Search by keyword, name or CAS number

A double salt of aluminium sulphate and aluminium chloride (AlClSO4.6H2O)

Over at ScienceMadness, member ‘CHRIS25’ got interested in a reference (link to atomistry) to an ‘aluminium chlorosulphate’, which states:

When aluminium sulphate is boiled with concentrated hydrochloric acid and the liquid cooled, crystals of aluminium chlorosulphate, AlClSO4.6H2O are deposited, which are decomposed by water.

I was sceptical about this claim, partly because ‘chlorosulphate’ in this context is a misleading term (it is not a salt of chlorosulphuric acid), partly because of the lack of almost any other authorative references to this substance and the fact I couldn’t find a CAS number for it (but I don’t have direct access to the CAS Registry).

CHRIS25 then prepared a white, crystalline material by boiling homemade aluminium sulphate hydrate (ASH) with conc. HCl until all sulphate had dissolved. On cooling the solution the material then crystallized out.

I did the same by using 20 g of commercial ASH (molar mass 600 g/mol, hydrate number n = 14.3) with 40 ml of 37 w% of HCl and also obtained on cooling and icing a mass of white, small crystals. These were filtered off and washed with small amounts of iced water, sucked dry, dried further on filter paper in a refrigerator and finally dried in a CaCl2 desiccator to constant weight. Based on the assumed formula AlClSO4.6H2O, actual yield was about 40 %.

A few quick tests showed that, contrary to the atomistry claim, the material dissolved in water with gusto and to a clear, colourless solution.

1. Demonstration of bound sulphate and chloride:

At this point it could not yet be excluded that the crystalline material was either simple aluminium sulphate hydrate or simple aluminium chloride hydrate.

To a solution of the material, assumed about 0.05 M, was added an excess of chloride-free barium nitrate. Strong precipitation of BaSO4 proved that the material contained chemically bound sulphate ions.

This slurry was filtered and to the sulphate free filtrate was added and excess of silver nitrate. Strong precipitation of AgCl proved that the material contained chemically bound chloride ions.

2. Determination of aluminium content:

On this first batch the aluminium w% was determined by back titration of a solution of this salt, to which a known excess of Na2EDTA was added. Back titration was done with standardised ZnSO4, using Eriochrome T as indicator. The zinc sulphate titrant had been standardised against reagent grade CaCO3 (dissolved in HCl). 9.4 w% aluminium was found.

3. Determination of chloride content:

Another batch was prepared in identical conditions (but twice the scale), this time for determination of chloride content. A slightly modified version of Mohr’s method, titration of chloride with silver nitrate and potassium chromate indicator, was used. Because at the required titration pH of 6.5 to 9, Al cations cannot exist in solution (at least not at 0.05 M), aluminium had to be removed from the sample solution prior to titration.

This was done by adding a precise amount of 33 w% ammonia solution (about 2 ml) to the sample solution very gradually, to take the pH up to 8. The Al then precipitated as Al(OH)3. It was filtered off on a Buchner filter and the filter cake washed repeatedly with water to a volume of about 200 ml filtrate. This filtrate was transferred quantitatively into a 250.0 ml volumetric flask and diluted to the mark. This solution was expected to be about 0.05 M in chloride (based on the AlClSO4.6H2O empirical formula).

0.1 M silver nitrate was standardised with recrystallised sodium chloride (0.1 M). The titration on the sample solution then yielded a Cl w% = 12.7 w%.

As an interesting side note, this titration showed that Mohr’s titration can be carried out in the presence of modest amounts of sulphate (here, with hindsight about 0.05 M) because although silver sulphate is poorly soluble it is far more soluble than silver chloride, so the sulphate does not interfere with the analysis.

4. Determination of empirical formula:

From the combined results Al = 9.4 w% and Cl = 12.7 w% could be deduced that the molar ratio Al/Cl was 0.995, or in other words 1 (one). From the neutrality requirement can then be deduced that the molar ratio Al/Cl/SO4 = 1/1/1.

From this the water content and hydration number could be estimated: the latter computes to 7. Considering it was obtained indirectly I’m inclined to give more credence to the n = 6 number quoted by atomistry.

That would make the empirical formula of the double salt:


Here’s about 10 g of this salt: