a survey of v nmr spectroscopy · review a survey of 51v nmr spectroscopy dieter rehder institut...

REVIEW

A Survey of 51V NMR Spectroscopy

Dieter Rehder

Institut fur Anorganische undAngewandte Chemie der Universitat

Martin-Luther-King-Platz 6D 2000 Hamburg 13, FRG

I.

II.

Introduction

III.

IV.

V.

VI.

ShieldingA. Theory

ApplicationsChemical Shifts Related to ESR g-factorsCorrelations between s H7 and 31P Shift ValuesTemperature Dependence of S'XV Shielding

Nuclear Spin-Spin Coupling ConstantsA. General and TheoryB. DiscussionC. Correlation between NMR and ESR Coupling

Constants (TI -C5H5V(C0)3PZ3 and Fe(N0)2(PZ3)Br)

Line Widths in Isotropic MediaA. Exchange BroadeningB. Quadrupolar BroadeningQuadrupole Effects in Anisotropic MediaA. Liquid CrystalsB. Solids

Concluding Remarks

References

Page33

343435535557

58586367

686868717174

78

79

I. INTRODUCTION

Vanadium-51 is a suitable nucleusfor NMR experiments using both Four-ier transform (FT) and wide-lineequipment. Nuclear data are: naturalabundance 99.76 %, nuclear spin 7/2,magnetic moment 5.139 (in units ofthe nuclear magneton), sensitivityrelative to XH 0.382 at constant

magnetic field and 5.52 at constantfrequency. Line widths of s *V NMRsignals in solution vary from <2 Hz(for vanadium compounds of 0h pointsymmetry) to ca. 3000 Hz (correspond-ing to relaxation times of ca. 10 to0.1 ms), but usually cover the range20-200 Hz, reflecting comparativelysmall quadrupolar interactions, whichallow observation of sufficiently

Vol. 4, No. 1/2 33

Table 1. s1V NMR Data of Carbonylvanadium and Related Complexes

Complex8

[V(C0)6]~[V(C0)5L]"L = NH3

C

PH3PF3P(0R)3

P(NR2)3

P(alkyl)3P(t-Bu3)P(aryl)3«P(C5CPh)3As(0Et)3AsEt3

AsPh3d

Sb(0Et)3SbEt3SbPhg"BiEt3psb(P)psb(Sb)

[{V(C0)5}2ii-L]2-

L = P2Cy4

P2Me4

P2Ph4

As2Ph4

Sb2Ph4

t-dpedppaarphos(P)arphos(As)dpae

[V(C0)sCN]2"

cis-[V(C0)4L]"L = 2 PH3

2 PF32 P(OR)3

2 P(alkyl)32 PPh2MePCy3 + P(0Me)3

dppm

P2[5],dppp, dppb

P3> P4arphos, pabdpae, aabdpap

6( 5 1V) b

[ppm]

-1952

-1670-1850-1961-1910 to -1920-1780 to -1840-1850 to -1890-1738-1805 to -1850-1780-1862-1850-1800-1885-1900-1881-1743-1814-1861

-1889-1867-1850-1745-1740-1836-1823-1847-1823-1824

-1850',-1720c

-1830-1967-1880 to -1910-1720 to -1780-1671-1890-1590-1755 to -1830-1700, -1720-1730 to -1830-1738, -1772-1730, -1754-1651

lJ(3 1p_Slv)

[Hz]

1V0488355-370210-230185-230

190-350

220

220207210

214232250

130498365195-215

193,381

225190

220,240

References

26,43

30312620,26,31-3331,3320,26,31-33e20,31-33e202020,32,34202033a203434

353535,3635353737383838

12,30

312626,31,3326,31,3339e39,4037,39,4039,4039-4238,3938,3939

36 Bulletin of Magnetic Resonance

Table 1. Continued

Complex8

diarsasbdpsp

cis-[{V(C0)4}2y-L2]2-

L = P2Ph4

dpae

[{V(C0)4}2(AsPh2)2]2"

mer-[V(C0)3L]-L = p3

P4

PP3P2P4

fac-[V(C0)3L)L = P2P4

3 PF3

trans-[V(C0)2(PF3)4r[V(PF3)6][V2(C0)8(CN)4]

u-Cp2V2(C0)5

CPV(CO)4

CpV(C0)3LL = PH3

PF3P(0R)3

P(NR2)3

PR3'PR3

j

P(aryl)3d

PPhCl2As(0Et)3

AsEt3

AsPh3d

Sb(0Et)3SbEtaSbPh3BiEt3P2Me4

P2Cy4

P2Ph4

As2Ph4

psb(P)

6(51V)[ppm]

-1750-1785-1781

-1780-1781

-1810

-1720-1810-1729-1724-1734

-1778-1964

-1948-1946-760-1666-1534

-1380-1567-1450-1340-1376-1250-1300-1390-1310-1350-1230-1310-1460-1429-1125-1381-1363-1407-1157-1306

b

tototototo

to

-1500-1360-1420-1300-1326

-1260

XJ(3XP-51V)[Hz]

205

224525

506510

155427260-330200-230150-160110-160130-170140

165140157

References

3934,e

35,38

35

41,40,424246

4612

1226,30e26

3126,26,31,26,3931,39202020,202033a20353535,3534

39

36

4242

43

•

3931,39,31,

39,

45

L

36

39,44,454439,44,45

44,45

Vol. 4 , No. 1/2 37

Table 1. Continued

Complex8

(CpV(C0)3CN]~

{CpV(C0)3}2y-LL = MePhP-PPhMe

t-dpedppaSb2ph4

trans-[CpV(CO)2L]L = 2 PF3

2 P(0R)3

2 PMe3

2 PPhMe2

cis-[CpV(C0)2L]L = 2 PF3

2 P(alkyl)3

2 PPhMe2

2 P(0R)3

2 P(NMe2)3

2 PPh2MedppmP2[5]dppbP3arphos, pabdpaer aabdpapdiarsdpspasb

L = P2Me4

P2Cy4

MePhP-PPhMet-dpedppa

6( s lV) b

[ppm]

-1440

-1390-1362-1343-1325

-1573-1210 to -1270-990-903

-1608-1160 to -1250-1195-1190 to -1390-1340-1152-870-1110 to -1143-1360-1176-995, -1030-937, -921-1260-992-1042-1008

]-1325-1076-1070-1143-1281

[Hz]

165

446340-370210185

430160

280-300240108

References

26

35373735

2631,3331,3331

2631,4431,3331,3331,333939,4037,39,4039,4026,4138,3938,393939e34,39

3535353737

CpV(CO)2n2-dppa -51

CpV(C0)LL = P3

k

3 PF3

CpV(PF3)4

[CpV(C0)3H]-

-950 ̂-156fc-1475-1730

450448

37

k121246


Table 1. Continued

Complex8

[CpV(C0)2(H)L]-L = PF3

P(0Et)3

PPh2Et

CpV(C0)(N0)2CpV(N0)2PZ3

1

T13-C3H5V(CO)3LL = dppm

dppedppparphosdiars

Tl3-C3H4MeV(C0)3dppeTI 3 -CrotylV(C0)3 dppe

HV{C0)4LL = dppm

dppedpppdppbarphos

cp3

HV(C0)3LL = p 3

m

P4CP3

PP3P2P4

trans-[HV(C0)2L]

L = P4

P2P4

cis-[HV(C0)2pp3]

{HV(C0)4}2V-p4

V(C0)3N0(PMe3)2

V(C0)3N0(dppe)

fi(51V)b

[ppm]

-1691-1603-1355

-1294-973 to -1265

-1355-1492-1348-1413-1461-1391-1445

-1593-1690-1600-1553-1663-1608

-1640-1665-1528-1690-1485

-1658-1500 to -1800

-1541-1685-1332-1382

lj(31p.51v)

[Hz]

562458

400-655

180

74

References

eee

47,4847,48

393939eeee

39,4939,4939,4939,4946,4946

464646

, 46

46

4646

46465050

a) Abbreviations for ligands: pab = o-C6H4(PPh2)(AsPh2),psb = o-C6H4(PPh2)(SbPh2), asb = o-C6H4(AsPh2)(SbPh2),aab = o-C6H4(AsPh2)2, diars = o-C6H4(AsMe)2 ; p2[5] = phosphines forming che-late 5-rings such as o-C6H4(PPh)2 , Ph2P(CH2)2PPh2 andcis-Ph2PCH=CHPPh2; t-dpe = trans-Ph2PC=CPPh2, dppa = Ph2PC=CPPh2, arp-hos = AsPh2(CH2)2PPh2; Ph2E(CH2)nEPh2: E = P (n = 1: dppm, n = 2: dppe,

Vol. 4, No. 1/2 39

n = 3 : dppp, n = 4 dppb), E = As (n = 2: dpae, n = 3: dpap), E = Sb (n = 3:dpsp); p 3 = PhP(CH2CH2PPh2)2, cp3 = CH3C(CH2PPh2)3, p4 = [Ph2PCH2CH2PPhCH2]2,pp3 = P(CH2CH2PPh2)3, p2p4 = (Ph2PCH2CH2)2PCH2CH2P(CH2CH2PPh2)2. b) Upfieldof VOCI3 (neat or in CD2C12) 0.1 - 0.2 M in THF or THF/CH3CN 1:1 at roomtemperature, if not indicated otherwise, c) At ca. 200°K in liquid ammonia,d) Including oligodentate ligands with EPh2 groups (E = P, As, Sb). e)Unpublished. f) In CH3CN. g) Geometry unconfirmed (see reference 42). h)2J(51V-19F) = 10.3 Hz. i) Tolman's cone angle < 140°. j) Tolman's coneangle 170c k) This complex was formerly assigned trans-[CpV(C0)2p3](26,41) on the basis of IR data. An X-ray analysis of this complex, however,shows that actually three CO groups are substituted (51). 1) Data for avariety of CpV(N0)2L complexes, where L is a ligand containing a C, N, 0 or Sdonor/acceptor function, are also reported in Ref. 48. m) Erroneously char-acterized as {V(CO)3p3}2 in reference 41. n) XJ(51V-31P) + ^ ( " V - lllN).

jChtoro Complexes

VOCyOR).VOCI(OR)2

VOtORJj

VO2* Derivatives

Vanadates (*V),iso- andHeteropoly vanadates W)

Peroxovanadiuml+V)

CpVINOJjL

CpV(H)IC0)2L

HV(C0)6J.n

6(5 ty ) relative to VOCI3 (ppml 0 -500 -tooo -1500 -2000

Figure 1. 5XV chemical shifts in ppm relative to VOC13. Not included inthis chart is V0Br3 [+ 434 ppm (19)] and current work on sulfidocomplexes [with low-field shifts down to ca. + 1400 ppm (53)].


onal-pyramidal structure of the(V(CO)4_n Ln} moiety. Shielding forthese complexes goes down as far asto that of dicyelopentadienyldithio-formiatovanadium-(+III) compounds(52) and V0F3 (19), but does notreach the extremely low shieldingvalues for VOC13, V0Br3 (19) and sul-fidovanadium complexes (down to +1400ppm (53)). The chart in.Figure 1, onfirst sight, imparts the impressionof a correlation between shieldingand formal oxidation state. Thiswould imply significant contributionsto variations in the overall shield-ing arising from the local diamag-netic term; we have ruled out thispossibility in the previous section.Hence, electronic (and steric) fac-tors have to be taken into considera-tion to account for these variations.

Qualitative and semi-quantitativeMO considerations on CpV(C0)3PZ3 and[V(C0)5PZ3]" complexes (12,26,32,45)have shown that 6(51V) values can beused to classify the phosphineligands PZ3 according to their ligandstrength (0-donor and ir-acceptorability): the ligand strengthincreases with increasing negativeshift values, i.e. with increasingoverall shielding or decreasing par-amagnetic deshielding due to enhancedLUM0-H0M0 splitting AE (equation 7),

P(NR2)3~P(aryl)3

<P(0R)3

P(alkyl)3~PH3<PF3

This ordering which - except forarylphosphines - is quite similar tothe sequence made up on the basis ofother (e.g. IR) spectroscopic parame-ters, also reflects a decrease inmetal-contribution to MOs taking partin LUM0-H0M0 transitions. The unus-ual position of PPh3 (and relatedphosphines) in this series ascompared to its IR-spectroscopicposition (where PPh3 is placedbetween alkyl- and alkoxyphosphines)has been explained by direct ligand-to-ligand interaction, making PPh3 agood n-acceptor on the IR but a poorone on the NMR scale (20).

Essentially the same ordering as

that for phosphines was establishedfor arsenic and antimony ligands,hence

EPh3~EPh2CH2R < EPhMe2~EEt3 <E(OEt),,

where E = P, As, Sb (20,39). Thisindicates that effects arising fromthe substituents on E usually over-ride those imparted by the donor/ac-ceptor function of E itself. For anidentical set of substituents Z, how-ever (20)

BiEt3 < AsZ3 < PZ3 < SbZ3

Z3 = Et3,Ph3,Me2Ph, (0Et)3

and this is paralleled by an analo-gous ordering of bidentate ligands(38,39):

Ph2As(CH2)2AsPh2 < Ph2As(CH2 )2PPh2<Ph2P(CH2)2PPh2

and (34,39)

,AsPh2 PPh2

SbPh2

This sequence is independent of thetype of complex, which allows an une-quivocal assignment of vanadium NMRsignals in isomeric mixtures.

While the ordering of phosphorusand arsenic (and bismuth) ligandsapparently reflects the order ofdecreasing ligand strength (decreas-ing AE), the position of antimonyligands such as SbPh3 needs specialattention. Here, AE appears to beoverpowered by the large nephelaux-

etic effect (small <r" 3 >3 d) induced3d

by SbPh3, probably connected with anincrease in covalency (small k' ) ingoing from P to Sb. A similar

Vol. 4, No. 1/2 41

conjecture has been made for cobaltcomplexes (54-56).

The close relationship between theshielding and the ligand strength wasalso employed to assign signals ofvanadium complexes obtained withdiphosphanes R2PPR2 (35), which formmononuclear, monosubstituted anddinuclear, monosubstituted complexes.The former should exhibit a greaterir-interaction due to the presence ofan uncoordinated PR2 group directlybonded to the ligated phosphorus andhence a high 6(51V) value. This isshown in Figure 2.

