor" xlink:href="1-1220048\d0f0aa45-bdf1-4ad6-b193-d01c1e52d78a.jpg"  xlink:type="simple"/></disp-formula><p>and</p><disp-formula id="scirp.35702-formula1617"><label>, (10)</label><graphic position="anchor" xlink:href="1-1220048\4e3d0c38-53fd-47cd-b9b2-2c9e0a0bfdec.jpg"  xlink:type="simple"/></disp-formula><p>Notice that two operators H and <img src="1-1220048\08b15eea-8209-4289-95d5-ac71fffb789d.jpg" /> are called equivalent based on the Equation (8).</p><p>The relevant change of variable consists in posing</p><p><img src="1-1220048\614bbd87-7bf9-476b-9929-48348c10be12.jpg" />. In particular the differential symbol <img src="1-1220048\5eb7cc39-89fc-4bed-a7bc-f98ee92bc008.jpg" /></p><p>is transformed into the following expression</p><disp-formula id="scirp.35702-formula1618"><label>(11)</label><graphic position="anchor" xlink:href="1-1220048\79d6db5f-cbf0-42de-8d42-7cc65379587d.jpg"  xlink:type="simple"/></disp-formula><p>We recall that for generic values of<img src="1-1220048\6300e5e8-4fe6-4363-b0c9-24e1193e7426.jpg" />, the Jacobi functions obey the following relations [<xref ref-type="bibr" rid="scirp.35702-ref11">11</xref>]:</p><disp-formula id="scirp.35702-formula1619"><label>(12)</label><graphic position="anchor" xlink:href="1-1220048\17a558f7-9460-4178-af8c-da35f1f25066.jpg"  xlink:type="simple"/></disp-formula><p>These identities as well as the following ones are useful to establish the gauge Hamiltonian (8) in the variable <img src="1-1220048\d8999371-280f-48d9-95ab-2835bc4152dc.jpg" /> after the prefactor including the Jacobi functions has been extracted [<xref ref-type="bibr" rid="scirp.35702-ref11">11</xref>]:</p><p>Referring to the above relations (12) and (13), for<img src="1-1220048\691490e0-dd14-4d27-9094-39ec2b19cd54.jpg" />, the second term and the third term of the operator <img src="1-1220048\09933616-8050-4025-89f5-32f8e75bfe55.jpg" /> of the Equation (9) are written as follows:</p><disp-formula id="scirp.35702-formula1620"><label>(14)</label><graphic position="anchor" xlink:href="1-1220048\40a3b7c9-f20a-4be3-8070-5f3a0dcee154.jpg"  xlink:type="simple"/></disp-formula><disp-formula id="scirp.35702-formula1621"><label>. (15)</label><graphic position="anchor" xlink:href="1-1220048\ce945034-edad-477a-af71-dc3527da41c1.jpg"  xlink:type="simple"/></disp-formula><p>Referring to the same relations used previously, for<img src="1-1220048\524b83ec-d6f3-41fc-9368-475e4ae2719b.jpg" />, the second term and the third term of the operator <img src="1-1220048\b9b03b1f-5cfa-453c-b460-bc6b7ace7726.jpg" /> of the Equation (9) are of the following form:</p><disp-formula id="scirp.35702-formula1622"><label>(16)</label><graphic position="anchor" xlink:href="1-1220048\5c260ceb-27d2-4bd1-b789-28362d62487e.jpg"  xlink:type="simple"/></disp-formula><disp-formula id="scirp.35702-formula1623"><label>. (17)</label><graphic position="anchor" xlink:href="1-1220048\6c555615-6999-4672-998f-abd4bf63105b.jpg"  xlink:type="simple"/></disp-formula><p>Replacing the terms of the components of the Hamiltonian <img src="1-1220048\726a5bd0-faf9-4cde-8a7b-294ba13e85fe.jpg" /> given by the Equation (9) by the expressions (11), (14)-(17) and considering the change of variable<img src="1-1220048\24d22a41-deab-49ff-9033-b728c0ef7565.jpg" />, one can easily check the following components in variable<img src="1-1220048\69899597-0d84-4a31-9bce-d58d6314075f.jpg" />:</p><disp-formula id="scirp.35702-formula1624"><label>. (18)</label><graphic position="anchor" xlink:href="1-1220048\fd190d3f-9083-4780-a199-92f6d0787380.jpg"  xlink:type="simple"/></disp-formula><p>The