Investigation of a sodium hypochlorite catholyte for an aluminum aqueous battery system

Article abstract of DOI:10.1021/jp982825p

A solution-phase sodium hypochlorite (NaOCl) cathode for use in an aluminum-based seawater battery is presented. This catholyte attains a high power density, a positive redox potential, and a high Faradaic current capacity. NaOCl is also inexpensive, stable, and environmentally safe/friendly. Four aluminum alloys and 99.999% pure aluminum, serving as anodes, were tested in 3.0, 5.0, and 8.0 M NaOH solutions at the temperatures 45, 55, 65, and 80 °C. The catholyte concentration and temperature effects were optimized, and the best electrocatalyst was chosen to carry out the NaOCl reduction. Optimization experiments were carried out on a half-cell basis using a potentiostat in a three-electrode configuration with Ag/AgCl as the reference electrode and a carbon rod as the counter electrode. Full-cell testing incorporating the optimum conditions was performed with a flowing electrolyte apparatus. A catholyte concentration of 0.70 M NaOCl proved optimum in combination with 3.0 M NaOH at the temperature 55 °C and 5.0 M NaOH at the temperature 65 °C. The temperaure 45 °C was not high enough for aluminum activation, the 80 °C data did not show an improvement over the 65 °C data, and polarization was not dramatically enhanced in 8.0 M NaOH. Nickel foil and porous nickel separately electroplated with palladium were found to be the most effective electrocatalysts. Full-cell data demonstrates a successful electrochemical system with current densities approaching 1200 mA/cm² and voltages above 1.0 V.

Full text of DOI:10.1021/jp982825p

9908
J. Phys. Chem. B 1998, 102, 9908-9914
Investigation of a Sodium Hypochlorite Catholyte for an Aluminum Aqueous Battery
System
†
,‡
Maria G. Medeiros and Cynthia G. Zoski*
Newport DiVision, NaVal Undersea Warfare Center (NUWC), Building 1302, Newport, Rhode Island 02841,
and Chemistry Department, UniVersity of Rhode Island, 51 Lower College Road, Kingston, Rhode Island 02881
ReceiVed: June 30, 1998; In Final Form: August 23, 1998
A solution-phase sodium hypochlorite (NaOCl) cathode for use in an aluminum-based seawater battery is
presented. This catholyte attains a high power density, a positive redox potential, and a high Faradaic current
capacity. NaOCl is also inexpensive, stable, and environmentally safe/friendly. Four aluminum alloys and
99.999% pure aluminum, serving as anodes, were tested in 3.0, 5.0, and 8.0 M NaOH solutions at the
temperatures 45, 55, 65, and 80 °C. The catholyte concentration and temperature effects were optimized,
and the best electrocatalyst was chosen to carry out the NaOCl reduction. Optimization experiments were
carried out on a half-cell basis using a potentiostat in a three-electrode configuration with Ag/AgCl as the
reference electrode and a carbon rod as the counter electrode. Full-cell testing incorporating the optimum
conditions was performed with a flowing electrolyte apparatus. A catholyte concentration of 0.70 M NaOCl
proved optimum in combination with 3.0 M NaOH at the temperature 55 °C and 5.0 M NaOH at the temperature
65 °C. The temperature 45 °C was not high enough for aluminum activation, the 80 °C data did not show
an improvement over the 65 °C data, and polarization was not dramatically enhanced in 8.0 M NaOH. Nickel
foil and porous nickel separately electroplated with palladium were found to be the most effective
electrocatalysts. Full-cell data demonstrates a successful electrochemical system with current densities
2
approaching 1200 mA/cm and voltages above 1.0 V.
Introduction
permanganate (158.03 g/mol). From a theoretical energy
capacity standpoint, the Al-NaOCl system has a higher
The aluminum silver oxide (Al-AgO) battery1,2 has been
theoretical energy capacity (1417 W h/kg) than Al-K3Fe(CN)6
widely used and recognized as one of the most high-power and
-
(285 W h/kg) but is comparable to Al-H2O2 (1760 W h/kg)
energy density systems for underwater propulsion systems. In
and Al-KMnO4 (1400 W h/kg). NaOCl is also environmentally
safe, inexpensive, and readily available.
terms of power and energy, this battery has the theoretical cell
3
potential 2.69 V versus a standard hydrogen electrode (SHE),
Each Al-catholyte system has distinct advantages and
the theoretical charge capacity 318 A h/kg, and the theoretical
energy density 855 W h/kg. The aluminum anode4-6 has been
chosen in previous studies due to its availability, weight
considerations, cost efficiency, and its high current-carrying
capability. However, a disadvantage of the Al-AgO system
is the cost of the AgO cathode. It is this drawback that has led
to further investigations of alternate cathodes to replace AgO
in an aluminum-based alkaline battery system.
7
disadvantages. The ferricyanide system is the most predictable
and reproducible and therefore is often used in electrochemical
modeling. However, in this system there is a mixture of
ferricyanide and ferrocyanide. Ferricyanide is unstable in
sodium hydroxide solutions and becomes increasingly insoluble
as the NaOH concentration increases. The presence of a mixed
species reduces the available material necessary for reaction,
so that the charge capacity is too low for high-energy-density
The selection of such a cathode must meet certain criteria,
which include the storage of large Faradaic current capacities,
high voltages, and the ability to deliver stored energy very
rapidly. These criteria have led to the development of solution-
phase cathodes, the so-called catholytes, as possible cathodes
for aluminum-based seawater batteries. Several catholytes exist
8
systems. The permanganate system can achieve high currents
2
with corresponding high voltages (e.g. 1.8 V at 1200 mA/cm ).
