HPLC testing laboratory · Davis, CaliforniaPublished pricesBook onlineinfo@openchemworks.com
Insights

How much sample do labs actually need? Amounts by sample type

Published October 20, 2026Updated November 19, 20266 min readOpenChemWorks Laboratory
Short answer

Send 200 µL of a liquid such as media, broth, wort, juice or a drink (400 µL preferred), 15 µL or more of plasma or urine, or at least 100 mg of a solid. The instrument measures a small prepared portion; the rest lets the lab prepare it properly and repeat a preparation if something looks off. For coarse or blended powders, particle size matters more than the label amount, so tell us what you have and we will confirm the amount before you ship.

Amounts by sample type

Sample typeSend per sampleNotes
Liquids: cell culture media, broth, wort, juice, drinks200 µL, 400 µL preferredMix well. Simple liquids that only need spinning down can run in larger numbers overnight.
Plasma and urineFrom 15 µLResearch use only, not for clinical diagnosis. Note freeze and thaw history.
Fine powder that dissolves in water or bufferAt least 100 mgNo prep charge for powders that simply dissolve.
Fine powder that needs extraction (peptones, protein powders, flours)At least 100 mgPREP-1 adds $30 per sample for weighing, extraction and filtration.
Coarse powder, granules, ground feed, seeds, plant materialAt least 100 mg for the analysis; tell us the particle sizeThe lab grinds and composites before taking its portion. PREP-1 applies. Fewer samples fit overnight.
Blends with visible componentsAt least 100 mg, plus a note naming the minor componentsThe minor component sets the sampling error. See the table below.
Gummies, capsules, bars, thick saucesAt least 100 mg after homogenisingWe confirm the preparation when you tell us what the product is.

Label each tube with the sample ID exactly as it appears on your sample sheet. Book online first; we confirm the start date by email and send the sample sheet and shipping address. Overnight turnaround available, with expedited service on request. Standard results in 3 to 5 business days from receipt. Tell us your sample count and type, and we will confirm timing.

Order samples See an example report

Why the measured portion is small

In AQC (AccQ-Tag) amino acid analysis with HPLC and dual UV and fluorescence detection, a portion of your sample is dissolved or extracted, the amino acids are chemically tagged so they can be detected, and a small volume of that prepared solution is what passes through the instrument. The 19 amino acids plus ammonia in the free panel, AA-1, are measured from that preparation against a calibrated standard. Cysteine is not part of that panel; it needs a separate AA-3 workup.

If the instrument consumes so little, why send more? Because measurement is the last step. Weighing, dissolving a powder evenly, diluting a strong or salty sample into range, and filtering a sauce all consume sample, and each step must be done well or the small final portion will faithfully misrepresent your product. Extra sample also allows a repeat preparation from retained material when a result does not fit the profile, and, where you commission it, a spiked portion to check recovery in a new matrix (PREP-2, $200 per assay, once per formulation, waived at 20 or more samples).

For powders, particle size sets how much is representative

A blend that is not perfectly mixed differs at the top of the jar from the bottom. The table below is a counting model, not a measurement: it asks how many particles a subsample holds when the component you care about is 5% of the particles. Particle mass scales with diameter cubed, so a coarse powder needs far more mass than a fine one to reach the same sampling error. The model uses an apparent particle density of 0.974 g/cm³ (a published value for whey protein isolate particles) and spherical particles.

Particle sizeRSD at 0.2 gRSD at 1 gRSD at 10 gMass needed for 1% RSD
60 µm (fine, like a whey isolate)0.32%0.14%0.05%0.021 g
250 µm (coarse powder)2.75%1.23%0.39%1.51 g
382 µm (coarsest commercial whey protein concentrate in the source)5.20%2.32%0.73%5.40 g

Worked example for 382 µm: particle mass = 0.974 x 0.5236 x (0.0382 cm)³ = 2.84 x 10⁻⁵ g. A 0.2 g subsample holds about 7,035 particles, and RSD = √(0.95 / (7,035 x 0.05)) = 5.2%. Real powders are size distributions, and segregation in shipping makes real error worse than this model, so treat the table as a floor. The practical lesson: grinding is the cheapest precision in the whole measurement, and a coarse powder sent with its particle size noted lets the lab grind and composite before it takes its portion.

Hydrolysis losses for total amino acids are usually single-digit percentages per amino acid, so for a coarse blend the subsampling error can be the larger one. If you have a segregating blend and the result will drive a batch decision, three independent subsamples reveal heterogeneity instead of hiding it. Ask us before you ship.

Sending it well

Use a clean, dry, sealed container for a powder and a leak-proof tube for a liquid, labelled with the sample ID from your sample sheet. Send the batch you mean to describe, well mixed, as sold or as eaten. Keep the batch identity for your own records so the report can be filed against the right batch. If you have very little of a precious sample, say so in your first message. If you have plenty, do not send a kilo hoping for a better result; more material does not improve a measurement limited by preparation and matrix.

Frequently asked questions

Can I send less than the amounts in the table?
Ask first. Tell us what you have and what question you need answered, and we will confirm what is feasible before you ship.

Why does the lab not just ask for the tiny amount the instrument uses?
Because preparation, representativeness, repeats and any spike-recovery work all consume sample before and beside the instrument portion. The measured volume is only the end of the process.

Does sending more sample make the result more accurate?
No. Beyond enough for good preparation and repeats, accuracy is governed by mixing, preparation, calibration and matrix behaviour, not by parcel weight. For coarse or blended powders, a larger well-mixed sample does help the portion represent the batch.

Is the amount different for a sauce?
A sauce is a liquid or a semi-solid. Send 200 µL to 400 µL of a well-mixed liquid, or at least 100 mg of a thick sauce, and tell us what it is so we can confirm the preparation.

References

  1. Waters: Derivatization of Amino Acids Using Waters AccQ-Tag Chemistry: Method background for AQC derivatization and HPLC-UV amino acid analysis.
  2. OpenChemWorks: Pricing: Published service and pricing copy (2026).
  3. Onwulata C.I., Konstance R.P., Tomasula P.M. (2004), Minimizing variations in functionality of whey protein concentrates from different sources, Journal of Dairy Science 87:749 to 756: commercial whey protein concentrate particle sizes of 53 to 382 µm.
  4. Whey protein isolate agglomeration study, ChemEngineering (2020): D50 58.5 µm and apparent particle density 974 g/L.
  5. Gy P. (1979), Sampling of Particulate Materials: sampling variance falls with sample mass and rises steeply with particle top size. Used for structure only.

OpenChemWorks Laboratory · Reviewed by the OpenChemWorks laboratory · Published October 20, 2026 · Updated November 19, 2026

Get the next one in your inbox

One email per post, the science, no spam, no list-selling. Unsubscribe anytime.

Ready to send samples?

Overnight turnaround available, with expedited service on request.
We use AQC derivatization with HPLC and dual UV and fluorescence detection. New clients can start with a $250 five-sample pilot, credited in full toward a first order of 20 or more samples within 90 days.