Special interest has been focussedon the question of whether PF or COare the better ir-acceptors(12,39,43). The 6(51V) values forthe complexes CpV(PF3)4 (-1475 ppm)and [V(PF3)6]' (-1946 ppm) aresmaller than for the corresponding

carbon monoxide compounds (CpV(C0)4 :-1534, [V(CO) ]": -1952 ppm) and thisshould suggest a higher ir-acceptorpower of CO (greater AE and smallerparamagnetic deshielding), providedthe factor <r~3>3dk'

2 is constant oroverpowered by influences in AE.Analogous results are obtained fromthe 93Nb NMR spectra of octahedralniobium compounds (43). In complexescontaining both CO and PF3 ligands,6(51V) are, however, clearly largerfor the PF3 complexes than for theparent carbonyl compounds, althoughdiminished symmetry should increasethe number of symmetry-allowed tran-sitions and thus opara • Figure 3gives a graphical presentation of thesituation. Where PF3 and CO competefor the metal iT-electron density, PF3appears to be the better IT-acceptor.This statement is valid only with the

-1000 -TI00 -1200 -1300 -U00 -1500 ppm

Figure 2. 23.66 MHz 51V NMR spectrum of the reaction products betweenT15-C5H5V(CO)4 and PhMeP-PMePh.


assumption of constant <r"3>k'2

contributions and neglect of aniso-tropy effects (29).

A symmetry/shielding correlationwas also proposed for a number ofcomplexes [V(C0)6 n(PZ3)n]~ andCpV(C0)4_n(PZ3)n (n"= 1 to 3) (26)with the phosphines P(OMe)

3 » PMe3 '

PPh2Me, and PhP(CH2CH2PPh2)2 (Figure4), which have substantially lowerIT-acceptor power than CO or PF3 .Shielding decreases, as CO is succes-sively replaced by the phosphine

ligand, which is explained by (i)decreased energy separation betweenHOMO and LUMO as the degeneracy ofthe MO taking part in transitions israised plus increased number of rele-vant transitions in the complexes oflower symmetry, and (ii) destabiliza-tion of o-type and/or Tr-type MO asV-PZ3 interaction becomes increas-ingly important relative to V-COinteractions. For [V(CO) 6_n (PZ3)n ]"complexes these relations are shownin Figure 5.

-20001

-1990

-1980

-1970

-1950-

-1950-

-1940-

-1590-

-1570-

-1550-

-1530-

-1510-

-1490 -

-1470-

6(51V) [ppml

Number of PF3 Ligands (n)

0 1 2' 3 4 5 6

6(51V| Ippm]

mono di

-1900

^700

H300

-1100

-900

V =

1 \ ' / V \ >>

= PhP(CH2CH2PPh2l2

trans-di mono cis-di

Figure 3. Chemical shifts vs. num-ber of PF3 ligands in carbonyltri-fluorophosphinevanadium complexes,showing the dependence of 6(5lV) uponthe local symmetry. Dashed linesconnect the higher symmetry isomers.The graph also illustrates the temp-erature dependence of 6(51V) for[Et4N][V(C0)(PF,)5](small squares).

Figure 4. Correlation between chem-ical shift, local symmetry, andligand strength for various phosphinederivatives of n5-CsH5V(C0)4 (dashedlines) and [V(CO) 1~ (solid lines).

6

Vol. 4 , No. 1/2 43

Generally,order

{V(CO)m} >

S(51V) decreases in the

{V(CO)m_, }EZ3 >) 2 > (V(CO)

m-3}(EZ3)3,

where {V(CO)m} = [V(CO)6]~CpV(CO)4; E = P,As,Sb,

1/2 di- or 1/3Further,

and EZ3 is a mono-,tridentate ligand.cis-[V(C0)m.2(PR3)2]trans-[V(C0)m_2(PR3) ] andfac-[V(C0)3(PR3)3] >mer-[V(C0)3(PR3)3] (R ji F).

An additional decrease in 5 lVshielding is observed with very bulkyphosphines. This steric effect ispronounced in the complexesCpV(N0)2PR3 (47) and CpV(C0)3PR3 (39)and illustrated for the latter inFigure 6, where Tolman's coneangle is employed as a measure forthe ligand size. While variations in6(51V) for phosphines of similar coneangles (PF3, P(0Me)3, PPhH2) againillustrate influences primarily elec-tronic in nature (full circles inFigure 6), the decreasing shieldingfor alkylphosphines of increasing

cone angle (shadowed squares andsolid line in Figure 6) is represen-tative for factors primarily stericin nature. The broken line in Figure6 connects methylphenylphosphines,where either effect has to be takeninto account. For bulky phosphines,overlap conditions with suitablevanadium orbitals are less favorabledue to steric crowding and increasedbond lengths. The validity of thelatter argument has been demonstratedby correlating shielding parametersand structural parameters obtainedfrom X-ray analyses of several phos-phine-platinum complexes (57-63).The net effect of hindered vanadium-phosphorus interaction is a decreasein AE, increase in k'2 and possiblyalso an increase in <r"3> (diminishedextention of the metal-3d system),all of which result in increased par-amagnetic deshielding and decreasedoverall shielding 0. A decrease ofshielding due to hindered V-P overlapwas also observed in strained che-late-ring structures (39,40) and willbe discussed in the context of6(31P)/6(51V) correlations in SectionII. D.

IV(CO)5L]" lV(CO)Al2r tV(CO)3L3r ID

(2)

(3)

W

yz

xy

b, [a*]

3d(V)Oi_

e

b2

\jrlA

\ a, ffd(PZ3)

Figure 5. The vanadium-3d system for [V(CO)6_nLn]" complexes (left), andthe vanadium-phosphorus interaction in [V(C0)5PZ3]~ (right). (1)Complex, (2) local symmetry, (3) transformation properties of theangular momentum operator, (4) relevant transitions. From refer-ence (26), slightly revised.


V

A

V

\

1

ift [

ppm

]C

hem

ical

Sh

V-5

1

--1550

-1500

--K50

- -U00

PH2

--1350

-1300

--1250

90

• P F

PM«

110

3

OMeK

^ - C 5 H

))

PMePh™^ O

PPh^

130 150

5VICO) 3PZ 3

u')3

o LTcy3

P(Bul)3Q

170

Tolman's Cone Angle t°]

2. Thioformiatovanadium(+III)Complexes

51V NMR spectra were obtained forseveral dicvclopentadienyldithioform-iatovanadiuni- ( + 111) complexes (Table2) which, in solution exhibit a tem-perature dependent equilibriumbetween a diamagnetic chelate struc-ture with S-S-coordination and anopen, paramagnetic form (52):

C-YCp Cp S (8)

Since the presence of paramagneticspecies usually prevents the observa-tion of vanadium NMR signals, theequilibrium for the compounds listedin Table 2 may be considered to lie,at the temperature indicated, almostexclusively on the side of the diam-agnetic forms.

The very similar 6(51V) values forY = OEt, NEt2, PCy2, PPh2, and AsPh2

support the assumption that this

Figure.5

6. 6(51V) orTl'-C5H5V(C0)3PZ3 complexes vs. Tol-man's cone angles of the phosphinesPZ3. The graph illustrates thedependence of 6(51V) on stericeffects (squares) and electroniceffects (full symbols).

group does not take part in coordina-tion to vanadium. The same appliesto the oxo-phosphorus and thio-phos-phorus groups: coordination via thephosphorus-bonded oxygen and sulfurwould result in a chelate five-ringstructure, for which shielding shouldbe higher than in the four-ring sys-tems (see discussion in Section II.D.). The lower shift values forthese complexes are in good agreementwith the known weak ligand strengthof the ligands [S2CP(S)R2] as com-pared to [S2CPR2] (64)

3. Oxovanadium(+V) ComplexesContaining the V0 3 + or V02*Group

Investigations in this fieldinclude compounds of the types V0X3

(X = F, Cl, Br, NR2) and V0Cl3.n(0R)n

(n = 1-3) (19,65), [V0F4r (66,67)and [V0Cl4_nFn«CH3CN]~ (n = 0-4)(68), and [V02]

+ derivatives (69).NMR data are summarized in Table 3.

No 6(51V) value was reported for

Vol. 4, No. 1/2 45

Table 2. s 1V NMR Data of Dithioformiatovanadium(+III) Complexes

Cp2V(S2CY)

Solvent Temperature 6( 5 1V)

[ppm]V2

[Hz]

OEtNEt 2

PCy2

PPh2AsPh2

P(O)Bz2

P(S)Ph2

-HaIf-width

THFTHF

CH2CI2

CH 2C1 2

CH 2C1 2

CH2 Cl2

CH 2C1 2

280300250300280280300250230

-1100

-1030

-1080

-960-1000

-1040

-1090

-895-780

>3000

>3000

640

ca. 1000

>3000

120

700

Table 3. s 2V NMR Data of V 0 3 + and V 0 + Complexes

1 . — — _ _

Compound

•

VOBr3

[VOCL, CH3CN]"b

VOC1 3

VOCl3(CH3CN)2

VOCl 3(PhCN) 2

VOC13/THFC

V0(SC10 3) 3d

! VO(NEt 2) 3

VOCl2OMe11

V0Cl20Et

V0Cl2(0-i-Pr)1

V0Cl2(0-n-Bu)

' V0Cl2(0-i-Bu)

V0Cl2(0-t-Bu)1

1 V0Cl(0Et)21 VOCl(O-i-Pr),

Phase

Br2neat

CCUCH3CNneat

THFCH3CN b

THFTHF

HSCIO3

CH3 CN

neat

THFneat

THFneat

THFneat

neat

neat

THFneat

neat

6( s lV) a

[ppm]

+440+432+417+23

0-32-114-35-32,-242-316-4-365

-290

-304

-300-316

-309-342

-290

-288-320

-404-414

-506

Av[Ez]

20

68

38

126

16

6328191977056

References

1919196819,219

19

19691919,e

1919,e

1919, e

19ee

19, e

e


Table 3. Continued

Compound

VQCl(O-n-Bu)2

V0Cl (0 ' - i -Bu) 2

VOCl(O- t -Bu) 2

V 0 C l ( a c a c ) 2f

V(OMe)3

V(OEt ) 3

VO(O-n-Pr ) 3

VO(Oi-Pr) 3

VO(O-n-Bu)3

VO(O-i-Bu) 3 .VOF 3

tVOCl3F(CH3CN)rt r a n s - [VOC12F2CH 3CNTfcis-[VOCl2F2(CH3CN)r[VOCIF 3(CH3CN)f

[VOF4(CHaCN)r[V0(02)]+

fvo2]+

[V02(H2O)4]+

[V02(C204)2]3"[VO2(EDTA)]3"[VO2Cl2r[V02F2]-

9

Phase

neatneatneatTHFTHFCH3CNneatneatEt2OheptaneTHFneatEt20THFC 6H i 1 C H 3neatTHFneatTHFCH 3CNCH3CNCH CN'CH CNCH 3CNCH3CNH20/H2O2/H+

H 2 O/H+H 2 O/pH 0H 2 O/pH 1H 2 O/pH 2H C 1 0 4 , HNO3

H C l , H 2 S 0 4 ,H 3 P 0 4

H20H20CH3CNHSFO3DMSO

6 ( 5 1 V ) a

[ppm]

-470-478-540-344-458-459-443-555-584-577-585-628-624-619-623-548-580-538-757-786-205-380-456-608-778-543-548-545-545-541-545

-520 to-587-533-513-301-66-64

A v i / 2

[Hz]

36021058

15182114

23

3401390-1730

850582400

600-1500250800606060

References

eee1919191919eee19, e656565e19191919686868686870716670,727369

696969696969

a) At room temperature if not indicated otherwise. Some of the shift valuesreported in reference 19 have been slightly revised. b) At 230°K. c) Thethree signals probably correspond to [VOC13](THF)n , [VOC12(THF)n]C1, and[VOCl(THF)n]Cl2. d) Tentative assignment for the signal observed when NH4V03

is dissolved in HSCIO3.tative assignment.

e) Unpublished. f) acach2 = Acetylacetone. g) Ten-

Vol. 4, No. 1/2 47

the [VOF4]" anion, the composition ofwhich was based on the observation ofa quintet in the vanadium NMR spec-trum. Since a quintet is in agree-ment with the magnetic equivalence ofthe fluorine ligands established forlocal C 4v symmetry, this geometry wassuggested by Hatton et al. (66).According to Moss and Howell (67),however, it is doubtful that thefield gradient at the vanadiumnucleus is sufficiently small toallow observation of a resolved vana-dium signal, and they suggest a rapidexchange amongst the F ligands eitherin a C4v or in a C2» symmetry or,alternatively, interconversionbetween these two geometries.

Buslaev and co-workers studied thesystems VOC13/CH3CN, VOC13/CH3CN/KC1,and VOC13/CH3CN/KF,HF (68). ForVOC13 dissolved in CH3CN, theyobserved a chemical shift of -114 ppmwhich they attribute to the knowncomplex VOC13»2CH3CN. The 6(51V)values reported by us for VOC13

#2CH3CN (-32) and, VOC13»2 PhCN (-35ppm)(19) dissolved in THF may hencecorrespond to VOC13/THF adducts (videinfra) formed by displacement of thenitrile for THF. On addition of KC1to a solution of VOC13 in CH3CN, asingle signal at -38 ppm is obtainedwhich is thought to indicate theposition of an exchange equilibriumof the kind

VOC13«2 CH3CN [VOC14'CH3CN]"

+ CH3CN (9)

On lowering the temperature to ca.230°K, two signals (-114 and +23 ppm)are observed (68).

Rapid exchange among various spec-ies also occurs in the VOC13/CH3CN/HFsystem, for which 6(51V) values coverthe range of -112 to -535 ppm (atroom temperature), depending on theVOCI3/HF ratio. At ca. 240°K, fivefluorine-containing complexes can becharacterized by distinct 5XV NMRsignals., where (i) .' shieldingincreases with increasing number ofF-ligands and (ii) shielding fortrans-[VOC12F2]~ is smaller than for

the cis-complex. These results arein accord with (i) the strongerligand field exhibited by the fluor-ide ligands and (ii) our findings onthe shielding in cis/trans isomers ofcyclopentadienylcarbonylphosphine-vanadium complexes (see Section B.1.). The relative intensities for aVOC13/HF ratio of 1:5 are[VOC14»CH3CN]" (0.26) /[VOC13F«CH3CN]- (0.27) /cis-[V0Cl2F2»CH3CN]~(0.11)/ trans-[VOC12F2 •CH3CNf (0.06)/[VOC1F3«CH3CN]" (0.16) / [V0F4»CH3CN]~(0.14). The rapid intermolecular

exchange in oxovandium fluorides atroom temperature is in contrast tothe behaviour of oxofluorochloridesof niobium, for which a 93Nb NMRstudy has been carried out (74).

The signal at -778 ppm dissolvesinto a quintet; the corresponding ion[V0F4»CH3CN] apparently is equiva-lent to that found by Hatton (66) andHowell (67). The shift valuereported for V0F3 dissolved in CH3CNis -786 ppm (19), which, within thelimits of temperature effects, coin-cides with the 6(51V) value of

[V0FJ- •Figure 7 gives a graphical repre-

sentation of the correlations betweenthe chemical shift values of oxovana-dium complexes V0Z3 and the elec-tronegativity x of the groups Z, onthe one hand, and the. bulkiness of Rif Z = OR, on the other hand (comparereference 19). Buslaev's data on[V0Cl4.nFn*CH3CN] have been included(68). There is a consistent upfieldtrend for the shift values as theelectronegativity of Z and the spa-tial crowding at the V0 3 + centreincrease. The high field shift withincreasing X is consistent withincreasing AE and decreasing cova-lency of the V-Z bond (equation 7).The steric influence is opposite tothat observed for CpV(C0)3PZ3 com-plexes (Figure 6). Lachowicz andThiele (75,76) have shown that thereis a definite tendency for vanadylesters V0(0R)3 to associate, and thistendency decreases with increasingbulkiness of the alkyl substituent R.


D-—"Hfr-—D-—-Q-voci^R

0- IVOCI3

!'= Me Et

VOZ3

O V0(0R'|3/ \ V0CI(0R')2D V0CI2(0R')

Increasing Ligand Bulkiness -—»

Figure 7. 6(51V) of V0Z3 compoundsvs. the Allred-Rochow electronegativ-ities X of the substituents Z (upperabscissa, solid line and shadowedcircles; R =n-Bu) and the bulkinessof Z if Z = OR' (lower abscissa, bro-ken lines, and open symbols). Datataken from reference (19) and

Bun Bu1 PH Bu*

We have confirmed this phenomenon byinvestigating the temperature depen-dence of the 5 XV shift values andhalf widths of the 5 *V signals.Accordingly, the shifts for the vana-dyl esters (and esterchlorides) indi-cate the equilibrium position ofrapid exchange equilibria of the kind

n VO(OR), {V0(0R)3}r (10)

where n very likely attains the value2. At temperatures below ca. 240°K,the two species exhibit two distinctsignals (65). If the signal at lowfield is assigned the oligomeric(dimeric) form, the high field shiftsfor vanadyl esters containing spa-cious groups R indicate that there islittle association for these com-pounds (see also the discussion inSection II. E.). An additional con-tribution to the high field shift mayarise from an expansion of the coor-dination sphere (small <r~3> value inequation 7).

Another interesting feature is thesubstantial upfield shift for V0Br3

and VOCI3 when dissolved in THF (19).The signal for V0Br3 is shifted from

n variesppm for a

+432 to -242 ppm, suggesting a com-plete rearrangement in the coordina-tion sphere with partial removal ofbromine ligands and formation of aspecies [V0Brn(THF)m]Br3_n, where nprobably = 1. With VOC13 two signalsare obtained. While the low-fieldsignal at -32 ppm may be interpretedin terms of a simple solvent shift(THF coordinated in the second coor-dination sphere), the high-field sig-nal is probably due to a complex[V0Cln(THF)]Cl3-n , wherefrom 2 (5(51V) = -242VOCI3/THF ratio of 1:1) to. a limitingvalue of 1 (6(S1V) = -316 ppm;VOCI3/THF = 1:10).

Howarth (70) has reported theshift values for the oxovanadium andoxoperoxovanadium cations ([V02] :-545; [V0(02)]

+: -543 ppm). Theshifts are identical within the lim-its set by the inaccuracy in measur-ing shift values, but [V0(02)]

+ isdistinct from [V02]

+ on the groundsof the different line widths and thebehaviour towards EDTA: the peroxocompound is not readily chelated byEDTA. The dependence of shift andline width for [V02]

+ upon concentra-

Vol. 4, No. 1/2 49

tion and nature of the acid employedfor the acidification of an originalmetavanadate solution has been dis-cussed by O'Donnell and Pope (69)(cf. Table 3).

4. Vanadates and Peroxovanadates(+V),Iso- and Heteropolyvanadates(+V).

Howarth (70) has also investigatedthe system metavanadate (i.e.[H2V04]~)/H202/0H~ and identified upto eleven peroxovanadates from theirresonances in the vanadium NMR spec-trum. The data are contained inTable 4; four of the species havebeen tentatively assigned. The com-position, i.e. the number of peroxoligands and the degree of protona-tion, depends on the number of equiv-alents of H 20 2 and base added to[H2V04]~ at a starting pH of 8. Thecomplexes formed with one equivalentof H 2 0 2 are [H2V03(02)]~ (tentativeassignment), [HV03(0 2)] 2~ and, at pH> 10, [HV0 2(0 2) 2]

2-. This species isthe only one present if two equiva-lents of H 20 2 plus one equivalent of

base are added to metavanadate, whilethe diprotonated complex[H 2V0 2(0 2) 2 ]" is formed without theaddition of OH". Addition of morethan 2 equivalents of H 2 0 2 yields[HV0(02)3]

2~ . All protonated formsare deprotonated to [V02(02)2]

3~ and[V0(02)3]

3" if the pH exceeds 13.Under extreme conditions (substantialexcess of H 20 2 and OH"), [V(02)4]

3"is obtained.

Shift values increase by ca. 50ppm or less, if protonation occurs([HVOJ 2- (-534) •> [H2V04]" (-574)),and increases by ca. 100 ppm for each0 2

2~ ligand replacing the 02~ ligand.These trends are explained by anincrease of the energy of theligand-to-metal charge transfer(increased AE in equation 7), in con-nection with an increase of covalencyof the V-0 bond (decrease of k 1 ) .

Four groups (66, 70/72/79, 73,69)have investigated the systems metava-nadate/OH" and metavanadate/H30*.The results can be summarized as fol-lows (see also Figure 8 and comparethe data contained in Table 4):

-550-

-500-

-450-

10

Figure 8. Composition and chemical shifts of vanadates(+V) between pH 0 and14. Vertical lines connect 5 H7 signals belonging to the samespecies. : .


Table 4. 51V NMR Data of Vanadates(+V), Peroxovanadates(+V), andIsopolyvanadates(+V)

Compound PH[ppm]

Av[Hz]

References

VO,

vo 31 7 o 3 -

14

V O ^

2 7

("VO - " )

V20H3"HV03(02)2"

V0 2 (0 2 ) 23 " d

HV02(02)22"

H2V0z(02)2*H[V0(02)2]2OV0(02)3

3

HV0(02)3

3-3 -

2 -

V0(02)2(NH3)v n 5"V 3 U 1 0

HV 0 5 '10 28

H V 0 '2 10 28

913-10

13913-913-9.5910-8<8-7

13-95.57.5

13-5.5

11-76.9-2.7

8.5-7.56.9-2.77

5.5

5-3.5

5.8-4

-536

-537-554 to -559

-534-574-553 to-527 to-562-623-621-760-766-697-757-845-733-734-746-493 to-510 to-572 to-407 to-510-492-418-511-495-426-510 to-500 to-420 to-520-513-427-512 to-495 to-421 to-519-502-422

•568•534

•504-523-574-419

<10-6027ca 100b

50c

508080-100

85

250240260650380160480165

150-180

-519-504-425

-517-500-422

150150350

69-737869667870,7219,7069,70667070707070707070707070

666966

72,79

73

69

73

79

79

a). In aqueous solution at room temperature; resonances cited in reference 70at 273°K. b) The slightly broader signals and higher shift values at lowerpH indicate protonation and exchange equilibria between protonated and unpro-tonated forms. c) Line width for the V-170 satellites in the 51V NMR spec-trum, d) Tentative assignment.

Vol. 4, No. 1/2 51

At pH 14, the only species presentis the orthovanadate ion [V04]

3"(-536 ppm) which, due to its tetrahe-dral geometry (field gradient at thenucleus ~ 0) , is characterized by anarrow line. As the pH is lowered,the signal broadens due to the forma-tion of dimeric [V207]''' (-553 ppm atpH 13) and protonated species[HV04]

2" (-534 ppm at pH 13) and[HV207]

3* (-562 ppm at pH 9).Exchange between these speciesaccounts for further line broadening.In the pH region of 8.5 to 7.5, oli-gomeric metavanadates [(V03)x]

x" with•x probably 3 and 4 are present (-5 72to -576 ppm), but the signal appear-ing at -574 ppm (pH = 9) is alsoassigned the diprotonated [H2V04]"("metavanadate [V03]""). Furtherlowering of the pH produces signalswhich have been associated with theoligomeric anions [V 30 1 0]

5" (two sig-nals at -493 and -510 ppm (pH =6.9)for the two different vanadium sitesin this anion) and tetra- or pentam-eric forms (ca. -410 ppm) (66).While these are tentative assign-ments, the existence of decavanadate[̂ 'iot-l28]6" below pH 7 is clearlyindicated by a set of three signals

[-418 (a), -492 (b), -510 (c) ppm atpH 7 (79)j in the intensity ratio1:2:2, corresponding to the threedifferent vanadium sites in thisanion (Figure 9). Finally, below pH5.8, the decavanadate is protonatedand the species[HVI0028]

5™ (-421, -496, -512 at pH5.8) and [H2V10 0 2 8 ] " " (-422, -501,-519 at pH 3.0) are formed before, instrongly acidic solution (pH < 2),they break down to [V02]

+ (-545 at pH0).

Special attention has been attrib-uted to the protonation of [V10028]

6"over the pH range of 9 to 2 (79).The first protonation occurs with anoverall pK of ca 5.5, the second pro-tonation at a pK of ca. 3.6. Sincethe signal (a) is only very slightlyaffected by the protonation, andsince the first protonation affects(c) to a lesser extent than (b) whilethe influence on (c) and (b) is oppo-site with the second protonation, theprotonation sites are considered tobe (b) for the first and (c) for thesecond protonation (cf. Figure 9).The authors also give strong evidencethat the apical out-of-plane, and notthe bridging oxygens are protonated,

© =V(Q)

Figure 9. Structure of the decavanadate ion [V 1 OO 2 8]6" [adopted from refer-

ence (77)], showing the three different vanadium and the two pro-tonation sites.

Bulletin of Magnetic Resonance

which is contrary to the conclusionmade on the basis of 170 NMR studieson [V 1 00 2 8]

6" (80).The 51V NMR spectra of heteropoly-

vanadates of general formulas[VnMe.n019 ] (

n*2)" (n = 1,2),[VnM13.n 040] (

n+2)" (n = 3,4),[VnM 2_n P040] (

n*3 Xn = 1,2) (includ-ing some short-living protonatedforms), where M = molybdenum andtungsten, have been investigated byfour groups (69,71, 73/81, 82), andrecently, these studies were extendedto [VnW18.n P2062] (

n + s )" (n = 1,3) andseveral heteropolyvanadates contain-ing silicon instead of phosphorus(83). 31P (69,81,82) and 1 70NMRstudies (82) have also been carriedout on selected compounds. 5 *V dataare compiled in Table 5.

Species containing two and morevanadium atoms usually show more thanone 5 *V NMR signal (an exception is[V 2W 40 1 9]"' (69), which is attributedto isomeric mixtures in solutionand/or different vanadium sites inthe complex structure. Thus, thelines at -553 ppm for [VifWgOito ] s " and-558 ppm for [V3W10Olt0 ]

5 ' are allo-cated to the central (tetrahedral)vanadium in the Keggin structure ofthese anions, while the lines at ca.-506 and ca. -530 ppm, respectively,are due to external (octahedral)vanadium atoms (69). Heteropolyvana-dates - like isopolyvanadates - showa pH-dependent shielding of the 5 *Vnucleus. This is again attributed toprotonation, and bridging oxygens areconsidered to be the protonationsites (69,73,81) which, in the lightof the investigations carried out byHowarth and Jarroli ontvio°28]e~ (79), seems doubtful.

Chemical shift values are higherfor tungsten than for molybdenumanions (compare [V2W^019]*": -513 and[V2Mo4019]*-: -502; [ VW, ̂ 0 , „ ] *" :-551 and [VMOj jPO^ ]"" : -532 ppm(73), and substantially higher forphosphorus-containing heteropolyvana-dates. This is rationalized byKasanskii (73) by considering thefirst CT transition; i.e. the energyseparation between non-bonding orbi-

tals, consisting chiefly of purep-orbitals localized on the bridgingoxygen atoms, and the vacant, basi-cally vanadium-3dx orbital. Theenergy of the latter depends on thedegree of its participation in thebond between vanadium and suitablebridging oxygens. The Mo-0 bond inmolybdenum vanadates is weaker thanthe W-0 bond in the tungsten vana-dates. Hence, in the former, thereis a more substantial participationof oxygen orbitals in the V-0 bondand, in turn, a lesser V-dxy contri-bution. The result is a destabiliza-tion of the dxy orbital which leadsto higher AE and smaller opara valuesand thus produces an increase of theoverall shielding (equations 7 and1).

C. Chemical Shifts Relatedto ESR g-Factors

The ESR-spectroscopic g-factor canbe expressed by

= -2X£(En-E0)<O|L2|n> (11)

where in the notation of equation 5,X is the spin-orbit coupling and L isthe angular momentum operator. Sincethe vanadium shielding constantdepends on the orbital angular momen-tum and the energy separation betweenground state, E Q, and the excitedstates, E n, (equation 5), a correla-tion between 6(51V) and g values isnot unexpected.

An approximately linear correla-tion has been shown to exist betweenthe 6(51V) values of the vana-dates (+V) [ W 5 0 1 9 ]

3 ~ , [VPMo uO^p]"" ,[H2VWX1 0 k0]

7" and [VPWX1 0kl} ] "' , andthe g-factors of the reduced formswith vanadium in the formal oxidationstate +IV (73): |5(51V)j increaseswith decreasing g. This correlationwas disputed by O'Donnell and Pope(69); it seems to be approximatelyvalid for compounds containing anisolated V atom. Thus, for

[aVSiWM0V P W

i6-

the respective g-values are 1.962,

Vol. 4, No. 1/2 53

reported for phosphorus-containingheteropolyvanadates(+V) (69,81,82)and for carbonylvanadium(-I and +1)complexes (39). In nearly all of thecompounds studied, trends in 5(31P)are opposite to those in 6( S 1V), i.e.where shielding of the 31P nucleusdecreases throughout a series of com-parable complexes, shielding of the5 *V nucleus increases. This is dem-onstrated for selected heteropolyva-nadates in Table 6 and for carbonyl-vanadium complexes in Figure 10.

