© 1996-1999 NATI Research JSC.

  home.gif (4269 bytes) obj_list.gif (4250 bytes)        Metrology   PPM-mineralogy

This part is interesting for mineralogists mainly. The part is important in case when researcher has to decide what the order of principle steps are required for getting a reliable result and for an opportunity of evaluation of received data completeness. The section will be not so necessary for other readers and we recommend them to go for results of study on different geological objects. (Go to Object List).

   Authors

Go Top    Metrological maintenance for mineral processing.
            High sensitive technology for precious minerals study:
            "PPM-mineralogy"

It is easy to denominate at least four reasons why the exact and complete information on the mineral forms of noble metals in rocks, ores and technological products of industrial ores processing has primary importance.

1. In spite of the fact that the concentration of noble metals in rocks is frequently lower than 0.5ppm, the information on the phase forms of noble metals allows to solve a lot of theoretical and genetic geological problem.

2.The information on phase forms of noble metals in rocks allows to establish complete mineral paragenesis accompanying mineralization which in turn can be used as direct search indicators for regional forecast valuation and for ore bodies revealing.

3. The information on phase forms of noble metals in ores is a necessary element for development or choice benefication technology and subsequent metal extraction at planning and economic estimation of ore processing factories.

4. The information on phase forms of noble metals in technological products of ore processing factories allows to provide a monitoring of noble minerals on the whole of technological process, to control and make fast technology adaptation at variations in initial ores composition, that results in increasing of useful components extraction.

However, it is necessary to admit that before the present time there were no rational, convenient and express technologies of such kind of information receiving. It is explained by complexity of the subject of investigation: the concentration of noble metals (NM) in rocks usually are lower 0.5ppm, and the grains size of NM mineral-bearers is at first tens or less than ten micrometer. Therefore, the finding even of individual grains of minerals NM is usually considered as doubtless researcher Luck. However these data without quantitative estimation of completeness are not a representative and, certainly, they are insufficient for solving above geological and especially technological problems on ore processing factories.

It is obvious, that metrological maintenance should based any technology of useful information getting. This maintenance has to provide at least evaluation of received information completeness, because in opposite case, when the evaluations are impossible, investigation technology is not well correct and all heuristic finds should be refer to researcher Luck only.

 

Go Top     Elements of metrology

Information on the phase forms of NM is a qualitative and quantitative information on physical properties (size, shape, weight, volume, chemical composition etc.) of mineral grains extracted from rock or ore, therefore completeness of this information is directly proportional to amount of the founded grains.

Following to definition of the term "mass concentration" (relation of weight of a component in a mixture to weight of a mixture) for rock with concentration of some chemical element C at localization of this element in k various minerals the next expression will be right:

          SUM(n' m D V)i
C = --------------------------- ,   where i = 1 ... k
                    H

or, for single mineral-concentrator of this element (i.e. when k=1):

           n' m D V
C = ------------------ ,                                 [ 1 ]
               H

where the expression in numerator is equal to weight of a chemical element in n' grains of a mineral with density D and volume of each grain V, at concentration of an element in a mineral m, and denominator equal to whole sample weight H.
The parameter n' is a phase equivalent of a concentration (PEC), that corresponds to amount of grains with fixed size, containing all concentration of a element in a sample C. It convenient for practical estimations to turn from volume V to linear mineral sizes i.e. to approximate the grain shape by cube (V=a3) or sphere (V=4/3 ÏR3). However it is necessary to keep in a mind that accuracy is lost during approximation and the choice of the form of approximation will be defined by working problem. It is convenient also to enter parameter N = 1/Da3, which has simple physical sense: amount of mineral grains at density D of the cubic shape with the size of cube edge a, contained in batch (sample weight) = 1g. It is possible to fasten accounts and estimations by substituting N in the above formulas, addressing to the diagram:


eteh_gr1.gif (9864 bytes)

The above formula [1] have strict mathematical sense, however, going to estimations of real technologies, where grains of NM minerals are extracting from some rock volume and are studying later, it is necessary to take into account that in practice the part, but not all of grains is always extracted only. Therefore it is useful to define a concept of extraction factor (I) connecting amount of the really extracted grains (n) with amount of theoretically presented grains (n'):

I = n/n'.

