whats the minimum a certificate has to have before you take it seriously
😂17⚠️16🎉16
#coa-reading 2025-07-06
Sunday52 messages10 participantstimes are UTC
Highlights from this day
- nhs_pathway_nell — which is why the vendor-submitted independent report is the strangest artifact of all. vendor picks the sample, vendor picks the lab, vendor pays, vendor decides… 19:13
- vial_ledger — publication bias with extra steps 19:25
- egfr_ed — Xi'an Mokemei sent me a sheet with the analyst signature scanned in and a phone-camera photo of the chromatogram printout. weirdly that felt more real than a clean PDF 20:17
- nhs_pathway_nell — Hubei Jianxiang gave me two sheets for two lots six months apart and the retention times differed by 0.3 min with a note about a new column. that is what a real rerun… 20:20
what does net peptide content mean next to gross mass
weigh the vial
⚠️15
if the certificate date precedes the lot date, ask. sometimes it is sloppiness and sometimes it is not
🙏12👍18😂2
sorry to jump in one decimal versus two is usually just house style, not precision
if you only read one line on a certificate, read the lot number. if you read two, read the assay
right so i asked MKM for the method and they sent it. that alone put them ahead of most, not advice obviously
residual solvents matter more than people think for anything that went through a lot of purification steps
peptidemeter-summary-e-2205.pdf
2 pages · 680 KB · not retained in the public archive
the honest summary is that a certificate tells you what the supplier says about a lot they chose to test
tfa counterion runs higher by mass than acetate. if the certificate does not say, you cannot do the mass balance
certificate is dated before the lot date, is that just sloppiness
why would a vial be 99.2 pure and still assay under label
the certificate from TFC is the one i use to teach this because it actually has a populated impurity table, we shall see
❤️1🤝13
not detected means nothing
ask for water content
does purity by hplc at one wavelength miss things
gross mass on a jewellery scale plus the water content is enough to catch most underfills
*PeptideMeter not the other one
thats a tell
a supplier certificate on its own is a starting point, not evidence. that is the whole reason the verification channel exists
tfa or acetate, how do i tell from the paperwork
19:04topic set to: seller-supplied COAs vs independent reports — what the gap actually is
starting point. a seller COA and an independent report answer different questions
the seller sheet says: this lot met our spec when we made it. the independent report says: this sample, which someone posted from somewhere, measured this
and both have a hole in them. the seller sheet has an incentive problem. the independent report has a chain of custody problem
whats the custody problem, isnt independent the gold standard here
the lab tests what arrived in their envelope. they cannot know it came out of the vial you say it came out of, or that it spent nine days at 34 degrees on the way
labs are honest about this. we are the weak link, not them
which is why the vendor-submitted independent report is the strangest artifact of all. vendor picks the sample, vendor picks the lab, vendor pays, vendor decides whether to publish
publication bias with extra steps
😂9💯7
not worthless though. a vendor who publishes independent reports on a schedule, including boring ones, is behaving differently to one who published one report in 2023
agreed. cadence is the signal, not the number
how do people spot a fake sheet
there is no single tell. there is a stack of small things and you look for two or more
the stack, roughly weighted
| What you notice | Weight | Innocent explanation |
|---|---|---|
| same chromatogram, different lot | very heavy | essentially none |
| retention times identical to 0.01 min across dates | heavy | none |
| one field in a different font | medium | reseller filled in a SKU |
| area% figures that do not sum near 100 | medium | truncated impurity list |
| lab name with no address or accreditation line | medium | small lab, real |
| PDF metadata author is the reseller | light | they scanned it |
| no lot number at all | heavy | genuinely careless vendor |
the sum-to-100 one caught something for me. purity 99.4, impurities listed 0.2 and 0.2, and then a total impurities line saying 1.9
so either the table is truncated or the numbers were typed by someone who was not reading them
i asked. got told the table shows impurities above 0.1% only, which is a legitimate reporting threshold
it is. and it also means 1.5% of that vial is a crowd of sub-threshold peaks nobody characterised
so not a forgery. just a document that reveals less than it appears to
im starting to think most of these are neither fake nor useful
that is the actual state of the market, yes
Xi'an Mokemei sent me a sheet with the analyst signature scanned in and a phone-camera photo of the chromatogram printout. weirdly that felt more real than a clean PDF
it is harder to fake sloppily than to fake cleanly, funny enough
Hubei Jianxiang gave me two sheets for two lots six months apart and the retention times differed by 0.3 min with a note about a new column. that is what a real rerun looks like
✅11
that note is the most reassuring thing in this entire conversation
so what do i do with the sheet that came with my Kerui order
file it, log the lot, do not treat it as a test result, and if the compound matters to you pay for your own. see #lab-requests for how the pooled submissions work
and check the vendor on VendorInvestigate before you spend money on a test, someone may already have
will do. thanks all
good conversation. pinning the tells table
21:06quiet.hours pinned a message to this channel