| Author | Message | ||
     Michael C Mound (mike) Junior Member Username: mike Post Number: 9 Registered: 7-2007 |
Thanks, Mark. I am personally familiar with all the technologies mentioned...used to sell the sparkers and ICP and mass specs, etc. The issue, though, is how to manage the analyses in real-time and on-line. Not so easy to do with sampling, digestion, etc. But, thanks, anyway. Best regards, Mike | ||
| Howard Mark (hlmark) Senior Member Username: hlmark Post Number: 130 Registered: 9-2001 |
Mike - since your main interest is in inorganics, you should definitely look into techniques like flame emission, plasma emission, spark emission, or one of the variations of gas and/or plasma absorption spectroscopy. None of those involve any radioactive materials. They are all classified as "atomic spectroscopy". Except for spark emission, most of them require the sample to be dissolved, ergo you might have to set up some automated sample preparation equipment. You probably should talk to somebody in the Analytical Chemistry department of a nearby university. Someone teaching an advanced analysis course should be familiar with these various technologies, and can recommend which one they think would be best. They also will likely have some suitable instrumentation that they could use to demonstrate the method to you. You could also contact the major instrumentation companies (Perkin-Elmer, Thermo, etc) that provide the spectrometers. Some of them will likely be at EAS this fall; and if you can wait a bit, all or nearly all of them will be at Pittcon next March. Then you can talk to their experts, as well as seeing the equipment. \o/ /_\ | ||
| Michael C Mound (mike) Junior Member Username: mike Post Number: 7 Registered: 7-2007 |
Hello, Everyone...and, thanks for your valuable input. Not to beat a dead (or dying) horse, but there are a few points I would still like to make. First of all, I have been deeply involved in various technologies used for coal characterization, including dual-gamma systems, prompt gamma neutron activation analysis (PGNAA), and various types of XRF systems. The general feeling among users is that XRF is good enough for off-line analyses, but only the neutron/gamma systems are useful for on-line. Secondly, the coal analytics I am concerned with are more process analytical techniques than lab versions. This means that the pulverized coal fed to commercial boilers is one of the subsets of the requirements. With the current concerns about using radioisotopic sources of neutron emitters (usually a neutron generator or an actual isotope, such as Cf252, there is sufficient justification to seek other, less hazardous materials. As CF252 is a by-product of plutonium generation, there is understandably a nervousness about relying upon systems that require excitation sources that are either or both politically and practically sensitive to populations. So, along comes some light at the end of the tunnel in the way of NIR, or even DRIFT, and I am wondering how to push the envelope, so to speak. Here are some links to what I am talking about, so if I am "barking up the wrong tree", perhaps I have company (???). The reference for DRIFT is from Talanta Volume 72, Issue 4, 15 June 2007, Pages 1423-1431 - Coal analysis by diffuse reflectance near.doc The references for NIR for coal are from http://www.informaworld.com/smpp/content?file.txt So, perhaps, I am trying to find the keys away from the light, and maybe I won't find them...but, it is fun to try. Thanks once again, good friends. Mike | ||
| Gavriel Levin (levin) Senior Member Username: levin Post Number: 51 Registered: 1-2006 |
Hi Guys, didn't think this topic will pop up again, thought it was done with. Michael has good points, and so is Howard. It is like we insist on looking for the keys under the street light, because that's where there is light, but if the keys aren't there we will not find them, light ot not. Ash is usually the residue from a combustion process of some type of fuel, and it is usually consisting of several minerals, depending on what's in the fuel. In that case, XRF would be useful, because carbon is practically non absorber at the energies typical of the metals because it is a lighter element (atomic number 6) and most of the metals of interest will be above atomic number of 12, and Mg (12) is detectable, so is Al (13) and heavier metals. There are many good XRF analyzers, and today their prices are no longer prohibitive, and with the superb software and the kind of electron (or photon) multipliers they have today, detection limits went down consideraly from what they were 20 years ago. I hope this is any help, Gabi Levin | ||
| Howard Mark (hlmark) Senior Member Username: hlmark Post Number: 129 Registered: 9-2001 |
