Synthesis of Hypervalent Pentavalent Carbon and Boron Compounds

Yohsuke Yamamotoa and Kin-ya Akibab a Department of Chemistry, Graduate Schoolof Science,Hiroshima University, b Advanced ResearchCenter for Scienceand Engineering ,Waseda University,

ReceivedAugust 9, 2004

Abstract:Various carbon and boron compoundsbearing a 1,8-disubstitutedanthracene skeleton were syn- thesizedand characterizedby X-ray analysis.They showedthree typesof structuresbased on the kinds of substituents.The first one is symmetricaland is a loosepentacoordinate structure, which has the sp2carbon or boron atom and the two weak apicalinteractions. The next one is an unsymmetricaltetracoordinate structure,which has the sp3 central atom. The last one is symmetricaland is a tight pentacoordinatestruc- ture, which resultedfrom the specialfeature of the fluorine substituentson the boron. The existenceof hypervalentinteraction was provedby the Atoms In MoleculesTheory, experimental electron distribution analysisand a comparisonamong the structuresof tight and loosepentacoordinate species. The synthesis and structuresof hypervalentcarbon compoundswith a 2,6-bis(aryloxymethyl)benzeneligand are also dis- cussed.

mation of a sulfonium salt, where one of the phenylthio I. Introduction groups stays unaffected. They concluded that the desired

Pentacoordinate carbon compounds should be classified pentacoordinate carbon species should be the transition state into electron-deficient and electron-rich species based on the in the 'bell-clapper' (bond-switching) rearrangement between number of formal valence electrons around the central car- the two sulfonium salts. Forbus and Martin claimed the syn- bon. Electron-deficient pentacoordinate carbon compounds thesis and direct observation of a pentacoordinate carbon such as the methonium ion (CH5+),1 transition-metal com- species using the dication of the 1,8-bis(tolylthio)anthracenyl plexes bearing C-H agostic interaction or a bridging alkyl ligand, where the carbocation at C-9 is contained in a ben- group,2 carboranes,3 and a carbon atom in a metal cluster zene ring bearing the methoxy group at the 2•Œ,6•Œ-positions.8 cage4 have eight electrons formally assigned around the cen- The 13C NMR chemical shift of the central carbon (ƒÂ 109 tral carbon, and the pentacoordinate structure is made possi- ppm; dication) was shifted to higher-field from the reduced ble by the interaction between a sigma bond (2 center 2 elec- electronically neutral anthracene compound (ƒÂ 118 ppm). tron bond: 2c-2e) with a vacant orbital of proton and so Although they carefully tried to prove the existence of penta- forth. In contrast, electron-rich pentacoordinate carbon coordinate carbon compounds with NMR, which showed a species such as the transition states of the bimolecular nucle- symmetrical spectrum reasonable as a pentacoordinate ophilic substitution (SN2) reaction have ten electrons formally species in solution, the compound had not been character- assigned around the central carbon atom based on an inter- ized by X-ray analysis. In 1984, Lee and Martin reported action of a vacant 2p orbital of the central carbon atom with direct observation of hypervalent penta- (ƒÂB -20 ppm) and two lone-pair electrons (4 electrons) or an interaction hexacoordinate (ƒÂB -123 ppm) boron compounds with the between a vacant C-X ƒÐ* orbital and a lone pair of a nucle- use of "B NMR.9 The "B NMR chemical shift ((ƒÂB -123 ophile. The interaction leads to the formation of a linear 3 ppm) was the highest value ever observed for any known center 4 electron (3c-4e) "hypervalent"5 bonding system; boron compound with only first-row elements bonded to the thus, they are called hypervalent compounds. central boron atom. Then, the Hojo group attempted to pre-

Theoretical calculations showed that the transition state pare a pentacoordinate carbon compound by stabilizing the of the SN2 reaction should be D3h trigonal bipyramidal carbocation bonded at the C-1 of fluorene by coordination

(TBP). Due to the fundamental importance of SN2, there of the two methylthio groups at C-9.10 However, in the solid have been a variety of efforts to stabilize the TS and even to state the two C+-S bond distances were different, with the prepare model compounds of TS. Breslow undertook to shorter one (1.94A) close to that of a covalent C-S bond solve the problem using trityl cation derivatives bearing some (1.81 A), forming a sulfonium salt.11 Kudo detected CLi6, o-methylthiomethyl substituents.6 The compound came out which was proposed by the Schleyer group,12 by Knudsen- as sulfonium salts, and no positive evidence for the presence effusion mass spectrometry.13 However, the details of the of pentacoordinate carbon compounds could be found. bonding nature of CLi5 and CLi6 have not yet been clarified.

Martin and Basalay attempted to stabilize the pentacoordi- Recently we reported the synthesis and the crystal struc- nate carbon species by the 1,8-bis(phenylthio)anthracenyl lig- ture of 1,8-dimethoxy-9-dimethoxymethylanthracene mono- and.7 At room temperature, the peak of two methyl groups cation (1) as the first fully characterized model compound for on the carbocation at C-9 was equivalently observed. When the SN2 transition state.14a In addition, we reported the syn- the temperature was lowered, two singlet peaks of the two theses and X-ray structures of several carbon and boron 14b-e methyl groups were separately observed because of the for- compounds (2) bearing a newly synthesized rigid tridentate

1128 (62) J. Synth . Org . Chem ., Jpn . anthracene ligand (Scheme 1). Here we summarize our recent ried out with lithium 4,4' -di-tert-butylbiphenylide (LDBB) research, especially on the structures of the carbon and the followed by the reaction with BrCF2CF2Br. Since the reac- corresponding boron compounds with several newly synthe- tion gave 4 in up to 30% yield even under careful reaction sized tridentate ligands. The bonding nature between the conditions, we recently improved the route via a 9-trifluo- central carbon (or boron) and the two coordinating atoms is roacetoxy intermediate. The yield of 4 was improved to 51%. elucidated and discussed based on the several X-ray struc- The synthesis of the boron compounds (2) is straightforward tures with several different substituents, theoretical DFT cal- as illustrated in Scheme 1. culations, and the accurate experimental X-ray electron-den- For the synthesis of 9-bromo-1,8-bis(dimethylamino)- sity distribution analysis. anthracene (8) a totally different route was necessary.41d Bro-

mide substitution15 of commercially available dichloroan- 2. Syntheses of Ligand Precursors Bearing the Anthracene thraquinone followed by reductionl 6 afforded dibromoan- Skeleton and the Corresponding Carbon and Boron Compounds throne. Deprotonation followed by methylation gave 1,8- dibromo-9-methoxyanthracene (9) in 58% yield. The Pd(0)-

