Monday, January 10, 2011

911 Truth and Nanothermite: I Tried...

This follows a previous post about the nanothermite paper of Harrit et al.: HERE

Following my critical expert review of the Harrit et al. paper, Niels Harrit agreed to respond and we had several email exchanges about the matter. These email exchanges included more than twenty prominent 911 Truth Movement proponents in cc (see posted exchanges).
The emails are quite instructive about how science operates, how some 911 truthers operate and about persistent uncertainties regarding the Harrit et al. paper.
Therefore, I have decided to make these email exchanges public: HERE-1, HERE-2, HERE-3, HERE-4. (fresh link to the email exchanges:

As you can see from the main exchange (HERE-1), I was trying to treat one point at a time in order to go down the list, in the order of Harrit's first response. 

The first point was: Is there conclusive measurement evidence for the presence of aluminum in the red layer? 

After several back and forth contributions, Harrit simply quit without providing any of the extra data that he had mentioned and without answering the crux of the aluminum question. He ended with "Enough. It ends here.

See last email in the main exchange. Here, my point is not that there is no aluminum in the red layer. My point is that the Harrit paper does not provide conclusive evidence that there is aluminum in the red layer. More and better measurements (without the measurement design problems used by Harrit et al.) are needed yet it has been more than two years now since the original work was done and the needed measurements have not been reported, nor has any other research group confirmed the findings. 

To be fair, Harrit is not an expert in electron microprobe analysis (EMPA/EDX/SEM) and this was the first time that he was co-author on a paper that used this measurement method whereas I have taught the technique at the graduate level and used it in my research for over a decade and had my own electron microprobe analysis instrument in my lab until 2008 at the University of Ottawa. 

Then we were going to move on to all the other points (one at a time, see email exchange) but we never got past the first point. For example, I would have been happy to correct Harrit's Newtonian physics errors (see email exchange) if he agreed to publicly exchange about these points. Physicist David Griscom (who was a referee on the Harrit et al. paper) was even more vitriolic than Harrit (see exchanges, especially HERE-4). 

None of the other more than twenty prominent 911 Truth Movement proponents said a peep on the email exchange, not even just to appeal to reason and calm.
In my opinion, the Harrit et al. paper has no merit as it stands and the exchange I proposed was aimed at helping the authors either correct the paper or correct their conclusions. Harrit's behaviour, reinforced by Griscom, gives the impression of a cover up rather than a dedication to truth seeking.
I think a movement based on truth should call for logic and reason in evaluating the scientific claims, not appeal-to-authority arguments, verbal intimidation, and a refusal to intellectually engage.


Darren Butts said...

Your debate on The Kevin Barrett Show was great. It is nice to hear logic in the 9/11 Truth Movement. I am in complete agreement with your stance on 9/11. People in the movement are missing the forest for the trees. In typical internet fallacious reasoning, they attack anyone who dissents from their own "official conspiracy theory". More people like you need to represent the movement before it is lost to history.

Thank you!

NonnyMus2 said...

I know I'm late to the party, but that first set of emails was hilarious! I've never experienced a scientist replying to questions about a paper with a long screed about some other topic!

Wouldn't Harrit have thought that rant might just show he couldn't defend his paper?

I also love how he never gives details on the supposed 'peer review'. Man, I could tell you long, detailed stories of peer review on every single paper I ever published! Some of them are even funny!

I think we can conclude there was no review, especially in light of the editor resignations and Phil Davis' experience with another Bentham Journal, The Open Information Science Journal. They didn't even catch that his home institution's name was "Center for Research in Applied Phrenology" (CRAP)!

I mean, srsly! Even scientists with anger-managment problems can stay on-topic when discussing their work. Harrit can't.

Anonymous said...

Denis, the observation of molten iron in the DSC falsifies your alternative rust hypothesis.

any serious discussion would start with the strongest evidence, speculative nit picking around the edges which i suspect is how your exchanges are viewed by most does nothing but hide from and obfuscate the strongest observations. people have limited time so focus on the meat and potatoes, not the gravy, how do explain the molten iron in the DSC?? any chance of a response?

Denis Rancourt said...

To Anonymous:

More chance of a response if your comment is not anonymous.

How do do Harrit's DSC measurements demonstrate the presence of metallic iron as spherules...? DSC = differential scanning calorimetry. Hello? They do not.

In any case, my critique has shown that Harrit has not presented reliable evidence even for metallic aluminum.

Sid said...

