Journal of Australasian Mining History, Vol.

2, September 2004

The Extraction of Gold by Amalgamation and Chlorination

By RALPH W. BIRRELL
University of Melbourne

T
he gold ores found above the waterline in the Lachlan Fold Belt of south
eastern Australia in the 1850s, were free milling, meaning that most of the
gold could be easily separated from the quartz by crushing, followed by
amalgamation with mercury.1 The particles of gold were relatively large and separated
cleanly from the quartz. Above the water line the sulphides of iron, arsenic, lead, silver,
antimony, bismuth and copper deposited with the ore had mostly been oxidised over
millions of years, leaving gold particles free of sulphides. This easily separated gold
could be amalgamated with mercury without difficulty. If, after crushing, small amounts
of these impurities had not been oxidised and were still attached to the gold particles,
the mercury was often sickened, resulting in it being coated with a black scum. This
prevented the gold amalgamating and both the mercury and some of the finer gold was
washed away with the water if wet crushing was used. Ores from below the water line
were not oxidised and the several mineral sulphides in the crushed ore caused sickening
or flouring of the mercury when it was separated into very small droplets during
crushing and these were also washed away in the water and lost.2
Historians of Australian mining have written little on the technical aspects of
amalgamation of gold with mercury and the use of the chlorination process in the
extraction of gold from its ores.3 In her publication on colonial technology, Jan Todd
criticised the slow acceptance of the chlorination process in Australia compared with the
short time it took to utilise the process in South Africa and New Zealand, stating that:

The transfer of cyanide gold extraction benefited from the sharing of common
institutions, language and culture and from the steady flow of people. A
multitude of colonial connections were able to smooth away early obstacles in
the patent system. More difficult was the distances between the practices of
Australian goldmining and the leading edge of the technology frontier which this
chemical process represented: despite some pockets of experiment with
chlorination, the vast bulk of the goldmining industry was ignorant of chemical
extractive processes and the scientific principles which governed them. These
were considered the province of the much maligned ‘theorist’.4

17
 Ralph W. Birrell

It is the intention of this paper to discuss the use of calcination as part of the
process of amalgamating the gold with mercury, the problems of amalgamating pyritic
gold ores, and the use of the chlorination process after the 1870s to improve the
extraction rate of the gold from the ore. The validity of the comment by Jan Todd on the
lack of knowledge in Australia about chemical extraction processes and chlorination
will be assessed as part of this discussion.

Calcination
When quartz mining commenced in eastern Australia from about 1854, many miners
believed that heating the quartz was beneficial in two ways: first, it made the quartz
softer and friable and more easily crushed by the primitive machines then in use;
second, if any sulphides were present they would be oxidised.5 Ore was mixed with dry
wood in piles or in simple kilns, the wood was ignited and the pile burnt, a process
called calcination. This oxidised most of the sulphur present to sulphur dioxide, which
combined with water vapour in the air to form sulphurous acid so that any arsenic and
antimony sulphides present were sublimed as oxides. Although the sulphides of other
metals present were oxidised, some of the oxides such as lead and copper remained with
the ore and made amalgamation difficult as they coated the gold particles. These ores
were said to be refractory.6 In 1860, a Mr. Wilkinson demonstrated a furnace at Ballarat
that used charcoal to decompose steam into hydrogen and oxygen that were then burnt
in a cupola containing quartz ore. He claimed the fine gold in the ore was heated to a
high enough temperature to form globules that amalgamated without trouble. The
Ballarat miners conducted tests on the plant and declared it to be uneconomical.7
The Port Phillip and Colonial Mining Company was formed in London in 1852
by a group of men who owned other companies mining for gold in South America.
After searching for a suitable mine in Victoria, the resident director, Rivett Bland,
leased private land at Clunes in 1857 and established a successful quartz mine. By the
early 1860s the company was mining sulphide ores from below the water line.8 Some of
the staff had treated gold ores in Brazil containing 15 per cent pyrites and although the
Clunes ores contained less than one per cent pyrites they included a kiln as part of the
first plant with which to heat all the quartz before sending it to the stampers. This was a
large bottle shaped kiln with work platforms top and bottom. Ore and timber were
thrown into the top of the kiln from trucks on rails, in alternate layers until the kiln was
full.9 Similar kilns had been used for many years for burning seashells to make lime.10

18
 The Extraction of Gold by Amalgamation and Chlorination

The timber was then ignited at the bottom and as the burnt quartz settled down to the
hearth it was quenched with water and shovelled into trucks that were then wheeled to
the stampers.11 This was designed to be a continuous process to make the ore more
friable for crushing, but as the output of the mine increased it became more difficult to
continue this process due to volume and cost. It was abandoned in 1862 and replaced
with a Chambers rock crusher that acted as a primary crusher before the stampers.12 At
many early quartz mines in Victoria the quartz ore was burnt in surface stacks or in
kilns before crushing and this practice was still in use at Stawell in 1869.13 The ratio of
pyrites in Stawell ores was said to be one in a thousand, so it seems calcining was used
there to make the ore more friable,14 although it is difficult to decide from the literature
whether the use of this practice was for that purpose or to alter minerals in the ore. It is
possible that old methods persisted without justification, although one witness at the
1890 Royal Commission on Mining claimed that calcining Maldon ores increased gold
production by 25 pennyweights per ton.15 The second reason for calcining the quartz
will be considered further when examining the treatment of pyrites.