next step is to establish the conditions such that the gauge operator becomes quasi-exactly solvable. The so called QES conditions help to give the values of the real parameters <img src="1-1220048\20383c9b-9f63-48c0-978c-33b9c9985a3a.jpg" /> and <img src="1-1220048\8d67e8e8-920f-461d-9644-2f88b6e5d781.jpg" /> in terms of <img src="1-1220048\6dfd2d24-5a5c-4fc3-9d98-f3d1816fc56f.jpg" /> and<img src="1-1220048\8fc40a1f-318a-4aae-a9fb-42262c3f7a7a.jpg" />. Indeed <img src="1-1220048\73ab3c0d-036b-4e8d-be64-5323927b8aa2.jpg" /> remain free parameters and <img src="1-1220048\d91d5aaa-ec53-42e0-bf6c-02ea085046bf.jpg" /> is an integer.</p><p>Let us decompose the operator <img src="1-1220048\95528c14-5700-433d-b167-25d542323544.jpg" /> given by its components (18) according to</p><disp-formula id="scirp.35702-formula1625"><label>(19)</label><graphic position="anchor" xlink:href="1-1220048\139cf80e-47a8-45db-abc8-e264fb38f42d.jpg"  xlink:type="simple"/></disp-formula><p>where</p><disp-formula id="scirp.35702-formula1626"><label>(13)</label><graphic position="anchor" xlink:href="1-1220048\5fa8f3b2-541f-41d3-a480-16f792e5b6d5.jpg"  xlink:type="simple"/></disp-formula><disp-formula id="scirp.35702-formula1627"><label>(20)</label><graphic position="anchor" xlink:href="1-1220048\38b2de55-5fbf-489d-9828-bb7858c3865d.jpg"  xlink:type="simple"/></disp-formula><p>The generic vector of the invariant vector space under the action of the Hamiltonian <img src="1-1220048\acbb74c5-cc72-4ddf-9238-701f770f22f8.jpg" /> has the following form as it is given by the Equation (2)</p><p><img src="1-1220048\7f8cab72-ee8d-4737-b2ee-771520b96626.jpg" /></p><p>as<img src="1-1220048\bb2e05f7-c3f0-44bf-a670-743ebfeaa2fc.jpg" />, the above wave function <img src="1-1220048\1ead7fcf-5969-40a3-83de-f00dacd599c1.jpg" /> is written as follows</p><disp-formula id="scirp.35702-formula1628"><label>(21)</label><graphic position="anchor" xlink:href="1-1220048\e76e933a-5581-4865-bc7c-56ffba11b897.jpg"  xlink:type="simple"/></disp-formula><p><img src="1-1220048\7c2ac979-e6a4-4ea6-816d-0c9a6fe30ca4.jpg" />acting on the wave function<img src="1-1220048\89b44aaf-2708-4bfd-8c2b-d519a876dcc5.jpg" />, he increases the degree by one unit,</p><p><img src="1-1220048\bc963c4e-5fb6-4d97-913a-9f3484053ae1.jpg" />doesn’t change the degree of the wave function<img src="1-1220048\dc83f2b3-8d45-410e-8580-6f720b26e04d.jpg" />, and <img src="1-1220048\b8a28513-13ef-4545-9ba0-6d92570cb68c.jpg" /> reduces the degree of the wave function <img src="1-1220048\e09984bb-b2af-4683-a37c-c707e4aaafee.jpg" /> by one unit.</p><p>Let the operator <img src="1-1220048\79445aa4-6004-4b64-8e32-5dc73044f245.jpg" /> acts on the above vector<img src="1-1220048\092e1dd2-97c4-49f5-8cd2-54a1c4db5db0.jpg" />, the components of the vector <img src="1-1220048\535c9cf5-dc3a-4004-9e5c-86127bfd04c2.jpg" /> are then polynomials in <img src="1-1220048\a7a17bd3-05e1-4d9f-8cb5-11c3e0527420.jpg" /> whose components are linear in the constants<img src="1-1220048\fe44c982-5992-4aea-8495-22830b71c3f4.jpg" />.