Thursday, August 30, 2012

Viscoelasticity: It's Not Just for Polymers Anymore

Viscoelasticity is that intriguing phenomona that blurs the lines between a solid and a liquid. Classic liquids are viscous and not elastic, and classic solids are elastic and not viscous, but modern materials such as molten polymers show parts of both and are called viscoelastic. When I first was learning of viscoelasticity many years ago, it was accepted on faith that all liquids would show viscoelastic behavior, even water, if examined closely enough. There was no data, just logic, to support this. That has now changed.

Researchers have now measured elastic behavior in water (free access until ~ September 14, 2012 with registration). I found the report fascinating for a number of reasons. First, the tests were completed on macroscopic samples. The published data was for 0.125 mm thick layers of water, but the researchers had gone to samples as thick as 0.5 mm. Secondly, the results were just what was expected. For instance, here is a strain sweep:
and just like any other viscoelasic material, G' (the storage modulus - i.e., solid characteristic) is greater than G" (the loss modulus - i.e., liguid characteristic) at small strains, with both functions decreasing nonlinearly as the strain increases, and also crossing over somewhere in the process. If I removed the numerical values from the y-axis, this would look just like plots that I make in the lab for any old polymer or gel or what have you. There are also these frequency sweep plots made at 2 separate strains. The first is at small strains showing that the water is solid-like
and the second made at very high strains showing that the water is purely liquid.
G' isn't in the picture - literally. What is so frustrating is that the authors only hint at what occurs between these plots as the strain is increased:
"It is interesting to note that the evolution from the low strain amplitude sinusoidal wave to the large strain amplitude sinusoidal-like wave does not correspond to a simple shift of the phase but to a strong modification of the signal."
More research is needed! And quickly please![*]

But lastly, the coolest part of this research is that is was done not with an ultrasophisticated instrument build just for this purpose but using a rather mundane and ordinary rheometer - an ARES2! Yes, it was modified slightly - the parallel plates were made from nonporous alumina (40 mm diameter of course), and the input and output signals were monitored by a 7-digit voltmeter rather than with the standard hardware/software, but that's it. Nothing more was done. This testing is something that pretty much any rheology lab in the world could measure. It's that simple.

[*] There is one plot in the article that was made at an intermediate strain, but to discuss it would require getting too involved in the rheological details, and besides, that one plot is only one place along the transition from the low to high strains. It's not enough. I still want to see the whole evolution from small to large strains and there are a lot of gaps to fill in.

Wednesday, August 29, 2012

Here Come the Plastic Cars

The US government announced that it had formally approved new automobile fuel economy targets this week. Even though the numbers were first announced some time ago, I still think they are eye-popping: 54.5 miles/gallon (4.3 l/100 km) by 2025.

An article in the Detroit News shows that there is some "softness" to these targets. For instance,
"Automakers can make improvements to air conditioning systems to reduce emissions without improving mileage. As a result, the standard will actually result in a fleet-wide average of 48.7 to 49.7 mpg by 2025."
And then reality will rear its ugly head too:
"Taking into account real-world driving, the actual average will be around 40 miles per gallon in 2025."

Regardless of the exact final numbers, these improvements will only come about as a result of replacing more and more metal components with plastic components. That is not the only way to acheive these numbers, as better engine designs with and without the use of hybrid drives and all-electric vehicles will also play a significant role in future cars. But there still will always be neverending siren call to reduce vehicle weight further and further, and that means plastics will play an ever increasing role in the cars of the future. It's rather ironic that petroleum-based materials will play a key role in reducing the use of petroleum, isn't it.

Tuesday, August 21, 2012

The Toughest Lesson from Grad School

The toughest lesson by far that I learned from my adviser in grad school was to never stop thinking about the results of an experiment. He never said anything explicitly to this; it was all done by example. I've never seen somebody before or since that was able to get so much analysis out of a single experiment. Again, it was never stated explicitly, but his attitude was that if you've taken the time and effort to do the experiment, get all you can from it.