Then, at approximation of the grains shape by cube, formula [1] becomes:

           n m D a3
C = -----------------
            H I

It is easy to recognize using the formulas what characteristics of studied rock will important significantly at a choice of technology of mineralogical works. For example, when you study 50g of rock sample with density at 2.8g/cm3 at the average grain size 50µm and with gold concentration at 0.1ppm (typical sample from black shale complex), it is easy to calculate that total amount of this size grains in the rock batch will make 143 million, and only 2 grains will presented by gold (PEC = 2). Even if all gold at this concentration will be located in one "huge" grain (PEC = 1), its size will reach 65µm (size of an cube edge) and investigation of larger granulometric fraction in this sample is mindless.

These data let us to understand a degree of complexity of a working problem before the study will started and to choose such technology of material treatment according of parameter I, which will lead to reliable result, independently of Fortune.

We have to note that PGM study when the element concentration in samples are lower, for instance, than 0.1ppm, fraught for researcher with some unusual and unpleasant surprises connected with sample contamination by an external material. It is easy to convinced using the formulas, that in rather small sample batch (n*100g) the low element contents will be concentrated in fine mineral grains (n - n*10µm). Amount of these grains will not exceed first tens, while total amount of this size grains in the sample batch will reaches up to 1012. It is easy to guess what the requirements are demanded to the equipment and labs for investigations. We are sure, that, mastering for the first time technique of high-sensitive mineralogical works, the researcher even once will ask himself in despair: " It’s impossible! What is it? Whence?", i.e. one of these endless rhetorical questions like "Were the dust from?", "Where do money disappear?"...

 

Go Top     "PPM-mineralogy" (Parts Per Million - mineralogy) -
            technology of high-sensibility mineralogical investigations

Rocks disintegration
Granulometric separation
Gravitational separation
Chemical analysis of NM on graviconcentraters (GC)
Electron microscopy and optics
Present to patient reader


The difference of any technology from a complex of separate methods and techniques consists in that all elements of technology are incorporated by the uniform purpose and each separate element corresponds to general requirements.

The main purpose of proposed technology of ppm-mineralogy is investigation of "heavy" minerals (including NM minerals) with mineralogical sensibility from 0.01- 0.001ppm. For this purpose following requirements are placed upon elements of technology:

1. All elements of technological process should be very sensitive to fine mineral phases of size at n*10µm and for some purposes - n*1µm;

2. All parts of technology and especially those connected with extraction and concentration of "heavy" phases should provide extraction even single "heavy" grains from the large number (billions) grains of disintegrated rock. It means that a reduction coefficient (RC) of initial sample should reaches up to hundreds of thousand or millions at acceptable parameter I;

3. All parts of technological process should exclude an uncontrolled sample contamination even by single grains of external "heavy" phases;

4. The technology as a whole should provides reproducible results and to be accompanied by presenting of reliably interpreted analogue and digital data.

The direct identification of noble metal phases (NM) in massive rock fragments by mineragraphic or microprobe techniques is corresponds to above requirements. However, for the example with black shale sample given in the previous section it means, that for revealing of native gold with the size 50µm at Au concentration in rock 0.1ppm, the area of one polished section should reaches up 1786cm2 (42 on 42 cm) that the unique gold grain (PEC = 1) has appeared on a surface. This is fair only when I = 1, that hardly it is possible to guarantee at that polished section size. It is obvious, that the given technology (collection of large monolithic samples, preparation polished section with size at 42 on 42cm, proper mineragraphic and microprobe studies) has only theoretical interest.

Technologies, providing rock crushing (milling) up to wanted size fraction and concentrating of the grains containing ore component, accompanying by subsequent analysis of mineral concentrates, are a rather more efficient.
Main elements of technology named "ppm-mineralogy" are given below. We have no goal to give the detailed description of specific steps (i.e. to write "the user's guide") of high-sensitive mineralogical study. Our purpose is: based on our experience to point attention at those aspects, which appreciably influence on final results. This approach is proved because of all parts of technology are interconnected and moreover it is impossible to foresee in advance specific characteristics of the used equipment: each researcher will define that concretely based on metrological characteristics of the available equipment.