</p><p>As a consequence the vector <img src="1-1220048\f72ce5e4-2b8f-4e38-81a1-c0a54c2c7f90.jpg" /> can be decomposed uniquely according to [<xref ref-type="bibr" rid="scirp.35702-ref9">9</xref>]</p><disp-formula id="scirp.35702-formula1629"><label>. (22)</label><graphic position="anchor" xlink:href="1-1220048\1272f595-1860-4c45-9e9d-e15277402fa8.jpg"  xlink:type="simple"/></disp-formula><p>This above vector defines in particular the constant 2 &#215; 2-matrices <img src="1-1220048\1362bb2c-9e38-4879-9947-728777700c10.jpg" /> and <img src="1-1220048\ad080e68-42ff-4cde-8c07-f225b9c5e28b.jpg" /> which are found as follows</p><disp-formula id="scirp.35702-formula1630"><label>(23)</label><graphic position="anchor" xlink:href="1-1220048\c79dd9f2-4229-427e-b586-7bbc52d3d802.jpg"  xlink:type="simple"/></disp-formula><p>One can easily find The three necessary QES conditions for the operator <img src="1-1220048\538d3b66-dc28-4b45-ad0b-071693206bb5.jpg" />to have a finite dimensional invariant vector space are successively obtained [<xref ref-type="bibr" rid="scirp.35702-ref9">9</xref>]:</p><p>1) The first QES condition is</p><disp-formula id="scirp.35702-formula1631"><label>(25)</label><graphic position="anchor" xlink:href="1-1220048\69ef3247-c8b9-4169-a41c-7898a0de2dea.jpg"  xlink:type="simple"/></disp-formula><p>2) the second QES condition is as follows</p><p><img src="1-1220048\b2ba13a9-d26f-47d9-8a9b-667028d01797.jpg" /></p><p>In this above equation replacing <img src="1-1220048\786de509-ff8a-42c2-8c26-852fc31fbccd.jpg" /> by its value (25) and after some algebraic manipulations, the second QES condition is obtained</p><disp-formula id="scirp.35702-formula1632"><label>(26)</label><graphic position="anchor" xlink:href="1-1220048\594b01ed-ca98-4d4e-95e3-16868a789212.jpg"  xlink:type="simple"/></disp-formula><p>3) finally the third QES condition for the operator <img src="1-1220048\babc9525-3950-41fe-ad6a-69a66286b0c9.jpg" /> to have a finite dimensional invariant vector space (i.e. the operator <img src="1-1220048\4e6b68f2-e31c-4837-82c9-5de8eb66c733.jpg" /> is said quasi-exactly solvable) is obtained by the condition involving the matrix <img src="1-1220048\a56e8fad-2c14-4a18-adb8-bae638c64aab.jpg" /> as</p><disp-formula id="scirp.35702-formula1633"><label>, (27)</label><graphic position="anchor" xlink:href="1-1220048\041bc03e-5178-4fe7-a4ce-8c13487cf469.jpg"  xlink:type="simple"/></disp-formula><disp-formula id="scirp.35702-formula1634"><label>(24)</label><graphic position="anchor" xlink:href="1-1220048\4e5c75cd-7dd7-431f-b41e-a40e05444b0d.jpg"  xlink:type="simple"/></disp-formula><p>where <img src="1-1220048\b547bb74-34eb-4e2f-9f8b-38f17ae50aa1.jpg" /> is a constant and</p><disp-formula id="scirp.35702-formula1635"><label>(28)</label><graphic position="anchor" xlink:href="1-1220048\ce7049ce-1110-481e-bbfa-5cf3506f39e9.jpg"  xlink:type="simple"/></disp-formula><p>The above expression is given by the first QES condition <img src="1-1220048\03d766b5-8a48-425f-8d6b-d3b85a1d047b.jpg" /></p><p>After some algebraic manipulations, the Equations (27) and (28) lead to the third QES condition</p><disp-formula id="scirp.35702-formula1636"><label>(29)</label><graphic position="anchor" xlink:href="1-1220048\8b42073b-e393-4759-b809-5fadcd947d60.jpg"  xlink:type="simple"/></disp-formula><p>Now, referring to the QES conditions given by the Equations (25), (26) and (29), we are allowed to conclude that the operator <img src="1-1220048\02d2b78f-bdad-4654-a8b3-c50398c3e883.jpg" /> (therefore<img src="1-1220048\dbb08b72-4fc5-4685-a280-5c76d5e4ec8b.jpg" />) is quasi-exactly solvable [<xref ref-type="bibr" rid="scirp.35702-ref9">9</xref>]. In other words, a finite part of the eigenvalues of the operator <img src="1-1220048\8e7b0977-03ae-4023-a1f0-33fad715ece5.jpg" /> can be computed algebraically. Note that the QES Hamiltonian constructed depends only on two free parameters <img src="1-1220048\8c6e288b-6ff6-4841-aa8e-cbf4bf229bfa.jpg" /> and on the non negative integer<img src="1-1220048\1dbc6806-4bdb-42bd-918b-4c2dbd17cfb4.jpg" />.