The synthesis of ligand precursors, 1,8-dimethoxy-9-tri- mediated coupling reaction with various nucleophiles fluoromethanesulfonyloxyanthracene (3) and 9-bromo-1,8- (Bu3SnNMe2, Me3SnNMe2, LiNMe2 and HNMe2) did not dimethoxyanthracene (4), is illustrated in Scheme 1. The give the expected 1,8-bis(dimethylamino)-9-methoxyan- introduction of a carbon atom at the 9-position was success- thracene (10); instead, only a mono-dimethylaminated prod- ful from 1,8-dimethoxy-9-trifluoromethanesulfonyloxyan- uct was obtained in most cases, and reduction of the C-Br thracene (3) by Pd-catalyzed CO insertion with DMSO as a bonds took place in some cases. However, 9 could be convert- solvent and Pd(PPh3)4 as a catalyst in 52% yield.14a The ed to 10 in 79% yield by heating a HNMe2-THF solution of expected cation (1) could be prepared by treatment of the 9 up to 150•Ž in a pressure-resistant vessel with a Ni(0)-cat- ester (5) with excess Me3O+BF4-. However, for introduction alyst, which is a modified Buchwald method.17 LDBB (lithi- of a boron atom, 9-bromoanthracene precursor (4) had to be um di-tert-butylbiphenylide) reduction of the methoxy group synthesized.14b 9-Bromo-1,8-dimethoxyanthracene 4 was at the 9-position of the anthracene skeleton worked well for synthesized by a selective C-O bond cleavage reaction from 10 to afford a novel anthracene ligand precursor, 9-bromo-

9-trifluoroacetoxy compound (6) as a key step (Scheme 1). 1,8-bis(dimethylamino)anthracene (8), in 51% yield after Initially, several reduction conditions were examined for reaction with BrCF2CF2Br.

9-phosphatel4b with a couple of one electron-reducing To synthesize an anionic hypervalent pentacoordinate reagents (K/NH3, lithium naphthalenide, etc.). The reduction carbon compound, we designed a novel ligand bearing two of the phosphate under K/NH3 (Birch reduction condition) deprotectable methoxymethoxy groups as a trianion equiva- gave a 9-protonated product, which should be generated by lent. Synthesis of 9-bromo-1,8-bis(methoxymethoxy)- the reaction of the 9-potassium derivative with ammonia. anthracene (12) was achieved with stepwise oxygenation with

To prevent the trapping of a proton, THE was chosen as an the oxaziridine reagent18 and gaseous O2 as illustrated in aprotic solvent, and the reduction of the phosphate was car- Scheme 3.14e

Scheme 1. Synthesis of new anthracene ligand precursors (4 and 5) bearing two oxygen atoms at 1,8-positions and their application to synthesis of hypervalent carbon (1) and boron (2) compounds

3 5 1

6 4 2

Vol.62 No.11 2004 (63) 1129 Scheme 2. Synthesis of a new anthracene ligand precursor (8) bearing two nitrogen atoms at 1,8-positions and its application for synthesis of boron compounds (7)

9 10

8 7

Scheme 3. Synthesis of new deprotectable ligand 12 and the anionic species (11)

12 13 11

After quantitative regeneration of the lithium derivative the plane of the anthracene, one of the lone pairs of each by the reaction of 12 with one equivalent of n-BuLi in ether, oxygen atom should be directed toward the empty p-orbital 9-lithio-1,8-bis(methoxymethoxy)anthracene was reacted of the central carbocation at the 9-position and the other with gaseous hexafluoroacetone to give the adduct alcohol. should be parallel to 7r-electrons of the anthracene. There- After deprotection and deprotonation with KH in the pres- fore, the geometry around the central carbon atom is TBP, ence of 18-crown-6 the desired anion species 11 was generat- which is only slightly distorted. The two C-O distances are ed (Scheme 3). almost identical [2.43(1) and 2.45(1)A], which are signifi- cantly longer than that of a covalent C-O bond (1.43 A)12 3. Structures of Hypervalent Pentacoordinate Carbon (the ratio of the observed C-O apical length of 1 to the sum Compound 1 of the covalent radius was 1.71), but shorter than the corre- The X-ray structure of 1 is shown in Figure 1. The coun- sponding apical C-O bond (2.64 A) of [H2O-C(CH3)3- teranion is B2F7- unexpectedly, but it is well separated from OH2]+, which is theoretically proposed as the transition state the cationic part. The structure clearly shows the symmetrical of SN2.19 Thus, the C-O bond in 1 is more tight than the nature of the compound. The sum of the angles (C9-C9-O3, theoretical model (the ratio of the apical C-O bond in [H2O- C9-C19-O4, and O3-C19-04) around the central carbon is C(CH3)3-0H2]+ to the sum of the covalent radius is 1.84).19

360.O•‹, indicating that the carbon is planar with sp2 To determine the 13CNMR chemical shift of the central hybridization. The angles around the oxygen atoms of the carbon of 1 decisively, 13C-I was synthesized. Esterification methoxy groups at the 1,8-position of anthracene are of 3 proceeded under 13C-enriched CO atmosphere to give 13Clabeled ester (13C 119.2(9)•‹ (CI-O1-C) and 120.3(9)•‹ (C8-O2-C16), showing -5) . The ester was similarly methylated that both oxygen atoms have sp2 hybridization. Since the car- by Me30+ BE: to give a 13C-1. In the 13CNMR spectra, the bon atoms of the methoxy groups at the 1,8-positions are in chemical shift of the central carbon shifted from (5 172.30

1

Figure 1. ORTEP drawing of 1 (10-C-5) (30% thermal ellipsoid).

1130 (64) J. Synth . Org . Chem ., Jpn . Table 1. Comparison of distances of some anthracene derivatives (13C-5) to 6 192.58 (13C-1) ppm. The difference in the chem- ical shifts (A6=20.28) was comparable to the difference (ƒ¢ƒÂ=

21.3) from methyl formate (ƒÂ 171.52)20 to the dimethoxymethyl cation (ƒÂ 192.8).21 Hence, the chemical shift change in 13C-1 cannot be definite evidence for the formation of the pentavalent carbon species (10-C-5).

