Evidence for the coexistence of two bond-stretch isomers in solution

Article abstract of DOI:10.1002/anie.200460475

Lowering the temperature breaks the σ bond in the interconversion of 1,3-dibora-2,4-diphosphoniocyclobutane-1,3-diyl and its bicyclo[1.1.0]butane bond-stretch isomer (see picture). Variable-temperature NMR and UV/Vis experiments demonstrate for the first

Full text of DOI:10.1002/anie.200460475

Communications
Bond-Stretch Isomers
Evidence for the Coexistence of Two Bond-
Stretch Isomers in Solution**
Amor Rodriguez, Ryan A. Olsen, Nima Ghaderi,
David Scheschkewitz, Fook S. Tham,
Leonard J. Mueller,* and Guy Bertrand*
The concept of “bond-stretch isomerism” has been intro-
duced by Stohrer and Hoffmann by using strained tricyclic
hydrocarbons (Scheme 1): “In the 2,2,2-system the optimum
alignment for through-bond coupling of radical lobes creates
the conditions for a new type of isomerism—two stable
conformations related by a simple bond stretching. These are
Scheme 1. Some key compounds in the debate on bond-stretch
isomerism.
[*] Dr. A. Rodriguez, R. A. Olsen, N. Ghaderi, Dr. D. Scheschkewitz,
Dr. F. S. Tham, Prof. L. J. Mueller, Prof. G. Bertrand
UCR-CNRS Joint Research Chemistry Laboratory (UMR 2282)
Department of Chemistry
University of California
Riverside, CA 92521-0403 (USA)
Fax: (+1)909-787-2725
E-mail: gbertran@mail.ucr.edu
[**] We are grateful to the NSF (CHE 0213510 and CHE 0349345) and
RHODIA for financial support of this work.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
4880
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/anie.200460475
Angew. Chem. Int. Ed. 2004, 43, 4880 –4883

Angewandte
Chemie
the normal tricyclic form II and the stabilized diradical I”.[1]
The early attempts to characterized bond-stretch isomers
either failed or were eventually rejected as crystallographic
artifacts,and therefore the existence of bond-stretch isomers
became questionable.[2,3] According to the most recent review
on this topic,[4] the 1,3-diphosphacyclobutane-2,4-diyl III and
1,3-diphosphabicyclo[1.1.0]butane IV reported by Niecke
et al.[5] are the first and only known stretch isomers that
have been isolated and independently characterized. It is
important to note that because of a trans-annular anti-
bonding p overlap,the thermal ring closure of III into IV is
forbidden. Herein we report the first experimental evidence
for the existence of two bond-stretch isomers that features a
trans-annular bonding p-overlap,which allows for the ther-
mal ring closure and opening processes.
13.58),which suggests some delocalization between the 2p(B)
orbitals and the p-ring systems.
The magic-angle-spinning solid-state NMR spectrum of
the purple crystals shows a single 31P signal at d = 5.9 ppm,
which is comparable to that observed for 1 both in solution
and in the solid state,thus confirming the diradical structure
in the solid state. However,the solution state NMR spectra of
the same compound (Figure 2) at room temperature reveal a
single 31P signal at d = À28 ppm and an 11B resonance at d =
À9 ppm,which suggest a different structural preference.
