Hydrogen generation under electrolysis process using Ni diffusion

catalytic membranes as electrodes

GENNADIY GLAZUNOV, ALEKSEY KONOTOPSKIY, DMITRIY ELISIEIEV,

IGOR GARKUSHA, SERGEY MAZNICHENKO

National Academy of Science of Ukraine,

National Science Center “Kharkov Institute of Physics and Technology”,

Institute of Plasma Physics,

61108 Kharkiv, Academichna str. 1

UKRAINE

Abstract: The work presents a setup scheme for studying the processes of hydrogen generation and analysis of

its penetration through diffusion-catalytic membranes directly during the electrolysis process, as well as first

experimental results obtained when testing this setup (volt-ampere characteristics, mass spectra of the mixtures

of gases, generated during electrolysis, etc.).

Key-Words: Electrolysis, nickel, diffusion catalytic membrane, hydrogen generation, mass-spectrum.

Received: November 9, 2022. Revised: September 19, 2023. Accepted: October 21, 2023. Published: November 22, 2023.

1 Introduction

In work [1] in order to obtain especially pure

hydrogen under the electrolysis process, it was

proposed to use special cathode designs. For

instance, it could be beneficial to use membranes in

the form of nickel tubes or other metals with high

hydrogen permeability, like palladium and its alloys.

In this case, one end of the cathode (diffusion-

catalytic membrane) is hermetically sealed by

argon-arc welding, and the other one is connected

through a valve to a vacuum volume or to a

hydrogen accumulator, e.g. similar as it was used in

works [2-3], for production of extra pure hydrogen

in the flame of combustion of hydrocarbons. Such a

device would be able to provide the possibility to

obtain ultra-pure hydrogen (with a purity of more

than 99.999% vol.) through the electrolysis along

with the production of technical hydrogen or a

combustible mixture (explosive gas, Brown's gas).

To prove the scientific concept for such technology

realisation, special installation was created and

experimentally tested. Presented below first results

experiments confirm the potential advantages of

such approach aiming at generation of pure

hydrogen to be used in various applications.

2 Experimental and Results

A schematic diagram of the experimental setup is

shown in Fig. 1. Experimental installation consists

of two chambers by 1 dm3 volume of each that made

of thick-walled stainless steel 12Cr18Ni10Ti. One

of them is served as electrolysis chamber (1) and the

other is used for measurements (2). In the first

chamber, a diffusion-catalytic membrane made of

Ni-99.95 (4(1)) is placed through a fluoroplastic

seal-insulator (3). It forms a tube with the diameter

of 5.5 mm, the thickness of 0.2 mm and overall

length of 400 mm, which is hermetically sealed by

argon-arc welding at one end. Opposite end of the

tube 4(1) is connected through the valve (5) to the

vacuum volume of chamber (2) for measuring the

pure hydrogen flow. Chamber (2) provides the

technical ability to place various samples for

studying the hydrogen permeability, including in-

situ measurements under conditions of exposure to

glow discharge plasma. For temperature

measurements inside the membranes, a chromel-

copel thermocouple (9) has been installed. The

pressure in the electrolysis chamber (1) and the

vacuum in chamber (2) are measured by pressure

gauge (10) and thermocouple gauge PMT-2 (8)

respectively. An electric potential can be supplied to

International Journal on Applied Physics and Engineering

DOI: 10.37394/232030.2023.2.17

Gennadiy Glazunov, Aleksey Konotopskiy,

Dmitriy Elisieiev, Igor Garkusha, Sergey Maznichenko

E-ISSN: 2945-0489

194

Volume 2, 2023

both membranes: in chamber (1) for electrolysis

process, and in chamber (2) for igniting a glow

discharge. This design of the experimental

installation allows not only studying the processes

during electrolysis and the hydrogen permeability of

diffusion-catalytic membranes, but also allows

investigations on the possibility of decomposition of

water vapour and electrolysis products under the

plasma discharge conditions.