</p></sec><sec id="s3_2"><title>3.2. Case δ = 2</title><p>Along the same lines applied for the previous case, i.e. for the case δ = 1, one has to perform a gauge transformation according to</p><disp-formula id="scirp.35702-formula1637"><label>, (30)</label><graphic position="anchor" xlink:href="1-1220048\6e64cfcb-e0c3-439b-a2c3-2b4c4197654c.jpg"  xlink:type="simple"/></disp-formula><p>after some algebraic manipulations, the components of the above Hamiltonian are of the following form</p><disp-formula id="scirp.35702-formula1638"><label>(31)</label><graphic position="anchor" xlink:href="1-1220048\77002c7e-30c8-4ba9-afaf-dd7e328e01e2.jpg"  xlink:type="simple"/></disp-formula><p>with</p><p><img src="1-1220048\96ce4bf6-6945-44cc-8335-99d03af3d8a3.jpg" /></p><p>and the operator <img src="1-1220048\3b0cf72b-b5c0-49f4-8dfe-e5c731f7024d.jpg" /> is given by the Equation (7).</p><p>Referring to the relations (12) and to the table of identities given by the Equation (13), the second term and the third term of the operator <img src="1-1220048\54a4a3fa-c0a0-43c8-b1c7-1c1e76b02576.jpg" /> (31) are of the following form</p><disp-formula id="scirp.35702-formula1639"><label>(32)</label><graphic position="anchor" xlink:href="1-1220048\0c51f98a-673f-4e60-86fd-31175cf79fcf.jpg"  xlink:type="simple"/></disp-formula><disp-formula id="scirp.35702-formula1640"><label>(33)</label><graphic position="anchor" xlink:href="1-1220048\3daff1da-93b3-42ed-a1e0-16e50e44a521.jpg"  xlink:type="simple"/></disp-formula><p>with<img src="1-1220048\ec7aa5c7-ebc1-4fc0-a673-b29446c09bba.jpg" />.</p><p>For<img src="1-1220048\8e85b58a-b08a-440b-bfea-db56d810178b.jpg" />, the same relations (12) and the same table of identities (13) lead to the following second term and the third term of the operator <img src="1-1220048\f42aae8f-1cab-4bd2-a30b-bdb9b3f7810d.jpg" /> (31):</p><disp-formula id="scirp.35702-formula1641"><label>(34)</label><graphic position="anchor" xlink:href="1-1220048\e2e1a861-95e1-4719-a867-40bc2c18ac06.jpg"  xlink:type="simple"/></disp-formula><disp-formula id="scirp.35702-formula1642"><label>. (35)</label><graphic position="anchor" xlink:href="1-1220048\85a7e71a-d36b-44ab-b5fb-eb199a885f78.jpg"  xlink:type="simple"/></disp-formula><p>Referring to the relations (11), (32), (33), (34), (35) and after performing the change of variable<img src="1-1220048\669e526d-5339-4af3-8d60-cb83da9fa90c.jpg" />, the different components of the Hamiltonian <img src="1-1220048\3c726b87-ec8e-4fb1-9315-0dfaa8a01815.jpg" /> given by the Equation (31) take the following form Decomposing now the above operator <img src="1-1220048\9f8f4029-9297-4a1e-b2a7-258984930388.jpg" /> according the Equation (11), we obtain</p><disp-formula id="scirp.35702-formula1643"><label>(36)</label><graphic position="anchor" xlink:href="1-1220048\4beaea20-d083-4f63-b1aa-a523b5f039ba.jpg"  xlink:type="simple"/></disp-formula><disp-formula id="scirp.35702-formula1644"><label>(37)</label><graphic position="anchor" xlink:href="1-1220048\6d2d608f-34db-4a13-8117-9a66ba879e77.jpg"  xlink:type="simple"/></disp-formula><p>The generic element of the invariant vector space <img src="1-1220048\b41b39be-c8bb-4f98-bb61-625dfd13ef63.jpg" /> under the action of the operator <img src="1-1220048\9f3f7d4d-c7b2-43bb-9bb0-ea0e2db9e2d1.jpg" /> is given by the Equation (2) as in the QES analytic method</p><p><img src="1-1220048\cd3c0373-c117-40cf-89f2-c6516ccad5f4.jpg" /><img src="1-1220048\4f011afc-34d9-40df-9917-4352a5f27246.jpg" />the case <img src="1-1220048\cafe287b-b80e-4403-b07a-c85fb4e89c05.jpg" /> leads to</p><disp-formula id="scirp.35702-formula1645"><label>. (38)</label><graphic