It is a well established fact thatin chelate complexes of the diterti-ary phosphines Ph2P(CH2)nPPh2 (dip-hos; n = 1: dppm, n = 2: dppe, n = 3:dppp, n = 4: dppb), the 31? nucleusis shielded less in the unstrainedfive-ring than in the strained four-ring structures (84). These rela-

tions are paralleled by a contrarytrend in the shielding of the metalnucleus (85-87) as exemplified forthe sXV nucleus of the complexesCpV(CO)2 diphos and HV(C0)4 diphos inFigure 10. Observations of this kindhave been discussed in the light ofphosphorus-metal-phosphorus bondangles and metal-phosphorus bondlengths obtained from X-ray data [see59-63) and literature cited in refer-ence (39)], suggesting that angledistortions (deviations from idealbond angles) and extended bondlengths in chelate systems subjectedto enhanced ring strains cause hind-ered metal-phosphorus overlap (hencea decrease in AE and an increase inmetal-d character of the relevantmolecular orbitals), and thusdecreased overall shielding of the

-1700-

-1600-

-1500-

-1400-

-1300-

1200-

-1100-

-1000-

-900-

-800-

dppm4

fi(51V]

\\

Q

dppm

dppe51

dppp6

O = HV(CO)4Ph2P(CH2)nPPh2

yb

dppe5

/

i

i

dppb7

10-

30-

40-

50-

60-

80-

90-

OcpVICOI2Ph2P(CH2)nPPh2 -

\ 1 1 0 -

dppp* dppb

6 7Chetate-ring Size n

Figure 10. Chelate-ring size vs. 6(51V)(left ordinate, solid lines andshaded symbols) and'6(31P)(right ordinate, broken lines and opensymbols). Shielding of the 5XV and 31P nuclei increases frombottom to top. The data taken for the dicarbonylcyclopenta-dienylvanadium 6-ring complex ('") are those ofTi5-CsH5V(C0)2Ph2(CH2)3PPh(CH2)3PPh2.


metal nucleus in dppm complexes.Simultaneously, bond angles at thephosphorus atom decrease (dppm) orincrease (dppp) with respect to the109° angle for sp3 hybridization,which is accompanied by an increaseof the o(s) electron density in thedonor function and hence an increasein 31P shielding. Hindered P^MTT-overlap and a resulting destabili-zation of the vacant phosphorus typeMO taking part in electronic tran-sitions relevant for the paramagneticshielding of 31P should contribute tothe high 3XP shielding in the dppmcomplexes (34,39).

Large 5 *V upfield shifts for dppe-and large 51V downfield shifts fordppm complexes have also beenobserved for the complexescis-[Et4N] [V(C0)4diphos] andTl3-allyl V(C0) diphos (39).

E. Temperature Dependenceof 51V Shielding

With increasing temperature,vibronic levels are increasingly pop-ulated and thus AE in equation 7 willdecrease, leading to an increase ofthe paramagnetic term and a decreaseof overall shielding (and |6(S1V);cf. also references (27) and (88)).This is verified for the phosphomo-lybdato- and silicotungstatovana-dates-(V), where 6(51V) decreases by3 to 7 ppm as the temperature israised from ca. 300° to ca. 350°K(82,83) (cf. Table 5). In carbonyl-vanadium(+I and -I) complexes, thereis usually a linear relationshipbetween 6(51V) and the temperature inthe temperature range 200-340°K(Table 7 and Figure 11; see also Fig-ure 3). Temperature gradients are

[Et4N][V(CO)5P(OF)3]

0,18 M in THF

Figure 11. 3D-Plot of 6(S1V) of [V(C0)5P(p-C6H4F)3 ]" vs. temperature andrelative amplitude. The doublet (51V'-31P coupling) collapses to a singleline below ca. 240°K due to increased line broadening.

Vol. 4, No. 1/2 57

around 0.3 ppm/degree for carbonylva-nadates(-I) and ca. 0.7 ppm/degree for carbonylvanadium(-t-I).

In contrast, vanadyltriisopropy-late V0(0iPr)3 exhibits a negativetemperature gradient; i.e. shieldingdecreases with decreasing tempera-ture. The temperature dependence of6(51V) (and the line-width) of thiscompound has been subject to anextensive study (65), according towhich the decrease of 6(51V) is theresult of increased formation of(probably) dimeric species of lowershielding relative to that of themonomer (equation 10 in Section B.3.). Below ca. 240°K, the two spec-ies emerge as two separate signals(Figure 12 a and b). Over a range ofca. 25 degrees,the signals shiftslightly to stronger field, and,starting at ca. 220°K, this trend isagain reversed by a second downfieldshift which may reflect furtheraggregation to {V0(0iPr)3 }n (n > 2)units.

HI. NUCLEAR SPIN-SPINCOUPLING CONSTANTS

A. introduction and Theory

The main contribution to nuclearspin-spin coupling is usually consid-ered the Fermi contact term (89,90)which, in the average energy approxi-mation, can be written in the form

| S ( 0 ) w | 2 | S ( 0 ) x | 2 o ( s ) 2 (13)

where o ( s ) = 2 £ c ( s ) M c ( s ) i X . I n s t e a d of

quoting J, it is sometimes convenientto compare the reduced coupling con-stant^ K = 2nJ/tiyM2rx.

3AE is the mean triplet excitation

energy, S(0)M and S(0)x are therespective valence-s wave functionsat the nuclei M and X. 0(s)2 repre-sents the o(s) contribution to theM-X bond. equation 13 does not con-tain parameters correlated withdirect ir-inf luences . Nonetheless, aswe will see, indirect n-contributionsare predominant at least in vanadium-phosphorus couplings of carbonylphos-phinevanadium complexes. The influ-ence of the |S(0)M|

2 term is demon-strated below for the M-P and M-Fcouplings of the octahedral complexes[M(PF3)6]" (M = V, Nb). For thesecomplexes, |S(0)x |

2 and - in a firstapproximation-AE "x and o(s)2 may betaken as constant. However, an addi-tional contribution to K(NbP) mayarise from the greater o-bond orderin the Nb-P bond.

One of the main problems encoun-tered with couplings in vanadium com-plexes - as in complexes of othernuclei with nuclear spin > 1/2 - isthe quadrupolar line broadening inall complexes but those of cubic andC3y symmetry, which quite oftencauses unresolved NMR signals. Thus,no coupling constants are reportedfor phosphorus-containing heteropoly-vanadates(+V) (an exception is[Bu4N]3[VW5019] (83), Table 8), whichmay, however, also be due to rapidexchange between different isomers ordifferent sites in the same molecule.For the same reason, vanadium-fluo-rine coupling in complexes such as[V0F4]" is only observed at low temp-eratures (66,67,74).

M S ( O )M(-I) 2K [Hz] (Ref.43)MF

V 0.869

Nb 1.788

154

340

1.3

7.7


?••*•-•

Table 7. Temperature Dependence of 5 1V Shift Values

Complex Phase Temperature

[°K]

3303203103002902803l+0 - 280

313303.2932832732632532U3233223203313 - 283283 - 203

323 - 193313 - 213313 - 213320 - 200330 - 193330 - 193

6(51V)

[ppm]

-19^.5-19^7.7-1950.9-195U.1-1957.0-1960.2

-1951.1-195^-2-1957.3-1960.3-1962.9-1965.2-1967.6-1970.2

-1972.5-197^.8

-1979.1

-1805/-1838-181+9/-1881-1702/-17U1-1957/-1995-1514/-1598-1300/-1396

Temp.-Gradient

[ppm/°]

0.313

0.320.2U

0.360.32

0.390.320.6l0.70

[Na(diglyme)2][V(C0)6] . glyme 0.2 M

[EtnN][V(CO)6] THF/CH3CN 1 : 1 , 0 .1 M

[Eti tN][V(CO)5P(p-C6HuF)3][EUN][V(C0)5PPhMe2]

THF 0.2 MTHF/CH3CN 1 : 1 , 0 .1 M

cis-[Et l tN][V(C0)4(PPhMe2)2] THF/CH3CN 1 : 1 , 0 . 1 MTHF 0 . 1 MTHFTHF 0.2 M

H2O/H3O+

CpV(C0KCpV(CO)3P(p-C6H1|F)3

3*+5 - 300 0.07 - 0.16

Vol. 4 , No. 1/2 59

Table 7- Continued

Complex Phase Temperature[°K]

353 and 303353 and 303

320 - 2^0

29827^21+9239229219209201+19919518918U180

298 - 180

6( 5 1V)[ppm]

-566/-571-569/-575

-627/-623

-619.7-618.1-612.0-607-2-601.6-590.9, -601.1-581.O, -596.7-577.1, -59^.8-567.8, -590.8

-563.7, -589.3-558.5, -587-552.6-5^7.6

Temp.-Gradient[ppm/°]

0.10.1

-0.05

-o.6id

V0(0i-Pr)3

V0(0i-Pr)3

D2O/D3O*D2O/D3O

neat

THF 0.86 M

a) Ref. (82); see also Table 5. b) Ref. (65). c) Ref. (83). d) Averaged for the low field signal. No linearrelation.


Sly chemical shift values of VQtOPr'b in EtaO(1.72 M)7 7 7 7 7 / / 7 7 7 7 / 7 7

n] -580 -590 -600 -610 -620 -630

•570 -580 -590 -600 -610 -620 -630 Ippm)

-630

•620

•610

-600

I 5 9 0

£-580~BM

•fi-570

-560

*0.39Ma 1.72M

180 200 220 240 260 280 300TEMPERATURE |K]

Figure 12. Temperature dependence of 51V shielding in V0(0i-Pr ) 3 . (a) Thesignal to low field ( T < 240°K) probably corresponds to adimeric species. Above 240°K, there is rapid exchange betweenmonomeric and dimeric forms. The Figure also shows the linebroadening with decreasing temperature. (b) At low concentra-tions (0.39 M), the inconsistency at ca. 240°K disappears.

Vol. 4, No. 1/2 61

On the other hand, vanadium NMRsignals in most [V(C0)6_n(PZ3) ]" andCpV(CO)4.n (PZ3)n (n = 1,2) complexesare sufficiently narrow so as toallow direct observation of vanadium-phosphorus coupling constants J(VP).This is somewhat surprising, sincethe local symmetry for these com-plexes is C,̂ at the best, anddecreases down to Cs in CpV(CO)3PZ .In complexes of low point symmetry,the electric nuclear field gradienttensor is fully effective and relaxa-tion times should be rather short.The relatively narrow 5XV signals maytherefore account for a small elec-tric quadrupole moment eQ of the S1Vnucleus as discussed in the introduc-tory section.

The feature of the spectra usuallydepends on the strength and the bulk-iness of the phosphine ligand, thelocal symmetry of the molecule andthe viscosity of the solution. Theeffect of viscosity is shown in Fig-ure 11: with decreasing temperature,the ti?o lines of the doublet broadenand finally melt to give one broadline. Strong iT-accepting ligandsgive rise to comparatively narrowsignals and large coupling constantsand thus favor well resolved spectra.In a series of complexes such as

[V(CO) L ]", line widths increasewith decreasing point symmetry, andthis dependence sometimes allows forthe distinction between geometricalisomers.

A three-fold axis present in themolecule usually decreases the relax-ation times to an extent, wherecoupling is not averaged to zero byrapid quadrupolar relaxation. Thisis shown in the s XV and 1H NMR spec-tra of {CpV(H)(CO) ]" (46) (Figure13). While 3

the vanadium spectrum exhibits adoublet, the eight equidistant sig-nals of equal intensity, which oughtto show for the hydride ligand in theXH NMR spectrum, overlap to give aplateau-like signal very typical of aspin 1/2 nucleus coupling to a quad-rupolar nucleus. An analogous XHsignal was observed for n7-C7H7V(CO)3

at room temperature (91). The signalsuccessively collapses into a broadLorentzian line as the temperature isdropped. Finally, in the hydridocomplexes HV(CO)npm (n = 2-4; p m =oligotertiary phosphine), the XH and5 XV nuclei are decoupled (an excep-tion is HV(C0)3MeC(CH2PPh2)3 whichagain contains a C3 axis), and theline pattern observed in the 1H - NMRspectrum (hydride region) solely

70 Hz

. 160 Hz

Figure 13. The 23.66 MHz S1V (A) and 270 MHz *H (B) NMR spectra of[Et4N][n

5-C5H5V(H)(C0)3](C3v point symmetry for the [V(H)(C0)3]moiety).


arises from 1E - 3XP coupling (46).31P NMR spectra of phosphinevana-

dium complexes generally exhibit sig-nals indicating no or only partialdecoupling, which are therefore hardto observe. Exceptions are cyclopen-tadienyl, allyl, and (to a certainextent) hydrido complexes, where suf-ficiently distinct 3*P signals can beobserved at 220°K (39-42, 46,92,93).In a few special cases[CpV(C0)PhP(CH2CH2PPh2)2](51) andallyl complexes (92,93) the 31P and5XV nuclei decouple at low tempera-ture, and a single, rather narrowline is seen in the 31P NMR spectrum.

If the symmetry is 0h or Trf, thefield gradient at the nucleus iszero, and quadrupolar relaxationbecomes ineffective. Thus, for[V(PF3)6]" , a well resolved

3 *P spec-trum is observed, showing eight(3ip-5iy coupling) superimposed1:3:3:1 quartets (31P-19F coupling)(43). The corresponding 5XV spectrumis illustrated in Figure 14. Aresolved eight-line pattern is alsoobserved in the 170 NMR spectrum of[V0.J3" (78) and in the 13C NMR spec-trum of [V(C0)6]-~ (94). Aqueous[NH^][VF6], however, exhibits broadsignals in both the 19F (width ca.200 Hz) and the 5 *V (ca. 170 Hz)spectra (95), indicating either rapid

exchange of the fluorine sites ornon-ideal octahedral symmetry. Inacetonitrile solution, and at 250°K,an eight-line pattern is observed forthis ion (14).

B. Discussion

Except for vanadium-phosphoruscouplings, information on couplingconstants involving vanadium isscarce. The known data are compiledin Table 8 (for J(VP) see also Table1). Two-bond couplings have beenreported in three cases only (2J(VF)in [V(PF ) J" : 10.3 Hz; 2J(VP) in[V(CO)5P MeJ": 40 Hz; 2J(VW) in[VW5019]'

3~: 11 Hz); they are smallerby about an order of magnitude thanthe one-bond coupling constantsJJ(VF) (88 to 140 Hz) and XJ(VP) (341Hz in [V(C0)5P2Melf]~).

1J(S1V-31P) coupling constantshave been extensively studied (31)and the results are similar to thoseobtained for J(PW), J(PPt) and J(PRh)(89).

Steric factors do not seem toimpart significant influences on thesize of J(VP), except where extremeconditions are encountered as withthe very bulky phosphines PCy andP(t-Bu), or with disubstituted,cis-configurated complexes of compar-

. 500 Hz

Figure 14. 23.66 MHz S1V NMR spectrum of [Et4N][V(PF3)6], ca. 0.1 M in THF,at room temperature. The fine structure splitting is due to2J(51V-19F) coupling.

