Clay Minerals (1985) 20, 367-387

NICKEL-BEARING CLAY MINERALS: I. O P T I C A L S P E C T R O S C O P I C STUDY OF NICKEL CRYSTAL CHEMISTRY A. M A N C E A U ,

G. C A L A S

AND A. D E C A R R E A U *

Laboratoire de MinJralogie-Cristallographie, UA CNRS 09, Universitds Paris 6 et 7, 4 Place Jussieu. 75230Paris Cedex 05, and *Laboratoire de Gdochimie des Roches S~dimentaires, UA CNRS 0723, Bat. 504, Universit~ de Paris-Sud, 91405 Orsay Cedex, France (Received 25 February 1985) A B S T R A C T: Nickel crystal chemistry was systematically studied in various phyllosilicates,

mainly the natural phases selected from the 'garnierites' of the supergene ore deposits of New Caledonia. Minerals which do not usually occur in New Caledonian parageneses were synthe sised, as they could represent intermediate phases of genetic importance. In the kerolitepimelite series, a linear relationship occurred between the ratio I(13,20~/I(02,11) of the hk bands and Ni-content. Diffuse reflectance spectra were used to derive the crystal chemical parameters of Ni. These confirmed its divalent character and its occupation of octahedral sites; the resulting structural distortion was slight and could not be detected in some minerals. There was no optical evidence for Ni atoms in 4-fold coordination. The two main parameters which showed significant variations among the studied phases were site distortion and crystal field stabilization energy (CFSE). Site distortion was at a maximum in trioctahedral smectites and sepiolite. CFSE depended on the mineralogy, crystallinity and chemical composition (Al-content) of the phase. Finally, clay minerals are classified according to the increasing stability of Ni in the octahedral sheet, which has been tentatively related to the geochemical distribution of this element. Secondary minerals are usually enriched vs. primary ones and among them are nepouite and kerolite which exhibit a high CFSE in contrast to sepiolite.

Although the green-coloured 'garnierites' have been studied for many years (Faust, 1966), their chemical and structural characteristics are not entirely clear. An accurate study of these hydrous Ni-Mg silicates requires investigations at various microscopic scales, from the study of the relations between the various layers to the determination of the precise nature of the sites occupied by Ni atoms. In this paper we present some results quantifying Ni crystal chemistry, which have been derived from the study of diffuse reflectance spectra of various phyllosilicates. Special emphasis is laid on parameters such as coordination number, site distortion, nickel-oxygen bond covalency and crystal field stabilization energy (CFSE). These results should promote a better understanding of Ni geochemistry during the formation of supergene ore deposits. A subsequent paper will be devoted to the intracrystalline distribution of Ni studied by X-ray absorption spectroscopy in the same phases. EXPERIMENTAL

Pure phases were selected on the basis of X-ray diffraction (XRD) and wet chemical analyses. The ignition losses of minerals previously heated at 110~ (H20 +) were 9 1985 The Mineralogical Society

A Manceau et al.

368

determined for the kerolite-pimelite series. A Philips EM300 Transmission Electron Microscope (TEM) was used with a X-ray dispersive energy spectrometer (XDS). Diffuse reflectance spectra were obtained with a Cary 17D spectrophotometer and an integrating sphere attachment. They were recorded by taking BaSO 4 as a reference compound. Reflectance spectra, R = Rsample/Rreference, were collected using a PDP 11/04 computer interfaced with the spectrophotometer by averaging ten measurements at 1.5 nm intervals between 1700 and 325 nm. CHARACTERIZATION

OF THE SAMPLES

The samples were mainly collected in New Caledonia whose nickel ore deposits have been the subject of several studies (Trescases, 1975; Troly et al., 1979; Pelletier, 1983).

Kerolite-pimelite series Chemical analyses. The chemical analyses and structural formulae are reported in Table 1. They are in fair agreement with those reported in the literature for the kerolite/pimelite Ni Ni+Mg 1-

P2

0.75--

P1

0.50-

K6 K5

L8

- 4 . ~

K3

0.25"

K4 L7

__L~_K,~_2

GM

K1 i

2

I

3

I

I

4

5

1(13.20)/I (02.11 .~ FIG. 1. Plot of Ni content vs. 13,20/02,11 intensity ratio in the kerolite-pimelite series (GM: sample from Goles mountain, Yugoslavia (Brindleyet al., 1977)) and in lizardites.

369

Ni crystal chemistry in clay minerals

z

~

~

~ _

~ 6 6 6 6

~6

6~

e~

ooo o io~176 I i ~

q io~

~II~ I~ c~

6

=~ o

r

o,I .~-

~4 g~

~

o

E

~11

4

0

0

4 I J4 0",

~

co

Z

66..~

~

Z.~

~

N

dNm'o

Z

~

o

~

.~ . . .~. ~.~. ~~, o,-, , , 6

~m

I I.~~

0

,.,4 J 6 , , 4

~>

[

I ~6

Z~

~4 ~4

~'

.'a

r

2

.-d.-d-..d rA ~

~mZZZ:~Z

• 0

0

0

ZZ>~z [-

0

A Manteau et al.

370

series (Brindley et al., 1979; Gerard & Herbillon, 1983). The deviations from ideal composition (tetrahedral atoms