position="anchor" xlink:href="1-1220048\311455d2-8418-4157-9ffd-827c74397fe7.jpg"  xlink:type="simple"/></disp-formula><p>Notice that the above operators <img src="1-1220048\8fba8fd0-7e8f-4c52-a43c-14ca49d8e1a4.jpg" /> and <img src="1-1220048\e8af269a-c6e7-4538-9a5b-0dcf00ee4d37.jpg" /> given by the Equations (37) are respectively the matrix operators which increases, preserves and reduces the degree of the above generic vector <img src="1-1220048\c486c8a4-3bb8-490e-8206-a30595fa8e2d.jpg" /> given by the Equation (38). As a consequence the vector <img src="1-1220048\3ab08679-34ae-4372-b4b7-b62963247c5b.jpg" /> can be decomposed as follows</p><disp-formula id="scirp.35702-formula1646"><label>(39)</label><graphic position="anchor" xlink:href="1-1220048\c31a3533-4dbd-4ef6-b73b-16cb17358a2c.jpg"  xlink:type="simple"/></disp-formula><p>where the constant 2 &#215; 2-matrices <img src="1-1220048\0a0463d9-49a8-43c7-bf38-6d467b217ac2.jpg" /> and <img src="1-1220048\cdae660e-04a2-4b26-aa11-9126739ed4a3.jpg" /> can be computed explicitly after a straightforward calculation</p><p><img src="1-1220048\e2ed2b96-799e-4253-849e-94a599625cec.jpg" />where</p><p><img src="1-1220048\f25f1ba0-c9f7-4b73-b807-bad577706a28.jpg" /></p><p>One can deduce the matrix <img src="1-1220048\1be0ad21-ef81-4ccf-ad5e-e6ab35d25b27.jpg" /> from the following expression</p><p><img src="1-1220048\753573e8-0016-46b7-9655-7c565fb6030c.jpg" />where</p><p><img src="1-1220048\16679d44-e0b3-43e1-8368-6f9381225226.jpg" /></p><p>finally the matrix <img src="1-1220048\7b8f6297-dc25-4eed-b82c-cdace42a9bb1.jpg" /> is easily found by</p><p><img src="1-1220048\e6409ecb-ae5c-4e32-be29-d8b75e37d2d4.jpg" />where</p><p><img src="1-1220048\78ac11c1-6ad7-43d2-8aa0-e94788d309d5.jpg" /></p><p>Along the same lines used in the QES analytic method, the three necessary conditions (4) for the operator <img src="1-1220048\f4876b77-7282-40cf-91b6-f72457a9ebfc.jpg" /> whose components are given by the Equation (36) to be quasi-exactly solvable are successively obtained:</p><p>1) the first QES condition is as follows</p><p><img src="1-1220048\e11568c4-33d7-4df3-b653-216792068701.jpg" /></p><p>2) the second QES condition is easily checked</p><p><img src="1-1220048\45a43761-599d-4be3-a1e6-c1af569e806a.jpg" /></p><p>3) Finally the third QES condition is found</p><disp-formula id="scirp.35702-formula1647"><label>(40)</label><graphic position="anchor" xlink:href="1-1220048\7dcafc82-e5f6-4aae-8b4b-c40494959d26.jpg"  xlink:type="simple"/></disp-formula></sec></sec><sec id="s4"><title>4. Conclusion</title><p>In this paper, we have applied the QES analytic method established in the Ref. [<xref ref-type="bibr" rid="scirp.35702-ref9">9</xref>] in order to construct a 2 &#215; 2-matrix QES Hamiltonian which is associated to a potential depending on the Jacobi elliptic functions. We have considered two cases: <img src="1-1220048\a0a51f8e-2c4d-45ff-9271-a889ef05a8a2.jpg" />and<img src="1-1220048\7a45ee0f-6128-4df9-ba35-255118a758c4.jpg" />. More precisely, the three QES conditions for the Jacobi Hamiltonian to have an invariant vector space are computed algebraically.</p></sec><sec id="s5"><title>5. Acknowledgements</title><p>I thank Pr. Yves Brihaye for useful discussions.</p></sec><sec id="s6"><title>REFERENCES</title></sec></body><back><ref-list><title>References</title><ref id="scirp.35702-ref1"><label>1</label><mixed-citation publication-type="other" xlink:type="simple">A. V. Turbiner, “Quasi-Exactly-Solvable Problems and sl(2) Algebra,” Communications in Mathematical Physics, Vol. 118, No. 3, 1988, pp. 467-474. 
