Criticism of RATE’s helium diffusion data
From CreationWiki, the encyclopedia of creation science
The RATE group’s helium diffusion work is the biggest threat the old Earth model has ever faced, since it strikes a fatal blow to radiometric dating. So it is no surprise that Anticreationists have attacked it viciously. Many of these attacks are personal attacks against Russell Humphreys himself. The degree of viciousness shows an intense desire to discredit this work at all costs, particularly the truth.
Lack of Publishing In Non-Creationist Peer-reviewed Journals
It had been said that Humphreys et al had not publish their work in non creationist peer reviewed journals. First of all, this is a slam on the quality of the peer review process in journals like Journal of Creation and the Creation Research Society Quarterly. The notion is that they are biased and of generally poor quality. This also ignores the bias against creation in mainstream journals. Anticreationists will often insult creationist journals when they refer to mainstream journals as “legitimized”, “real”, or “authentic” scientific journals, implying that creationist journals are not legitimate, real, or authentic scientific journals.
That said, the simple fact is that Humphreys et al presented a paper on their work at the American Geophysical Union 2003 annual meeting. To get the opportunity to present this paper it had to pass peer review by mainstream geologists. So Humphreys’ work has been successfully peer reviewed by secular scientists. Radioisotopes and the Age of the Earth, Volume II pg 25 
Humphreys et al Misidentified Their Rock Samples
This claim is based on the term “Jemez granodiorite” a descriptive name used by Humphreys for lack of an existing name for the formation. It is claimed that some of the the rocks are gneiss and others are granodiorite.
According to Dr. Baumgardner; a geophysicist and one of Humphreys co authors; based on a personal examination of the GT-2 core from Fenton Hill:
Yes, there are occasional veins of material other than the coarse-grained granodiorite that forms the vast majority of the core. In making the selections I made of what samples to use, I purposely avoided these occasional veins. In fact I tried to select sections of the core well removed from such veins. So at least from my vantage point, the samples of core we used for the helium diffusion measurements were indeed coarse-grained granodiorite, not gneiss. 
The claim that some of the rocks are gneiss is based on a couple of papers, one of which has a labeling of the upper portion as gneiss, but this conflicts with a paper from Los Alomos labs that shows much of that portion to be granite and granodiorite. This paper clearly shows that Gentry’s samples 1, from a depth of 960 m (3149.6 ft), was in granodiorite. Now it does label the rock of RATE’s two new samples, depth 750 m (2461 ft) and 1490 m (4888ft) core, as granite. But that does not mean that Dr. Baumgardner was wrong. First of all, the word granite  is often used refer to rocks other than true granite and so this may be granite but may contain granodiorite as well. Furthermore granite and granodiorite are almost the same type of rock.
Granodiorite: An intrusive igneous rock similar to granite, but contains more plagioclase than potassium feldspar. It usually contains abundant biotite mica and hornblende, giving it a darker appearance than true granite. Mica may be present in well-formed hexagonal crystals, and hornblende may appear as needle-like crystals. 
Not only does this show how similar granite and granodiorite are but is also shows that contains abundant biotite mica making even a small vein the most likely pace to find biotite mica samples.
This claim takes several forms but the basic idea is the same. The idea is that helium found in RATE’s zircons is a result of outside contamination. However Humphreys had already considered this possibility and eliminated it.
- The He concentration of the zircons was 1320 nmol/cm3 and the He concentration of the surrounding biotite is only 6.6 nmol/cm3. This means that the He concentration of the zircons is 200 times that of the biotite. Since diffusion only goes from higher to lower concentrations, this fact alone eliminates the possibility of contamination
- The amount of He in the surrounding biotite is of the same order of magnitude as that lost by the zircons.
In addition to this Q/Q0 decreases with depth as predicted by the zircons being the source, on the other hand, contamination would tend to produce the opposite pattern since the deeper zircons would have higher diffusion rates, it would tend to accumulate quicker in those at shallower depths.  Radioisotopes and the Age of the Earth, Volume II pg 59 
Fudging Soviet Helium Diffusion Data
This is mainly a personal attack against Humphreys since Magomedov’s data was only used to help set up the model.