In common with the ferricyanide system, the permanganate
solubility decreases with increasing NaOH concentration. Ad-
ditionally, permanganate converts to manganate at high NaOH
concentrations, temperatures above 100 °C are required for cell
activation, and precipitation of manganese dioxide limits the
7
and have been previously investigated, including ferricyanide,
8
9
permanganate, and hydrogen peroxide. Sodium hypochlorite
NaOCl) as a solution-phase cathode is another possibility which
9
system performance. Hydrogen peroxide represents a more
(
recent and simple system with a lightweight catholyte; the
system is still currently under investigation. Unfortunately, there
are safety issues which must be addressed when handling
concentrated H2O2 (>50%). Concomitantly, hydrogen peroxide
decomposes with increasing temperature, producing gases which
must be monitored. Compared to the previous systems, sodium
hypochlorite is a low-cost, environmentally safe, yet possibly
high-performance system with simple electrochemical reactions
and no safety issues.
has not been investigated. In terms of molecular weight, that
of sodium hypochlorite (74.44 g/mol) is lower than those of
both potassium ferricyanide (215.02 g/mol) and potassium
*
To whom correspondence should be addressed.
Naval Undersea Warfare Center. E-mail: medeirosmg@code80.npt.
†
nuwc.navy.mil. Fax: 401-841-6202.
‡
University of Rhode Island. E-mail: czoski@chemnt1.chm.uri.edu.
Fax: 401-874-6818.
1
0.1021/jp982825p CCC: $15.00 © 1998 American Chemical Society
Published on Web 10/22/1998

Sodium Hypochlorite Catholyte
J. Phys. Chem. B, Vol. 102, No. 49, 1998 9909
TABLE 1: Comparison of Al-Catholyte Electrochemical
intermediate aluminate species, which is water soluble. The
sodium aluminate produced ultimately leads to the production
of aluminum hydroxide, which can cause clogging of the battery
system and degraded cell performance.
Systemsa
theoretical
theoretical
theoretical
cell potential charge capacity energy density
system
Al-H
(V vs SHE)
(A h/kg)
(W h/kg)
The goal of this investigation was to reduce parasitic reactions
7 and 8 while maintaining cell potentials close to theoretical
values and simultaneously achieving high Coulombic efficien-
cies for both the anode and cathode. Reaching this goal required
optimization of the concentrations of sodium hydroxide/seawater
2
O
2
3.23
3.15
2.94
2.69
2.83
681
450
476
318
101
1760
1417
1400
855
Al-NaOCl
Al-KMnO
Al-AgO
4
Al-K
3
Fe(CN)
6
285
1
3-15
electrolyte and sodium hypochlorite catholyte
temperature.
and of
a
Theoretical refers to per kilogram of stored reactant.
Table 1 compares theoretical cell potentials, charge capacities,
Experimental Section
and energy densities for several Al-catholyte systems including
-
the Al-OCl and Al-AgO systems. These calculations were
Chemicals included sodium hydroxide (ACS grade, 99.9%,
Aldrich Chemical Co.), sodium hypochlorite (10-13 vol %,
George Mann Inc.), and sea salt (ASTM-D 1141-52, Lake
Products). The maximum stabilized concentration of com-
mercially available NaOCl is 1.4 M (13%).
Aluminum alloys and pure aluminum (ALCAN International
Limited, Kingston, Ontario, Canada) were used as working
electrode materials. Metal foils (nickel, silver, and copper,
Aldrich Chemical Co.) and porous nickel (RETIC Ni, ELTECH
Inc.) were used as electrocatalysts. RETIC Ni contains 80 pores
performed assuming standard conditions of 1 atm of pressure
and 25 °C. Table 1 illustrates that, on a theoretical basis, the
aluminum hypochlorite system should surpass the aluminum
silver oxide system in terms of cell potential, charge capacity,
and energy density. Additionally, the total weight of the
aluminum hypochlorite system is less on the basis of kilograms
of active material required for the reaction than that of the
aluminum silver oxide system.
5,10
The redox potentials versus SHE associated with the Al-
-
2
OCl system are as follows.
per inch, is 1.52 mm thick with a density of 0.4 gm/in. , and
has a surface area equal to 10 times that of planar nickel. Both
Ni foil and porous Ni plated with palladium (Tanuary Plating
Company) were also used as electrocatalysts.
-
-
cathode: 3OCl (aq) + 3H O (l) + 6e f
2
-
-
3Cl (aq) + 6OH (aq)
E° ) 0.81 V (1)
An EG&G PAR Model 371 potentiostat and an EG&G PAR
-
-
2
anode: 2AlO (aq) + 4H O(l) + 6e f
Model 379 digital coulometer were used for the 0.5 cm half-
2
2
-
cell tests. External potential and current readings were observed
using a Model 2830 digital multimeter by B+K Precision. Data
were acquired using the software system Superscope, manu-
factured by GW Instruments of Somerville, MA.
The Model 371 potentiostat was used in a three-electrode
configuration. The working electrodes (WEs) are described in
a preceding paragraph. A silver-silver chloride electrode (Ag/
AgCl, Fisher Scientific) served as the reference electrode (RE),
and the counter electrode (CE) was a carbon rod (type L-3089/
Grade AGKSP, Bay Carbon, Inc.).
