1、78035AzetidinesRing expansionintonow0C211 2009 Elsevier B.V. All rights reserved.thylHammerand the stereospecificity of this rearrangement, as well as the pos-sible introduction of external nucleophiles able to compete withthe chloride anion 4ce has been experimentally established,thus offering a ve
2、ry efficient entry to 3-substituted pyrrolidines(Fig. 1). Though the mechanism of the former rearrangement hasbeen extensively studied and has led to many synthetic applica-lytical gradient optimization method. Stationary points were char-acterized by frequency calculations. The minimum energy react
3、ionpaths were obtained via the gradient descent method from thetransition states in both direct and reverse directions of the transi-tion vectors.The solvents effects has been considered in the context of polar-izable continuum model (PCM) 811 and DMSO was chosen as arepresentative polar solvent (e
4、= 46.7), which has been frequentlyused in such rearrangements. All calculations were done in IEF-PCM approximation 1215, with dielectric constant at infinite* Corresponding author. Tel.: +33 139254455; fax: +33 139254452.Journal of Molecular Structure: THEOCHEM 908 (2009) 2630Contents lists availabl
5、eStruE-mail address: coutychimie.uvsq.fr (F. Couty).ion 2 formed through an internal backside nucleophilic substitu-tion (SNib), which is the rate determining step. This putative inter-mediate 2 is next opened through a clean SN2 reaction by thechloride anion at the bridgehead carbon to give the mor
6、e stablepiperidine 3. Furthermore, addition of an external nucleophile, ableto compete with the chloride anion allows the synthesis of various3-substituted piperidines in a stereospecific manner, a reactionthat has found many applications in the synthesis of bioactivepiperidinic compounds 3. On the
7、other hand, the rearrangementof lower homologues analogous to 2-chloromethyl azetidine 4 into3-chloro-pyrrolidine 6 has been studied much more recently 45 a viable intermediate considering its strained nature comparedto 2? It is worthy to note that closely related (N-ethyl) bicyclicaziridinium 2 hav
8、e been successfully isolated and characterized2a, which is not the case, to the best of our knowledge, for lowerhomologue 5.2. Computational methodDFT calculations were performed using B3LYP exchange-corre-lation functionals 5,6, together with the standard 6-31G*basisset 7. Geometry optimization was
9、 carried out using Bernys ana-1. IntroductionRing enlargement of 2-chloromechloro-piperidine 3 is a venerable reactionby Fuson 1. It has been proposed throughdata and kinetic measurements fromreactions takes place through an intermediate0166-1280/$ - see front matter C211 2009 Elsevier B.V. Alldoi:1
10、0.1016/j.theochem.2009.04.035pyrrolidine 1 into 3-that was first reportedaccurate experimentalet al. 2 that thisbicyclic aziridiniumtions, little is known about the intimate mechanism of the secondrearrangement. Surprisingly, none of these reactions have yet beenstudied in silico. Therefore, conside
11、ring the synthetic importanceof these ring expansions and the lack of accurate kinetic data forthe azetidine case, we have undertaken a DFT simulation of bothrearrangements, starting from N-isopropyl derivatives 1 and 4.One particular point was expected to be answered: is aziridiniumPyrrolidinesAzet
12、idinium ionsazonia-bicyclo2.1.0pentane intermediate 5 , while the pyrrolidine still rearranges in a single stepprocess.Ring expansion of 2-chloromethyl pyrrolidinetheoretical investigationFranois Coutya,*, Mikhail KletskiibaInstitut LAVOISIER, UMR CNRS 8180, Universit de Versailles, 45, Avenue des E
13、tats-Unis,bDepartment of Chemistry, Southern Federal University, 7, Zorge St., 344090 Rostov-on-Don,article infoArticle history:Received 24 March 2009Received in revised form 27 April 2009Accepted 29 April 2009Available online 7 May 2009Keywords:abstractDFT calculation were performedpyrrolidine and
14、azetidinetively. These ring enlargementsthe synthesis of nitrogen heterocycles.a polarizable continuum model.reactions, with a single transitionand the two reactions differJournal of Molecularjournal homepage: www.elsevirights reserved.or azetidine: ComparedVersailles Cdex, FranceRussian Federationi
15、n order to simulate the ring enlargement of N-isopropyl-2-chloromethylthe corresponding 3-chloro piperidine and 3-chloro pyrrolidine, respec-are prototypes of reactions that have found growing applications inBoth calculations were conducted in gas phase and in DMSO, usingIn gas phase, these ring enl
16、argements were found to be synchronousstate, but in DMSO, significant decrease of energy barrier is observed,to a large extent, since the azetidine rearranges through a strained 1-at ScienceDirectcture: THEOCHEM equal to 2.179 and solvent radius (in Angstroms) equalto 2.455.3. Results and discussion
17、Performed quantum mechanical gas phase calculations showthat both chlorides 1 and 4 rearrange through a concerted one stepelectrocyclic process (Figs. 2 and 3, red line). In the gas phase reac-barrier for the rearrangement of N-ethyl-2-chloromethyl pyrroli-dine into the corresponding piperidine perf
