08-08-2011, 03:42 PM
Contrasting uniformity and non-uniformity of WTC dust sample results of the Harrit et al. paper "Active Thermitic Material Discovered in Dust from the 9/11 World Trade Center Catastrophe"
http://www.bentham-open.org/pages/conten...7TOCPJ.SGM
T Mark Hightower
8/8/2011
ABSTRACT
Although the Harrit et al. paper claims great uniformity of results among all samples based on all the tests apart from the Differential Scanning Calorimeter (DSC) tests, this is in significant contrast to the tremendous lack of uniformity reflected in the DSC results.
Also, although the paper contains a rather doubtful statement regarding its findings in one part of the paper, it concludes with a strikingly confident conclusion in the final statement of the paper.
THE UNCERTAINTY OF THE DIFFERENTIAL SCANNING CALORIMETER (DSC) RESULTS
There are some weaknesses in the Differential Scanning Calorimeter (DSC) data that is presented in the paper (Harrit et al.) for the 4 dust samples.
There were 4 dust samples tested. Page 9 of the paper identifies them as
1 MacKinlay
2 Delassio/Breidenbach
3 Intermount
4 White
The DSC data is presented in the text of page 19, section 3, which refers to graphical results of Fig. 19 on page 20. One of the DSC traces is compared to a trace of published nanothermite data in Fig. 29 on page 25. The results are presented in Fig. 30 on page 27 in the form of bar graphs reporting units of kJ/g.
There is an oddity I want to point out before I get into the first weakness.
The data referred to on page 19 and in Fig. 19 appears to have not included sample 2 Delassio/Breidenbach, but instead has a MacKinlay 2 sample in its place to give a total of 4 samples tested. In Fig. 30 the 4 samples are clearly labeled as 1, 2, 3, and 4, so here it seems to imply that the 2 Delassio/Breidenbach sample is included. Interestingly, the numerical value of 3 kJ/g given in the text on page 19 for the MacKinlay 2 sample of Fig. 19 is the same or at least close to the same as the 2 (presumably) Delassio/Breidenbach sample of Fig. 30. Clarification from the authors should be sought to clear up this confusion.
So there were 4 separate dust samples, with multiple red/gray chips in each sample.
I will use the numerical values of energy release given in the text (page 19) as representing the values in the bar graph of Fig. 30 for the 4 WTC chip samples. These would be
Sample 1: 1.5 kJ/g
Sample 2: 3 kJ/g
Sample 3: 7.5 kJ/g
Sample 4: 6 kJ/g
As these four DSC data points are all we have, it is of note that there is tremendous scatter in this data. The average value is 4.5 and the standard deviation is 2.7. As two standard deviations is usually what is used when referring to a value being +/- some uncertainty, in rough terms, we would then say that the DSC data gives an average value of 4.5 kJ/g, with an uncertainty of +/- 100%.
It appears that only one red/gray chip was selected from each sample for DSC testing. If DSC tests had been done separately for multiple chips in each sample, then the question of whether the scatter in the data was present similarly within each of the samples could also have been addressed. You cannot find what you do not look for.
The paper offers some possible explanations for the scatter in the data. From page 19, section 3, it states,
"Variations in peak height as well as yield estimates are not surprising, since the mass used to determine the scale of the signal, shown in the DSC traces, included the mass of the gray layer. The gray layer was found to consist mostly of iron oxide so that it probably does not contribute to the exotherm, and yet this layer varies greatly in mass from chip to chip."
Page 29, Conclusion 10. offers an explanation for higher total energy release than can be explained by the classic thermite reaction (true for samples 3 & 4)
"The carbon content of the red material indicates that an organic substance is present. This would be expected for super-thermite formulations in order to produce high gas pressures upon ignition and thus make them explosive. The nature of the organic material in these chips merits further exploration. We note that it is likely also an energetic material, in that the total energy release sometimes observed in DSC tests exceeds the theoretical maximum energy of the classic thermite reaction."
If citing gray layer variation and organic content is not enough to explain the tremendous variation in the results, another explanation is added on page 27, section 6. I will quote an entire paragraph so you can appreciate the context.
