Lot Numbers and the Complaint That Arrives in Month Nine

The email arrived in week 38: the gel has turned to mush, the client is withholding payment, and the contractor wants to know what went wrong. The site had received four deliveries over five months, all from the same supplier, all the same grade. Nobody could say which delivery went into which block. The bags had been stored in a steel shed that reached 47 °C in July. By the time anyone thought to test, the only material left was in the ground.

Nine months after application, a performance complaint is almost impossible to resolve without records. With records it takes about a week, because the question narrows quickly from “your product failed” to a specific lot, a specific storage condition or a specific application fault.

What a lot number can and cannot tell you

A lot number identifies a production run: the reactor charge, the date, the drying conditions and the QC result. If you kept a 500 g sample from that run, you can retest it today and know whether the material left the factory within specification. That is the entire evidential value, and it is enough to settle most disputes in one direction or the other.

What a lot number cannot tell you is what happened after delivery. Storage above 40 °C for weeks, moisture pickup in a humid shed, and mechanical damage from double-stacking all leave the polymer weaker without changing its identity. This is why the retained sample has to be stored properly, in a sealed jar, out of sunlight, at moderate temperature. A sample kept in a site container for nine months proves nothing except that the container gets hot.

The minimum record set

Five records, and four of them cost nothing but discipline. The temperature log can be a 15 data logger in the shed; without one, a complaint about a hot summer becomes an argument about whether the shed was hot.

Reading a month-nine failure

Three signatures are worth knowing. If the gel is soft, cloudy and disintegrating, suspect heat and moisture during storage, or a lot at the low end of crosslink density. If the gel is hard, shrunken and glassy, suspect repeated drying without re-wetting, which is an irrigation fault rather than a product fault. If the gel is simply absent from the block, suspect the application, not the material.

Take samples from three depths in three pits per block, photograph them next to a scale, and put half of each in a sealed bag for the laboratory. Then pull the retained sample for the lot that went into that block and test both in the same water on the same day. That side-by-side is the whole case.

In four of the six month-nine complaints we have been involved with, the retained sample tested within specification and the field sample did not. The product was fine at the factory gate and degraded on site. The contractors who had storage logs won those arguments; the ones without logs paid, even when the storage had obviously been the problem.

The reverse also happens, and it is worth being fair about. One lot in 2023 left the plant at 6.9 percent moisture and 240 g/g, passed QC, and performed poorly in the field because the crosslink was at the bottom of its tolerance band. The buyer had a retained sample, the retest confirmed the number, and the supplier replaced the lot. Without that jar, the conversation would have been about the weather.

Set it up once

Buy a box of 500 ml jars, a marker and a cheap logger. Sample every delivery, label with lot, date and initials, and store the jars in an office rather than a container. Add the lot number to the application map as a matter of routine, and photograph the label on the pallet before it is cut open. The whole system takes ten minutes per delivery and it is the difference between a claim and a write-off. The lot coding and documentation we supply with each SAP powder shipment is designed to line up with this record set.