Spectroscopy applications in Tribology and Metal Working Fluids

[TheBlackFedora]

Greetings.

I am a new member of this forum, and I joined specifically for assistance on the following topic.

 

 

 

The company I work for is setting up a Tribology laboratory to test various metal working fluids.

The required tests include an elemental breakdown to determine fluid additive levels as well as contaminants from dirt and wear metals.

 

Standard test results offer levels (in PPM) of: Iron, Chromium, Nickel, Aluminum, Copper, Lead, Tin, Cadmium, Silver, Vanadium, Silicon, Sodium, Potassium, Titanium, Molybdenum, Antimony, Maganese, Lithium, Boron, Magnesium, Calcium, Barium, Phosphorous, and Zinc.

 

The industry standard appears to be an ICP (Inductively Coupled Plasma) spectrometer, however some references include the use of an optical emission spectrometer and an FTIR (Fourier Transform Infared) spectrometer.

 

We have a UV/VIS spectrometer available to us, but I am unsure if this instrument can duplicate or replace any of the abilities of the other types used in Tribology laborator testing.

 

So my question is... what type of spectrometers (or other instruments such as mass spectrometers, NMR spectrometers, gas chromatographs, or High Pressure Liquid chromatographs) would provide such an elemental an analysis, and could our available UV/VIS spectrometer provide any useful data?

 

Cost effectiveness is a major concern in this project. The projected costs (initial start-up, operating, and maintenance) will determine whether or not our laboratory comes into being. So the more economical instruments are prefered.

 

My knowledge of the subject is limited, and I'd appreciate any useful advice or suggestions I can get.

[TheBlackFedora]

Anyone... Anyone...?

 

Bueller...?

[John Cuthber]

In many cases, the answer depends on what your labour costs are.

With an ICP I can pretty much dilute the sample in some acid and run it into the machine. It will tell me the concentrations of those elements and it will measure then down to parts per billion (typically).

 

with a UV/Vis machine I can measure iro, but to do so I have to add some reagent which gives a colour with iron.

Here's a "recipe"

http://www.sas.upenn.edu/~kimg/mcephome/chem506/felab.pdf

For chromium I could make the sample alkaline, oxidise the chromium to chrome (VI), and then follow this method

http://www.baaqmd.gov/~/media/Files/Records/MOP/vol%203/MOP-34.ashx?la=en

and so on, I could go through the list and find methods for most of the metals you are asking for.

You could do the analyses, but it would take days.

 

That's the simple reason why companies pay for better machines.

[TheBlackFedora]

Thank you, that does seem like the logical way to look at this.

 

Essentially, the more money the company puts into the lab, the better the results. ("Better" as in accuracy and speed of results.)

 

The problem I'm facing at the moment is one of economy. An old saying I've heard is that "Anyone can design a Rolls Royce. It takes skill to design a Volkswagen."

 

While my company could invest the money in an ICP spectrometer, I was given the task of finding a variety of options available to us. So I need to consider the initial cost as well as the operating cost (supplies, chemicals, energy), and finally the labor cost (something that takes days would be an issue to consider, as you pointed out.)

 

One item I came across was a brocure for a "turn-key Tribology Lab" offered by a particular company. (I'm not posting the name because I don't want to promote anyone or anything on a forum I've only just joined.)

 

Anyway, this "turn-key lab" setup included several instruments which I hadn't thought about using, and it DID NOT include the ICP spectrometer. It included an optical emission spectrometer and a FTIR spectrometer. These were the only instruments they listed to analyse fluid contaminants and wear metal content.

 

This set my mind on the current track, which is... can we do an elemental breakdown of the sort that an ICP spectrometer provides with alternate instruments?

 

This brought me to thinking about the possibilities of a gas chromatograph. A question from a friend who is a molecular biologist suggested an NMR spectrometer. After looking around and doing some more research... I ended up with more questions than answers.

 

That is what brought me here.

[John Cuthber]

I'd need to check, but optical emission spectroscopy isn't a bad way to do elemental analysis.

[TheBlackFedora]

Thank you for the input. I really have very little experience with such things, and my online research has been less than completely clear.

