An assay is the standard procedure for determining the metal content of a mineral sample. Drill core, RC and channel chips as well as bulk samples are all assayed to determine the metal content, and therefore value, of the rock being evaluated for a mining operation. The problem is that not all assays are created equal. This can be illustrated by the strange example of Prophecy Platinum, no, I meant Wellgreen Platinum, no again, I meant Nickel Creek Platinum (TSX:NCP). Whatever they call themselves these days, the company is engaged in the development of the Wellgreen Ni-Cu-PGM deposit in the Yukon. And man, do these guys and gals have a lesson to teach us. It seems few people to this day seem to understand what, why or how the disappointment at Wellgreen happened ... possibly not even some of the people at the company. So I thought maybe it's time to pull back the curtain and look a bit closer.
I first questioned the assay results of what was then Prophecy Platinum in 2011 after the company acquired the Wellgreen project. A previous operator, Coronation Minerals, had published a resource estimate in 2008. It looked like this [Source: July 15, 2008 Technical Report, Watts, Griffis and McQuat]:
I'll just focus on the nickel grade for reasons that should shortly become obvious. It's 0.43% for the (small) Indicated resources and 0.29% for the (larger) Inferred. This resource is based on zones near the historical underground workings because the plan at that time was to evaluate an underground operation.
So far so good. Except Coronation quickly got bored with Wellgreen -- it simply didn't appear sexy enough for their tastes -- and so they decided to try their luck in sexier Guyana. Wellgreen was returned to the original owner, Northern Platinum, which promptly merged with John Lee's Prophecy Resource Corp.
None of the intervening activity at Wellgreen could possible presage the immense resource that was soon revealed by Prophecy in 2011. The "updated" resource was received with widespread acclaim as follows [Source: July 21, 2011 Technical Report, Wardrop]:
Holy wow! The Indicated resources have doubled in size but with a big pop in nickel grade to 0.69%. But the real action was in the Inferred resources, which exploded to a massive 289 million tonnes at 0.38% nickel (that's nearly the grade of the original, small Indicated resources). They must surely have done a lot of drilling between 2008 and 2011 to get such a massive increase, right?
Not really. Northern Platinum completed 16 holes for 4000 meters in 2009-2010 and Prophecy drilled one hole in 2010. Yes, one measly hole. So how is it possible that the resource grew by such a large multiple? Wardrop tried to explain, but as we'll see in a bit, this is just part of the deceptive story:
That's all fine and good, it's their prerogative. But there is a big problem. Look at the two resource tables and see if you can spot it before reading further. If you look carefully, you'll notice that the Inferred resources have grown massively larger by supposedly incorporating the "disseminated sulphides" (which by definition are lower grade than "semi-massive to massive" sulphides) and yet the nickel grades (copper too, but not by as much) have actually increased. This is very strange. Never in the history of the mining industry has such a magic act been pulled off whereby lower grade material increased both the tonnage AND the grade over what they started with! The very foundations of math require that it be one or the other, but not both.
And yet here was Prophecy management on a conference call telling me exactly that, even making the circular argument that the resource was higher grade simply because they used a higher cutoff. Okay, it's now 0.4% NiEq cutoff compared to 0.2% NiEq cutoff. But ... but ... uhmmm ... not only the NiEq increased (for reasons Wardrop quoted above). The Nickel grade itself increased! I tried to reason with Prophecy management on that conference call to no avail, they simply would not reveal the trick. Worse yet, the other analysts on that conference call seemed to prefer blissful ignorance as they counted all those billions of dollars of nickel, copper, platinum, oh my!
So how was this magic pulled off? A combination of fortuitous circumstances that likely has been repeated at other companies and projects though never in my recollection to such spectacular effect. And it seems to have its source with rock types and assay types.
First let's look at rock types. There are different ways a metal (nickel in this case) can be incorporated in the chemical structure of various types of rock. Simply put, some rocks have nickel that can't be recovered by standard processing. For example, the mineral olivine often has nickel in its molecular structure (giving olivine its green color), but olivine is not broken down during processing (milling and flotation) and therefore it (and the nickel locked up in olivine) ends up in the tailings, not the concentrates.
Nickel-rich olivine at Wellgreen is found more frequently in lower grade disseminated sulphides rather than in the semi-massive to massive sulphide zones (which mainly contain another silicate rock called gabbro that has a lot less olivine). By contrast, the nickel in the semi-massive to massive sulphide minerals is mainly contained in the sulphides that can be readily floated and recovered in sulphide concentrates. This probably explains why the inferred grade in the bulk disseminated portion of the Wellgreen deposit was boosted in nickel ... the magical "extra" nickel was in the olivine, not the (disseminated) sulphides.
