Hydrogen production by methanol-water solution electrolysis with an alkaline membrane cell

Abstract

The basic concept of methanolewater solution electrolysis with an alkaline membrane and the experimental
studies are presented. The measurements were performed with a membrane electrode assembly
consisting of an anion exchange membrane, Pt/C cathode and either PtRu/C or Pt/C anode. Hydrogen
production efficiency was measured and the effects of temperature and methanol concentration on the
electrolysis performance were investigated. PtRu was found to be a better catalyst at low potentials and
the hydrogenwas produced with a high efficiency. The results indicate that the favored kinetics and mass
transport at high temperatures result in increased hydrogen production. Additionally, it was found that,
increasing methanol concentration decreased the performance of the electrolysis as the membrane
morphology changes as a function of the concentration.

Full text

Hydrogen Production by Methanol-Water Solution
Electrolysis with an Alkaline Membrane
Sami Tuomi, Annukka Santasalo-Aarnio, Petri Kanninen, Tanja Kallio*
Research Group of Electrochemical Energy Conversion and Storage, Department of Chemistry, Aalto University, P.O.
Box 16100, 00076 Aalto, Finland
*Corresponding author. Tel.: +358 50 563 7567, Fax: +358 9 470 22580, E-mail address: tanja.kallio@aalto.fi
Abstract
The basic concept of methanol-water solution electrolysis with an alkaline membrane and the
experimental studies are presented. The measurements were performed with a membrane electrode
assembly consisting of an anion exchange membrane, Pt/C cathode and either PtRu/C or Pt/C
anode. Hydrogen production efficiency was measured and the effects of temperature and methanol
concentration on the electrolysis performance were investigated. PtRu was found to be a better
catalyst at low potentials and the hydrogen was produced with a high efficiency. The results indicate
that the favored kinetics and mass transport at high temperatures result in increased hydrogen
production. Additionally, it was found that, increasing methanol concentration decreased the
performance of the electrolysis as the membrane morphology changes as a function of the
concentration.
Keywords: Hydrogen production; Alkaline; Methanol; Electrolysis
1

1. Introduction
Hydrogen is considered as one of the most promising energy carriers for future electrical devices
owing to its environmentally friendly production possibilities and high energy density. There are
various methods for hydrogen production but electrolysis is the most suitable for on-site production,
e.g. producing hydrogen for mobile or transportable electrical devices using fuel cells, although it is
not limited by the size of the applications and can be also utilized for industrial hydrogen
production. Water electrolysis is the conventional form of electrolysis but the energy consumption is
high and therefore an interest for hydrogen production by methanol electrolysis has arisen due to
various advantages over conventional water electrolysis [1-5]. The standard potential of methanol
electrolysis is only 0.02 V which is significantly lower than the 1.23 V of water electrolysis
resulting in decreased power consumption, though in practice the operation voltage of methanol
electrolysis is above 0.4 V [1, 3, 5] and that of water electrolysis above 1.4 V [6-8]. The cost for
long term hydrogen production by water electrolysis for large-scale applications is mainly
determined by the energy consumption as the cost of electrolysis equipment becomes insignificant.
It has been estimated that the energy saving with methanol electrolysis is about 70 % compared to
water electrolysis and when the price of methanol is taken into account methanol electrolysis is
expected to be 50 % cheaper than water electrolysis [9].
When considering a small scale portable application also the expense of the equipment is relevant.
Electrolysis is closely related to fuel cells and one important area in the fuel cell research has been
finding cheaper materials and techniques to manufacture the cells. For instance, the costs of Nafion®
proton exchange membranes are high due to the complicated manufacturing process and thus anion
exchange membranes provide a possibility for a significant decrease in the electrolyte price.
2

