Metallurgical Analysis Intro

Introduction to Metallurgical Analysis

• Properties of minerals heavily rely on their composition and component amounts.
• Key parameters include identification, determination, and estimation.

Identification

• Depends on fundamental and consistent characteristics.
• Employs specific properties to differentiate minerals.
• Depends on the chemical nature of the mineral.
• An example of a minerals chemical composition is quartz which is SiO2
• Relies on external crystallographic features.
• Microscopic external crystallographic features are considered
• Relies on Internal atomic/ionic arrangements within the crystal lattice.
• Considers physical and chemical changes within the mineral.

Determination

• The result of definitively ascertaining the position, nature, or amount through measurement, weighing, or counting.

Estimation

• The result of approximately ascertaining numbers, quantities, and magnitudes without exact enumeration, weighing, or measurement.

Sampling

• Securing a representative small weight or fraction from a large mass.
• Removing part of the bulk material to reduce quantity while minimizing alterations to characteristics.
• The guiding principle is a gradual reduction in bulk alongside a gradual reduction in the size of the pieces.

Elements of Sampling Problem

• Precise definition of the quality to be tested is required.
• The characteristics of the quality under test must be determined.
• Characteristics of the tested material is important.
• Relation of the quality to the material is required.
• Necessary degree of accuracy.
• Characteristics of the sample to supply information about the quality.
• The way in which the parent lot occurs.
• Size of the sample itself.
• Suitable sampling method is needed.
• Mechanisms involved for the sampling method.

Sampling Considerations

• Uniformity of Composition of the Bulk needs to be considered.
• Differently-graded ores need to be accounted for.
• Select portions with value require attention.
• Account for ores with different sizes and specific gravity.
• Ratio between different grades must match the original bulk.
• Thorough mixing to solve issues with ores in different sizes or specific gravity.
• Fineness of the division must be considered.
• The size of the single pieces forming the bulk before a cut for the samples must be known.
• Size of the bulk controls the cutting.
• Mode of Occurrence of the Metallic constituents.
• Sampling is easy with a very large bulk and uniform composition
• Sampling is difficult when metal is carried by a comparatively few pieces in a large mass of barren material.
• The ratio of weight of sample to the fineness of division is important.
• A definite ratio must exist between the weight of the sample and the size of the ore particle.
• "The weight taken for sample shall be proportional to the square of the diameter of the largest particle” according to Prof. R.H Richards (MIT).
• The Ratio of weight of the 1st Sample to the total weight of the bulk should factor in the uniformity of composition and fineness of division of the bulk material.
• A General Sampling Practice is to cut out the 1st sample from 5% to 20% of the ore bulk.
• Methods of cutting out the Sample should obtain a fair average proportion of all the different qualities and sizes of the parent bulk material.

Sampling Practices - Objectives

• Exposing the determinative variables relevant to the operations.
• Explaining the independence of the variables.
• Controlling sampling operations
• Predicting the results
• Sampling practices are very prone to error

Sampling Practices - Perfect Sampling Technique

• Yield samples such that any deviation from complete representation of quality to be tested in the sample are ascribed to chance causes

Statistics of Sampling Error

• Sampling error occurs when an analyst does not select a sample that represents the entire population of data.
• there will be a deviation in the sampled value versus the population value.
• Population - the total number of discrete units in a bulk material.
• Parameter - numerical characteristics of the population.
• Statistics - selected samples which may be practically measured and examined and which give an estimate of the parameters of the whole universe or population.
• Arithmetic Mean µ or x is used.
• Measures the central tendency of a certain parameter.
• Does not completely describe the distribution of the parameter over the population.
• Variance, s2 or σ2, is used.
• This is the mean square deviation of the observed parameter from the mean.
• Describes how possible values are spread around the expected value
• Confidence interval is the probability that a certain parameter will fall between set values.
• These produce Confidence limits which are upper and lower boundaries

Control of Sampling Error

• No sampling system can be so perfect
• Choose a method which will give random rather than systematic errors
• Cost considerations should never be allowed to reduce the precision of sampling
• Continuous automatic sampling is more preferred than manual

Sampling Methods

• MANUAL SAMPLING / Hand Sampling is more expensive and waste labor in continuous stream sampling, it includes Personal biases.
• Hand Sampling: Grab Sampling- This is the simplest form involving.
• Taking small equal portions at random or at regular intervals.
• This method is cheap and is quick.
• This is difficult to get full representation of all particle sizes.
• Applicable for particles that are less than 0.5" in uniform ores.
• Hand Sampling: Coning and Quartering is a standard method used in US.
• Applicable to practically all classes of ore, 50-ton lots
• Used in ores with <2" largest particle size.
• Hand Sampling: Shovel is a quick and cheap method.
• Every alternate, or every third, fourth, fifth, or tenth shovel-full is taken for the sample, depending upon the sample size required
• Applicable to much larger lots
• Not applicable for samples with big lumps (>2"); susceptible to ore segregation; subject to manipulation
• Hand Sampling: the Trench Method.
• Consists of cutting trenches along a flattened cake of ore for the samples
• Good for sample material degree of mixing.
• Hand Sampling: Pipe and Auger Sampling uses pipes with a sharpened end and handles in the other end.
• Pipes should be long enough.
• The pipes used are long enough to reach the bottom of heap.
• This has low cost and quick results
• Errors due to ore clogging and insufficient mixing

Machine Sampling

• Devices designed to substitute mechanical processes for the undesirable human element in hand sampling
• Lessens or eliminate accidental and intentional errors due to personal factors
• Shortens the time required for sampling and minimizes operating labor
• Machine sampling can be Stationary and Moving

Sampling Methods - Stationary Machine

• Whistle-pipe sampler uses Vertical pipe with notched openings cut hallway through the pipe.
• Each opening spaced 90 degrees horizontally from the preceding
• Cheap to install quick reduction in bulk with no power needed
• This however suffers Ore clogging and segregation due to metal wear
• Jones Riffle Sampler is also a Stationary machine that allows for Continuous diversion of ore.
• Using an even number of equally sized chutes, adjacent chutes discharging at opposite ends
• Used for cutting down large samples to assay weight, although;
• Largest ore particle size should be smaller than chute width by at least a factor of 3

Sampling Methods - Moving Machine

• Snyder Sampler consists of a pan-shaped plate with one or more sample spouts for stream sample cuts from a feeding chute
• It’s simple, easy to maintain but it cant handle sticky ore streams
• Umpire Sampler.
• Used for cutting-down dry samples for assay
• Consists of two counter-rotating buckets placed one over the other, each divided into 4 compartments with alternating openings