Application of one-dimensional nutation nuclear magnetic resonance to 51V in ferroelastic BiVO4

Application of one-dimensional nutation nuclear magnetic resonance to 51V in ferroelastic BiVO4

SOLIDSTATE Nuclear Magnetic Resonance Solid State Nuclear Magnetic Resonance 3 (1994) 231-236 Application of one-dimensional nutation nuclear magn...

390KB Sizes 0 Downloads 19 Views

SOLIDSTATE Nuclear Magnetic Resonance Solid State Nuclear

Magnetic

Resonance

3 (1994) 231-236

Application of one-dimensional nutation nuclear magnetic resonance to 51V in ferroelastic BiVO, Pascal P. Man *ya,S.H. Choh b, J. Fraissard a ’ Laboratoire de Chimie des Surfaces, CNRS URA 1428, Uniuersite’Pierre et Marie Curie, 4 Place Jussieu, Tour 55, 75252 Paris Cedex OS, France ’ Department of Physics, Korea Unillersity, Seoul 136 701, South Korea Received

20

April 1994; accepted 26 April 1994

Abstract

The central line intensity of a spin I = 7/2, excited by a radio frequency (rf) pulse, is calculated by taking into account the first-order quadrupolar interaction during excitation. Thus, the result is valid for any ratio of quadrupolar coupling to pulse amplitude. The quadrupolar coupling of the nuclei vanadium 51V in a single crystal of ferroelastic BiVO, is determined using this one-dimensional (1D) nutation method. Key words:

5’V nuclear magnetic resonance;

Spin-7/2;

One-dimensional

1. Introduction Recently, we extended the study of time domain response of half-integer quadrupolar spins to I = 7/2 111.Some nuclei. for examnle 43Ca 121. 45sc [3],‘49Ti-[4], 51V [5-loj, 59C~ [l,i1,12], 13G [13] and ‘39La [14-161, are widely used as nuclear magnetic resonance (NMR) probes for investigating local environments in inorganic compounds. The physical parameters (quadrupolar coupling constant e2qQ/h, asymmetry parameter 77) are mainly obtained from static or spinning powder lineshape analysis [2,4-161 when the second-order quadrupolar interaction A$’ dominates. Sidebands of satellite transitions, if detected, are also

* Corresponding

author.

0926-2040/94/$07.00 0 1994 Elsevier SSDI 0926-2040(94)00017-7

Science

nutation;

Ferroelastic

BiVO,

used [6]. However, if this is not the case, a nutation experiment (1D or 2D) is required [1,3,5,8, 11,17-231. The 1D nutation experiment consists simply of acquiring a series of free induction decays obtained with increasing pulse length t. The fit of the line intensities with an analytical expression allows us to determine the quadrupolar coupling oo in a single crystal, or e’qQ/h and 77in a polycrystalline sample. Two systematic studies of half-integer spins, up to I = 9/2, undertaken by Kentgens et al. [3] and Samoson and Lippmaa [17], have given the principal trend. However, the main effects of the first-order quadrupolar interaction 28) on the line intensity have been analyzed in great detail for spin3/2 [19,21,23] and -5/2 [18,22] systems only. In this work, we deal with the relative intensity of the central line by considering explicitly A?&“ during the excitation of the spin-7/2 system. We

B.V. All rights reserved

232

P.P. Man et al. /Solid State Nuclear Magnetic Resonance 3 (1994) 231-236

apply the 1D nutation method using the nuclei vanadium 5’V in a single crystal of ferroelastic

/’

, I-7/2>

BNO, .

I8>

w. i 60,

I

t

2. Relative central line intensity of spin I = 7/2 The static magnetic field B, is supposed to be strong enough so that A?‘$‘, which is inversely proportional to B,,, becomes negligible compared with z&r’, which is independent of B,. In this condition, the dynamics of a spin I = 7/2 system, excited by an x pulse (Fig. l), is described by the density matrix p(t) expressed in the rotating frame associated with the central transition: p(t) = exp( -i2Fa)t)Iz

exp(iA?E’a’t)

I

;;, l-3/2>

16> -

t 4

00 +20,

I wo

b’

(I)

where &oCa)= -w,fZx + z&r’

A?($’= w

-

I\

I\ I_ \15/2>

;oQ[3zj - I( I + l)] 3e*qQ

Q - 81(21-

1)h

17/2>

X [3 (30s~p - 1 + n sin’ p cos 2a]

orfIx

Fig. 2. Energy levels, their shifts and the two forms of eigenstates of a spin-7/2 Wz means Zeeman interaction).

which are related by 0 = T-‘MT, Eq. 1 can be rewritten as:

M,

p(t) = T exp( -iflt)T-‘p(O)T

are found,

exp(iW)T-’ (3)

Knowledge of the density matrix enables us to determine the relative intensity of the central line F4g5(t) (the superscripts 4 and 5 indicate the central transition): F4s5( t)

-ELv-4

I l>

(2)

The Euler angles (Y and p orientate B, in the principal axis system of the electric field gradient tensor. The eigenstates of I,, I m) (Fig. 2), are redefinedas:Ii)=IZ-m+l),soi=1,...,21+ 1. In particular, the energy levels 14) and 15) are those of the central transition. The matrix representation M of A?@),expressed in the eigenstates of Zz, is not diagonal. When the matrices of eigenvalues L? and eigenvectors T associated with

3&y-

l2>

I I

time

Fig. 1. Hamiltonians associated with the one-pulse excitation.

