Laboratory Testing & General Mineral Processing Engineering

Laboratory Testing & General Mineral Processing Engineering

  • To participate in the 911Metallurgist Forums, be sure to JOINLOGIN
  • Use Add New Topic to ask a New Question/Discussion about Mineral Processing or Laboratory Work.
  • OR Select a Topic that Interests you.
  • Use Add Reply = to Reply/Participate in a Topic/Discussion (most frequent).
    Using Add Reply allows you to Attach Images or PDF files and provide a more complete input.
  • Use Add Comment = to comment on someone else’s Reply in an already active Topic/Discussion.

Moisture effects on iron ore Standards analysed by lithium borate fusion - ICP (11 replies)

Victor Bergman
8 years ago
Victor Bergman 8 years ago

I am trying to determine the cause of errors in analysis of iron ore standards where the analytical technique involves lithium borate fusion followed by ICP-AES. Manufacturers of the standards specified they must be oven dried prior to fusion to avoid moisture effects. Does anyone know what the magnitude of such moisture effects is likely to be, i.e. by what percentages will the analyzed grade under- or over-report?

O
OberstGruppen
8 years ago
OberstGruppen 8 years ago

Can't help you on the exact effects, but given that analytical determinations are always related back to the mass of the sample, I'm guessing excess water in the sample may bias your results low. There may be effects on the flux during fusion but you would have to ask an analytical chemist on that one. Hopefully one of them will pop in here and post 🙂

However, most modern analytical labs have a drying step as part of the sample preparation process, and if your standards are inserted in the sample train then there should be little 'moisture' left in them. This is different from structural water of course.

(unknown)
8 years ago
(unknown) 8 years ago

The issue is whether the Fe ore standards (I prefer reference materials) are hygroscopic. It is particularly problematic for any lateritic materials and it can also depend on the climate where the test work is being done (i.e. it’s not a problem in the Atacama but it is in the Amazon). The issue is that the materials were certified on a dry weight. If the RM adsorbs moisture and it is not dried prior to fusion then results will be lower. You can think of it as the RM being "diluted" by adsorbed water. The question is then, if the RM is matrix-matched with samples, do you have the same effect for samples? You could ask the lab to do a moisture test on the RMs which will tell you the % to which the RMs may be biased.

(unknown)
8 years ago
(unknown) 8 years ago

Make sure you are drying at ASTM temperature. While I do not know the characteristics of the reference standards you are using, I made the unfortunate discovery, using reference standards in a different application context, that over-temperature drying for those standards volatilized something and the material, for ever after, assayed 5% high.

I normally assay gold ore or metallic scrap. I am just initiating assays on material with a matrix similar to iron ores. Would you mind sharing a copy of your lithium borate fusion method? I am testing peroxide-carbonate and peroxide - lithium metaborate fusion methods, but would like to see what someone more experienced in iron ores is using. Could also share the names/sources of the iron ore reference standards you are using in the review?

O
Obergruppenfuhrer
8 years ago
Obergruppenfuhrer 8 years ago

For XRF analysis, the effect of not drying the flux depends on how much moisture the flux absorbs, which is determined by the laboratory environment (humidity) and time. You can dry a sample of flux to see how much moisture is actually absorbed and then calculate the error.

(unknown)
8 years ago
(unknown) 8 years ago

Loss of Moisture (LOM), (drying at105°C) will account for hygroscopic moisture while Loss (or gain) on Ignition (LOI) (drying at 1000°C) arises from the loss of water of crystallization, combustion of organic matter or decomposition of carbonates, which can be significantly higher.

During fusion, you will encounter both of these effects; however, hygroscopic moisture can be much more unpredictable. Most labs should perform these tests alongside the fusion. If done correctly, with a full-suite analysis you should get a "Total", (sum of elemental composition) close to 100%.
I hope this helps.

(unknown)
8 years ago
(unknown) 8 years ago

Generally under normal condition hematite Iron ore pulps (-100mesh) inherent moisture can be 0.8 to 1.5 percent. Accordingly Fe(t) weight percent may under estimate 0.36 to 0.50 percent provided Fe(t) = 65%.

