# What is the maximum number of water molecules that has been recorded till date to present as water of crystallization in a compound?

Context: I was casually looking through some old questions when I found this question. The OP wanted a compound that has all the hydrates ranging from monohydrate to decahydrate. I found chromium(III) sulfate [$$\ce{Cr2(SO4)3.x(H2O)}$$ where $$x=0,\ldots,18$$] to be suitable. But, the actual formula of the compound is $$\ce{[Cr(H2O)6]2(SO4)3·6H2O}$$, so 6 of the 18 molecules are present as water of crystallization. So, I wondered, what is the maximum number of water molecules that can present in a compound as water of crystallization?

I can agree to what @BuckThorn said in his answer. It might be protein crystals. Wikipedia also supports the fact:

Compared to inorganic salts, proteins crystallize with large amounts of water in the crystal lattice. A water content of 50% is not uncommon for proteins.

Now, the question is what protein has the maximum number of water molecules as water of crystallization?

I did some cursory searches and looked for inorganic hydrates (see edit history: 36-hydrate is maximum I could find). I am also looking for such compounds which is in inorganic domain.

• There are gas hydrates which are pretty much mostly ice, but I guess they don't count since technically they are clathrates, not hydrates. Apr 12 at 7:17
• Would it make more sense to do wt percent instead of number of water molecules so you aren’t biased to larger unit cells? Apr 14 at 0:22
• I edited tags to reflect that your interest is in inorganic hydrates. I sort of apologize for my answer. I realized proteins weren't perhaps what you were looking for, but I'll leave it up anyway as food for thought. Hint: look for a really big and spherical inorganic molecule.... Apr 19 at 7:20
• @BuckThorn No. I am not interested in inorganic hydrates (I was initially) but I edited the question to reflect your POV (+ a quote from Wikipedia) and that I am interested in any compound(organic or inorganic) that has the maximum number of water molecules in its water of crystallization (which happens to be proteins). I will accept your previous answer once the bounty period is over (I was waiting for other answers). Regarding that link, you happened to find the answer (it would have been a duplicate if I hadn't edited the question). +1 to both of your answers. Apr 19 at 8:04
• However, I am removing the 'inorganic chemistry' tag to show that I am not bounded under the 'inorganic domain. I am just looking for a canonical answer which describes a compound of inorganic domain. Apr 19 at 8:06

## 2 Answers

It might be a protein crystal. Proteins are large and the interstitial space is filled to a large extent with water. The volume of a protein crystal occupied by the solvent can be 50% or so. However the fraction of waters observed in crystals (relatively fixed) is often low, on the order of 10%.

For instance, from a report on the crystal structure of crambin (chosen arbitrarily, see Ref. 1):

The volume of solvent (32%) is low but not atypical for proteins [values of 30-78% solvent are found.] Crambin is small (Mr 4720), and most residues are solvent accessible. Although crystals were grown from 60% ethanol, only 7%(vol/vol) ethanol has been located. Of the 85 water molecules expected (based on the density and the presence of 2 ethanol molecules), the model includes 64 [...]

It's not clear whether the 85 expected or the 64 found water molecules would be considered crambin's water of crystallization (presumably all count, even if not observed).

As another example, the model crystal structure of BPTI contains 63 waters (observable waters, see PDB structure 5PTI).

While I was looking for the protein crystal structure with greatest number of waters in its model, Karsten Theis suggested as an example the pyruvate-bound crystal structures of ADP–glucose pyrophosphorylase from the bacterium Agrobacterium tumefaciens (pdb 5W5T, see the file for a description of the solvent model), which has 3752 waters in the crystallographic model (59.12% of the volume).

Many waters may have irregular positions, and some may form clathrate-like solvation structures (as in gas hydrates, as andselisk comments), and some can be placed at regular locations.

References

1. M. M. Teeter. Water structure of a hydrophobic protein at atomic resolution: Pentagon rings of water molecules in crystals of crambin. Proc Natl Acad Sci U S A. 1984 Oct; 81(19): 6014–6018. doi: 10.1073/pnas.81.19.6014
• You might also take a look at pubs.rsc.org/en/content/articlehtml/2016/sc/c5sc03447g which is an ab initio study of crambin, and considers the water associated with the crystallised protein Apr 12 at 16:39
• At room temperature, most of the water in typical protein crystals is in the liquid rather than the solid state. So it is not clear if you want to count those, or only the ones that are fairly stationary. Apr 12 at 21:00
• @KarstenTheis I've been looking around for the protein crystal structure with the largest reported (static?) number of interstitial waters. You have more experience and may know? Apr 12 at 21:16
• I know that photosystem I crystals have 90% water with a huge unit cell. There are currently 8 structures in the PDB that have more than 10,000 water molecules in the model. One of these structures is 5W5T, if you are interested in an example. Apr 12 at 21:37
• Apr 13 at 5:55

This feels like cheating, since andselisk did all the hard work already and could have provided you with the answer, but I would think

$$\ce{Na48[H_xMo368O1032(H2O)240(SO4)48] · \text{ca.}~1000 H2O}$$

is a good candidate for the largest inorganic hydrate. Note the "circa" 1000 waters. Unfortunately I don't have access to the Angewandte Chemie article and could not confirm whether any waters were observable spectroscopically or by scattering, clarifying how many waters are in fact crystalline.

The reasoning why this would be the largest hydrated inorganic compound follows from my other answer: the larger and more spherical the inorganic compound the more interstitial space is available; more interstitial space means more room is available for interstitial water. Crystallization however usually requires regular specific interactions with the inorganic core, at least in order to observe coherent scattering (water may instead be crystalline but irregularly distributed).

Please see the answer to "What is the largest, noncrystalline, non-polymer inorganic molecule?" for more on the compound.

• The crystal-water content has been determined thermogravimetrically. Also, some oxygen atoms from the water molecules have been allocated on the electron-density map as seen with the deposited structure (CSD-412283). As for posting the answer, I didn't think it actually counts as a record since I figured OP is asking about normalized water content per metal center, which, being calculated for $\ce{Mo386}$ cluster, is not that impressive. Apr 19 at 12:03