These chemical shifts are very comparable to those observed
both in solution and in the solid state for the bicyclic
derivative 2 (same substituents at the P atom and duryl
instead of phenyl at the B atp,). As the 31P chemical shift was
found to be temperature dependent and moves towards a
lower field as the temperature decreased,these results as a
whole suggest a fast interconversion between a diradical
structure 6 and the corresponding bicyclic stretch isomer 7,
the latter being favored at higher temperatures. This hypoth-
esis is consistent with the observed changes in the NMR line
shape[12] as the solution is cooled. But instead of two slow-
exchange P signals,one for 6 and one for 7,three resonances
were observed at À1458C. The signal at d = 4.0 ppm can be
easily assigned to the open form, 6,whereas the signals at d =
We have recently reported the synthesis of a 1,3-dibora-
2,4-diphosphoniocyclobutane-1,3-diyl 1,[6–8] as well as several
1,3-dibora-2,4-diphosphoniobicyclo[1.1.0]butanes
2–5[9]
(Scheme 1); they differ by the nature of the boron and
phosphorus substituents and therefore are not bond-stretch
isomers. Derivatives 1 and 2–5 feature very different spectro-
scopic properties,which at first glance could be used as a
fingerprint for the diradical versus the bicyclic structure. As
observed for related carbon-based singlet 1,3-diradicals,[10]
compound 1 is strongly colored [lmax (toluene) = 446 nm,
e = 2200],whereas all the derivatives 2–5 are colorless.
Because the phosphorus is in a three-membered ring and
the boron is tetracoordinated,the 31P and 11B NMR signals for
2–5 appeared at a much higher field than those for 1.
In the course of a systematic study on the influence of the
nature of the substituents on the ground state structure of the
PBPB system,we have prepared a derivative that features iso-
propyl at phosphorus and phenyl groups at boron. This
compound has been isolated in 57% yield as very-air-
sensitive,but thermally highly stable purple crystals (m.p.:
1058C). In line with the strong coloration,X-ray diffraction
analysis[11] (Figure 1) revealed that in the solid state this
compound adopts a very similar structure to that observed for
À
1: a planar,almost square PBPB ring,with a very large B
B
interatomic distance of 2.57 Š. Interestingly,the phenyl rings
are almost coplanar to the PBPB skeleton (torsion angle
Figure 2. 31P spectra of the interconverting singlet diradical 6 and bicy-
clic stretch isomers 7 as a function of temperature. Spectra were
obtained on a 400 MHz (1H) Bruker DMX spectrometer equipped
with a 5 mm high-resolution double resonance probe and referenced
through an external solution of 1% phosphoric acid at 308C. The tem-
perature was measured by using an internal probe thermocouple cali-
brated versus a standard methanol chemical shift thermometer. The
signals at À20 ppm are tentatively assigned to impurities.
Figure 1. Molecular view of 6 in the solid state (H atoms are omitted).
Selected bond lengths: P1–B1, 1.8943Æ17 Š; P1-B1a, 1.8915Æ16 Š;
B1–C1, 1.557Æ2 Š; B1-P1-B1a, 85.23Æ78, P1-B1-P1a, 94.77Æ78, P1-
B1-C1, 129.47Æ118; P1a-B1-C1, 135.75Æ128.
Angew. Chem. Int. Ed. 2004, 43, 4880 –4883
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4881

Communications
À32.2 and À41.8 ppm could not be readily explained and
stretch form 6. This conclusion is confirmed by variable-
temperature UV/Vis spectroscopy,[16] which also tracks the
population shift from the colorless bicyclic form 7 to the
colored diradical species 6 as the temperature is decreased
(Figure 4).
prompted us to perform ab initio calculations.[13]
For the planar structure,local minima that correspond to
eight different arrangements of the isopropyl groups,includ-
ing the crystallographically observed conformer 6,were
found. The calculated 31P NMR chemical shift (+ 6.0 ppm)
agrees well with both the solid-state and the low-temperature
liquid-state NMR data. For the bicyclic structures,ten local
minima were found,among them 7a and 7b were the lowest
and equivalent in energy (Figure 3). Due to the presence of
Figure 4. UV/Vis spectra showing the preference for the colored
open form 6 over the colorless bicyclic form 7 as the temperature is
lowered.
It can be concluded that the order of stability of the bond-
stretch isomers 6 and 7 is strongly entropy driven. The
diradical isomer with the coplanar phenyl group has fewer
degrees of freedom than the bicyclic isomers in which free
rotation of the phenyl groups and inversion at boron are both
allowed. This is certainly a unique example of a reaction in
which the breaking of a s-bond is induced by decreasing the
temperature and the bond formation is entropically favored.
Figure 3. Optimized structures of the two most-stable bicyclic
conformers 7a and 7b.
two inequivalent phosphorous nuclei,AX systems are calcu-
lated at d = À36.3 and À43.1,and À31.9 and À40.3 ppm,for
7a and 7b,respectively. The observed signals of equal
intensity and line width at d = À32.2 and À41.8 ppm fit well
with a rapid low-temperature interconversion between 7a and
7b,although the participation of other conformers that
preserves the AX system cannot be excluded.
À
It is noteworthy that the B B interatomic distance between 6
[257 (exp),258 pm (calcd)] and 7 [186 pm (calcd)] varies by
40%. Combined with the phenomenon of temperature-
dependent interconversion,these results open interesting
perspectives for “molecular muscles”[17] as well as electrical
switch devices.[18]
As the temperature is raised above À1458C,the singlet
diradical and bicyclic bond-stretch isomers exchange. Initially,
the signals at high field broaden more rapidly due to the
inequivalent exchange,which favors 6 over 7 by a 3:1 ratio at
À1458C. Rates extracted from the line width of the diradical
peak in the initial broadening regime (from À145 to À1108C)
give a free energy of activation[14] of 6.6 Æ 1.8 kcalmolÀ1 at
À1308C for the pathway between 6 and 7. The transition
through intermediate exchange (À110 to À858C) is somewhat
convoluted as the equilibrium constant 6/7 also changes
dramatically in this region with preference switching from 6 to
the bicyclic forms 7 (1:2 at À858C). Above À658C,only a
single fast-exchange resonance is observed with a temper-
ature-dependent chemical shift that reflects the changing
populations of the bond-stretch isomers,which reaches a 6/7
ratio of 1:7 at room temperature. We can extract a free-energy
difference between the stretch isomers from these data by
using a three-site exchange model,which gives DH = 1.4 Æ
0.2 kcalmolÀ1 (6 being the most stable isomer) and DS = 7.2 Æ
1.6 calmolÀ1 K. Although,ab initio calculations of the energy
difference between the biradical and bicyclic forms favor the
bicyclic isomer by 0.8 kcalmolÀ1,this discrepancy is within the
limit of error expected for such a comparison.
Experimental Section
All manipulations were performed under argon by using standard
Schlenk techniques. Dry,oxygen-free solvents were employed.
Synthesis of [(iPr)2PB(Cl)Ph]2: (iPr)2PSiMe3 (3.05 g,16.05 mmol)
was added to a solution of commercially available PhBCl2 (2.55 g,
16.05 mmol) in toluene (30 mL) at À808C. The reaction mixture was
heated overnight at 1008C. All the volatile products were removed
under vacuum. Single crystals were obtained by cooling a saturated
solution of product in toluene to À308C. m.p.: 2688C,decomp;
13C{1H} NMR (125.8 MHz,CDCl 3): d = 132.9,127.7,126.8 (s,C aro),
24.4 (pseudo-t, JPC = 15.2 Hz,PCH),21.4 (s,CH CH3),20.3 ppm (s,
CHCH3), ipso-C atoms are not observed; 1H NMR (300 MHz,
3
CDCl3): d = 7.72 (d, JHH = 6.0 Hz,4H,Ph- o-CH),7.23 (m,6H,Ph-
3
m,p-CH),3.02 (m,4H,P CH),1.28 (dd, JHP = 13.8 Hz, JHH = 7.2 Hz,
12H,CHC H3),0.75 ppm (dd,
J
HP = 13.8 Hz, 3JHH = 7.2 Hz,12H,
CHCH3); 31P{1H} (CDCl3) d = À5.2 ppm; 11B{1H} (CDCl3) d =
À1ppm.