The main technological elements of the ppm-mineralogy are:

Rock disintegration

We have to note that it means rock phases individualization but not a rock crushing despite of term “crushing” we use further. It is important for subsequent parts of a technological process that the mineral grains were individualized maximally and had a physical properties of individual phases but not of average characteristics of physical properties, that is inherent for mineral intergrowths, to be produced during traditional crushing techniques.

There is no technique of rock disintegration providing complete phases individualization. However the best percentage of mineral grains extraction from massive rock with preservation of the primary mineral forms is achieved at shock-free disintegration techniques. Traditional crushing by using of jaw crusher and roll crusher with subsequent grinding on ball crusher or ring crusher is much less effective.

It was seems before that there was no alternative to electric pulse disintegration technique, however, subsequently we had to refused it for the following reasons: 1. The electric pulse discharge melts both material of sample and material of electrodes down, therefore sample (heavy graviconcentrates especially) it become oversaturated or consisting from balls or spherical drops with the sizes from 1 up to n*100µm, that strongly complicates the further material treatment and furthermore data interpretation. (See also: Agaphonov, Velinsky et al., "Electric pulse crushing as the probable source for geological misinformation", Doklady AC USSR, 1991, v. 318, N 6). 2. The useful information is lost due to unpredictability of specific characteristics of material melting. The possibility of metrological estimations of method is becomes problematic, that is unacceptable for many geological, and furthermore technological problems. 3. The balls solder to working volume of the chamber and contaminate subsequently samples for a long time. As has experience shown, the complete chamber cleaning is impossible. There are some more disadvantages of this method: 1. The complete and homogeneous disintegration of the whole rock sample is not achieved for many rock types, that requires in increasing of sample volume (considerably, reducing parameter I) and resulted in some uncertainty of metrological estimations; 2. This disintegration technique has not high efficiency, needs high voltage (up to 250KV) and is complicate in operation.

Using the disk mills with increased working gap and subsequent numerous sieving and removing of a fine material to overgrinding prevention is useful for typical problems of the "ppm-mineralogy". This method is practical, but the grains are not individualized all perfectly sometime and parameter I may be reduced.

The way of crushing is important in connection of overgrinding problem. Let's assume the investigation task as a two samples study: rock (ore) with mineral grain sizes 40–200µm and a crushed material for subsequent flotation with grains size less than 44µm. Assume also, that crushing - is the subsequent impacts to solid particles. First impact upon each of 100µm mineral grains will resulted in abundant of fine grains (less than 50µm) and several fragments of a primary grain, with size more than 50 and less than 100µm. The subsequent impacts upon fragments 50-100µm will resulted in subsequent increasing of fine grains amount, and maximum amount of grains will appear in finest granulometric fraction of crushed sample (Histogram A).


eteh_his.gif (4645 bytes)In case if the minerals in primary rock have grains size not 100µm, but at 200µm, that more significant enrichment by the crushed grains will take place in finest granulometric class (histogram B). This is a typical example of a overgrinding when the largest minerals in the crushed rock will placed to the finest granulometric class. Besides of that, the only grains with natural size less than 50µm will (may) be safe but other minerals will situate in fine granulometric fractions as a grain fragments: so we lose additional information. Mineralogical works with these fine fractions (less than 20µm) are extremely complicate because of difficulty of grains extraction from fine-dispersed material: the equipment providing very high factors of initial material reduction (RC = n*104-106) at appropriate parameter I is strongly required. It means, that the material prepared for the flotation is not a simple object for mineralogical study, and a choice of mineralogical toolkit - serious problem.

Granulometric selection of crushed rock samples

The typical way for selection of disintegrated rock material on various granulometric fraction is a dry sieving. However, for achievement of maximal mineralogical sensibility it is necessary often to analyze granulometric classes less than 20mm, that requires using of wet sieving or classificators.