= -$ Tr[ p( t)4Z,?5] =

& c ,$vm+Xj_(7xm+y_+ m-l

5ym+zj_

J-1

+32,+7_+

Vnl+Xj_) sin(w,+-wj_)t

(4)

P.P. Man et al./Solid

The terms X, +, Y, +, Z, *, V, +, w, + and their associated parameters are defined in Table 1. The relative line intensity F4,5(t) is the sum of sixteen sine curves of different amplitudes and frequencies. Fig. 3 represents F4.5(t) LWSUSthe pulse length t for three values of wo with w,/27r = 50 kHz, a typical magnitude provided by modern spectrometers. The maximum of F4.5(t) decreases, as well as the associated pulse length, when wo increases from a small value to a large one, but both reach a limiting value which is one quarter of those when wo G 0. In other words, when wo B mrf, the magnetization of the central transition precesses around the radiofrequency magnetic field four times as fast as the opposite case (ho < w,r). A loss of line intensity by a factor of 4 occurs. There appears also a linear region, defined by t < 0.5 ps, where the

0

2

4

-p’,

cos p,=

RI += i

+ Qr,]“’

Pi cos - 3

8

10

Pulse length t (ps)

line intensity is proportional to r and independent of ho. This linear region is therefore available for a distribution of wo. This occurs in a

9 f = 32&(8~~

+ F - 4)

+36p+r++549”, J/Z (~2, + 12r,)-

p,+[p”,

6

Fig. 3. Relative intensity of the central line F?t) L’ersus the pulse length t for three values of wQ/2rr: (a) 0 kHz; (b) = 50 kHz. 50 kHz; (c) 1 MHz. w,/Zr

Table 1 Parameters defined for an x pulse (m = 1, 2. 3. or 4, and E = wo/wrf) p *= 8w;~(21~’ T Se + 5) r k= lbw~~(lOS~’ T 74~” + 59~’ T 18~ + 9)

233

State Nuclear Magnetic Resonance 3 (1994) 231-236

“?

6 1 S+=[;(P~-~)-R;~]“’

w2+= &$ w.,+= k:-

1 + $R,,-D,) ;CR,i+SiJ

P.P. Man et al. /Solid State Nuclear Magnetic Resonance 3 (1994) 231-236

234

powdered sample. This excitation condition must be applied in order to obtain quantitative determination of spin populations in powdered compounds. A11 these properties have already been proven for spins I = 3/2 [19] and 5/2 [l&22], and generalized for any integer or half-integer spin 124-261.

3. Experimental We checked our theoretical result with a single crystal of ferroelastic BNO,. It has the shape of a quarter of a disk whose radius and thickness are 7 and 3 mm, respectively. The quadrupolar coupling of “V in this experiment, which is defined by half the frequency separating two consecutive lines in the spectrum of the single crystal (see Fig. 21, was oo/2a = 27.8 kHz. The static “V NMR spectra, acquired with one-pulse excitation, were obtained with a Bruker ASX-300 multinuclear spectrometer operating at

78.897 MHz. The high-power static probehead was equipped with the standard (10 mm diameter and 30 mm length) horizontal solenoid coil. The amplitude of the pulse, determined using an aqueous solution of NaVO, contained in a glass tube having the same size as the solenoid coil, was 0,/27r = 27.77 kHz corresponding to a 7r/2 pulse length of 9 ps. Each spectrum was obtained with a recycle delay of 20 s, eight scans, a spectral width of 2.5 MHz and a dead time of 8 pus. The pulse length was increased from 1 to 10 ps by 0.5~ps steps. 51V Spectra on the absolute intensity scale, obtained with increasing pulse length, are represented in Fig. 4. Due to the large spectral width, only the central line in all the spectra was properly phased. In Fig. 5, the curve corresponds to a fit of the experimental integrated line intensities (full circle) with Eq. 4. The simplex fit procedure was used. The results for the quadrupolar coupling and the amplitude of pulse are 28.6 and 35.4 kHz, respectively. The value of the

9P

I

I~~~~~~~~~I.~~~~III~,.~~~~~~~~l~~~..IlII,I~~~~~~~~’

2000

1000

-1000

-Z&Xl

CPA Fig. 4. 5’V NMR spectra from bottom to top.

of a single crystal

of BiVO,

for increasing

pulse lengths

t from 1 to 10 KS by 0.5~s

steps;

t increases

P.P. Man et al. /Solid State Nuclear Magnetic Resonance 3 (1994) 231-236

235

static magnetic field, whereas lineshape analysis requires spectra obtained with several static magnetic fields of different strengths in order to introduce the second-order quadrupolar effect on the spectra.

