Chelating Agents for Heat Exchanger and Evaporator Descaling

A heat exchanger tells you it is fouled in a way that is hard to argue with. The approach temperature widens, the steam valve opens further for the same duty, and the product outlet drifts below specification. By the time someone writes the work order, the plant has usually paid for several weeks of extra energy and, in an evaporator, some loss of capacity that shows up as a shorter run between washes. Descaling is the correction; how it is done decides whether the metal survives it.

Fouling is a heat-transfer problem before it is a cleaning problem

Scale on a heat transfer surface is not just dirt sitting on metal. It is a mineral layer with a thermal conductivity far below that of the tube wall, so a film measured in tenths of a millimetre changes the overall heat transfer coefficient noticeably. The economic loss is easy to state: the same duty needs more steam, or the same steam delivers less duty, and the gap widens as the layer thickens.

That also explains why cleaning is judged on performance rather than appearance. A bundle that looks clean after a wash but still shows a wide approach temperature was not cleaned, it was rinsed. The verification section below deals with how to tell the difference before the unit goes back into service.

Reading the deposit: carbonate, sulphate, silicate, organic

Descaling starts with knowing what the layer is made of, because the four common groups behave differently in a cleaning solution.

Carbonate scale, mostly calcium carbonate with some magnesium, is the easiest. It dissolves readily in acid and softens in a chelant solution. Sulphate scale, calcium sulphate in its various hydration states, is the difficult one, because its solubility is low and it does not respond to the acid strengths that plants use for carbonate. Silicate and aluminosilicate deposits are a third case: they are not dissolved in the usual sense, they are broken up and dispersed, and they need alkaline conditions and a dispersant rather than acid. Organic and biological films sit on top of all three and are removed by caustic with a surfactant.

A plant can identify the layer without a laboratory. A sample of the deposit, dried and weighed, loses its organic fraction on ignition. The remainder fizzes in dilute acid if carbonate dominates and stays quiet if sulphate or silicate dominates. A drop of acid on a sulphate-rich layer often produces a white cloudy suspension rather than brisk effervescence. Ten minutes with a sample and a dropper bottle saves a wasted cleaning cycle.

Why acid alone is fast and why plants still add a chelant

Acid is the direct answer to carbonate, and no chelant dissolves calcium carbonate faster than a dilute mineral or organic acid at the right temperature. The reason plants add a chelant to the circuit is not speed, it is control.

The first reason is iron. Many deposits contain iron oxide or iron carbonate, and iron behaves badly during an acid clean: as the acid is consumed and the pH rises, dissolved iron can precipitate as hydroxide or re-deposit as oxide on the cleanest surfaces in the circuit, which are often the ones you most wanted to protect. A chelant holds the iron through the pH swing.

The second reason is the rinse. Acid cleaning leaves a surface that is momentarily very reactive, and hardness in the rinse water can lay down a thin film before the unit is back in service. A chelant in the final rinse keeps that hardness in solution.

The third reason is the mixed deposit. Where carbonate sits over sulphate or silicate, the acid stage removes the top layer and exposes the rest, and a chelant in the same circuit begins to work on the exposed mineral instead of letting it settle elsewhere in the loop.

For an overview of how the different agents compare on binding strength and alkaline stability, the article on GLDA versus MGDA versus IDS sets out the practical differences.

Plate exchangers, shell-and-tube units and evaporators

The cleaning method follows the geometry, not the chemistry.

A plate exchanger is the most forgiving to clean and the least forgiving to damage. The pack is opened, plates are inspected, and cleaning is often done by circulation in a purpose-built rig or by immersion. Gaskets set the temperature and chemical limits: elastomer gaskets tolerate a narrower range than the stainless plates, and a hot alkaline soak that is fine for the steel can soften or swell a gasket. Plates are also thin, so mechanical cleaning with a scraper is a last resort.

A shell-and-tube unit is cleaned in place, usually by circulating through the tube side. The constraint is velocity distribution: a large bundle has tubes that see more flow than others, and the slowest tubes are the ones that stay fouled. Alternating the circulation direction and allowing a soak period helps, and so does accepting that a badly fouled bundle may need a second cycle rather than a stronger solution.

An evaporator adds a further constraint: the deposit forms on the heated surfaces at temperature, some of it is not pure mineral scale but a baked-on organic-mineral composite, and the equipment often has tight clearances and small nozzles. Cleaning has to remove the layer without attacking the tube surface, and the rinse has to be thorough because any residue ends up in the product. In dairy and food evaporators the cleaning regime is also constrained by the hygiene requirement, which pushes the programme towards alkaline and acid stages with food-contact-compatible chemistry.