"How do Harrit's DSC measurements demonstrate the presence of metallic iron as spherules...?"

Hi Denis,
That wasn't my question, nowhere did I say "DSC measurements"

Let me clairfy: the observation of molten iron seen in the DSC post ignition falsifies your alternative rust hypothesis.

From the Harrit paper:
"4. Observation of Iron-Rich Sphere Formation Upon Ignition of Chips in a differential Scanning Calorimeter

In the post-DSC residue, charred-porous material and numerous microspheres and spheroids were observed. Many of these were analyzed, and it was found that some were iron-rich, which appear shiny and silvery in the optical microscope, and some were silicon-rich, which appear transparent or translucent when viewed with white light; see photographs taken using a Nikon microscope (Fig. 20)."
"After igniting several red/gray chips in a DSC run to 700 °C, we found numerous iron-rich spheres and spheroids in the residue, indicating that a very hightemperature reaction had occurred, since the iron-rich product clearly must have been molten to form these shapes. In several spheres, elemental iron was verified since the iron content significantly exceeded the oxygen content"

Molten elemental iron in the DSC when the DSC was run to 700C!

1. The melting temperature of iron oxides are ~1300-1500C.
2. There was too little oxygen in the molten iron rich spheres.

Those two observations rule out rust, so why are you claiming it's rust?

Denis Rancourt said...


I see, you meant seen (by optical microscopy and SEM-EMPA) in the DSC-treated samples rather than seen by DSC. OK.

"Iron-rich" regions "seen" in SEM electron microprobe analysis (EMPA) and that are reported to contain oxygen cannot be "molten" or metallic iron if we believe the reported Fe/O ratios.

Also the Fe/O ratios in all such EMPA measurement on such granular and inhomogeneous samples (and that contain steel!) are not quantitative (reliable).

There has been no demonstration of the presence of metallic iron formed by the DSC reaction.

The SEM-EMPA work of Harrit is very poor and that is an understatement. I taught SEM-EMPA at the graduate physics level for many years and these guys would have failed the course.

Sid said...

"Iron-rich" regions "seen" in SEM electron microprobe analysis (EMPA) and that are reported to contain oxygen cannot be "molten" or metallic iron if we believe the reported Fe/O ratios.

fig 24 shows a specturm for spheres generated post ignition from commercial thermite, as you can see the commercial thermite molten residue contains oxygen. compare with the fig 25 spectrum showing spheres formed during ignition of red/gray chip in the DSC.
your assertion that the DSC residue cannot be molten or metallic iron cannot be true when the spectra from fig 24 and fig 25 both contain similar iron and oxygen ratios. are you therefore asserting that the fig 24 spectra for commerical thermite spheres as shown in Harrit cannot contain molten iron?

Also the Fe/O ratios in all such EMPA measurement on such granular and inhomogeneous samples (and that contain steel!) are not quantitative (reliable).

if you are claiming the material is molten rust rather than molten residue from a thermite reaction, how can you explain the molten rust at 700C? rust and iron have similar melting temperatures around 1500C. what reaction are you proposing occured in the DSC that created a molten iron-rust material at 700C?

There has been no demonstration of the presence of metallic iron formed by the DSC reaction.

from the Harrit paper:
"After igniting several red/gray chips in a DSC run to 700 °C, we found numerous iron-rich spheres and spheroids in the residue, indicating that a very hightemperature reaction had occurred, since the iron-rich product clearly must have been molten to form these shapes. In several spheres, elemental iron was verified since the iron content significantly exceeded the oxygen content"
what other form of iron compound are you suggesting? there is not enough oxygen in the DSC iron rich spheres to be rust.

Denis Rancourt said...

Hi Sid:

You make good points.

If you completely and verifiably identify yourself (using an account and with your full name) then I will continue the exchange (which Harrit quit and did not want to continue).

I just think readers should know both parties in an exchange like this.

It would also be good if you disclosed any affiliations with Harrit et al. Again, so we know who the two parties are.

Otherwise, I have many other things to do and will leave it as is.

Sid said...

Denis, I have no affiliations with Harrit, or any other truth organisation other than I know I am not a liar by knowledge or predjuced delusion, so my full name and address are not required in order to understand the soundness of your rust hypothesis.

I have listened to your debate with harrit and read the follow ups from yourself and harrit and also the consequent emails. I am posting here to understand where you are coming from, I do not see how your rust explanation works for the reasons I have mentioned. without an exothermic reaction the sample in the DSC cannot be rust since the DSC ran only to 700C, so what exothermic reaction are you proposing that produced temperatures around 1500C that would be required to melt the sample if it were rust?