Amalgamation of gold
On 21 December 1893, a meeting of the Institute of Mining and Metallurgy was held at
the Museum of Practical Geology in Jermyn Street, London, at which the President,
George Seymour called on a member, C.G. Warnford Lock to present his paper on
‘Gold Amalgamation’, commenting that there was no issue in mining that offered so
large a field for controversy. The ensuing discussion recapitulated debates conducted
over the previous forty years in California, Australia, South Africa and Great Britain but
what nobody disputed was that nine-tenths of the gold produced worldwide was
recovered by the amalgamation process.16 Included in the discussion was a survey
conducted in 1874 by the Victorian Mines Department, when a wide range of people
involved in mining, plus people living near processing plants and doctors of medicine
were interviewed.17 The Board members directed their questions to amalgamating
methods then in use and the problems of using those methods for the separation of gold
from pyrites. The report also discussed a chemical method then in use in London that
separated gold and silver from ores. Roasting of the pyrites after separation followed by
amalgamation was recommended but no comments were made about the use of the
chemical method.18

19
 Ralph W. Birrell

The process of amalgamation of gold and silver by mercury (quicksilver) had

been known for centuries. The Romans knew that gold would amalgamate with mercury
but it is not certain they used the process to separate gold from ores. Agricola described
the process of gold amalgamation in use in Europe in 1556 and it is likely that the
process was first used by the Spaniards to separate silver from its ores in Peru in 1566.19
The basic process as described by Agricola was to fine-grind the ore between two
rotating flat stones, similar to flour mills at that time, and wash the pulp into a series of
three barrels containing mercury. In each barrel the pulp and mercury were stirred by
paddles, so the gold and mercury amalgamated in a continuous process. The amalgam, a
lumpy grey mass, was later removed from the barrels, placed in a chamois, or fine
cotton bag, and squeezed so that most of the mercury oozed out, leaving the gold and
some mercury behind. He also described the collection of the gold by sluices, canvas
strakes and buddles but said the use of the amalgamating barrel was the most efficient
for collecting gold.20 It is now believed that amalgamation is a physical process rather
than a chemical one. The ‘catching’ of the gold is nearly instantaneous when the particle
of free gold, liberated from the gangue by the stampers, is wetted by the mercury and
because of its higher specific gravity sinks into the mercury.21
Just as the speakers at the meeting in 1893 had debated their preferred method of
amalgamation, so every millman treating gold ores in the 19th century had his own
preferred method, as he did on the best operating procedures for operating machinery to
separate the gold from the ore. In the 1850s, in Victoria and New South Wales, Berdan
pans, Chilean mills and stampers were tried for crushing. Berdan pans were circular
bowls with a heavy steel ball rolling in the bowl to crush the ore. Chilean mills
consisted of a heavy roller, rotated in a circular path in a shallow trough holding the ore.
In both machines mercury was added to amalgamate the gold freed by crushing in a
batch operation and this proved costly,22 for Chilean mills were slow, and the balls wore
rapidly in Berdan pans. By the end of the 1850s, the preferred method was to wet crush
with stampers in a continuous process, then run the slurry through a pool of mercury or
several pools in succession, on an inclined plane. The spongy amalgam was removed
from the pools with a small scoop, at regular intervals. The variation between stampers
was substantial as each millman sought to deal with different ores by empirical
methods. William Kelly operated a stamper mill at Bendigo in 1855 and experimented
with different types of deep wells after the mortar box so as to force the slurry to mix
with the mercury to assist amalgamation.23 Other millmen passed the slurry over copper

20
 The Extraction of Gold by Amalgamation and Chlorination

plates coated with mercury that caught the gold and any silver particles.24 Others
followed Agricola and used blankets or cow hides to catch the gold, washing the
blankets every few hours to collect the gold and sand particles that were then treated in
an amalgamation barrel.25 This was a small rotating barrel, up to four feet diameter by
six feet long [1.22m x 1.83m], usually made of cast iron, containing small iron rods or
balls for grinding. Blanket sand containing gold and pyrites, together with amalgam
from cleaning up mortars and plates, was placed in the barrel with mercury and rotated
for several hours until all the gold had amalgamated.26 The amalgam, which contained
30 to 45 per cent gold, was then removed and most mercury was extracted by squeezing
in a chamois bag. The remaining amalgam was then placed in an iron retort, heated to a
dull red heat so the remaining mercury distilled off, and was then recondensed and
collected.27 The remaining gold cake containing some mercury was sold to a bank
where it was refined. Some operators placed mercury in the mortar box for
amalgamation to take place but others criticised this practice. They argued that it
reduced the efficiency of the reduction of the ore and also reduced the mercury to very
small particles that were poor amalgamators. It also caused flouring of the mercury
which broke up into small droplets and was washed away with the water. The process
worked best with free milling ores with no pyrites but by the 1860s, when mining was
below the water line, the presence of pyrites at many mines caused problems with the
sickening and flouring of the mercury and the loss of fine gold distributed in the pyrites.
Amalgamation as an effective process of gold separation for free milling ores, after the
removal of the pyrites, persisted in Australia into the second-half of the 20th century.28
Because the process had been known for centuries it was never subjected to a scientific
analysis of its properties as was later done with sodium cyanide as a dissolver of gold.
Thus, millmen used empirical processes in determining the best way for its use - hence
the continuous debate throughout the 19th century.

The solution of the pyrites problem

In the period after 1851 most advances in milling technology proceeded in small steps
as managers and operators adjusted and improved equipment. For example, through
rotating the stamps by altering the cam shape and increasing the weight of individual
stamps. Occasionally, one mine would develop a change in technology that would have
a major effect in improving the efficiency of the gold extraction process. Such a

21
 Ralph W. Birrell

development occurred in the early 1860s at the Port Phillip and Colonial Gold Mining
Company’s mine at Clunes in Victoria.
By 1860 the quartz mines in Victoria and New South Wales were extracting ores
from below the water line. Stampers were being improved and in the free milling ores at
Bendigo, Ballarat, Castlemaine and Clunes much of the gold in the ore was separated
cleanly from the quartz gangue during the crushing. It was then collected by mercury
troughs, copper plates and canvas strakes that were washed to amalgamate the gold in
barrels. The Port Phillip Company was one of the few companies that regularly tested
their tailings for gold.29 In 1861, the works manager, Henry Thompson and their
chemist George Latta, realised the tailings contained six percent of the gold in the ore
and proved that most of this gold was in the pyrites but could be separated by fine
crushing after roasting the pyrites. The flow sheet was altered and a jaw crusher used to
reduce ore from the mine to a uniform size. This was fed to a mortar box, the pulp
exited from the screens and passed through mercury troughs for amalgamation, across
blanket strakes to collect the pyrites, through a final mercury trough to catch any
remaining free gold, and the tailings were then discharged into a creek. The initial
roasting of all ore was abandoned as uneconomical. The small amount of pyrites mixed
with sand from the blankets was treated in a Borlase buddle, modified by another
member of the staff, John Munday, who patented a new type to finally separate the
pyrites.30 The sand was discharged to the tailings and the pyrites, about one per cent of
the original ore, was roasted in a small reverberatory furnace. The arsenic oxide and
sulphur dioxide were removed from the gases, leaving the furnace by water sprays
before the gases were vented up the chimney. The oxidised pyrites was crushed finer by
Chilean mills and amalgamated in steam amalgamating barrels.31 The Company
completed these experiments and built a reverberatory furnace to roast the pyrites in
1865.32 Thomas Carpenter also experimented with pyritic ores at Bendigo in 1862 and
after crushing he used canvas blankets to separate the pyrites and then calcined the
pyrites in a reverberatory furnace before amalgamation.33 His experiments do not appear
to have been as comprehensive or influential as those of the Port Phillip Company.
Of particular significance in this development work at Clunes was the first use
of a jaw crusher in gold mining in Australia in 1862, the separation of pyrites by blanket
strakes and later by buddle, and the trials of copper plates and mercury troughs for
amalgamation.34 Fortunately for the company, its ores contained a small percentage of
pyrites but being largely iron sulphide this did not sicken the mercury after roasting. As