Synthesis of derivative 6/7: A freshly prepared solution of Li-
naphtalene (6.4 mL,0.8 m,thf) was added dropwise to a toluene
solution (15 mL) of [(iPr)2PB(Cl)Ph]2 (2.5 mmol) at À808C. The
reaction mixture was warmed to room temperature and stirring was
maintained for about 30 minutes. The solvents were immediately
removed under vacuum and the residue was dissolved in pentane
(30 mL). Salts were removed by filtration and pentane was removed
under vacuum. Naphtalene was sublimed by heating to 808C under
vacuum for 30 minutes. Purple single crystals (57% yield) were
obtained by cooling saturated solutions of product in pentane to
À308C. m.p.: 1058C; 13C{1H} NMR (125.8 MHz,C 7D8): d = 143.8 (br,
We note the excellent agreement of the limiting chemical
shifts for the diradical 6 and bicyclic forms 7 with the related
compounds 1 and 2,respectively,as well as the agreement
with the ab initio calculated shifts. These results give us
confidence that the observed dynamic NMR behavior is not
due to isopropyl group rotations[15] within the planar bond-
4882
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 4880 –4883

Angewandte
Chemie
i-Caro),136.3,128.5,126.3 (s,C aro),28.6 (pseudo-t, JPC = 22 Hz,PCH),
21.4 ppm (s,CH CH3); 1H NMR (300 MHz,CDCl 3): d = 7.48 (d,
3JHH = 7.2 Hz,4H,Ph- o-CH),7.01 (pseudo-t, 3JHH = 7.2 Hz,4H,Ph-
SAINTPLUS program. The Bruker SHELXTL program[19c] was
used for direct methods of phase determination and structure
refinement. Atomic coordinates,isotropic and anisotropic dis-
placement parameters of all the non-hydrogen atoms of the two
compounds were refined by means of a full-matrix least-squares
procedure on F2. All H atoms were included in the refinement in
calculated positions riding on the atoms to which they were
attached. C24H38B2P2, Mr = 410.10,crystal size 0.51 ” 0.30 ”
0.15 mm3,monoclinic,space group P21/c, a = 8.8947(12) Š, b =
m-CH),6.89 (t, 3JHH = 7.2 Hz,2H,Ph- p-CH),1.73 (d. sept., JHP
4.2 Hz, 3JHH = 7.2 Hz,4H,P CH),0.80 ppm (dd, JHP = 16.5 Hz, 3JHH
7.2 Hz,24H,CHC H3).
=
=
Received: April 28,2004
11.1419(15) Š,
c = 12.6125(18) Š,
b = 102.553(3)8,
V=
(l =
Keywords: boron · NMR spectroscopy · phosphorus · radicals ·
UV/Vis spectroscopy
1220.1(3) Š3, 1calcd = 1.116 gcmÀ3
,
2qmax = 52.748,Mo
.
Ka
0.71073 Š),low temperature = 223(2) K,total reflections col-
lected = 6967,independent reflections = 2474 (Rint = 0.0216,
Rsig = 0.0239,redundanc y = 2.8,completeness 100%) and 2151
(86.9%) reflections were greater than 2s(I),index ranges À10 %
h % 11, À13 % k % 13, À15 % l % 9,absorption coefficient m =
0.186 mmÀ1; max/min transmission = 0.9727/0.9112,150 param-
eters were refined and converged at R1 = 0.0359, wR2 = 0.0954,
with intensity I > 2s(I),the final difference map was 0.389/
À0.143 eŠÀ3. CCDC-237053 contains the supplementary crys-
tallographic data for this paper. These data can be obtained free
of charge via www.ccdc.cam.ac.uk/conts/retrieving.html (or from
the Cambridge Crystallographic Data Centre,12,Union Road,
Cambridge CB21EZ,UK; fax: ( + 44)1223-336-033; or deposit@
ccdc.cam.ac.uk).
[1] W. D. Stohrer,R. Hoffmann, J. Am. Chem. Soc. 1972, 94,779.
[2] a) G. Parkin, Chem. Rev. 1993, 93,887; b) J. A. Labinger, C. R.