EG&G flat specimen cell holders (Model K0105) were used
to obtain half-cell data. The button cell holder or WE contains
a hole (0.800 cm in diameter) in the center, exposing 0.5 cm2
of the anode material. A rubber “O” ring and the electrode
sample were placed in the holder. A stainless steel nut was
then screwed into the liner to hold the electrode disk tightly
against the face of the holder and to make good contact. The
holder was then sealed with a cap. A stainless steel rod, 10 in.
long, was inserted through the top of the holder to make contact
with the metal liner. A glass sleeve, extending from the holder
to 2 in. from the top of the rod, was placed around the rod. The
top of the steel rod is threaded, which allows for a nut to screw
down to the rod to hold the glass sleeve securely in place. It is
at the stainless steel rod that the connection is made to the
potentiostat. Once the cell holder was loaded with clean and
weighed material, it was lowered into the electrolyte. All
connections (WE, RE, CE) were made to the potentiostat prior
to testing.
2Al(s) + 8OH (aq)
E° ) -2.34 V (2)
-
-
overall: 2Al(s) + 3OCl (aq) + 2OH (aq) f
-
2
-
2
AlO (aq) + H O(l) + 3Cl (aq)
E° ) 3.15 V (3)
2
-
Reactions 1-3 demonstrate that the Al-OCl system is capable
of attaining a higher potential than that associated with the Al-
AgO system.
-
cathode: 3AgO(s) + 3H O(l) + 6e f
2
-
3
Ag(s) + 6OH (aq)
E° ) 0.35 V (4)
-
-
anode: 2AlO (aq) + 4H O(l) + 6e f
2
2
-
2Al(s) + 8OH (aq)
E° ) -2.34 V (5)
-
overall: 2Al(s) + 3AgO(s) + 2OH (aq) f
-
2
2
AlO (aq) + H O(l) + 3Ag(s)
E° ) 2.69 V (6)
2
Unfortunately these theoretical open circuit potentials for both
systems are reduced and the electrochemical performance is
inhibited by the following parasitic reactions.5
-
corrosion reaction: 2Al(s) + 2H O(l) + 2OH (aq) f
2
-
2
AlO (aq) + 3H v(g) (7)
2
2
-
-
direct reaction: 2Al(s) + 3OCl (aq) + 2OH (aq) f
-
2
-
2
AlO (aq) + H O(l) + 3Cl (aq) (8)
Seawater electrolyte solutions consisted of either 3.0, 5.0, or
2
8
.0 M NaOH and 40 g/L of sea salt. The prepared solutions
Corrosion reaction 7 occurs due to the direct interaction between
sodium hydroxide and the aluminum anode, resulting in the
were placed in a 650 mL beaker with a stirring bar.
A
thermolyne (Type 1000 Series) hot plate was used to heat the
solutions to temperatures of 45, 55, 65, or 80 °C. Once the
desired temperature was reached, catholyte was added and
testing began. The temperature of 80 °C is the maximum
1
1,12
production of sodium aluminate and hydrogen gas.
The
direct reaction 8 also produces sodium aluminate in addition to
chloride ions. Thus both parasitic reactions produce the

9910 J. Phys. Chem. B, Vol. 102, No. 49, 1998
Medeiros and Zoski
Figure 1. Flowing electrolyte apparatus.
temperature that is appropriate on a total systems basis. Beyond
this temperature, electrolyte boiling occurs.
Anodic polarizations were conducted by applying a potential
ranging from -2.2 to 0.0 V versus Ag/AgCl and recording the
current in milliamperes. The potentials were scanned at 10 mV/
s. The maximum current carrying capacity of the potentiostat
was 1 A. The potential range applied for the electrocatalyst
and hypochlorite polarizations was 0.4 to -1.0 V versus Ag/
AgCl. The measured current for half-cell experiments was
converted to current density by dividing the current by the
2
2
electrode area of 0.5 cm , resulting in mA/cm . The data
collected included half-cell voltages, current, initial and final
temperatures, and initial and final open circuit voltages. Some
of the data are presented as polarization graphs where potential
(
V vs Ag/AgCl) is displayed on the x-axis and current density
2
(mA/cm ) on the y-axis. All experiments were performed in
triplicate and the data averaged.
Full-cell testing was performed utilizing a flowing electrolyte
system, as shown in Figure 1. This apparatus consists of a 2-L
electrolyte tank set into an electrothermal unimantle. An
Isomatec pump system is used to pump the seawater electrolyte
through the flow loop. A Polystat constant-temperature circula-
tor is used to maintain a constant temperature of the seawater
electrolyte. The cell holder is made of plexiglass and has
electrode dimensions of 1.5 in. by 8 in., as shown in Figure 2a.
Figure 2. Flow cell: (a) 1.5 in. by 8 in. cell holder; (b) single-cell
configuration.
3
The apparatus also contains a gas-monitoring device (cm /min),
pressure gauges to monitor seawater electrolyte flow and to alert
when blockages occur, multimeters, and a data acquisition
system. Data were collected using Laboratory Notebook TM
and analyzed using Microsoft Excel, Cricket Graph, and Quatro
Pro.
are set up side by side with a fixed gap between them. A
separator, which is not shown in Figure 2b, maintains the cell
gap between the electrodes. The electrode area for the full cell
2
is 77.4 cm . The effective electrode area, excluding the masking
2
by the separator, is 68.3 cm ; this is the number that was used
At the start of an experiment, seawater electrolyte is pumped
to convert current to current density in full-cell experiments.