It was a great lesson to learn, as it is something that I perceive few other people learning it. Someone gives me a stack of data and some conclusions and I immediately go to work on what else can be gathered from it. It's not something that people are inclined to do - look for what else might be lying out there.

The pithiest slogan I've ever seen that captures this is: "A conclusion is the point at which your mind stops thinking."

So why was this the toughest lesson I learned in grad school? Because when I was finishing up and thinking I'm done, my advisor kept saying something to the effect of: "no, there is more analysis we can do with the results". It got to be quite unnerving. You keep thinking you can write it up and yet there is still more to do. Not more experiments like most people suffer from - more analysis, more thinking, more conclusions. When this is thrown at you at a moment of weakness ("I wanna be done and get out of here!"), you can't help but remember it.

So know you know why I was perplexed yesterday with what to do with that flask of polypropylene crystals floating in tetralin. We had already done the intended experiment and analysis, but there's that nagging voice that keeps pushing me to see what else we can do with it. I just can't toss it and I haven't yet. What to do, what to do...

Monday, August 20, 2012

Cleaning Up the Lab

It's been a long summer with the intern, and for a while things were running faster than we could keep up. The exciting experiments outpaced the dull maintenance aspects of lab life. For reasons beyond our control, that pace has now lagged. Looking around, it looked like the morning after an all-night party. Flasks, beakers and jars everywhere. It was time to clean up.

Much as I would like to claim wisdom and foresight, it is rather the hard-and-painful knocks from experience that made sure everything was labeled, so it was easy to firmly decide on keep/toss for all the containers and samples, except for one flask.

It wasn't that the flask had an unknown sample in it - I knew exactly what is was - a sample of polypropylene crystals floating in tertralin (1,2,3,4-tetrahydronaphthalene). We had been extracting some material from it, and tetralin in a great solvent for dissolving PP as it is non-polar but has a sufficiently high boiling point to allow for the solvation of the PP. (There are no solvents that dissolve PP anywhere near room temperature. The heat of crystallization has put the sample so far down the thermodynamic well that only a tremendous amount of heat will be able to crank the windlass and raise it up.)

So we extracted what we want, and let the flask cool. As expected, the PP crystallized out and now floats in a thick layer near the top (PP has a density of about 1, while the tetralin has a density of 1.54.) So the question, still unanswered, in my head is this: is there still something we can learn from those crystals?

Thursday, August 16, 2012

The Omics of Polymers

The Wall Street Journal had an article a few days ago about how the suffic "-omics" is now becoming overused. It originally started out as a nice word in the field of biology, most prominently with "genomics" (identifying the genes of a life form), then moved on to proteomics (identifying the proteins of a life form) and then has gone crazy to the point where over 400 examples in biology alone exist. Worse yet, the suffix has now leaped the field-of-study barrier, to include fields far removed from biology such as patentomics (all of patent law) and culturome (all of human culture).

Before it's too late, polymers should get into the game, so I propose the following:
  • Polymerome - the totality of all polymers
  • Thermoplasticome - the totality of all thermoplastics
  • Thermosetome - the totality of all thermosets
But those are easy ones. Now let's fire up some words to really get your toungue tied:
  • Viscoelasticitiome - the totality of all viscoelasticity phenmona
  • Mark-Houwink-Sakurada constantome - the totality of all Mark-Houwink-Sakurada equation constants
  • Time-Temperature Superpositionome - the totality of all time-temperature superpositions
  • and finally,
  • Non-linear differential corotation modelome - the totality of ________, yeah, that.
Please start using these and others omic words in your next journal submissions, grant applications and conversations with neighbors. It will defiantly show that you are a cutting edge researcher that needs to be taken seriously. And also be sure to check out the badomics generator, a webpage that automatically generates omics such as what I have proposed here, and also shows what they could look like in an article title.

