METE 256 ASSAYING ore reserve calculations control of processes - - PDF document

1/5/2017 1

Kwame Nkrumah University of Science & Technology, Kumasi, Ghana

METE 256 ASSAYING

• Dr. Anthony Andrews

Department of Materials Engineering Faculty of Mechanical and Chemical Engineering College of Engineering

www.knust.edu.gh

Course Objective

• Determination of the constituents of ores and

metallurgical products for:

– prospecting – ore reserve calculations – control of processes (gravity concentration) – recovery calculations – smelter schedule – bullion sales, etc

www.knust.edu.gh

Course Outline

• Sampling

– Methods of sampling – Sampling dividing techniques – Weight of samples relative to size of particles

• Statistical evaluation of data
• Metallurgical testing

– Bottle roll test, Column leach test, Acid digestion, Fire assaying, Diagnostic leaching

• Characterization and instrumental methods of analyses

www.knust.edu.gh

Course Assessment

• Quizzes – 10 points
• Mid Exam – 20 points
• Final Exam – 70 points

www.knust.edu.gh

Fire Assaying - Introduction

• The particular fire assay method under discussion is

aimed only at measuring Gold and Precious Metals

• Variations of fire assay can be used for other metals,

however, in most instances other analytical methods are favoured

www.knust.edu.gh

Fire Assaying - Background

• Many methods have been developed and refined over the years,

but “Fire Assay” remains a favoured method for determining the total gold content of a sample.

• In this method, a pulverised mineral sample is dissolved using heat

and fluxing agents.

• Precious metals are extracted from the melted material using

molten Lead (Pb).

• The precious metals are then separated from the Lead in a

secondary process called “cupellation”.

• The gold content of the precious metals collected is then

determined, using a variety of analytical techniques.

1/5/2017 2

www.knust.edu.gh

Fire Assaying – Applications

• Soil samples
• Exploration drill samples
• Grade control
• Mill solutions
• Tailings

www.knust.edu.gh

Traditional Fire Assay Method

(After Sample Preparation)

• 1. Sub-sampling & Catch-weigh
• 2. Fluxing
• 3. Firing
• 4. Cooling & Separation
• 5. Cupellation
• 6. Parting & Dissolution
• 7. Analysis

www.knust.edu.gh

Sampling

• A process of taking a portion from a bulk of material and

using that portion to represent the bulk of material. Or

• A sample is a small amount of material removed from a

bulk, such that it contains all the components in the proportion in which they occur in the original lot.

• Why Sample???

www.knust.edu.gh

Significance of Sampling

• Convenience in size for transportation and testing
• Obtain the desired information at the smallest cost
• Entire bulk may be inaccessible, too massive or too

dangerous to deal with. E.g human blood

www.knust.edu.gh

Important Considerations in Sampling

• Representative of the bulk
• Results from analysis of the sample should be appropriate to

predict the behaviour of the bulk

• No sample can provide absolute information about the bulk
• Statistical technique – provide an estimate within probability limit
• All the components in the bulk should have equal chance of

reporting into the sample

• Pre-sampling preparation to reduce biasness

www.knust.edu.gh

Categories of Sampling

• Exploratory

– Samples taken during prospecting, exploration and proving of a mine

• Controlled

– Samples taken to determine the content of specific constituents in a given environment

1/5/2017 3

www.knust.edu.gh

Principles of Sampling

• The distribution of values in an ore body is never

uniform

• The results of the sampling shall represent as truly as

possible the average metallic content of the ore/bulk material

• Each single sample must represent a true average of that

portion of bulk from which it is taken

www.knust.edu.gh

Methods for Sampling Material in a Lab

Stratified or Unstratified

• When is this sampling

technique used?

• Where will you take a sample

from?

www.knust.edu.gh

Methods for Sampling Material in a Lab

Random – chance

• Where will you take a sample

from?

Systematic – orderly

• Where will you take a sample

from?

www.knust.edu.gh

Methods for Sampling Material in a Lab

Grab sample

• Simplest, quickest, and most

flexible method

• It can be carried out on small

quantities using spatulas, or on large quantities using shovels

• This method uses the least

equipment, but also is the most prone to human biases and has a higher variance between samples than other methods.

Mixing a sample on a rolling mat. Mix by first drawing corner A so that the sample rolls towards C, then drawing corner B to corner D, then drawing corner C to corner A, then corner D to corner B, then repeat.

www.knust.edu.gh

Methods for Sampling Material in a Lab

Composite sample

• Individual samples combined

as single sample

www.knust.edu.gh

Sample Dividing Methods Scoop sampling

• The sample does not pass

through the sample device and hence prone to error

• Sample is taken from the

surface where it may not be typical of the mass.

• Shake sample before

sampling.