Mike - I think the concensus is growing that you're trying to do an analysis that is difficult under the best conditions, and wanting to do that analysis under impossible conditions. I hope you're not being a victim of the "I have a hammer so I want everything to be a nail" syndrome. Valuable and of wide application though NIR is, there are things it can't do. Ash and mineral analysis by NIR is problematic under the best conditions. Yes, there are sporadic reports of cases where someone was able to develop an analysis for a non-organic component, but those reports are few and far between. And if you look closely at them, the nature of the sample usually made it a favorable situation. But when you say you want to do the analysis when the sample is coal, which is almost impossible to either get light through by transmission, or to get light back from by reflection, you've set yourself up for failure; even the Marines couldn't do it. But the concensus is (and in my opinion, I agree with the concensus) that you're not going to be able to measure ash in coal by NIR; you should look for a different technique. XRF was mentioned, but if XRF is not a viable method, then you need to look for something else: an analytical technique that will be appropriate for your situation. If nothing exists, then you may have to create one. "Necessity is the mother of invention" as they say. Even modern NIR analysis grew out of a need: to make a long story short, Hayward Hunt at the USDA needed a way to detect water in apples; that led him to Karl Norris and the rest is history. BTW - there are no real "ground rules" for what may be discussed on this forum except that it should appropriate for scientific gentlemen (and ladies), both in topic and language. But by the nature of the discussion group, you can't expect to find special expertise on any subject except NIR analysis, either. We can all come up with ideas, and if they're mentioned here then they're public, so you can try using them. Here's one: automated pretreatment of the sample, combined with some type of atomic spectroscopy. Instruments and chemometrics can do wonderful things, but sometimes even they need a little help! \o/ /_\ | ||
| Michel Coene (michel) Senior Member Username: michel Post Number: 43 Registered: 2-2002 |
Frankly, I think you are looking in the wrong direction. Remember, even reflection spectroscopy still has to answer to Lambert-Beer. This is not some unfortunate inconvenience, it is the basis of spectroscopy itself. If you put something interesting that absorbs little into something boring that absorbs almost everything, you can choose between a microscopic pathlength or looking in the dark. Either you find a wavelength where the boring stuff absorbs less, or you set up a complex sample pretreatment to "de-blackify" your sample. In your case that would almost certainly mean burning it, unless you can think of some kind of liquid extraction method. | ||
| Michael C Mound (mike) Junior Member Username: mike Post Number: 6 Registered: 7-2007 |
Thanks to all, but I have a slight disagreement here. XRF has been used with varied degrees of success, but is not really viable for on-line characterization. The use of other technologies, not a subject within the ground rules of this forum, have been on- and off-useful. What is important is that the blending of thermal and metallurgical coals for an optimization of washeries and power applications relates to the balance between sulfur content and non-combustible mineral constituents (sometimes called the "bone" of coal content). The point is to arrive at a blend, or mix of low- and high-sulfur and low- and high-ash content coals. The calorific values are based on accepted formulae once the constituent balances are known. If the ash mineral constituents can be characterized (why not with NIRS, as they are spectrally known in various libraties?), then both the CV and sulfur can be known and controlled. Some minerals, notably pyrite, are the sources of the S content, so measuring the pyrite mineralization leads to patterning of the amount of contained mineralogically based S. I know of no other source of S in coals, so this is an important point. Since Fe values from other than pyrite can also exist in some coals, it is critical to know from which mineral source (pyrite or other) the Fe derives. Once all this can be established, it would be theoretically possible to know ash, S, CV, and Fe, and thus, the application. My question is how to deal with the high absorption of the dark coal itself, or is this not a problem? Theories welcome. Thanks, again. Mike | ||
| Gavriel Levin (levin) Senior Member Username: levin Post Number: 45 Registered: 1-2006 |
Howard makes a good point - what is the purpose, but at the end the purpose can not be served by a method that will not deliver as Keneth says. Yes, knowledge of sulfur in coal is very important because it is needed for calculations on scrubbing capacity, how to run the scrubbers that "collect" the undesired chemicals from being released to the air. CO2 is still being released as it is not a toxic chemical in the simple sense of toxic. People are talking about scrubbing the CO2 from power stations, well, they will need some efficient method, and water based scrubbing is not a very good one for that. Thanks, Gabi Levin | ||