S
Standartenfurer
8 years ago
Standartenfurer 8 years ago

As manufacturers of CRMs we are mindful of the critical importance of correct moisture handling procedures for iron ore CRMs and for the samples they are used to monitor. If moisture is not corrected for and the sample is assayed ‘as received’, the under-reporting of constituent values will be directly proportional to the moisture content at the time of weighing and this can be very significant, up to 10% in highly hygroscopic laterite based samples. Moisture (which can be defined as H2O) is dependent on the particle size (surface area) of the CRM, the hygroscopy of the various mineral constituents and ambient temperature and humidity. Errors can be especially large for high LOI iron ores and nickel laterites. I’ve heard stories where highly hygroscopic samples were dried and then stored with desiccant prior to analysis. The desiccant was then robbed of its moisture by the sample! This illustrates the importance of using fresh desiccant if samples are not immediately fused.

Following ISO 2596 overcomes these problems and stipulates: “weighing of analytical test samples shall be performed in parallel with hygroscopic moisture sampling and preparation operations; otherwise erroneous moisture corrections will result. The determination of hygroscopic moisture shall be performed whenever a constituent is reported to a dry basis. The hygroscopic moisture values shall not be averaged, but shall be used individually to correct the corresponding constituent values.”

Extreme care should therefore be taken with hygroscopic moisture correction and/or drying procedures as LOI determinations are often a major source of error in high LOI iron ores. To avoid these errors the following is recommended:

Equilibration of sample in lab atmosphere (minimum 2 hours) with all analyzes corrected to dry basis after moisture analysis at 105°C on a separate aliquot (this is the preferred method and is stipulated by ISO TC 102/SC 2 N 1550 for samples containing greater than 2% LOI) OR

Removal of H2O- by drying at 105°C for a minimum of 6 hours or until constant mass is achieved, and then placed in desiccators with fresh desiccant until weighed for analysis.

(unknown)
8 years ago
(unknown) 8 years ago

Note that mass losses can occur well before the full LOI temperature of 1000°C mentioned by before. I expect that iron ore reference materials are less likely than gold ores to have low temperature volatiles, but, for the gold ore reference material problem traced to staff applying excess drying temperature, permanent mass losses occurred at a drying temperature of just 163°C. This is less than 50°C over the ASTM recommended maximum.

Victor Bergman
8 years ago
Victor Bergman 8 years ago

Thanks for all the comments on this query. I don't know if moisture effects are a cause of error in this case, just trying to understand what the errors might be. The samples are being analysed at an international lab, so it really shouldn't be a problem.

Gruppen
8 years ago
Gruppen 8 years ago

I'm interested in your comments. Which ASTM standard specifies the sample drying protocol? Is your example of a problem with gold assays published anywhere online?

(unknown)
8 years ago
(unknown) 8 years ago

The ASTM moisture method that specifies appropriate drying temperature is ASTM D2216-10. Perhaps ASTM has details supporting their specified temperature within their development history. I am not aware of any published documentation concerning improper drying and handling of certified standards - I just avoid doing it.

I simply encountered a case where an over-zealous tech dried a subsample of a bulk reference standard used in routine QC. I traced the sudden reproducible bias in the lab's assays for that standard to the drying temperature applied. Same bottle, portion not high-temp dried, bias gone. Problem quietly removed. In Nevada labs, moisture weight gain for properly handled standards (although carbon can have issues) is often minimal (desert air).

I expect that many certified gold ore standards are not negatively impacted by improper drying. I apply the drying recommendations which come with some of the certified standards I’ve used to all standards. Carbon standards are more obvious examples of the potential impact of excess temperature drying (or exposure to humidity). If a lab weighs ore standard controls out of bulk bottles or sealed packets "as received"; over-drying is not an issue. If the ore standard is not hygroscopic, the source material was high-temp dried prior to certifying, protected in sealed containers and handled with reasonable care, the impact of air-acquired moisture or over-temp losses can be negligible (particularly in Nevada’s dry air). However, verification is good policy - a few percent biases can add up to significant money in some applications.

In high humidity locations (e.g., 80% relative), a LIMS programmer friend noted that a client had implemented instrument moisture reads, made at time of weigh-up, to correct assay weights for carbon fire assays. During the time delay between removal from sealed packets or storage in desiccators, moisture weight gain, at this lab, generated assay bias relative to certified carbon assay values. Activated carbon is very hygroscopic; apparently precautions that work in high desert were not enough at that laboratory. In any case, once this instrument correction went live, assay bias on certified carbon standards apparently ended. LIMS programmer’s client was pleased with outcome, but not interested any publicity.

Please join and login to participate and leave a comment.