1/5/2017 4

www.knust.edu.gh

Sample Dividing Methods Coning and quartering

www.knust.edu.gh

Sample Dividing Methods

Chute-Type Riffle Sampler

www.knust.edu.gh

Sample Dividing Methods

Rotary Riffle Splitter

www.knust.edu.gh

Comparison of Lab Sample Devices

Sampling Method Standard Deviation of Samples (%)

Cone & Quarter 6.81 Grab Sampling 5.14 Chute-Type Sample Splitter 1.01 Rotary Riffle 0.125

www.knust.edu.gh

Sampling Problems and Requirements

• Degree of representativeness is based on heterogeneity
• Issues with variations in the distribution of components within the

bulk such as: – Size segregation – Mineralogy – Chemical composition – Grade – Moisture content – Weight – Shape

www.knust.edu.gh

Sampling Problems and Requirements

• Problems in sampling centers on:

– Nature and efficiency of sampling process – Weight reduction in the lab – Correctness in the interpretation of data – Reliability of results – Accuracy of results – Precision of results – Biasness in sampling and measurement

• Incorrectness of the above will result in sampling error

1/5/2017 5

www.knust.edu.gh

Size effect on sample integrity

• Mineralogy, grade and moisture content may vary with size
• Bulk material …Gross sample…Lab…Measurement

– Samples for lab measurement are obtained by standard techniques – Samples for lab measurement can be size-biased

• Coarse samples presents challenges in size volume reduction
• Smaller volume samples are more representative when particle

size is fine

www.knust.edu.gh

Sampling Calculations using Gy’s Method

• This method is a general-purpose calculation to determine the

minimum size of sample needed to ensure that it will be representative of the whole lot, within specified limits. Before using, approximate estimates of the following will be needed:

• The content of the species of interest in the lot (assay)
• The general shape of the particles
• The densities of the various species and phases present
• The particle size distribution
• The degree of liberation, and the grain size

www.knust.edu.gh

Sampling Calculations using Gy’s Method

Basic Equation:

www.knust.edu.gh

Sampling Calculations using Gy’s Method

Basic Equation:

When W is much larger than M, the equation is simplified to:

www.knust.edu.gh

Gy’s Equation – Working out C

𝑇2 = 𝑔𝑕𝑚𝑛𝐸3 1 𝑁 − 1 𝑀 Where C is fglm

• f= particle shape factor (describes the shape of the particles)
• g= granulometric factor (describes how much variation there is

in the size of particles)

• l = liberation factor (how close to liberation the material has

been ground)

• m = mineralogical composition factor (describes how much of

a rock is made up of the element of interest at a given grade)

www.knust.edu.gh

Calculating with incomplete information

Make the following conservative assumptions:

• f = 0.5 (normal blocky particles);
• g = 0.75 (narrow size distribution. Use g = 1 if the sample is
• bviously monosized and 0.25 for broad size distribution);
• l = 1 (grains are as large as the particles)
• The value of m will still need to be calculated, based on your best

estimate of the assay of the sample and the densities of the components of interest.

1/5/2017 6

www.knust.edu.gh

Calculating with incomplete information

• The liberation factor, l, is a measure of the degree of dispersion
• f the valuable material through the bulk, and of the homogeneity
• f the material.
• It is calculated from the expression:

𝑚 = 𝑀 𝑒 Where: L = the size where the values are essentially completely liberated (grain size), cm d = sieve size

www.knust.edu.gh

Calculating with incomplete information

• The composition factor (m), is calculated from the formula:

𝑛 = 1 − 𝑏 𝑏 1 − 𝑏 𝑠 + 𝑏𝑢 Where: r = specific gravity of the valuable component t = specific gravity of the remainder of the material a = fractional average assay of the valuable substance

www.knust.edu.gh

Gy’s Equation

• Simplified version of Gy’s equation:

𝑋 ≥ 125000𝑒3 W = weight, g d = diameter of the largest particle (cm)

www.knust.edu.gh

Work Example

• A sample of 200 g is to be taken and used for fire

assaying from a bulk sample of weight 5 kg with average particle size 10 mm. How fine should the material be crushed before a representative sample can be taken?

www.knust.edu.gh

Home Work

Bulk Materials Parameters:

• Materials of Interest: CuFeS2 in a silica matrix, 1.5% Cu

(4.3318% CuFeS2); Top Size = 1.5 cm; CuFeS2 grain size = 0.01 cm.

• Desired sampling accuracy: ±0.02% Cu, certainty of 0.99 (2.576

standard deviations)

• CuFeS2 specific gravity = 4.2; Overall specific gravity = 2.8;

Broad size distribution. Determine the minimum sample weight (in grams) needed for testing.

www.knust.edu.gh

Important Terminologies

• Replicates: - samples of the same size that are carried through an

analysis in exactly the same way.

• Precision: - the closeness of data to other data that have been
• btained in exactly the same way.
• Accuracy: - the correctness of measurement or closeness of a result

to its true or accepted value.

• Outlier: - an occasional result in replicate measurements that
• bviously differs significantly from the rest of the results.
• Bias: - a measures of the systematic error associated with an

analysis.

1/5/2017 7

www.knust.edu.gh

Important Terminologies

• Error:- a measure of deviation of the observed or calculated value

from the true value

• Absolute error:- the difference between the measured value and

the true value. 𝐹𝑏 = 𝑦𝑗 − 𝑦𝑢

• Relative error:- absolute error divided by the true value

𝐹𝑠 = 𝑦𝑗 − 𝑦𝑢 𝑦𝑢 × 100

www.knust.edu.gh

Random Error

• Caused by unknown and unpredictable changes in the experiment
• Examples of causes of random errors are:

– electronic noise in the circuit of an electrical instrument, – irregular changes in the heat loss rate from a solar collector due to changes in the wind.

• Random errors often have a Gaussian normal distribution

The Gaussian normal distribution. m = mean of measurements. s = standard deviation of measurements.

www.knust.edu.gh

Systematic Error

• Occur with measuring instruments when the calibration of the

instrument is not known correctly.