Vol. 4, No. 1/2 63

Table 8. Nuclear Spin-spin Coupling Constants

Complex

[CpV(H)(C0)3]-

[V(CO)sf(Me3Si0)3V=Nt-Bu[V04]

3"[V0F4]" 263°K

253°K193°K

[VF6f (MeCN, 253°K)

V(C0)3(N0)(PMe3)2

[V(PF3)6]-

[{V(C0)s}2y-P2Me4]2"

[V(C0)5P2Me4r

cis- [V(CO)4PCy3P(0Me)3 ]"

[VW5019]3"

Coupling Constant

J I, nXJ(13C->J(14N-\J(l70-XJ(19F-

1 T / 1 3T»

XJ(31P-aJ(31P-2J(19F-XJ(31P-1J(31P-2J(31P-

W)51V)5 1 V ) e

51V)S1V)

S1V)S 1 V ) /51V)51V)51V)siy)5 1V)5 1V)

1J(31PCy-51V)1JC31P(OMe)-slV)2J(51V- 1 8 3W)

[Hz]

20.3±0.21669561.6±2.5a

116±512014088

74b

51010.32073414019338111

References

4694,9696a7866676714

5043433535ccc83

[V(C0)5PZ3 =

PZ3]: PF3

P(0Me)3

P(NMe2)3

PMe3

PPh3

PH3

cis-[V(C0)4(PZ3)2rPZ3 =

CpV(CO)

PZ3 =

= PF3

P(0Me)3

PMesPH33PZ3= PF3

P(0Me)3

P(NMe2)3

PMe3

PCy3P(t-Bu)3

PH3PPh3

cis-[CpV(C0)2(PZ3)2

PZ 3 == PF3

P(0Me)3

P(NMe2)3

PMe3

xJ( s lP- 5 1V) d

488370230210375170

498370200130

427300230150110160155157

430300240160

26263126c31

26262631

2626312631c3144

26263126

,31,33,31

,31,31

,31,33,31

,31,33,31

,33

533

,33,33

,33,45,33,45

,33

,33


Table 8. Continued

Complex Coupling Constant [Hz] References

trans -

PZ3

[CpV(C0) 2 (PZ 3 ) 2 ]

= PF3

P(0Me)3

PMe3

CpV(NO)2PZ3

PZ3 = P(0Me)3

PMe3

P ( t - B u ) 3

PPh3

446370210

623403342403

2626,3126,31

48474747

a) Obtained from the 170 NMR spectrum. The coupling constant obtained fromthe 51V NMR spectrum of 170 enriched V04

3" is 63 ± 5 Hz. b) Eight-line pat-tern (relative intensities 1:1.6:2.3:2.5:2.5:2.5:2.3:1.6:1). c) Unpublishedor revised. d) Selected data. See also Table 1. e) 1:1:1 Triplet.

atively spacious phosphines(cis-[Et N][V(C0) (PPhMe) ] andcis-[CpV(C0)2(PPh2Me)2J (39)), orwith chelate four-ring structures,where hindered o-overlap may beresponsible for a small o(s)2 term inequation 13 and small coupling con-stants (unresolved spectra).

More pronounced is the dependenceof J(VP) upon the type of complex orlocal symmetry, respectively.Coupling decreases in the order

[CpV(H)(C0)2L] CpV(N0)2L » [V(CO)SL]"»cis-[V(C0) l fL2]"> CpV(C0)3L~ t rans-

[CpV(C0)2L2] »cis-[CPV(CO) L ]

3'for a given phosphine L such as PMeP(0Me)3) P(NMe2)3, PH3 (cf. Table 8).The larger coupling in the pseudo-oc-tahedral anionic complexes relativeto that in the square-pyramidal neu-tral complexes may be due to thehigher s-contribution to. the hybrideorbitals of the former (d2sp3) ascompared to those of the latter(d3sp3) [see, e.g., the discussion of

variations in J(RhP) for differenthybridization states of rhodium inreference (97)]. The ratio of J(VP)for [V(CO)5PZ3r/CpV(CO)3PZ3 variesfrom 1.1 : 1 to 1.4 : 1. The greaterJ(VP) values in trans than incis-[CpV(CO)2(PZ3)2] parallels ananalogous observation for pentacar-bonylphosphinetungsten-(O) complexesand was explained by enhancedIT-interaction (giving rise toenhanced synergetic o-donation of thephosphine ligand) in the trans com-plex (via two orbitals, namely thedxz and d z ) (98). However, a con-trary trend (J cis > J ,„„. ) isobserved in most Pt(89,99,100).

The most obviousJ(VP) originate from electronic fac-tors. J(VP) values correlate withthe Allred-Rochow electronegativitieslXz of the substituents Z in thephosphine PZ3, and similarily withthe inductive Taft constants Z(a,)z(31). Very bulky phosphines excluded(Tolman's cone angle > 170°), the PZ3

trans

and Rh complexes

alterations in

Vol. 4, No. 1/2 65

can be arranged in order of increas-ing coupling (R = alkyl)

PH3 > PR3 > P(NR2)3 ~ PPh3 > P(0R)Ph2» P(0R)2Ph > P(OR)3 > PF3

The increase of coupling withincreased eiectronegativity of theligand or the substituents on thedonor function of the ligand (seealso Figure 15) has been describedfor a variety of metal-X couplingconstants (89, 101-105). From IRdata it is well known

12-

£ 9 -

w 6

CpV(CO)3PZ3

»epto

-10

,PPhH2

100 150 200 250 300 350 400

1 j (51y31 p ] COUPLING CONSTANT (Hz)

Figure 15. 'Jj(s H'-3*P) couplingconstants of TIN!-C5H5V(C0)3PZ3 com-plexes vs. the Allred-Rochow elec-tronegativities X of Z. epto = 4-ethyl-2,6,7-trioxa-l-phosphabicyclo-[2.2.2.]octane.

that increasing eiectronegativity ofZ increases the ir-acceptor power ofPZ3. Since the Fermi term (equation13) does not contain parametersdirectly connected to ir-electron den-sity distributions in the bond underconsideration (except, perhaps, forAE), interpretation of trends as thatshown in Figure 15 has to take intoconsideration indirect influence ofthe dominant Tr-effect on the parame-

ters13:

(i)

(ii)

|S(0)|2 and o(s)2 in equation

Substantial n-delocalizationinto suitable accepting orbi-tals located on the phosphineincreases the s-contributionto the o-bonding electron pairon phosphorus via the a/ti syn-ergism, and hence the o(s)2

term.

V -* P Tr-delocalization desh-ields the 5XV nucleus, whichresults in a contraction ofthe 4s wave function; i.e.|S(0)T7|

2 increases.

(iii) Strongly electronegative Zsimilarily cause a contractionof the phosphorus 3s wavefunction and thus increases|S(0)p|

2.

(iv) Electronegative Z should alsoincrease the s-character ofthe donating electron pairthrough rehybridization.

The net effects are large couplingconstants for phosphines such as PF3,and P(OR) , which are usually consid-ered as weak o-donors, and compara-tively small coupling constants foralkylphosphines and PH3, which areclassified as good o-donors. Foraminophosphines, R2N ~*.P p -d back-bonding, as discussed by some workers(106) should be taken into account.


A further influence should arisefrom AE-1, which depends on both,0- and TT-interaction. Strongn-accepting ligands (PF , P(OR) )give rise to large HOMO-LUMO split. -tings and hence should diminishcoupling - contrary to observation.Transitions may therefore berestricted to o-type molecular orbi-tals, or effects via AE are overrid-den by the factors mentioned above.This parameter may, however, accountfor the unexpectedly large couplingencountered with the weak ligandsPPh3 in [V(CO)5PR3]* (375Hz).

Ma-f/l-f

C. Correlations between NMR and ESRCoupling Constants (CpV(CO)3PZ3

and Fe(NO)2(PZ3)Br)

The relationship between nuclearand electronic spin resonance parame-ters with special reference to thecorrelation between NMR coupling con-stants XJ(VP) of CpV(CO)3PZ and iso-tropic ESR hyperfine coupling con-stants Aiso(P) of Fe(N0)2 (PZ3)Brcomplexes has been reported in refer-ence (44). Correlations of thiskind, describing the interactingproperties of the phosphine ligands,are not unexpected, since the iso-tropic ESR hyperfine coupling - likethe nuclear spin-spin coupling - isdescribed by a Fermi term

A.so (P)=const.ci(P)|To(0)|2 (14)

where |f0(0)|2 = A o is the s-electron

density at the nucleus of the freeatom. The coefficients cs(P) are ameasure for the a(s)(PZ ) contribu-tion to the ground state. Theydepend on the o-donor strength and -directly via a stabilization of theHOMO, or synergetically - on their-acceptor strength of PZ 3 (107).

If the bonding and antibondingo-type MO for the M-P o-interaction(M = V, Fe) are I^and ¥*, respec-tively, then

Ta * -Ma

Pa

Pa

where

(15a)

(15b)

and

ry. Now, equation 13can be written in the form

J(MP)=const.•ISp(0)I2 I SM(0) |

2

(16)

Assuming that changes in M-Po-bonding (M = Fe, V) impart equaleffects on the variables a2(l-£2) inCpV(CO)3PZ3 and CS

2(P) inFe(N0)2(PZ,)Br, i.e. assuminga(l-^2)^ « CS(P), equation (16)becomes

J(VP)=const'.|S (0)|2|S (0)|2

P V3AE"1[c|(E2a2-c|)b2/a2] (17) (17)

£ indicates contributions from otherorbitals than those considered above.

Assuming further that|S(0) |2|S(0) |2AE"1and the s-coeffi-cients a and D are approximately con-stant, one arrives, by combination ofequations (14) and (17), to a "joint"

equation for J(VP) and A (P):iso

J(VP)«A (A-A ),iso iso

where A=£2a2|Tn (0) | 2 (18)

Equation (18) relates the phospho-rus-s-orbital contribution to themolecular orbitals relevant for thecoupling. Changes in coupling con-stants can hence be inferred fromchanges in a single ground statevariable, a. The relation is inrather good agreement with findingson 19 phosphines, for which theexperimental relation is

J(VP) = 0.82A (P) +iso

2.2»10"3[Aiiso

(19)

The complexes cis-[CpV(CO)2(PZ3)2]

also fit this curve.

Vol. 4, No. 1/2 67

IV. LINE WIDTHS IN ISOTROPIC MEDIA

Line widths Avj^ of S1V NMR sig-nals for all but the carbonyl com-plexes are contained in Tables 2 to5. For selected Avi values of car-'sbonylvanadium complexes see Table 9.It should be mentioned that the lowerlimit for an accurate determinationof line widths on conventional wide-line spectrometers is ca. 50 Hz atbest.

A. Exchange Broadening

Under isotropic conditions, quad-rupolar effects are restricted tobroadening of the NMR signals byquadrupolar relaxation, which is themain relaxation mechanism in systemswith quadrupolar nuclei. A secondcontribution may arise from thedynamic behavior of a system, origi-nating either from inter- or fromintramolecular exchange processesprovided the exchange is sufficientlyfast to suppress detection of thespecific forms taking part in theequilibrium.

In various systems containingvanadates, iso- and heteropolyvana-dates, line broadening (plus anupfield shift) is observed on acidi-fication. This has been attributedto protonation (giving rise to anincrease of the electric field gradi-ent tensor at the nucleus; videinfra), together with rapid exchangebetween protonated and non-protonatedspecies. Equilibria of this kindinvolve the systems [VOJ 3" / [HVOJ 2"(72), [HVO ]2-/[V20 I"" (72), [V207]

4"/[HV2O7]

3" (72), 2[HV01J2-/[HV:>07]

3-69 70) [HVO^] ~/[H2V01(] /[V03]70)', [HxV2O7]"-

x/[(VO3)x]x- (x pos-

sibly = 3,4) (69),

(7-x)'10 28

(6ST,75,79),[V 12P0 36]

7" / [HXV 12P0 361 vy'_ *

; (73) ,

Internal/external proton exchange wasdiscussed for [1L W . ,0 ̂ ] 7~ by O'Don-nell (69). To account for theresolved 5XV NMR spectrum of [VOF.] ,

Howell and Moss have proposed rapidinter- or intramolecular siteexchange of F~ ligands (averaging ofthe field gradient to a small value)(67).

Exchange broadening contributes tothe line widths of the vanadium sig-nals of V0(0R)3/[V0(0R)3]2 and[V0(0R) 3] 2/[V0(0R) 3]n (n > 2) equi-libria (compare V0(0iPr)3 in Figure12). The main contribution, however,arises from quadrupolar broadening.

B. Quadrupolar Broadening

Line widths of nuclei possessingan electric nuclear quadrupole moment(nuclear spin > 1/2) are dominated bya quadrupolar relaxation mechanismwhich originates from the interactionof the electric nuclear field gradi-ent tensor with unsymmetrical elec-tric fields produced by polar mol-ecules in motion. For the extremenarrowing case (2ITV OT C» 1 , vo = NMRfrequency, xc = molecular correlationti le), the relaxation mechanismdescribed by

_L_ _ _1_"V T, " T2

is

10 -f (i).

,1 2

Q"<1+ M (20)

with Q' = (e2q Q)/h the electricquadrupole coupling constant (qzz

=

field gradient in zz-direction, andeQ = quadrupole moment). Tj and T2are the longitudinal and transverserelaxation times, respectively; theyare connected with the half-widththrough T2"' = TTAV , and with thepeak-to-peak width of the firstderivative of the NMR signal throughT 2"

x = /3TTAV 1/2. TI is the asymmetryparameter (TI = 0 for axial symmetry),and xc correlates with the interac-tion between solute and solvent mol-ecules; f(I) = (2I+3)/I2(2I-l) is0.136 for the 5 XV nucleus.

From equation 20, it is quiteclear that Av ^ will increase withdecreasing point symmetry (where qbecomes more effective; q = 0 for 0h

and Td symmetries) and increasing x(increasing viscosity of the


Isolution). Further, there should bea sizable electronic effect on q,imparted by the electronic structurein the coordination sphere of thenucleus. Thus, for MABg systems(e.g. [V(CO)5PZ3]"), q is related(108) to the metal-to-ligand IT- andthe ligand-to-metal o-interaction via

C-IT 4-trB A

a +4 oB 3 A

(21)

In Table 9, selected line width dataon carbonylvanadium complexes arelisted. No quantitative interpreta-tion of the trends can yet be carriedout. Qualitatively, the results maybe summarized as follows:

(i) A decrease of symmetry (0h •*C4v •* C 2 v -> Cs; C4v -* C.)results in a substantialincrease in Av

1/2

(ii) Increase(PPhMe2

in ligand* PPh, -+

bulkinessP(t-Bu)j)

increases the solvent-soluteinteractions and thus A

(iii) Line widths increase as thebonding orbitals of thedonor/acceptor function becomemore diffuse (P -* As -»• Sb) .