Wednesday, August 15, 2012

Helmet Gels to Reduce Head Injuries in Sports

Head injuries in full-contact sports such as hockey and (American) football are becoming ever more common as are discussions of their long-term impact. While this has been mostly chatted up in the sporting community, it has now crossed over to the general public after the noted political columnist George Will highlighted some of the new concerns in a recent editorial. [*]

While rule changes have been made to try and reduce this problem, there are also approaches taking advantage of technology, specifically gel rheology. A Georgia company, Guardian now makes a gel-filled cap that covers a football helmet and, at least based on initial, anecdotal reports from players, seems to pretty much eliminate head trauma. Gels are well-known for absorbing and dissipating energy repeatedly, making them the ideal choice for this application. From a rheological perspective, soft gels have a dominant viscous behavior and a minor elastic behavior. The viscosity is important as it is the sink for the impact energy. Elasticity is recoverable energy, something that should be minimized, but if there is too little of it, the gel would not recover its shape repeatedly. (Imagine how unworkable a cap filled with water or even a very viscous liquid would be. It would all flow to the bottom and provide no protection up high.) A key question is how long these gels maintain their integrity, as all the impact energy will eventually break down the elastic component, making them of little use.

Since the caps go over the helmets, the caps are only used in practice and taken off for games. But players practice more than play actual games, so while not perfect, the impact(grin) of the caps is significant. It seems like the caps are too thick for them to be be placed inside the helmet and therefore be used all the time. That would be a nice second generation product or something for a competitor to invent.

[*] While the column is almost entirely about chronic head injuries, I was much more taken aback by this quote:"For all (football) players who play five or more years, life expectancy is less than 60; for linemen it is much less." I used to think my sport, bicycle racing, was tough because of the old saying in the peleton that riding the Tour de France takes a year off your life. But for football players, it looks like a year in the NFL takes 3 years off your life. Yikes.

Tuesday, August 14, 2012

Polymerizing Antioxidants

The various defense labs run by the military get to work on some really cool stuff at times, such as hypersonic jets that cruise along at Mach 6. But sometimes, the cool stuff is more mundane, such as preserving food. Napoleon long ago recognized that "an army marches on its stomach" and offered a prize for invention of a safe food preservation process. The winning technology is still around today - canning. While modern soldiers ride more and march less than in the past, keeping them fed is still as much of a concern as in the past, and food preservation is just as important.

A few weeks ago, the US military announced that researchers at the Natick Soldier Research, Development and Engineering Center had discover that polymerization of a natural antioxidant, hydroxytyrosol, allowed it to become an even more effective antioxidant. A quick review of the patent shows that horseradish peroxidase, an enzyme that as you might guess is extracted from horseradish, [*] is used as the initiator - hence it is a free radical polymerization, and in this case, a green, water-based approach as well. Sadly, the details of the polymers structure are not provided at all, although the PR release hints at a conjugated backbone in the product. That would certainly be helpful in providing increased antioxidation potential.

I've never worked with horseradish peroxidase, but others have, such as to polymerize methylmethacrylate (free access) or as was seen here today, polyphenols ($). The general approach is that the enzyme catalyzes a reaction (as enzymes have a tendency to do) between some substrate (not necessarily the monomers) and hydrogen peroxide. That generates a free radical which then starts the whole polymerization. A little biochemistry never hurt anyone, did it?

[*] The enzyme however, is not what gives horseradish its pungency. That comes from allyl isothiocyanate instead.

Monday, August 13, 2012

A New Perspective on the Great Garbage Patch

When a blog post is entitled The Eighth Continent: Searching of the Great Pacific Garbage Patch, I expected more of the usual about how huge the amount of garbage is and how bad plastic is and... After all, calling it the Eight Continent creates a pretty strong visual image, and not a nice one at that.

