.. _example-predominance-phreeplot-02: Cu-S-C Predominance =================== Copper in a water that also contains sulfate, carbonate and chloride -- example 2 of the PhreePlot manual. Where :ref:`example-predominance-phreeplot-01` had one element to speciate, this system has several ligands competing for the same metal, and the diagram is correspondingly less tidy. It also makes a point that predominance diagrams make badly on their own, and so comes with two slices through it. The system ---------- .. raw:: html :file: study_02 - Cu-S-C System.html 0.1 mol/l each of copper and sulfate, in a 0.1 mol/l NaCl background at 20 °C, with CO2 fixed at a log fugacity of -3.5 -- roughly atmospheric. The starting pH of 1.8 is set just below the bottom of the grid, so that fixing pH always means adding base rather than acid. ``Fix_H+`` and ``O2(g)`` impose the two axes as before: pH 2 to 10, log fO2 from -80 to 0. The input carries a pointed comment, ``N.B. no minerals``: only aqueous species compete here. A copper system this concentrated would precipitate in reality, and leaving the solids out is a decision about what the diagram is asking, not an oversight. The diagram ----------- .. figure:: PredominanceDiagram_op_ex2_cp_.svg :alt: Predominance diagram of copper species against pH and oxygen fugacity :align: center Six copper species hold territory. Cu+2 occupies the oxidised, acid corner; Cu2(OH)2+2 and Cu(OH)2 follow as pH rises and the metal hydrolyses; Cu(CO3)2-2 takes the alkaline side, where carbonate out- competes hydroxide. Below the redox boundary copper is Cu(I), as Cu+ and as the chloride complex CuCl2-. That CuCl2- appears at all is the thing to notice. Chloride is here only as a background electrolyte, and it still wins a region of the diagram: Cu(I) binds chloride strongly enough that what was meant to be inert ionic strength becomes part of the chemistry. As before, the strips marked ``O2(g) > 0.21 atm`` and ``CH4(g) > 1 atm`` are outside what can occur. What the diagram cannot show ---------------------------- A predominance diagram shows the winner and nothing else. It gives no sense of by how much, and no sense of what the other species are doing -- and in a system with this many complexes, the dominant one may hold only a modest share. The two slices cut through the diagram at log fO2 = -63, in the reducing part, and plot concentrations directly. .. figure:: CuSpeciesSlice.svg :alt: Concentrations of all copper species against pH at log fO2 = -63 :align: center Every copper species along that line: eighteen of them, against pH. The predominance diagram showed two regions here. This shows the handover between them as what it is -- curves crossing, with several species within an order of magnitude of each other through the transition, and a whole population of minor complexes that never appear on the diagram at all. Read the two together. The diagram says where; the slice says how much, and whether "dominant" meant 90 % or 35 %. .. figure:: ClSpeciesSlice.svg :alt: Concentrations of chloride species against pH at the same slice :align: center The chloride species on the same line. Free Cl- dominates throughout, as it must at this ionic strength, and below it CuCl+, CuCl2- and CuCl3-2 trade places as pH changes. This is the quantitative version of the previous point: the copper- chloride complexes are a small fraction of the chloride but a large fraction of the copper. Which of those two statements you make depends entirely on what you divide by, and a diagram drawn per element cannot make both. Try it ------ * Move the slice to an oxidising fugacity and compare which species are in play. * Drop the chloride to 1 mmol/l and watch the CuCl2- region shrink or vanish. * Let a copper mineral into the equilibrium phases and see how much of the diagram survives. Source ------ * Kinniburgh, D. G. and Cooper, D. M. (2011). *PhreePlot: Creating graphical output with PHREEQC.* This is example 2 of the PhreePlot manual. See `the PhreePlot website `_. * The thermodynamic data is ``wateq4f.dat``, distributed with PHREEQC (Parkhurst and Appelo, 2013).