doi:10.1007/BF01466727</mixed-citation></ref><ref id="scirp.35702-ref2"><label>2</label><mixed-citation publication-type="other" xlink:type="simple">A. G. Ushveridze, “Quasi-Exactly Solvable Models in Quantum Mechanics,” Institute of Physics Publishing, 1995.</mixed-citation></ref><ref id="scirp.35702-ref3"><label>3</label><mixed-citation publication-type="other" xlink:type="simple">A. V. Turbiner, “Lame Equation sl(2) Algebra and Isospectral Deformations,” Journal of Physics A: Mathematical and General, Vol. 22, 1989, pp. 1-144. 
doi:10.1088/0303-4470/22/1/001</mixed-citation></ref><ref id="scirp.35702-ref4"><label>4</label><mixed-citation publication-type="other" xlink:type="simple">M. A. Shifman and A. V. Turbiner, “Quantal Problems with Partial Algebraization of the Spectrum,” Communications in Mathematical Physics, Vol. 126, No. 2, 1989, pp. 347-365. doi:10.1007/BF02125129</mixed-citation></ref><ref id="scirp.35702-ref5"><label>5</label><mixed-citation publication-type="other" xlink:type="simple">A. González-López, N. Kamran and P. J. Olver, “Normalizability of One-Dimensional Quasi-Exactly Solvable Schr?dinger Operators,” Communications in Mathematical Physics, Vol. 153, No. 1, 1993, pp. 117-146.  
doi:10.1007/BF02099042 </mixed-citation></ref><ref id="scirp.35702-ref6"><label>6</label><mixed-citation publication-type="other" xlink:type="simple">A. González-López, N. Kamran and P. J. Olver, “Quasi-Exactly Solvable Lie Algebras of Differential Operators in Two Complex Variables,” Journal of Physics A, Vol. 24, No. 17, 1991, p. 3995. 
doi:10.1088/0305-4470/24/17/016</mixed-citation></ref><ref id="scirp.35702-ref7"><label>7</label><mixed-citation publication-type="other" xlink:type="simple">R. Zhdanov, “Quasi-Exactly Solvable Matrix Models,” Physics Letters B, Vol. 405, No. 3-4, 1997, pp. 253-256. 
doi:10.1016/S0370-2693(97)00655-2</mixed-citation></ref><ref id="scirp.35702-ref8"><label>8</label><mixed-citation publication-type="other" xlink:type="simple">Y. Brihaye and P. Kosinski, “Quasi Exactly Solvable Matrix Models in sl(n),” Physics Letters B, Vol. 424, No. 1-2, 1997, pp. 43-47. doi:10.1016/S0370-2693(98)00167-1</mixed-citation></ref><ref id="scirp.35702-ref9"><label>9</label><mixed-citation publication-type="other" xlink:type="simple">Y. Brihaye, A. Nininahazwe and B. P. Mandal, “PT-Symmetric, Quasi-Exactly Solvable Matrix Hamiltonians,” Journal of Physics A: Mathematical and Theoretical, Vol. 40, No. 43, 2007, pp. 13063-13073 
doi:10.1088/1751-8113/40/43/014</mixed-citation></ref><ref id="scirp.35702-ref10"><label>10</label><mixed-citation publication-type="other" xlink:type="simple">Y. Brihaye and A. Nininahazwe, “Extended Jaynes-Cummings models and (Quasi)-Exact Solvability,” Journal of Physics A: Mathematical and Theoretical, Vol. 39, No. 33, 2006, pp. 1-14. doi:10.1088/0305-4470/39/31/011</mixed-citation></ref><ref id="scirp.35702-ref11"><label>11</label><mixed-citation publication-type="other" xlink:type="simple">Y. Brihaye and B. Hartmann, “Quasi-Exactly Solvable N × N-Matrix Schrodinger Operators,” Modern Physics Letters A, Vol. 16, No. 29, 2001, pp. 1895-1906. 
doi:10.1142/S0217732301005242</mixed-citation></ref><ref id="scirp.35702-ref12"><label>12</label><mixed-citation publication-type="other" xlink:type="simple">Y. Brihaye and M. Godard, “Quasi Exactly Solvable Extensions of the Lamé Equation,” Journal of Mathematical Physics, Vol. 34, No. 11, 1993, p. 5283.  
doi:10.1063/1.530304</mixed-citation></ref></ref-list></back></article>