"It is striking that some of the red/gray chips release more energy in kJ/g than does ordinary thermite, as shown in the blue bar graphs above. The theoretical maximum for thermite is 3.9 kJ/g [27]. We suggest that the organic material in evidence in the red/gray chips is also highly energetic, most likely producing gas to provide explosive pressure. Again, conventional thermite is regarded as an incendiary whereas super-thermite, which may include organic ingredients for rapid gas generation, is considered a pyrotechnic or explosive [6, 24]. As this test was done in air it is possible that some of the enhancement of energy output may have come from air oxidation of the organic component."
So the DSC tests were done in air so extra oxygen was present to help liberate energy from any organics that might be present.
I know there is very little data to go on, just 4 DSC scans of red/gray chips, but with the scatter in the data and the explanations offered to explain it, I get the impression that the red/gray chips are tremendously lacking in uniformity. Or else there is much inherent error in the experimental apparatus.
By having air and therefore a source of oxygen present in the DSC seems to be an error in method since it allows for the input of energy from outside the substance that is itself being measured for its energy content. From Fig. 30 Chip 3 liberates considerably more energy than the high explosive HMX, and Chip 4 also exceeds the HMX value, but only by a little. The HMX provides its own oxygen within its chemical makeup, so it would not depend upon additional oxygen present to liberate its full energy. (Be sure not to confuse energy release with detonation velocity, an issue I am not dealing with in this write-up.)
I would like to see what a DSC trace of pure HMX would look like. It would probably look quite different because it would start to release its energy at a lower temperature, the deflagration temperature of HMX being 287 deg C. (page 238, "Explosives," 6th edition, Meyer et al., 2007)
In summary, the DSC data is extremely limited with much scatter and has a potential method error. Drawing firm conclusions from it is extremely dubious.
QUANTIFYING EXPLANATIONS FOR SCATTER IN DSC DATA
Variability in proportions of gray layer within the red/gray chips and organics present in the red layer are cited as explanations for the scatter in the DSC data. Let's quantify these explanations to see what kind of variability of the specimens might account for the scatter in the data.
The energy release for thermite is cited as 3.9 kJ/g. For the lowest value from the DSC tests, 1.5 kJ/g for sample 1, let's first assume for the sake of illustration that this specimen had a low value because it had no organics in the red layer (in other words it is essentially pure thermite), and the gray layer being predominantly iron oxide as the paper says, acted as excess reactant and therefore was essentially inert providing no energy in the DSC test. In this case, the red layer would have to be present at 38 % by weight and the gray layer present at 62 % within the specimen. The math is 1.5/3.9 = 0.38. In summary, this would be assuming no organics, 38% red layer, and 62% gray layer.
The above calculation is done as a base case for comparison, even though its assumption of no organics in the red layer goes against the major thrust of the conclusion of the Harrit et al. paper that the red layer is a form of nanothermite that includes organics.
The next calculation is an attempt to quantify the high end of the DSC data, 7.5 kJ/g for sample 3. The presence of organics is cited as the explanation for the high value by the paper. Pure high explosive HMX has a energy release of 5.2 kJ/g, so even if sample 3 were 100 % HMX, this could not account for the higher value of 7.5 kJ/g obtained by the DSC test. So, for the sake of illustration, I am going to assume that the organic present in the red layer has an energy release of twice that of HMX, or 10.4 kJ/g. To help account for the high value of 7.5 kJ/g for sample 3, I am also going to assume that it contains no gray layer. In other words I am going to assume that sample 3 is 100 % red layer material. Setting x = weight fraction thermite in the red layer, the math is 3.9x + 10.4(1-x) = 7.5. Solving for x gives 0.45. So based on the above assumptions the red layer would have to contain 45 % thermite by weight and 55 % organic. In summary, this would be 100% red layer, no gray layer, with the red layer made up of 45% thermitic material and 55% organics.
Let's now go back to the low value case, 1.5 kJ/g for sample 1, and assume that it is made up of red layer (containing 45 % thermite and 55 % organic from the previous calculation) plus gray layer of inert excess iron oxide reactant. In this case then, the specimen would have to contain 20 % red layer and 80 % gray layer. The math is 1.5/7.5 = 0.2. In summary, this would be 20% red layer (made up of 45% thermite and 55% organic) and 80% gray layer.
Therefore, a very high degree of variability among the red/gray chips is necessary to explain the scatter in the DSC test data, unless there is significant error in the experimental apparatus and technique.
This is in contrast to these statements from the paper concerning the great uniformity of results.
