Emerging Graphene Technology Company
Hydrothermal Graphite Deposit Ammenable for Commercial Graphene Applications
Message: Re: must read
| From: BumblebeeRR | 5/13/2015 11:09:16 PM |
| 2 Recommendations Read Replies (1) of 30 | |
| the_Chief, I am going to share something from a friend. Now, I cannot speak for what you heard from Agilent, Perkin Elmer, etc. I am not a statistics expert, or an analytical chemist, although I understand a few concepts. I do not have personal knowledge of the intimate details of how SGS did Zenyatta's study. However, given all of this discussion, I asked a friend, who is a laboratory technician in an ISO 14001 certified-lab, for some context and comment, as the "1.4% error" (or now even higher ... 6%??) did not sound right ... for other reasons I will add later. My friend and his lab also use a Leco for carbon analysis regularly, and from what I understand, do the same kind of analysis being performed by SGS for Zenyatta. My friend informed me that their control chart for their method has 0.007% as one standard deviation. That means that three standard deviations would be 0.021%. From what I can remember of statistics, if the data are normally distributed, 3 standard deviations would be expected to account for 99.73% of all data points. Even if the data are not normally distributed, the 3-sigma rule of thumb would still state that 98% of all data points are within 3 standard deviations. For some background ... basically to measure samples, and to ensure their results are accurate, they first create a control chart, by doing, from what I understand, about 200 samples that are within the range they want to test. They take these samples, which haveknown values as they are considered "standards" (i.e. certified reference materials from a certified ISO 14001 lab), and run them on the Leco, and compare the results from the machine to the actual, known values. From this, they use statistics to produce a control chart, with a mean value and up to three standard deviations, above or below the mean. Once established, when they go to run a new sample, they would first run control samples, with concentrations likely above and below the expected value of the sample being tested. The results for the control samples (which again, have known carbon concentrations) would have to be within 3 standard deviations of the mean, high or low. If the control results aren't in that span, then that means something is wrong, because the machine is producing inaccurate results. For their particular setup, that would mean that if they tested a known carbon sample of 99.96%, in order for any subsequent results to be accepted, the machine would have to spit out a number between 99.939% and 99.981% (i.e. the actual, known value, of 99.96, plus or minus 3 x 0.007%). Per ISO 14001, they would be forbidden to report results, if their control results are not coming back within the control chart properly. Having validated the machine is working properly (controls are within the acceptable range), they would then run the sample(s) they wanted tested (whose value(s) is/are unknown). The reported results for the samples would then be considered accurate, with excellent confidence that the results reported are within +/- 0.021% of the actual value. I believe a final round of analysis is done on one or more controls, to simply verify that the machine is still running accurately (i.e. nothing bad happened between the time the first controls were run and now). So you have verification on both the front and back end, giving you confidence that the tests in the middle are good. None of this may be new to you. If it is not, I do not mean to condescend by providing what may be considered "of course I know that". It is not bad to state it anyway, for those perhaps less familiar that may read this. Anyway, getting to my first point ... if my friend's lab can achieve this level of accuracy, then it is likely that SGS was able to achieve something similar for Zenyatta, at least based on my understanding. Again, I do not claim to be an expert on this stuff, and simply am trying to share what a friend shared in context of how the analysis is completed, as best I can. ZEN's purity was quoted as 99.9% or higher. Someone, perhaps yourself, asked where is the reporting of the error or confidence. I believe you will find that is built into the number being reported, through the wording. Likely, when they report 99.9% or higher, they are reporting the lowest result achieved for any sample minus the 3 standard deviations (or perhaps more) they were using ... thus giving you the lower confidence limit. So basically they are saying, we know all the samples were at least this high (with some high level of statistical certainty, at least 3 sigma and perhaps even higher). That's point number one. Bear with me and my long-winded post. Point number two. What would be the point of running a test that had 6% accuracy (or even 1.4% accuracy) when such a wide range of uncertainty would basically preclude you from ever being certain you had material that could produce an economical deposit? From my own standpoint, without even getting into any statistics or analytical chemistry, if the method was that inaccurate, you would find another way. That would be no different, in principle, than using Method X to calculate gold concentration in a core sample where method X has results that are within +/-20 g/t and then reporting results of 20.15 g/t over 100 metres in a news release. That means you could have 100 metres of amazing ore (40.15 g/t) or 100 metres of useless ore (0.15 g/t). The method must obviously be chosen carefully taking into account the material nature of the results. Such a potential difference from the reported value would clearly be material and would lead to total lack of confidence in the data, not to mention a very misleading press release. From the Zenyatta press release: The metallurgical test work is being performed under the supervision of Alex Mezei, M.Sc., P.Eng., Director, Engineering Technical Services at SGS Lakefield, independent consultants to Zenyatta, and Peter Wood, P.Eng., P.Geo., VP Exploration of Zenyatta. Peter Wood and Alex Mezei are the Qualified Persons under National Instrument 43-101 who supervised the preparation of the scientific and technical information that forms the basis for the disclosure contained in this news release and they have reviewed this news release. SGS performed analyses of all purified Albany graphite samples by direct ash analysis using a platinum crucible, according to a validated method that also accurately quantifies key trace level impurities by subsequent ICP analysis. This allowed SGS not only to accurately characterize the purified graphite, but also to support the metallurgical balancing for the purpose of the PEA study Some key points in there. Mezei and Peter Wood supervised the study. The testing was done using a validated method. Both of them reviewed the release. Mezei is director of engineering and technical services for SGS, whose reputation I don't think anyone is going to question. He has a master's degree in science or engineering by the looks of it, and is a licensed professional engineer. His job is to make sure results are reported accurately, with a proper method. I feel pretty confident that he knows what he is doing ... because if he didn't know what he was doing, he wouldn't be employed at SGS as a director. For them to use a method that has a "1.4% error" or a "6% error", and then report that result as 99.9% or higher ... basically they would be publishing hugely misleading information that would have highly material consequences on the economics of the deposit. Doing such a thing, would, in my opinion, destroy the credibility of SGS, open SGS and Zenyatta, Mr. Wood and Mr. Mezei as well as the directors/officers of Zenyatta to huge legal liabilities. Even if you were to assign Mr. Wood zero credibility because he is not independent (I don't think that's warranted, he is a professional as well, but just to make a point) ... even then, a third-party provider of services is not going to review a news release and allow it to be published with his name, his company name, etc., when the NR contains material misrepresentation of the facts. SGS is paid to produce accurate results. The integrity of the "system" depends on these independent labs producing results that investors can rely on. If SGS says the purity is 99.9% or higher ... then that's what it is. To think otherwise is, in my opinion, to basically say both Zenyatta and SGS have no integrity and are out to mislead people. I for one certainly do not believe that. Respectfully, BUMBLEBEE - As an Appendix to this long-winded piece, I will also attach the SGS Code of Ethics section on Integrity of Services ... All SGS services must be undertaken professionally and honestly in accordance with agreed standards, methods and policies. SGS maintains its independence of judgment and does not surrender to pressure and inducements to misrepresent findings or alter the results of its inspections, certifications, audits or testing. All findings must be adequately documented and no untruthful or misleading reports or certificates issued. All findings and results must be accurately documented and must not be changed improperly. Findings and opinions issued by SGS are supported by true and accurate job files and activity reports maintained in accordance with relevant SGS Group policies. |
28 Recommendations
Loading...
Loading...
New Message
Please
login
to post a reply