4. Elucidation of the Attractive C-O Interaction in 1

In order to elucidate the property and the degree of inter- action between the central carbon atom and the two oxygen atoms in 1, the side view of the crystal structures and selected distances of some derivatives are shown in Figure 2 together with 1. In the structure of the tribromo derivative, the three than the experimental distances (2.44(1)A). It should be bromine atoms are placed out of plane of the anthracene noted that the bond paths are found between the central car- skeleton in order to avoid large steric repulsion. In the struc- bon atom and the two oxygen atoms, clearly showing that ture of the triflate, the two methoxy oxygens are only slightly these atoms are bonded.23 The bond is weak and ionic based shifted from the plane; however, triflate oxygen is apparently on the small p(r) value of 0.022 e/a03 (ao = 0.529177 A) and shifted from the anthracene plane, probably due to the lone- the small positive V2p(r) value of 0.080 e/a05 at the bond crit- pair repulsion among the three oxygens. In the case of the ical points. These values are consistent with those of axial nitrile and the cation 1, the atoms attached to the 1, 8, and 9 C-H bonds in CH5- (p(r): 0.067 e/a03, •Þ2p(r): 0.009 e/a05).24 positions are almost perfectly in the plane of the anthracene. A large value (0.219) of the ellipticity of the C-O bond

The comparison of the distances between the atom at the 9- reflects the electron donation to the central carbocation from position of the anthracene and the atoms at the 1,8-positions the two lone pairs of the oxygen atoms of the methoxy is also worthy of note (Table 1). In the cases of the tribromo groups at the 1,8-positions within the anthracene plane. In derivative and the triflate the averaged distances of a and conclusion, the anthracene carbocation 1 is the first fully d'(3.2698(6) A, 2.572(2) A) are longer than those of b and characterized hypervalent pentacoordinate 10-C-5 com- b'(2.566(6) A, 2.550(3) A), while in the nitrile, the averaged pound for a model of the transition state of the SN2 reaction. distance of a and a' (2.531(3) A) is comparable with that of b Although we have not yet been successful in obtaining an and b' (2.540(5) A). In contrast, in 1 the averaged distance of X-ray structure of other hypervalent pentacoordinate carbon a and a' (2.44(1) A) is clearly shorter than that of b and compounds bearing the anthracene skeleton, recently we were b'(2.51(2) A). Therefore, it can be concluded that the interac- able to obtain many hypervalent pentacoordinate boron com- tion between the central carbon atom and the two oxygen pounds including a nitrogen-donating system using the same atoms in 1 should be clearly attractive. anthracene skeleton.

To elucidate the nature of the interaction between the 5. Synthesis and Structures of Boron Compounds Bearing central carbon atom and the two oxygen atoms at the an Anthracene Skeleton 1,8-positions of the anthracene skeleton, the geometry of 1 was fully optimized with the hybrid nonlocal density func- The synthesis of the boron compounds is illustrated in tional theory (DFT) at the B3PW91 level using the Gaussian Schemes 1 and 2. The distances between the central atom

98 program.22 The structural optimization afforded the sym- and the two coordinating atoms are shown in Table 2, and metrical Cs structure of 1 as the energy minimum. The two the numbering of atoms in 2 and 7 is shown in Figure 3.

C-O distances are identical (2.472 A) and slightly longer

3 1

Figure 2. Comparison of side view of anthracene compounds.

Vol.62 No.11 2004 (65) 1131 2 7

7b 7c 2a 2b 2c 2d 2e 2f 7a

Figure 3. The numbering of atoms in 2 and 7.

The boron compounds (2) with the OMe-ligand system B-O distances in 2e and 2f [4.59AA 2e and 4.48A for 2f] showed pentacoordinate structures, which are similar to the was shorter than the distance between the two oxygen atoms carbocation 1 shown above. Only one exception was the of 1,8-dimethoxyanthracene [4.752(3)A], which may be due

dichloro substituted boron compound 2f, which showed a to the stronger interaction between the central boron atom tetracoordinated boron structure (Table 2). The sum of the and the two oxygen atoms in 2e and even in 2f. Surprisingly, bond angles around the central boron atom of 2a-2c is although all the structures of 2d-2f were quite different, the 11B 360.0•‹, which indicates that the central boron atom is planar NMR chemical shifts of 2d-2f are almost similar (δ with sp2 hybridization. Thus, one of the lone pairs on the 2733 for 2d, 20-24 for 2e and 2225 for 2f). Such insensitiv-

oxygen atoms at the 1,8-positions in 2a-2c interacts with the ity of NMR chemical shifts to the structural difference has empty p orbital of the central boron atom, which can be never been observed in common hypervalent main group ele-

regarded as a slightly distorted TBP structure. The two B- O ment compounds. lengths (B1-O1, B1-O2) are identical (2.436(2) A) in 2a, The boron compounds can be classified to three types of

2.379(2) and 2.441(2)AA 2b, and 2.398(4) and 2.412(4)A in structures based on the difference in B-O distances, that is, 2c. The lengths are longer than those of the B1-O3 and B1-O4 loose pentacoordinate boron compounds (two relatively long

(1.397(2) and 1.403(2)AA 2b, 1.394(3) and 1.400(3)A in 2c) similar B-O lengths, 2a-2d), tight pentacoordinate boron but shorter than the sum of the van der Waals radii (3.48 compound (two relatively short similar B-O lengths, 2e) and A).12 All of the boron compounds bearing the two OMe- tetracoordinate boron compounds (two very different B- O

donating skeleton including the unsymmetrical 2f showed a lengths, 2f, 7a-7c). In loose pentacoordinate boron com- symmetrical anthracene pattern (two doublets, a triplet and a pounds (2a-2d), the B-O distances were similar to the 1,9- singlet) and a very sharp singlet for the two OMe groups peri-distance (2.46(2) A)26 of nonsubstituted anthracene and

(6H), which indicated that the two OMe groups were equiva- the O-(B)-O distance was slightly longer than the distance lent in their 1H NMR spectra even at -90•Ž. The 11B NMR between the two oxygen atoms of 1,8-dimethoxyanthracene spectra of the compounds 2a (6 55-68), 2b (5 28-39), and 2c [4.752(3) A. The distances indicate that the B-O interaction