But, what about the title of this commentary? What do assays have to do with all of this? Well, you might wonder why an assay would test for nickel if it can't be actually recovered during mining. And you'd be right to wonder that. There is a particular type of assay, called 4-acid or "near-total" digestion. The "near-total" refers to the fact that this method actually dissolves not only the ore-bearing minerals of interest (sulphides) but also the silicate host rocks (olivine is a silicate rock). And when we examine the sampling procedures for Wellgreen, we are informed that in fact both Coronation and Prophecy used 4-acid digestion in assaying for nickel. Only the short-lived Northern Platinum 16 hole drill program was assayed using partial digestion by Aqua Regia (which does not dissolve silicates). But this represents only a tiny portion of the drill database used for the resource model. So why was 4-acid digestion used? It's easier and more informative most of the time (except for nickel in silicates).
And it's not that Wardrop was not aware of potential problems with assaying nickel that could never be recovered from olivine silicate rocks during mining. In the Data Verification section of the 2011 Technical Report, they state:
Interestingly, subsequent technical report updates on Wellgreen no longer referred directly to the possibility of the 4-acid digestion problem of assaying nickel in silicates. The reports did continue to note, however, the consistent if small nickel grade shortfall in check assays (as well as cobalt, another metal that tends to be found in olivine). There was even a failure of the "Paired-T" statistical hypothesis for nickel and cobalt indicating a consistent difference between assay lab results. While both the original lab and the check assay lab used 4-acid digestion, they probably have slightly different protocols. In particular, they may use different digestion times and acid proportions resulting in different rates of "near-total" digestion. This could show up as consistent small assay differences for those metals found in silicates (nickel and cobalt), which is precisely what we see.
We had other subtle confirmation in subsequent technical reports that nickel in the silicate olivine (which is also found in peridotite and its altered state, serpentine) might be a problem:
Eventually, the mounting evidence could no longer be ignored when further mineralogical testing in 2017 revealed the obvious problem in retrospect, forcing at least the technical team to come partly clean:
Ahhh. Much better. Nickel grades back down to 0.26% to 0.28%. Consistent with the original, much smaller Coronation resources. Of course they had to obfuscate by changing the cutoff from a Nickel-Equivalent to NSR. Otherwise they risk the deception becoming clear to all.
So, in conclusion, here is what happened. Coronation built a resource on the higher grade underground zones in gabbro rocks where the nickel was predominantly found in sulphides. Then, Prophecy got hold of the project and published a resource using mainly the historical drilling, and used the raw 4-acid digestion assays from the larger bulk portion of the deposit in olivine rocks where a substantial portion of the nickel is locked up in the silicates. This resulted in both a big grade AND tonnage boost against all odds. Prophecy enjoyed a big market cap boost while the few people who actually try to understand things in mining were left scratching their heads. In the subsequent 7 years, the simple problem was slowly revealed and a lesson has been taught to those who care to listen and learn:
I first questioned the assay results of what was then Prophecy Platinum in 2011 after the company acquired the Wellgreen project. A previous operator, Coronation Minerals, had published a resource estimate in 2008. It looked like this [Source: July 15, 2008 Technical Report, Watts, Griffis and McQuat]:
I'll just focus on the nickel grade for reasons that should shortly become obvious. It's 0.43% for the (small) Indicated resources and 0.29% for the (larger) Inferred. This resource is based on zones near the historical underground workings because the plan at that time was to evaluate an underground operation.
So far so good. Except Coronation quickly got bored with Wellgreen -- it simply didn't appear sexy enough for their tastes -- and so they decided to try their luck in sexier Guyana. Wellgreen was returned to the original owner, Northern Platinum, which promptly merged with John Lee's Prophecy Resource Corp.
None of the intervening activity at Wellgreen could possible presage the immense resource that was soon revealed by Prophecy in 2011. The "updated" resource was received with widespread acclaim as follows [Source: July 21, 2011 Technical Report, Wardrop]:
Holy wow! The Indicated resources have doubled in size but with a big pop in nickel grade to 0.69%. But the real action was in the Inferred resources, which exploded to a massive 289 million tonnes at 0.38% nickel (that's nearly the grade of the original, small Indicated resources). They must surely have done a lot of drilling between 2008 and 2011 to get such a massive increase, right?
