Oil Analysis and Wear Particle Size

Not open for further replies.


Jul 16, 2002
Alamogordo, NM
I've read from a couple of sources that the inexpensive type of oil analysis we do can read wear particles as large as 5 microns.(Terry claims 10 microns.)

Molakule once said that when wear occurs, atomic-sized particles are released into the oil, and that these atomic-sized particles are the ones read by oil analysis. He went on to say that the amount of "wear atoms" released was proportional to wear.

I've been thinking about this lately, and I wonder if the wear particles that FTIR-style analysis sees really are no bigger than atoms. When an alloy of two or more metals is made, they mix at the atomic level; that is, the different metals occupy interstitial positions of each other's atoms. So, if I file a bronze doorknob, for example, I don't get a small pile of copper and tin. I get bronze filings. If there is any copper or tin in the pile, then the particles are obviously atomic-sized. See what I'm getting at?

Since most of the metals in an engine are alloys, and since FTIR-style analysis reads only elements, then the particles that are read must be no larger than atoms. Is that right? A .1 micron-sized wear particle of lead babbit bearing material, for example, would never be read by oil analysis because FTIR-style analysis can't read babbit--only lead and tin. Since steel is an alloy of iron and carbon, only atomic-sized particles of iron can be read? No?

Please feel free to comment if you can shed any light on this subject.
Oil analysis has been done for over 60 years. In all that time no one has data that demonstrates that the wear metals in a UOA reflect actual wear in an engine. No one in 60 years. Until I see some "proven" relationship as far as I am concerned it is oil astromony.

There are several studies that demonstrate little to no difference in actual wear between different brands, dino and synthetic, and OCI differences. They all look to be about the same.

FTIR is not used to determine the wear metals during an oil analysis. FTIR is used to determine things such as water content, oxidation, nitration, and sulfation. It is most appropriate for dealing with the organic side of the analysis.

Metals determination today is done usually by one or the other of Direct Coupled Plasma (DCP) emission or by Inductively Coupled Plasma (ICP) emission. In either of these techniques an argon plasma is generated at a temperature of 8,000-10,000 degrees Kelvin. That is the same temperature as the surface of the Sun. This temperature is hot enough to dissociate any molecule or alloy back to their individual atoms. A 5 um particle of any alloy will be read as the individual atoms of which the alloy is composed.

The basics of plasma emission analysis are: Some of the electrons in the atoms are moved to higher energy state by the heat of the plasma. When an electron jumps energy states light is emitted. Each atom emits a characteristic set of wavelengths when an electron that is excited into a higher state. By reading the intensity of the wavelength(s) emitted by the elements of interest you can quantify those elements. The amount of light emitted is proportional to the concentration of the element in the plasma, and therefore relates to the concentration in the original sample.

The particle size limitation stems from two main sources, the residence time the sample spends in the plasma, and the spray chamber used to introduce the sample into the plasma. Particles larger than about 10 um are not in the plasma long enough to absorb enough heat to be fully dissociated. The spray chamber is specifically designed to prevent particles over a 10 um aerodynamic diameter from making it to the torch. This prevents larger particles that may only be partially dissociated from interfering with the accuracy of the results. Only about 2-3% of the sample aspirated into the instrument makes it into the plasma.

I hope that helps you understand what is being measured, and how, when checking for wear metals.

Ed - excellent explanation! you took the words right out of my mouth.

I disagree with Terry on the size and think it's smaller, more in the 3-5 microns range. This is based on my microscope studies of particles in the oil where I have seen many samples that have so many 5 micron iron particles that you can pick up the membrane filter with a magnet yet the iron reading is very low. Some larger particles probably to make it through but the majority are the little ones. (to see some good examples look at my post under Transmission Oil Analysis)


Originally posted by Ugly3:
Oil analysis has been done for over 60 years. In all that time no one has data that demonstrates that the wear metals in a UOA reflect actual wear in an engine. No one in 60 years. Until I see some "proven" relationship as far as I am concerned it is oil astromony...

Ugly3 - what in the world are you talking about?
We tear down engines all the time because of the results of the oil analysis, and find problems every day. We require it on all used machines we take on trade and it is also required for warranty purposes. Properly used it saves our customers millions!

Perhaps I misunderstand your meaning here? if you were to say that is is of questionable value for comparing oil then I would tend to agrees, unless you look as many many samples.

I am surprised that I've never seen anyone here using oil analysis to check out a used vehicle. We do this all the time for ourselves and our customers and have found more coolant entry problems than you can shake a stick at.

I also agree on the particle size, especially with ICP. DCP looks to have a better solid handling capability and possibly a longer residence time. I don't have any oil analysis experience, so didn't want to contradict Terry on that point.

The spray chambers are designed to have a cut point of 10 um. That's an aerodynamic diameter of 10 um. Aerodynamic diameter is density dependent. A metal particle would have to be much smaller than a solvent droplet to make it to the plasma.

Are you using DCP? Seems that the Beckman DCPs would be long in the junk yard except for the huge number of them in CAT labs. They really are a great tool for oil analysis.

Hi Ed,

Terry may be speaking of DCP or Rotrode, there are different opinions from knowledgeable people about size so I can't speak for them, only of my own experience. We are using a Perkin Elmer about a year old, before that we used a Leeman and that is when I came to the conclusion about the small particles. I haven't really thought about it much lately so perhaps it's time to reevaluate what the new instrument is seeing. I never heard about the 10 micron design limit before but that makes perfect sense. When you look at the spray chamber on the Leeman it amazing that anything gets to the torch.

Those DCP's are a real workhorse and see to last forever. I know a bunch of people using them and there is a guy in the southwest (I think) that buys old ones and refurbishes them.
Ed, Stinky, thanks so much for the explanation! I realized that the type of analysis that determines metal particles wasn't FTIR (I had been corrected before about this.), but I didn't know what to call it.

I'm familiar with optics and lasers, so much of your explanation makes sense.

I'm going to have to visit an oil analysis lab one day and actually see the equipment, and see how it works. That would fill in a lot of holes in my understanding.

Thanks again.

I thought it was an excellent explanation as well...thanks! I was going to post in response, but I don't think I could have explained this concept nearly as clearly.

I think most labs with modern equipment can detect some particles up to 10 um, but it's really the smaller ones < 1 um that determine the wear metal concentrations....This is why by-pass filters have little or no effect in skewing oil analysis results.

I'm not sure where the Ugly3 comment came from either, but it's certainly way off base....I'd also do an oil analysis on the engine oil and ATF from any used vehicle I was thinking of buying. You can tell a great deal about the mechanical condition of the engine - compression and ring seal - as well as the state of "tune" just from looking at this data.


[ April 25, 2005, 06:42 AM: Message edited by: TooSlick ]
Well guys here are my thoughts.

I think too many people think oil A with 12 PPM iron is better than oil B with 19 PPM iron. And if oil C produced 27 PPM iron their engine will fall apart tomorrow. I don't think so.

If I had 10 readings from the same engine and the same oil in the range of 15 PPM iron and the next two readings were 40 PPM I might think something is going wrong (or maybe they changed the oil formula).

If I had thousands of readings of wear metals from some model engine using a particular oil and my engine deviates significantly from the norm I might think something is wrong.

I fully agree that a UOA can tell you an enormous amount of information about the condition of the engine, it is just that wear metal readings are the least valuable portion of the data. Oil condition is much more significant.

I don't know if my engine with a life of 50 PPM iron will die and sooner that your engine with 25 PPM iron.
Not open for further replies.