(ƒÂ 30-37) showed a broad peak over the range of normal cat- was relatively weak. In tight pentacoordinate boron com- echolborane derivatives (cf. phenylcatecholborane: 6=32.1 in pound 2e, the B-O distances were clearly shorter than that of THF),25 indicating that the interaction between the central the loose one and the O-(B)-O distance (4.59A) was shorter boron atom and the oxygen atoms at the 1,8-positions is very than the distance [4.752(3)A] between the two oxygen atoms

weak. In contrast, upfield shifts were observed in the 11B of 1,8-dimethoxyanthracene. These apparently short B- O NMR spectra of the boron compounds prepared by Lee and distances of 2e may prove that the B-O hypervalent interac-

Martin (-20 to -41 ppm).9 The difference in the chemical tion of 2e is strengthened from the loose pentacoordinate shifts between the compounds of Lee and Martin and 2a-2e species and the hypervalent interaction really exists. In tetra-

may be a consequence of the charge difference (the former coordinate boron compounds (2f, 7a-7c), the shorter B-O being anionic systems, while 2a-2e are electronically neutral (or N) lengths were similar to the sum of the covalent radii,12 systems). The X-ray structure of 2d showed similar B-O dis- on the other hand, the longer B-O (or N) lengths were slight- tances [2.44 A; average of the three independent molecules] ly shorter than the sum of van der Waals radii or close to the

to those of catecholate derivatives 2a-2c. Although the sum. These distances indicate that one of the two OMe (or structure of 2e was also symmetrical, their B-O distances NMe2) groups interacted strongly with the central boron were shortened [2.29 A; average of the two independent atom, and the other one interacted very weakly with the cen- molecules]. On the other hand, the structure of 2f was quite tral boron atom or did not interact at all. Based on the clas- unsymmetrical, and the B-O distances [1.686(7) and 2.790(7) sification by bond lengths and configuration of the central A] were similar to those of the derivatives 7a-7c bearing the boron atom, our carbon compounds 1 (cationic species) and

two NMe2 groups at the 1,8-positions of anthracene. It is 11 (anionic species) were classified as a loose pentacoordinate interesting to compare the sum of the B-0 distances in 2a- group and a tetracoordinate group, respectively. 2c. The sum of the B-O distances of 2b was 4.881 A (aver- The apparent difference between the oxygen-donating age), which was similar to the distance between the two skeleton and the nitrogen-donating one is probably due to hydrogen atoms at the 1,8-positions of nonsubstituted the stronger stabilizing energy by the formation of one B-N anthracene [4.92 A]26 and longer than the distance between bond in comparison with that of one B-O bond. Hence, the the two oxygen atoms at the 1,8-positions of 1,8- large stabilizing energy of the B-N bond overrides the desta- dimethoxyanthracene [4.752(3)A]. However, the sum of the bilizing energy by the formation of the strained five-mem-

1132 (66) J. Synth . Org . Chem ., Jpn . Table 2. Classification of the compounds based on the X-ray structures

bered tetracoordinate boron structure. The F3B-NMe3 tetracoordinate dichloroboron 7c and the pentacoordinate bonding energy (26.6 kcal/mol) was reported to be much one 7c1, which should be the transition state of the bond- stronger than the corresponding F3B-OMe2 bonding energy switching process, must be very small. The small activation

(13.9 kcal/mol).27 energy in 7c indicates that our newly prepared rigid The discussion is consistent with the X-ray crystallo- anthracene ligand system stabilizes the pentacoordinate graphic analysis of 7a-7c, which showed the unsymmetrical dichloroboron transition state (7c). structures with coordination of only one NMe2 group toward the central boron atom. The shorter N-B bond lengths are Scheme 4. Very rapid N-B bond-switching equilibrium in 7c

1.809(2) A in 7a, 1.739(2) A in 7b, and 1.664(3) A in 7c, and the longer N-B bonds are 2.941(2) A in 7a, 3.124(3) A in 7b, and 3.129(3)A in 7c. Therefore, the shortest N-B bond dis- tance is observed in 7c, where the two electronegative chlo- rine substituents on the central boron atom should lead to a 7c 7c‡ 7c strong N-B bond.

But even in 7c, the destabilizing energy by formation of the strained five-membered tetracoordinated structure lessens the energy difference between the unsymmetrical tetracoordi- The difference among the substituents on the boron, nate structure and the symmetrical pentacoordinate one. In namely OMe (2d), F (2e), and Cl (2f in the oxygen-donating fact, the variable temperature 1H NMR (CDCl3 or CD2Cl2) ligand, can be explained by a similar delicate balance of 7a-7c showed symmetrical anthracene patterns in the aro- between the stabilizing energy by the formation of two rela- matic region (two kinds of doublets, a triplet, and a singlet) tively weak B-O bonds and the stabilizing energy by the for- and a sharp singlet signal of the two NMe2 groups at room mation of one strong B-O bond and the destabilizing energy temperature (6 2.65 in 7a, 2.83 in 7b, and 3.04 in 7c in by pyramidalization. It has been well established by several CDCl3). The peaks maintained their sharpness and the sym- different experimental methods that the Lewis acidities of the metrical pattern even at -80•Ž. This is the case even in the boron trihalides are in the order of BF3 < BC13 < BBr3.29 This most unsymmetrical dichloro compound 7c (CD2C12). The order is unexpected in view of the decrease in electronegativi- fact contrasts with the behavior of BC12[2,6-(Et2NCH2)2 ty from F to Cl to Br, which predicts a decreasing positive C6H3]28 in 1H NMR, which showed two kinds of NEt2 groups charge density on the boron in the order of BF3>BCl3> in C6D6 or THE-d8 solvents at 25°C. The NMR data of 7c BBr3. In addition, theoretical calculation supported the indicates that the very rapid "bond-switching" accompanied experimental trend of the charge density of the central boron with the inversion at the central atom (bell-clapper rear- atom.3° The generally accepted explanation for the anoma- rangement)7 takes place in solution as illustrated in Scheme lous order of Lewis acidity is based on the strong back- 4. Since the energy barrier of the N-B bond-switching pro- donation of electrons from one of the lone pairs (2p orbital) cess in 7c was too small to be measured by the coalescence of the fluorine atom to the vacant 2p, orbital of the boron method, the energy difference between the unsymmetrical atom, which leads to some double-bond character for the

Vol. 62 No.11 2004 (67) 1133 B-F bond in BF3.29'30 In BCl3 and BBr3, this type of back- were found between the B-O bonds as shown in Figure 4. donation is much less important because the longer B-Cl and The shapes of the bond paths of the two bonds are almost