Another reason to investigate alkaline membrane cells is the enhanced methanol oxidation kinetics
in alkaline media [10, 11] enabling decreased catalyst loadings and there is even a possibility to
utilize non-precious metal catalysts for methanol oxidation [12-14]. Furthermore, methanol cross-
over from an anode to a cathode in the Nafion membrane methanol electrolysers increases fuel
consumption and generating impurities in the produced hydrogen, consequnetly membranes have
been developed further to overcome this problem [15-17]. Anion exchange membranes, however,
are more resistant towards methanol cross-over, for example a Fumapem® FAA-2 membrane has a
cross-over rate only 16 % of the Nafion-115 membrane [18].
In this paper the basic concept of methanol-water solution electrolysis with an alkaline membrane is
introduced. In addition, experimental studies with this electrolysis method, including efficiency
measurements, were performed and the effects of temperature and methanol concentration on the
electrolytic performance are presented.
2. Methanol-water solution electrolysis with an alkaline membrane
A schematic of methanol-water solution electrolysis with an alkaline membrane is shown in Fig. 1:
the electrolytic cell is composed of an anode, a cathode and an anion exchange membrane
electrolyte placed between the electrodes. Supported metals are used as the electrocatalysts on both
electrodes which are connected to each other through an electrical circuit via a DC power source.
3

Methanol-water solution is fed to the anode where methanol is oxidized in a reaction with
hydroxide ions transferred through the anion exchange membrane. The products at the anode are
carbon dioxide, water and electrons according to Eq 1. The electrons are circulated to the cathode
through the DC power source.
−−
++→+
e 6OH 5COOH 6 OHCH
223
(1)
SHE vs.V 81.0
0
a
=E
Water is transported through the membrane to the cathode where it is reduced releasing hydrogen
gas and forming hydroxide ions according to Eq. 2. Hydroxide ions are transported through the
anion exchange membrane to the anode compartment.
22
HOH 2e 2OH 2
+→+
−−
(2)
SHE vs.V 83.0
0
c
−=
E
The resulting overall reaction for the methanol-water solution electrolysis is expressed in Eq.3.
4

2223
H 3COOH OHCH
+→+
(3)
V 02.0
0
−=
E
It is crucial that no oxygen is present at the cathode in order to avoid the more advantageous fuel
cell type reaction displacing hydrogen evolution.
3. Experimental
Carbon supported Pt catalyst (60 % Pt/C, Alfa Aesar) was used as the cathode catalyst and either
carbon supported Pt-Ru alloy (40 % Pt – 20 % Ru /C, Alfa Aesar) or the aforementioned Pt/C as the
anode catalyst. The catalyst ink was prepared by mixing the catalyst with a 2:1 solution of
isopropanol and water, and it was ultrasonicated in a water bath for 30 min to form a homogeneous
mixture. Alkaline ionomer polyvinylbenzyltrimethylammonium (PVBTA) (MW 100 000) from
Scientific Polymer Products Inc. and divinylbenzene (DVB) cross-linker from Merck were mixed
into a 2:1 solution of isopropanol and water, and heated in a boiling water bath for 15 min to
activate the polymerization. Catalyst ink and polymer solution were mixed together followed by
heating for 15 min in a boiling water bath prior painting the ink. The membrane used was
Fumapem® FAA-2 anion exchange membrane by FuMa-Tech. The membrane electrode assembly
(MEA) was made by spraying catalyst ink onto both sides of the membrane. After painting, the
MEA was dried in vacuum oven at 60 °C for 60 min and subsequently hot pressed for 120 s at 30
5

MPa and 80 °C. The MEA, with active electrode area of 7.29 cm2 and catalyst loading of 0.5 mg
cm-2 at both electrodes, was sandwiched between gas diffusion layers and polytetrafluoroethylene
gaskets backed up by graphite blocks with serpentine flow field channels. Finally the cell was
clamped together with eight bolts that were tightened with a torque of 10 Nm.
The cell was stabilized overnight both before starting the measurements and after changing the fuel
concentration. In order to measure the difference between Pt and PtRu catalysts at the anode, the
cell performance was analyzed with both catalysts. The hydrogen production rates at various current
densities were analyzed by measuring the gas flow volume from the cathode outlet, and the
performance of the methanol electrolysis was analyzed at different temperatures and methanol
concentrations. All the potential sweeps were performed with a low sweep rate, 0.5 mV s-1, to
ensure stable conditions through the potential scale.
4. Results and discussion
Methanol oxidation has shown enhanced activity on PtRu alloy catalysts [11, 19, 20] and therefore
the performance of the electrolyser with a 1 mol dm-3 methanol solution was studied with Pt and
PtRu catalysts at 30 °C temperatures (Fig. 2a). In low potential region (<1.2 V) the PtRu catalyst
provides a higher performance in comparison to Pt due to a higher resistance to CO poisoning as
CO oxidation from catalyst surface occurs at lower potentials on the PtRu surface than on pure Pt
[21]. However, at high potentials (>1.2 V) the obtained current with the Pt catalyst approaches the
current observed with PtRu because at these potentials water also dissociates on pure Pt sites
enabling CO oxidation and moreover the density of active sites towards methanol oxidation is a
higher in the pure Pt catalyst. Nevertheless, for real applications the lower potential region is of
interest and therefore PtRu was selected as the catalyst for further studies.
6