– Instrument has linear response

• Two types of systematic errors

– Offset or zero setting error – Multiplier or scale factor error

Systematic errors in a linear instrument (full line). Broken line shows response

• f an ideal instrument

without error. The accuracy of measurements is often reduced by systematic errors

www.knust.edu.gh

Gross Error

• Gross errors occur occasionally, are often large, and may

cause a result to be either high or low.

• Gross errors lead to outliers
• Gross errors can be avoided by using two suitable

measures

• 1. Proper care should be taken in reading, recording and

calculating data.

• 2. By increasing the number of experimenters

www.knust.edu.gh

Sample Extraction Methods

• Samples can be taken by hand or by using machines

– Hand sampling is slow and prone to bias

• Automatic sampling is done by mechanically driven

sampling cutters designed to cut the falling ore or pulp at predetermined intervals

• Bulk sampling can be done in continuous streams or

stationary systems

• General rule in sampling:

Whenever possible, a sample should be taken when the material is in motion

www.knust.edu.gh

Stationary and Continuous Streams

Stationary

• Heaps
• Drums
• Trucks
• Bars
• Bags
• Bucket conveyor
• Slurry in container

Continuous

• Ore on conveyor belt
• Slurries in pipes

1/5/2017 8

www.knust.edu.gh

Sampling from a moving streams

• The ratio of the cutter width to the diameter of the largest particle

should be made as large as possible with a minimum value of 20:1

• The collecting device should cover the whole stream
• The device/cutter should be presented at right angles to the

stream.

• The speed of the cutter should be constant

www.knust.edu.gh

Sampling from a static slurry/liquid

Eg, Slurry in a buckets

• First agitate to suspend particles
• Scoop from various sections or pour whole content

depending on number of containers and stage of sampling

• Filter and dry slurry

Eg, Reagent or water in container

• First agitate to homogenize system
• Scoop from various sections or pour whole content

www.knust.edu.gh

Sampling from heaps and dumps

Eg, heap leach pads

• May demand the use of pipe and

auger sampler

• Pipe should be long enough to

reach the bottom of the heap to be sampled

• Samples can be taken at various

depths

• After sampling, pipe is

withdrawn, and sample discharged

www.knust.edu.gh

Sampling from drums and bags

• First sample may be taken

randomly

• Additional samples

systematically

www.knust.edu.gh

Sampling from metals and alloys

• Chipping the corners of the bar
• Drilling holes in the bar and using the material that

comes out of it is as sample.

• Sawing through the bar and using the dust as sample.

www.knust.edu.gh

Grab Sampling

• Taken with a scoop, shovel, hand, bottle , pipe, etc
• Can follow fixed pattern
• Rapid and cheaper but unscientific
• For bulk sampling, composite may be better
• For lab samples, standard dividing techniques may

provide a more representative sample

1/5/2017 9

www.knust.edu.gh

Automatic Vs Manual

• When?
• Where?
• Advantages?
• Disadvantages?

www.knust.edu.gh

Sampling on metallurgical plants

• To acquire information on ore entering the plant for

treatment.

• To inspect the condition of the ore at selected points

during its progress through the plant.

• To check the performance of the plant against set targets
• To correct malfunctions, reduce losses, and improve upon

recovery

www.knust.edu.gh

Typical Sampling Points in a Plant

• Head feed
• Comminution
• Crushing

– product

• Grinding circuit samples

– Ball mill discharge – Classifier return – Classifier overflow

• Concentration
• Flotation products
• Gravity concentration

– Concentrate – Tailings

• Leaching
• Solution
• Solids
• Loaded carbon
• Desorption
• Eluate
• Barren carbon
• Pretreatment
• Roasting
• Biooxidation
• Smelting
• Bullion

www.knust.edu.gh

Sampling from Metallurgical Plant

Head sample

• Location:

– Crusher product and mill feed

• Data taken/Test performed:

– Moisture content to correct for dry tonnage – Size analysis to check crusher performance and correct size for mills – Tonnage by a weightometer

www.knust.edu.gh

Sampling from Metallurgical Plant

Grinding circuit samples

• Location:

– Ball mill discharge, classifier overflow and underflow

• Data taken/Test performed:

– Pulp density/solid-liquid ratio to control coating of balls. – Density and particle size from classifiers to check classifier performance and effect on mill throughput, leaching feed, etc.

www.knust.edu.gh

Sampling from Metallurgical Plant Flotation products

• Location:

– Concentrate – Tailings

• Data taken/Test performed:

– Grade by fire assaying – Partial chemical analysis

1/5/2017 10

www.knust.edu.gh

Sampling from Metallurgical Plant Biooxidation

• Location:

– Primary reactors – Secondary reactors

• Data taken/test performed:

– Acidity, – Temperature, – Fe2+/Fe3+ ratio, – Sulfide, carbon, arsenic, gold grade, – Bacterial activity, redox potential,

www.knust.edu.gh

Sampling from Metallurgical Plant Leaching/cyanidation/adsorption circuit

• Location:

– Head and tail tanks – All tanks – Carbon recovery screen

• Data taken/Test performed:

– pulp density, – pH – cyanide level, – dissolved oxygen, – carbon content and grade Loading of carbon

www.knust.edu.gh

Sampling from Metallurgical Plant Elution/desorption/stripping circuit

• Location:

– Soaking/preheating tanks – Stripping tank

• Data taken/Test performed:

– Gold in solution – Loaded and barren carbon grade – Caustic-cyanide strength

www.knust.edu.gh

Sampling from Metallurgical Plant Electrowinning

• Location:

– Cell solution – Cathode

• Data taken/Test performed:

– Gold in solution – Loaded cathode – pH – Temperature

www.knust.edu.gh

Sampling from Metallurgical Plant Smelting

• Location:

– Smelter feed – Furnace – Smelter products

• Data taken/Test performed:

– Purity of gold bullion – Gold value in slag – Gold value in calcined cathode – Temperature

www.knust.edu.gh

Basic Statistical Analysis

• A subset of the population is used to estimate the population
• The sample will therefore be a representative of the population

Population

Sample

1/5/2017 11

www.knust.edu.gh

Measure of Central Tendencies

• Three common ways to measure central tendency:

– Mean – Median – Mode

• Mean is based on quantitative data whereas median is

based on position and mode is based on frequency

www.knust.edu.gh

Measure of Central Tendencies

• Find the mean, median and mode.
• The sample mean, y, is given by:
• where n is the sample size and yi are the measurements

www.knust.edu.gh

Measure of Central Tendencies

Difference between precision and accuracy

www.knust.edu.gh

Symmetrical Distribution

• Mean, Median and Mode are all the same, mound shape,

no skewness

www.knust.edu.gh

Right Skewness

• Mean to the right of the Median
• Long tail on right

www.knust.edu.gh

Left Skewness

• Mean to the left of the Median
• Long tail on left

1/5/2017 12

www.knust.edu.gh

Measures of Variability

WHAT IS V

ARIABILITY?

• Variability refers to how "spread out" a group of scores is.

Bar chats of two quizzes Quiz 1 Quiz 2

www.knust.edu.gh

Range

• The range is the highest score minus the lowest score

Examples

• 1. What is the range of the following group of numbers: 10,

2, 5, 6, 7, 3, 4?

• 2. Here’s a data set with 10 numbers: 99, 45, 23, 67, 45, 91,

82, 78, 62, 51. What is the range?

www.knust.edu.gh

Interquartile Range

• The interquartile range (IQR) is the range of the middle 50% of the

scores in a distribution.

• It is computed as follows:

QR = 75th percentile - 25th percentile Quiz 1 Quiz 2

www.knust.edu.gh

Variance

• The variance is defined as the average squared difference
• f the scores from the mean

where σ2 is the variance, μ is the mean, and N is the population where s2 is the estimate of the variance and M is the sample mean

Population variance Sample variance

www.knust.edu.gh

Variance

www.knust.edu.gh

Standard Deviation

• The standard deviation is the square root of the variance
• The symbol for the population standard deviation is “σ”;

the symbol for an estimate computed in a sample is “s”

1/5/2017 13

www.knust.edu.gh

Standard Deviation

Normal distributions with standard deviations of 5 and 10.

www.knust.edu.gh

GOLD

• Classification of gold ores
• Host materials
• Characteristics of gold ores

– Equipment used – Parameters to observe

• Analyses

www.knust.edu.gh

GOLD ORES

• Most noble metal, native occurrence
• Also associated with silver, tellurium, bismuth and

PGM’s

• Typical ore grades: 0.5 to 20 g/t
• Primary gold source

– ores

• Secondary gold sources

– gravity concentrates – flotation concentrates – plant tailings – refinery tailings – recycled gold

www.knust.edu.gh

Classification of Gold Ores

Classification of gold ores and typical recovery with traditional methods

www.knust.edu.gh

Most common causes for refractoriness and double refractory ore appearance

www.knust.edu.gh

Classification of Gold Ores

• Non-refractory;

– placer, – free-milling, – oxidized

• Refractory

– Ultrafine gold particles in the matrix of sulphide minerals – Carbonaceous materials

1/5/2017 14

www.knust.edu.gh

Types of Gold Deposits

• Placer ores
• Oxidized ores
• Primary ores

www.knust.edu.gh

Gold Ore Types

• Main ore types
• placers
• oxidized
• free milling
• silver rich
• iron sulphide bearing
• arsenic sulphide bearing
• carbonaceous
• copper bearing
• antimony bearing
• gold telluride bearing

easy processing refractory

www.knust.edu.gh

Gold in its ore (host material)

• Tiny particles (< 75 μm)
• Minute concentration (<0.001%)
• Highly disseminated in the gangue (unwanted) materials

(99.999%)

• Recovery depends on particle size of gold and degree of

association with unwanted materials

www.knust.edu.gh

Traditional Gold Recovery

• Recovery rate of refractory gold ores can be improved

through roasting.

• Laboratory roasting is done in an electric furnace.
• Use of lead and other reagents in laboratory smelting.

www.knust.edu.gh

Characteristics of Gold Ores

• Gold grain size distribution
• Type of gangue minerals
• Mineral associations and alterations
• Mineralogical mode of occurrence
• Variations of the above items within the same ore body

www.knust.edu.gh

Characterization Methods

• X-ray fluorescence (XRF)
• X-ray diffractometer (XRD)
• Optical microscope (OM)
• Scanning electron microscope (SEM)
• Infra-red spectrophotometer (IRS)
• Raman spectrometer (RS)
• X-ray photoelectron spectroscopy (XPS)
• Atomic emission spectrophotometer (AES)
• Volumetric titrator

1/5/2017 15

www.knust.edu.gh

Parameters to look out for

• Qualitative and quantitative identification of:

– Elements – Minerals – Compounds, etc.