B-Br bonds form a poorer overlap. These results suggest that identical. The bond lengths of the two B-O bonds are the oxygen and the fluorine substituents donate electrons essentially the same but those of the B1-O2 bond are slightly toward the central boron atom stronger than the chlorine smaller than those of the other (B1-O1 2.4353(7), B1-O2 substituent. Indeed, in the case of 2d and 2e,both averages of 2.3791(7)A). The bond critical points are also found on each the B-X distances [1.36Afor 2d, X= OMe; 1.34A for 2e, X bond path, and the points are closer to the central boron

=F] were 88% of the sum of the covalent radii (1.54A for B- atom than the O atoms because of the polarities of the

O, 1.52 A for B-F),12 while the B-Cl distance (1.83 A, identi- bonds. The small electron densities (p(r): B1-O1 0.022, B1-O2 cal) of 2f was similar (98%) to the sum of the covalent radii 0.032 e/ao3) and the small positive Laplacian (V2(r): B1-O1

(1.87 A for B-C1).12 However, the Brinck group suggested 0.073, B1-O2 0.087 e/ao5) values at the bond critical points that the back-donation was not essentially important based indicate the electrostatic character of the bonds. The differ- on the data of the coefficient of HOMO, which corresponded ence of p(r) and V2(r) between B1-O1 and B1-O2 can be relat- to the pƒÎ-pƒÎ interaction between B and X.30a They conclud- ed to the difference between the bond lengths since the ed that the Lewis acidity of BX3 was related to their destabi- decrease in the p(r) and V2(r) values may imply weakening of lization energy on the pyramidalization, which was defined as the bond. On the other hand, the bond ellipticity, ƒÃ, values the energy difference between the sp2 and sp3 state of BX3 of the two B-O bonds are quite different (ƒÃ: B1-O1 0.09, B1- without a base, not to the charge density of the central boron O2 0.74).The difference in the e values indicates that the atom. That is, an increasing electronegativity from Br to Cl donation from the lone pair in the p orbital is more sensitive to F affects the polarity of the B-X bond; as a result, in to the distance than those in the sp2 orbital because of the addition to the B-X bond lengths, the repulsion of the par- directionality of the p orbital on the two oxygen atoms. The tial negative charge of X increased from Br to Cl to F when orders of these values are in agreement with the results from

BX3 was pyramidalized. the DFT calculation (p(r): 0.022 e/ao3, V2(r): 0.058 e/ao5, ε:

In our compounds 2e and 2f, the rigidity of the boron sp2 0.220) described below. plane of 2e may be stronger than that of 2f because of the repulsion of the partial negative charge in pyramidalization, too. In consequence, 2f was pyramidalized to obtain a larger stabilization energy by the formation of a single bond between the central boron atom and the oxygen atom, while

2e kept its planarity of the sp2 plane because a large stabiliza- tion energy by the formation of a single bond is not large enough to overcome the destabilization energy of pyramidal- ization. In the case of 2d and 2e, the planarity of 2d was strong probably due to the same reason as that of 2e. How- ever, a difference in the electronegativity between the fluorine atom and the methoxy group leads to the difference in the partial positive charge of the central boron atom, which directly affects the strength of the apical B -O interaction.

Because of the reason for the foiniation of tight pentacouidi- nate boron compounds, the synthesis of tight pentacoordi- Figure 4. The bond paths between the central boron atom and the nate carbon compounds using some substituents, which can two oxygen atoms at 1 and 8 positions and the static keep the planarity of the carbocation, and some electron- model map of 2b on the anthracene plane based on the withdrawing groups is currently under investigation. experimental detailed X-ray analysis. Contour interval is 0.1 e A-3. 6. Experimental Electron Density Distribution Analysis of

2b

7. Elucidation of the Interaction Between the Central The static model map of 2b is shown in Figure 4 based Boron Atom and Two Oxygen Atoms of 2b Based on on the experimental accurate X-ray electron density distribu- the DFT Calculation tion analysis. The character of the B-O bonds is clearly

shown in the static model map.31 The hole of the 2p orbital The attractive interaction between the central boron atom

appeared as electron-deficient regions at both sides of the and the oxygen atoms at the 1,8-positions is also confirmed central boron atom and has lobes extended perpendicular to by the hybrid nonlocal density functional theory (DFT) at the sp2 plane of the boron containing the catechol ligand. the B3PW91 level using the Gaussian 98 program.22 The

The lone pairs in the sp2 orbital on each O atom are on a co- optimized geometry of 2b is symmetrical, where the two B- O plane of the anthracene and directed to the empty 2p orbital bond lengths are identical (2.431A) and are slightly longer of the B atom. Such geometry of the orbitals clearly indicates than the experimental data (2.3791(7) and 2.4353(7)A). The

the electron donations from the lone pairs in the sp2 orbitals topological analysis of 2b shows that the bond path is found

on the 0 atoms to the empty 2p orbital on the central boron between the central boron atom and the two oxygen atoms.

atom to form a three-center four-electron bond. The B-O bond is weak and slightly ionic as shown by the

For quantitative treatment of the characterization of the small value of the electron density [p(r): 0.022 e/ao3] and the

bonds, topological analyses were performed. The bond paths small positive Laplacian value [•Þ2p(r): 0.058 e/ao5] at the

1134 (68) J. Synth . Org . Chem ., Jpn . bond critical points. These values together with the large cating that the carbon is planar with sp2 hybridization. The value of the ellipticity (E=0.220) are well consistent with the angle (O1-C1-O2) is 162.3(1)•‹ and the angle around the oxy- data from the accurate X-ray analysis and very similar to the gen atoms is 116.3(2)•‹ (C5-O1-C6 and C7-O2-C8), showing values for the C-O bond in 1 [p(r): 0.022 e/ao3, •Þ2p(r): 0.080 that both oxygen atoms have sp2 hybridization. Since the e/ao5, ƒÃ: 0.219]. The calculation also showed that the sym- central benzene ring and the atoms consisting of the hyperva- metrical pentacoordinated structure was a global minimum in lent carbon and the ligand system (O1, C1,O2, C2, C5, C6, C7,

2e, but in 2f both the pentacoordinated structure and the and C8) are almost completely in a plane, the sp2 lone pairs tetracoordinated one were found as minimums. The tetraco- of each oxygen atom should be directed toward the empty p-

ordinate structure was only 0.14 kcal/mol more stable than orbital of the central carbocation; thus, the geometry around

the pentacoordinated one in 2f. It should be noted that the the central carbon atom is TBP which is only slightly distort-

