2 Ferruginous rocks of Pagala (Togo) and Tin Akoff (Burkina Faso)

2.6 Geochemical study

The firm presence and association of different trace elements obviously give gossans their well known originality in comparison with other ferruginous rocks (Kosakevitch, 1979; Ryall and Taylor, 1981; Blot and Magat, 1990; Scott et al., 2001; Blot, 2002, 2004 …).

 

The factors governing the abundance and associations of these elements have endogeneous and exogeneous causes, the respective part of which is not known and not easy to evaluate. It can be noted that present-day climatic conditions should permit a climatic discrimination in the removal of elements i.e. more ancient leaching at Tin Akoff than at Pagala where it is presently stronger.

 

At Pagala, the most spectacular anomalies have concerned Zn, Ba and Mn, reaching and going beyond 1%, i.e a similar content as for some major elements. However the polymetallic aspect has been highlighted in detailed studies and in a more synthetic way (Blot and Magat, 1989).

 

The geochemical study of Togbé does not emphasize any difference between the different facies but points out two distinct geochemical parageneses around Zn (with Mn and HRE) and Ba (with P and Cu) surprisingly gathered as originating from a same protore.

 

After a statistic geochemical study (hierarchic ascending classification), Deschamps et al., hold 12 families of gossans among which the ones testify to massive sulfidic mineralizations and the others to disseminate or of Fe-carbonated origin mineralizations. The highest contents concern Zn, Ba, Pb (Cu, As).

 

BHP, Coronation and Ravensthorpe Inc. left little information about their work; however, facies of gossans developed from more or less sulfidic mineralization and facies generated from the zinciferous sideritite were defined at that time.

 

Mikhaïloff (2000) had doubt about the conclusiveness of the denomination of gossans, but without giving fair interpretation to these supergene markers of mineralization.

 

Theron (2002) discriminated between pyritic, zinciferous or lead-bearing gossans and others originating from oxides of BIF type and from remobilizations (“wrong” gossans). The place of lead is rather ill-defined, this element being considered as exotic in order to account for supergene behaviour different from that of zinc.

 

At Tin Akoff, the contents of trace elements do never reach 1%, but the variety of anomalies was emphasized, whilst in the leaflet of the synthetic map of Burkina Faso on 1/1,000 000, Castaing et al. (2003) only mention the Zn anomalies.

 

Analyses of gossans are always difficult to compare as well because of the large variability of mineral compositions and because of unsufficient knowledge on alterations and alteration rate. Moreover, analytical techniques may vary according to the laboratories and in the course of time. The variations on similar objects are consequently very important and the images provided by different studies for a unique site are often not easily comparable.

 

For Pagala, the author’s own results are available and complemented by analyses performed by Togbé in the same laboratories (ULP, Strasbourg and BRNM Togo for As and Pb) and subsequently by successive mining operators (BRGM Orléans, France, Anglo American, ACME Vancouver). The element populations in Pagala’s gossans are really identical only for several elements, by comparing average values: Mn, Pb and (As, Pb) are equivalent as considered in the case of ORSTOM, BRGM, Anglo-american. Parallaly very important differences are pointed out in major elements and in trace elements (table 19). Apart from these two observations, it was decided to present comparison between the elements analyzed by Anglo American because more completed and better appropriate for Pagala and the elements analyzed by IRD at Tin Akoff. So, 52 elements were compared in both sites.

 

Table 19

Comparison between analytical results of some elements of the Pagala gossans, from three different laboratories

 

We present contents of non-constitutive elements according to two methods, comparison of average values and comparison of contents normalized to the Clarke. For comparison of the average values, the lack of homogeneous sampling is obviously no quantifiable factor. The comparison of contents’ distribution suffers because of the problem of proportioning’s limits according to the laboratories and epochs. The calculation of average values permits to get a geochemical picture of the two populations concerned, whilst contents’ distribution permits to specify the decisive features (table 20).

 

Table 20 

Principal elements of Pagala and Tin Akoff in ppm and ppb*,

1 = average, 2 = average of 10 % of the highest contents

 

Comparison of average values (test of average values’comparison)

At Pagala where weathering is rather juvenile, 19 elements are characterized by significantly higher average value, whilst at Tin Akoff only 12 of such elements are detected and 21 do not display any significative difference of the average values.

 

Comparison of the elemental content distribution

Geochemical background:

Two different situations can be discriminated from the mean elemental contents which really compose the geochemical background of the ferriferous outcrops. For a Clarke (average abundance of an element) higher than 1, the geochemical background is considered as high and this concerns about 20 elements. About 30 elements are characterized by average contents lower than the Clarke, especially the principal major elements except iron and manganese, since ferruginous rocks are of weathering origin. However, some elements with large deficit on the population scale and less concerned by removal of major elements can display important significative anomalies (Ag, Hg, Ni, S and W at Tin Akoff, Pt, W and above all Ba at Pagala, where barytine occurs frequently).

 

Standard deficit (average < 1 Clarke)

  • B, Ba, Ce, Cr, Cs, Ga, Ge, Hf, In, La, Li, Nb, Rb, Sn, Sr, Ta, Th, Tl, W, Zr, Al, Ca, Mg Na, K, Ti

Standard geochemical background (average > 2 Clarkes)

  • As, Au, Be, Bi, Cd, Co, Cu, Fe, Mn, Mo, Pb, Pd, Re, Sb, Se, Te, U, Y, Zn

 

High contents

 

The main anomalies relevant to the Clarke correspond to the components of the geochemical background but also to regionally less represented elements. Here, the definition of anomaly is an average calculated from 10% of the highest contents superior to 2 and 5 times the Clarke value. 27 elements display their mean highest contents superior to 2 Clarkes. It can also be noted that for about fifteen elements the content is higher than that of iron which is by definition a residual element.

 

Standard high contents (average of 10% of the analyses > 2 Clarkes)

  • Ag, As, Au, Be, Bi, Cd, Co, Cu, Fe, Hg, Mn, Mo, Ni, P, Pb, Pd, Pt, Re, S, Sb, Se, Te, U, W, Y, Zn

Standard anomalies (average of 10% of the analyses > 5 Clarkes)

  • As, Au, Bi, Cd, Co, Fe, Mn, Mo, Pb, Pd, Re, Sb, Se, Te, U, Zn

Specific high anomalies of Tin Akoff

  • Be, Pt, V

Specific high anomalies of Pagala

  • Ag, Ba, Hg, P, S