• Mineral associations
• Many other things like:

– Shape/size – Texture – Crack propagation – Presence of microcracks

www.knust.edu.gh

XRD Analysis

www.knust.edu.gh

Analysis

• Determination of:

– Moisture content and pulp density – Dissolved oxygen, pH and lime addition – Cyanide consumption

• Aqua regia
• Acid digestion
• Cyanidation
• Bottle roll test
• Column leaching
• Diagnostic leaching
• Fire assaying

www.knust.edu.gh

Determination of Moisture Content

𝑁𝑝𝑗𝑡𝑢𝑣𝑠𝑓 𝑑𝑝𝑜𝑢𝑓𝑜𝑢 % = 𝑁𝑝𝑗𝑡𝑢𝑣𝑠𝑓 𝑑𝑝𝑜𝑢𝑓𝑜𝑢 (𝑕) 𝑋𝑓𝑢 𝑥𝑓𝑗𝑕ℎ𝑢 (𝑕) × 100% 𝑁𝑝𝑗𝑡𝑢𝑣𝑠𝑓 𝑑𝑝𝑜𝑢𝑓𝑜𝑢 𝑕 = 𝑋𝑓𝑢 𝑥𝑓𝑗𝑕ℎ𝑢 𝑕 − 𝐸𝑠𝑧 𝑥𝑓𝑗𝑕ℎ𝑢 𝑕

• If 100 tonnes of ore is treated and the grade is 0.002 g/t,

find the quantity of metal recovered assuming the moisture content is 5%.

www.knust.edu.gh

Determination of Pulp Density

Determining solids fraction

• To determine the percent solids of a slurry from the

density of the slurry, solids and liquid ∅𝑡𝑚 = 𝜍𝑡(𝜍𝑡𝑚 − 𝜍𝑚) 𝜍𝑡𝑚(𝜍𝑡 − 𝜍𝑚)

• ∅𝑡𝑚 = 𝑡𝑝𝑚𝑗𝑒𝑡 𝑔𝑠𝑏𝑑𝑢𝑗𝑝𝑜 𝑗𝑜 𝑡𝑚𝑣𝑠𝑠𝑧 𝑛𝑏𝑡𝑡
• 𝜍𝑡 = 𝑡𝑝𝑚𝑗𝑒𝑡 𝑒𝑓𝑜𝑡𝑗𝑢𝑧
• 𝜍𝑚 = 𝑚𝑗𝑟𝑣𝑗𝑒 𝑒𝑓𝑜𝑡𝑗𝑢𝑧
• 𝜍𝑡𝑚 = 𝑡𝑚𝑣𝑠𝑠𝑧 𝑒𝑓𝑜𝑡𝑗𝑢𝑧

www.knust.edu.gh

Determination of Pulp Density

• Liquid mass from mass fraction of solids

∅𝑡𝑚 =

𝑁𝑡 𝑁𝑡𝑚 × 100

𝑁𝑡𝑚 =

𝑁𝑡 ∅𝑡𝑚 = 𝑁𝑡 + 𝑁𝑚

𝑁𝑚 =

𝑁𝑡 ∅𝑡𝑚 − 𝑁𝑡 ∅𝑡𝑚 = 𝑡𝑝𝑚𝑗𝑒 𝑔𝑠𝑏𝑑𝑢𝑗𝑝𝑜 𝑗𝑜 𝑡𝑚𝑣𝑠𝑠𝑧 𝑁𝑡 = 𝑛𝑏𝑡𝑡 𝑝𝑔 𝑡𝑝𝑚𝑗𝑒𝑡 𝑁𝑚 = 𝑛𝑏𝑡𝑡 𝑝𝑔 𝑚𝑗𝑟𝑣𝑗𝑒 𝑁𝑡𝑚 = 𝑛𝑏𝑡𝑡 𝑝𝑔 𝑡𝑚𝑣𝑠𝑠𝑧

1/5/2017 16

www.knust.edu.gh

Determination of pH and Lime

𝑞𝐼 = −log[𝐼+]

• Use to determine the pH of cyanide solutions.

– Ex: pH > 10.5

• Precipitation of salts

– pH < 8.5 – Decreases cyanide efficiency

• pH meter and a probe are used to measure pH

– Calibration required (Buffer solution)

www.knust.edu.gh

Determination of Cyanide Consumption

• Leaching of gold with cyanide

– Rolling bottle with perforated lid – Columns (cylinders with perforated base) – Miniature tank with a stirrer – Beaker placed on a shaker – Bottle/beaker/container with magnetic stirrer

• pH of solution is raised to about 11, before cyanide

addition

www.knust.edu.gh

Bottle Roll Test

• Weigh 1 kg of sample
• Prepare 50% pulp density
• Adjust pH
• Add cyanide
• Agitate by rolling bottle for 72 hrs
• Take solution samples at time intervals (1,2,4,12, 24 hrs)
• Take 100 g samples to determine the tailings grade

www.knust.edu.gh

Column or Percolation Leach Test

• Design parameters used for heap leaching
• Crush ores
• Mount in columns
• Irrigate with cyanide
• Several columns mounted to determine the appropriate

particle size, strength of cyanide etc.