DFT calculation using B3LYP showed that the pentacoordi- ed. X-ray analysis of 14f-ClO4 showed two independent

nated structure was more stable than the tetracoordinate one molecules where the two C-O distances (1st molecule:

in 2f. 2.776(4), 2.855(5) A; 2nd molecule: 2.617(4), 2.827(4) A) are

significantly different in the two molecules, indicating the 8. Hypervalent Pentavalent Carbon Compounds with energy difference between the two molecules is quite small 2,6-Bis(p-tolyloxymethyl)benzene Ligand and the hypervalent bond becomes balanced and is yet still

Recently we succeeded in the synthesis and isolation of flexible. In addition, it is quite interesting to note that the

carbocations (14) bearing a 2,6-bis(p-tolyloxymethyl)ben- central carbon of 14g-ClO4 showed the tetracoordinate struc-

zene as a flexible ligand. The synthesis of 14 is illustrated in ture where only one of the two oxygen ligands is weakly

Scheme 5. The carbocations were somewhat unstable at coordinated with the central carbon atom and the other is

room temperature in solution but the dark green color per- free. Since the aromatic xanthylium cation should be more

sisted up to 0•Ž for several hours. Figure 5 shows the stable than the thioxanthylium cation (the Hammett a+ val-

ORTEP drawing of a single molecule of 14b-ClO4. X-ray ues: -0.78 for OMe and -0.60 for SMe), the C-O attractive

structural data of several derivatives are shown in Table 3. interaction in 14g-ClO4 should be weaker than that in 14f-

In 14b-ClO4, the bond distance between C1-O1 and C104. The X-ray results imply that 14g-ClO4 corresponds to

C1-O2 is the same (2.690(4) A), which is even longer than the initial stage of the interaction where the planar xanthyli-

those of the C-O distances {2.43(1), and 2.45(1)A)} in 1 and um cation changes the structure to tetracoordinate-like.

is significantly longer than that of a covalent C-O bond (1.43 According to the increase of interaction, the central carbon

A),12 but is still shorter than the sum of the van der Waals becomes a pentaccordinate structure as in 14f-ClO4 and 14b-

radius (3.25 A).12 The sum of the angles (C2-C1-C4, C2-C1- ClO4. The small inductive effect of the methoxy group was

C3, and C3-C1-C4) around the central carbon is 360.0•‹, indi- observed in 2,6-bis(p-methoxyphenyloxymethyl)benzene lig-

Scheme 5. Synthesis of a new tridentate ligand precursor, 1-bromo-2,6-bis(p-tolyloxymethyl)benzene, and its application to the synthesis of hypervalent carbon compounds

14 a) b)

Figure 5. ORTEP drawing of 14b-C104 (30% thermal ellipsoid) and the numbering of atoms.

Vol.62 No.11 2004 (69) 1135 Table 3. X-ray structures of carbocations bearing the 2,6-bis(p-tolyloxymethyl)benzene and 2,6-bis(p-methoxyphenyloxymethyl)benzene ligands

A.; Prakash, G. K. S.; Williams, R. E.; Field, L. D.; Wade, K. and since the C-O bond length of 14d-ClO4 (X=F) is elon- Hypercarbon Chemistry; John Wiley & Sons: New York, 1987. gated slightly (0.028 A) in comparison with 14b-ClO4. The (c) Olah, G. A.; Prakash, G. K. S.; Sommer, J. Superacids; John geometry of 14b-ClO4 was fully optimized by the hybrid Wiley & Sons: New York, 1985. (d) Marx, D.; Parrinello, M. Nature 1995, 375, 216. (e) Schreiner, P. R.; Kim, 5.-J.; Schaefer, nonlocal density functional theory (DFT) at the B3PW91 H. F.; Schleyer, P. v. R. J. Chem. Phys. 1993, 99, 3716. (f) Scuse- level using the Gaussian 98 program.22 The basis sets ria, G. E. Nature, 1993, 366, 512. (g) Olah, G. A.; Klopman, G.; employed were 6-31(d) for C, O, and H. The calculation Schlosberg, R. H. J. Am. Chem. Soc. 1969, 91, 3261.

indicates that the symmetrical C2 structure is the energy min- 2) (a) Braunstein, P.; Boag, N. M. Angew. Chem. Int. Ed. 2001, 40, 2427 and references therein. (b) Besveld, W. M.; Hitchcock, P. imum. The two C-O distances are identical (2.689 A) with B.; Lappert, M. F.; Liu, D. S.; Tian, S. Organometallics 2000, 19, the experimental distances (2.690(4) A). It should be noted 4030. (c) McGrady, G. C.; Turner, J. F. C.; Ibberson, R. M.; that the bond paths are also found between the central car- Prager, M. Organometallics 2000, 19, 4398. (d) Klosin, J.; Roof,

bon atom and the two oxygen atoms, clearly showing that G. R.; Chen, E. Y.-X.; Abboud, K. A. Organometallics 2000, 19, 4684. (e) Yu, Z.; Wittbrodt, J. M.; Heeg, M. J.; Schlegel, H. B.; these atoms are bonded. The bond is weak and ionic based Winter, C. H. J. Am. Chem. Soc. 2000, 122, 9338. (f) Chen, E. on the small value of the electron density (p(r)) (0.017 e/a.3) Y.-X.; Abboud, K. A. Organometallics 2000, 19, 5541. (g)

and the small positive Laplacian value (•Þ2p(r)) (0.049 e/a.5) Vanka, K.; Chan, M. S. W.; Pye, C. C.; Ziegler, T. Organometallics 2000, 19, 1841. (h) Chan, M. S. W; Ziegler, T. at the bond critical points. These values, including a large Organometallics 2000, 19, 5182. (i) Schrock, R. R.; Casado, A. value of the ellipticity (e) (0.090), are very similar to the val- L.; Goodman, J. T.; Ling, L.-C.; Bonitatebus, Jr. P. J.; Davis, ues for the C-O bond in 1 (p(r): 0.022 e/ao3, •Þ2p(r): 0.080 W. M. Organometallics 2000, 19, 5325. (j) Duncan, A. P.;

