It is becoming common knowledge that ketchup is "thixotropic". More and more people are beginning to talk about it, which means that this is a good time to put the idea to the test. In addition, the Grand CENtral blog is sponsoring a #FoodChem carnival this week, so this is my entry.
What most people have observed is that ketchup is a thick fluid, and in glass bottles in particular, very difficult to get flowing. However, once the flow starts, it flows quite readily, usually leading to too much ketchup ending up on your burger and fries. The question is how do rheologists describe this behavior and is this sudden onset of flow really due to "thixotropy"?
Before we can get into that discussion, there are a few terms that I will define as they all play a role in the rheology of ketchup. But before I can get to those terms, I'm going to discuss what I mean by shear and shear rate. While there are mathematical definitions, I'll skip them for today and just describe these ideas qualitatively.
For any liquid that is flowing, the velocity of the liquid that is right up against the wall is zero , while the velocity of the fluid elsewhere is not. This means that the fluid is being sheared. The faster the fluid is moving, the higher the shear rate is, but also, the smaller the gap in which the fluid is moving, the higher the shear rate.
For a run of the mill liquid such as water, the viscosity of it is constant regardless of the shear rate. That makes it a Newtonian liquid. If the viscosity is not constant, then it is a non-Newtonian liquid. Non-Newtonian behavior comes in many flavors, but I'm only to going to discuss three options today.
The first is "shear-thinning". A shear-thinning liquid is one where the viscosity decreases as the shear rate increases.
Second is the le mot du jour, thixotropy, which is similar to shear-thinning but also decidedly different. At a constant shear rate, a thixotropic material will show a decrease in viscosity over time.
Last is "yield stress". This is the idea that certain materials need a minimal amount of force applied to them in order for flow to start. If less force is applied nothing happens.
It is possible for a non-Newtonian fluid to exhibit any combination of these characteristics. Ketchup in fact shows all three behaviors. But enough with the academic terms, let's get on with the show.
Like most places of work with a refrigerator in the cafeteria, there is an old bottle of ketchup sitting in it, bought some time ago for the company picnic that gets used every once in a while to spice up a tater-tot hot dish (a.k.a. tater-tot casserole if you never learned to speak Minnesotan). This morning I grabbed the bottle, got out the big 45 mm plates for the rheometer , squirted a portion out and took some data. (So does this qualify for #RealTimeChem too?)
The plot below shows the viscosity of the ketchup as the shear rate increases.
The next plot shows what happens when I subjected the ketchup to a constant shear rate.
I would have loved to have taken the data for this last plot, but I have the wrong type of rheometer for it , so I am borrowing a plot from TA Instruments.
So the question is this: when you get a big glug of ketchup shooting out of the glass bottle, is it due to the shear-thinning, the thixotropy or the yield stress?
Despite the trendiness of the term, the thixotropy is the easiest candidate to eliminate. Look at the small drop in viscosity. It's just not significant. And while the shear-thinning plots shows a large drop in viscosity, you need high shear rates to achieve that and you won't find that in the big part of the bottle - it's just too big a gap. So that leaves yield stress as the winner. Look at the plot of yield stress again. The viscosity suddenly drops by a factor of 1000 just by reaching a critical stress! That's why you suddenly get a massive amount all at once - you finally reached the yield stress.
While ketchup is indeed thixotropic and I am glad that more and more people are becoming familiar with rheology, that phenomenon is the least of your concerns in getting the ketchup out of the bottle.
 This is a very fundamental concept in fluid mechanics known as the no-slip boundary condition.
 This stuff is a pretty soft gel, so by using larger diameter plates and their associated larger torque, I am generating a larger signal for the instrument to pick up. Also, the plates were coated with 600 grit sandpaper in order to minimize slip at their surface.
 My rheometer subjects the sample to a strain (deformation) and measures the stress (force). The type of instrument I need for a yield-stress plot is one that subjects the sample to a stress and measures the strain. Santa, I've been a good little rheologist this year. Can I please have a controlled-stress rheometer for Christmas?