Not really. Northern Platinum completed 16 holes for 4000 meters in 2009-2010 and Prophecy drilled one hole in 2010. Yes, one measly hole. So how is it possible that the resource grew by such a large multiple? Wardrop tried to explain, but as we'll see in a bit, this is just part of the deceptive story:
"The fundamental difference between the 2008 WGM resource and the 2011 Wardrop resource is that the 2008 WGM resource concentrated on the resource material in close proximity to the semi-massive to massive nickel sulphide pods that had been exploited by Hudson Yukon compared to the 2011 Wardrop resource targeting a large tonnage low grade open pit geological model. In the 2010 Wardrop technical report (McCracken, 2010), the potential of a large tonnage, low grade resource was identified by the presence of a significant amount of hanging wall material, up to 350 m core length, of very fine grained disseminated sulphides which contain a considerable amount of platinum, palladium and gold. Although this material had been assayed in the past, it was not incorporated in the 2008 WGM resource model. Using this new geological model as well as with the inclusion of cobalt, gold, platinum and palladium to the Nieq formula, which was not incorporated n the WGM Nieq value, drastically increases the volume of material in the resource estimation compared to the 2008 WGM resource. "In plain English, Prophecy changed the model from an underground mine targeting the higher grade semi-massive to massive sulphide mineralization to a bulk scale project that includes disseminated sulphides over a large area that could potentially be mined as an open pit.
That's all fine and good, it's their prerogative. But there is a big problem. Look at the two resource tables and see if you can spot it before reading further. If you look carefully, you'll notice that the Inferred resources have grown massively larger by supposedly incorporating the "disseminated sulphides" (which by definition are lower grade than "semi-massive to massive" sulphides) and yet the nickel grades (copper too, but not by as much) have actually increased. This is very strange. Never in the history of the mining industry has such a magic act been pulled off whereby lower grade material increased both the tonnage AND the grade over what they started with! The very foundations of math require that it be one or the other, but not both.
And yet here was Prophecy management on a conference call telling me exactly that, even making the circular argument that the resource was higher grade simply because they used a higher cutoff. Okay, it's now 0.4% NiEq cutoff compared to 0.2% NiEq cutoff. But ... but ... uhmmm ... not only the NiEq increased (for reasons Wardrop quoted above). The Nickel grade itself increased! I tried to reason with Prophecy management on that conference call to no avail, they simply would not reveal the trick. Worse yet, the other analysts on that conference call seemed to prefer blissful ignorance as they counted all those billions of dollars of nickel, copper, platinum, oh my!
So how was this magic pulled off? A combination of fortuitous circumstances that likely has been repeated at other companies and projects though never in my recollection to such spectacular effect. And it seems to have its source with rock types and assay types.
First let's look at rock types. There are different ways a metal (nickel in this case) can be incorporated in the chemical structure of various types of rock. Simply put, some rocks have nickel that can't be recovered by standard processing. For example, the mineral olivine often has nickel in its molecular structure (giving olivine its green color), but olivine is not broken down during processing (milling and flotation) and therefore it (and the nickel locked up in olivine) ends up in the tailings, not the concentrates.
Nickel-rich olivine at Wellgreen is found more frequently in lower grade disseminated sulphides rather than in the semi-massive to massive sulphide zones (which mainly contain another silicate rock called gabbro that has a lot less olivine). By contrast, the nickel in the semi-massive to massive sulphide minerals is mainly contained in the sulphides that can be readily floated and recovered in sulphide concentrates. This probably explains why the inferred grade in the bulk disseminated portion of the Wellgreen deposit was boosted in nickel ... the magical "extra" nickel was in the olivine, not the (disseminated) sulphides.
But, what about the title of this commentary? What do assays have to do with all of this? Well, you might wonder why an assay would test for nickel if it can't be actually recovered during mining. And you'd be right to wonder that. There is a particular type of assay, called 4-acid or "near-total" digestion. The "near-total" refers to the fact that this method actually dissolves not only the ore-bearing minerals of interest (sulphides) but also the silicate host rocks (olivine is a silicate rock). And when we examine the sampling procedures for Wellgreen, we are informed that in fact both Coronation and Prophecy used 4-acid digestion in assaying for nickel. Only the short-lived Northern Platinum 16 hole drill program was assayed using partial digestion by Aqua Regia (which does not dissolve silicates). But this represents only a tiny portion of the drill database used for the resource model. So why was 4-acid digestion used? It's easier and more informative most of the time (except for nickel in silicates).