e/ao5, ƒÃ: 0.219). In conclusion, the 2,6-bis(p-tolyloxymethyl) Mullins, S. M.; Arnold, J.; Bergman, R. G. Organometallics 2001, 20, 1808. (k) Schumann, H.; Keitsch, M. R.; Demtschuk, benzene ligand stabilized carbocations with very weak but J.; Molankier, G. A. J. Organornet. Chem. 1999, 582, 70. (1) definite C-O attractive interaction to form hypervalent car- Schaverien, C. J. Organometallics 1994, 13, 69. (m) Chen, E. bon compounds (10-C-5). The X-ray structures of the pre- Y.-X.; Marks, T. J. Chem. Rev. 2000, 100, 1391. (n) Song, X.; Thornton-Pett, M.; Bochmann, M. Organometallics 1998, 17, sent sterically flexible system (in comparison with the 1004. (o) Kleinhenz, S.; Seppelt, K. Chem. Eur. J. 1999, 5, 3573. anthracene compounds) showed very interesting sensitivity to (p) Kulzick, M. A.; Andersen, R. A.; Muetterties, E. L.; Day, the electronic effect on the central carbon atom, illustrating a V. W. J. Organomet. Chem. 1987, 336, 221. (q) Schwartz, D. J.; model for the SN2 reaction coordinate. Ball, G. E.; Andersen, R. A. J. Am. Chem. Soc. 1995, 117, 6027. (r) Shin, J. H.; Parkin, G. Chem. Commun. 1998, 1273. (s) Chan, Acknowledgments The authors gratefully acknowledge the M. C. W; Cole, J. M.; Gibson, V. C.; Howard, J. A. K. Chem. contributions of many coworkers and collaborators, in partic- Commun. 1997, 2345. (t) Bursten, B. E.; Cayton, R. H. ular, Prof. Shigeru Nagase (Institute for Molecular Science), Organometallics 1986, 5, 1051. (u) Ozawa, F.; Park, J. W.;

Dr. Nozomi Takagi (Institute for Molecular Science), Prof. Mackenzie, P. B.; Schaefer, W. P.; Henling, L. M.; Grubbs, R. H. J. Am. Chem. Soc. 1989, 111, 1319. (v) Dawkins, G. M.; Fujiko Iwasaki (The University of Electro-Communications, Green, M.; Orpen, A. G.; Stone, F. G. A. J. Chem. Soc. Chem. Rigaku), Prof. Daisuke Hashizume (The University of Elec- Commun. 1982, 41. (w) Calvert, R. B.; Shapley, J. R. J. Am. tro-Communications, Riken), Dr. Makoto Yamashita (The Chem. Soc. 1978, 100, 7726. (x) Hamilton, D. H.; Shapley, J. R. Organometallics 2000, 19, 761. (y) Schultz, A. J.; Williams, J. M.; University of Tokyo), Mr. Yuji Moriyama (Nissan Chem.). Calvert, R. B.; Shapley, J. R.; Stucky, G. D. Inorg. Chem. 1979, This work was supported by a Grant-in-Aid for Scientific 18, 319. (z) Dutta, T. K.; Vites, J. C.; Jacobsen, G. B.; Fehlner, Research (Nos. 11304044, 12304044, 14340199) provided by T. P. Organometallics 1987, 6, 842. (aa) Adams, R. D.; Horvath, the Ministry of Education, Culture, Sports, Science and I. T. Prog. Inorg. Chem. 1985, 33, 127. 3) (a) Grimes, R. N. Carboranes; Academic Press, New York, Technology of the Japanese Government. 1970. (b) Olah, G. A.; Wade, K.; Williams, R. E. Electron Defi- cient Boron and Carbon Clusters; John Wiley & Sons: New York, References 1991. (c) Kabalka, G. W. Current Topics in the Chemistry of Boron; The Royal Society of Chemistry: Cambridge, 1994. (d) 1) (a) Olah, G. A.; Laali, K. K.; Wang, Q.; Prakash, G. K. S. Siebert, W. Advances in Boron Chemistry; The Royal Society of Onium Ions; John Wiley & Sons: New York, 1998. (b) Olah, G. Chemistry: Cambridge, 1997. (e) King, R. B. Boron Chemistry

