Question 8 of 8: Environmental protection

Question 8: Environmental protection (10 marks)

Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.

Two descriptions earn the marks; all four are given here so that the answer covers the whole of the examinable material. Under Canadian practice the governing instruments are the federal Reduction of Carbon Dioxide Emissions from Coal-fired Generation of Electricity Regulations, the Canada-wide Standards for particulate matter and ozone administered through the CCME, and provincial approvals issued under the applicable environmental protection Act.

Method 1 — collection and handling of fly ash

About 80 per cent of the ash in a pulverised-coal furnace leaves entrained in the flue gas as fly ash: for the 600 MW plant of Question 5 that is roughly 190 tonnes a day of particles from below one micrometre to about one hundred micrometres. The dominant collector is the dry electrostatic precipitator, because it removes even the fine fraction at a pressure drop of only 100 to 250 pascals, which matters when the induced-draught fan is handling a million cubic metres an hour.

The physical principle is electrostatic drift. Rows of thin discharge electrodes are held at 40 to 70 kilovolts negative direct current between earthed collecting plates spaced 250 to 400 millimetres apart. The field near the wire exceeds the breakdown strength of the gas and a corona forms, filling the inter-electrode space with free electrons and negative ions. Particles passing through acquire charge by two mechanisms: field charging, in which ions follow field lines onto particles larger than about one micrometre, and diffusion charging, in which thermal ion motion charges the sub-micrometre fraction. A charged particle then experiences a Coulomb force balanced by Stokes drag and drifts sideways at a terminal migration velocity w of a few centimetres per second, proportional to the particle radius and to the square of the field strength. Integrating that drift along the passage gives the Deutsch-Anderson relation, η = 1 − exp(−wA/Q), where A is the collecting area and Q the volume flow: efficiency depends exponentially on the specific collecting area, so 99.5 per cent collection is reached by making the casing large rather than by increasing the field. Ash resistivity is the critical variable. Between about 10⁸ and 10¹¹ ohm-centimetres the layer discharges freely; above that it holds its charge, the layer surface potential rises and back-corona destroys performance, which is why low-sulphur coals are often conditioned with a few parts per million of sulphur trioxide.

The equipment is a wide, low-velocity casing — the gas is slowed to about one metre per second by an inlet plenum with perforated distribution screens — divided into three to five independently energised fields in series, each with its own transformer-rectifier set and automatic voltage control that hunts just below the sparking limit. Collected ash builds into a layer several millimetres thick and is dislodged by rappers, tumbling hammers or electromagnetic impactors that strike the plates on a programmed sequence, starting at the outlet field where re-entrainment matters least. The ash falls into pyramidal hoppers fitted with heaters and level probes and is withdrawn through rotary or dome valves into a dry pneumatic conveying system that carries it to silos. From there it is either sold — a good low-carbon fly ash is a valuable supplementary cementitious material under CSA A3001 and displaces Portland cement, saving carbon dioxide in its own right — or conditioned with water and landfilled in a lined, monitored ash cell. Where the ash is very fine or resistivity is unfavourable, a pulse-jet fabric filter is used instead, which achieves lower outlet loadings independently of resistivity at the cost of a much higher pressure drop and periodic bag replacement.

Method 2 — capture and extraction of sulphur dioxide

Sulphur in coal burns almost entirely to sulphur dioxide, which is the principal precursor of acid deposition and of secondary sulphate particulate. A coal at 1 per cent sulphur burned at 46 kg/s releases about 0.9 kg/s of sulphur dioxide, and Canadian approvals now require 90 to 95 per cent of it to be removed. The dominant technology worldwide is wet limestone scrubbing with forced oxidation, used on roughly 85 per cent of installed desulphurisation capacity because limestone is cheap and abundant and the product is saleable.

The chemistry is acid-base neutralisation in the aqueous phase followed by oxidation. Sulphur dioxide is absorbed into an alkaline slurry droplet, where it hydrolyses to bisulphite; limestone dissolves and neutralises the acid, precipitating calcium sulphite hemihydrate and releasing carbon dioxide; and air blown into the reaction tank oxidises the sulphite to sulphate, which crystallises as gypsum. The overall reactions are SO₂ + CaCO₃ → CaSO₃ + CO₂ and CaSO₃ + ½ O₂ + 2 H₂O → CaSO₄·2H₂O. Because absorption is gas-film controlled, the rate depends on interfacial area and on keeping the droplet surface alkaline; the slurry pH is therefore held near 5.5 by limestone addition, since a higher pH scales the internals with calcium carbonate and a lower one stalls the absorption.

The equipment is a large open spray tower, ten to eighteen metres in diameter on a unit of this size, lined with rubber or flake glass and standing directly in the flue gas path downstream of the precipitator. Flue gas enters near the bottom, passes upward through three to five spray levels fed by dedicated recirculation pumps, and leaves through a two-stage mist eliminator that returns entrained droplets. The liquid-to-gas ratio is 10 to 20 litres per cubic metre, which gives a droplet surface area of hundreds of square metres per cubic metre of tower and a gas residence time of three to five seconds. Below the sprays, the integral reaction tank holds several hours of slurry inventory with agitators and oxidation air lances; a bleed stream goes to hydrocyclones and a vacuum belt filter that produce gypsum at 90 per cent solids for wallboard manufacture or cement retarder, with the filtrate returned to the tank. The cleaned gas leaves saturated at 50 to 60 °C and is either reheated, passed through a gas-to-gas heat exchanger, or sent to a wet stack designed for condensate. The penalties are real and must be quoted: 1 to 2 per cent of plant output for the recirculation pumps, oxidation blowers and the extra draught, plus significant make-up water. Where water is scarce or sulphur low, a spray dryer absorber using lime slurry, or dry sorbent injection ahead of a fabric filter, achieves 70 to 90 per cent removal with a dry, landfillable product instead.