22
 The Extraction of Gold by Amalgamation and Chlorination

the pyrites contained up to four ounces of gold per ton, the recovery of this gold could
mean the difference between a profit or a loss for any company mining ores less than
five pennyweights per ton of ore overall.35 For this reason the experiments by the Port
Phillip Company that were completed and publicised by 1867, had been watched
closely by many other companies in Australia.36 The Superintendent, Rivett Bland,
welcomed inspection of the plant by other companies and published the results widely,
leading to the adoption of similar methods by other mines.37 Unfortunately, the results
did not solve the problems of mines such as St Arnaud where there were highly
refractory ores containing small amounts of silver, copper and lead, or those at
Costerfield containing antimony that sickened the mercury. These mines continued to
experiment with gravity methods of separating the components in their ores and with
smelting.38
The Port Phillip Company treated pyrites concentrates from other companies in
Victoria and New South Wales by amalgamation.39 In the early 1870s, millmen who
had worked at Clunes, and others, began to erect pyrites treatment plants on other gold
mining fields where they treated parcels of concentrate either for a fee or by purchase.
The mixing of the concentrates from various mines to give more efficient processing
was also adopted.40 In so doing, they were following procedures developed at Freiberg
in Saxony and at Swansea in Wales, where ores from all over the world were purchased
on the basis of assay and smelted to recover a wide range of minerals. Concentrates
from St Arnaud were sent to Freiberg for treatment until well into the 1880s and those
from Lucknow in New South Wales were sent to Swansea from 1880.41 In eastern
Australia plants were erected in most districts from 1871 for the contract treatment of
pyrites, and individual mines built pyrites plants for treating their own ores.42 The delay
between the successful operation of the process by the Port Phillip Company and its
general adoption by 1871 was probably because of water shortages caused by drought
after 1865, but can also be put down to the conservative attitudes of the Cornish mine
managers who were then becoming dominant in the local industry.43
Parallel with these developments in the treatment of pyrites during the 1860s and
1870s was continuous modification and improvement of stampers in California and
Australia, leading to comparisons being made between machine efficiencies on various
fields.44 The efficient operation of a battery depended on several variables: the type of
ore, the weight of the stamp, the rate of stamping, the distance the stamp fell, the size of
holes in the screen and the position of the screens relative to the dies. If the ore was free

23
 Ralph W. Birrell

milling and there was no pyrites in the ore, the time in the mortar box had to be kept
short or energy was wasted. If the screen holes were too large, the pulp passing through
the screen would have particles in which gold had not been separated from the quartz,
resulting in no amalgamation. If the screen holes were too small, the slurry particles
would remain in the mortar box and be crushed too fine, resulting in slimes from which
it was difficult to separate the fine gold which floated off in the current of water. Heavy
stamps gave rapid crushing but because of the inertia could not be raised too high
because power would be wasted. Heavy stamps required slow speed, while light stamps
allowed higher speeds. Despite the large number of variables, battery operators in
California and Victoria reached similar conclusions on best practice by 1871. In that
year, American writer, W. Raymond, presented comparisons of Australian and
Californian practice:

Table 1: Comparison of Californian and Victorian Battery performance c.1871.

Goldfield Wt. of Stamp lb. Drops per Min. Drop inches Tons
daily/H.P.
California
District
Tuolumne 500 40-80 6-10 1.26-2.2
Amador 600 70-79 9-11 1.07-1.9
Eldorado 300-650 65-80 7-10 1.24-3.93
Placer 600-800 60-75 7-12 1.24-1.53
Plumas 400-850 35-70 7-11 1.15-2.83
Shasta 300-600 60 6 1.83-2.75
Victoria
District
Ballarat 400-850 50-85 7-10 1.66
Beechworth 442-775 40-90 5-14 2.13
Bendigo 500-800 25-75 6-18 1.5
Maryborough 450-800 50-75 6-22 1.33
Castlemaine 450-800 35-75 6-15 1.7
Ararat 500-675 60-72 7.5-10 1.83
Gippsland 600-750 60-80 7-10 1.58
Pt. Phillip Co. 600-800 60-80 7-10 2.42-3.3

Source: R.W. Raymond, Mines. Mills and Furnaces of the Pacific States and Territories, J.B. Ford & Co.,
New York, 1871.

The final column can be considered a measure of the efficiency of crushing. As

weighing was not very accurate at most mines in the 1870s and there is no guarantee
that a ton measured in California was the same as in Victoria, the second decimal place

24
 The Extraction of Gold by Amalgamation and Chlorination

in this column can be ignored. The Shasta and Port Phillip results are for individual
mines while the others are averages for several mines. The figures show the averages of
efficiencies were equal if not better at Victorian mines and the results at the Port Phillip
Company were better than most. Later in the 19th century Rickard reported that at mines
in Gilpin County in Colorado, the stamps weighed 500 to 600 pounds [226.8 to
272.2kg], the drop was 16 to 20 inches [40.6 to 50.8cm] while the speed was an average
of 30 drops per minute. In California at that period the stamps weighed 750 to 1000
pounds [340.2 to 453.6kg], the drop 4 to 6 inches [10.2 to 15.2cm] and the average drop
rate 90 per minute. The difference between the fields was that the Colorado mines had
about 15 per cent pyrites in their ores and required fine crushing, while most Californian
and Victorian ores contained less than two per cent pyrites and coarse crushing was
sufficient to separate most of the gold.45 Unfortunately, comparative statistics for
Australian mines or mills treating high pyritic ores are not available.

Chlorination of gold ores

The use of water containing dissolved chlorine gas as a solvent to dissolve gold from
ores was first suggested by Dr John Percy of London as a result of experiments he made
in 1846. His proposal was developed into an industrial process by Professor Plattner and
his assistants at the Royal Freiberg Smelting Works in Saxony in 1848 to treat ores
containing iron oxides and arsenic from which most of the arsenic had been removed.
However, the residue contained about one ounce of gold to the ton, the separation of
which proved intractable. The gold was dissolved from the pyrites at ordinary air
pressure in earthenware pots, the liquid was run off into other earthenware jars and the
gold precipitated from solution by hydrogen sulphide gas. Plattner later suggested the
method would work on a greater scale using large wooden vats coated with pitch instead
of earthenware jars and his method was introduced to Califonia by C. Deetkin in 1857.46
In August 1864, Alan Cameron Lyster De Lacey, a civil engineer living in
Melbourne was granted Victorian Patent No. V 748 for ‘Improvements in machinery
and processes for pulverising gangue and the extraction of gold from auriferous
matters’. This was the eighth patent granted to De Lacy in Victoria since 1857, the
others being related to the preservation of timber, iron and steel and to crushing of ores.
An examination of patent records shows he was a prolific inventor, but few other
references to his career have been found. He may have been ahead of his time and the
foundries in Australia might not have been able to implement his ideas. Patent No. 748

25
 Ralph W. Birrell

related to the use of a solution of chlorine in water for extracting gold. Pressures above
atmospheric were obtained by generating excess chlorine, with the crushed pyrites being
placed in a barrel. This was dampened with water, chlorine gas injected and the barrel
rotated to speed up processing. The liquid was then removed from the barrel and the
gold precipitated by adding iron sulphide.47 This was not the first barrel process for
chlorination, for Duflos had used this method in Breslau in 1848.48 No information on
the use of De Lacy’s patent has survived but the patent documents show it was well
thought out and would probably have worked well, although it would have been
expensive as it was a batch process treating four tons of pyrites per batch.
De Lacy was living in Bendigo and had a mine there in 187149 and suggesting he
might have influenced local technology was the construction of a plant in 1876 by the
United Pyrites Company at Pinchgut Gully. In 1879, the Bendigo mining registrar
reported the company had replaced an amalgamation plant using arrastras (Chilean
mills) for fine grinding, by a chlorination plant with a four ton vat, earthenware gas
generators heated by steam and with a large number of earthenware vessels to hold the
auriferous solution - a close description of De Lacy’s patent.50 The 1879 report said the
company was treating sulphide ores from St Arnaud and Bourke’s Flat successfully and
could handle 40 tons per week.51 In his The Metallurgy of Gold, Rose said this patent
was tried and then forgotten.52 No documentation has been found to prove that this
installation at the United Pyrites Company used De Lacy’s patent but it seems likely it
did so. If true, it is a good example of the early efforts made to adapt an overseas
invention to local ore processing in Australia in the second-half of the 19th century.
In 1880, Victorian patent cover was granted to James Howell Mears, an
American inventor living in Philadelphia, for a process in which chlorine was used
above atmospheric pressure. This was introduced under a law that allowed overseas
inventors to be protected without a full submission if the patent was already approved in
the country of residence.53 De Lacy had covered the same process but his patent was
now void due to effluxion of time. James Cosmo Newbery and Claude Vautin were
awarded Victorian Patent No. 4484 in 1886 for a chlorination process which was similar
to those of De Lacy and Mears, except that the excess pressure of four atmospheres
above atmospheric was produced by compressed air.54 The solution was heated to speed
the process and charcoal was used as the precipitating medium.55 Newbery and Vautin
had worked closely with Joel Deeble and his staff at the United Pyrites Company plant
at Pinchgut Gully, Bendigo and this was apparently the first time wood charcoal was

26
 The Extraction of Gold by Amalgamation and Chlorination

used to precipitate the gold. The charcoal was then burnt at high temperature to recover
the gold.56 Newbery was Victorian Government Analyst, a graduate of Harvard
University in the USA who worked for Professor McCoy at the National Museum in
Melbourne, with the right to private practice, while Vautin was a metallurgist with
experience at the Cobar copper mines.57 Parcels of refractory gold ores from Australia
and other countries were treated at Bendigo and at Ballarat until the second decade of
the 20th century using this process.58 Chlorination plants were built at Daylesford,
Casillis, and Melbourne in Victoria, in the late 19th and early 20th century. Both Granya
and Bethanga treated very refractory gold ores containing copper.59 Although little
documentation remains, there is evidence that the Oswald mine and the South German
mine at Maldon, were operating chlorination plants up to World War I and there were
most likely others.60 A plant at Harden in New South Wales, the Carringar Chlorination
Works, advertised its ability to treat refractory ores by chlorination and in 1899 there
were two chlorination plants at West Wyalong, one at Wellington and one at Adelong.61
In Queensland, the Pyrites Works of Curtis and Company was established in
Gympie around 1880 using the Plattner process to treat pyritic ores from local mines.
Soon afterwards, the Aldershot Smelting Works commenced work in Maryborough to
treat auriferous concentrates and pyrites and to smelt lead ores containing gold.62 It is
uncertain whether this plant used chlorination. The Summary of Gold Mining Statistics
issued by the Queensland Mines Department in 1892 listed six chlorination vats at
Cloncurry, 49 at Charters Towers, 48 at Ravenswood and 132 at Rockhampton (Mt
Morgan). The Charters Towers field was discovered in 1871, the ore being found in
quartz veins in fault fissures in grano-diorite. The ore contained lead, copper and iron
minerals, the oxides of which caused no difficulties with the amalgamation process in
the early years. The flowsheet for the rich ‘brownstone’ deposits of the oxidised zone
was stampers, copper plates with mercury wells, Mundays buddle or Brown and
Stanfield concentrator, Wheeler’s pan (a modernised version of Agricola’s stone
grinders for fine grinding), and a tank to settle out any gold particles and amalgam. The
lighter particles were stirred and run off to a Berdan pan and fine ground with mercury.
When the water level in the mine was reached there was a tendency for the sulphides in
the ore to slime in the attempt to fine grind. A modified flow sheet was introduced by
D.A. Brown at the Charters Towers Pyrites Company. This saw a rockbreaker crush the
ore for the stampers, followed by the concentrator, Wheeler’s pan, settling tank, Berdan
pan, and a second settling tank to collect the pyrites. These were then roasted with salt

