Comment on Hultquist et al. ‘‘Water Corrodes Copper’’
[Catal. Lett. 132 (2009) 311]
Lars O. Werme
•
Pavel A. Korzhavyi
Received: 21 November 2009 / Accepted: 3 February 2010 / Published online: 23 February 2010
Ó Springer Science+Business Media, LLC 2010
The authors claim that ‘‘water corrodes copper’’. They
support that claim using first principles calculations, data
from experiments and archaeological samples, in this case
coins from the Vasa ship. The purpose of this letter is not to
discuss Hultquist et al.’s experimental results in the present
and previous papers. In order to be able to do so, much
more detailed information is necessary than what there is
normally published in journal papers. We would, however,
like to address two points in the present paper: (a) the first
principles calculations and (b) the conclusions drawn from
experiments and the Vasa coins.
(1) The possibilities of forming solid phases in the
Cu–O–H system have been addressed using a thermo-
dynamic approach and ab initio molecular dynamics.
The text states that OH
-
was used in the simulation
of the copper–OH interaction, since it is always present
in water. However, judging from the calculated
energies of adsorption, a charge-neutral isolated OH
•
molecule, i.e. the OH radical rather than pure water,
has been assumed in the initial state. There is a large
difference between the hydroxyl ion and the OH-
radical. The OH-radical, with an O 2p vacancy, is
extremely reactive and, consequently, an extremely
strong oxidant quite capable of removing electrons
from both copper and gold as is, indeed, shown by
Hultquist et al. The concentration of the OH-radical in
water is negligibly small and the presented calculations
have, therefore, no relevance for the possibility of
copper corrosion in pure water. By no means can the
adsorption energy of an OH-radical to the Cu(001)
surface be used as an estimate of the standard free
energy of formation for a hydrogen containing corro-
sion product (CuOH) that is claimed to be stable in
pure O
2
-free water. Using electronic and phonon
structure calculations [1], we have modeled the
structure and derived the enthalpy and the free energy
of formation for possible copper oxyhydrides, and
copper hydroxide (CuOH). None of them is found to be
stable with respect to Cu
2
O. The oxyhydrides, such
as Cu
4
O
1
H
2
, are unstable relative to Cu
2
O and CuH
(a metastable phase). The free energy of CuOH is
calculated to be 50 kJ/mol higher than the average of
the free energies of Cu
2
O and water. Thus, Cu
2
O is the
most stable of these Cu(I) compounds. The corrosion
of copper in pure O
2
-free water is not confirmed by first
principles calculations involving the relevant species
for that system, which does not include the OH-radical.
(2) The extrapolation of corrosion attack by Hultquist
et al. on nuclear waste containers rests heavily on two
corroded coins from Vasa. The coins are undoubtedly
corroded, but the corrosion product, if it has been
identified, is not mentioned in the text. Neither has the
environment in which the coins have been corroded
been discussed. It is also worth mentioning that over
4,200 coins were found at the Vasa site, many of them
in far better condition than the two shown in the paper
/http://samlingar.maritima.se/marketstore/index.jsp
(the site is in Swedish)/. The corrosion products on
the coins have, however, not been analyzed.
L. O. Werme (&)
Department of Physics and Materials Science, Uppsala
University, Box 530, 75121 Uppsala, Sweden
e-mail: lars.werme@fysik.uu.se
P. A. Korzhavyi
Applied Materials Physics, Department of Materials Science and
Engineering, Royal Institute of Technology (KTH), 10044
Stockholm, Sweden
123
Catal Lett (2010) 135:165–166
DOI 10.1007/s10562-010-0307-2