(iv)

(v)

Strong ii-accepting ligands(PF3) and o-donating ligands(PPnMe2) give rise to narrowlines.

Enhanced strains inring systems (5-ring-* 4-ring) increase AV

chelate--* 6-ring

The relation between the electricfield gradient tensor q and the localsymmetry at the vanadium nucleus haseffectively been employed in theassignment of NMR signals of vana-dates(+V) • and polyvanadates(+V).O'Donnell and Pope (69) summarize therelations as follows:

Site of symmetry of V Range of line widths (Hz]

tetrahedraltetragonal (no neighboring Vas in [VW5O19 ]3")

tetragonal (with neighboring Vas in [V10028]

6~)

rhombic

60 - 100

60 - 80

100-200

200-800

Vol. 4, No. 1/2 69

Table 9. Selected slV Line-width Data of Carbonylvanadium Complexes

Complex LocalSymmetry

/[Hz]

6(S 1V)[ppm]

(1) Effect of point symmetry and donor/acceptor function (coordinatedatoms underlined)

(a) CpV(CO) derivatives

CsCsCsCsCsCsCs

CpV(C0)3o-C6H4(PPh2)2

CpV(CO)3o-C6Hlt (AsPh2 ) (SbPh2 )CpV(CO)3o-C6Hlt (AsPh2 ) (SbPh2 )CpV(C0)2o-C6Hlt (AsPh2 ) (SbPt^ )CpV(CO)3o-C6Hl4 (Ph2P) (BiPh2)cis-[CpV(C0)2PhP(CH2CH2PPh2)2]CpV(C0)PhP(CH2CH2PPh2)2

(b) [V(C0)6]" derivatives

[V(CO)6]"

c i s - [V(C0) l t (PPh 2 Et) 2 ] -mer- [V(CO)3 PhP(CK, , ^ 2

H3-C3HgV(CO)3PhzP(CH2)2PPh2

C2vCs

12130245320440500430620

1.4±0.312077519002400

-1534-1312-1236-1382-1008-1260-1176-970

-1952-1842-1685-1736-1492

(2) Ef fec t s of po in t symmetry and l igand s t r e n g t h

(a) PPhMe complexes

[V(CO)5PPhMe2]"cis-[V(C0) [ t (PPhMe ) ] *CPV(C0)3PPhMe2

(b) PF complexes

[V(CO) PF3]~c i s - [ V ( C 0 ) l t ( P F 3 ) 2 ] 'CpV(CO)3PF3

cis- [CpV(C0) 2 (PF 3 ) 2 ]

c

C2v

Cs

Cs

(3) Effect of ligand bulkiness in [V(C0).PZo] complexes

PZ3 = PPhMe2 cone angle 122 CPPhjMe 132 C,"

PPh3PCyP(t-Bu),

145170182

325960

40483131

s

3216013689420

-1855-1710-1396

-1961-1967-1565-1608

-1855-1839-1805-1854-1738


* . . - • - . < •

Thus, the narrowness of the signalsfor [VO.J3" and [V207]*" is consis-tent with the symmetrical electricalenvironment of the vanadium nuclei inthese anions. Protonation destroysthe high point symmetry, and the sig-nals broaden. On the basis of symme-try considerations, the two compara-tively narrow lines (ca. 150 Hz) of

[V10°28]6~ a t ca- ' 5 0° a n d " 5 2° PPm

(see Section II. B. 4. and Table 5)can be assigned the "outer-plane"vanadium sites with terminal oxygen(b and c in Figure 9), while thethird signal at ca. -420 ppm (Av =350 Hz) corresponds with the ' in-plane" (site a in Figure 9) vanadiumatoms (69,79). Line widths changesfor the three signals on lowering thepH from 7 to 2 were also employed todetect the protonation sites in thedecavanadate anion (79).

On the grounds of line-width argu-ments, the [VOJ* (or [V02(H20)lf]

+)ion present at pH values < 2(Av . ~400 Hz) was assigned a cisoidarrangement of the two 02"° ligands(69), while the additional broadening( -> 1600 Hz) in HC1, H^O^ and H3POifmay indicate the formation of pseudo-octahedral, cis-configurated (C2v)complexes with the incorporation ofchloro and sulfato ligands. InHSC103 and HSF03, the signals narrowto 60 Hz, which was tentativelyattributed to the formation of thepseudo-tetrahedral anions [V02F2]~and [VO2C12]~ (69); the local symme-try remains C 2 v, though. Likewise,it was argued on the basis of line-width differences between [VW5019]

3~(34 Hz) and [V2Vv

74 O19 ]*" (68 Hz) that

the two V06 groups in the latteroccupy cis-positions in the overallstructure (73).

Since the correlation time T isclosely related to the viscosity,line widths increase as the tempera-ture decreases (Figures 11 and 12 a).In O'Reilly's quasi-lattice model(109) relaxation takes place byjump-reorientation into empty sitesof a hypothetic lattice. Evaluationof T in the line of this model (110)leads to

, exp(E/RT) (22a)

where E is the activation energy ofrotational reorientation, and vf isthe "free volume" (110). The Av, /Tcorrelation obtained from equation22a,

Av1/2=AT"l/2exp(B/T) (22b)

is in excellent agreement with exper-imental data for V0(0i-Pr)3 dissolvedin THF (temperature range 300 - 180K, Avi/2 = 10 - 600 Hz; A = 0.39404,B = 1767.56, activation energy forthe rotation barrier 14.7 kJ) (65).

Gillen and Noggle (111) investi-gated the temperature dependence ofthe relaxation times of the 35C1 and5 H7 nuclei in neat VOCI3, for whichanisotropic reorientation is assumed.From the data (summarized in Table10), they calculated theArrhenius activation energies E forthe appropriate nuclei and also therotational diffusion constants forrotation about the C3 axis (D,,) and

perpendicular to it (D.). RecentT, -studies on neat VOC13 have con-firmed that relaxation in this systemis purely quadrupolar (2,10).

V. QUADRUPOLE EFFECTS INANISOTROPIC MEDIA

The shape of spectra obtained inanisotropic media (liquid crystals,polycrystalline powders, single crys-tals) is mainly defined by quadrupoleinteraction (line broadening, firstand second order quadrupole split-ting) and anisotropic chemical shifteffects. The broadening of lines insolids can additionally be caused bymagnetic dipolar coupling. Werestrict our attention to effectsoriginating in quadrupolar interac-tions .

A. Liquid Crystals

In anisotropic mesophases, the N'MRsignals of quadrupolar nuclei splitdue to the non-vanishing electric

Vol. 4, No. 1/2 71

Table 10. Relaxation Data for Neat VOCI3

339.2316.6297-T273.9255.0

35Cl (Ref, 113)

E =

21k.6

.91 ± 0.17

T2*105 [s]

2U.2 ± 0.719.i ±0.1+15.3 ± 0.1+11.5 ± 0.38.53 ± 0.25.81 ± 0.23-87 ± 0.1

51V (Ref. 113)

DM(303°K) =

D, (303°K) =

298.1279.9255.023^-0217.5199.0

E = 8.2i+ ± 0.25 kJ/xMol

7.O'1O10 s"1

1+.9-1010 s"1

Ti'103 [s]

23.2 ± 0.319-2 ± 0.313.8 ± 0.29.-72 ± 0.156.89 ± 0.10I4-.U2 ± 0.07

T!(V)Ti(35Cl)T2(

51V)T2(

35C1)TcRef.

18 ± 2

16

15(2)

16.30.202.6(150)

0.12

O.lU3.2 .(151)

10-12

msmsmsmss


20 kHz

72/4 A

Zeemann-| Quadrupole-Splitting

Ase?qQ{3cos2e-1)/8I(2I-1J

5 kHz

Figure 16. First order quadrupolesplitting of the 5 H? NMR signal(first derivative) of V0(0i-Pr)3 dis-solved in the liquid crystal MBBA.Two different orderings (anisotropyaxes) are shown. The central linecorresponds to the +1/2 -*• -1/2 tran-sition. Magnetic field = 1 T, fre-quency = 11.2 MHz.

Figure 17. Energy level diagram forthe Zeemann splitting in an externalmagnetic field and first-order qua-drupole perturbation for a nucleus ofnuclear spin 7/2 (eQ > 0). The scalefor the quadrupole splitting is exag-gerated by a factor of ca. 103. Theincrease from AE1 to AE2, AE2 to AE3etc. is 24/4 A, hence the equidistantsignals in Figure 16.

3e2qQSfield gradient. Accordingly, the 5*VNMR signals of vanadyl esters dis- E=YhBo m+ 4 W 2 I - I ) (3cos*e ~:

solved in liquid crystals show aseven line pattern resulting from afirst order quadrupole splitting asshown in Figures 16 and 17, anddescribed by equation 23 (112):

[3m -1(1+1)] (23)

The first term corresponds to theZeemann splitting in the external

Vol. 4, No. 1/2

magnetic field Bo , and the second tothe quadrupole splitting. 8 is theangle between the director in theliquid crystal and Bo, and S is anordering factor, denoting the degreeof order with respect to a selectedmolecular axis: S = 1/2 < 3cos' - 1>(for cylindrical symmetry), with <f>the angle between the director andthe molecular axis (i.e. the largestcomponent of the field gradient eq).Assuming that (3cos2$ - 1) is aver-aged to one (113), the quadrupolesplitting Av between two adjacentlines becomes

(x = 0 - 0.08) (124) were studied by51V NMR; and phase transfer in solidsolutions was investigated in

B-Na 0.33 V 20 s (125) and V02 containingadditives of niobium, chromium andferric oxide (119; see also Table11).

A typical S1V NMR spectrum (poly-crystalline VOC13, (150)) is shown inFigure 18. The line pattern origi-nates from first order quadrupolesplitting. A similar spectrumwas reported for single crystals

Av = 3e2qQS/4hI(2I-l) (24)

If S is known, quadrupole couplingconstants can be calculated (videinfra). Figure 16 illustrates thesituation for a solution of vanadyltriisopropylate in the nematic liquidcrystal N(p-methoxybenzyliden)p-n-bu-tylaniline (MBBA). The splitting Avdepends on the concentration and thetemperature and is further a functionof the mesophase and the vanadylester employed (150).

B. SoSids

5*V NMR results of oxidic vanadiumcompounds have been reviewed recentlyby Pletnev (16). We thereforerestrict our treatment of this widearea to a compilation of data (Table11) and a brief mention of severalfeatures.

Direct 5XV NMR investigations havebeen carried out on V 20 5 (115-117),VnOa,,., (n = 3-8) (118), V02

(119-122), V01<25 (123), V 20 3 (124),vanadium bronzes (125), metavana-dates (126-132), divanadates(126,133,134), orthovanadates (126,135-145), VB (146) and V3Si (147a),mostly employing polycrystalline pow-ders. However there are also somestudies on single crystals(115,116,130,133,134, (141-143) andon amorphous systems (117,131). Thephase diagram for FexNbxV2-2x O4(x = 0 - 0.09) (121) and V203+x

100 KHz

Figure 18. 11.20 MHz sXV NMR spec-trum of VOCI3 at ca. 90°K, the lowerpart showing a section with the splitcentral line and the first satellite(150). Computer simulation fore2qQ/h =5.7 MHz indicates an asymme-try parameter in=0.08 (compare theresults by Allerhand (149), obtainedfrom the evaluation of the two compo-nents of the central line:e2qQ/h =5.4 MHz, n< 0.4). The spec-trum was obtained on a Varian DP 60spectrometer at 0.9935 T (sweep width0.0415 T), with the RF field set tominimal power (10yA on the RF controlunit V-4210A). Modulation 0.5 mT; 85scans.


Table 11. 51V NMR Data of Crystalline Vanadium Compounds8"

Compound

V205 .v 2 o 5

b 'c

V02 ,Cr Nb V 0

X X 2-2X it

(x = 0.01-0.09)

Fe Nb V 0 d

X X 2~2X It

(x = 0.01-0.09)LI3VO4Na3VO.lt C

Ca3(V0O2 fSr3(V0O2Ba3(VO<t)2

ScVOitScVOitYVO1+YV0.+LaVOitCeV04

CeVOuPrVOitYbV04

CrVOit

FeVOit

Co3(VOi+)2

Ni3(V0if)2

Ni 2V z0 7

e2qQ/fi

0.8 ± 00.8056.72 ±

6.72 to

6.72 to

1.51 ±1.05 ±2.05 ±0.53 ±0.75 ±3.80 ±

4.75 ±

5-21 ±5.23 ±5-33 ±5.11 ±4.22 ±2.19 ±2.81 ±2.15 ±1.68 ±2.162.48 +2.15 ±2.58 ±3.35 ±

[MHz]

.06

0.14

4.76e

4.34e

0.010.030.030.020.020.04

0.04

0.070.090.070.090.040.0U0.076

0.040.146

0.086

0.0Uo.oie

0.06

n

0.040.51±0.

0.74 to

0.50 to

0

0

0.69 ±0.69 ±

0.76 ±0

02

0.73

0.79

0.010.03

0.03

2.367-0.33

1.0 ±

0.6 ±

3.5 ±

ggggg

0 . 2

•0.2

ii

i

0.04

02

-5.934-0.39

1.6

1.6

5-5

03

3.567-0.12

0 .8

0 .8

-20.7

i s o

-0.28-0.28

to -0.26-0.19to -0.27

-0 .1 ± 0.5+0.3 ± 0.6+0.1 ± 0.5-0.3 ± 0.5-0.1+ ± 0.5

0.96

0.970.7 ± 0.50.5 ± 0.3

-1.251.1 ± 0.04

0.31

2.00

1.08

O.76 ± 0.02

0.56 ± 0.01

Refs.

115116119,122

119

119

135135135135135136137136137138138137,144138136126139126139126139126139126

Vol. 4, No. 1/2 75

Table 11. Continued

Compound e2qQ/h [MH] (73 a.ISO

Refs.