27
 Ralph W. Birrell

to oxidise the pyrites and to convert any zinc to zinc chloride in a three stage furnace
built on a hillside. The roasted sand was fed into wooden vats with a filter of broken
glass at the bottom, dampened with water and chlorine gas forced in from below. A
weak solution of chlorine gas in water was added. When the gold had dissolved, the
solution was run through a vertical column with counter current high-pressure steam to
drive off the chlorine gas for reuse. The gold was precipitated with ferrous sulphate onto
beds of charcoal and sawdust, which were burnt to smelt the residue. The method was a
technical success but it was an expensive plant and was superseded after 1892 by the
cyanide process.63
The largest chlorination plant in the world was erected at Mount Morgan to
recover the fine particles of gold in ironstone.64 In the oxidised zone, amalgamation
recovered only about 40 per cent of the gold, the rest being washed over the plates
without being caught. A Plattner process installed by A. Lymburner of Gympie was
tried in 1884 but did not recover more than 75 per cent of the gold. The Newbery-
Vautin patent was installed in 1885 and gave satisfactory results but was fragile for the
quantity of ore to be handled. It was partly discarded later that year when a plant
installed and designed by Henry Trenear, who had experience in Mexico, was installed
to give 98 per cent recovery. Some of the Newbery-Vautin components were
incorporated in the revised plant at the ‘Lower Works’, which was later enlarged to a
capacity of 500 tons per week. It was then realised that the orebody was larger than
anticipated and a new plant, the ‘Upper Works’, of 1000 tons per week was erected in
1887. After coarse crushing, drying and dry crushing between rolls, the fine ore was
roasted for three hours in a reverberatory furnace, cooled and fed to hardwood
chlorination barrels containing one ton of ore and mixed with water, calcium chloride
and sulphuric acid, to produce chlorine (the Mears Process). After being rotated for
three hours the barrels were discharged into leaching vats holding three tons where the
mixture was drained with vacuum assistance for 36 hours. The residue was shovelled
out while the gold chloride solution was passed through charcoal, which absorbed the
gold. The charcoal was burnt and the gold smelted with appropriate fluxes to produce
gold bullion. This plant was successful and the company paid substantial dividends but
there was continual pressure to reduce costs. After much experimentation, the manager,
Wesley Hall, and the chief metallurgist, George Richards, a graduate of the Ballarat
School of Mines, developed what was called the Hall-Richard process. This involved
the ore being crushed by orebreakers, dried in rotary furnaces (each 6ft 8inches diameter

28
 The Extraction of Gold by Amalgamation and Chlorination

by 30ft 5inches long [2.3 by 9.3m]), fine crushed in No. 5 Krupp ball mills. Following
this, the crushed material was passed through wire screens with 400 perforations to the
square inch, was roasted in rotary furnaces, cooled and carried by small rail trucks to
vats each containing 100 tons of ore. The ore in the vat was saturated with water and a
solution of 80 grams of chlorine in a gallon of water was run in to the top and allowed
to percolate through the ore to convert the gold to gold chloride, the solution being run
into cement lined brick tanks from where it was passed over charcoal beds to precipitate
the gold. The charcoal was removed periodically, burnt in a reverberatory furnace and
then reduced in a smelting furnace to gold bullion. Mixing sodium chloride produced
the chlorine, which was dissolved when manganese dioxide and sulphuric acid were
passed up a scrubbing tower down which water was percolating. The total cost of
treating the ore to extract the gold was just over 10 shillings per ton in 1911, including
crushing, roasting and chlorination. This final plant, the West Works, with an initial
capacity of 2,500 tons per month, was built progressively in four similar parallel
sections in 1896-7 to a capacity of 10,000 tons per month. It was closed in 1911 when
the ore for which it had been built was worked out.65
As the returns were overshadowed by the cyanide process after 1900, the
chlorination process has been mostly ignored by mining historians but the use of the
process was important in the development of gold mining technology in eastern
Australia from 1876 when the process was first used on pyritic ores from Bendigo.
Although the tonnages of pyrites treated and the gold recovered were not as large as that
recovered by amalgamation, the revenue from chlorination was additional income not
otherwise available to a company mining refractory ores. Often it was the difference
between paying a dividend, or struggling to survive. As an example, the Lord Nelson
Mine at St Arnaud began paying dividends in 1888, one year after Zebina Lane (Snr)
was appointed manager. The company had previously sent pyrites concentrates to
Freiberg at a profit but using Wheeler’s pans, Lane introduced fine grinding followed by
amalgamation.66 By 1890 he was sending the residue from this process to Bendigo for
further treatment by chlorination.67 Early statistics of the mine production are not
available but the Mines Department published some figures of the mine output between
1905 and 1914, the year the mine closed. These figures for 1905-14 have been collated
as 75,203oz by amalgamation, 13,549oz by chlorination, 20,186oz by cyaniding old
tailing dumps, and in 1905 a figure of 572oz by pan amalgamation of pyrites was given.
Apparently, pan amalgamation was abandoned soon after 1905 and all pyrites treated by