When RATE first started this research, Magomedov’s zircon diffusion data was all that was available to use for developing a model for their research. However, Magomedov’s zircon diffusion data as presented was somewhat ambiguous. Rather than placing it in a chart, it was given only in a graph with the diffusion rate scaled to ln(D). In most mathematic and scientific circles, this would be a reference to a natural logarithm and Humphreys interpreted it as such. The extremely high rates would take only 23 years to reach 58% helium retention. This caused them to think that the biotite mica was the main hindrance to helium diffusion out of zircons and so RATE began what turns out to be the first known diffusion rate measurements of biotite mica. These experiments confirmed that the diffusion rates in biotite were several orders of magnitude lower than Humphreys thought Magomedov’s data indicated for zircons. This confirmed that, if zircon diffusion rates were that high, the biotite diffusion rate would be the major factor in determining a helium diffusion date out of the zircon.
In 2001 Humphreys et al received a paper by Reiners et al that published newer zircon diffusion data from Nevada. Humphreys noticed that these new zircon diffusion rates were orders of magnitude slower than he thought Magomedov’s were. Because is some Russian circles “ln” is used to denote a common logarithm, Humphreys considered the possibility that Magomedov was using a common (base 10) logarithm and found that it was consistent with the Nevada data, and later their own data from Fenton Hill. As a result, Humphreys concluded that he had erred in interpreting ln(D) as a natural logarithm rather than a common logarithm. Even if wrong, Humphreys was not fudging, but correcting what he thought was misinterpretation, because the data is somewhat ambiguous. The worst Humphreys would be guilty of is an honest mistake. Furthermore, Humphreys et al does not use Magomedov’s data in their calculations, but only mention it in more recent papers to explain the change in the model.
This argument rests on the assumption the natural logarithm is correct and that Magomedov’s stated activation energy of 15 kcal/mole applies the natural logarithm verson. To test the claim it was necessary to estimate the values based only on graphs. The following tables provide the estimated values for both temperature and diffusion rates.
Estimated Magomedov’s data - natural logarithm
Estimated Magomedov’s data - common logarithm
- Diffusion formula
- The universal gas constant (R) is 1.986 calories per mole-kelvin
To find the activation energy (E) one plugs the diffusion rate (D) and temperature in Kelvin (T) of two points into the diffusion formula and then uses them to solve for E.
Using samples 2 and 5 to calculate the activation energy, the intrinsic line yields 17.15 kcal/mole for the natural logarithm version and 39.94 kcal/mole for the common logarithm version.
At first, this would seem to show the claim correct on this point, but that assumes the Magomedov was using the intrinsic line. The fact that neither version yields ~15 kcal/mole suggests that Magomedov may have been using a best fit line. Now the best fit line that can be drawn between two data points would be between samples 3 and 7. When this is used to estimate the activation energy, the result is 7.29 kcal/mole for the natural logarithm version and 15.53 kcal/mole for common logarithm version. So it would seem that the common logarithm is correct and that the 15 kcal/mole activation energy is a result of the use of a best fit line.
The result is that Humphreys was correct to switch from using the natural logarithm to common logarithm for interpreting Magomedov’s data.
Helium Diffusion Experiments Performed In A Vacuum
According to this claim, the results of Humphreys et al are invalid because the helium diffusion experiments were performed in a vacuum and not under the subsurface pressures found at Fenton Hill. The claim is that under pressure the diffusion rates would be decreased by several orders of magnitude.
Actually, Humphreys had already considered this problem, and after researching it, he found that in the case he was dealing with, the difference in pressure would affect the diffusion rates by less than 1%, placing it within the margin of error.
This claim is based on argon diffusion experiments done under pressure in soft mica, but this does not match Humphreys' experiments. Zircons are very hard and compress very little under pressure. Furthermore, argon is both an order of magnitude larger and heavier than helium, these factors would nearly eliminate any pressure effect under the conditions found at Fenton Hill.
The argon diffusion experiments were done using layered soft mica. Under pressure the layers of mica are easily squeezed together, reducing diffusion between the layers. Furthermore the pressure was supplied by water, and not the weight of overlaying rock. It turns out that the water not only supplied pressure, but it also gets between the layers of mica and chemically bonds with it, thereby reducing the diffusion rate considerably more. In fact, the water content of mica had far more of an effect on diffusion than did the pressure.
