The mechanochemical reduction of AgCl with metals : RRRRevisiting an experiment of M. Faraday

Article abstract of DOI:10.1007/s10973-007-8479-8

Faraday induced the mechanochemical reduction of AgCl with Zn, Sn, Fe and Cu in 1820, using trituration in a mortar. This experiment is revisited, employing a mortar-and-pestle and a ball mill as mechanochemical reactors. The reaction kinetics depends both on the thermochemical properties and the hardness of the reactants. When using Zn as the reducing agent, Faraday likely observed a mechanically induced self-sustaining process (MSR), or at least he came very close to doing so.

Full text of DOI:10.1007/s10973-007-8479-8

Journal of Thermal Analysis and Calorimetry, Vol. 90 (2007) 1, 81–84
THE MECHANOCHEMICAL REDUCTION OF AgCl WITH METALS
Revisiting an experiment of M. Faraday
L. Takacs*
University of Maryland, Baltimore County, Department of Physics, Baltimore, MD 21250, USA
Faraday induced the mechanochemical reduction of AgCl with Zn, Sn, Fe and Cu in 1820, using trituration in a mortar. This
experiment is revisited, employing a mortar-and-pestle and a ball mill as mechanochemical reactors. The reaction kinetics depends
both on the thermochemical properties and the hardness of the reactants. When using Zn as the reducing agent, Faraday likely
observed a mechanically induced self-sustaining process (MSR), or at least he came very close to doing so.
Keywords: Faraday, mechanochemistry, self-sustaining reactions
Faraday’s contribution to mechanochemistry
reduction by ‘nascent’ hydrogen. In order to do so, he
dispersed silver chloride in dilute acids and generated
hydrogen in situ by adding zinc. Hydrogen was
liberated and some metallic silver was obtained, but
Faraday was not convinced about the role of
hydrogen. First he repeated the process using water as
the dispersing medium. The reduction of silver
chloride still occurred, although no obvious genera-
tion of hydrogen could be noticed. Then he directly
mixed silver chloride with zinc, without any source of
hydrogen. When the mixture was heated, a violently
exothermic reaction took place and silver was
obtained. That was a direct solid state reaction;
hydrogen could not play any role. Finally, Faraday
turned to mechanochemistry. As he reported:
It is customary to associate the emergence of
mechanochemistry with the papers of Carey Lea,
published at the end of the 19th century. He was the
first scientist to clearly point out that heat and
mechanical action can induce different reactions in
the same system [1]. As mentioned by Prof. Boldyrev
recently [2], Michael Faraday was aware of some
effects of mechanical activation much earlier. Indeed,
in his book published in 1827 on ‘Chemical Manipu-
lations’, a twenty-page chapter is dedicated to mortars
and comminution. It begins with stating, that ‘the
division of matter is often highly advantageous in
facilitating chemical action’ [3]. Faraday’s book ends
with 318 practical exercises; the 58th of them reads:
‘Make an intimate mixture of equal parts by mass of
muriate of ammonia and quick lime; observe the
ammonia evolved even at common tempera-
tures…’ [4]. He refers back to the chapter on tritura-
tion for the way to make the ‘intimate mixture’. In
modern terms, he asks for affecting a mechano-
chemical reaction between NH4Cl and CaO.
A very direct reference to a mechanochemical
process appears in a paper written in 1820 on the
decomposition of silver chloride [5]. He signed the
paper as ‘M. Faraday, Chemical Assistant’. Indeed,
that was his official title at the time. He was only 29,
working at the Royal Institution of London, mainly
under the supervision of Sir Humphrey Davy [6]. This
is not Faraday’s first paper; he began publishing
already in 1816 [7]. But this was the first time he was
dealing with the mechanism of a reaction. He wanted
to decide, whether hydrogen was capable of reducing
silver chloride. He found that a stream of hydrogen
gas had no such effect. Then he attempted to cause the
‘… if dry chloride of silver in powder be
triturated in a mortar with zinc filings, the two bodies
immediately act, and a heat above that of boiling
water is produced.
Zinc is not the only common metal which thus
rapidly decomposes chloride of silver, in the dry way.
Tin acts even more powerfully when triturated with it:
and copper and iron have both of them affinities for
chlorine strong enough to produce the same effect.’
This passage clearly describes a mechano-
chemical process, dated well before the better-known
papers of Carey Lea from the 1890’s [8]. Moreover,
Faraday mentions the ‘dry way’ of causing a reaction
very casually, suggesting that the method was quite
familiar to him. He does not emphasize it as a new
discovery. Instead, he continues the paper with the
interpretation of the results, stating that metals can
reduce silver chloride directly, without the presence
of hydrogen as an intermediary. He leaves open the
question whether direct reduction by hydrogen is
possible.
*
takacs@umbc.edu
1388–6150/$20.00
Akadémiai Kiadó, Budapest, Hungary
Springer, Dordrecht, The Netherlands
© 2007 Akadémiai Kiadó, Budapest