Method 3 — reduction of nitrogen oxides in burners and exhaust gas

Nitrogen oxides from a coal furnace come from two sources. Thermal NO_x is formed by the Zeldovich mechanism, in which atmospheric nitrogen is attacked by oxygen atoms in the hottest part of the flame; its rate has an activation energy of roughly 300 kJ/mol, so it is negligible below about 1500 °C and rises exponentially above it. Fuel NO_x comes from nitrogen chemically bound in the coal, which is released with the volatiles and oxidises readily; on a bituminous coal it accounts for the majority of the total. The two mechanisms respond to the same lever, oxygen availability at high temperature, which is why combustion modification is always the first line of attack.

A staged low-NO_x burner delivers the pulverised fuel with only part of the combustion air, creating a fuel-rich primary zone with a local equivalence ratio below unity. In that reducing atmosphere the volatile nitrogen is driven off as ammonia and hydrogen cyanide, which react with any nitric oxide already formed and reduce it back to molecular nitrogen rather than oxidising it; at the same time the flame is longer, cooler and less luminous, which suppresses the thermal route. Secondary and tertiary air is admitted through swirl registers further out to complete burnout without ever creating a high-temperature, oxygen-rich zone. Overfire air ports above the burner belt extend the same idea to the whole furnace, holding the burner zone at 80 to 90 per cent of stoichiometric air and admitting the balance high in the furnace where the gas has already cooled. Flue gas recirculation into the windbox lowers the peak flame temperature by dilution. Together these measures cut emissions by 40 to 60 per cent, but they trade against unburned carbon in the ash and against furnace-wall corrosion in the reducing zone, so there is a practical floor.

Where a lower limit must be met, post-combustion treatment is added. Selective catalytic reduction injects ammonia or urea into the flue gas upstream of a catalyst bed, where it reacts selectively with the nitrogen oxides: 4 NO + 4 NH₃ + O₂ → 4 N₂ + 6 H₂O. The catalyst is vanadium pentoxide with tungsten trioxide on a titanium dioxide support, extruded as a honeycomb or coated on parallel plates with channels wide enough to pass fly ash, and it is installed as two or three layers in a large box between the economiser outlet and the air heater, where the gas is at the 320 to 400 °C window in which the reaction is fast but ammonia oxidation is not. An injection grid with static mixers must deliver the ammonia to within a few per cent of uniformity, because the ammonia that does not react passes through as slip and forms ammonium bisulphate that fouls the air heater. Removal of 80 to 90 per cent is routine. Selective non-catalytic reduction, injecting the same reagent directly into the furnace at 870 to 1100 °C without any catalyst, is far cheaper but achieves only 30 to 50 per cent and demands careful control of the injection temperature window.

Method 4 — capture and disposal of carbon dioxide

Carbon dioxide is not a trace contaminant but a bulk product: the plant of Question 5 emits over four thousand tonnes a day of it, and unlike sulphur dioxide it cannot be converted to a saleable solid at any plausible cost. Three routes exist — post-combustion capture from the flue gas, pre-combustion capture after gasifying the fuel to hydrogen and carbon dioxide, and oxy-fuel combustion in oxygen and recycled flue gas to give a nearly pure carbon dioxide stream — but only the first can be retrofitted to an existing station, and it is the one demonstrated at commercial scale in Canada at SaskPower’s Boundary Dam Unit 3.

The chemistry is reversible chemical absorption. Flue gas at 40 to 60 °C is contacted with an aqueous alkanolamine, classically monoethanolamine at 30 per cent but now proprietary sterically hindered blends, and the carbon dioxide reacts to form a carbamate and a protonated amine: CO₂ + 2 R-NH₂ ⇌ R-NH₃⁺ + R-NH-COO⁻. The reaction is exothermic and therefore favoured at low temperature; raising the solution to 110 to 120 °C in a stripper reverses it and releases nearly pure carbon dioxide. Because the flue gas is at atmospheric pressure with only 12 to 14 per cent carbon dioxide, a chemical solvent is essential — a physical solvent obeying Henry’s law would need a partial pressure an order of magnitude higher.

The plant is a large absorber-stripper pair. Flue gas is first cooled and polished, since amines are degraded by sulphur dioxide, nitrogen dioxide and fly ash, so essentially complete desulphurisation is a prerequisite. It then rises through a packed absorber against descending lean solvent; treated gas leaves the top through a water wash that recovers amine vapour. Rich solvent is pumped through a lean-rich cross exchanger, which recovers most of the sensible heat, into the stripper, where a steam-heated reboiler drives off the carbon dioxide; the overhead is condensed to remove water and the product gas is compressed in four to six stages, dehydrated and delivered at 10 to 15 MPa as a dense supercritical fluid. Disposal is by pipeline to a permanent sink: a deep saline aquifer beneath an impermeable caprock, a depleted oil or gas reservoir, or an enhanced oil recovery operation, in every case with monitoring, measurement and verification obligations that persist long after injection stops. The dominant engineering issue is the energy penalty. The reboiler alone consumes 3 to 4 MJ of low pressure steam per kilogram of carbon dioxide, and with the compression duty added the parasitic load removes 8 to 12 percentage points of plant efficiency — enough to turn the 37 per cent plant of Question 5 into a 27 per cent one and to raise its fuel burn, its cooling water demand and its ash production by a third for the same sent-out power.