18、ormed in benzene(22 2 kcal/mol at 25 C176C) 2b, since a value of 18.8 kcal/mol is cal-culated here for the N-isopropyl analogue in DMSO. The strong sol-vation effect on the transition states involved in the ringexpansions can be explained by high polar character of these struc-tures (see the values
19、of dipole moments in Table 1).Another striking difference between the two rearrangements isthe high energy gain in the case of the ring expansion of chloro-methyl azetidine (16.7 kcal/mol) compared to the chloro pyrroli-dine (0.5 kcal/mol, DMSO results). This important differencedemonstrates that if
20、 thermodynamic control is established in thesereactions, then the equilibrium should favor exclusively 6 versus 4,NClNClNCl1 23NClNCl5NCl64Fig. 1. Ring expansion of 2-chloromethyl pyrrolidine or 2-chloromethyl azetidine.F. Couty, M. Kletskii/Journal of Molecular Structure:tions 1?3 and 4?6 are accom
21、panied by significant energyexpenditures, with activation barriers of 34.8 and 41.7 kcal/mol,respectively (Figs. 2 and 3, red line).On the other hand, in DMSO environment, the rearrangementof 4 propagates as a real staged process via the bipolar interme-diate 50through two different transition state
22、s TS1 and TS2(Fig. 3, blue line) which is in sharp contrast with the results ob-tained in gas phase. On the other hand, chloride 1 still rearrangesin DMSO through a single synchronous process via pericyclictransition state 2 (Fig. 2, blue line). This definitely proves thatstrained 1-azonia-bicyclo2.
23、1.0pentane 50is a viable intermediatein this reaction.In the general case, solvation effects lead to significant energybarrier decrease. This is indeed the case here, were the energy bar-rier for the limited stage of reaction 4?50and for reaction 1?3Fig. 2. Energy profiles for the rearrangement 1?3.
24、 Red line corresponds to the gasphase process and the blue line corresponds to the reaction in DMSO. (Forinterpretation of the references to color in this figure legend, the reader is referredto the web version of this article.)decrease by 17.7 and 16.0 kcal/mol, respectively, in DMSO. In thispolar
25、solvent, the difference of energy barrier required for thesetwo rearrangements is also lowered (5.2 kcal/mol higher for theazetidine case) compared to the gas phase (6.9 kcal/mol higherfor the azetidine case). It should be mentioned that these theoret-ical data fit well with the experimental estimat
26、ion of the energyFig. 3. Energy profiles for the rearrangement 4?6. Red line corresponds to the gasphase process and the blue line corresponds to the reaction in DMSO. (Forinterpretation of the references to color in this figure legend, the reader is referredto the web version of this article.)THEOC
27、HEM 908 (2009) 2630 27but only slightly 3 versus 2 (the difference of 0.5 kcal/mol repre-sent a 3/7 ratio of 3/2 at 20 C176C).It is interesting to note that the intermediate 50is a real ion pair(Table 1) which consists of two subsystems, the bicyclic cation andchloride anion. On the other hand, in g
28、as phase the total energy offree cation and chlorine anion is much greater than the energy ofintermediate 50(Table 1). Calculated characteristics of all station-ary points of both potential energy surfaces are represented inTable 1 and in Figs. 47.4. ConclusionThis work demonstrates that in polar me
29、dium (DMSO), bothrearrangements are easy processes, with medium activation en-ergy barriers. It highlights two important differences betweenthe pyrrolidine and the azetidine case. First, contrary to our expec-tations, the 1-azonia-bicyclo2.1.0pentane 20is not too high in en-ergy in DMSO, and in this
30、 case, the rearrangement takes placeStructure:28 F. Couty, M. Kletskii/Journal of Molecularthrough a two-steps mechanism via this intermediate. Secondly,the important difference of energy between the reactants andproducts in the azetidine case reflects the strain associated withTable 1Total energies
31、 (E), relative energy values (DE) and dipole moments (l, D) of corresponding systemsFor a reference, the sum of total energies of reagents is accepted. ZPE is a zero-point energySystem Gas phaseE, a.u. E + ZPE, a.u. DE, kcal/mol DE + ZPE, kcal/mol1 C0829.453 0.02 C0829.397 34.83 C0829.452 0.24 C0790
32、.114 C0789.908 5 C0790.048 C0789.846 41.7 41.36 C0790.139 C0789.934 C015.6 C014.5Anion ClC0C0460.252 Bicyclic cation C0329.657 TS1 TS2 Fig. 4. Main bond lengths () and bond angles (in degrees) of reagent 1, transition statefrequence in cmC01.Fig. 5. Main bond lengths () and bond angles (in degrees)
33、of reagent 1, transition statefrequence in cmC01.THEOCHEM 908 (2009) 2630the starting four membered ring and implies a total reaction incase of equilibrium establishment, contrary to the pyrrolidine case.Further calculations are in progress in order to understand the reg-involved in the reactions 1?