The separator used is VEXAR, a polyethylene woven material
that is 0.635 mm thick. NaOH seawater electrolyte and NaOCl
catholyte enter the inlet at the bottom of the cell. Hypochlorite
reacts with the catalytic surface of the electrocatalyst to carry
3
at the fixed flow rate 250 cm /min from the reservoir to an inlet
located at the bottom of the cell. The seawater electrolyte exits
the cell at the top and returns to the reservoir, where it is
recirculated. The temperatures of the electrolyte reservoir, cell
inlet, and cell outlet are recorded. Changes in cell inlet and
outlet temperatures were especially monitored; this temperature
change indicates how efficiently the cell is operating. Large
temperature changes are associated with high gassing, high
corrosion, and low efficiencies. Temperature changes of 10 °C
are reasonable for high-rate systems such as the one reported
here.
-
-
-
out the reduction of OCl , thereby producing Cl and OH ions.
The hydroxide ions react with the aluminum anode, producing
aluminate. The NaOH seawater electrolyte/catholyte is continu-
ously pumped through the cell, exiting through the outlet at the
top of the cell.
Results and Discussion
Figure 2b illustrates the configuration for a single cell. The
dimensions of the electrodes (anode and electrocatalyst) are 1.5
in. in width and 8 in. in height. The anode and electrocatalyst
Anode Characterization: Seawater Electrolyte Concen-
tration and Temperature Variations. Experiments were

Sodium Hypochlorite Catholyte
J. Phys. Chem. B, Vol. 102, No. 49, 1998 9911
2
TABLE 2: Current Densities (mA/cm ) at 3.0, 5.0, and 8.0
M NaOH for Each Aluminum Anode Considered at the
Designated Temperaturea
temp
Al
NaOH
45 °C
55 °C
65 °C
80 °C
XA5P
3 M *440 ( 3 *620 ( 10 1770 ( 15 *1820 ( 38
5
8
M
M
*680 ( 12 1820 ( 5
1900 ( 46 2000 ( 5
2000 ( 3
*560 ( 9 *700 ( 14 1980 ( 6
3 M *350 ( 10 *350 ( 27 *700 ( 13 1950 ( 19
DF50V
pure Al
Alcoa
5
8
3 M
5
8
3 M
M
M
*200 ( 9 *580 ( 20 *1450 ( 9
*350 ( 10 *600 ( 14 *220 ( 4
1980 ( 3
*400 ( 5
1200 ( 3
450 ( 7
650 ( 13
780 ( 6
M
M
480 ( 24 800 ( 28 1100 ( 7 *1200 ( 12
*400 ( 14 650 ( 16
450 ( 7 600 ( 15
900 ( 38 *1000 ( 16
700 ( 13 *900 ( 22
5
8
M
M
500 ( 15 650 ( 13 1020 ( 7 *1300 ( 8
*380 ( 6
380 ( 9
600 ( 24
590 ( 10
900 ( 16 *1380 ( 15
800 ( 20 1200 ( 3
980 ( 13 1500 ( 9
900 ( 18 1450 ( 26
potline Al 3 M
5
8
M
M
420 ( 10 650 ( 26
*350 ( 6
590 ( 8
Figure 3. Polarization curves for several aluminum anodes in 3.0 M
NaOH and at 65 °C.
a
An * is used to indicate where passivation occurred. ( signifies
the standard deviation.
undertaken to determine if additives or alloying materials in
the aluminum anode were critical or if more inexpensive grades
of aluminum would be acceptable. Additives such as Sn, Ga,
Mg, and Pb alloyed to aluminum have been found to aid in
Table 2 summarizes the polarization results. The highest
current density achieved during a polarization experiment is
recorded as a function of temperature and sodium hydroxide
concentration. Thus, for each aluminum sample and for a
specific temperature, three current densities are listed which
correspond to the measured current densities at 3.0, 5.0, and
activation through depassivation of the oxide layer and gas
reduction.1
6,17
Passivation is the result of the formation of an
impermeable oxide layer on the metal surface which protects
the underlying metal from further oxidation. Current density
decreases as the voltage is varied at the onset of passivation
and then reaches a constant value when passivation of the metal
surface is complete. Gallium increases the voltage range over
which aluminum is activated by breaking down the oxide layer.
Lead increases the atmospheric/chemical corrosion resistance
and Coulombic efficiencies of the aluminum; however, it also
decreases the voltage. The high overpotential of tin for the
production of hydrogen gas suppresses corrosion reaction 7 and
improves the voltage. Magnesium renders the aluminum sample
more malleable.
8
.0 M NaOH, respectively. The standard deviation is indicated
by the ( value. The numbers with asterisks (*) signify the
highest current density achieved before passivation.
Aluminum alloy XA5-P passivated at 45 °C in the three
NaOH concentrations considered with low current densities
ranging from 440 to 680 mA/cm . At 55 °C, passivation
occurred only in 3.0 and 8.0 M NaOH, whereas in 5.0 M NaOH
the highest current density of 1820 ( 5 mA/cm was achieved
for this temperature. No passivation occurred at 65 °C;
however, similar current densities of 1900 ( 46 and 1980 ( 6
mA/cm were achieved in 5.0 and 8.0 M NaOH, respectively.
