by David Davies

1    The Gross Process

The Gross Process is a method of catalytic distillation in which a metal is extracted from an alloy or purified from an impure source by heating with a gaseous halide.  The most important case is that in which aluminium is extracted from an alloy or purified from an impure form by heating it with aluminium trichloride.

It was invented by Dr Philipp Gross (PG) when he was working at International Alloys in Slough.

Early November 1943

PG is “beginning to be concerned with the problem.” We don’t know what exactly the problem was. A discussion took place between PG and the addressee of a memo (ref: V501) discussed in detail below; this discussion is referred to in the memo.

16th February 1942

Fortunately Dr Gross wrote down his thinking in a memo, the first page of which we have in our archive (Ref: V501).  I have copied it below, with my added explanatory notes.

The memo is incomplete; only the first page survives. It doesn’t say  to whom it is addressed. I (DD) was the one who entered its record into our references spreadsheet. I am sure that it came from Ron Lewin’s collection and it was almost certainly Ron who told me that the addressee was Mr W.E. Prytherch (WEP). This certainly makes sense; PG was, at the time, the senior scientist at International Alloys (INTAL) and WEP was Chief Metallurgist at High Duty Alloys (HDA), a sister organisation. The tone of the memo is that of a communication between peers.


From: Dr. Gross,                                              16th February 1944

                       On my visit to the Patent Office library on February 11th, I have seen for the first time the original paper by Klienen and Voss on Distillation of Aluminium with Aluminium fluoride. (Zeitscher. Anorg.Ch.251, 233, (1943)).

                       This publication was not available in this country before. As I concluded from the abstract in American Chemical Abstracts the authors have carried out an experimental investigation into the distillation of Aluminium with aluminium fluoride on lines very similar to those which I suggested to you early in November last, when I was concerned with the problem for the first time.

Klienen and Voss appear to be testing the invention of Willmore, see below)

(Incidentally Kleinen, who is one of the best inorganic chemists in Germany, carried out the work on behalf of “Vereingte Aluminium Werke”, the large German aluminium combine).

                       The results are : From aluminium trifluoride (AlF3) and aluminium, an aluminium mono fluoride (A1F) is formed which distils at about 700°C in a high vacuum according to:

                                               2Al + AlF3 ↽=====⇀ 3AlF

and decomposes again into aluminium tri-fluoride (AlF₃) and aluminium metal below 700°C.

                       The weight ratio of aluminium to aluminium fluoride in the distillate under the proper conditions is therefore 54 parts aluminium to 84 parts aluminium fluoride or 39 per cent. of the whole distillate is free aluminium metal.

                      The paper by Kleinen and Voss is concerned only with the inorganic aspects of distillation and does not deal with the physical chemical side. From the given distillation temperature I would estimate that the energy consumption for formation and distillation of one AlF is about 55,000 cal., its heat content at distillation temperature is about 12,000 cal. The power consumption – not considering losses from these data would be 4,500 KW h per ton aluminium, or the same as that of a cell running with 100 per cent. current-efficiency at a voltage of 1.5 v.

The cell that PG is referring to is the electrolysis cell used in the Hall process in which alumina, dissolved in molten cryolite at about 1000°C, is electrolysed to produce aluminium.  This was and still is the standard method of extracting aluminium from purified bauxite, its ore.

If the thermal efficiency is only half, which should not be too difficult to achieve the power consumption of 9000 KW h would be the same as that of a cell running at 3 v.

                      So far only spectroscopic indications for the existence of an aluminium subchloride are not controversial. I think it probable, however, that it is formed under the same conditions as the monofluoride. Aluminium chloride sublimes at 180°C at atmospheric pressure.

PG is proposing to use the chlorine analogue of the fluorine experiment carried out by Kleinen and Voss, backing his hunch that AlCl is stable at elevated temperature.

                                                               1000°C                                                                                                                                                              2Al + AlCl3 ↽=====⇀ 3AlCl                                                                                                                                                                        cooling

It will therefore be necessary for the investigation to pass aluminium chloride vapour under low pressure (sublimation point 99°C at 1 mm. Mercury) over aluminium or aluminium ferro-silicon at about 1000°C. The aluminium subchloride if formed, will distil off. On cooling the aluminium subchloride should decompose condensing aluminium and aluminium trichloride in different zones.

                      If this method, which as far as I can see would not come within the scope of U.S. Patent 2,184,704 (Ch.B.Wellmore) be successful, it would have the great advantage that aluminium powder and aluminium chloride could be separated very easily.