Setting the concentration, temperature and contact time

The instinct is to raise concentration when a clean underperforms. In practice, contact time and temperature move the result more, and concentration is the variable most likely to cause damage.

For chelant-assisted descaling, a working solution in the range of 0.5 to 2 per cent by weight is typical, with the higher end used for short circulation and heavy load. Temperature is capped by the weakest material in the circuit, which is usually the gasket or a plastic component rather than the metal. Contact time is set by monitoring rather than by the clock: the solution is sampled during circulation, and the clean is continued while the dissolved metal concentration is still rising.

Two practical rules come out of this. Circulate before soaking, so the solution reaches the whole surface before it starts working; and keep the loop vented, because a closed circuit with active acid produces gas and a pressure build-up that has to go somewhere.

Material compatibility and the corrosion watch

Descaling puts metal in contact with an aggressive solution for hours, which is the opposite of normal service. The risks are specific and predictable.

Carbon steel is fine in alkaline chelant solutions and is attacked by acid, so acid stages need an inhibitor and a time limit. Austenitic stainless steels are attacked by chlorides under acid conditions, which is why the chloride content of the chelant grade matters and why hydrochloric acid is a poor choice for a stainless circuit. Aluminium and its alloys are attacked by strong alkali as well as acid, so temperature and concentration both need to be conservative. Copper alloys sit in the middle: they tolerate many chelant solutions but not ammonia or amine-based additives.

The check that catches problems early is a coupon. Put a weighed coupon of the circuit's main metallurgy in the cleaning loop, weigh it afterwards, and calculate the corrosion rate. A rate that looks small over four hours becomes a wall thickness loss over a decade of annual cleaning, and the coupon is the least expensive way to see that before it becomes a leak. The companion note on chelant corrosion risk across plant metallurgy covers the mechanisms in more detail.

Verifying the clean

Three readings decide whether the unit can go back into service. The approach temperature is the direct measure: at a defined flow and duty, the difference between the hot side outlet and the cold side inlet should return close to the commissioned value. The pressure drop across the unit should return to baseline, and in an evaporator the achievable concentration or the run length between washes should recover.

If the approach temperature improves but does not return, the clean was partial. If it returns and then degrades within days, the deposit is reforming faster than the programme assumes, and the question moves upstream to water quality, antiscalant dose or operating temperature.

Waste, neutralisation and discharge

A spent descaling solution is a concentrated metal-bearing waste, not a rinse. It carries the metals that were on the surface, it is at a pH that will not pass a discharge consent, and its volume is small enough to handle deliberately. Neutralise to the permitted range, let the metal hydroxide settle, and dispose of the sludge through the route your permit allows. In many plants the spent solution goes to the trade effluent system at a controlled rate rather than to surface water.

The chelant itself matters here. EDTA in a spent solution keeps metals in solution and defeats the settlement step, which makes the waste harder to treat. Agents designed to mineralise under OECD 301 conditions behave better in a biological treatment plant, and that is increasingly the reason a plant switches rather than the price per kilogram.

Frequently asked questions

Can a chelant descale a calcium sulphate deposit on its own? Only slowly. Sulphate scale has a low solubility ceiling and needs either a converter stage or a long soak with a sequestrant, and sometimes the honest answer is mechanical removal.

How do I know the acid stage is finished? Monitor pH and dissolved calcium in the circulating solution. When the pH stops rising and the calcium curve flattens, the reaction is complete. Adding more acid after that point only attacks the metal.

Is a food-grade chelant needed for a dairy evaporator? The cleaning chemical should be suitable for food-contact surfaces and the rinse must be verified. Ask for the documentation rather than relying on the word food-grade on the label.

Does cleaning damage gaskets? It can. Check the elastomer's temperature and chemical limits before choosing the programme, and inspect gaskets after the first clean rather than after the fifth.

How often should a plate pack be opened? At least once a year for inspection, and whenever a clean fails to restore the approach temperature. Circulation cleans the surface it can reach; only opening the pack shows the rest.

For a descaling programme review, a deposit sample assessment or a bench trial on your own scale, write to [email protected] or call +86-537-3739818. Product data for water treatment and chelating agents is listed at yuanlianchem.com, with mechanism notes on the technical site at yuanlianchelating.com.