But this excerpt from Andrew Blackwell's book Visit Sunny Chernobyl: And Other Adventures in the World's Most Polluted Places is different. It is an honest look at the situation and possible solutions.
"These Garbage Patch conversations tend to follow a certain profile. First there is the flash of recognition, embedded with nuggets of misinformation and cliché: Right! The giant plastic island! The one the size of Texas! It's not an island, you say. You also want to interrogate them on the subject of Texas. Why must the Garbage Patch’s size must [sic] always—always—be measured in Texas units?

Ok, it’s not an island, they say, backing away a little. It's more of a pile. You narrow your eyes. Seriously, how do you pile anything on the ocean? Eventually, with coaxing, they let go of the island imagery, of impractical notions of how things pile, of Texas. Then comes the inevitable question: Can it be cleaned up?

A lot of people have considered this question, and a broad consensus has emerged among scientists and environmentalists. I'm happy to summarize: GET REAL. I know everybody loves the pipe dream of a technological fix, but we're talking about the ocean here. Even assuming that it’s possible to drag nets back and forth across hundreds of thousands of square miles of ocean, and that it would be worth the massive use of fuel … even granted these improbabilities, there remains the intractable fact of the confetti."

Andrew was on the crew for a boat that was collecting trash in order to understand the flow lines, the particular streams that take trash into the vortex, and also discovered what drives many environmentalists.
"The point, we realized, was that our goal was not to study the debris in any useful way, but simply to find it. We were looking for what Mary referred to as "current lines" of trash, narrow bands of high density. Mary spoke again and again of the current lines, and began to suspect that if the Kaisei returned to port heavy with trash, it would serve to validate Project Kaisei’s dream of cleanup. But for that, we would have to find the mother lode. It was the paradoxical symbiosis that can afflict any activist. You come to depend on the problem you’re fighting. That we were so focused on finding the Garbage Patch in a concrete and spectacular form was tragic—particularly because it isn’t a visually spectacular problem. As we would discover once we reached the Gyre, the Great Pacific Garbage Patch doesn’t actually look like much—unless you’re paying attention. The plastic confetti are invisible unless you scrutinize the surface of the water. And the millions of plastic bottles and laundry hampers and snarls of old fishing tackle are not clumped into a single mass. Yet the Garbage Patch is indeed a problem of vast scale and implications.

This conflict between the reality of the problem and its non-visual nature is at the root of the myth of the plastic island. We hunger for a compelling image to help us understand the issue. But depending too much on spectacular imagery can actually limit our understanding. We create islands where none exist, and then waste our time searching for them. We become Ahabs without a whale."

Getting back to the imagery of the "garbage patch", Andrew creates what I think is the best one yet:
"If we absolutely must have an image to use in thinking of the Garbage Patch, it should be that of a galaxy. The Garbage Patch is like the Milky Way, an impossibly massive spiral that, because of its very vastness, is also phenomenally diffuse. You could pass right through it without ever bumping into a star or a planet. But that doesn’t mean it’s not there. And please note: Your galaxy is many times the size of Texas."(emphasis added)
That's imagery that needs to become far more popular - not an island or a continent [*], but a galaxy. Very dilute, and extremely large, and in the case of this galaxy, something that just shouldn't be there at all.

[*] How did BoingBoing come up with "The Eight Continent" for the title of this excerpt? Didn't they read the article? No, I guess they didn't.

Friday, August 10, 2012

Friday Funnies

Just for giggles, I decided to check my Klout account today. I got the giggles alright:
See! Not only am I influential about the Nobel Prize ("Honey! It's Sweden on the phone again!" "Tell them to give it to Whitesides this year and leave me alone!"), but I'm also quite the fashionista! I'm assuming this is all because of the post PPE for the Fashionable Polymer Chemist that I wrote for National Lab PPE day. One humorous post and tweet about fashion. Back in June. And suddenly it's my fourth most influential topic. Not polymers. Not rheology. Not engineering.

Looks like it's time to start working on my Fall line of clothing. Thanks Klout, for pointing this career path out for me.