29
 Ralph W. Birrell

chlorination at the mine.68 The percentage of pyrites in the ore was up to 20 per cent69
and the figures show that 18 per cent of mine production in this period was produced by
the chlorination process. In 1897, the South German mine at Maldon was also operating
a chlorination plant for pyrites and recovering 97 percent of the gold and additionally
used a cyanide plant to extract gold from tailings. Charcoal was used to precipitate the
gold in both cases.70 Figures printed in 1900 state that nearly 450,000oz of gold had
been obtained from pyrites in Victoria up to that date. However, this figure apparently
included gold from fine crushing, plus amalgamation, as well as chlorination and
cyaniding. It also probably included pyrites sent from elsewhere in Australia and
overseas for treatment at Ballarat and Bendigo.71 Between 1884 and 1911, the major use
of the process in eastern Australia was at Mt Morgan where it was used to separate fine
gold not recoverable by amalgamation. During that period, the peak production there in
one year was 300,000oz of gold.72
While in eastern Australia from 1876 there were a limited number of
chlorination plants compared with the number of cyanide plants after 1900, the
experience gained by metallurgists and mine managers who worked with the former
process would have given them an understanding of how chemistry could be relevant to
the efficient extraction of gold from complex refractory ores.
As early as 1864, Dr Ottway, an American chemist, was experimenting with
lixiviation to separate silver from gold in the refractive St Arnaud ores and he later
experimented with the Plattner process on Blackwood ores.73 In the same year De Lacy
patented a variation of the Plattner process.74 By 1876, as discussed above, plants
established in Bendigo and Ballarat to separate gold from pyrites by fine crushing and
amalgamation indicate that this method failed to extract half of the gold. The Plattner
process proved not very successful on the more complex ores. In the mid 1880s further
experiments at Bendigo with the Newbery-Vautin patented variation of the Plattner
process were more successful but the method was costly. This process was tried in 1885
at Mt Morgan on oxidised ironstone ore in which the gold was finely distributed, but the
equipment could not process the amounts of ore to be treated and failed mechanically.
Some of the equipment was incorporated in a modified barrel leaching process in 1887
and this was successful.
From about 1890 chlorination plants were distributed in strategic locations on
the goldfields of eastern Australia to treat refractory ores. As the amount of pyrites in
most ores was of the order of a few percent of the ore crushed, the number of plants did

30
 The Extraction of Gold by Amalgamation and Chlorination

not need to be large to treat all the pyrites available. By the end of the 19th century some
mines with very refractive ores were installing chlorination plants. The process replaced
the system of fine crushing the pyrites followed by amalgamation, and sending the
residue still containing gold to the pyrites plants for further treatment. At the same time,
they installed cyanide plants to treat old oxidised tailings, and later fresh tailings were
treated very cheaply by the newer method. On some fields, such as Charters Towers,
cyaniding had replaced chlorination by 1897 because the ore was suitable for the
cyanide process. However, very refractory ores found elsewhere were still being treated
by chlorination until World War I.
Additionally, by 1890, most mines with refractory ores were crushing the pyrites
fine enough to be experiencing a sliming problem and were experimenting with very
sophisticated mechanical methods such as vanners to separate the pyrites and the
contained gold from the quartz. These slimes were so fine that they ‘packed’ in the vats
used for chlorination and cyaniding. In order to get the chemicals to attack and dissolve
the slimes to recover the gold, Joel Deeble experimented in Bendigo with an agitator to
stir the slimes in the vat. He later sold some nine of these plants to mines in the
colonies, including two to Kalgoorlie and two to Maldon where they were fine crushing
their ores for cyanide plants, although it is likely he first developed the plants for
chlorination.75
Todd concedes that colonial Australia showed an ability to change course, to
cross the divide between old and new trajectories and to transfer and adapt
technologies.76 Although there were undoubtedly some mine managers and others who
were ignorant of the chemical methods, after 1890 it is unlikely they were the majority.
In fact, chlorination and cyaniding were adapted in parallel, but with chlorination plants
being somewhat earlier and distributed widely enough to educate mine managers and
their staffs in chemical methods of ore treatment. In the first quarter of the 20th century,
some mines with refractory ores operated both chlorination and cyanide plants - the
chlorination plant to extract fine gold from pyrites and the cyanide plant to extract fine
gold from tailings. Ignorance was not the reason the cyanide process was adopted
slowly, more important was the complexity of many gold ores, the cost of the royalty,
the uncertainty of the validity of the cyanide patents, the 1890s depression in Victoria
that saw a squeeze on capital for new developments, and the need for time to adapt the
cyanide process to complex ores. The sponsors of the cyanide process had to prove that
their system was the cheaper, something they did not always succeed in doing.77