 

To further illustrate my lack of knowledge, is there a difference between "optical emission spectroscopy" and the Visible side of the VIS/UV spectrometer we already have available?

[mississippichem]

I'd need to check, but optical emission spectroscopy isn't a bad way to do elemental analysis.

 

Does optical emission spectroscopy refer to AES "atomic emission spectroscopy". I've been told that AES can't really compete with AAS in terms of sensitivity because at any given time the sample will have a much smaller population in the first excited state than in the ground state. In other words it is easier to measure the decrease in intensity of the incident light (AAS) than it is to measure the weak signal that results from the emission in AES. I have no experience with AES though so you may know something I don't. I don't know if you still need element specific cathode lamps for AES like you do for AAS but I would guess that you would.

 

TheBlackFedora:

 

I agree with all the advice John Cuthber gave you in his first post. I think that AAS through ICP methods might be a winner but keep in mind that cost will be high. Your friend who suggested an NMR gave poor advice.

 

As mentioned earlier the UV-vis analysis would take days as you'll have to complex each one of these metals to get a species that has any measurable absorption. Also, IIRC UV-vis can't really hold a candle to AAS, AES, or AFS (atomic fluorescence spectroscopy) in terms of sensitivity and lower detection limit. How low of concentrations are you looking to detect? Are we talking a few ppm or several hundred ppm?

[John Cuthber]

OES can cover a number of things from flame tests to ICP/OES.

It's impossible to say which is meant here.

It's true that the ground state is most populated.

On the other hand the absorbtion method involves measuring a very small change in a large light intensity, which is also difficult.

[TheBlackFedora]

The "turn-key" tribology lab may have been referring to an ICP spectrometer when they listed the "optical emission spectrometer". This would explain what the ICP was missing from their lab brocure.

 

I was a bit confused by my friend's suggestion of NMR, as it was mentioned nowhere in relation to the type of testing we are expecting to do. But as I said, I am a novice and do not yet know helpful advice from usless advice. Thanks for the input.

 

Our current testing reports list concentrations of metals from as low as 0 and 1 PPM up to above 900 PPM. Most of these are contaminants (such as silicon and sodium), while others are deposits of "wear metals" from bearings and machine tools. Still others are chemical additives originally used in the metal working fluids.

 

With the help of this converstaion, I am pretty sure that our UV/VIS spectrometer is not going to be useful in our laboratory for replicating the tests we are currently outsourcing. It would be too time consuming and therfore not economical according to the intention of this project. However, I expect it won't be abandoned as it may be useful for something else in our shop. (Other than a doorstop or paperweight.) If the event we can't use it at all, then it will be sold to help raise capital for the lab.

 

While I understand that the mention of "optical emission spectrometer" could indicate several things, I am relatively sure it means ICP spectrometer due to that instrument being referenced by other laboratories which do similar testing. Unless I come across something else (such as a functional Star Trek tricorder), I am going to recommend we aquire an ICP spectrometer. Though it won't be cheep initially, it will duplicate the tests exactly (which we currently outsource) and it will do so efficently.

But... if anyone actualy has access to a Star Trek tricorder...

 

As this is an on-going project, I will still be open to suggestions and advice for quite a while before I reach a "point of no return", and can no longer put the input to good use.

[John Cuthber]

ICP OES is certainly a reasonable enough idea.

It's how they distinguish the kit from ICP/MS.

 

I do know how to build a Tricorder, but it would put me and my fellow analytical chemists out of a job so I'm keeping quiet.

 

You may want to look at the price of spectroscopic grade argon.

Edited June 12, 2012 by John Cuthber

[TheBlackFedora]

That's a very good point. Consumables for operating an ICP spectrometer will likely put our costs beyond the original expectations. This project is starting to look too expensive for us to pursue.

 

Thanks anyway on the Tricorder. I doubt we could afford that either.

 

I have a bit more information now.

 

The "turn-key" tribology lab includes a "Rotating Disk" spectrometer which is rated to test for all the contaminants, wear metals, and additives we would require. This is a new one for me, and I'm going to have to figure out a whole new set of of questions to have answered.

 

The more research I do, the more questions I have.

[John Cuthber]

Unlike the tricorder, this "The more research I do, the more questions I have." is what keeps scientists in a job.