If the technique of granulometric fraction getting is more or less clear, the definition of upper and lower limits of granulometric fraction (sieve module) is a special problem. It is caused by granulometric selection like an element of a technological process is submitted to the certain requirements of the whole work technology. For example, let consider a case when fraction <250µm is investigated and of interest are "heavy" phases with densities more 6g/cm3. It is obvious, that gravitational separation of rock follows by granulometric selection, and it's final result is defined directly by a previous type of treatment. If the rock sample of class <250µm has undergone by gravitational separation, depending on RC the final result for the different size phases with densities >6g/cm3 may be next:

 

eteh_gr2.gif (4395 bytes)It is obvious, that the gravitational separation in fraction <250µm can much deforms a relationship between minerals with densities more 6g/cm3, if these minerals are represented by different grains size.
     To prevent possible sample contamination during granulometric selection, each sieve is cleaned separately in a ultrasonic bath during 15-30 minutes prior to the beginning of sieving and after sieving. The grains which have dropped out on bottom of a bath after sample sifting are combined with the appropriate fractions.
According to own experience each mineralogists will define with time what kind of sieve (from which Company-manufacturer) and what the ways are cleaned and what sieves noway are never cleaned!

Gravitational enrichment of "heavy" phases

Gravitational enrichment of phases with density more than 5-6g/cm3 with grain size at first or first tens µm and RC factor n*103 – n*106  for many reasons excludes an application of traditional enrichment techniques.

In technology of ppm-mineralogy the gravitational hydroseparators of original engineering are used (patented by "NATI" Research JSC.,1996, Russia). These hydroseparators allow to perform the enrichment of "heavy" phases in a continuous mode with RC factor up to 1000-5000 (I = 0.6 - 0.8), as well as in mode of single batch separation of a sample up to 500g with RC factor up to n*106 (I = 0.7 - 0.9).

The continous mode with RC factor up to 1000-5000 may be applied successfully in industry, but for the most of mineralogical tasks this RC value is insufficient: for mineralogical tasks it needs to have more high mineralogical sensitivity and more high value of I parameter.
The same context, wide advertising modification of NATI's hydroseparator (and failed, for our opinion), based on our old version of equipment 1996 (so called now the HS-01 by Rudashevsky) don't supported metrologically, don't able to work with appropriate sensitivity and RC parameter. This tool not applied to reasonable mineralogical investigation of material, but for heuristics guessing of results: the most mineralogical examples we discuss in this website - could not be obtained using insufficient machine: HS-01.

The degree of sample reduction of NATI's equipment can be adjusted for that the weight of "heavy" concentrate corresponds to the requirements of the subsequent stages of sample investigation. The mineralogical investigations are desired to have the minimal weight of a concentrate (first milligrams, that is maximal quality of a product in relation to ore component). At the subsequent determination of element chemical composition of a concentrate the sample weight should corresponds to requirements of pertinent analytical method.

If the metrological characteristics of applied gravitational separator can be determined by practical way and therefore this way of sample treatment could be characterized from metrological aspect, the material collection after gravitational separation represents "a thing in itself ". The point is that for a high-sensibility mineralogical investigations (for example, at C=0.05ppm and PEC=n*1, see equation [1]), the RC of an initial rock sample with weight up to 100g should reaches up to n*104-5. It means, that after gravitational enrichment the batch with weight at 0.001-0.005g will obtained, and the  grains of heavy minerals (HM) will have the size n*1 - n*10µm. The question is: HOW and WHERE to collect 3mg of grains at the size <20µm and to lose no more than 50 % of material and to have an opportunity to continue study of HM with the rest of graviconcentrate (GC)? Beginner mineralogists should find perfect answers for these problems beforehand the investigations.

In a context of the previous paragraph the requirement to separator are emerged: its frame should exclude sample contamination even by individual fine grains of an external rock or materials, and the material unloading should occur without essential losses.

Electromagnetic separation

Electromagnetic separation of granulometric or gravitational fractions is carried out by traditional techniques on the serial equipment at obligatory observance of cleanliness of all operations. We have to emphasize what application of electromagnetic separation during study of samples with platinum should performs carefully because of the native ferrous platinum has strong magnetic characteristics.

