What goes wrong

The number is wrong and nothing looks wrong

Updated October 2026

On this page8

High results point at leaching Low results point at adsorption or permeation Run blanks in the containers you actually use A new supplier is a new variable Check that the container matched the method You cannot fix it afterwards Keep one container from each lot References

The defining failure here is a result that is wrong while everything looks right, and the container is one of the few places it comes from.

High results point at leaching

If a trace analyte reads higher than it should, ask whether the container could be supplying it. Sodium, boron and silica from glass, and trace metals or plasticizers from plastics, are the usual suspects. A blank run in the same container type answers it directly.

Low results point at adsorption or permeation

Metals lost onto the wall, which is what acid preservation prevents, or volatiles lost through a permeable bottle or past a poor seal. Both show as results that drift low with time between collection and analysis.

Run blanks in the containers you actually use

A method blank in a clean container from the same lot is the test that distinguishes a container problem from an analytical one. It is routine practice in regulated work and it is often skipped in informal work, where it would be most informative.

A new supplier is a new variable

Containers from a different source can have a different cleanliness and a different polymer formulation. If results shifted and nothing in the method changed, the box the bottles came in is worth looking at.

Check that the container matched the method

The simplest explanation is frequently that somebody grabbed the available bottle. Table II specifies the material per parameter, and a polyethylene bottle used for an organic determination is a documented reason for a low result rather than a mystery.

You cannot fix it afterwards

There is no correction to apply and no way to recover the true value. The sample is spent and the answer is to recollect, which is why the container decision sits before the work rather than inside it.

Keep one container from each lot

If a container problem is ever suspected, an unused bottle from the same lot is what lets you test the hypothesis. Without it the lot is gone and the question stays open.

References

40 CFR 136.3, Table II — Required Containers, Preservation Techniques, and Holding Times, which prescribes the container material, the preservation and the maximum holding time per parameter. Footnote 1 defines the codes: “P” is polyethylene, “FP” is fluoropolymer (PTFE), “G” is glass, “PA” is any plastic made of a sterilizable material. law.cornell.edu

Written from federal regulation, not from a bench. Where a method specifies a container, a preservative or a holding time, the method governs and nothing here overrides it. Read the method you are running rather than a summary of what methods generally say.