Ni(VO3)2NH4VO3

NaVO3NaV03

KVO3KVO3 (glass)KVO3 b

KVO3CsVO3RbVOaMg(VO3)2Ca(V03)2

n7-c7H7v(co)3VsSi1

VOF3VOCI3VOC13VOCI3 "VOCI3 P

V0Cl2(0i-Pr)P

V0Cl2(0i-Bu)P

V0Cl2(0Me)P

V0Cl(0i-Pr)z P

V0(0i-Pr)3VO(Oi-Bu)3

q

V0(0n-Pr)3q

mn

4.03 ± 0.062.88 ± 0.062.76 ± 0.033.65 ± 0.063.94 ± 0.044.05 ± 0.05; 2.14.36 ± 0.063.354.22 ± 0.154.34 e

3.84 ± 0.044.33 I6.19 I2.81 e

2.4 ± 0.22.924 ± 0.0158.9 ± 0.45.4 0.22.985.7 ± 0.27.5 ± 0.24.34.76.07.14.4 0.3

0.3

0.3 ±0.37 ±0.60 ±0.64 ±0.22 ±0.75 ±

0.650.770.630.720.630.60

0.100.050.100.050.030.10

± 0.15

± 0.05

0ca 0^ 0.4$ 0.4

0.08

-1.4 ± 0.2 -0.7-1.7 ± 0.2 -0.7-1.4 ± 0.2 0.8-2.0 ± 0.2 -1.9

-2.3 ± 0.2 -0,3-1.2 -0.3

-0.6 -0.9-2.3 ± 0.2 -1.5-1.4 -1.7-2.4 -2.4

6.45.0 0.4

0.23 ± 0.01 1262.7 1273.0 1283.5 1273.2 128

1292.1 1273.1 131

1302.2 1323.3 1281.7 1320.9 132

132

1481471491492150150150150150150150150150


a) See text for explanation of sym-bols. Standard for the shift valuesis, if not indicated otherwise, aque-oous KV03. Data, except where statedotherwise, at room temperature. b)Single crystal study. c) Standard =Na3V04/aq. d) At 250°K and 315°K5respectively, for details see Ref.119. e) By multiplication of vQ by14. f) Standard = Sr3(V04)2/aq. g)0 aniso values are as follows: +0.8,-0.9, -0.7, -0.3, -0.4 ± 0.5 h) Dataon other MV04 (M = Pr, Nd, Sm, Eu,Gd, Tb, Dy3 Ho, Er, Tm, Yb, and Lu)are also reported. i) oanjso =-0.3s

of V2O5 (115), where the splitting ofthe central line (m = +l/2^m = -1/2)is, however, caused by anisotropicshift interactions. Polycrystallineand amorphous V 20 5 was investigatedby Legrand and co-workers (117). Atfrequencies >6 MHz, the increase of^v 1/2 with the applied frequencyindicates predominance of magneticinteraction (more substantial incrystalline than in amorphous V2O5),while at frequencies <6 MHz, thedecrease of line widths with theapplied frequency both for crystal-line and amorphous V2O5 indicatespredominance of quadrupolar effects.This leads to the assumption of iden-tical local structures at the S1Wsites in either species.

An extensive discussion of thetheoretical background and analysisof the spectra is given by Baugher etal. (127). The evaluation is carriedout according to the following equa-tions (152)

-0.3, +0.2; =-0.5, -0.4, -1.4 in

A1 = 3e2qQ(l-Ti)/2I(2I-l)h

A, = 25[I(I+l)-3/4](e2gQ)2

64[I(2I-l)]zhzvo

K 135 45 ;

(25)

(26)

A1 is the splitting of the centralline, A2, the distance between thecentral line and the first satellite,and vo is the generator frequency.It has become common in solid state

the order LaV03, CeV03, PrV03.j) Twotypes of non-equivalent vanadiumsites. k) Indirectly determined insolution from the 5XV spin-latticerelaxation time and from the *H lineshape, respectively. 1) At 28°K. m)At 167°K. n) Determined indirectlyfrom relaxation measurements (cf.text). o) At ca. 90°K. p) Deter-mined from e2qQ/h(3SCl), Av%(35C1-NMR) and Av^(slV NMR). q)Determined indirectly from quadrupolesplittings in nematic liquid crys-tals.

51V NMR, to relate the shift valuesto aqueous KV03 (6(51V) = -576 ppmupfield of VOC13) and to quote valuesin units of 1G*4 %.

The results for metavanadates showthat the tetrahedral symmetry of thebasic V0 4 tetrahedron is distorted.It is expected that the primary con-tribution to the quadrupole couplingconstant will be due to substantialTT-bonding between vanadium and oxy-gen. The nther large variations ine2qQ/h for alkaline metal metavana-dates (1.51 - 4.36 MHz) are, however,not easily explained by a iT-bondtheory. In the sodium metavanadatedihydrate structure, the vanadiumatoms are located in two types ofnonequivalent sites with differentelectric field gradients at the vana-dium nuclei (129). A *H and 23Na NMRstudy was also carried out on thiscompound (129). In two papers(132,135) the electric field gradienttensor q has been obtained fromsemi-empirical Mulliken-Wolfsberg-Helmholtz calculations. The diago-nals are

[V0, [VO, 3-

(Ref. 132)107.9461.87

-169.83

(Ref. 128)-43.16480.000123.000

The results for orthovanadates show asimilar range of e2Qq/h values. Itwas noted that the molecular orbitalresponsible for the hyperfine inrer-actions at the 5XV nucleus is a

Vol. 4, No. 1/2 77

Ana(s) type orbital (139). For twocompounds, ESR spectroscopic g-fac-tors are reported: CrV04, g = 1.98;FeVO4, g = 1.96 (126).

First-order quadrupole splittingsof the S1V NMR signals of singlecrystals of TmV04 and GdV04 wereobserved at temperatures close toabsolute zero. For TmVO at 4.5°K, aseven line pattern is obtained whichsplits into a "double" spectrum at1.5°K, due to the two principal axesof Jahn-Teller distortion in the(001) plane (141,142). The quadru-pole splitting is 160 kHz. A four-teen line system is also observed inGdV04 (below 2.495°K) which arisesfrom the two different vanadium sitesproduced by magnetic ordering by Gd3+

(143).Apart from vanadates(+V), informa-

tion on quadrupole coupling constantsfor vanadium compounds is scarce(148-150, 2). We have determinede2qQ/h values for VOC13 and severalvanadyl ester chlorides indirectlythrough the 35C1 quadrupole couplingconstants measured at 77°K, and slVand 3SC1 NMR line widths in the neatsamples at room temperature (150).Comparison between e2qQ/h for VOC13obtained by this method (7.5 MHz) andfrom powder spectra [ca. 5.6 MHz(149,150)] show that the former pre-sumably is too large by approximately40%. One has, however, to keep inmind that in a frozen liquid such asVOC13, partial orientation is likelyto occur. The e2qQ/h value of VOC13

was also estimated from relaxationdata [2.98 MHz, (2)], using theDebye-Stokes-Einstein equation forthe calculation of xc; it seemsdoubtful, however, that this equationcan be applied to a neat liquid. Fora few vanadyl triesters, couplingconstants were calculated from qua-drupole splittings of the NMR linesin nematic liquid crystals, takinge2qQ/h[V0(0i-Pr)3] as obtaineddirectly from a powder spectrum forthe determination of the orderingfactor (150).

Vf. CONCLUDING REMARKS

Among the quadrupolar transitionmetal nuclei, the vanadium nucleusmay be considered the most favorableone for carrying out NMR experiments.This is due mainly to its rathersmall electric nuclear quadrupolemoment (ca.|0.05|*10"28 m 2 ) , keepingquadrupolar relaxations rates suffi-ciently low to allow observation ofwell resolved spectra even for com-pounds of low point symmetry. At thesame time NMR parameters arising fromquadrupolar interactions (line widthsin isotropic media, quadrupole split-tings under anisotropic conditions)can be obtained in addition to theparameters commonly observed withspin 1/2 nuclei (chemical shifts andnuclear spin-spin coupling con-stants), providing a comprehensiveset of data for different fields ofapplication.

Due to the wide range of chemicalshift values (caused by variations inthe 2nd order paramagnetic deshield-ing contributions), spanning about2500 ppm, the chemical shift is asensitive probe to alterations in thefirst and also in the second coordi-nation sphere of diamagnetic coordi-nation compounds of vanadium in theformal oxidation states, -I, 0, +1,+11I and +V. Applications go, how-ever, beyond the identification ofspecies, structural assignment andinvestigations of mechanisticaspects.

One of the areas of great promiseencompasses methodological-theoreti-cal statements on the basis of semi-quantitative analyses of 51V NMRparameters, which will become morepredominant as the NMR instrumenta-tion is improved continuously andfeasible vanadium complexes areincreasingly prepared. The impor-tance of vanadium NMR will thus shifttowards a better understanding of theelectronic interactions between vana-dium and its ligand systems, provid-ing results which may well apply tocoordination compounds of other tran-sition metals which are less accessi-ble to NMR.


References

*V. Saraswati, J. Chem. Phys. Soc.Jpn. 23, 763 (1967).

2M. A. Habayeb and 0. E. Hileman,Jr., Can. J. Chem. 58, 2115 (1980).

3W. J. ChiIds and L. S. Godman,Phys. Rev. 156, 64 (1967).

fcK. Murakawa, J. Phys. Soc. Japan21, 1466 (1966).

5R. A. Bennett and H. 0. Hooper,J. Chem. Phys. 52, 5485 (1970).

eK. Murakawa, J. Phys. Soc. Japan11, 422 (1956).

7J. 0. Artman, Phys. Rev. 143. 5 4 1

(1966).8E. Ertmer, U. Johann and G.

Meisel, Phys. Lett. B_8_5, 319 (1979).9W. J. Childs, 0. Poulsen, L. S.

Godman and H. Crosswhite, Phys. Rev.A19, 168 (1979).

10O. Lutz, W. Messner, K. R. Mohn,and P. Kroneck, Z. Phys. A300, 111(1981).

li:H.-Ch. Bechthold and D. Rehder,J. Organometal. Chem., 206>, 305(1981); ibid. 233, 215 (1982).

12H. Ch. Bechthold, A. Kececi, D.Rehder, H. Schmidt, and M. Siewing,Z. Naturforsch. B37, 631 (1982); A.Kececi and D. Rehder, Z. Naturforsch.B36, 139 (1981).

13A. Yamasaki, Denki Tsuhin DaigakuGakuho 27, 291 (1977).

^R. G. Kidd and R. J. Goodfellow,in NMR and the Periodic Table, R. K.Harris and B. E. Mann, Eds., AcademicPress, New York, 1978, p. 195.

15R. G. Kidd, in Annual Reports onNMR Spectroscopy, G. A. Webb, Ed.,Academic Press, New York 1978, Vol.10A, p. 1.

16R. N. Pletnev, V. A. Gubanov, andA. A. Fotiev, YaMR v Oksidnykh Soye-dineniyakh Vanadiya, Nauka, Moscow,1979.

17A. Saika and C. P. Slichter, J.Chem. Phys. 22, 26 (1954).

18K. A. K. Ebraheem, G. A. Webb,and M. Witanowski, Org. Magn. Reson.

Vol. 4, No. 1/2

8, 317 (1976).19D. Rehder, Z. Naturforsch. B32,

771 (1977).20R. Talay and D. Rehder, Chem.

Ber. m , 1978 (1978).21N. F. Ramsay, Phys. Rev. 7̂8. 699

(1950).22N. F. Ramsay, Phys. Rev. 86, 243

(1952).23J. A. Pople, Discuss. Farad. Soc.

34, 7 (1962).2"J. A. Pople, J. Chem. Phys. 37,

53 (1962).2 5T. Nakano, Bull. Chem. Soc. Japan

50, 661 (1977).2 6H. Schmidt and D. Rehder, Trans.

Met. Chem. 5, 214 (1980).27J. S. Griffith and L. E. Orgel,

Trans. Farad. Soc. 53, 601 (1957).28J. S. Griffith, Ed., The Theory

of Transition-Metal Ions, CambridgeUniversity Press, New York/London,1964, p. 284.

29N. Juranic, M. B. Celap, D.Vu6elic, M. J. Malinar, and P. N.RadivojSa, J. Magn. Reson. 35, 319(1979).

30D. Rehder, J. Organometal. Chem.37, 303 (1972).

31D. Rehder, W. L. Dorn, and J.Schmidt, Trans. Met. Chem. 1, 233(1976).

32D. Rehder and J. Schmidt. J.Inorg. Nucl. Chem. 36, 333 (1974).

33D. Rehder, J. Magn. Reson. 25,177 (1977).

33aR. Borowski, D. Rehder, and K.von Deuten, J. Organometal. Chem.220, 45 (1981).

3*R. Talay and D. Rehder, Z. Natur-forsch. B36, 451 (1981).

35H. Baumgarten, H. Johannsen, andD. Rehder, Chem. Ber. 112, 2650(1979).

36D. Rehder, J. Organometal. Chem.137, C25 (1977).

37H.-Ch. Bechthold and D. Rehder,J. Organometal. Chem. 172, 331C1979).

38W. R. W. Roose, D. Rehder, H.Liiders, and K. H. Theopold, J. Orga-nomet.ai. Chem. 157, 311 (1978).

39D. Rehder, JT~ Magn. Reson. 38,419 (1980).

U0D. Rehder, L. Dahlenburg, and I.

79

Miiller, J. Organometal. Chem. 122,53 (1976).

klI. Milller and D. Rehder, J. Orga-nometal. Chem. 139, 293 (1977).

<l2D. Rehder and U. Puttfarcken, J.Organometal. Chem. .184, 343 (1980).

113D. Rehder, H.-Ch. Bechthold, andK. Paulsen, J. Magn. Reson. 40, 305(1980).

**D. Rehder and J. Schmidt, Trans.Met. Chem. 2, 41 (1977).

U5D. Rehder, W. L. Dorn, and J.Schmidt, Trans. Met. Chem. j., 74(1976).

1(SU. Puttfarcken and D. Rehder, J.Organometal. Chem. 1&5, 219 (1980);D. Rehder and U. Puttfarcken, Z.Naturforsch. B3_7, 348 (1982).

*7F. Naumann and D. Rehder, J.Organometal. Chem. 204, 411 (1981).

fc8M. Herberhold, and H. Trampisch,Z. Naturforsch. B37, 614 (1982).

*9U. Puttfarcken and D. Rehder, J.Organometal. Chem. 157, 321 (1978).

50J. Schiemann, E. Weiss, F.Naumann, and D. Rehder, J. Organome-tal. Chem. 232, 219 (1982).

S1K. von Deuten, D. Rehder, and W.Roose, J. Organometal. Chem. 214, 71(1981).

52U. Mollers, Dissertation, Ham-burg, 1979.

5 3 0 . W. Howarth, private communica-tion.

54R. L. Martin and A. H. White,Nature 223, 394 (1969).