Fig.1. Scheme of the experimental electrolysis

installation: 1 – electrolysis chamber (E-1); 2-

measuring (plasma) chamber; 3- insulators; 4(1) and

4(2) – Ni membranes; 5 – valves; 6 – fore vacuum

pump 2PVR-5D; 7, 8 – vacuum gauges PMT-2; 9-

chromel-copel thermocouples; 10 - pressure gauge;

11 – replaceable anode.

The electrolysis chamber (E-1) is connected through

a valve (not shown in the diagram) to the inlet

system of the high-vacuum installation GAS (Fig.

2), which allows, using the mass – spectrometry to

determine the composition of the mixture of gases

formed during the electrolysis process [4,5].

Hydrogen is produced by electrolysis in following

way. At first, the internal volumes of the membrane

4(1), the measuring chamber (2) and the connecting

pipes were pumped out to the pressure of about 10-3

Torr. This pressure is measured by vacuum gauge

(8) and it is considered as initial pressure P0. Then

the electrolysis process is started in chamber (1),

shown in Fig. 1, by switching on a constant voltage

between anode and cathode. The voltage source (0-

200V) was a rectifier with four D132-80 diodes

connected through a current-limiting resistor. The

electrolyte used was 1% solution of NaHCO3 (soda)

in filtered tap water. Mineralization measured using

a TDS Meter-3 device has achieved 140 ppm. The

current-voltage characteristic of the electrolysis

process is presented in Fig.3. The temperature of the

electrolyte (cathode) increases monotonically over

time, reaching 50°C after half hour of operation of

the electrolysis chamber (Fig. 4).

Fig.2. Scheme of the GAS experimental setup: 1 -

vacuum chamber; 2- cryogenic pump; 3- mass

spectrometer MX-7304; 4- sample; 5- diffusion

pump M-500; 6 - fore-vacuum pumps 3NVR-1D; 7-

fine adjustment valve; 8, 9, 10 - high-vacuum

valves; 11-hydrogen balloon, 12-helium balloon; 13

- gas reducers; 14, 15, 16 - high pressure valves; 17

– ionization gauge PMI-10; , 18, 19 – thermocouple

gauges PMT-2; 20 – ionization gauge PMI-2; 21-

pressure gauge.

Fig.3. Current-voltage characteristic at electrolysis

of a solution of soda in water (1 wt.%).

At the same time, the current increases up to 4.3-

4.6A, and the voltage between the electrodes

decreases from 10V to - 7.3V. Gas pressure in

electrolysis chamber reached 4 atmospheres (free

volume in electrolysis chamber is about 0.5 dm3, so

we have ~ 2 dm3 of gas at the normal conditions). It

is quite enough to carry out precise measurements of

mass-spectra.

International Journal on Applied Physics and Engineering

DOI: 10.37394/232030.2023.2.17

Gennadiy Glazunov, Aleksey Konotopskiy,

Dmitriy Elisieiev, Igor Garkusha, Sergey Maznichenko

E-ISSN: 2945-0489

195

Volume 2, 2023

Fig.4. Dependence of the membrane temperature on

the time of electrolysis

To determine the composition of the gas mixture,

formed during electrolysis, vacuum chamber (1) of

the GAS installation was pumped out to a pressure

of 1-2·10-6 Torr. Through the valve 5 (Fig. 1), and

through valves 14, 7 (Fig. 2) the gas, generated

during the electrolysis process, has been released

into vacuum chamber (1) of the GAS device to a

pressure of 0.1-1∙10-4 Torr. The mass spectra

measured with MX-7304 mass spectrometer and

recorded, using the WAD-AIK-BUS analogue

module and a computer. Figure 5 (a, b) shows the

obtained spectra before and during the puffing of

gas generated under electrolysis into the vacuum

chamber. In this case, the pressure in the electrolysis

chamber was about 4 atm, and the electrolyte

temperature was ≈45°C. For comparison, the

spectrum is shown in Fig. 5c when pure hydrogen

(99.99% by vol.) is injected from a balloon.

