Simulated surface tension of three commercial fluids, under a bar set in advance

We simulated the surface tension of three fluids sold for electronics cooling and graded each against the value its maker publishes, under a pass band written down before the runs. All three came out high, and our own control run missed too. The method failed this test, and we publish it as a fail.

Under bands written down before the runs, our simulated surface tensions for three sold fluids all came out too high, and our own control run missed as well.

Each row is one sold fluid. The shaded band is the pass band around the maker’s published value (the white line); the hollow dots are our three runs and the orange dot their mean. All three means land above their bands.

A dotted underline marks a number read straight from a published file when this page was built.

In this post
  1. What it shows
  2. What we found afterwards
  3. Why it matters
  4. Who should care
  5. The limits, in the lab’s own words
  6. Prior art
  7. Evidence
  8. Related results

What it shows

Each fluid passed if the mean of three simulation runs fell within ±15% of its maker’s published surface tension. The band was written down before the runs started. Surface tension is in millinewtons per metre:

  • Novec 7100: the maker publishes 13.6; we simulated 18.5, which is 36% high.
  • Fluorinert FC-72 (as n-perfluorohexane): the maker publishes 10.0; we simulated 13.3, which is 33% high.
  • Chemours Opteon SF33 (HFO-1336mzz-Z): the maker publishes 13.0; we simulated 18.4, which is 41% high.

None of the three landed in the band. A control run on pentane, meant to show the method reproduces our own earlier value, missed its bar too. So the result says the simulation is in question, not the fluids.

What we found afterwards

After the runs, we checked a property we had not planned to grade: density. The simulated liquids came out far less dense than the real ones, short by between 26% and 59%. That is a likely cause of the high surface tensions. It is not a proven one, and a fix has not been run.

Why it matters

Our screen leans on simulated surface tension. Our blends are designed around a difference in surface tension between their two liquids, an effect our models assume and no one has yet shown for them. If the simulation runs high on fluorinated fluids, numbers built on it for those fluids carry that error.

A test that can fail is the only kind worth running. This one was set up in advance, it failed, and the failure tells us what to fix first.

Who should care

  • Engineers who use simulated fluid properties: this puts a size on one kind of error, against outside data.
  • A partner lab: surface tension and density are the first properties to measure on any blend.
  • An investor: the misses are published with the band they missed, so a reader can see exactly how far the method was off.

The post Why our surface-tension simulations came out high tells the same story at length.

The limits, in the lab’s own words

Quoted word for word from the published file named under each quote. Read the result with them.

Not pre-registered, added after the runs: the NPT stage starts every box at 550 kg/m3 and runs 0.5 ns, which left these fluorinated liquids far below their published density; a likely cause of the high surface tensions, not a proven one.

From The surface-tension runs for three commercial fluids, graded against published values.

Prior art

The published work this result uses or is checked against. Each link was opened and checked when this page was written.

Evidence

Each number with a dotted underline was read from one of these published files, field by field, when this page was built.

All results

Ask about a result, or check one yourself

Every result on this site links the published file it was read from. Licensing, acquisition and partnership enquiries go to one address, and a person reads it.

Write to us Read the results

How the numbers on this page are checked
Each underlined number is read from a published file when the site is built, and the build stops if the page and the file disagree. A number would be marked if
  • its file were still loading in your browser,
  • the question behind it had not been looked into,
  • it had been looked for and not found,
  • its file held no value for it,
  • its file were missing or had changed,
  • two files disagreed about what it describes,
  • its sample were too small for the claim,
  • two files gave it different values,
  • its file could not be published,
  • the file it is read from had last changed more than thirty days before this build,
  • the question did not apply to this page or
  • the program that makes it had stopped with an error; no underlined number on this page is marked.
A number in plain text, without the underline, is not checked this way and carries no mark; the date each file last changed is in the list of published files.