This was surely a typo. Patent US 2,184,704A was entitled “Enameling Machine” and was totally unrelated. It should have read US 2,184,705A “REFINING AND PRODUCING ALUMINUM”, inventor C.B.Willmore, Assigned to ALCOA. According to this invention, aluminium is distilled from a mixture containing impure aluminium and a halide, preferably a fluoride, such as aluminium fluoride, magnesium fluoride or cryolite, in an atmosphere inert to aluminium at a temperature such as 900°C or 1000°C. Patent applied for 1939-09-18 granted 1939-12-26.

This could be done either by distilling the aluminium chloride off at about 100°C or preferably by keeping the aluminium chloride vapour in a complete cycle.

This is PG’s essential insight. Because AlCl3 readily vaporises it is much easier to separate from aluminium than is AlF₃ which is solid up to 1290°C.

The theoretical…..

End of surviving part of the document.

We could plausibly suggest that the lost continuation page 2 might begin with composition and energy calculations on the chlorine analogue experiment.


18th April 1945

US 2470305 Process for the Production and Refining of Aluminum
Inventor Philipp Gross
US patent Applied for

19th April 1945

CH255758A Process for preparing aluminum
Swiss patent applied for by INTAL

16th November 1945

Formation of Almin Ltd announced, including International Alloys Ltd (Ref: F142a)

7th May 1947

PG begins a new laboratory notebook (Ref: V864)

On the inside front cover

P.GROSS, 17,Clifton Rd Slough

The first page is blank

What would be page 2 has the following table of enthalpies and entropies:-


ALCl3 hₒ -139000 sₒ illegible
h₁ₒₒₒ -119500 s₁ₒₒₒ illegible
AlCl hₒ -12800 sₒ 53.73
h₁ₒₒₒ -4020 s₁ₒₒₒ 66.80
Al hₒ 2440 sₒ 9.03
h₁ₒₒₒ 9230 s₁ₒₒₒ 19.19


This shows that PG was continuing to work on his process for aluminium chloride catalytic distillation of aluminium.

2nd July 1947

The following is an extract from the Brochure for the official opening of Fulmer Research Institute (Ref: F308)

“Metal Distillation

Sponsor- International Alloys Ltd,

Some time ago, the existence of monohalides of aluminium was deduced from spectroscopic data, and calculations by Dr. Gross, based on reasonable assumptions and the application of thermodynamics, suggested that the monohalides of aluminium would be stable at low pressure and high temperatures. That these deductions were correct was proved experimentally by showing that aluminium could be distilled catalytically at temperatures in the region of 900-1000°C, despite the fact that the vapour pressure of aluminium at 1000°C is of the order of only 0.0002 mm and at 1550°C. of the order of only 1 mm. The catalytic method of distillation involved leading the vapour of the normal trivalent halide over aluminium or aluminium alloys at temperatures of the order of 900°C and pressures of the order of 1 mm. when the monohalide was formed. The chloride was mainly used for this experimental work. The reaction

2 Al + AlCl3 = 3AlCl

moves to the right at high temperatures and low pressures, and on cooling the monochloride vapour decomposes into the normal trivalent aluminium chloride and aluminium, i.e. the reaction proceeds to the left. The aluminium produced in this way has been shown to be of much higher purity than the original alloy, and the basis of a process for purifying aluminium of high impurity content and possibly also for the extraction of aluminium from comparatively dilute aluminium alloys which may be produced by electrothermal reduction. Before the economic possibilities of such a process can be assessed the establishment of the energy requirements and equilibrium of the reaction had to be established in the laboratory, and the kinetics (rate at which the reaction proceeds) also determined. Work is still in progress on the kinetics of the reaction and on various problems connected with experimental plant design. The experimental work in connection with this problem is in progress in the vacuum laboratory.”

The fact that the process is described in detail implies that Fulmer and INTAL had confidence in its patent protection.

15th July 1948

CH255758A Process for preparing aluminum
Swiss patent Granted.

17th May 1949

US2470305 Process for the Production and Refining of Aluminum
Inventor Philipp Gross. Assignee International Alloys Ltd,
US patent  Granted

21st February 2018

Entry in frhg minutes

Ron Lewin explained that he had learned from Veryan Scott-Hayward, daughter of Margaret Harkness (MLRH), that she had at one time been a colleague of Dr Gross, possibly at Cardiff.