TAKACS
Results and discussion
It seems that this form of mechanochemistry was
common knowledge to him and most likely others
before him. Chemistry – and alchemy before-
hand – used heat as the primary means of stimulating
chemical reactions. But grinding and mixing in a
mortar is often part of the preparation. It is very likely
that chemical changes were also caused in the mortar,
intentionally or perhaps unnoticed. The decomp-
osition of cinnabar by rubbing was already known to
the ancient Greeks [9]. Whether other written records
of mechanochemical processes exist in the early
literature is yet to be seen.
According to Faraday’s description, the reaction
between silver chloride and zinc is fast and highly
exothermic, raising the possibility that he actually
observed a mechanically induced self-sustaining
reaction (MSR) [10]. As the mechanochemical
reduction of silver halides has not been studied since,
the reinvestigation of the reactions mentioned by
Faraday is of interest for more than historical curiosity.
In order to determine whether the ‘powerful’
reactions observed by Faraday could be mechanically
induced self-sustaining reactions (MSR), the
thermochemical parameters have to be consulted first.
Table 1 summarizes the reaction equations, the
reaction heat associated with the exchange of one
mole of chlorine atoms, and the simplified adiabatic
temperature, i.e. the ratio of the reaction heat to the
room temperature heat capacity of the products,
DH/C. This quantity correlates well with the true
adiabatic temperature [11] and is often used to predict
whether MSR is possible in a given system. A value
close to or above 2000 K is needed to support MSR in
oxide-metal thermite systems and simple combination
reactions [10]. In order to obtain the most exothermic
reactions possible, the reaction stoichiometry
providing the highest value of DH/C was chosen in
the cases when the reducing metal has multiple
oxidation numbers. Al was added to include a
reaction where MSR can clearly be anticipated.
Experimental
Figure 1 presents the temperature of the milling
vial during the reduction of AgCl with Zn in a ball mill.
The sudden temperature increase after about 90 s of
milling suggests MSR, in spite of the relatively low
value of DH/C. This is not entirely unexpected, as
MSR was observed during the reduction of Ta5Cl by
Mg before, although the adiabatic temperature of that
Nowadays the preferred equipment employed to
cause mechanochemical reactions is the ball mill.
A SPEX 8000 Mixer Mill with round-ended hardened
steel vial and steel balls was used in these
experiments. The typical charge was 6 g of AgCl-
metal mixture, containing 20% more metal than
required by stoichiometry, in order to compensate for
partial oxidation of the metals. Two 12.7-mm and
four 9.5-mm balls were used. The milling vial was
filled in air, as Faraday did his experiments without
the protection of an inert atmosphere as well.
The temperature of the milling vial was recorded
using a thermocouple, in order to see, whether a
self-sustaining reaction occurred, and if it did, to
determine the ignition time.
Table 1 The equation, reaction heat per exchanged chlorine,
DH/n, and simplified adiabatic temperature, DH/C of
the investigated reactions
Equation
DH/n/kJ mol–1
DH/C/K
2AgCl+Zn®ZnCl2+2Ag
2AgCl+Sn®SnCl2+2Ag
2AgCl+Fe®FeCl2+2Ag
AgCl+Cu®CuCl+Ag
80.6
38.5
43.9
10.2
108.2
63.1
1310
607
688
137
3AgCl+Al®AlCl3+3Ag
3AgCl+Cr®CrCl3+3Ag
1933
1124
Following Faraday, the reduction of AgCl with
Zn, Sn, Fe and Cu powders was investigated. Ball
milling was also performed with Al and Cr for
comparison. The purity of the starting materials was
98% or better.
The reactions were also induced by trituration in
a small porcelain mortar. The temperature was
measured with a thermocouple taped to the pestle
close to its active surface. While the information
obtained this way is semi-quantitative at best, it still
provides useful clues about the reaction kinetics.
The phase composition of the samples was
determined using an X’Pert (Philips) X-ray
diffractometer equipped with a Cu tube. The fraction
of Ag in the chloride and metal phases was deter-
mined by comparing the line intensities to those of a
standard with known phase composition.
Fig. 1 Temperature of the milling vial during the reaction of
AgCl with Zn in a ball mill
82
J. Therm. Anal. Cal., 90, 2007

