Spherulites, gels, phase separations…

1. Spherulites

Spherulites can be difficult to distinguish from oils unless you probe them with a whisker or needle. (Oils are dense liquid phases, kind of like honey; spherulites will be crunchy, solid.) Another feature that differentiates them: look closely at the edges of the spherulites. They are not completely round, as would be a liquid-liquid phase (oil). See further down for pictures of liquid-liquid phase separations.

2. Spherulites

If you can't see the difference between an oil and a spherulite, don't worry. In practice, it doesn't really make any difference: treat the oils or spherulites in the same manner. Use them as seeds in a new drop. Of course you can also screen around the conditions. Maybe all that is required is fine-tuning the concentrations of precipitant, etc.

3. Spherulites


Spherulites can also be quite huge, as shown here. The biggest one is about 100 microns  in diameter. This is a test protein, bovine trypsin inhibitor (BPTI).

4. Not a spherulite

But don't be fooled. Not all spherulites are spherulites—they can, in fact, be crystals. One day, back in September 2000, Matthias Dreyer, then at University of Wurzburg,  sent me an email to tell me about his unusual-looking spherulites.  When he opened up the drop, these "spherulites"  shown on the left, flipped over on their side and revealed a half-dome shape. These were not spherulites at all, but crystals that diffracted to 1.92Å. It's not a common space group: F432. The protein is fuculose-1-phosphate aldolase and you can read more about the story behind these crystals here. (Picture is taken from the figure in this paper.)

I've lost track of Matthias, but if he sees this, I say hi and thank him for bringing my attention to this interesting example.

I have run into crystals of this shape maybe five or six times over the last 35 years,  so  they are not common.

Moral of the story: no matter what it looks like, put it in the beam and shoot it.

5. Spherulite or dome-shaped crystal?

Here is an example from my own work. In this view, the crystal looks like a half-moon or dome-shaped crystal (just like the fuculose adolase crystals above), but flipped on its side, it looks like a flat spherulite. I shot it last month (Oct 2019) at the beamline at MAX IV (Lund, Sweden). It is mounted in a 0.1 mm Lithloop. This crystal of LpxH did not diffract however.

Sea urchins

These often originate as spherulites and thin needle-like crystals grow from the single nucleation site.

These will require some optimization to turn them into fine, single crystals, but are certainly worth pursuing.

For more pictures and information, go to tutorial 4 "Crystals with Problems".

1. Gel

Synonyms: Gelatinous protein, gelled protein, protein gels


This is my favorite example of a gel because the protein is red (it is a heme-binding protein). When this gel  was used to streak-seed new drops, good crystals grew.

How is that possible? Read more in the next block.

2. Gel

Drop (200 nl volume) with gelated protein.

These gels are dense phases of the protein. Back in the 1980s, we argued at conferences about the nature of these gels. Were they denatured protein? Oxidized? Whatever they were, they were considered a dead-end in terms of crystallization.

Recent research, however, shows that they are not so amorphous as they seem. Indeed, XFELs and microED demonstrate these gels contain short-range ordered structures, even nanocrystals. Read more here and here.

This explains why seeding with the protein gel would work.


3. Gel

Close-up view of gelation and precipitate (the darker regions).

4. Gel

Another example of gelled protein, this time a close-up view. When you see this in your drops, you are very close to the right conditions; screen around the conditions that give you this.

For example, you might also consider including a reducing agent such as beta-mercaptoethanol, TCEP, or DTT in the protein solution. Does your protein have free cysteines? You want to keep them reduced.

But my primary recommendation would be to use the entire drop as a seed stock and seed with it. It may contain nanocrystals. (See above, 2. Gel).

5. Gel

Transparent, irregular regions in the drop. You are very close to the right conditions. In this example, the large crystal in the center of the drop grew some weeks after the gel. Some of the gel is covering the right side of the crystal.

6. Gel

Another example of what gels look like, this time in a microbatch drop. Compare the difference between gels and amorphous precipitates. Gels have at least some kind of short-range order to them. Do a grid screen or an additive screen around the condition where you obtained the gel.

1.Liquid-liquid phase separation (LLPS); "oiling out"

A classic example of a liquid-liquid phase separation in the kitchen is salad dressing (oil & vinegar).

LLPS in the crystallization drop will often appear as hundreds of small droplets. If a lot of the small droplets coalesce (merge together), you will get a great big drop, like the one you see here in the middle.

The protein will preferentially "oil out" into one of the liquid phases more than the other. (You can only have two phases simultaneously.) When the protein "oils out", it forms a very dense liquid. Hence, liquid-liquid phase separation and "oils" are two descriptors of the same phenomenon.

Click to learn more about phase separation and get tips on how to optimize it. 

2. Liquid-liquid phase separation (LLPS)


In this example, a huge crystal (>0.3 mm in the longest direction) grew some weeks after the  phase separation. So, yes, you can get crystals in drops with phase separation, just as you can get crystals growing in drops full of precipitate. 


3. Liquid-liquid phase separation (LLPS)


Crystals growing inside the  LLPS

Here we clearly see that the protein has preferentially sequestered itself into one of the phases. It becomes highly concentrated in the phase, and the high levels of supersaturation in this phase are more conducive to nucleation than in the other phase. These crystals show up beautifully in the UV imaging (see next image).

4. Liquid-liquid phase separation (LLPS)


UV image with LLPS

What to do next though? Obviously these crystals are too small to mount, so why not use them as seeds. Or stir the drop. That redistributes the tiny crystals in the drop and might kickstart some new growth. Or do an additive around the condition that produced this, and use these as seeds.

What did I do with these crystals? See the result and video in tutorial 5, seeding.

5. Liquid-liquid phase separation (LLPS)

LLPS droplets on the crystal itself

References coming soon.



6. Liquid-liquid phase separation (LLPS)

Just joking.

It's a painting by the Russian artist Wassily Kandinsky (1866-1944). But I hope you can agree that it looks like a drop of phase separation that someone colored in.

Art imitates nature. Or is it the other way around?

6. Beam me up, Scottie

(That's a line from Star Trek.)

But as everyone should know by now—given her recent popularity and Guggenheim exhibitionHilma af Klint  (1862-1944) is the Swedish artist whose work preceeded Kandinsky's abstract art by at least a decade. This painting (1902) is from a series called  "Duvan" (The Dove).  I leave it to you to figure out why I felt compelled to include it on a site about crystallization.