From page 15, right column, it states, "From these data, it is determined that the red/gray chips from different WTC dust samples are extremely similar in their chemical and structural makeup. It is also shown that within the red layer there is an intimate mixing of the Fe-rich grains and Al/Si plate-like particles and that these particles are embedded in a carbon-rich matrix."
From page 23, upper left column, it states, "The results clearly show the similarities of the red/gray chips from the different dust samples from all four sites."
ANOTHER WEAKNESS IN THE DSC DATA
On page 25 of the paper, it says
"The red layer of the red/gray chips is most interesting in that it contains aluminum, iron and oxygen components which are intimately mixed at a scale of approximately 100 nanometers (nm) or less. Now we compare a DSC trace obtained for a WTC red/gray chip with a DSC trace obtained for known super-thermite (see Fig. (29))."
Note that in Fig. 29 the trace of the WTC sample is really not all that similar to the known super-thermite. The WTC sample cited in Fig. 29 is the MacKinlay sample, although it does not say whether it is the MacKinlay 1 or MacKinlay 2 sample. But from Fig. 19, where 4 WTC DSC traces are plotted, for the four samples, MacKinlay 1, MacKinlay 2, Intermont, and White, the two MacKinlay samples are the lowest energy release traces. The other two, Intermont (sample 3), and White (sample 4) have much higher energy release, and deviate even more from the known super-thermite trace of Fig. 29.
STATISTICAL WEAKNESS OF THE DATA PRESENTED IN THE PAPER
With the tremendous scatter in the DSC data presented in the paper, and its implication for extreme non uniformity of the red/gray chips that I have tried to quantify, the question needs to be raised on all of the other tests that were performed to characterize the red/gray chips, and whether additional tests should have been done on other of the chips within the samples, to see if those tests also would have shown such high levels of non uniformity.
OTHER QUOTES OF NOTE FROM THE PAPER
From page 25 of the paper, right column, first paragraph, the final sentence is
"We make no attempt to specify the particular form of nano-thermite present until more is learned about the red material and especially about the nature of the organic material it contains." This statement expresses a lot of doubt about the findings.
From page 29 of the paper, final paragraph.
"Based on these observations, we conclude that the red layer of the red/gray chips we have discovered in the WTC dust is active, unreacted thermitic material, incorporating nanotechnology, and is a highly energetic pyrotechnic or explosive material." This final conclusion statement of the paper expresses much less doubt.
http://www.bentham-open.org/pages/conten...7TOCPJ.SGM
T Mark Hightower
8/8/2011
ABSTRACT
Although the Harrit et al. paper claims great uniformity of results among all samples based on all the tests apart from the Differential Scanning Calorimeter (DSC) tests, this is in significant contrast to the tremendous lack of uniformity reflected in the DSC results.
Also, although the paper contains a rather doubtful statement regarding its findings in one part of the paper, it concludes with a strikingly confident conclusion in the final statement of the paper.
THE UNCERTAINTY OF THE DIFFERENTIAL SCANNING CALORIMETER (DSC) RESULTS
There are some weaknesses in the Differential Scanning Calorimeter (DSC) data that is presented in the paper (Harrit et al.) for the 4 dust samples.
There were 4 dust samples tested. Page 9 of the paper identifies them as
1 MacKinlay
2 Delassio/Breidenbach
3 Intermount
4 White
The DSC data is presented in the text of page 19, section 3, which refers to graphical results of Fig. 19 on page 20. One of the DSC traces is compared to a trace of published nanothermite data in Fig. 29 on page 25. The results are presented in Fig. 30 on page 27 in the form of bar graphs reporting units of kJ/g.
There is an oddity I want to point out before I get into the first weakness.
The data referred to on page 19 and in Fig. 19 appears to have not included sample 2 Delassio/Breidenbach, but instead has a MacKinlay 2 sample in its place to give a total of 4 samples tested. In Fig. 30 the 4 samples are clearly labeled as 1, 2, 3, and 4, so here it seems to imply that the 2 Delassio/Breidenbach sample is included. Interestingly, the numerical value of 3 kJ/g given in the text on page 19 for the MacKinlay 2 sample of Fig. 19 is the same or at least close to the same as the 2 (presumably) Delassio/Breidenbach sample of Fig. 30. Clarification from the authors should be sought to clear up this confusion.
So there were 4 separate dust samples, with multiple red/gray chips in each sample.