The dependency of the hydrogen production rate on current density has been studied in order to
verify the efficiency of the electrolysis (Fig. 2b). The almost linear hydrogen production rate, which
is close to a theoretical amount, shows that the electrolytic cell performs with a high efficiency and
reached values higher than 90 %.
Electrolysers are preferred to be operated with a concentrated fuel in order to minimize the required
fuel volume and therefore the effect of methanol concentration was studied at 30 °C (Fig. 2c). Fuels
with a concentration of 1 and 2 mol dm-3 perform very similarly but with concentrations of 5 and 10
mol dm-3 the performance decreases. This phenomenon results from the changes in the membrane
morphology and properties as it has been reported [22] that the membrane swells and the ion
conductivity decreases when methanol concentration increases. In addition, the self-diffusion
coefficient of water in the membrane is lower with 10 mol dm-3 methanol than in the studied
solutions with lower concentrations. These changes explain the differences obtained in the
electrolysis with varying methanol concentrations. Moreover, methanol cross-over increases with
higher concentrations and although methanol does not react at the cathode, the cross-over increases
fuel consumption and methanol can vaporize as an impurity in the produced hydrogen.
The effect of concentration has been studied also at higher temperatures and similar kinds of
behavior with slight differences were obtained: 1 and 2 mol dm-3 fuels showed comparable
performance at all the temperatures. At 50 °C and 70 °C there is an evident drop in the performance
with the 5 and 10 mol dm-3 methanol solutions compared to the fuels with low concentration, even
though at 30 °C only slightly weaker performance was obtained with the high concentration fuels,
and currents obtained with the 5 mol dm-3 methanol even exceeded those with 1 mol dm-3 methanol
7

at high voltages. The morphology of the membrane is assumed to change at the elevated
temperatures resulting in changes in transportation which explains the obtained expanding gap
between various concentrations when temperature is increased.
Polarization of the alkaline membrane electrolyser has been studied at temperatures of 30 °C, 50 °C
and 70 °C with 1 mol dm-3 methanol as a fuel and the results are shown in Fig. 2d. It can be clearly
seen that the current densities rise with increasing temperature. At 1 V potential the currents are
10.1 mA cm-2, 24.0 mA cm-2 and 48.9 mA cm-2 at temperatures of 30 °C, 50 °C and 70 °C,
respectively. This phenomenon is similar to that observed with direct methanol fuel cells [23] as the
increased temperature enhances mass transport and reaction kinetics resulting in reduced activation
and concentration overpotentials. This behavior has been reported previously for methanol
electrolysers utilizing acidic membranes [1, 4]. In addition, it can be seen that the onset potentials
decreases with increasing temperature and the onset potentials are approximately 0.72 V, 0.70 V and
0.65 V at temperatures of 30 °C, 50 °C and 70 °C, respectively. These values are higher than
reported for acidic membrane electrolysers [1, 3, 4] and this reflects the different methanol
oxidation reaction mechanisms and kinetics in alkaline and acidic environment.
5. Conclusion
The concept of methanol-water solution electrolysis in a fuel cell equipped with an alkaline
membrane for hydrogen production is introduced, offering the possibility for lower power
consumption in hydrogen production when compared to traditional water electrolysis. The alkaline
membrane electrolyser is an attractive alternative to the conventional electrolysers with proton
8

exchange membranes enabling the utilization of low-cost membrane materials offering savings with
material expenses.
PtRu was found to be a better catalyst than Pt for methanol oxidation in alkaline electrolysis cell in
low potential region (<1.2 V). The system is capable to operate with a high efficiency producing
almost the theoretical amount of hydrogen and the efficiency increases with increasing current
which is essential for real applications.
Increasing methanol concentration resulted in decreased performance in the electrolysis due to
morphological changes in the membrane structure resulting in a decreased ion conductivity and
water self-diffusion coefficient. The results indicate that the membrane requires improvements
before operating the cell with a highly concentrated fuel to minimize the needed fuel volume for
practical applications.
This novel method opens a new area for investigation in hydrogen production and enables
alternative material choices to produce commercially viable methanol electrolysers. Non-precious
metal catalysts are a very attractive option which may be utilized in this system.
Acknowledgments
9