Thursday, August 09, 2012

Swimming and Viscosity, Part 2

Yesterday's post on "Swimming in Jello" stirred up a real bee's nest over at reddit/physics, so I thought I would reply to the most common complaints about my post.

First, I apologize to all that I didn't get a chance to catch the firestorm earlier when I could have done more to control it. I now realize that my original post is lacking a key piece of information - adding Jello (gelatin) to ~ room temperature water will not turn it into the thick solid gel that we all have eaten. I made some up this morning, but even after 2 hours, the water is only slightly thicker but still extremely runny. The water is also very cloudy from all the undissolved gelatin. In order to get the solid gel, the gelatin needs to be "cooked" to about 180 F or so and then cooled, something that would not be possible in a swim pool.

Having worked with numerous gelling agents over the decades, I knew this, but I should have been clearer about it upfront. Again, I apologize. So when I saw the cartoon characters dumping Jello into the pool, I falsely assumed that no one would think that it would turn to a solid, but rather just to a slightly thicker liquid, and that further that the cartoonist was making the assumption that the increase in viscosity would slow Michael Phelps down.

The article that I linked to (yes, I have fixed the link) about swimming in guar gum takes advantage of the fact that it will gel water without needing to be heated. Jello is protein based, but guam gum and other gums are carbohydrates. There are also synthetic thickeners such as carbopols, pluronics and crosslinked polyvinyl alcohol. These latter actually work best if you first disperse them in ice water and let the ice melt, certainly a viable option for gelling a swim pool.

By the way, if anyone is interested in repeating the guar gum pool, you better be very rich or willing to wait a few years. Guar gum is in extremely short supply right now as it is the preferred gelling agent for fracking, and there's a lot of extremely rich farmers in India as a result. In a few years, the crop will be larger and prices will fall to normal.

Gelled water is non-Netwtonian, a very broad term that I can tell from more than a few of the comments is also not well understood. Gelled water will be not only shear-thinning (the viscosity will drop as the shear rate increase) but also thixotropic (the viscosity at a constant shear rate will drop over time, but will then increase after the shearing has stopped). All of these characteristics will lower the viscosity as someone swims through it, so if you want to use viscosity to slow down a swimmer, you need the opposite approach, something that makes water dilatant and rheopectic. by making into Oobleck - just add cornstarch.

Sadly, the cartoonist missed that simple and fun approach as the real way to slow down Michel Phelps.

Wednesday, August 08, 2012

Swimming and Viscosity

I really love the comic XKCD, but his output today is just plain wrong.

In it, two people are trying to catch a swimming Michael Phelps with a net, but can't because he is too fast. So they decide to dump Jello in the pool to slow him down.

Unfortunately, this will do nothing to slow him down. While the viscosity of the water will increase and hence the drag, he will also be able to push off the viscous water more effectively. The net effect will be that nothing changes.

This has in fact been demonstrated back in 2004 by one of my previous chemical engineering professors here at the University of Minnesota, Ed Cussler. Himself an avid swimmer, he was somehow able to convince the maintenance people here to let him gel the pool with guar gum. Additionally, he was able to convince the swim team to swim in the goo. A comparison of times in water vs. goo showed no statistical difference, and in doing so, won an Ignobel Prize in 2005.

Back to the cartoon, the mouseover shows that the two pursuers were unsuccessful in catching Phelps anyway, as he ate all the Jello, a nod to his well known appetite. Right result, but wrong reason.

UPDATE: I have additional comments on this matter here.

Tuesday, August 07, 2012

Self-Immolative Polymers

Polymer degradation usually means that the polymer chains are being broken somewhere in the middle. This is the most common degradation mechanism for polymers and occurs with biodegradable polymers that are composted, most cases of thermal degradation, oxidation, exposure to UV light, etc. Much more rare is degradation of a polymer proceeding from the ends towards the middle. People are first exposed to this idea when learning about the ceiling temperature for a polymerization - the temperature at which the forward reaction (polymerization) is in equilibrium with the reverse reaction (depolymerization).