31
 Ralph W. Birrell

Endnotes
1
R.M. Serjeant, 'Report of the Board to Investigate Methods of Treating Pyrites', Parliament of Victoria,
Papers Presented to Parliament 1874 [hereafter PPP Vic 1874], no. 96, Government Printer, Melbourne,
1874, p. 62. Appendix E includes the following data for gold recovered by simple amalgamation after
crushing: Llanberris Co., Ballarat 93 per cent; Port Phillip Co., Clunes, 96 per cent; Walhalla Co.,
Walhalla, 97 per cent; Caledonian Co., Maldon, 81 per cent. At each mine the remaining percentage was
recovered after treating the pyrites. Losses in the tailings are not included but would have been about one
per cent or more.
2
Ibid., p. 13.
3
Exceptions are: D. Menghetti, ‘Extraction Practices and technology on the Charters Towers Goldfield’,
North Australia Research Bulletin, September 1982; J. Kerr, Mt Morgan, Gold, Copper and Oil, J.D. Kerr
and R.S. Kerr, 1982, pp. 55-58.
4
J. Todd, Colonial technology: Science and the transfer of innovation to Australia, Cambridge UP, Hong
Kong, Melbourne, 1995, p. 206.
5
G. Agricola, De Re Metallica, 1556, translated by H.C. Hoover and L.H. Hoover, New York, Dover,
1950, pp. 118-20, 273. Agricola described how firesetting was used to heat hard rock underground to
soften it. This firesetting technique was gradually superseded after gunpowder was introduced at
Schemnitz in 1627 but it was still in use in English mines until about 1700. Agricola also discussed how
burning ores made them easier to crush with stampers.
6
Refractory ores were found at Maldon, St Arnaud, Bethanga and Granya in Victoria and at Harden,
Lucknow and West Wyalong in New South Wales and at Ravenswood and Charters Towers in Qld.
7
Mr. Wilkinson, 'Lecture on the Extracting Gold from Quartz,' Colonial Mining Journal, July 1860, pp.
175-6. This furnace was an early adaptation of science to gold mining technology.
8
J. Woodland, Sixteen Tons of Clunes Gold, Clunes Museum, Clunes, 2001, pp. 15, 71.
9
The Mining Journal (London), 11 September 1858.
10
J. Birmingham, I. Jack, and J. Jeans, Industrial Archaeology in Australia - Rural Industry, Heinemann
Publishers Australia, Richmond, 1983, p. 74.
11
The Mining Journal, 11 September 1858.
12
Woodland, Sixteen Tons of Clunes Gold, pp. 68, 77.
13
‘Reports of Mining Surveyors and Surveyors, Victoria', Mines Department, Melbourne, 1869, p. 33.
14
Serjeant, Report of the Board to Investigate Methods of Treating Pyrites, p. 59. It is odd that Stawell
witnesses, members of the local Council, doctors and mining men said the percentage of pyrites in
Stawell ores was very small and the fumes from roasting did not affect health, while recent geological
studies state that the Cross reef, which was being worked in 1874 was sulphide rich. See D.C.
Fredericksen and M. Gane, 'Stawell Gold Deposits,' in D.A. Berkman (ed.), Geology of Australian & New
Guinean Mineral Deposits, Australasian Institute of Mining and Mineralogy, Melbourne, 1998, p. 538.
Perhaps the locals were trying to protect the reputation of their town. Witnesses from Castlemaine were
much more forthright about the detrimental effects of the fumes.
15
Report of the Royal Commission on Mining, Government Printer, Melbourne, 1891, para. 9539.
16
C.G.W. Lock, 'Gold Amalgamation,' Transactions of the Institute of Mining and Metallurgy vol. 1,
1893, pp. 205-37. Lock gave examples of practice in the Transvaal, Australia and the United States.
17
The Board was set up on 28 June 1873 and submitted its report the following year. PPP Vic 1874, report
no. 96. The membership was; four mine managers, the Government Analyst J. Cosmo Newbery, a
geologist, and an industrial chemist.
18
A written questionnaire was sent to each person called to give evidence and he was then questioned
about his replies and other matters. The replies given in the record of the evidence show there was just as
much difference in the 1870s about amalgamation as there was in the 1890s. Each manager or millman
had a preferred method for treating the ore with which he was working.
19
Agricola, De Re Metallica, pp. 293-99. See also the translators’ note pp. 297-300 on the history of
amalgamation.
20
Ibid., p. 298.
21
T.T. Read, 'The Amalgamation of Gold-Ores,' Transactions of the American Institute of Mining
Engineers XXXVII, 1906, p. 70. The specific gravity of gold is 19.32, that of mercury 13.55.
22
Woodland, Sixteen Tons of Clunes Gold, pp. 8-9.
23
W. Kelly, Life in Victoria, London, 1857, pp. 195-231.
24
Read, 'Amalgamation of Gold-Ores,' p. 57. Read believed copper plates were developed in California
but were not patented and the inventor is unknown.
25
Woodland, Sixteen Tons of Clunes Gold, p. 66.

32
 The Extraction of Gold by Amalgamation and Chlorination

26
A. F. Taggart, Handbook of Mineral Dressing, John Wiley and Sons, New York, 1945, p. 149.
27
T. K. Rose, The Metallurgy of Gold, Charles Griffin & Co., London, 1906, pp. 140-41.
28
Australian Development Ltd., 'Amalgamation', in J.T. Woodcock (ed.), Mining and Metallurgical
Practices in Australasia, Australasian Institute of Mining and Metallurgy, Melbourne 1980, pp. 484, 88,
517. The carbon in pulp process replaced amalgamation in the 1980s.
29
Woodland, Sixteen Tons of Clunes Gold, p. 71.
30
Ibid., p. 74.
31
Ibid. Woodland gives a comprehensive description of the work done by the Port Phillip Company in the
development of improved crushing techniques and the treatment of pyrites of which only a brief outline is
given here. Serjeant, 'Report of the Board to Investigate Methods of Treating Pyrites'. In the report the
Board recommended the use of the modified Borlase buddle, of Cornish origin, to separate the pyrites
from the sand.
32
Woodland, Sixteen Tons of Clunes Gold, p. 73.
33
Catherine Reef United Company, ‘Mine Manager’s Reports’, 25 September, 9 October, 16 October, 11
December 1862. I am grateful to James Lerk for access to his files on mining in Bendigo.
34
Woodland, Sixteen Tons of Clunes Gold, pp. 67, 73.
35
In each 1 ton of gold ore of average grade of 5 dwt overall and containing 1 per cent pyrites there are
0.99 tons of ore with free gold which can be amalgamated and 0.01 tons of pyrites with attached gold
which must be fine ground. The 0.01 ton of pyrites contains 4/100 ounces of gold = 80/100 dwts = 0.8
dwts. The free gold is therefore 5 - 0.8 = 4.2 dwts. When assumed all the free gold is amalgamated, the
pyrites gold is 0.8/4.2 = 19 per cent of the free gold. By extracting the extra 19 per cent in the pyrites, a
mine which had extracted the free gold and barely made a profit, could turn into a good profit maker.
36
Ibid., ch. 14.
37
In 1868, several mine managers from Bendigo inspected the Port Phillip plant and soon afterwards
requested the permission of the Sandhurst City Council to erect treatment plants to roast pyrites. See
Bendigo Advertiser, 6 June 1868. Several people, including medical men, objected that the fumes from
such plants were poisonous.
38
Report of the Royal Commission on Mining, 1891, pp. 460-65. Many managers and owners, including
Fred Stahl and John Rew, managers of St Arnaud mines, who appeared before this body, reported on
experiments with vanners and other devices being tested.
39
Woodland, Sixteen Tons of Clunes Gold, p. 76. Pyrites from the Amelia reef in the Blue Mountains of
New South Wales were sent to the Port Phillip Company for treatment in 1867.
40
These plants were erected outside municipal boundaries on leases granted by the Mines Department for
this purpose. People living near these plants complained to the Board on Pyrites about the noxious fumes
but several medical men said the fumes were not dangerous.
41
Report of the Royal Commission on Mining, 1891, p. 460. ‘The Lucknow Goldfield, Report No. 30',
New South Wales Geological Survey, Sydney, 1920, evidence by Fred Stahl and L.F. Harper.
42
Victorian Mines Department Reports and Statistics 1859 - 1893, Victorian Mines Department,
Melbourne, 1859. These reports contained statistics on the tonnage of pyrites treated in each mining
district from 1869. In most cases the returns were between two and five ounces of gold to the ton of
pyrites. In the first quarter of 1869 the Port Phillip Company produced 87 tons and the New North Clunes
Company 35 tons of pyrites respectively, with other districts producing negligible amounts. In the first
quarter of 1872 the Ballarat mining district produced 414 tons mainly from Clunes, the Beechworth
district 96 tons, the Bendigo district 396 tons, Maryborough district 108 tons and the Gippsland district
137 tons, respectively. In later years Bendigo was usually the largest producer and processor of pyrites.
43
F. Cusack, Bendigo a History, Heinemann, Melbourne, 1973, p. 135.
44
W.P. Blake, Mining Machinery, Charles Clasfield & Co., New Haven, 1871. Blake included excerpts
from the Reports of Registrars and Surveyors of the Victorian Mines Department.
45
T. A. Rickard, Stamp Milling of Gold Ores, Engineering and Mining Journal, New York, 1906, p. 2.
46
Rose, Metallurgy of Gold, ch. XIII.
47
A.C.L. De Lacy, 'Improvements in Machinery and Processes for Pulverising Gangue and the Extraction
of Gold', in Victorian Patent V748, Victoria, 1864. The use of a rotating barrel appears to be a transfer
from the amalgamation process.
48
Rose, The Metallurgy of Gold, p. 257.
49
Register of Claims, Victorian Mines Department. Claim no. 3976 for a residence licence associated
with the Plantagent Gold Mining Company mine at One Tree Hill, Bendigo.
50
W.B. Kimberley, 'Thomas Edwards', in Bendigo and Vicinity, Ballarat, 1895, pp. 246-7. Edwards had
commenced roasting pyrites at Clunes in 1864. He probably learnt these skills at the Port Phillip works.
51
'Reports of Mining Surveyors and Surveyors, Victoria', Second Quarter 1879, Sandhurst Division.

33
 Ralph W. Birrell

52
Rose, Metallurgy of Gold, p. 257.
53
See Victorian patent application no. 2814 by E. Waters, Patent Attorney, 14 April 1880.
54
D. Clark, Australian Mining and Metallurgy, Critchley Parker, Melbourne, 1904, p. 269. Rickard,
Stamp Milling of Gold Ores, p. 200. Clark said the patent was not sound scientifically as the amount of
chlorine dissolved in the water was not increased by increasing the total pressure, but only by increasing
the partial pressure of the chlorine gas. The patents of De Lacy and Mears did this. I am thankful to an
anonymous referee for the information that Vautin later had a chequered career in England and South
Africa and was imprisoned for six months for fraud in the latter country in 1899.
55
See Victorian Patent Application, no. V4484, 29 March 1886.
56
Joel Deeble had been a partner with Thomas Edwards in establishing the Pinchgut Gully plant in 1872
and later bought out his partner. He was the expert in metallurgy while Edwards concentrated on the
furnace side of the business and in running his plant at Ballarat. In the 1890s Edwards built a new plant,
near the Pinchgut plant and started chlorinating ores in opposition to Deeble who up to that time had had
a monopoly in Bendigo. Unlike Edwards, Deeble was not a self publicist and there is little documentation
on his development work.
57
Biographical Sketches, Bulletin no. 23, Victorian Mines Department, 1910. W. Meyerriecks, Drills and
Mills, Will Meyerriecks, Tampa, 2001, p. 162, states that the Newbery Vautin process was licensed for
use in the United States.
58
See Supplement to the Weekly Times, Melbourne, 10 September 1898, p. 57.
59
Reports to the Minister of Mines, Victoria, Mines Department Victoria, Fourth Quarter, 1890, 1903.
Clark, Australian Mining and Metallurgy, pp. 308-9, 316, 335.
60
The Oswald Gold Mines NL at Maldon was operating a chlorination plant in 1913 on an intermittent
basis to treat its pyrites (see Oswald Mine, Managers Reports, copy with author), and it is likely the Lord
Nelson Company at St Arnaud had a plant working at the same time as it regularly reported on the
amount of gold produced by chlorination. Clark, Australian Mining and Metallurgy, p. 327.
61
Australian Mining Standard, 24 April 1889. Reports to Minister of Mines, New South Wales,
Government Printer, Sydney, 1899.
62
W. Lees, The Goldfields of Queensland, 1899, Reproduced by the Mines Department, Brisbane 1986.
63
D. Menghetti, 'Extraction Practices and Technology on the Charters Towers Goldfield', North Australia
Research Bulletin, September 1982.
64
Rose, Metallurgy of Gold, p. 281.
65
J. Kerr, Mt Morgan, Gold, Copper and Oil, J.D. & R.S. Kerr, Brisbane, 1982, pp. 55-58. R.F. Boyle,
R.W. White, and O.A. Wilson, 'Mount Morgan, Queensland - Marvellous Mountain', paper presented at
the AIMM Conference, 1993, pp. 330-31. C.H. Humphreys, 'West Works - Mount Morgan (Queensland)
Chlorination', Proceedings of the Australasian Institute of Mining Engineers, vol. XV no. 2, 1911, pp.
363-77.
66
Y.S. Palmer, Track of the Years, MUP, Melbourne, 1955, p. 263.
67
Report of the Royal Commission on Mining, 1891, p. 455.
68
Reports to the Minister of Mines Victoria, 1904 - 1914.
69
Report of the Royal Commission on Mining, 1891, p. 456.
70
J. Mactear, 'Notes on the South German Mine, Maldon, Victoria', Transactions of the Australasian
Institute of Mining and Metallurgy, 1897, pp. 43-51.
71
Reports to the Minister of Mines, Victoria, 1900, p. 10.
72
Kerr, Mt Morgan, Gold, Copper and Oil, p. 62.
73
G.H.F Ulrich, 'Gold and Silver Bearing Reefs of St. Arnaud', Victorian Geological Survey, Melbourne,
1864, p. 3. Report of the Royal Commission on Mining, 1891, para. 4619, p. 192.
74
De Lacy, ‘Improvements in Machinery and Processes for Pulverising Gangue’.
75
Todd, Colonial Technology, p. 177. The Australian Mining Standard, 1 July 1897, p. 994, 5 August
1897, p. 2091, 1 June 1899, p. 120.
76
Todd, Colonial Technology, p. 207.
77
Ibid., p. 133.

34