HIGH RESOLUTION NMR OF 29Si IN AMORPHOUS HYDROGENATED SILICON

(1)

HAL Id: jpa-00220811

https://hal.archives-ouvertes.fr/jpa-00220811

Submitted on 1 Jan 1981

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

HIGH RESOLUTION NMR OF 29Si IN AMORPHOUS HYDROGENATED SILICON

B. Lamotte, A. Rousseau, A. Chenevas-Paule

To cite this version:

B. Lamotte, A. Rousseau, A. Chenevas-Paule. HIGH RESOLUTION NMR OF 29Si IN AMOR-

PHOUS HYDROGENATED SILICON. Journal de Physique Colloques, 1981, 42 (C4), pp.C4-839-C4-

841. �10.1051/jphyscol:19814185�. �jpa-00220811�

(2)

JOURNAL DE PHYSIQUE

ColZoque C4, suppZ6ment au nOIO, Tome 4 2 , octobre 1981 page C4-839

B. Lamotte, A. Rousseau and A. Chenevas-Paule

Section de Re'sonance Magrze'tique and LETI-NCE, CENG, 85 X, 38041 GrenobZe Cedex, France

Abstract.- The new techniques of High Resolution NMR in solids by cross-polarization with protons and magic angles rotation of the sample have been applied to the study of amorphous hydrogenated Silicon in order to contribute to the charac- terization of this material. This method gives the possibility to distingyish (by their "chemical shifts") the different kinds of environments of each Si nu- cleus, in relation with the bonds in which these atoms are involved.

1- Introduction

In spite of the fact that amorphous hydrogenated silicon raises considerable _.- . - interest for solar photovoltaic applications and for device applications and that it has been very much studied in the last years, its structure is still not well known.

Therefore, new methods aiming to characterize this material have much interest.

We have applied here the new techniques of High Resolution NMR in solids which re originally introduced by A. Pines, M.G. Gibby and J. Waugh

[I]

for the study of

''

C in natural abundance in molecular solids. This NMR method which combines in its pulse sequence cross polarization of the "rare" spins - those spins which are observed

-

with protons, proton spin-decoupling and rapid rotation of the sarfgle at t magic angle, has roved very successful in the last years, mainly for C and

''

^P

nuclei, and the 2g Si isotope is a good candidate because of its spin of 142, its natural abundance of 4.7 %, and its magnetic moment of 20 % less than of C.

Our aim in applying this new method to the study of amorphous hydrogenated silicon is to try to fharacty~ize this material in the same way as a chemist identi- fies a molecule by H or C NMR in solution. He will use the chemical shift information to characterize the different chemical groups in the molecule. In the same way, we will try here to characterize the different kinds of bonding and elec- tronic environments of silicon :

by their chemical shifts.

The pulse sequence used f o ~ ~ t h e s e experiments

[I]

is a double irradiation method where both the proton and Si spin systems are excited. First the pr ons are polarized. Their polarization is immedialy transfered to the2gpins of the

''

^Si

nuclei. In a third period, the free precession signal of the Si nuclei is detecr ted and the sum Fourier transformed to give the spectrum. Moreover, the anisotropy of the chemical shifts r e p r e s e n t i n ~ a c ~ m p l i c a t i n ~ factor in such complex compounds is removed by rapid rotation of the sample - around 3000 Hz - atVthe magic angle".

2- Samples

Several different samples prepared by reactive sputtering in an Ar-Hz mixture

Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:19814185

(3)

C4-840 JOURNAL DE PHYSIQUE

and deposited at different temperatures at the speed of about 1 p per hour have been studied. F i l m s of 20 to 40 p were formed either on Aluminium or Molybdenum foils, giving samples of 100 to 400 mg. In the first case, Aluminium was dissolved by di- luted hydrochloric acid. In the second case, the samples were recovered by folding the Molybdenum foil over and over.

3- Results : 29 Si High Resolution NMR spectra Spectra of two kinds were obtained :

1) spectra showing a partial resolution, such as in Fig 1 .

Fig I

2) spectra not resolved at first glance.

The lines obtained in these spectra remained relatively broad because of two factors a) due to the amorphous nature of the material there exists a continu- ous distribution of chemical shifts for each given kind of bonded silicon.

b) there exists a superposition of several peaks which can be clearly seen in the spectrum of Fig 1.

This spectrum can be identified as the sum of five different peaks centered at respectively :-5?p.p.m./T.M.g (T.M.S. is Tetramethylsilane, the zero reference of chemical shif~s in NMR of Si)

-

which is a broad peak

-,

- 69 p.p.m./T.M.S.,

-

83 p.p.m./T.M.S., - 99 p.p.m./T.M.S. and - 1 1 1 p.p.m./~.M.S.

A theoretical reconstruction of this spectrum with gaussian lines can be made which fits quite satisfactorily with the experimental spectrum.

Moreover, by applying techniques of resolution enhancement by convolution of the free precession signals with a gaussian shape (a technique commonly used by NMR spectroscopists of proteins) it was possible to see that the spectra not resbl- ved at first glance contained the same peaks as those observed in the sample shown in Fig 1 , but with different intensities, the peaks at - 99 p.p.m and - 1 1 1 p.p.m.

being much weaker in these spectra.

4- Identification of the main peaks

The broad peak at

-

⁵⁷p.p.m. could be identified to Silicon bonded to four other Silicon, by comparison with the chemical shifts measured for crystalline and also for amorphous not hydrogenated silicon.

A first possibility for identifying the remaining peaks was to compare with 29 Si chemical shifts of parent molecules measured in sblutioni :

i 6 =

-

⁹⁶p.p.m.1 T.M.S. _[2]

(4)

Such a comparison suggests that the peaks at

-

99 p.p.m. and

-

111 p.p.m.

would respectively be attributed to Si H and Si H bondings.

However, using a variation of the NMR pulse sequence giving the possibility to de- 2 termine silicon atoms directly bonded to protons from others, it was possible to prove unambiguously that the peaks at

-

69 ppm and

-

⁸³ppm corresponded to Silicon atoms directly bonded to protons, these peaks being identified respectively with Si H and Si H, bonds.

We think-that the peaks at

-

99 ppm and

-

111 ppm observed in the spectrum of the sample shown in Fig 1 can correspond to the result of an oxidisation process of this finely ground sample by moist air, in relation with the existence of Si-OH bonds. We will try soon to verify this hypothesis.

References :

[I-) A. Pines, M.G. Gibby and J. Waugh J. Chem. Phys.,

2,

569 (1973)

[2) "Silicon 29 NMR Spectroscopy" by E.A. Williams and J.D. Cargioli in Annual Reports on NMR Spectroscopy Vol 9 (1979)