At 80 °C, XA5-P passivated only in 3.0 M NaOH and a current
density of 2000 mA/cm was observed in both 5.0 and 8.0 M
NaOH. The inconsistency in behavior observed in 3.0 M NaOH
and at 80 °C is under further investigation.
The DF50V aluminum alloy passivated at all temperatures
in 8.0 M NaOH and also at 45, 55, and 65 °C in 3.0 and 5.0 M
NaOH. Current densities were generally low, ranging from 200
to 700 mA/cm , except at 65 °C in 5.0 M NaOH, where a current
density of 1450 ( 9 mA/cm was observed. At 80 °C, no
passivation was observed in 3.0 and 5.0 M NaOH and large
current densities of 1950 ( 38 and 1980 ( 3 mA/cm ,
respectively, were achieved.
No passivation of pure aluminum occurred in 3.0 M NaOH
at the specified temperatures; current densities increased with
temperature, reaching a maximum of 1200 ( 3 mA/cm at 80
°C. In 5.0 M NaOH, pure aluminum passivated only at 80 °C,
where a maximum current density of 1200 ( 12 mA/cm was
2
2
2
Five different aluminum anodes were evaluated. Four of the
2
18
18
anodes were aluminum alloys: DF50V, XA5-P, Alcoa #500,
and potline aluminum. Aluminum alloys DF50V and XA5-P
contain the additives Mg, Sn, and Ga. Alcoa #500 aluminum
contains numerous additives in addition to Mg, Sn, and Ga.
The exact amounts/percentages were not available from the
manufacturer (Alcan International) for proprietary reasons.
However, the manufacturer would confirm that the Ga percent-
age was greater in XA5-P, that the Mg percentage was greater
in DF50V, and that the Sn percentage was the same in both
XA5-P and DF50V. Potline aluminum is an impure form of
aluminum obtained from mined bauxite. The fifth anode was
2
2
2
1
9
9
9.999% pure aluminum, containing no alloying additives.
Experiments were also designed to determine the optimum
NaOH concentrations20 in addition to the optimum operating
temperature. Sodium hydroxide concentrations of 3.0, 5.0, and
2
2
8
.0 M were evaluated at temperatures of 45, 55, 65, and 80 °C
at each aluminum anode. Polarization curves were used to
compare current density (mA/cm ) versus potential (V) at the
attained. Within experimental error, this is the same current
density which was achieved in 3.0 M NaOH, where passivation
was absent. Passivation also occurred in 8.0 M NaOH at
temperatures of 45 and 80 °C, where the largest current density
was 1000 ( 16 mA/cm at the latter temperature. The
performance of Alcoa aluminum was similar to that of pure
aluminum.
2
three NaOH concentrations and the four temperatures at each
aluminum anode. Comparison of these polarization curves
enabled the identification of the best aluminum anodes and
optimum NaOH concentration and temperature. Figure 3 is an
example of polarization curves recorded at each aluminum anode
at 65 °C and in 3.0 M NaOH. Passivation occurs only at the
DF50V aluminum anode under these experimental conditions.
2
Potline aluminum passivated only at 45 °C in 8.0 M NaOH
2
with a current density of 350 ( 6 mA/cm . A maximum current

9912 J. Phys. Chem. B, Vol. 102, No. 49, 1998
Medeiros and Zoski
2
TABLE 3: Current Densities (mA/cm ) at Pure Aluminum,
XA5-P Aluminum Alloy, and DF50V Aluminum Alloy for
Each Catholyte Concentration at the Designated
Temperaturea
aluminum, and DF50V aluminum, respectively. The standard
deviation is indicated by the ( value. The numbers with
asterisks (*) signify the highest current density achieved before
passivation.
temp
The first number under each temperature in Table 3 corre-
sponds to current densities recorded at a pure aluminum anode.
For this anode, there is no activation at any of the designated
temperatures; aluminum passivation occurs at current densities
NaOCl
Al sample
45 °C
55 °C
65 °C
80 °C
0.35 M
0.70 M
1.40 M
pure Al
XA5-P
DF50V
pure Al
XA5-P
DF50V
pure Al
XA5-P
DF50V
*30 ( 1 *35 ( 1 *35 ( 3 *48 ( 1
200 ( 1 220 ( 3 220 ( 2 260 ( 1
200 ( 4 200 ( 5 275 ( 2 320 ( 4
*28 ( 1 *25 ( 0 *22 ( 2 *22 ( 3
280 ( 3 350 ( 3 400 ( 3 440 ( 7
340 ( 1 360 ( 4 440 ( 1 480 ( 4
*50 ( 1 *48 ( 2 *48 ( 2 *50 ( 1
550 ( 6 560 ( 7 680 ( 3 700 ( 8
600 ( 4 650 ( 6 690 ( 1 780 ( 4
2
between 22 and 50 mA/cm . Thus, some type of impermeable
layer must be forming on the surface of pure aluminum to show
passivation at all temperatures and NaOCl concentrations
considered. The exact nature of the surface layer is under
investigation.