Chemical study of "heavy" concentrates (HC)

As gravitational separation increases concentration of "heavy" phases on some orders (hence, and concentration of "heavy" elements), therefore using of HC for determination of chemical elements by various analytical methods "raises" the sensibility of elements detection proportionally to factor of reduction (RC). For example, at low limit sensibility of DCP (Direct Current Plasma atomic emission spectrometry) of Pd at about 1ppm with RC=1000, the total sensibility will makes up 0.001 ppm, and the meaning of parameter I allows to determine an error of the obtained concentration.

The positive difference of Flame Atomic Absorption Spectrometry (FAAS) method of HM concentration determination from DCP is better metrological characteristics and much less uncertainty in determination of elements concentration. To this moment considerable experience in determination of Au, Pd and Pt concentrations in small amounts of material (n*10 - n*100 mg) by FAAS with preliminary gravitational enrichment is gathered during the investigations in co-operation with Laboratory of noble metals JSC “Mechanobr-Analyt” (the head of laboratory L.A.Ushinskaya). Simplifying the work scheme, it is possible to tell, that the fire assay (FA) stage was replaced by gravitational enrichment, that has reduced the price and simplified significantly traditional FA(Pb)FAAS. Because of uncertainty in determination of elements by FAAS is well defined and parameter I is established during at gravitational enrichment, the received data are very reliable and are proved by independent methods (mineralogical as well as chemical-analytic methods).
Briefly say, the detection limit depends on the initial sample weigh only! We attain detection limit for gold at 0.2ppb and prove this value: two gold grains of ~12µm was obtained from sample weight 300g.

It is very important, that the application of this combined method of determination of noble metals concentration in samples is an independent type of the analysis and gives an additional information: at such method the really extractable part of noble metals from ores is defined, that, in turn, is very important for technical characteristics of ores and for choice the industrial technologies of ore enrichment.

There is an opportunity of HC getting during prospecting: high amount sample processing by both mechanical and chemical component of the analysis at the rather low price make a method very attractive for geological survey as well as for prospecting works. Furthermore, if HC obtained from initial rock sample about 100-200g and FAAS is getting - the representativity and reliability of data is more high, because the usual sample weight for FAAS is about 10-30g.

It is important to emphasize, that at using of HC as a material for element concentration determination it is necessary to utilize the whole HC material. During the quartering of small amount of HC (1 - 40mg), containing an individual grains of ore phases, the strong distortion of the obtained concentration of chemical elements is possible. For an example, one grain of gold at the size 50µm in pyrite concentrate with the HC weight at 10mg (all about 20000 grains of this size) has a gold concentration at 250ppm.

 

Optical and microprobe analyses of
"heavy" concentrates (HC)

Derived "heavy" concentrates can be investigated by optical as well as microprobe methods, however advantage of microprobe study is apparent especially for identification and documentation of fine mineral phases.

Optic (mineragraphic) and microprobe analyses start with preparation of microprobe samples (MS) where "heavy" phases grains are mounted in one layer without overlapping and are cemented by organic compound. Then MS is polished slowly until all grains exposed on surface up to relationships between the total grains surface and compound surface will maximal, but grains crumbling don't occur. The subsequent MS polishing should provide uniform grains removing as well as compound removing (it’s softer), otherwise grains will raise above compound and risk of their loss increases considerably.

If to take into account the importance of each grain in a HC and that the grains size can be less than 10µm, it is obvious that the MS preparation is one of the most delicate and risky procedure in all technological process. Even at the works experience is a quit great, during the microprobe mount preparation for fraction <22µm, only for this procedure parameter I can be equal 0.5! It allows to remind once more: the work with HC requires special carefulness.

The final results, quality and time of mineragraphic and microprobe studies depend directly on two main reasons: 1.feasibility and real parameters of ore microscope or scanning electron microscope (microprobe)   and 2.talent for observation and analysis, qualification and operator skills of the geologist-mineralogist (but not a machine operator), working with device.

If the first factor determines detail of observation and material documentation quality, second one defines received data substantiality and opportunity on this basis to carry out the geological interpretation.

Depending on a goal, the mineragrafic or microprobe study can be carried out as on one of fractions (for example, on the finest and "heaviest" fraction – last one is most informative and representative), as and for all combinations of granulometric, electromagnetic and gravitational fractions of initial rock.