1136 (70) J. Synth . Org . Chem ., Jpn . at the Millennium; Elsevier Science, Amsterdam, 1999. (f) David- mer, D.; Kraka, E. J. Am. Chem. Soc. 1983, 105, 5061. (d) son, M.; Hughes, A. K.; Marder, T. B.; Wade, K. Contemporary Bader, R. F. W J. Phys. Chem. (A) 1998, 102, 7314. Boron Chemistry; The Royal Society of Chemistry: Cambridge, 24) Carroll, M. T.; Gordon, M. S.; Windus, T. L. Inorg. Chem. 2000. 1992, 31, 825. 4) Scherbaum, F.; Grohmann, A.; Milner, G.; Schmidbaur, H. 25) Farfan, N.; Contreras, R. J. Chem. Soc. Perkin Trans. 2 1987, Angew. Chem. Int. Ed. Engl. 1989, 28, 463. 771. 5) (a) Musher, J. I. Angew. Chem. Mt. Ed. Engl. 1969, 8, 54. (b) 26) (a) Brock, C. P. Acta Crystallogr. Sect. B 1990, 46, 795. (b) Akiba, K.-y. Chemistry of Hypervalent Compounds; Wiley-VCH: Mason, R. Acta Crystallogr. 1964, 17, 547. New York, 1999. 27) Sana, M.; Leroy, G.; Wilante, C. Organometallics 1992, 11, 781. 6) (a) Breslow, R.; Garratt, S.; Kaplan, L.; LaFollette, D. J. Am. 28) (a) Schlengermann, R.; Sieler R.; Hey-Hawkins, E. Main Group Chem. Soc. 1968, 90, 4051. (b) Breslow, R.; Kaplan, L.; LaFol- Chem., 1997, 2, 141. (b) Toyota, S.; Oki, M. Bull. Chem. Soc. lette, D. J. Am. Chem. Soc. 1968, 90, 4056. Jpn. 1990, 63, 1168. 7) (a) Basalay, R. J.; Martin, J. C. J. Am. Chem. Soc. 1973, 95, 29) (a) Brown, H. C.; Holmes, R. R. J. Am. Chem. Soc. 1956, 78, 2565. (b) Martin, J. C.; Basalay, R. J. J. Am. Chem. Soc. 1973, 2173. (b) Shriver, D. F.; Swanson, B. Inorg. Chem. 1974, 10, 95, 2572. 1354. (c) Cotton, F. A.; Wilkinson, G. Advanced Inorganic Chem- 8) (a) Forbus, T. R., Jr.; Martin, J. C. J. Am. Chem. Soc., 1979, istry; John Wiley and Sons: New York, 1980. 101, 5057. (b) Forbus, T. R., Jr. Ph. D. Thesis, University of 30) (a) Brink, T.; Murray, J. S.; Politzer, P. Inorg. Chem. 1993, 32, Illinois, 1981. (c) Forbus, T. R., Jr.; Martin, J. C. Heteroatom 2622. (b) Robinson, E. A.; Johnson, S. A.; Tang, T.-H.; Gille- Chem. 1993, 4, 113. (d) Forbus, T. R., Jr.; Martin, J. C. Het- spie, R. J. Inorg. Chem. 1997, 36, 3022. (c) Robinson, E. A.; eroatom Chem. 1993, 4, 129. (e) Forbus, T. R., Jr.; Kahl, J. L.; Heard, G. L.; Gillespie, R. J. J. Mol. Struct. 1999, 485, 305. Martin, J. C. Heteroatom Chem. 1993, 4, 137. (f) Martin, J. C. 31) (a) Coppens, P. X-ray Charge Densities and Chemical Bonding; Science 1983, 221,509. Oxford University Press, New York, 1997. (b) Wilson, A. J. C. 9) (a) Lee, D. Y; Martin, J. C. J. Am. Chem. Soc. 1984, 106, 5745. International Tablesfor Crystallography, Vol. C; Kluwer Academic (b) Lee, D. Y., Ph. D. thesis, University of Illinois, 1985. Publishers, Dordrecht, 1992. 10) Hojo, M.; Ichi, T.; Shibato, K. J. Org. Chem. 1985, 50, 1478. 11) Schleyer, P. v. R.; Wurthwein, E.-R.; Kaufmann, E.; Clark, T. J. Am. Chem. Soc. 1983, 105, 5930. PROFILE 12) Dean, J. A. Lange's Handbook of Chemistry, 11th Ed.; McGraw- Hill: New York, 1973, pp 3-119-3-123. Yohsuke Yamamoto is Professor of 13) Kudo, H. Nature, 1992, 355, 432. organic chemistry at Hiroshima Univer- 14) (a) Akiba, K.-y.; Yamashita, M.; Yamamoto, Y; Nagase, S. J. sity. He was born in Mie in 1954 and Am. Chem. Soc. 1999, 121, 10644. (b) Yamashita, M.; Yamamo- received his B. Sc. (1977) and Ph. D. to, Y; Akiba, K.-y.; Nagase, S. Angew. Chem. Int. Ed. 2000, 39, degree (1982) from the University of 4055. (c) Yamashita, M.; Watanabe, K.; Yamamoto, Y; Akiba, Tokyo. He joined the department of K.-y. Chem. Lett. 2001, 1104. (d) Yamashita, M.; Kamura, K.; chemistry, Hiroshima University, as an Yamamoto, Y.; Akiba, K.-y. Chem. Eur. J. 2002, 8, 2976. (e) Assistant Professor in 1982 and was pro- Yamashita, M.; Mita, Y; Yamamoto, Y.; Akiba, K.-y. Chem. moted to Professor in 2001. In 1989- Eur. J. 2003, 9, 3655. 1990 he worked as a postdoctoral fellow 15) Benites, M. del R.; Fronczek, F. R.; Hammer, R. P.; Maverick, at Vanderbilt University with Professor A. W. Inorg. Chem. 1997, 36, 5826. J. C. Martin. In 1988, he received the 16) Biehl, R.; Hinrichs, K.; Kurreck, H.; Lubitz, W Mennenga, U.; Chemical Society of Japan Award for Roth, K. J. Am. Chem. Soc. 1977, 89, 4278. Young Chemists. His research interests 17) Wolfe, J. P.; Buchwald, S. L. J. Am. Chem. Soc. 1997, 119, 6054. are in the area of main group element 18) Davis, F. A.; Towson, J. C.; Vashi, D. B.; ThimmaReddy, R.; chemistry and oxidized porphyrins. McCauley, J. P., Jr.; Harakal, M. E.; Gosciniak, D. J. J. Org. Chem. 1990, 55, 1254. 19) Yamabe, S.; Yamabe, E.; Minato, T. J. Chem. Soc. Perkin Trans. 111983, 1881. Kin-ya Akiba is Professor of Advanced 20) SDBSWeb : http://www.aist.go.jp/RIODB/SDBS/menu-j.html Research Center for Science and Engi- 21) Olah, G. A.; Hartz, N.; Rasul, G.; Burrichter, A.; Prakash, G. neering, Waseda University and Profes- K. S. J. Am. Chem. Soc. 1995, 117, 6423. sor Emeritus of Hiroshima University. 22) Gaussian 98, Revision A.5, Frisch, M. J.; Trucks, G. W.; He was born in 1936 in Tokyo and Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. received his B. Sci. (1959) and Ph. D. R.; Zakrzewski, V. G.; Montgomery, J. A.; Stratmann, R. E., degree (1964) from the University of Jr.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniel, A. D.; Tokyo. He joined Hiroshima University Kudin, s, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, as a Professor in 1980 after being Asso- V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, ciate Professor in the University of C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Tokyo. He received Divisonal Award Cui, Q.; Morokuma, K.; Malick, D. K.; Rabuck, A. D.; (Organic Chemistry,1986) and Award Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; (1998) of the Chemical Society of Japan, Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, Research Award of Humboldt Founda- I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al- tion (Inorganic Chemistry,1997), Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.; Chugoku Culture Medal (1998), Medal Challacombe, M.; Gill, P. M. W; Johnson, B.; Chen, W; Wong, with Purple Ribbon (2000), and Reinvi- M. W.; Andres, J. L.; Gonzalez, C.; Head-Gordon, M.; tatioin of Humboldt Foundation (2004). Replogle, E. S.; Pople, J. A. Gaussian, Inc., Pittsburgh PA, His research interests are in the chem- 1998. istry of main group elements, hyperva- 23) (a) Bader, R. F. W. Atoms in Molecules-A Quantum Theory; lent compounds and heterocycles. Oxford University Press: Oxford, 1990. (b) Bader, R. F. W Chem. Rev. 1991, 91, 893. (c) Bader, R. F. W; Slee, T. S.; Cre-

Vol.62 No.11 2004 (71) 1137