www.knust.edu.gh

Acid Digestion

• Dissolution of metals/elements using acid
• High temperature and/pressure
• Fume chamber with extractor is required
• Small quantities of material can be used

www.knust.edu.gh

Acid Digestion

• Perchloric acid (HClO4)

– Used for wet ashing when sample contains carbonaceous material

• Hydrofluoric acid (HF)

– Used to digest silica to release occluded minerals

• Aqua-regia

– Used to determine gold in samples – Mixture of HNO3 and HCl

1/5/2017 17

www.knust.edu.gh

Diagnostic Le stic Leaching

• 1. Water leaching (if tailings material)
• 2. Cyanide leaching of liberated gold
• 3. Digestion with dilute hydrochloric acid to break down weak

components like carbonates, followed by cyanide leaching

• 4. Digestion with nitric acid to break down/oxidize components like

sulfur, followed by cyanide leaching

• 5. Roast at 750oC to decompose carbonaceous matter, followed by

cyanide leaching

• 6. Fire assaying of final tailings to determine gold in quartz, or

leaching with hydroflouric acid (teflon beaker)

• 7. Add all the gold to get the calculated head grade

www.knust.edu.gh

Conversion of grade from g/l of solvent to g/t of ore

Question Suppose a 50g sample was digested with acid and then filtered into a 100 ml volumetric flask and topped to the mark with distilled water. If the AAS reading is 3.5 mg/l, estimate the grade of ore in g/t.

www.knust.edu.gh

Cyn Cynanidation

hrs

www.knust.edu.gh

Cyn Cynanidation

Leaching time (hrs)

www.knust.edu.gh

Gold Gold Reco ecover ery

Question A Bottle roll test was conducted on 2 kg of a soil sample at 50%

• solids. 50 g of the tailings was digested with aqua-regia and topped to

10 ml. The data obtained after AAS is presented in Table 1. Find the head grade and percent recovery for each period and plot a suitable

• graph. Are there preg-robbers in the sample?

Time, h Gold in solution, mg/l 2 1.75 4 1.52 8 3.10 16 4.52 24 4.35 Tailings 0.92

www.knust.edu.gh

What t is s Fire Ass ssaying ying?

• Quantitative method for the

determination of Au, Ag, Sn, Cu, Hg, Pb and the platinum group of metals

• Consist of crucible fusion of

weighed amount of sample with suitable reagents

• Two major stages are fusion and

cupellation

1/5/2017 18

www.knust.edu.gh

Objecti tives s of Fire Ass ssaying ying

• The valuation of a mining property
• The basis for buying and selling various materials
• In plant quality control
• Accounting and inventory requirements
• Environmental considerations.

www.knust.edu.gh

Step teps s in Fire Ass ssaying ying

• 1. Pulverizing/Sampling
• 2. Mixing of sample and fluxes
• 3. Crucible fusion (assay furnace)
• 4. Cupellation (assay/cupellation furnace)
• 5. Parting and Annealing OR
• 6. Acid digestion and AAS finish

www.knust.edu.gh

Fire Reagents ts

• Borax (sodium tetraborate)
• Silica
• Soda ash (sodium carbonate)
• Litharge (lead oxide)
• Carbon (in the form of flour or charcoal)
• Nitre (potassium nitrate)

www.knust.edu.gh

Borax (N (Na2B4O7)

• Anhydrous borax melts at 741oC to form a viscous glass,

which becomes more fluid at elevated temperatures

• It is a strongly acidic and readily dissolves almost all

basic metal oxides

• The borax melts to form a colourless transparent glass

𝑂𝑏2𝐶4𝑃7 → 𝑂𝑏2𝐶2𝑃4 + 𝐶2𝑃3 𝑎𝑜𝑃 + 𝐶2𝑃3 → 𝑎𝑜𝐶2𝑃4

www.knust.edu.gh

Soda Ash sh (N (Na2CO CO3)

• Na2CO3 is a strong basic flux and fuses most readily with

silica to form fusible slag

• It is used as an oxidising and desulphurising reagent
• Na2CO3 melts at 850oC, and at 950oC, dissociates

partially evolving CO2 and liberating some free alkali 𝑂𝑏2𝐷𝑃3 → 𝑂𝑏2𝑃 + 𝐷𝑃2 𝑂𝑏2𝑃 + 𝑇𝑗𝑃2 → 𝑂𝑏2𝑇𝑗𝑃3 𝑂𝑏2𝐷𝑃3 + 𝑂𝑏2𝑇𝑗𝑃3 → 𝑂𝑏𝑇𝑗𝑃4 + 𝐷𝑃2

www.knust.edu.gh

Litha tharge (Pb PbO)

• Litharge is a readily fusible basic flux
• It acts as an oxidising and desulphurising agent
• It melts at 883oC and reacts with the reducing agent to

liberate metallic lead

• This metallic lead provides the lead rain, which collects

the noble metals in the sample to form a lead button upon solidifying

1/5/2017 19

www.knust.edu.gh

Red Lead (Pb (Pb3O4)

• Used as alternative to litharge
• Additonal oxidising effect during fusion
• More expensive than litharge and hence not commonly

used

www.knust.edu.gh

Nitr tre (K (KNO3)

• KNO3 is a powerful oxidising reagent that melts at 339oC
• It decomposes at about 400oC, liberating oxygen
• Nitre oxidises sulphides to sulphates, and arsenides to arsenates.
• It is used most commonly for converting metallic sulphides to
• xides
• The disadvantages of nitre are the possibility of oxidising silver

and the tendency to cause boiling of the charge

www.knust.edu.gh

Reducing Agent

• Serves two purposes:

– They reduce sufficient litharge to produce the Pb button, which collects the values – They reduce any ferric oxide present in the sample to the ferrous state so that it can be slagged

• Sources of reducing agents:

– Those added to the charge – Those already present in the charge

• Most effective reducing agent is carbon

www.knust.edu.gh

Theoreti tical Reducing Power

• Denotes the amount of grams of metallic Pb that is

produced from 1 g of reducing agent added to PbO 2𝑄𝑐𝑃 + 𝐷 → 2𝑄𝑐 + 𝐷𝑃2 [C = 12; Pb = 207] 𝑆𝑓𝑒𝑣𝑑𝑗𝑜𝑕 𝑞𝑝𝑥𝑓𝑠 = 414 12 = 34.5 Flour or starch as reducing agent 12𝑄𝑐𝑃 + 𝐷6𝐼5𝑃10 → 12𝑄𝑐 + 5𝐼2𝑃 + 6𝐷𝑃2 𝑆𝑓𝑒𝑣𝑑𝑗𝑜𝑕 𝑞𝑝𝑥𝑓𝑠 = 2486 162 = 15.3

www.knust.edu.gh

Oxidising sing Agent

• The presence of sulphides, arsenic or antimony makes it

necessary for oxidising agents to be added

• Examples of oxidising agents:

– Red lead (Pb3O4), – Manganese dioxide (MnO) and – Nitre (KNO3)

www.knust.edu.gh

Theoreti tical Oxidising sing Power

• Denotes the amount of grams of Pb prevented from being

reduced by 1 g of oxidising agent 2𝐿𝑂𝑃3 → 𝐿2𝑃 + 2𝑃2 2𝑃2 + 4𝑄𝑐 → 4𝑄𝑐𝑃 𝑃𝑦𝑗𝑒𝑗𝑡𝑗𝑜𝑕 𝑞𝑝𝑥𝑓𝑠 = 828 202 = 4.1

1/5/2017 20

www.knust.edu.gh

Fluxes

Reagent Reaso sons Litharge Used to provide lead to collect the precious metals. It is also a strong basic flux and reacts with metallic

• xides and silica to form a slag. By far the most

expensive component of a fire assay flux. Soda Ash A powerful basic flux that is usually the principal component of fire assay flux. It reacts with silicates to form a slag Borax An acidic flux that lowers the fusing point of all slags. It forms fusible complexes with limestone and magnesite

www.knust.edu.gh

Fluxes

Reagent Reaso sons

Silica An acidic flux that forms the principal component of many samples. Small amounts are present in the flux to prevent attack on the fire assay crucibles when assaying samples deficient in silica. Nitre A powerful oxidising agent added to the flux when assaying samples containing sulfides Flour A source of carbon used to reduce the litharge to lead. Silver A small amount is added to the flux to provide a collection medium for the precious metals

www.knust.edu.gh

Cruc ucible ble Fu Fusion sion

• Heating (fusion) of the charge in a

crucible to attain two products; Pb button and slag

• Litharge is added as a flux (except Pb
• re)
• PbO undergoes reduction to produce

lead which contain the noble metals

• Formation of matte(s) or speise must

be avoided since either of these would attempt to collect some of the values

www.knust.edu.gh

Crucible Fusion sion – Poss ssible Layers

www.knust.edu.gh

Characteristi teristics s of Meta Metallic Pha hase se

• 1. A minimum amount of impurities
• 2. A bright, soft, malleable button
• 3. A button close to the desired weight
• 4. A complete recovery of the noble metals

www.knust.edu.gh

Sl Slag Pr Properties

• 1. It should have a comparatively low formation temperature.
• 2. It should be pasty at its formation temperature.
• 3. It should be thin and fluid when heated to somewhat above its

melting point.

• 4. It should have a low capacity for noble metals.
• 5. It should allow a complete decomposition of the sample by the

fluxes.

1/5/2017 21

www.knust.edu.gh

Slag Properti ties

6. It should not attack the material of the crucible to any great extent. 7. Its specific gravity should be low. 8. When cold, it should separate readily from the lead, and be homogeneous. 9. It should contain practically all the impurities of the sample.

www.knust.edu.gh

Prepar eparation ion of

• f Pb

Pb But utton f

• n for
• r

Cup upella ellation ion

• After fusion, the melt is poured into a mould and allowed

to cool until the Pb solidifies.

• The lead is detached from the slag by striking at the

junction of the Pb and slag with a hammer

• The cone-shaped button is hammered into a rough cube
• n an anvil.

www.knust.edu.gh

Cup Cupellatio tion

• Cupellation is the oxidation of the lead and subsequent absorption

into a small shallow porous cup called a cupel

• The cupel should have a smooth surface and readily absorb its own

weight or a little more than its own weight of Pb without cracking

• Cupel must be dry before being placed into the muffle furnace and

then raised to cupellation temperature of 950oC before button is added

• The furnace should have an ample supply of air for the oxidation
• f Pb

www.knust.edu.gh

Step teps s in Cupellati tion

• 1. Preheating of the cupels to drive off any water, organic

matter, and carbon dioxide.