Both of these mechanisms usually take a lot of energy to completely degrade a polymer, needing a continual stream of energy input - lots of heat or UV or whatever. In neither case can you initiate the reaction and just let it go, there are too many obstacles (side reactions) in the way.

But all that is changing. A recent report ($) [1] tells of polymers ("self-immolative") that degrade from the ends after UV light (or 2-photon near-IR (NIR) light) removes the endcap of the polymer. The monomers of the polymer have a strong tendency to depolymerize but only the endcaps prevent it. Remove them and the whole polymer unzips all on it's own as the monomers fall off via a cyclization mechanism. The researchers are most interested in using it for an approach to controlled release within the body, hence the interest in the NIR mechanism [2].

But what if this approach could be applied to more mainstream plastics? The plastics would only need a brief exposure to sunlight before starting to degrade. While that may be appealing, there would be a certain element of the game "hot potato" to it as well, as the plastic would be continually degrading and you could never be sure when it will have degraded too much until it fails.

[1] There are a couple of articles with free access about this technology, one be a brief review similar polymers, the other being a report on dendrimers.

[2] NIR is able to penetrate the body without significant absorption, but it the photons are not energetic enough to break off the endcaps. If 2 NIR photons are absorbed at the same location however, the energy is sufficient. An extra advantage of this method is that the absorbing site can be extremely localized if the 2 photons arrive from different directions. No tissue along either pathway absorbs enough light for damage, but the polymer located at their intersection does.

Monday, August 06, 2012

The Secrets of Oobleck Revealed - Partially

I was surprised that there wasn't more of a splash in the general news about this research, but when the article is entitled Impact-activated soldification of dense suspension via dynamic jamming fronts ($), it's pretty tough for anyone to recognize this is an article about Oobleck, isn't it?

Oobleck is a term stolen from Dr. Seuss to describe dilatant/rheopectic materials, the most common of which is cornstarch/water suspensions. (Being a suspension, the cornstarch particles are not dissolved in the water, but are in it as solid particles with water in between them.) Although the suspensions are nominally of low viscosity (think interior house paint), when stressed rapidly, they become extremely stiff. Youtube has a number of videos of the stuff, with people walking on it, or bouncing it on speakers or other fun stuff. It's easy to make this stuff up and home with about a 50/50 w/w mix of cornstarch and water. My favorite trick is to roll a ball of it between my hand and then stop the rolling just to watch is dribble out.

Using a nicely instrumented tank with lasers and x-rays to look at the surface and interior of the fluid, University of Chicago Professor Heinrich Jaeger and his student Scott Waitukaitis looked at what was occurring when an aluminum rod impacted the Oobleck. They found that during the stress, the rod pushes the particles together and quickly displaces water, something they called the snowplow effect.
You may then think that this column of highly concentrated cornstarch beneath the rod contacts the bottom of the tank to provide the needed support, but that in fact is not the case. Certainly in shallow tanks it can't be avoided, but even in extremely deep tanks, the solid forms and resists the impact.

What is actually occurring is that the volume being displaced extends as a cone beyond the area of impact:
With such a large volume of mass being displaced, the solidity of the suspension now is a momentum transfer problem - you are transferring the momentum of the rod to this very large mass of suspension and it just isn't going to be going anywhere very quickly.

The title of this post states that the secrets of Oobleck are partially revealed and that is true. What still remains unknown is how this system relaxes. As you can see from the linked videos, the suspensions remains stiff, but only for a short period of time. For reason still unknown, it relaxes and returns to its fluid state, meaning that if you are walking on this stuff, you better keep walking. Standing still is not an option.

Thursday, August 02, 2012

"Improvements" to the Melt Flow Test that Aren't

The melt flow test is THE workhorse test for the polymer industry. Despite it being a test that academics and other well-educated people love to hate, it is pretty much the entire basis for the sale and purchase of billions of pounds of thermoplastics annually. That one little number pretty decides if a new batch of polyethylene is in spec or out.

The test in a nutshell is this: The sample is placed into a heated bore, and the molten plastic is driven out the bottom by a piston that descends because of weights placed on it. The temperature and the mass of the weights are prescribed by ASTM D1238 or agreement. The mass of material that is extruded through a small hole at the bottom over a known time interval is then converted to grams/10 minutes. And that's it. It really no different than measuring how long it takes for water to drain from a bucket [*].

So given a test like this that has been around for decades, well before computers and electronics were around, there is an endless temptation for people to bring the test into the modern/computer age by adding copious amounts of sophistication to the simple test. In some cases, it is helpful. Anything that can help put on and remove the weights is great. The weights are in the range of 1 - 21.6 kg (!) and need to be lifted up high over hot equipment, so any mechanical assistance is desirable particularly for short operators or anyone with weak arms.

But in other cases, the additions are non-productive or even counterproductive, such as when a computer calculates the shear stress, shear rate, viscosity, phase of the moon, and tonight's winning lottery numbers. Any number for the shear stress, shear rate and viscosity are meaningless. The die through which the polymer leaves the piston is too short - it's only 4 times the die's diameter. The length needs to be at least 20 times the diameter in order for the polymer to have "forgotten" its previous flow conditions, such as when it underwent extensional flow going from the large diameter piston down to the low diameter exit die. The flow in the melt flow die is shear that still has remembrance of extension, and yet all the equations used to in the calculations assume that there is only shear flow and no extensional flow.

If I ever see those numbers associated with a melt flow test, I ignore them. That's why I earlier called them "non-productive". But for too many people, they numbers become real and they put faith in them as being representative of the polymer. That's counterproductive or even dangerous. Ignore the shear stress, shear rate and viscosity. The test was designed to measure melt flow rate and only melt flow rate and that's all the more it will ever measure, and no computer-base calculations will ever change that.

[*] Please don't think for a minute that draining water from a bucket or the like is a moronic test. Glass capillary viscometry is also draining water from a bucket, but it is a test with plenty of theoretical support - and the results show it. However, that's another story for another day.

Gels are not just protein-based

The American TV Chef Alton Brown was recently asked his opinion about the famous "Cupcake-Frosting-Is-A-Gel" incident, in which a TSA agent confiscated a woman's cupcakes because 1) the frosting was a gel, and 2) the amount of the gel was over the legal limit of 3 ounces.

Alton had this to say:
"You know what: Icing is not a gel. By definition, icing is not a gel. Because a gel—we get that word from gelatin, which implies coagulated proteins, so it is not. Technically frosting is a condensed syrup, so I would argue with TSA that they were out of their freaking mind."
I admire Alton quite a bit. His TV shows are both entertaining and informative, but in this case, Alton is clearly wrong. While it is true that the word gel does have its origins in the gelatin , the word has expanded it's meaning well beyond those limited origins. As was noted in the link above, hair gel as a word arose in the late 1950's, and many hair gels are not protein based. Further, Paul Flory in his Introductory Lecture ($) to the 1974 Faraday Discussions noted that there were 4 different types of materials that were considered gels"
  1. Well-ordered lamellar structures, including gel mesophases
  2. Covalent polymeric networks; completely disordered
  3. Polymer networks formed through physical aggregation; predominantly disordered, but with regions of local order
  4. Particulate, disordered structures
What this really means is something that rheologists have known too well for too long - gel is a poorly defined scientific term, one that is used to describe materials with similar macroscopic physical behavior even as the underlying microscopic physiochemical make-up varies wildly. This is approach taken by the TSA - they are concerned about materials with macroscopic gel-like properties, not what the chemical constituents are.

Lastly, it is possible that the frosting on these cupcakes was in fact a gel, as royal icing has egg whites in it. Egg whites have lots of water, but also the protein albumin, meaning that even under Alton's limited definition of gel, it would be a qualify.