I will use the numerical values of energy release given in the text (page 19) as representing the values in the bar graph of Fig. 30 for the 4 WTC chip samples. These would be
Sample 1: 1.5 kJ/g
Sample 2: 3 kJ/g
Sample 3: 7.5 kJ/g
Sample 4: 6 kJ/g
As these four DSC data points are all we have, it is of note that there is tremendous scatter in this data. The average value is 4.5 and the standard deviation is 2.7. As two standard deviations is usually what is used when referring to a value being +/- some uncertainty, in rough terms, we would then say that the DSC data gives an average value of 4.5 kJ/g, with an uncertainty of +/- 100%.
It appears that only one red/gray chip was selected from each sample for DSC testing. If DSC tests had been done separately for multiple chips in each sample, then the question of whether the scatter in the data was present similarly within each of the samples could also have been addressed. You cannot find what you do not look for.
The paper offers some possible explanations for the scatter in the data. From page 19, section 3, it states,
"Variations in peak height as well as yield estimates are not surprising, since the mass used to determine the scale of the signal, shown in the DSC traces, included the mass of the gray layer. The gray layer was found to consist mostly of iron oxide so that it probably does not contribute to the exotherm, and yet this layer varies greatly in mass from chip to chip."
Page 29, Conclusion 10. offers an explanation for higher total energy release than can be explained by the classic thermite reaction (true for samples 3 & 4)
"The carbon content of the red material indicates that an organic substance is present. This would be expected for super-thermite formulations in order to produce high gas pressures upon ignition and thus make them explosive. The nature of the organic material in these chips merits further exploration. We note that it is likely also an energetic material, in that the total energy release sometimes observed in DSC tests exceeds the theoretical maximum energy of the classic thermite reaction."
If citing gray layer variation and organic content is not enough to explain the tremendous variation in the results, another explanation is added on page 27, section 6. I will quote an entire paragraph so you can appreciate the context.
"It is striking that some of the red/gray chips release more energy in kJ/g than does ordinary thermite, as shown in the blue bar graphs above. The theoretical maximum for thermite is 3.9 kJ/g [27]. We suggest that the organic material in evidence in the red/gray chips is also highly energetic, most likely producing gas to provide explosive pressure. Again, conventional thermite is regarded as an incendiary whereas super-thermite, which may include organic ingredients for rapid gas generation, is considered a pyrotechnic or explosive [6, 24]. As this test was done in air it is possible that some of the enhancement of energy output may have come from air oxidation of the organic component."
So the DSC tests were done in air so extra oxygen was present to help liberate energy from any organics that might be present.
I know there is very little data to go on, just 4 DSC scans of red/gray chips, but with the scatter in the data and the explanations offered to explain it, I get the impression that the red/gray chips are tremendously lacking in uniformity. Or else there is much inherent error in the experimental apparatus.
By having air and therefore a source of oxygen present in the DSC seems to be an error in method since it allows for the input of energy from outside the substance that is itself being measured for its energy content. From Fig. 30 Chip 3 liberates considerably more energy than the high explosive HMX, and Chip 4 also exceeds the HMX value, but only by a little. The HMX provides its own oxygen within its chemical makeup, so it would not depend upon additional oxygen present to liberate its full energy. (Be sure not to confuse energy release with detonation velocity, an issue I am not dealing with in this write-up.)
I would like to see what a DSC trace of pure HMX would look like. It would probably look quite different because it would start to release its energy at a lower temperature, the deflagration temperature of HMX being 287 deg C. (page 238, "Explosives," 6th edition, Meyer et al., 2007)
In summary, the DSC data is extremely limited with much scatter and has a potential method error. Drawing firm conclusions from it is extremely dubious.
QUANTIFYING EXPLANATIONS FOR SCATTER IN DSC DATA
Variability in proportions of gray layer within the red/gray chips and organics present in the red layer are cited as explanations for the scatter in the DSC data. Let's quantify these explanations to see what kind of variability of the specimens might account for the scatter in the data.
The energy release for thermite is cited as 3.9 kJ/g. For the lowest value from the DSC tests, 1.5 kJ/g for sample 1, let's first assume for the sake of illustration that this specimen had a low value because it had no organics in the red layer (in other words it is essentially pure thermite), and the gray layer being predominantly iron oxide as the paper says, acted as excess reactant and therefore was essentially inert providing no energy in the DSC test. In this case, the red layer would have to be present at 38 % by weight and the gray layer present at 62 % within the specimen. The math is 1.5/3.9 = 0.38. In summary, this would be assuming no organics, 38% red layer, and 62% gray layer.