The next experiment was carried out to test the

device in regime of pure hydrogen generation. The

hydrogen flow through a diffusion-catalytic nickel

membrane has been measured by following

procedure. Hydrogen, which is formed on the Ni-

membrane surface facing to the electrolyte, absorbs,

diffuses through the membrane volume, and then

desorbs to the internal volume of the membrane.

Only hydrogen is able pass through the diffusion-

catalytic nickel membrane, which separates the

hydrogen from other gases with larger mass. Thus,

the generated hydrogen, as confirmed by our

measurements, has a high purity – better than

99.999 % vol. As soon as hydrogen desorbs, the

pressure P in the vacuum chamber (2) increases.

The measurements of hydrogen flow through Ni

membrane (productivity) were carried out with the

method of constant pressure, similar to the described

previously [6, 7].

Fig.5. Mass spectrum of gases in the GAS chamber:

(a) before inlet, (b) during inlet of gas from the

electrolysis chamber (E-1); (c) hydrogen puffing

from hydrogen balloon.

The hydrogen flow Q in our case corresponds to the

hydrogen amount generated per time unit. If the

membrane surface area is taken into account, one

possible to obtain the specific hydrogen flow –

hydrogen amount generated per time unit from

membrane surface area unit. We used the following

units: normal cm3 (cm3 of gas at the atmosphere

pressure and at the room temperature) of hydrogen

per second (Ncm3/s) or liters of hydrogen gas at the

International Journal on Applied Physics and Engineering

DOI: 10.37394/232030.2023.2.17

Gennadiy Glazunov, Aleksey Konotopskiy,

Dmitriy Elisieiev, Igor Garkusha, Sergey Maznichenko

E-ISSN: 2945-0489

196

Volume 2, 2023

atmosphere pressure per hour (l/hour). The specific

hydrogen flow q(N.cm3/s.cm2)=Q/F, where F is the

membrane work surface area (faced to electrolyte),

was used, too. For technical reasons, the cathode

temperature not monitored in this experiment, but

the temperature of the chamber wall, measured by a

thermocouple, was ~170°C. According to the

estimates, the cathode temperature could reach 200-

300°C. The measured pressure in the electrolysis

chamber was about 16 atm. and the measured

pressure in the vacuum chamber (2) was 3∙10-3

Torr. The hydrogen flow through Ni membrane to

vacuum chamber is:

Q = 1.3·(P-P0)·S,

where P0 =2∙10-3 Torr is the initial pressure, P =

3∙10-3 (Torr) is the measured final pressure, S= 5

l/s is the pumping speed. So, pure hydrogen flow

through Ni membrane is Q≈ 0.006 Ncm3/s or

about 21 Ncm3/h.

3 Discussion

From Fig. 3 it is follows that the electrolyte

temperature after 50 minutes of the process reaches

60°C, at which the transformation of Na bicarbonate

into soda ash begins:

2NaHCO3 = Na2CO3+CO2+H2O.

Therefore, measurements of the spectral

composition of the generated gas were carried out at

temperatures below 50°C. Processing of the

instrumental spectra curves shown in Fig. 5 allows

conclusion that in our case the gas generated during

electrolysis consists of ~94% hydrogen (Fig. 5b). It

has been taken into account that the sensitivity of

the mass spectrometer for hydrogen is

approximately two times lower than for air. The

spectrum of the generated gas differs from the

spectra of pure hydrogen released from a balloon

(Fig. 5c) and initial one in vacuum chamber of GAS

device by the noticeable increase of water (≈3%

vol.), oxygen and CO (≈1.5 % vol. each). These

results are very different from the available

literature data on water electrolysis (reaction

equation 2Н2О=2Н2+О2: 66% H2+34% O2). To

clarify the physicochemical reasons for this,

additional investigations are necessary; which we

are going to carry out in the nearest future. Here we

primarily consider why such low values registered

in the testing experiments on measuring the pure

hydrogen flux through Ni diffusion-catalytic

membrane.