And it's not that Wardrop was not aware of potential problems with assaying nickel that could never be recovered from olivine silicate rocks during mining. In the Data Verification section of the 2011 Technical Report, they state:
"A NI-OG46 package was completed on the samples as well. This was to test for nickel in sulphides using a partial digestion (aqua regia). The partial digestion method will easily digest all the sulphides in the sample and only a minor amount of the silicate minerals may get digested. This will provide an idea as to the amount of nickel that is in the sulphides compared to how much nickel is in the silicate minerals such as olivine. The samples returned comparable results to the previous assays and indicate that over 90% of the nickel is likely found within the sulphides as showing in Table 12.2."Unfortunately though, if we look at the data in Table 12.2 we can't determine how representative the check assays were. We do find that in 2 out of the 8 samples the partial digestion resulted in significantly lower nickel compared to the Prophecy assay (0.494% vs. 0.643% and 2.21% vs. 2.83%). Without knowing where in the deposit the 8 samples came from, however, it is not possible to conclude that "over 90% of the nickel is likely found within the sulphides." It's more probable that over 90% of the nickel in the semi-massive to massive sulfides hosted by gabbro -- also a silicate rock but one that doesn't have a lot of olivine -- is likely found within the sulphides. As for nickel in the olivine host rocks? It's quite possible less than 90% of nickel is in sulphides.
Interestingly, subsequent technical report updates on Wellgreen no longer referred directly to the possibility of the 4-acid digestion problem of assaying nickel in silicates. The reports did continue to note, however, the consistent if small nickel grade shortfall in check assays (as well as cobalt, another metal that tends to be found in olivine). There was even a failure of the "Paired-T" statistical hypothesis for nickel and cobalt indicating a consistent difference between assay lab results. While both the original lab and the check assay lab used 4-acid digestion, they probably have slightly different protocols. In particular, they may use different digestion times and acid proportions resulting in different rates of "near-total" digestion. This could show up as consistent small assay differences for those metals found in silicates (nickel and cobalt), which is precisely what we see.
We had other subtle confirmation in subsequent technical reports that nickel in the silicate olivine (which is also found in peridotite and its altered state, serpentine) might be a problem:
"Mineralogical study of the composites indicated that peridotite domain samples could be characterized by high serpentine content, lower nickel grades, and a lower portion of the contained nickel present as pentlandite, as compared to samples from the gabbro domain."Moreover, metallurgical recovery of nickel to concentrates continues to be a vexing problem at Wellgreen to this day. The most likely explanation for this is that a large amount of nickel is locked up in silicates and therefore lost to the tailings.
Eventually, the mounting evidence could no longer be ignored when further mineralogical testing in 2017 revealed the obvious problem in retrospect, forcing at least the technical team to come partly clean:
"Analysis of the peridotite composites indicated that the domain is approximately 60-70% serpentine by mass. Other gangue minerals include actinolite, clinopyroxenite, chlorite, magnetite and Cr-spinels. EPMA revealed that nickel is found in trace amounts in solid-solution in most of these minerals, and this contributes to the non-recoverable nickel in the sample. The estimated grade of non-recoverable nickel in the peridotite domain ranges from 0.06% to 0.11%."All said and done, what does the latest resource tell us? Here it is (Source: June 26, 2017 Technical Report, Independent Mining Consultants):
Ahhh. Much better. Nickel grades back down to 0.26% to 0.28%. Consistent with the original, much smaller Coronation resources. Of course they had to obfuscate by changing the cutoff from a Nickel-Equivalent to NSR. Otherwise they risk the deception becoming clear to all.
So, in conclusion, here is what happened. Coronation built a resource on the higher grade underground zones in gabbro rocks where the nickel was predominantly found in sulphides. Then, Prophecy got hold of the project and published a resource using mainly the historical drilling, and used the raw 4-acid digestion assays from the larger bulk portion of the deposit in olivine rocks where a substantial portion of the nickel is locked up in the silicates. This resulted in both a big grade AND tonnage boost against all odds. Prophecy enjoyed a big market cap boost while the few people who actually try to understand things in mining were left scratching their heads. In the subsequent 7 years, the simple problem was slowly revealed and a lesson has been taught to those who care to listen and learn:
IGNORE 4-ACID DIGESTION ASSAY RESULTS OF NICKEL IN OLIVINE HOST ROCKS. IT'S PROBABLY A SCAM.
The End.
Comments
Post a Comment