MECHANOCHEMICAL REDUCTION OF AgCl WITH METALS
reaction is also quite low [12]. Closer inspection shows
with Zn or Sn, in spite of Faraday’s claim to the
contrary.
Local melting may also contribute to the high
that although the reaction starts very early, it is more
sluggish than a typical MSR; the temperature ‘jump’
takes 15–20 s, while it happens within 2 s in oxide-
metal and many other MSR systems [10]. Although the
reaction clearly accelerates due to self-heating, it does
not propagate across the entire charge easily. Stopping
the mill at the beginning of the process would probably
arrest the reaction. This behavior may relate to the
softness of AgCl, that makes it easy to achieve close
contact between the reactants. As a result, local
reactions take place due to shearing early, before
sufficient mixing could take place to propagate the
reaction through the charge. According to X-ray
diffraction (XRD), the reduction of AgCl is practically
complete after 20 min of milling (Fig. 2); about 3% of
the Ag remains in the chloride phase, probably due to
partial oxidation of the Zn powder. As anticipated, the
reduction of AgCl with Al also takes place in the
self-sustaining way and it results in practically full
transformation in a short time.
No temperature jump is observed when Sn is
used as the reducing metal, but the reaction is
practically complete after 20 min of milling as shown
by the XRD pattern (Fig. 2). If the reaction of AgCl
appeared more ‘powerful’ with Sn than with Zn to
Faraday, it was not because of the total amount of
evolved heat. What distinguishes Sn is its softness
that increases the efficiency of trituration.
The importance of hardness is clearly brought out
by an attempt to reduce AgCl with Fe. Even after 60 min
of ball milling, only about 12% of the silver was in the
metallic state (Fig. 2), although the reaction is about as
exothermic as the reaction with Sn. Although Cr is
much more reactive than Fe, it is also harder.
Consequently, 60 min of milling with Cr reduced only
13% of silver in AgCl, essentially the same as Fe.
Copper is less reactive but softer; as a result, ball milling
with copper reduced 88% of the silver in 60 min and
already 28% in 20 min. In any case, the reaction of AgCl
with Fe or Cu is far from being as intense as the reaction
reactivity between AgCl and Zn. The DH/C value for
their reaction is 1310 K, much higher than the melting
points of AgCl (728 K) and Zn (693 K) and the
product, ZnCl2 (591 K). The reaction between AgCl
and Al is also exothermic enough to melt the
reactants. The other investigated reactions are less
exothermic or, in the case of Cr, the melting temper-
ature is very high, thus melting is not possible. The
fast rate of the reaction between Ta5Cl and Mg was
also explained by local melting [12].
It would be interesting to compare the above
results with data from the literature. However, the only
long-term program on MSR and related reactions of
halides is that of the group of McCormick [12–14]. The
author of this paper is not aware of any recent study on
the mechanochemical reactions of Ag halides. In
general, halides are not convenient model systems, as
they are often hygroscopic and consequently difficult to
work with. On the other hand, the fact that many halides
dissolve in water or alcohol easily provides an oppor-
tunity to prepare metal [13, 15] or oxide [14] nano-
powders using halide precursors.
Ball milling is more suitable for obtaining
reproducible results than manual trituration in a mortar,
but the mechanical actions are different. Therefore, it
has to be verified whether conclusions drawn from ball
milling experiments are also applicable to experiments
employing hand trituration. Figure 3 shows the temper-
ature measured close to the active surface of the pestle
during the processing of 3-g batches of powder mixture.
The powder was rubbed vigorously for 60 s, followed
by 60 s rest, repeated three times. When AgCl was
rubbed with Fe, the temperature increased only slightly
and quite smoothly. Much of the increase came from the
warmth of the hand; the little undulation is probably due
to the heat evolved by friction. According to XRD, only
about 2% of the silver could be reduced even when the
trituration was continued for 15 min. Cu gave somewhat
better results, about 7% transformation in 15 min, con-
sistent with the ball milling experiments. Clear heating
and cooling periods could be observed when Sn was
used, showing the effect of the heat generated by the
reaction. 40% transformation was achieved by 3 min of
trituration. The temperature change is larger with Zn
due to the larger reaction heat. The transformation rate is
much higher as well, more than 90% of the reaction is
complete after 3 min of working. Yet, the fast temper-
ature increase stops as soon as the trituration is paused
and re-starts only when the rubbing starts again. Under
our conditions, the reaction is fast, but requires constant
mechanical agitation; it does not become self-
propagating.
Fig. 2 X-ray diffraction patterns of ball-milled AgCl – metal
mixtures. The symbols indicate the lines of – AgCl,
*
+ – Ag, ¡ – Sn and ¸– Fe. The unmarked lines of the
middle pattern correspond to SnCl2
J. Therm. Anal. Cal., 90, 2007
83

