The Swell Number on the COA: Five Things It Leaves Out

Two technical data sheets land on the buyer’s desk. Grade A swells to 450 g/g in the certificate of analysis. Grade B claims 320. The price difference is 12 percent, the logic seems airtight, and the buyer orders Grade A. By August the field crew is describing Grade A as ‘slime that will not stay where we put it,’ and Grade B, trialled the following season out of frustration, outperforms it in every watering-interval measurement that matters.

The headline swell number is measured in deionised water, in a tea-bag test, with no load on the gel and no salts around it. It is close to the best possible condition the polymer will ever see, and it says nothing about the four or five properties that decide how the product behaves in a hole in the ground. A very high raw swell figure often signals a lightly cross-linked grade: enormous appetite for pure water, weak gel strength, poor behaviour under the weight of wet soil, and a tendency to collapse into mush through the season.

The five numbers to ask for instead

• Absorbency under load, or AUL: capacity with a weight pressing on the gel, which is what wet backfill actually does. A grade that holds 350 g/g free but 15 g/g under load will disappoint in any root zone.
• Swelling in saline solution, typically a standard salt solution: the number that predicts performance in real irrigation water and fertilised soil.
• Particle size distribution: fines dust away and can crusting; over-size granules hydrate slowly and sit unused in a short season.
• Residual monomer content on the certificate of analysis: the compliance number, and a quick marker of how well the plant is run.
• Moisture content of the powder as supplied: you are paying for water if this runs high, and dosing by volume drifts.

The counter-intuitive rule buyers learn eventually: within a supplier’s range, the best field grade is rarely the highest swell number. The grade that keeps working through a second dry cycle, drains rather than soups, and re-wets after drying is almost always a middle-of-the-range swell figure with good load numbers. Field trials sorted this out on the two-grade comparison above in one season: Grade B’s lower raw figure came with roughly double the under-load capacity, and under load is where the tree roots live.

How to read a certificate without being fooled

Check three things on any certificate of analysis: that the reported tests match the batch numbers on the sacks delivered, that the methods are named rather than just numbers quoted, and that the saline and under-load figures exist at all. A certificate with only a deionised swell number and a moisture line is telling you the supplier expects to be compared on price. Asking for the fuller panel is not being difficult; any serious producer produces it routinely, and the product specification pages for the SAP powder grades show what a complete data set looks like when a supplier has nothing to hide.

A worked comparison from one tender

One municipal buyer re-scored two shortlisted grades on the five-line panel instead of the swell headline. Grade A, the 450 g/g specimen, held 90 g/g in saline and 14 under load, with a fines fraction near 12 percent. Grade B, the quiet 320, held 210 g/g in the same saline solution and 31 under load, with fines under 4. On the buyer’s own watering-interval model, Grade B delivered roughly twice the field days per kilo despite the smaller number on the front page.

The exercise changed more than the order. The buyer’s spec template now requires the full panel for any shortlisted grade, and the tender scoring weights the under-load and saline figures above the raw swell number, which quietly removed the datasheet vanity contest from every subsequent bid. Suppliers adjusted within one tender cycle; a spec that asks better questions gets better answers.

The practical purchasing sequence costs one small trial: two candidate grades, ten paired trees each on representative soil, one season, scored by watering interval and spade test rather than by datasheet. It replaces the swell-number beauty contest with something a crew can defend. Keep the certificates in the job file either way, because when a grade underperforms, the first question any consultant asks is which numbers the buyer actually had before the order went out.