HPMC Viscosity Does Not Match the COA: What to Check Before Rejecting a Lot

A receiving laboratory reports a different HPMC viscosity from the supplier’s certificate of analysis (COA). Before making a rejection decision, establish what the two numbers actually describe: the same lot, the same solution definition, and the same written test method.

Start by classifying the evidence as unknown, different, invalid, or matched. Missing settings call for records; different methods call for alignment; an invalid measurement calls for investigation. Only then can a valid result be assessed against the applicable viscosity limit.

Use this checklist to structure a technical discussion between supplier and receiving laboratory. It is not a laboratory SOP, a contractual acceptance rule, or a substitute for your approved quality procedure.

Why the method label belongs beside the number

Consider a published product specification, rather than a disputed batch. LANDU’s HPMC K100 technical specifications list two viscosity ranges:

The linked K100 technical data sheet gives the same method-labelled ranges. These are manufacturer specifications, not paired measurements from a real lot or evidence of a customer complaint.

The practical implication is to compare a receiving-lab result with the limit attached to its agreed method. These ranges do not establish a universal NDJ-to-Brookfield conversion factor. Dividing their endpoints would not validate a conversion for another grade, setup, or sample.

The published viscosity entries identify the instrument family, concentration and temperature, but leave other method details unstated. Ask for the full written procedure rather than filling those gaps with familiar laboratory defaults. In particular, “2% solution” alone does not tell the reader which concentration basis the supplier used.

Build one record for both laboratories

LANDU laboratory technician handling a powder sample with a metal funnel
Laboratory sample handling at LANDU. This photograph illustrates laboratory work; it does not show a viscosity measurement.

Begin with the documents that identify the received material: grade, lot number, sample identifier, COA number and the purchasing specification or agreed TDS revision. Record which viscosity limit and method apply. If those identities do not line up, resolve that discrepancy before comparing readings.

For measurement comparability, the Brookfield DVNext manual, sections 3.2 and 3.4 identifies instrument, spindle, speed, vessel, temperature and test duration. It also calls attention to the container and whether a guard leg is fitted. This is general instrument guidance; it does not supply the missing K100 test settings.

Use this worksheet to record both laboratories’ conditions side by side. Enter the supplier’s written requirement beside what the receiving laboratory actually did. Mark each row matched, different, or unknown; leave a missing value unknown until someone supplies evidence. Add an owner and next action for every unresolved row.

CheckSupplier method or COA recordReceiving-lab actual recordComparison and next action
Identity and limitGrade, lot, sample, COA and specification revisionReceived material and sampled lotConfirm identities and applicable limit
Solution definitionConcentration and stated mass/volume, dry or as-received basisActual powder/water quantities and calculation basisResolve missing definitions
PreparationProcedure revision, dissolution and conditioning sequenceMixing, rest/storage history, visible lumps or airRecord deviations for technical review
TemperatureRequired measurement temperatureActual sample temperature at readingCompare the recorded conditions
Instrument setupModel/range, spindle, speed, vessel/volume, guard-leg stateActual setup and accessoriesIdentify every setup difference
Reading and validityReading time, units and applicable validity checksIndividual readings, timing, instrument status and check recordsRetain evidence; flag invalid or unresolved results

Classify the discrepancy before choosing a retest

The four outcomes below suggest what to check or request next. Apply your agreed procedure when deciding sample handling, investigation, retest authorization and lot disposition.

Unknown: the record does not establish comparability

For example, a COA might state viscosity and temperature while the accompanying record omits spindle or speed. Do not treat those blank fields as agreement with your own setup. Ask the supplier for its complete method and ask the receiving laboratory for its actual test record.

Resolve the concentration basis just as explicitly. Record how the solution was calculated and prepared; do not silently assume that both laboratories interpreted a short concentration label the same way. Keep the comparison open until the missing information is obtained or the responsible quality team determines how to proceed.

Different: the laboratories used different conditions

Compare the differences with the written method before scheduling another run. Brookfield’s temperature and shear-rate guidance explains that these conditions can affect the accuracy and reliability of viscosity results. It does not provide an HPMC correction coefficient.

For a useful comparison, agree on one complete procedure. Matching rpm alone does not demonstrate that two instrument and spindle arrangements are equivalent. If different instruments must be used, request a documented correlation acceptable to both parties, including its applicable materials and conditions. Do not turn a product specification table into that correlation.

Invalid: a result fails the applicable test-validity checks

Keep method differences separate from an invalid measurement. A result should be classified as invalid when the applicable instrument instructions or laboratory procedure establish that it cannot be used. An unfamiliar value alone is not that evidence.

Review the original record, instrument warnings and required checks. Record any preparation observations, such as persistent lumps or visible air, for investigation under the method. Where a repeat is authorized, retain the original reading, the reason for repeating and the conditions of the repeat. Do not average unlike procedures to make the discrepancy disappear.

Matched: compare the valid result with the agreed limit

Once identity, method conditions and validity are established, compare the result with the corresponding method-specific specification. A valid result within that limit addresses the viscosity requirement under those conditions. It does not automatically satisfy every other release requirement.

If a valid matched-method result is outside the agreed limit, send the complete evidence to the responsible quality team for disposition under the applicable procedure. Do not replace that limit with an invented percentage tolerance. This checklist supplies no universal repeat count, hydration period, spindle, speed or acceptance allowance.

Keep the technical query specific

Send a comparison package containing the grade and lot identifiers, COA, applicable specification revision, completed worksheet, written method, individual readings and relevant observations. State the unresolved question precisely: a missing preparation instruction, a setup difference, a validity concern, or an out-of-limit result after method matching. That gives both laboratories a concrete issue to investigate.

For LANDU materials, the quality-control and assurance team describes customer-specific QC reporting and technical troubleshooting support. Use that contact to request the relevant method documentation and discuss the records you have collected.

Finally, keep the complaint itself in view. If the concern is mortar open time, workability or another application property, assess that property through the separately agreed application tests. For the broader qualification discussion, see how to evaluate an HPMC supplier for dry-mix mortar applications. Resolving the COA comparison should produce a documented viscosity finding; the appropriate quality team can then combine it with the other evidence needed for the lot decision.