The above calculation is done as a base case for comparison, even though its assumption of no organics in the red layer goes against the major thrust of the conclusion of the Harrit et al. paper that the red layer is a form of nanothermite that includes organics.
The next calculation is an attempt to quantify the high end of the DSC data, 7.5 kJ/g for sample 3. The presence of organics is cited as the explanation for the high value by the paper. Pure high explosive HMX has a energy release of 5.2 kJ/g, so even if sample 3 were 100 % HMX, this could not account for the higher value of 7.5 kJ/g obtained by the DSC test. So, for the sake of illustration, I am going to assume that the organic present in the red layer has an energy release of twice that of HMX, or 10.4 kJ/g. To help account for the high value of 7.5 kJ/g for sample 3, I am also going to assume that it contains no gray layer. In other words I am going to assume that sample 3 is 100 % red layer material. Setting x = weight fraction thermite in the red layer, the math is 3.9x + 10.4(1-x) = 7.5. Solving for x gives 0.45. So based on the above assumptions the red layer would have to contain 45 % thermite by weight and 55 % organic. In summary, this would be 100% red layer, no gray layer, with the red layer made up of 45% thermitic material and 55% organics.
Let's now go back to the low value case, 1.5 kJ/g for sample 1, and assume that it is made up of red layer (containing 45 % thermite and 55 % organic from the previous calculation) plus gray layer of inert excess iron oxide reactant. In this case then, the specimen would have to contain 20 % red layer and 80 % gray layer. The math is 1.5/7.5 = 0.2. In summary, this would be 20% red layer (made up of 45% thermite and 55% organic) and 80% gray layer.
Therefore, a very high degree of variability among the red/gray chips is necessary to explain the scatter in the DSC test data, unless there is significant error in the experimental apparatus and technique.
This is in contrast to these statements from the paper concerning the great uniformity of results.
From page 15, right column, it states, "From these data, it is determined that the red/gray chips from different WTC dust samples are extremely similar in their chemical and structural makeup. It is also shown that within the red layer there is an intimate mixing of the Fe-rich grains and Al/Si plate-like particles and that these particles are embedded in a carbon-rich matrix."
From page 23, upper left column, it states, "The results clearly show the similarities of the red/gray chips from the different dust samples from all four sites."
ANOTHER WEAKNESS IN THE DSC DATA
On page 25 of the paper, it says
"The red layer of the red/gray chips is most interesting in that it contains aluminum, iron and oxygen components which are intimately mixed at a scale of approximately 100 nanometers (nm) or less. Now we compare a DSC trace obtained for a WTC red/gray chip with a DSC trace obtained for known super-thermite (see Fig. (29))."
Note that in Fig. 29 the trace of the WTC sample is really not all that similar to the known super-thermite. The WTC sample cited in Fig. 29 is the MacKinlay sample, although it does not say whether it is the MacKinlay 1 or MacKinlay 2 sample. But from Fig. 19, where 4 WTC DSC traces are plotted, for the four samples, MacKinlay 1, MacKinlay 2, Intermont, and White, the two MacKinlay samples are the lowest energy release traces. The other two, Intermont (sample 3), and White (sample 4) have much higher energy release, and deviate even more from the known super-thermite trace of Fig. 29.
STATISTICAL WEAKNESS OF THE DATA PRESENTED IN THE PAPER
With the tremendous scatter in the DSC data presented in the paper, and its implication for extreme non uniformity of the red/gray chips that I have tried to quantify, the question needs to be raised on all of the other tests that were performed to characterize the red/gray chips, and whether additional tests should have been done on other of the chips within the samples, to see if those tests also would have shown such high levels of non uniformity.
OTHER QUOTES OF NOTE FROM THE PAPER
From page 25 of the paper, right column, first paragraph, the final sentence is
"We make no attempt to specify the particular form of nano-thermite present until more is learned about the red material and especially about the nature of the organic material it contains." This statement expresses a lot of doubt about the findings.
From page 29 of the paper, final paragraph.
"Based on these observations, we conclude that the red layer of the red/gray chips we have discovered in the WTC dust is active, unreacted thermitic material, incorporating nanotechnology, and is a highly energetic pyrotechnic or explosive material." This final conclusion statement of the paper expresses much less doubt.