S5A. Yamasaki, F. Yayima, and S.Fujiwara, Inorg. Chim. Acta 2, 39(1968).

56A. Yamasaki, F. Yayima, and S.Fujiwara, J. Magn. Reson. 1, 203(1969).

57G. G. Mather, A. Pidcock, and G.J. M. Rapsey, J. Chem. Soc. DaltonTrans., 2095 (1973).

S 8P. B. Hitchcock, B. Jacobson, andA. Pidcock, J. Chem. Soc. DaltonTrans., 2038 (1977).

59J. MacLeod, L. Manojlovic-Muir,D. Millington, K. W. Muir, D. W. A.Sharp, and R. Walker, J. Organometal.Chem. 97, C7 (1975).

6DP. S. Braterman, R. J. Cross, L.Manojlovic-Muir, K. W.- Muir, and G.

B. Young, J. Organometal. Chem. 84,C40 (1975).

S1G. J. Palenic, M. Mathew, W. L.Steffen, and G. Beran, J. Am. Chem.Soc. 9_7, 1059 (1975).

62W. L. Steffen and G. Palenic,Inorg. Chem. 15, 2432 (1976).

6 3P. B. Hitchcock, B. Jacobson, andA. Pidcock, J. Chem. Soc. DaltonTrans., 2038 (1977).

6<tR. Lenck, Dissertation Hamburg(1973).

6SK. Paulsen, D. Rehder, and D.Thoennes, Z. Naturforsch. A33, 834(1978).

66J. V. Hatton, Y. Saito, and W. G.Schneider, Can. J. Chem. 43, 47(1965).

67J. A. S. Howell and K. C. Moss,J. Chem. Soc. A, 270 (1971).

68Yu. A. Buslaev, V- D. Kopanev, A.A. Konovalova, S. V. .Bainova, and V.P. Tarasov, Dokl. Akad. Nauk SSR 243,1175 (1978); Dokl. Chem. 243, 583(1978).

6 9S. E. 0'Donnell and M. T. Pope,J. Chem. Soc. Dalton Trans., 2290(1976).

7 0 0 . W. Howarth and J. R. Hunt, J.Chem. Soc. Dalton Trans., 1388(1979); E. Heath and 0. W. Howarth,ibid. 1105 (1981).

7 1C. M. Flynn, Jr., M. T. Pope, andS. O'Donnell, Inorg. Chem. 13, 831(1974).

7 2 0 . W. Howarth and R. E. Richards,J. Chem. Soc, 864 (1965).

7 3L. P. Kazanskii and V. I. Spit-syns Dokl. Akad. Nauk. SSR 223, 381(1975); Dokl. Phys. Chem. 223, 721(1975).

7fcYu. A. Buslaev, V. P. Tarasov, S.M. Sinitsyna, V. G. Khelebodarov,and V. D. Kopanev, Koord. Khim. 5,189 (1979).

75A. Lachowicz and K.-H. Thiele, Z.Anorg. Allgem. Chem. 434, 271(1977).

7GK. Witke, A. Lachowicz, W.Bruser, and D. Zeigan, Z. Anorg. All-gem. Chem. 465, 193 (1980).

77A. G. Swallow, F. R. Ahmed, andW. H. Barnes, Acta Cryst. 21, 397(1966).

7 80. Lutz, W. Nepple, and A. Nolle,


Z. Naturforsch. A31, 1046 (1976).7 9 0 . W. Howarth and M. Jarrold, J.

Chem. Soc. Dalton Trans. , 503 (1978).80W. G. Klemperer and W. Shum, J.

Am. Chem. Soc. 99, 3544 (1977).8 1L. P. Kazanskii, M. A. Fedotov,

M. N. Ptushkina, and V. I. Spitsyn,Dokl. Akad. Nauk. SSR 240, 117(1978); Dokl. Chem. 240, 422 (1978).

82R. I. Maksimovskaya, M. A. Fedo-tov, V. M. Mastikhin, L. I. Kuznet-sova, and K. I. Matveev, Dokl. Akad.Nauk. SSR 240, 117 (1978); Dokl.Chem. Proc. Acad. Sci. 240, 422(1978).

8 3S. P. Harmalker, M. A. Leparulo,and M. T. Pope, XXI. ICCC, Toulouse1980, and M. T. Pope, private commu-nication.

8 3 aK. Ihmels and D. Rehder, J.Organometal. Chem. 218, C54 (1981).

8I(P. E. Garrou, Chem. Rev.81, 229 (1981).

8 5T. G. Appleton, U. A. Bennett,and I. B. Tomkins, J. Chem. Soc. Dal-ton Trans., 439 (1976).

8 6S. Hietkamp, D. J. Stufkens, andK. Vrieze, J. Organometal. Chem.169, 107 (1979).

8 7S. 0. Grim, W. L. Briggs, R. C.Barth, C. A. Tolman, and J. P. Jes-sen, Inorg. Chem. 13, 1095 (1974).

88R. Freeman, G. R. Murrary, and R.E. Richards, Proc. Roy. Soc. A242,455 (1957).

89J. F. Nixon and A. Pidcock, Adv.Nucl. Magn. Reson. 2, 345 (1969).

9 0P. S. Pregosin and R. W. Kunz, inNMR - Basic Principles and Progress,P. Diehl, E. Fluck, and R. Kosfeld,Eds., Springer Verlag, Berlin, 1979,Vol 16, p. 1.

91G. M. Whitesides and H. L. Mit-chell, J. Am. Chem. Soc. 91, 2245(1969).

92U. Franke and E. Weiss, J. Orga-nometal. Chem.153, 39 (1978).

33U. Franke, Dissertation Hamburg1978.

9*G. M. Bodner and L. J. Todd,Inorg. Chem. 13, 1335 (1974).

35H. S. Gutowsky, D.W. McCall, andC. P. Slichter, J. Chem. Phys. 21,279 (1953).

9 6P. C. Lauterbur and R. B. King,

J. Am. Chem. Soc. 87, 3266 (1966).9 6 aW. A. Nugent and R. L. Harlow,

Chem. Commun., 342 (1979).9 7E. M. Hyde, J. D. Kennedy, B. L.

Shaw, and W. McFarlane, J. Chem. Soc.Dalton Trans., 1571 (1977).

9 8S. 0. Grim and D. A. Wheatland,Inorg. Chem. 8, 1716 (1969).

" W . McFarlane, J. Chem. Soc. Dal-ton Trans., 324 (1974).

10°H. C. E. McFarlane, W. McFar-lane, and R. J. Wood, Bull. Soc.Chim. 85, 864 (1976).

101J. D. Kennedy, W. McFarlane, R.J. Puddephatt, and P. J. Thompson J.Chem. Soc. Dalton Trans., 874 (1976).

102R. Meij, D. J. Stufkens, K.Vrieze, E. Roosendaal, and H. Schenk,J. Organometal. Chem. .155, 323(1978).

103J. Chatt, R. Mason, and D. W.Meek, J. Am. Chem. Soc. 97, 3826(1975).

10ltD. S. Milbrath, J. G. Verkade,and R. J. Clark, Inorg. Nucl. Chem.Lett. 12, 921 (1976).

1 0 5 S . 0. Grim, R. R. McAllister,and R. M. Singer, Chem. Comm., 38(1969).

1 0 SC. E. Jones and K. J. Coskran,Inorg. Chem. 10, 55 (1971).

107J. Schmidt, Z. Naturforsch. B27,600 (1972).

108J. M. Bancroft, H. C. Clark, R.G. Kidd, A. T. Rake, and H. G. Spin-ney, Inorg. Chem. 12, 728 (1973).

1 B 9D. E. 0'Reilley and G. E.Schacher, J. Chem. Phys. 39, 1768(1968).

110R. T. Obermeyer and E. P. Jones,J. Chem. Phys. 58, 1677 (1973).

1X1K. T. Gillen and J. H. Noggle,J. Chem. Phys. 53, 801 (1970).

112A. D. Buckingham and K. A.McLauchlan, in Progress in NMR Spec-troscopy, J. W. Ems ley, J. Feeney,and L. H. Sutcliffe, Eds., PergamonPress, Oxford, 1967, Vol. 2, pp.63-73.

1 1 3D. Lindman and S. Forsen, in NMR- Basic Principles and Progress, P.Diehl, E. Fluck, and R. Kosfeld,Eds., Springer Verlag, Berlin, 1976,Vol. 16, pp. 233-238.

1 1 4D. Rehder, K. Paulsen, and H.

Vol. 4, No. 1/2 81

Lechert, Z. Naturforsch. A33, 1597(1978).

115P. W. France, J. Magn. Reson.34, 585 (1979); ibid. 40, 311 (1980).

1 1 6S. D. Gomostansky and C. V.Stager, J. Chem. Phys. 46, 4959(1967).

116aJ. L. Ragle, J. Chem. Phys. 35,753 (1961).

I17L. Facchini, C. Sanchez, J.Livage and A. P. Legrand, JointISMAR-AMPERE Conference, Delft 1980.

118R. N. Pletnev, A. A. Sidorov, V.A. Peveljaev, and Yu. M. Beljakov,Dokl. Akad. Nauk. SSR 245, 144(1979).

119R. N. Pletnev, A. A. Fotiev, L.V. Zolotuchina, and V. K. Kaputskin,Dokl. Akad. Nauk. SSR 241, 873(1978).

120A. G. Petrov, R. N. Pletnev, L.V. Zolotukhina, and V. A. Gubanov,Izv. Akad. Nauk SSR, Neorg. Mater.16, 678 (1980).

121L. V. Zolotukhina, R. N. Plet-nev, A. A. Fotiev, and V. K. Kaputs-kin, Zh. Neorg. Khim. 23, 27 (1978);Russ. J. Inorg. Chem. 23, 15 (1978).

12ZK. Narita, J. Umeda, and H.Kusumoto, J. Chem. Phys. 44, 2719(1966).

1 2 3S. Tagagi, Y. Kitaoka, H.Yasuoka, T. Ohtani, K. Kosuge, and S.Kachi, J. Phys. Soc. Jpn. 49, 521(1980).

12"Y. Ueda, K. Kosuge, and S.Kachi, J. Solid State Chem. 3_1, 171(1980) .

12SK. Maruyama and H. Nagasawa, J.Phys. Soc. Jpn. 48, 2159 (1980).

126R. N. Pletnev, W. N. Lisson, Yu.M. Beljukov, and J. J. Miller, Inst.Khim. (Akad. Nauk. SSSR Ural'skijFilial) Trudy. Sverdlovsk 32, 32(1975).

127J. E. Baugher, P. C. Taylor, T.Oja, and P. J. Bray, J. Chem. Phys.50, 4914 (1969).

128R. N. Pletnev, V. A. Gubanov,and A. K. Chirkov, Zh. Strukt. Khim.H , 938 (1976); J. Struct. Chem. 17,805 (1976).

129R. N. Pletnev, V. V. Gorshkov,and A. A. Ivakin, Zh. Neorg. Khim.21, 331 (1976); Russ. J. Inorg. Chem.

21, 178 (1976).1 3 0S. D. Gomostansky and C. W.

Stager, J. Chem. Phys. 48, 1416(1968).

13 XS. L. Segel and R. B. Creel,Can. J. Phys. 48, 2673 (1970).

132V. A. Gubanov, N. I. Lazukova,and R. N. Pletnev, Zh. Neorg. Khim.23, 655 (1978); Russ. J. Inorg. Chem.23, 362 (1978).

1 3 3S. I. Andronenko, L. V. Dmi-trieva, N. G. Molodchenko, V. V. Mos-kalev, and Z. N. Zonn, Fiz. Tverd.Tela 21, 914 (1979); Sov. Phys. SolidState 21, 535 (1979).

13UV. A. Ioffe, L. V. Dtnitrieva, A.P. Vereshchagina, Z. N. Zonn, and S.I. Andronenko, Fiz. Tverd. Tela 1J9,1204 (1977); Sov. Phys. Solid State19, 704 (1977).

135R. N. Pletnev, N. I. Lazukova,and V. A. Gubanov, Zh. Fiz. Khim. 5_1,2359 (1977); Russ. J. Phys. Chem. 51,1382 (1977).

136V. A. Gubanov, R. N. Pletnev, V.N. Lisson, and A. K. Chirkov, Spectr.Lett. H), 527 (1977).

137A. A. Sidorov, V. N. Lisson, R.N. Pletnev, and W. A. Gubanov, Fiz.Tverd. Tela JL7, 2427 (1975).

138G. V. Basuev, R. N. Pletnev, W.K. Slebuchin, V. N. Lisson, and G. P.Shveykin, Izvest. Neorg. Materialy.14, 711 (1978).

139R. N. Pletnev, A. A. Sidorov, V.N. Lisson, and A. K. Chirkov, Dokl.Akad. Nauk. SSR 236, 1159 (1977).

^"R. N. Pletnev, V. N. Lisson, G.V. Basuev, and G. P. Shveikin,Radiospektroskopiya (Perm), 28(1978).

llllB. Bleaney and M. R. Wells,Proc. Roy. Soc. (London) A370, 131(1980).

lft2B. Bleaney and M. R. Wells,Joint ISMAR-AMPERE Conference, Belft1980.

llt3B. Bleaney, A. C. de Oliveira,and M. R. Wells, Joint ISMAR-AMPEREConference, Delft 1980.

^"M. Bose, S. Ganguli, and M.Bhattacharya, Phys. Rev. B_19, 5535(1979).

lfc5B. Bleaney, F. N. H. Robinson,and M. R. Wells, Proc. Roy. Soc.


A362, 179 (1978).li>6R. B. Creel and E. von Meerwall,

J. Chem. Phys. 73, 1538 (1980).147A. C. Gossard, Phys. Rev. 149,

246 (1966).lk7ak. V. Skripov, A. P. Stepanov,

V. A. Marchenko, V. M. Pan, and A. D.Shevchenko, Zh. Eksp. Teor. Fiz. 77 >2313 (1979); Sov. Phys. 50, 1113(1980).

1 4 8P. K. Burkert, M. F. Eckel, andH.-P. Fritz, Magn. Reson. Related

Phenomena, Proc. Congr. AMPERE, 17th(1972), p. 263.

l l l 9 A. Al lerhand, J . Chem. Phys. 5_23

2162 (1970).15°K. Paulsen and D. Rehder, Z.

Naturforsch. A3_7, 139 (1982).15lA. Abragam, The Principles of

Nuclear Magnetism, Oxford UniversityPress, Oxford, 1961, p. 348.

1 5 2P. A. Casabella, J. Chem. Phys.40, 149 (1964).

Vol. 4, No. 1/2 83

a survey of v nmr spectroscopy · review a survey of 51v nmr spectroscopy dieter rehder institut...

Documents