Current densities are significantly larger for aluminum alloy
XA5-P, as shown by the second number under each temperature
in Table 3. No passivation was observed at the three NaOCl
concentrations and four temperatures considered. At a specific
NaOCl concentration, there was an approximate 1.1-fold
increase in current density for each 10 °C increase in temper-
ature. However, the effect of NaOCl concentrations at indi-
vidual temperatures led to an approximate 1.6-fold increase in
current density from 0.35 to 0.70 M NaOCl and a 1.7-fold
increase from 0.70 to 1.40 M. A 4-fold change in NaOCl
concentration from 0.35 to 1.40 M led to an approximate 2.8-
fold increase in current density at each specified temperature.
For aluminum alloy DF50V, current densities are indicated
by the third number under each temperature in Table 3. These
current densities are generally slightly larger than those observed
for XA5-P. Similar to the case for alloy XA5-P, no passivation
was observed at the three NaOCl concentrations and four
temperatures considered, and the increase in current density was
approximately 1.1-fold at a specific NaOCl concentration.
Additionally, the effect of NaOCl concentrations at individual
temperatures led to an approximate 1.6-fold increase in current
density from 0.35 to 0.70 M NaOCl and a 1.7-fold increase
from 0.70 to 1.40 M. A 4-fold change in NaOCl concentration
from 0.35 to 1.40 M led to an approximate 2.8-fold increase in
current density at each specified temperature. These observa-
tions were also noted for XA5-P.
a
An * is used to indicate where passivation occurred. ( signifies
the standard deviation.
2
density of 1500 ( 9 mA/cm was observed in 5.0 M NaOH at
8
0 °C, where no passivation occurred.
There is a clear trend of an increase in current density, and
hence activation, with temperature at NaOH concentrations
evaluated for XA5-P aluminum, pure aluminum, Alcoa alumi-
num, and potline aluminum. For DF50V aluminum, current
density remained essentially the same at 45 and 55 °C before
increasing at 65 and 80 °C in 3.0 M NaOH. In 5.0 M NaOH,
there was an increase in current density with temperature at the
DF50V aluminum anode. The behavior of DF50V aluminum
in 8.0 M NaOH was sporadic as the temperature increased.
In general, experiments performed at the 8.0 M NaOH level
resulted in a decrease in current density of the aluminum samples
evaluated at a specific temperature. At 80 °C, slight enhance-
ments in current density were observed over those at 65 °C.
Overall, maximum current densities occurred in the presence
of 3.0 and 5.0 M NaOH and at the temperatures 55 and 65 °C.
Of the aluminum samples evaluated, aluminum alloy XA5-P
outperformed the others. Alloy DF50V passivated the most as
the NaOH concentration and temperature were changed; the only
exceptions occurred at 80 °C in 3.0 and 5.0 M NaOH. Pure
aluminum, potline aluminum, and Alcoa aluminum samples
performed similarly and exhibited the lowest current densities
as NaOH concentration and temperature were varied.
The enhanced cell performance observed for XA5-P can be
attributed to the greater percentage of gallium than found in
DF50V. Although gallium is also present in Alcoa aluminum,
the presence of many other additives appears to hinder the cell
performance of this sample. Potline aluminum did not perform
well under the test conditions due to its impure state.
Pure aluminum also demonstrated lower current densities.
However, the monitoring of its performance was continued in
order to serve as a benchmark against which XA5-P and DF50V
could be compared as catholyte concentrations and electrocata-
lysts were investigated.
The data in Table 3 demonstrate that none of the NaOCl
concentrations considered are large enough at any temperature
to activate the pure aluminum anode. However, NaOCl
concentrations of 0.7 and 1.4 M do show current densities
2
between 440 and 780 mA/cm at 80 °C for aluminum alloys
XA5-P and DF50V. Since these current densities are not large
enough for high-rate applications where current densities greater
2
than 1000 mA/cm are required, the effect of both NaOH and
NaOCl on current density magnitude was investigated for
increases in current density.
NaOCl and NaOH Concentration Effects. Table 4 sum-
marizes the polarization results for XAP-5 Al, DF50V Al, and
pure aluminum at 55 °C as a function of NaOCl concentration
in the absence and presence of 3.0 M NaOH. For each
aluminum sample and for a specific NaOCl concentration, two
current densities are listed which correspond to the measured
current densities in the presence of 0 M NaOH and 3.0 M
NaOH, respectively. The numbers with asterisks (*) represent
the highest current density achieved before passivation, and (
signifies the standard deviation.
Catholyte Concentration Characterization. A series of
experiments at the temperatures 45, 55, 65, and 80 °C were
performed to determine an optimum NaOCl concentration in
the absence of NaOH. NaOCl concentrations of 0.35, 0.70, and
.40 M were used in the presence of seawater. Pure aluminum,
XA5-P aluminum, and DF50V aluminum served as the anode.
Polarization graphs similar to Figure 3 were generated. Table
1
3
summarizes the polarization results. The highest current
densities achieved during the polarization experiments are
recorded as a function of temperature and catholyte concentra-
tion. Thus, for each NaOCl concentration and for a specific
temperature, three current densities are listed which correspond
to the measured current densities at pure aluminum, XA5-P
For XAP-5 aluminum, catholyte alone activates the anode to
2
a current density of 220 mA/cm compared to the passivation
2
which occurs at a current density of 620 mA/cm in the presence
of 3.0 M NaOH only. When a combination of 0.35 M NaOCl
and 3.0 M NaOH is used, no passivation occurs and a

Sodium Hypochlorite Catholyte
J. Phys. Chem. B, Vol. 102, No. 49, 1998 9913
2
TABLE 4: Current Densities (mA/cm ) at 0 and 3.0 M
NaOH for Each Aluminum Anode at the Designated NaOCl
Concentration at 55 °Ca
NaOCl
Al sample
XAP-5
NaOH
0 M
0 M
0.35 M
0.70 M
220 ( 2
1780 ( 14
200 ( 5
1380 ( 13
*35 ( 1
*350 ( 10
350 ( 3
1780 ( 13
360 ( 4
1420 ( 18
*25 ( 0
*420 ( 12
3
M
*620 ( 10
*350 ( 27
650 ( 13
DF50V
pure Al
0 M
3
M
0 M
3
M
a
An * is used to indicate where passivation occurred. ( signifies
the standard deviation.