I am asserting that the observation of melting in the DSC as reported in the Harrit paper falsifies your rust hypothesis.

Unknown said...

Denis, I work with molecular and macromolecular visualizations so I am very familiar with looking at the micro- and nano-sized world. I am commenting to tell you that it appears to me that the SEM imagery unambiguously falsifies your rust hypothesis and I would like you to respond specifically to this point. I'd like to keep this discussion away from the chemical make-up, and strictly on the observed morphology of the chips.
Please look at Fig. 8a and Fig. 9. (same cross section, different rendering).
These pictures clearly show that there are three distinct materials: two particle types are embedded in a third material. The two particle types are distinct and easily visually identifiable; there are hexagonal platelets ~1 micron across, and smaller faceted crystals ~100 nm across. The two particle types show excellent dispersion and mixing on the sub-micron scale. These properties are very difficult to achieve on purpose, and are not present in rust. Lawrence Livermore National Labs, describing the production of nano-thermite said, "We believe that the intimacy of mixing between oxidizer and fuel is an important factor in the behavior of these materials. One problem, all nanomaterials suffer from, is the tendency for agglomeration into larger aggregates... Even though nanosized components are used, there is no guarantee that the sol-gel composite will have such mixing." -- Nanostructured Energetic Materials with Sol-Gel Methods
In Figure 8c we can see some of these aggregates, both of the rhomboidal crystals and of the hexagonal platelets. Fig. 5a-d and Fig. 4 confirm that the largest of these aggregates are ~1-2 microns and that these aggregates are clearly not as common as the evenly dispersed particles surrounding them.
My 2-part question to you is this (and I would like reliable sources for your answer, as I provided with the LLNL quote):
Are you suggesting that the red layer in it's entirety is composed of rust?
- If so, how do you account for three distinct morphologies with submicron dispersion and mixing?
- If not, are you suggesting that one of the three morphologies is rust and the other two are other materials? Do you say, like Harrit, that the small faceted crystals are the "rust"? How did a big chunk of rust somehow fracture into nanoscale crystals, disperse itself on the nanoscale, and become intimately mixed with another particle type, consistently and throughout the volume of the red material?
I hope my comment will be accepted, it is an earnest attempt at civil discourse. Thanks!

Denis Rancourt said...


Here is what I stated in my review of the Harrit et al. paper:

"When steel rusts in a humid building environment it grows a crust composed of layers of different Fe-oxides and Fe-oxyhydroxides. These are stratified micro-layers with successive layers of different Fe-oxides species (wustite, maghemite, hematite, etc.). In a humid atmosphere the outer layers will be Fe-oxyhydroxides such as goethite, lepidocrocite and akaganeite. The latter three Fe-oxyhydroxides have the same chemical formula: FeOOH, and differ only in their crystal structures.

These Fe-oxyhydroxides typically form as nanoparticles and have the same needle and nanoflake-like morphologies as observed here.

When these Fe-oxyhydroxides are heated in a DSC they undergo a solid to solid exothermic reaction of dehydroxilation (loss of OH) and transform from FeOOH to Fe2O3 (hematite) at a temperature of approximately 400 C. The temperature of the transformation can vary depending on exact chemical composition, and on the crystal structure, but it is always at approximately 400 C.

Looks like our boys may have been discovering the properties of rusted steel. Steel contains C and Si which would end up in its oxidation products, especially in the oxyhydroxides.

I may be wrong but if I had been a reviewer I would have required that the authors prove the presence of metallic Al nano-flakes by X-ray diffraction (or electron diffraction) and that they be much more careful in their EDXA work. I think they would have seen FeOOH not metallic Al."

Furthermore, I have never heard of aluminum metal being synthesized as hexagonal shaped crystals...?! That would be a materials science discovery in its own right. Whereas the hexagonal platelet shape is known in Fe-oxyhidroxides.

Unknown said...