The experience of application of ppm-mineralogy has shown, that if the works are carried out under the complete scheme without infringement of technology, that I as the total parameter of quality of all works cycle makes 0.7 - 0.8 (not below 0.6). Moreover the real value of I on all technological units is possible to be determined over the formula: I = n/n'. The experimental definition of I parameter is very important, because extraction quality only gives an objective estimation of all processing and completeness of the received information.

It is very important to pay attention on that if the technology of mineralogical works provides sufficient mineralogical sensitivity and is created on reliable mineralogical basis (any technology, independently of value of I parameter), in practice of geological works a basis for really independent control of concentration elements definitions by various chemical-analytical methods appears. Furthermore, in working process the mineralogical assurance of HM concentration occurs. This provide the transition from a statistical level of the description of object using concentration of chemical elements, to phase one, i.e. level creating a basis for functional (cause-consequential, genetic) description of object. (Sir J.W.Gibbs shows still in 1876 that only phase level of study of material objects allows to definite functional connections between conditions of processes proceeding and materialized results of their proceeding).

 

Present to the patient reader

If you agree to consider the decision of puzzles and tasks as a kind of intellectual amusement, the given below example will help you to relax.

If geophysical (magnitometric) and mineralogical (approbation of river sediments for detection of minerals - satellite of diamond) methods are considered as the basic search methods of primary diamonds deposits, the next question will pertinent. What should to do for increasing of sensibility of prospecting methods on minerals - satellites and for an opportunity of finding of them on significant distance from primary sources?

Let's carry out mental (but having a real basis) experiment and assume that there is one pyrope grain by the size 1mm in heavy mineral fraction sample of weight at 10kg. Believing, that the average size of the grains in the sample is approximately the same, it is possible to fix a conventional sample representativity relatively of the pyrope grain. In 10kg of alluvial sand of density 2.7g/cm3 will contain 3.7 million grains of this size (the grain shape is approximated to the cube) and, hence, conventional sample representativity on pyrope will makes 1: 3700000.

It is interesting to compare, what there should be a sample weight with same representativity in the relation of pyrope, if the average grains size in sample is 50µm? It is easy to calculate that for the grains size of 50µm at conventional representativity 1: 3700000 the weight of sample will be equal to 1.25g! Further, it is reasonably to make a next inference. As the dependence between the linear grains size and their volume is an characteristic function of the third degree and the number of fine grains resulted from crushing of the large one is increased proportionally of third degree, the study even small weight of sample (up to 100g) fine-grained friable (alluvial, prolluvial etc.) material can gives the significant advantage in sensitivity of mineralogical works at mapping of dispersion haloes of minerals - satellite of diamond. Themselves haloes of fine-grained minerals - satellite under similar conditions of grain transport will occupy considerably large areas, than haloes of their coarse-grained varieties!
Is there any mistake in our reasoning? If there is no mistake, why is no prospecting potential of fine fractions used!? Dear Reader, you are welcome to answer on these questions and to solve the proposed task!

The above mentioned conclusions have a real basis: the experiment was put on and two parts of one sample of lamproites (one has weight at 800 g and another one is fraction <44µm having weight about 8g) were investigated. The large sample has experienced treatment by a traditional thermochemical method (the sample was investigated by V.A.Ezersky), and smaller one was treated by technology of ppm-mineralogy with extraction of gravitational concentrates with limit values of density about 3.3 -4.3 g/cm3 (search of Cr-diopside, pyrope, diamond) and more than 4.3 g/cm3 (search picroilmenite).
After thermochemical decomposition of the 800-gram part of sample only the grains of pyrope and Cr-diopside (4 and 6 grains) were revealed. Contrary in the 8-gram part of sample except of pyrope and Cr-diopside (18 and 14 grains, correspondingly) the tens grains of picroilmenite and phlogopite to be typomorphic for lamproites were founded out.

 

Above we have dwelled on the brief description of most substantial elements of high-sensitivity mineralogical works. For acquaintance with results of works on ppm-mineralogy technology in application to geological objects of different genetic types and some products of industrial ores processing we refer to the table of contents of the basic site part "by pressing" the button "Object List".

                 home.gif (4269 bytes) obj_list.gif (4250 bytes)       Go Top   Metrology   PPM-mineralogy