The authors thank the Starting Grant of Aalto University, Multidisciplinary Institute of
Digitalisation and Energy (Aalto University), Academy of Finland for funding the research and Dr.
Ben Wilson for proofreading this paper.
References
[1] S. Uhm, H. Jeon, T.J. Kim, J. Lee, J. Power Sources 198 (2012) 218-222.
[2] T. Take, K. Tsurutani, M. Umeda, J. Power Sources 164 (2007) 9-16.
[3] G. Sasikumar, A. Muthumeenal, S.S. Pethaiah, N. Nachiappan, R. Balaji, Int. J. Hydrogen
Energy 33 (2008) 5905-5910.
[4] Z. Hu, M. Wu, Z. Wei, S. Song, P.K. Shen, J. Power Sources 166 (2007) 458-461.
[5] C.R. Cloutier, D.P. Wilkinson, Int. J. Hydrogen Energy 35 (2010) 3967-3984.
[6] E. Rasten, G. Hagen, R. Tunold, Electrochim. Acta 48 (2003) 3945-3952.
[7] D.L. Stojić, M.P. Marčeta, S.P. Sovilj, Š.S. Miljanić, J. Power Sources 118 (2003) 315-319.
[8] S.A. Grigor’ev, V.I. Porembskii, Chem. Petrol. Eng. 40 (2004) 606-610.
[9] S.R. Narayanan, W. Chun, B. Jeffries-Nakamura, T.I. Valdez, US Patent 6,299,744 (2001).
[10] J. Prabhuram, R. Manoharan, J. Power Sources 74 (1998) 54-61.
[11] A.V. Tripković, K.D. Popović, B.N. Grgur, B. Blizanac, P.N. Ross, N.M. Marković,
Electrochim. Acta 47 (2002) 3707-3714.
[12] C. Xu, Y. Hu, J. Rong, S.P. Jiang, Y. Liu, Electrochem. Commun. 9 (2007) 2009-2012.
[13] M.A. Abdel Rahim, R.M. Abdel Hameed, M.W. Khalil, J. Power Sources 134 (2004) 160-169.
[14] M. Jafarian, R.B. Moghaddam, M.G. Mahjani, F. Gobal, J. Appl. Electrochem. 36 (2006) 913-
918.
10

[15] H. Deligöz, S. Yılmaztürk, T. Karaca, H. Özdemir, S.N. Koç, F. Öksüzömer, A. Durmuş, M.A.
Gürkaynak, J. Membr. Sci. 326 (2009) 643-649.
[16] L. Jörissen, V. Gogel, J. Kerres, J. Garche, J. Power Sources 105 (2002) 267-273.
[17] W.C. Choi, J.D. Kim, S.I. Woo, J. Power Sources 96 (2001) 411-414.
[18] A. Santasalo-Aarnio, P. Peljo, E. Aspberg, K. Kontturi, T. Kallio, ECS Trans. 33 (2010) 1701-
1714.
[19] S. Sen, F. Sen, G. Gokagac, Phys. Chem. Chem. Phys. 13 (2011) 6784-6792.
[20] V. Selvaraj, M. Alagar, Electrochem. Commun. 9 (2007) 1145-1153.
[21] J.S. Spendelow, G.Q. Lu, P.J.A. Kenis, A. Wieckowski, J. Electroanal. Chem. 568 (2004) 215-
224.
[22] A. Santasalo-Aarnio, S. Hietala, T. Rauhala, T. Kallio, J. Power Sources 196 (2011) 6153-6159.
[23] K. Scott, W.M. Taama, S. Kramer, P. Argyropoulos, K. Sundmacher, Electrochim. Acta 45
(1999) 945-957.
11

Figure 1. Schematic of methanol-water solution electrolysis with an alkaline membrane.
Figure 2. Comparison of Pt and PtRu anode catalysts at 30 °C with 1 mol dm-3 methanol (a),
hydrogen production rate dependency of the current density (b), the effect of fuel concentration at
30 °C (c) and the effect of temperature with 1 mol dm-3 methanol (d).
12