TAKACS
and could not perform phase analysis using XRD.
Given the chemical knowledge of his time and the
techniques available to him, his assessment of the
observed phenomena is quite remarkable. Neverthe-
less, the inaccuracies of his paper show that results
from the early literature have to be considered with
caution and reproduced if possible..
When reducing silver chloride with zinc via
trituration in a mortar, Faraday induced a self-sustaining
reaction, or at least he came very close to doing so. The
systematic investigation of mechanically induced
self-sustaining reactions did not begin until the works of
Tschakarov 160 years later [17].
Fig. 3 Temperature close to the end of the pestle during
trituration of AgCl with Fe, Sn and Zn powders in a
mortar. (The Sn and Zn curves are shifted up by a
degree for clarity.)
Acknowledgements
Helpful discussions and encouragement by Dr. Sandra Herbert
are thankfully acknowledged.
The question remains whether Faraday induced a
true MSR, or only a fast, highly exothermic reaction. He
claimed to produce temperatures above 100°C, our
temperature never got even close to that level. Higher
temperature requires less significant heat loss to the
mortar and the air, suggesting that Faraday was using
larger quantities of the materials. In order to deliver
sufficiently intense mechanical action, he had to labor
with the pestle very hard. Faraday was a strong young
man. During his bookbinding apprentice years, he
‘could strike 1000 blows with the mallet in succession
without resting’ [16]. He was also a skilled in chemical
preparation and knew how to use the mortar and pestle
efficiently. May be he did see a real MSR after all. The
true test would have been to interrupt the rubbing and
see whether the reaction continued thermally by its own
heat. Faraday did not report such an experiment.
References
1 M. Carey Lea, Am. J. Sci., 3rd Ser., 47 (1894) 377.
2 V. V. Boldyrev, Russ. Chem. Rev., 75 (2006) 177.
3 M. Faraday, Chemical Manipulations; being Instructions
to Students in Chemistry on the Methods of Performing
Experiments of Demonstrations or of Research, with
Accuracy and with Success, W. Phillips, London 1827,
p. 147.
4 M. Faraday, Chemical Manipulations; being Instructions to
Students in Chemistry on the Methods of Performing Experi-
ments of Demonstrations or of Research, with Accuracy and
with Success, W. Phillips, London 1827, p. 597.
5 M. Faraday, The Quarterly Journal of Science, Literature
and the Arts, 8 (1820) 374.
6 G. Cantor, D. Gooding and F. A. J. L. James, Michael
Faraday, Humanity Books, Amherst, N.Y. 1996, p. 10.
7 M. Faraday, Experimental Researches in Chemistry and
Physics, Taylor and Francis, London 1859.
8 L. Takacs, J. Mater. Sci., 39 (2004) 4987.
9 L. Takacs, J. Metals, 52 (2000) 12.
Conclusions
Michael Faraday used trituration in a mortar as a
natural tool to induce reactions the ‘dry way’,
showing that mechanochemistry was probably used
earlier and more widely than usually believed. It is
possible that while the written record of early chem-
istry emphasizes the magic of changing materials by
fire, distillation, or by mixing solutions, a lot of
practical knowledge accumulated on the use of the
mortar as well. It is possible that more signs of this
early knowledge will be discovered in the future.
Although Faraday’s paper from 1820 is an interesting
early result, it probably does not mark the beginning
of mechanochemistry.
One can claim that Faraday was mistaken about
many details: the effect of Sn is certainly weaker than
the effect of Zn and the reducing ability of Fe and Cu
is almost negligible in comparison. Yet, it would be
unfair to overemphasize this fact. Faraday did not
know the thermochemical properties of the materials
10 L. Takacs, Prog. Mater. Sci., 47 (2002) 355.
11 J. A. Rodrigues, V. C. Pandolfelli, W. J. Botta F.,
R. Tomasi, B. Derby, R. Stevens and R. J. Brook,
J. Mater. Sci. Lett., 10 (1991) 819.
12 H. Yang and P. G. McCormick, J. Mater. Sci. Lett.,
12 (1993) 1088.
13 T. Tsuzuki and P. G. McCormick, J. Phys. D: Appl. Phys.,
29 (1996) 2365.
14 T. Tsuzuki and P. G. McCormick, Mater. Sci. Forum,
343–346 (2000) 383.
15 E. G. Baburaj, K. T. Hubert and F. H. (Sam) Froes,
J. Alloys Compd., 257 (1997) 146.
16 L. P. Williams, Michael Faraday, Basic Books, Inc.,
New York 1987, p. 11.
17 Chr. G. Tschakarov, G. G. Gospodinov and Z. Bontschev,
J. Solid State Chem., 41 (1982) 244.
DOI: 10.1007/s10973-007-8479-8
84
J. Therm. Anal. Cal., 90, 2007