Figure 5. Measured voltages as a result of applying a constant current
at 55 °C. The full cell included 3.0 M NaOH, seawater electrolyte,
0
.70 M NaOCl, XA5-P as the anode, and either Ni foil with Pd or
porous Ni with Pd as the electrocatalyst. The lines were generated using
linear regression.
0
.4 to -1.0 V. Cu foil, Ni foil with Pd, and porous Ni all
demonstrated an increase in cathodic current density, as expected
over the potential range. Copper was not investigated further
due to its excessive gassing during testing. Porous nickel and
nickel foil were both chosen as the preferred backbones for the
electrolytic palladium surface in the sodium hypochlorite
reduction.
Full-Cell Testing. The previous sets of experiments indicate
that optimum parameters include [NaOH] ) 3.0-5.0 M,
[NaOCl] ) 0.35-0.70 M, temperatures ) 55-65 °C, electro-
Figure 4. Electrocatalyst determinations in 3.0 M NaOH and 0.35 M
NaOCl at 55 °C.
catalyst ) Ni as a substrate, and XA5-P as the anode. Several
combinations of optimum conditions were tested in a full-cell
configuration.
significantly enhanced current density of 1780 mA/cm2 is
observed; this is approximately three times larger than the
current density observed in 3.0 M NaOH alone and ap-
proximately eight times larger than the current density observed
in 0.35 M NaOCl alone. Fourfold and 7-fold increases were
noted for DF50V aluminum, which achieved a current density
Figure 5 shows full-cell polarization profiles at the temper-
ature 55 °C, 3.0 M NaOH, and 0.70 M NaOCl when the
electrocatalyst is either Ni foil or porous Ni plated with Pd.
Open circuit voltages between 2.0 and 2.4 V were measured.
Ni foil/Pd demonstrated a current density range from 0 to 700
2
of 1380 mA/cm under identical combination conditions. For
2
pure aluminum, a current density of 350 mA/cm was observed
2
mA/cm with corresponding voltages of 2.4-0.8 V. In com-
in a 0.35 M NaOCl and 3.0 M NaOH combination. An increase
in NaOCl concentration to 0.70 M, with NaOH concentration
fixed at 3.0 M, did not result in significant changes in current
density.
parison, porous Ni/Pd demonstrated an increase in current
2
density from 0 to 1200 mA/cm with corresponding voltages
of 2.0-1.0 V.
Figure 6 illustrates the constancy of the measured voltage
2
The goal of achieving 1000 mA/cm was attained with 3.0
2
with time at a constant current of 28 A (410 mA/cm ) for a cell
M NaOH and 0.35 and 0.70 M NaOCl. Any additional increase
in the NaOH or NaOCl concentrations would only impact
negatively on the system in terms of corrosion and decomposi-
tion reactions.
consisting of the electrocatalyst Ni foil with Pd, a XA5-P Al
anode, [NaOH] ) 3.0 M, [NaOCl] ) 0.70 M, and T ) 55 °C.
A steady voltage of 1.27 V was observed for the 17 min (1020
s) test duration. This measured voltage is consistent with the
results shown in Figure 5. The efficiency of the XA5-P Al
anode was calculated to be 52% from Coulombic calculations
and initial and final aluminum weights. The measured total
Electrocatalysts. Silver foil, nickel foil, copper foil, and
porous nickel were investigated as possible catalysts for the
reduction of NaOCl. Nickel foil was first electroplated with
9
3
palladium, and porous nickel was used with and without
gas generated for 17 min was 330 cm , leading to a gassing
3
palladium. Porous Ni was used to determine if this high-surface-
area material, representing a 10-fold increase in surface area
over that of a strictly planar surface, would significantly reduce
mass-transfer-limiting processes occurring at the catalyst surface
compared to the anode surface.
rate of 19.4 cm /min.
An experiment similar to that shown in Figure 6 was
performed with a porous Ni with Pd electrocatalyst. For a
2
constant current of 55 A (800 mA/cm ), a steady voltage of
1
.35 V was measured, which is verified by the data shown in
Figure 4 demonstrates that Ag foil performed ideally from
.4 to -0.1 V before passivation occurred at -0.1 V. Ni foil
showed zero current-carrying capability over the potential range
Figure 5. The aluminum efficiency was found to be 61%, and
3
0
the total gas generated was 391 cm for 17.3 min or a gassing
3
rate of 22.6 cm /min.

9914 J. Phys. Chem. B, Vol. 102, No. 49, 1998
Medeiros and Zoski
These results demonstrate the viability of Al-NaOCl electro-
chemistry in developing a seawater battery for high-rate
underwater applications.