| David Russell (russell) Senior Member Username: russell Post Number: 35 Registered: 2-2001 |
I seem to recall that Sulfur content is used as a correction factor in empirical formulas for the BTU value of fuels. | ||
| Howard Mark (hlmark) Senior Member Username: hlmark Post Number: 118 Registered: 9-2001 |
Venkynir doesn't say why he needs to know the various analytes he mentions, and it may not matter too much since we're all agreed that NIR is probably not suitable for his needs. But looking at the list of analyses, especially the "calorific value", implies that ultimately he wants to know what's going to happen when the coal is burned, not transformed into other fuels or chemicals, for example. Knowledge of the sulfur content is important, of course, in assessing the quality of the coal, as regards its effect on air quality. \o/ /_\ | ||
| Kenneth Gallaher (ken_g) Advanced Member Username: ken_g Post Number: 25 Registered: 7-2006 |
Being an application chemist by trade who actually has to deliver, I tend to be reality based. Gavriel clearly shares that attribute. | ||
| Gavriel Levin (levin) Senior Member Username: levin Post Number: 44 Registered: 1-2006 |
Kenneth is absolutley right, I forgot sulfur was also mentioned - even sulfur containing organic molecules such as are present in fuels which are primarily mercaptan type compounds - or in more conventional chemists language they are thiols - i.e., compounds in which there is a sulfur atom bound to a carbon atom and a hydrogen atom bound to the sulfur - do not provide a very strong signal in the NIR and in refineries they would have loved to determine the sulfur by NIR because they have it on line anyway - still they need sulfur analyzers to do so. Some of the mercaptans have sulfur with a double bond to carbon and these do not show at all as a significant signal in the NIR. So, NIR can do what it can, lets not ask it to deliver the moon. Gabi Levin Brimrose Corp. | ||
| Howard Mark (hlmark) Senior Member Username: hlmark Post Number: 117 Registered: 9-2001 |
If the interest is in total elemental sulfur rather than particular sulfur-containing species, then the XRF suggestion is a good one. XRF tends to be better for heavier elements, and sulfur is heavy enough to be a good analyte. But once we are talking about identifying and measuring particular elements, there are many types of atomic spectroscopy that can and should be considered. \o/ /_\ | ||
| Kenneth Gallaher (ken_g) Advanced Member Username: ken_g Post Number: 24 Registered: 7-2006 |
Ditto for sulfur. Sulfur looks too much like carbon in IR or NIR. Any NIR sulfur analyses you see are based on dangerous secondary correlations. For high levels of sulfur try XRF. | ||
| venkatarman (venkynir) Senior Member Username: venkynir Post Number: 44 Registered: 3-2004 |
What Mr.Gabi Levin suggest is correct .I have not seen much applicaiton on NIR in Coal As Gabin Levin staments we have think about carbon . | ||
| Gavriel Levin (levin) Senior Member Username: levin Post Number: 43 Registered: 1-2006 |
I must say, that from my past efforts in materials that are based on coal, carbon black etc, there is no useful signal coming back from these even for applications that are much less demanding then doing ash, and other minerals that do not have a nice NIR signature to begin with. Thus, if it were left to me I would not spend time trying NIR for that. Gabi Levin Brimrose | ||
| Gary Rollason (nvisible) New member Username: nvisible Post Number: 3 Registered: 2-2007 |
Generally speaking, the minerals (inorganics) do not absorb in the NIR region of the spectrum, and any would be so weak as to be 'lost' by the absorption by the coal itself. | ||
| Michael C Mound (mike) New member Username: mike Post Number: 4 Registered: 7-2007 |
Thanks, folks. What I really was interested in was the mineral constituents (the ash...or non-combustibles), and the sulfur content. | ||
| Howard Mark (hlmark) Senior Member Username: hlmark Post Number: 116 Registered: 9-2001 |
Coal also has the characteristic that, since we know the origin of it (ancient plant tissue), we can expect that there might be some remnants of the organic origin, i.e., -CH bands corresponding to actual functional groups. For the softest coals, that have been mauled least by nature's processes, (e.g., peat, lignite) they could conceivably show -C-O- bands of various sorts, in the mid-IR, with possible hydrogen-bonding changes to any -CH or -C-OH in the coal. \o/ /_\ | ||
| Kenneth Gallaher (ken_g) Advanced Member Username: ken_g Post Number: 23 Registered: 7-2006 |
Coal is a whole lot like coke (baked refinery bottoms - not the drug) that was discussed a week ago. Pretty close to a black body. There has been some Mid-IR done but it is rough. | ||
| Michael C Mound (mike) New member Username: mike Post Number: 3 Registered: 7-2007 |
Has anybody a suggetion as to how to set up an NIR quantitative analysis for coal? The idea would be to characterize mineral constituents (ash), bitumen, moisture, and calorific values. Also, would both black and white reference standards be necessary? Thanks, Mike Mound |