Hi Denis, thanks for your response.
Am I correct to assume that when you say "These Fe-oxyhydroxides typically form as nanoparticles and have the same needle and nanoflake-like morphologies as observed here." you are insinuating that both particle types are rust?
If this not the case, please correct me. If this is the case, you still have two fundamental problems:
1) I do accept that "rust" can form different morphologies in nature, including nanoparticles. However what you ignored from my previous post (despite me mentioning it thrice) is that in nature, these particles are not dispersed on the sub-micron scale and are not mixed with each other on the sub-micron scale. For example, Fig 1a of "Monitoring iron mineralogy and contaminant mobility using geoelectrical approaches" shows hexagonal platelets of iron oxide. However, importantly this is the only material present - there are no other particle types mixed in and no other material in which it is embedded. Which brings us to...
2) Rust forming the two distinct particles types still does not account for what the amorphous material is in which these two particles types are embedded, or how these two particles types became embedded throughout the volume of said material in a well-dispersed and well-mixed configuration. Indeed, this question of "how" is of fundamental importance to the manufacture of nano-thermite as my previous source showed. Here's another source describing a way of achieving these dispersion and mixing properties by precipitating the iron oxide directly in solution when making nano-thermite: "The Fe2O3 was prepared by the use of an organic epoxide which was added to an Fe(III) salt solution resulting in the formation of nanoscale crystalline and amorphous Fe2O3. The reaction to produce Fe2O3 was done in solution which already contained the UFG aluminum. In this case, the nanoparticle aluminum was sonnicated (suspended in isopropanol and placed in an ultrasonic bath to break up any aluminum aggregates) before mixing with the Fe(III) salt solution." -- Nanoenergetics: An Emerging Technology Area of National Importance
It would seem that my sources indicate that the specific properties I have described are consistent with nano-thermite and are not consistent with rust. As you pointed out, "Rusting steel is one of the most studied materials science problems in engineering." yet despite this being the case I can find no sources at all to corroborate your hypothesis, even after hours of searching.
I am willing to grant you that I may be ignorant on the subject, but I am not willing to merely take your word for it. Please provide references, as I have been courteous enough to do the same for you.

Denis Rancourt said...


You state: "in nature, these particles are not dispersed on the sub-micron scale and are not mixed with each other on the sub-micron scale."

This is completely incorrect. You should keep searching until you find the hundreds of scientific articles that show the opposite of your misconception.

Use Google Scholar. Include the environmental and geochemical literature, because the metallurgists tend to limit themselves. You can even add my name in the author search field if you like.

My point was to provide a plausible alternative.

You have not found an article showing hexagonal platelets of metallic aluminum, yet you seem to believe that? Cite such an article and then we can continue this conversation.

Unknown said...

Hi Denis, I appreciate the quick response.
You claim that there are "hundreds of scientific articles" which would elucidate your point... yet you are unable to provide me with even one, despite the fact that I was courteous enough to reference my own posts? It's your blog so if that's how you want to play it, fine. However, I do need a clarification from you before I continue my search:
Again, I ask you to please clarify: is your assertion that the entire red material (including all 3 distinct morphologies) is rust? Or is your assertion that the two distinct particle types are rust and that this rust is embedded in a third, unknown material? Give me this simple clarification, and perhaps I will be able find a suitable reference.
As to your "metallic aluminum" point - I did stipulate in my first post that I am not interested in discussing the chemical make-up. In the interest of not getting off track, let's keep the conversation strictly to finding out if rust can have the specific morphologies and distributions as seen in Harrit et al's micrographs.

Unknown said...

Nevermind Denis, I have no need of any further clarifications from you: I have found reliable sources which appear to conclusively disprove your rust hypothesis. Perhaps you should have read more papers about metal and fewer about soil? In your defense, "it can be said the recognition of composition and structure of the rust later is not obtained commonly" [1] ... but here we see the importance of using references! Anyone can make mistakes :).
NOTE: emphasis will be added to the quotes to demonstrate their relevance. All papers deal directly with rust layers formed on steel over several years.
You have stated that rust can account for the physical characteristics seen in Harrit et al's micrographs. Upon cursory inspection, it appears you could be correct, in that rust on steel is indeed a combination of an amorphous morphology combined with crystalline morphologies:
"In the rust layer on all specimens, a-FeOOH, b-FeOOH, c-FeOOH, Fe3O4 and so-called amorphous rust were found." [1]
But, as I questioned before, what about when we get specific about the DISTRIBUTION of these morphologies? That's where your hypothesis falls apart. In Harrit et al's paper, they found that "the size and presence of the particles was found to be consistent throughout the layer" and that "all of the particles are embedded in an unstructured matrix". The same is NOT true of rust:
"Amorphous rust is LOCATED AT THE BOTTOM of the rust layer irrespective of steel types." [1]
Additionally, the relatively flat shape of the interface between red and gray layers (Harrit Fig. 5b, 5d) does NOT match the interface between rust and steel:
"The cross-section of rust layer is WAVY AND UNDULATING, and appears thin and thick parts. Thus, thickness of rust layer is quite uneven. There will be no argument about the fact that the origin of unevenness comes from NON-UNIFORM DISTRIBUTION of physical and chemical factors"
In addition to the distribution, I stated before that rust crystals would NOT be dispersed on the submicron scale due to aggregation, and the sources agree with me despite your objection:
"the protective effect of the rust layer is due to the DENSE AGGREGATION of fine crystals of Cr-Goethite" [2]
Indeed, this aggregation is key to how the rust layer itself forms:
"dissolved Fe ions turn into Fe(OH)x, separate out from the solution, and finally form rust layers through GRAIN GROWTH" [3]
Fig 6 of [3] also gives us a scale of these aggregate grains - they are on the MILLIMETER scale! This is orders of magnitude larger than the largest aggregates seen in the red layer of Harrit et al's paper.

