D5 · Publication Volume 21

Mineral Processing, Metallurgy and Mining Economics Foundations

Connects mineralogy and ore variability to processing performance, metal accounting and economic decisions.

Purpose and boundary of this book

This book explains how mineral properties become process responses, products, wastes and economic consequences. Its purpose is to help geoscientists, engineers, data practitioners and decision makers exchange evidence across the boundary between an orebody model and a processing value chain. It develops the reasoning needed to ask whether a mineral can be liberated, separated or dissolved; whether a sample supports that conclusion; how mass and contained value move through a flowsheet; and how uncertainty changes a decision.

The book is not a process design, operating procedure, product contract, valuation, investment recommendation or substitute for competent laboratory, engineering, environmental, commercial and financial review. A real flowsheet requires representative material, verified methods, current product requirements, safe operating envelopes, site-specific water and residue studies, and professionally accountable approvals. Equations and diagrams here expose relationships; they do not certify equipment, recovery, scale-up or commercial viability.

General and institution-neutral scope

This is a general, institution-neutral tutorial. It has no relationship to, affiliation with, sponsorship by, endorsement from or curriculum dependency on any company or individual. It is not written for a named owner, operator, laboratory, consultancy, university, government programme, software product, commodity trader, property, deposit, mine or private database. Every unnamed ore, sample, test, stream, recovery, price, cost and decision in an example is synthetic teaching material.

The website only hosts and delivers this tutorial. It is not the publisher, scientific authority or subject of the curriculum, and hosting creates no technical, commercial or professional affiliation. Named organisations and authors appear only in source sections to identify traceable material. Their inclusion does not imply approval of this tutorial or make any jurisdictional requirement universal.

The integrated ore-to-value model

The organising chain is geological material → mineral assemblage and texture → liberation or exposure → unit-operation response → mass and component streams → product and residue quality → revenue, cost and risk → model feedback. Grade is one input, not the value model. Two parcels with the same assay can behave differently because the payable element occurs in different minerals, grain sizes or associations; hardness changes throughput; clay changes handling; an impurity reports to product; or recovery depends on a narrow operating window.

Each link has a support and an uncertainty. A polished section represents a tiny physical support. A variability test represents a prepared sample under a stated procedure. A process model predicts a stream over a time interval. A geological block represents a spatial volume. A discounted cash-flow scenario represents dated assumptions. Moving a value between these supports requires an explicit aggregation, prediction or allocation rule and a record of what information was lost.

Learning outcomes

After completing the volume, the learner should be able to:

  • translate mineralogy, texture, grain size and association into testable liberation and extraction hypotheses;
  • interpret crushing, grinding and classification without treating a single size metric as a complete particle distribution;
  • compare gravity, magnetic, dense-media and sensor-based separation using property contrasts and partition behaviour;
  • explain flotation through surface condition, collision, attachment, entrainment, froth transport, kinetics and selectivity;
  • structure a leaching and hydrometallurgical pathway using thermodynamic feasibility, apparent kinetics, solution chemistry and impurity control;
  • design a staged testwork programme that distinguishes representativity, repeatability, variability, scale and decision purpose;
  • define geometallurgical domains and prediction models without averaging nonlinear responses or hiding sparse support;
  • connect product specifications and deleterious elements to blending, penalties, rejection risks and residue consequences;
  • close mass and component balances, calculate recovery and identify measurement or inventory contradictions;
  • reason about cut-off, margin and net present value with explicit timing, units, scenarios and uncertainty; and
  • specify data contracts linking samples, tests, models, blocks, schedules and observed process performance.

Prerequisites, notation and units

Learners should be comfortable with minerals and rocks, geological domains, sampling, algebra, proportions, probability, uncertainty and basic discounted cash flow. Prior study of the scientific foundations, Earth materials and resource estimation volumes is useful. The tutorial uses dry mass unless moisture is stated, mass fractions rather than ambiguous percentages in equations, and internally consistent monetary units rather than a real currency.

For stream i, dry mass flow is M_i, component grade as a mass fraction is g_{i,j}, and contained component flow is C_{i,j}=M_i g_{i,j}. Recovery of component j to product p is R_{p,j}=C_{p,j}/C_{f,j}. Particle sizes such as P_{80} are defined by cumulative passing mass, not by visual diameter. Discounted value is written NPV=\sum_{t=0}^{T} CF_t/(1+r)^t. Every calculation must declare basis, units, period, moisture convention and treatment of inventory.

Synthetic teaching system

The recurring case is a fictional polymetallic mineralised system divided into three provisional domains. Domain North contains relatively coarse valuable sulfide grains and competent silicate gangue. Domain Central contains finer intergrowths, more variable hardness and a minor penalty-bearing mineral. Domain South contains partly oxidised material, more clay and a different extraction response. The names are spatial labels only; they do not describe a real site.

Synthetic samples progress through characterisation, size reduction, physical preconcentration, flotation or leaching alternatives, product and residue assessment, and an illustrative value model. Numerical values are deliberately simplified and are never default design factors. Every result remains tied to a sample identifier, preparation history, test method, conditions, uncertainty and applicable domain. Later lessons revisit earlier assumptions so learners can see how a new impurity result or recovery relationship changes the model.

Evidence artefacts and quality gates

Each lesson produces a controlled artefact: ore-to-product hypothesis, comminution data sheet, separation partition curve, flotation response surface, leach diagnostic, testwork matrix, geometallurgical domain register, product-quality ledger, reconciled mass balance, economic sensitivity sheet or geological-interface contract. A useful artefact states purpose, material identity, support, preparation, conditions, units, method version, detection or reporting limits, uncertainty, exclusions and review status.

The common gates are sample identity, representativity, method fitness, mass closure, component closure, calibration, repeatability, domain coverage, scale relevance, product acceptability, environmental boundary, economic basis and version lineage. Precision cannot repair biased sampling. A balance that closes only after an unexplained adjustment is not validated. A high recovery that produces an unacceptable product is not a successful result. A positive value in one deterministic scenario is not proof of robustness.

Assessment and completion standard

Completion requires an integrated synthetic study. The learner must compare two process pathways for contrasting domains, show at least one mass and component balance, explain why average grade is insufficient, propagate a material change through recovery and product quality, construct a transparent discounted-value comparison, and return appropriate attributes and uncertainty to a geological model. The submission must distinguish measurement, prediction, assumption and decision.

A passing record is reconstructable and falsifiable. Another reader can identify the material tested, reproduce the formulas, inspect rejected data, locate each assumption, determine what evidence would reverse the conclusion and see where qualified review is required. Unsupported recovery factors, hidden product penalties, current-looking prices without dates, unnamed inventory adjustments or domain labels without test support fail the standard.

Core sources