Chim. 2002, 5,235.
[3] The question of bond-stretch isomerism has been considered
theoretically with different combinations of heteroatoms. For
examples,see: a) P. V. Sudhakar,O. F. Guner,K. Lammertsma,
J. Phys. Chem. 1989, 93,7289; b) J. A. Boatz,M. S. Gordon, J.
Phys. Chem. 1989, 93,2888; c) P. V. Sudhakar,K. Lammertsma,
J. Phys. Chem. 1992, 96,4830; d) T. Kudo,S. Nagase, J. Phys.
Chem. 1992, 96,9189.
[4] M. M. Rohmer,M. Benard, Chem. Soc. Rev. 2001, 30,340 – 354.
[5] a) E. Niecke,A. Fuchs,M. Nieger, Angew. Chem. 1999, 111,
3213; Angew. Chem. Int. Ed. 1999, 38,3028; b) E. Niecke,A.
Fuchs,F. Baumeister,M. Nieger,W. W. Schoeller, Angew. Chem.
1995, 107,640; Angew. Chem. Int. Ed. Engl. 1995, 34,555; c) O.
Schmidt,A. Fuchs,D. Gudat,M. Nieger,W. Hoffbauer,E.
Niecke,W. W. Schoeller,W. W. Angew. Chem. 1998, 110,995;
Angew. Chem. Int. Ed. Engl. 1998, 37,949; d) W. W. Schoeller,C.
[12] a) Dynamic Nuclear Magnetic Resonance Spectroscopy (Eds.:
L. M. Jackman,F. A. Cotton),Academic Press,New York, 1975;
b) Dynamic NMR Spectroscopy (Ed.: J. Sandstrom),Academic
Press,London, 1982.
[13] Structures minimized (B3LYP/6-31G(d)) and chemical shifts
calculated (B3LYP/6-311G(2d,p)) in the Gaussian98 (Revi-
sionA.7),M. J. Frisch,G. W. Trucks,H. B. Schlegel,G. E.
Scuseria,M. A. Robb,J. R. Cheeseman,V. G. Zakrzewski,
J. A. Montgomery,R. E. Stratmann,J. C. Burant,S. Dapprich,
J. M. Millam,A. D. Daniels,K. N. Kudin,M. C. Strain,O. Farkas,
J. Tomasi,V. Barone,M. Cossi,R. Cammi,B. Mennucci,C.
Pomelli,C. Adamo,S. Clifford,J. Ochterski,G. A. Petersson,
P. Y. Ayala,Q. Cui,K. Morokuma,D. K. Malick,A. D. Rabuck,
K. Raghavachari,J. B. Foresman,J. Cioslowski,J. V. Ortiz,B. B.
Stefanov,G. Liu,A. Liashenko,P. Piskorz,I. Komaromi,R.
Gomperts,R. L. Martin,D. J. Fox,T. Keith,M. A. Al-Laham,
C. Y. Peng,A. Nanayakkara,C. Gonzalez,M. Challacombe,
P. M. W. Gill,B. G. Johnson,W. Chen,M. W. Wong,J. L. Andres,
M. Head-Gordon,E. S. Replogle,J. A. Pople,Gaussian,Inc.,
Pittsburgh,PA, 1998.
[14] P. Pechukas, Annu. Rev. Phys. Chem. 1981, 31,159.
[15] Distinct conformers due to isopropyl rotations have been
observed in slow exhange by NMR in unrelated systems. For a
recent example: J. E. Anderson,A. de Meijere,S. I. Kozhush-
kov,L. Lunazzi,A. Mazzanti, J. Am. Chem. Soc. 2002, 124,6706.
[16] The same approach has been elegantly used to provide evidence
for the existence of bishomoaromatic semibullvalene. a) H.
Quast,M. Seefelder, Angew. Chem. 1999, 111,1132; Angew.