2    William E Prytherch

Mike Dewey has written extensive notes on William Prytherch (WEP) in two papers which are now designated Ref: V136 and Ref: V135.  This section is supplementary to those documents and only includes their contents when relevant.


WEP is living at Meads, Stoke Green, Stoke Poges in the household of W.C.Devereux. We believe he is working in a senior position at HDA at that time.

Our evidence for WEP’s work at HDA comes from a number of research reports donated by Veryan Scott-Hayward, the daughter of Margaret Harkness (MLRH) who worked closely with WEP at HDA and later.

Here are the subjects that we know WEP and his group reported on:-

3rd December 1941 Manufacture of Light and Heavy Magnesium Carbonates (Ref: H020)
28th December 1942 Investigation of Bicarbonation of Magnesium Hydroxide Slurry (Ref: H024)
27th January 1943 The Extraction of Alumina from Redditch and Intal (Slough) Dross (Ref: H022)
24th May 1943 Production of Magnesium Fluoride from Magnesium Sulphate and Sodium Fluoride (Ref: H004)
27th September 1943 Further Work on the Production of Alumina from Intal Dross (Ref: H005)
17th April 1944 Production of Alumina from Shale (Ref: H003)
12th July 1944 Production of Magnesium Chloride (Ref: H001)
25th June 1945 Production of Drinking Water from Sea Water (Ref: H006)

The reports that we happen to have also give some evidence of career progression for WEP.  In 1941 and 1942 the reports, even when authored by WEP, have been signed off by R.Kreissman Metallurgist. Starting in 1943 WEP signs off his own department’s reports as W.E.Prytherch Metallurgist while other managers sign off other reports as e.g. H.G.Warrington Metallurgist. Finally, in 1945 WEP signs off Ref: H006 as W.E.Prytherch Chief Metallurgist.

Also relevant is a letter to WEP from J.H.Anderson, a colleague of WEP, who was designing and building a pilot plant for producing alumina from dross.  The tone of this letter (Ref: H010) is very much that of a peer rather than a subordinate.

In 1943-4 Colonel Devereux tried and failed to persuade the HDA board to diversify away from the aircraft industry. He sold his HDA shares and raised further capital to back his new group Almin.

31st August 1945

The following news item appeared in the Slough, Eton and Windsor Observer (Ref: F171)


Technical Director to Act As Consultant

Mr. W. E. Prytherch., M.Sc., F.R.Ae.S., has relinquished his position as technical director of High Duty Alloys Ltd., to become an independent research consultant.

He will act in that capacity for the group of companies which Col. W. C. Devereux is at present forming.

16th November 1945

Formation of Almin Ltd announced, including International Alloys Ltd (Ref: F142a)

WEP, together with MLRH, seems to have set up CML Limited at Dunloe Lodge, Taplow, Bucks, specifically to continue the research work they had been doing at HDA for the Ministry of Supply on the Production of drinking water from brine.

They also either set up or at some stage joined the Board of Xenit Products Limited. This company supplied mouldings and moulding powders (and batteries) at 95 Farnham Road Slough.

2nd April 1946

A letter to WEP at Little Hockeridge, Ashley Green, Chesham

re the Novation of the MoS contract on PRODUCTION OF DRINKING WATER FROM BRINE USING BASE-EXCHANGE RESINS (Formerly with HDA) in favour of CML Limited, Dunloe Lodge, Taplow, Bucks (Ref: H201)

29th August 1946

Letter to MoS

re Contract 6/Floatation/1/CF9A
Margaret Harkness signs as Company Secretary of CML Limited (Ref: H202}

2nd July 1948

W.E.Prytherch Director

9th November 1953, 22nd January 1954, 8th March 1954, 5th April 1954, 16th February 1955

M.E.Prytherch M.Sc., F.R.Ae.S. and M.L.R. Harkness BSc. are both listed as directors of Xenit Products Ltd 95 Farnham Rd Slough

Covering letters for minutes of Board Meetings and other documents (Ref: H391), (Ref: H394), (Ref: H395), (Ref: H396), (Ref: H397)

M.E.Prytherch M.Sc is living at Parsonage Farm, Warehorne, Ashford, Kent

3    Conclusion

We have no evidence that WEP made any contribution to PG’s thinking on catalytic distillation of aluminium.  What the two had in common was a commercial interest in reducing the great cost of aluminium production: PG by inventing a new process; WEP by economising on dross wastage.

Apr 2023

FRHG Ref: V127