• 2. Increase temperature to 950oC and place lead buttons.
• 3. Lead button melts covering the cupel with a dark scum

composed mostly of litharge.

• 4. Molten litharge slide off the surface of the lead and

absorbed by the cupel

www.knust.edu.gh www.knust.edu.gh

Cup Cupellatio tion

• Most of the Pb is absorbed by the cupel in the form of PbO but part

is volatilised and carried away as Pb fumes

• In the course of cupellation, the furnace door may be opened

slightly to allow air into the furnace to aid the process

• To prevent cracking due to colder air, the front row of cupels are

not fed with Pb buttons

• The furnace temperature has to be regulated as all the metal will

volatilize if temperature is too high

1/5/2017 22

www.knust.edu.gh

Lo Loss ss of

• f Silver

er in n Cup upel el

• Pb/Cu ratio should be around 16 at least, to ensure the absence of

free copper in the system Cu2O-PbO phase diagram

www.knust.edu.gh

Parti ting of the the Gold Beads

• This is the process of separating the Au and Ag obtained after

cupellation

• In the determination of gold and silver, the doré beads derived

from cupellation are weighed

• Separation achieved by dissolving Ag-Au alloy in acid

– HNO3 or concentrated H2SO4

• Au residue is washed, dried and annealed in muffle furnace until is

bright red

• Weigh Au and determine weight of Ag

www.knust.edu.gh

Parti ting of the the Gold Beads

• The concentration of Ag should be high. If low, more silver should

be added (Ag:Au at least 3)

• Addition of Ag to the fusion is termed inquartation

– Ag-Au ratio should be known

• Recupeling the doré bead with three times its weight in silver

www.knust.edu.gh

Ann Annea ealing ng

• After parting, the bead is annealed and weighed
• Annealing is a heat treatment done to:

– avoid weighing extraneous materials – provide opportunity to examine the gold bead for impurities – destroy porosity and so prevent the absorption of moisture and gasses

www.knust.edu.gh

Weighing

• Sensitive balance used (1/500 of a miligram)
• Maximum load capacity is 1 or 2 g
• Conducive environment
• Express weights in proportion in the material sampled
• 1 Assay Ton = 29.166 g of ore
• 0.001 g of gold in a sample weighing 29.166 g

www.knust.edu.gh

Fi Fire e Ass Assay ay

Ins nstrument umental al anal analysis is - AAS AAS

• Some laboratories complete fire assaying by recourse to

digestion of the prill and AAS analysis instead of parting, annealing and weighing

• This method reduces problems associated with parting

such as:

– handling of tiny gold beads, – incomplete silver dissolution, and – inefficiencies associated with weighing on a 4-6 d.p electronic balance.

1/5/2017 23

www.knust.edu.gh

Fir Fire e As Assay ay

Ins nstrument umental al anal analysis is - AAS AAS

• In this technique, silver-gold prills are digested with

boiling nitric acid and hydrochloric acid

• Gold content in the resulting solution may be determined

by AAS analysis

• Detection level of gold by fire assay/AAS method is

0.01ppm

• The grade per tonne of ore can then be calculated

www.knust.edu.gh

Fir Fire e As Assay ay

Ins nstrument umental al anal analysis is – ICP-OE OES

• ICP-OES has the advantage of being

able to analyze gold and other elements such as PGEs in one reading

• Results compared to known and

verified standards

• Detection level Gold by Fire Assay /

ICP-OES method = 0.01ppm

www.knust.edu.gh

Dif Difficu ficult Samp Samples

Eleme ment Co Comme mment Copper May be reduced to the metal during fusion and reports in the lead button. This can then inhibit cupellation, making it impossible to recover the precious metals. Alternatively, it may react with pyrite to form a matte that will preferentially absorb gold. Nickel Reacts similarly to copper with regards to cupellation, but will create problems at far lower concentrations (> 0.5% in the lead button). A combination of nickel and copper will create far greater problems than either of the elements individually. Antimony Completely miscible with molten lead. More than 2% antimony in the lead button may cause cracking of the cupel, resulting in low gold recovery. It will also form antimonides with copper, nickel or iron (speiss) which will preferentially absorb gold Arsenic Will form arsenides with copper, nickel or iron (speiss), resulting in low gold recovery.

www.knust.edu.gh

Dif Difficu ficult Samp Samples

Eleme ment Co Comme mment Tellurium Extremely detrimental to the recovery of precious metals during cupellation when present in amounts of > 0.5% in the lead

• button. It lowers the surface tension of the precious metal prill,

allowing some of the doré to be absorbed by the cupel. Selenium Behaves similarly to tellurium. Sulphur (Sulfides) Can cause problems by forming mattes with copper, nickel or iron compounds, resulting in low gold recoveries. Will also produce large lead buttons if using a high litharge flux. This can be controlled by adding a calculated amount of oxidant. Carbon (organic matter) Can cause major problems during fire assaying due to the formation of lead shot within the slag, leading to low lead button

• weights. This will result in low gold recoveries

www.knust.edu.gh

Applica pplication ion of

• f Fir

Fire e Assay ay Tec echniques hniques

• In the precious metal industry, fire assay techniques are usually

applied to ores, metal alloys and solutions Ores

• A general principle is that a siliceous ore requires a basic flux, and

a basic ore needs an acid flux Bullion

• The determination of precious metals in metallic alloys is referred

to as bullion assaying

• Received in the form of shot, borings, granules