So the experiments on biotite mica without water showed a much lower change. At a pressure of 14 kilobars, the diffusion was reduced by about two orders of magnitude. At a pressure of 1 kilobar, the effect comes to about one order of magnitude. 
It is further suggested that under pressure defects in the zircon crystals would close, eliminating the defect bend. Not only is there no evidence to support this claim, but there is evidence against it. First of all, most of the defects in zircons are radiation damage that would not close under pressure. Second, original zircons used to predict the diffusion rates not only show a defect bend, but accurately predicted its location and slope, thus showing that defect bend did occur under pressure in situ.
Humphreys Changed Robert Gentry Helium Retention Data
While Dr. Humphreys did correct two typographical errors in the absolute amounts of helium retained it, the correction were done after a consulting Gentry, who agreed with the corrections. Furthermore the corrections did not affect Humphreys et al’s results since they were based on the fraction of retention; which were correct and not the absolute retention.
The errors were discovered when Humphreys was doing the retention calculations for RATES sample. He noticed an order of magnitude discrepancy in the absolute helium amounts. When he contacted Gentry, Gentry agreed that they probably were typographical errors.
There is nothing devious about this, it is simply a case were an error was made, discovered and corrected. 
Samples Five and Six
The criticism here is that Humphreys et al  used sample 5 while discarding sample 6 of Gentry’s zircons when they both should have been treated the same because they retained the same percentage of Helium.
The reason for discarding sample 6 is that at its original temperature and depth, with the recorded concentrations of Helium, it would have been in equilibrium with its surrounding biotite, whereas sample 5 would not. This fact was demonstrated by how well sample 5 fits with the other data points. It is also supported by total amount of helium found in sample 6.
Finally discarding sample 5 would not significantly affect the results since the resulting age of the other three samples would be 5133 ± 3500 years. This places it within the margin of error of Humphrey’s age of 6000 ± 2000 years. The fact that eliminating sample 5 increases the margin of error further supports its inclusion. Also the date without sample 5 would still be consistent with accelerated decay during both creation and the Genesis Flood, while being totally incompatible with an old Earth.  
Humphreys‘ Formulas Produce Inconsistent Data
Any formula in science is only valid within the model from which it is derived. If data is placed in a formula from the wrong model the results will be bogus. For example placing the observed diffusion rates into the formula from Humphreys’ Uniformitarian model for the zircons would naturally produce erroneous dates since that model assumes a greater difference between the diffusion rates between the zircons and biotite than is actually observed. Likewise plugging Humphreys’ original value for zircon radius (a=22μm), presented in his first model, into the formula used in his second model instead of its a=30μm, would produce erroneous dates because once again the wrong data is being plugged into the model.
The accuracy of Humphreys’ creation model and its formula is supported by how accurately it predicted the diffusion rates in their respective temperature ranges.
Humphreys et al Used the Wrong Q0
This claim seems to be based mainly on a misunderstanding of Gentry et al’s description of the percentage of alpha particles (helium nuclei) escaping the zircons immediately as a result of being near the surface. The claim assumes a Q0 = 41 ncc/µg while Humphreys et al used a Q0 = 15 ncc/µ. The difference in these values seems to be mainly a result of different values for initial escaping alpha particles.
If Q0 = 41 ncc/µg were correct the result on the time calculation would be about 2 orders of magnitude or 600,000 years. In a young Earth model this could be accounted for by a heating event, but it does not even come close to being constant with 1.5 billion years.
That said, the test of assumptions on science is the ability of the resulting model to make accurate prediction and Humphreys et al’s predictions of helium diffusion rates nearly perfect. 
- HELIUM DIFFUSION RATES SUPPORT ACCELERATED NUCLEAR DECAY.
- Helium Evidence for A Young World Remains Crystal-Clear
- Recently Measured Helium Diffusion Rate for Zircon Suggests Inconsistency With U-Pb Age for Fenton Hill Granodiorite
- CVO Website - Granite
- W. D. Purtymun et all, Geology of Geothermal Test Hole GT-2 Fenton Hill Site, July 1974.
- Geology and Geochemistry of Samples from Los Alamos National Laboratory HDR Well EE-2, Fenton Hill
- Helium Diffusion Age of 6,000 Years Supports Accelerated Nuclear Decay
- Helium Evidence for A Young World Overcomes Pressure
- CVO Website - Granodiorite
- Have the decay rates been constant?