34、2?3 and 4?5?6 in the gas phase and DMSO.correctionDMSOl, DE, a.u. E + ZPE, a.u. DE, kcal/mol DE + ZPE, kcal/mol l, D2.8 C0829.456 0.0 2.813.9 C0829.426 18.8 13.43.0 C0829.456 C00.5 3.02.2 C0790.117 C0789.913 2.913.3 C0790.089 C0789.885 17.5 17.7 18.12.3 C0790.143 C0789.938 C016.7 C015.7 2.6C0460.362
35、 C0329.727 C0790.078 C0789.884 24.0 23.3 13.6C0790.085 C0789.882 19.8 19.5 16.32 and product 3 for the gas phase rearrangement 1?2?3. Unique imaginary2 and product 3 for the rearrangement 1?2?3 in DMSO. Unique imaginaryFig. 6. Main bond lengths () and bond angles (degrees) of reagent 4, transition s
36、tate 5 and product 6 for the gas phase rearrangement 4?5?6. Unique imaginaryfrequence in cmC01.Fig. 7. Main bond lengths () and bond angles (degrees) of reagent 4, transition states TS1, TS2, intermediate 50and product 6 for the rearrangement 4?TS1?50?TS2?6in DMSO. Unique imaginary frequencies in cm
37、C01.F. Couty, M. Kletskii/Journal of Molecular Structure: THEOCHEM 908 (2009) 2630 29ioselectivity observed during the nucleophilic opening of 20or 5with an added external nucleophile.AcknowledgmentsUniversity of Versailles St-Quentin en Yvelines and CNRS areacknowledged for generous support.Referen
38、ces1 R.C. Fuson, C.L. Zirke, J. Am. Chem. Soc. 70 (1949) 2760.2 (a) C.F. Hammer, S.R. Heller, J.H. Craig, Tetrahedron 28 (1972) 239;(b) C.F. Hammer, J.D. Weber, Tetrahedron 37 (1981) 2173.3 (a) J. Wilkem, M. Kossenjans, W. Saak, D. Haase, S. Pool, J. Martens, Liebigs Ann.(1997) 573;(b) J. Cossy, C.
39、Dumas, D. Gomez-Pardo, Synlett (1997) 905;(c) J. Cossy, C. Dumas, D. Gomez-Pardo, Bioorg. Med. Chem. Lett. 7 (1997)1343;(d) P. Michel, A. Rassat, J. Org. Chem. 65 (2000) 2572;(e) J. Cossy, O. Mirguet, D. Gomez-Pardo, Synlett (2001) 1575;(f) M. Mena, J. Bonjoch, D. Gomez-Pardo, J. Cossy, J. Org. Chem
40、. 71 (2006) 5930.4 (a) F. Outurquin, X. Pannecoucke, B. Berthe, C. Paulmier, Eur. J. Org. Chem.(2002) 1007;(b) F. Couty, F. Durrat, D. Prim, Tetrahedron Lett. 44 (2003) 5209;(c) W. Van Bradandt, R. Van Landeghem, N. De Kimpe, Org. Lett. 8 (2006) 1105;(d) F. Durrat, M. Vargas-Sanchez, F. Couty, G. Ev
41、ano, J. Marrot, Eur. J. Org.Chem. (2008) 3286;(e) B. Drouillat, F. Couty, O. David, G. Evano, J. Marrot, Synlett 9 (2008) 1345.5 A. D Becke, J. Chem. Phys. 98 (1993) 5648.6 C. Lee, W. Yang, R.G. Parr, Phys. Rev. B 37 (1988) 785.7 W.J. Hehre, L. Radom, P.v.R. Schleyer, J.A. Pople, Ab initio Molecular
42、 OrbitalTheory, Wiley, New York, 1986.8 B.Y. Simkin, I.I. Sheikhet, Quantum Chemical and Statistical Theory ofSolutions: A Computational Approach, Ellis Horwood, London, 1995.9 E. Cances, B. Mennucci, J. Tomasi, J. Chem. Phys. 107 (1997) 3032.10 M. Cossi, V. Barone, R. Cammi, J. Tomasi, J. Chem. Phy
43、s. Lett. 255 (1996) 327.11 V. Barone, M. Cossi, J. Tomasi, J. Comput. Chem. 19 (1998) 404.12 M.T. Cancs, B. Mennucci, J. Tomasi, J. Chem. Phys. 107 (1997) 3032.13 M. Cossi, V. Barone, B. Mennucci, J. Tomasi, Chem. Phys. Lett. 286 (1998) 253.14 B. Mennucci, J. Tomasi, J. Chem. Phys. 106 (1997) 5151.15 M. Cossi, G. Scalmani, N. Rega, V. Barone, J. Chem. Phys. 117 (2002) 43.30 F. Couty, M. Kletskii/Journal of Molecular Structure: THEOCHEM 908 (2009) 2630