Acknowledgement We thank Dr. P. Tougne for his help with the NMR experiment.

0

2

4

6

8

10

Pulse length (ps) Fig. 5. 51V in central mQ /2a

Experimental (0) integrated central line intensities of a singIe crystal of BNO, from Fig. 4 and calculated line intensities F4%) with the following parameters: = 28.6 kHz and o,r /2~r = 35.4 kHz.

quadrupolar coupling is in good agreement with that determined from line splitting. As already observed [1,27], the amplitude of the pulse is higher than that of the experimental one: 35.4 instead of 27.77 kHz. This may be due to the fact that the size of the crystal is much smaller than that of the solenoid coil, and the difference of quality factor of the coil between the solid sample and the aqueous solution.

References [I] P.P. Man and P. Tougne, Mol. Phys., in press. [2] A. Trokiner, L. Le Not, A. Yakubovskii, K.N. Mykhalyov and S.V. Verkhovskii, Z. Naturforsch., 49a (1994) 373. 131A.P.M. Kentgens, J.J.M. Lemmens, F.M.M. Geurts and W.S. Veeman, J. Magn. Reson., 71 (1987) 62. [4] SF. Dee, M.F. Davis, G.E. Maciel, C.E. Bronnimann, J.J. Fitzgerald and S.S. Han, Inorg. Chem., 32 (1993) 955. [S] H. Eckert and I.E. Wachs, J. Phys. Chem., 93 (1989) 6796. [6] J. Skibsted, N.C. Nielsen, H. Bildsoe and H.J. Jakobsen. Chem. Phys. Lett., 188 (1992) 405. [7] O.B. Lapina, V.M. Mastikhin, A.A. Shubin, V.N. Krasilnikov and K.I. Zamaraev, Prog. Nucl. Magn. Reson. Spectros., 24 (1992) 457. [S] M.L. Occelli, R.S. Maxwell and H. Eckert, .I. Cutal., 137

(19921 36. [9] T.H. Yeom, S.H. Choh, K.J. Song and M.S. Jang. J. Phys.: Condens. Matter. 6 (1994) 383.

4. Conclusion We have shown, in the particular case of a spin-7/2 system, that the quadrupolar coupling in a single crystal can be determined also from the dependence of the central line intensity on the excitation pulse length. More generally, this 1D nutation method represents an alternative way for determining the quadrupolar parameters in a polycrystalline sample if the lineshape analysis fails. This happens when the lineshape is dominated by the first-order quadrupolar interaction and/or the chemical shift anisotropy. This situation occurs in high static magnetic field measurement. This 1D nutation method requires only one

[lo] J. Davis, D. Tinet, J.J. Fripiat, J.M. Amarilla, B. Casal and E. Ruiz-Hitzky, J. Mater. Rex, 6 (1991) 393. [ll] T. Eguchi, H. Nakayama, H. Ohki, S. Takeda, N. Nakamura. S. Kernaghan and B.T. Heaton. J. Orgunometal. Chem., 428 (1992) 207.

1121J. Hirschinger, P. Granger and J. Rose, J. Phys. Chem., 96 (1992) 4815. [13] W.P. Power, R.E. Wasylishen, S. Mooibroek, B.A. Pettitt and W. Danchura, J. Phys. Chem., 94 (1990) 591. [14] R. Dupree, M.H. Lewis and M.E. Smith, J. Am. Chem. Sot., 111 (1989) 5125. 1151T.J. Bastow, Solid State Nucl. Magn. Reson.. 3 (1994) 17. [161 B. Herreros, P.P. Man, J.M. Manoli and J. Fraissard, J. Chem. Sot., Chem. Commun., (1992) 464. [17] A. Samoson and E. Lippmaa, J. Magn. Reson.. 79 (1988) 255. El81 J.A.M. Van Der Mijden, R. Janssen and W.S. Veeman. Mol. Phys., 69 (1990) 53. [19] P.P. Man, J. Magn. Reson., 67 (1986) 78.

P.P. Man et al. /Solid State Nuclear Magnetic Resonance 3 (1994) 231-236

236

[20] A.P.M. Kentgens, J. Magn. Resort. Ser. A, 104 (1993) 302. [21] L. Pandey, S. Towta and D.G. Hughes, J. Chem. Phys.,

85 (1986) 6923. [22] P.P. Man, Mol. Phys., 78 (1993) 307. [23] R. Janssen and W.S. Veeman, J. Chem. Sot. Faraday Trans. I, 84 (1988) 3747.

[24] J. Haase and E. Oldfield, J: Magn. Reson., Ser A, 104 (1993) 1. [25] P.P. Man, A&. Magn. Reson., 4 (1993) 65. [26] P.P. Man, J. Klinowski, A. Trokiner, H. Zanni and P. Papon, Chem. Phys. Lett., 151 (1988) 143. [27] P.P. Man, Solid State Nucl. Magn. Reson., 2 (1993) 165.