[1] "In-depth distribution of rusts on a plain carbon steel and weathering steels exposed to coastal–industrial atmosphere for 17 years" K. Asami et al.
[2] "Nanostructure of Protective Rust Layer on Weathering Steel Examined Using Synchrotron Radiation X-rays" Masato Yamashita et al.
[3] "Fe(O,OH)6 Network Structure of Rust Formed on Weathering Steel Surfaces and Its Relationship with Corrosion Resistance" Masao Kimura et al.

Denis Rancourt said...


Your argument is in logical error.

"In experiments on steel rust that I found, the rust looked like "this", therefore rust on steel must always look like "this"."

Wrong. It depends on the kind of steel and the conditions under which the steel rusted. And conditions under which this rust could have been modified by post-rusting conditions, etc.

For example, often steel is "painted" with a rust layer to protect it from further rusting, etc.

Suggesting that the observed micro-structure of rust on steel is always the same is, well, ridiculous. If this were true, metallurgists and materials scientists could quit their jobs.

Anonymous said...

Hello Mr. Rancourt.

The melting of the Fe2O3 grains in the chips into spheres requires more than 1400C.

The DSC maxed out at 700C and shows a spike at 430C.

What reaction starts at 430C and provides 1400C?

What is the explanation for the reduction of the Fe2O3 grains as they are melted into spheres with Fe:O ratios up to 4:1?

The transformation of solid Fe-oxyhydroxides into solid Fe2O3 at 400C is hardly a satisfactory response to this data.

Aren´t you the least bit curious?

Denis Rancourt said...


A temperature does not "provide" another temperature. That is incorrect physics.

Rather, an exothermic reaction, even if initiated at a low temperature, provides energy (heat and other forms), which increases the temperature of immediately surrounding material. The latter increase in temperature is not limited by the temperature that initiates the said exothermic reaction. Rather the local temperature depends on local energy delivery and local transport of energy away from the local region, all this occurring out of equilibrium.

The whole truther "logic" of equilibrium (quasi-static) melting temperatures for materials subjected to zero stress is inapplicable on the micron and sub-micron scales in the presence of reactions (and also on larger scales in a burning building).

Similarly, a given fuel may have a certain flame temperature in air under specific experimental conditions but this does not limit what temperatures can be locally and dynamically generated/attained from combustion of that fuel in complex circumstances!

There is simply a lot of scientific ignorance in many truther statements about physics.

One could write a book about it.

I have great respect for truth seeker political activists but I wish you would all learn enough physics to know that you are talking physics rubbish most of the time.

"Near-free fall", "no debris", "holograms", "tones of explosives", blah blah. It's all just crazy. I have tried to explain it. Read my stuff. And, yes, I agree about building 7. Concentrate on that.

Anonymous said...

Mr. Rancourt, I did not say a "temperature provides another temperature"; I asked you about the reaction that provides the temperature needed to melt the iron oxide. The second part of the question is about the simultaneous reduction of the iron oxide.

There are no "complex circumstances", only a very simple DSC air ignition of Fe/O-Al-C ingredients with known behaviors and melting points.

I am simply asking: do you have an alternative explanation for the melting and reduction of the iron oxide in the given circumstances?

Anonymous said...

So that is it? You are blocking further comments?

Anonymous said...

Why don´t you just call for someone to publish a rebuttal to Harrit?

If paint/rust chips can replicate those spheres in a DSC, let´s see it in a paper.

Don´t you agree this is the way to settle science?

Anonymous said...

"...let´s see it in a paper. Don´t you agree this is the way to settle science?"

Mr. Rancourt?