Chem. Int. Ed. 1999, 38,1064; b) A. C. Goren,D. A. Hrovat,M.
Seefelder,H. Quast,W. T. Borden, J. Am. Chem. Soc. 2002, 124,
3469.
Begemann,E. Niecke,D. Gudat,
J. Phys. Chem. 2001, 105,
10731; e) M. Sebastian,M. Nieger,D. Szieberth,L. Nyulaszi,E.
Niecke, Angew. Chem. 2004, 116,647; Angew. Chem. Int. Ed.
2004, 43,637.
[6] a) D. Scheschkewitz,H. Amii,H. Gornitzka,W. W. Schoeller,D.
Bourissou,G. Bertrand, Science 2002, 295,1880; b) H. Amii,L.
Vranicar,H. Gornitzka,D. Bourissou,G. Bertrand,
J. Am.
Chem. Soc. 2004, 126,1344.
[7] For calculations see: a) M. Seierstad,C. R. Kinsinger,C. J.
Cramer, Angew. Chem. 2002, 114,4050; Angew. Chem. Int. Ed.
2002, 41,3894; b) W. W. Schoeller,A. Rozhenko,D. Bourissou,
G. Bertrand, Chem. Eur. J. 2003, 9,3611; c) Y. Jung,M. Head-
Gordon, J. Phys. Chem. A 2003, 107,7475; d) M. J. Cheng,C. H.
Hu, Mol. Phys. 2003, 101,1319.
[8] See also: a) H. Grützmacher,F. Breher, Angew. Chem. 2002, 114,
4178; Angew. Chem. Int. Ed. 2002, 41,4006; b) P. P. Power,
Chem. Rev. 2003, 103,789.
[9] D. Scheschkewitz,H. Amii,H. Gornitzka,W. W. Schoeller,D.
Bourissou,G. Bertrand, Angew. Chem. 2004, 116,595; Angew.
Chem. Int. Ed. 2004, 43,585.
[10] a) W. Adam,W. T. Borden,C. Burda,H. Foster,T. Heidenfelder,
M. Heubes,D. A. Hrovat,F. Kita,S. B. Lewis,D. Scheutzow,J.
Wirz, J. Am. Chem. Soc. 1998, 120,593; b) M. Abe,W. Adam,T.
Heidenfelder,W. M. Nau,X. Zhang, J. Am. Chem. Soc. 2000,
122,2019; c) M. Abe,W. Adam,M. Hara,M. Hattori,T. Majima,
M. Nojima,K. Tachibana,S. Tojo, J. Am. Chem. Soc. 2002, 124,
6540; d) M. Abe,W. Adam,W. T. Borden,M. Hattori,D. A.
Hrovat,M. Nojima,K. Nozaki,J. Wirz, J. Am. Chem. Soc. 2004,
126,574.
[17] M. J. Marsella, Acc. Chem. Res. 2002, 35,944.
[18] a) A. Rajca, Chem. Rev. 1994, 94,871; b) J. A. Berson, Acc.
Chem. Res. 1997, 30,238.
[19] a) SMART Software Reference Manual,Version 5.054,Bruker
Analytical X-Ray System,Inc.,Madison,WI,
1997–1998;
[11] The Bruker SMART-1000[19a] X-ray diffraction instrument with
Mo radiation was used for data collection of compound 6. All
data frames were collected by using w-scan mode (À0.38w-scan
width,hemisphere of reflections) and integrated by using Bruker
SAINTPLUS program.[19b] The intensity data were corrected for
Lorentzian polarization and absorption corrections were per-
formed by using the SADABS program incorporated in the
b) SAINTPLUS Software Reference Manual,Version 6.02A,
Bruker Analytical X-Ray System,Inc.,Madison,WI 1997–1998;
c) SHELXTL Software Reference Manual,Version 6.10,Bruker
Analytical X-Ray System,Inc.,Madison,WI, 2000.
Angew. Chem. Int. Ed. 2004, 43, 4880 –4883
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4883