Acknowledgment. M.G.M. acknowledges the NUWC In-
dependent Research (IR) Program for its funding and co-workers
Eric Dow and Christopher Egan. C.G.Z. gratefully acknowledges
the award of NSF Research Planning Grant 93-130. We are
indebted to Allen Bard and the reviewers for their helpful
comments.
References and Notes
(
1) Anderson, G. Al-AgO Primary Battery. U.S. Patent #3, 953, 239,
976.
2) Linden, D. Handbook of Batteries, 2nd ed.; McGraw-Hill Publish-
ers: New York, 1995.
3) Latimer, W. Oxidation Potentials, 2nd ed.; Prentice Hall: New
York, 1953.
1
(
(
(4) Scamans, G.; Hunter, J.; Tuck, C.; Hamlen, R.; Fitzpatrick, N. In
Power Sources, XII; Killy and Baxter, Eds.; Elsevier Science Inc.: New
York, 1988; pp 363-369.
Figure 6. Voltage versus time in response to a constant current applied
at 55 °C. The full cell included 3.0 M NaOH, seawater electrolyte,
(
5) Hunter, J.; Scamans, G.; Tuck, C.; Hamlen, R.; Fitzpatrick, N.
0
.70 M NaOCl, XA5-P as the anode, and Ni Foil with Pd as the
Aluminum Batteries. European Patent Application #0354752, Alcan Int.
Ltd., 1989.
electrocatalyst.
(
6) Cotton, F. A.; Wilkinson, G. AdVanced Inorganic Chemistry, 5th
ed.; Wiley-Interscience: New York, 1988.
7) Licht, S.; Orndorff, C. M. An Aluminum Ferricyanide Battery. Navy
Case 74011, 1991.
Summary
(
The use of sodium hypochlorite in an aluminum-based
seawater battery system has been demonstrated. The battery
system consisted of an aluminum anode, a sodium hydroxide
and seawater electrolyte, an electrocatalyst, and a sodium
hypochlorite catholyte. Several aluminum alloys and pure
aluminum were considered as possible anodes. The aluminum
alloy XA5-P was found to consistently attain the highest anodic
current densities in 3.0 and 5.0 M NaOH solutions at the
temperatures 55 and 65 °C. Optimum NaOCl concentrations
of 0.35 and 0.70 M were demonstrated. Nickel foil and porous
nickel electroplated with palladium proved to be the most
effective electrocatalysts for the NaOCl reduction. Crucial to
the performance of this particular aluminum-based battery is
the combination of NaOH and NaOCl electrolytes. Sodium
hypochlorite alone was insufficient to activate the aluminum
anode. However, when combined with sodium hydroxide, an
(8) Licht, S.; Marsh, C. An Aluminum Permanganate High Current
Density Battery. Navy Case 73985, 1991.
(
9) Marsh, C.; Seebach, G.; VanZee, J.; Bessette, R.; Meunier, H.;
Medeiros, M. G. Preparation of an Electrocatalytic Cathode for an
Aluminum-Hydrogen Peroxide Battery. U.S. Patent #5, 296, 429, 1994.
(10) Bard, A. J.; Parsons, R.; Jordan, J. Standard Potentials in Aqueous
Solution; Marcel Dekker Inc.: New York, 1985.
11) Scamans, G.; Hunter, J.; Holroyd, N. Treatise Mater. Sci. Technol.
1989, 31, 1121-1124.
12) MacDonald, D.; Lee, K. H.; Moccari, A.; Harrington, D. Corrosion,
988, 44, 652-657.
13) Piskin, B.; Turkun, M. J. Endodontics 1995, 21, 253-255.
(14) Johnson, B. R.; Remeikis, N. A. J. Endodontics 1993, 19, 40-43.
15) Skarzewski, J.; Siedlecka, R. Org. Prep. Proced. Int. 1992, 24, 623-
47.
16) Tracy, G.; Mazgaj, R. Aluminum Alloys and Associated Anodes.
(
(
1
(
(
6
(
U.S. Patent #4808, 489, 1989.
(17) Tuck, C.; Hunter, J.; Scamans, G. J. Electrochem. Soc. 1987, 134,
2
970-2981.
(18) Perkins, J.; Medeiros, M. G.; Atwater, D.; Dow, E. A Comparison
8
-fold increase in current density over that observed in sodium
Study of Aluminum Alloys AB50V and DF50V: The Effects of Temper-
ature, Constant Power and Electrolyte. NUWCDIVNPT Technical Memo-
randum #932138, 1993.
hypochlorite alone was obtained. Full-cell testing resulted in
current densities up to 1200 mA/cm2 with corresponding
voltages above 1.0 V. With Ni foil/Pd as the electrocatalyst,
an aluminum efficiency of 52% was calculated and minimal
gas was generated. The most promising result was obtained
with porous Ni/Pd as the electrocatalyst, where 61% aluminum
(19) Marsh, C.; Seebach, G.; Bessette, R.; Muenier, H.; Van Zee, J.;
Medeiros, M. G. Electrolytically activated aqueous alkaline battery having
a pure aluminum anode and a method for providing a multi-performance
battery thereform. U.S. Patent #72909, 1990.
(
20) Fontana, M. Corrosion Engineering, 3rd ed.; McGraw-Hill, Inc.:
New York, Chapter 5.
21) Newman, J. S. Electrochemical Systems; Prentice Hall: New York,
1973.
efficiency was found in response to a constant current of 55 A
(
2
(800 mA/cm ), which generated a steady voltage of 1.35 V.