One need to stress, that penetration of hydrogen

through the wall of Ni cathode (membrane) is a

complex process consisting of a number of

sequential reactions: dissolution (absorption) in the

metal, including hydrogen molecules dissociation

and ionization, diffusion of atoms and ions in the

metal volume, release to the surface of the back side

of the membrane, hydrogen atoms recombination,

desorption of hydrogen molecules. In contrast to H2

permeation from the gas phase, in the case of

electrolysis, hydrogen will reach the membrane

surface in various states: H2+, H+, H. Ions, atoms and

charged hydrogen molecules can easily overcome

the potential barrier at the surface-metal interface.

Therefore, the surface processes cannot be the

limiting stage of the process in this case. Moreover,

we expect that under such conditions the near-

surface layer of the metal can be saturated with

hydrogen to high concentrations. Diffusion in the

volume of the metal, according to first Fick’s law, is

proportional to D(n0-n)/d, where n0 is the hydrogen

concentration on the inlet side of the membrane, n is

the one on the other side, D is the diffusion

coefficient. Since in our case n0>>n, and the

thickness of the membrane is small (d=0.2mm), we

can assume that in our case K ~ n/d~P0.5/d, where K

is the permeability coefficient, P is the hydrogen

pressure on the inlet side of membranes. On the

other hand, K=S∙D, where S is the solubility

coefficient of hydrogen in nickel. Both coefficients

depend exponentially on temperature, i.e. К=К0∙е-

Е/kT, where E= Ed +Es is the activation energy of

hydrogen permeability (Ed is the activation energy

of diffusion, Es is the activation energy of

solubility). It can be assumed that it is the

temperature dependence of permeability that is the

limiting factor. Indeed, according to literature data,

to obtain significant hydrogen flows through metals,

temperatures above 350°C are required [4]. So, in

order to obtain essential hydrogen flows through Ni

membrane it is necessary to increase its temperature,

at least, up to 400°C.

4 Summary

We have designed, manufactured and tested the

combined experimental installation, which allows

not only studying the processes during electrolysis

and the hydrogen permeability of diffusion-catalytic

membranes (cathodes), but also research on the

possibility of decomposition of water vapour and

electrolysis products under the impact of plasma

discharges.

International Journal on Applied Physics and Engineering

DOI: 10.37394/232030.2023.2.17

Gennadiy Glazunov, Aleksey Konotopskiy,

Dmitriy Elisieiev, Igor Garkusha, Sergey Maznichenko

E-ISSN: 2945-0489

197

Volume 2, 2023

The current-voltage characteristics of the

electrolysis process and the spectra of the generated

gas have been measured. It is shown that at

temperature of H2O + 1% NaHCO3 electrolyte of

45°C and a pressure in the electrolysis chamber of 4

atm, the generated gas is 94% vol. consists of

hydrogen, 3% H2O and 1.5% each of CO and

oxygen. These results appeared to be different from

the available literature data on water electrolysis

(reaction equation 2Н2О=2Н2+О2: 66% H2+34%

O2). To clarify the physicochemical reasons for this,

additional investigations are required; which are

going to be performed by our team in the nearest

future.

The diffusion flux of pure hydrogen through a

nickel membrane (cathode) has been measured. It is

shown that for the pressure in the electrolysis

chamber of 16 atm. and electrolyte temperature of ~

200-300°C, a pure hydrogen flow Q≈ 0.006 Ncm3/s

or 21 Ncm3/h is achieved. To further increase the

flow significantly, it is necessary to grow up the

cathode temperature, at least, up to 400°C.

Further upgrade of the experimental device is

planned within the next step in this research to reach

the required parameters, in particular, improvement

of electrolysis chamber in order to increase pressure

and temperature parameters and, as sequence, to

increase pure hydrogen flow through the Ni

diffusion-catalytic membrane. For the same purpose

it is planned to test other materials for producing of

diffusion-catalytic membranes (cathodes) and

anodes.

References:

[1] G.P. Glazunov, A.S. Bodnar, A.L. Konotopskiy,

D.M. Vinogradov, S.M. Maznichenko and I.E.

Garkusha. Erosion behavior of nickel diffusion

membranes in conditions of water electrolysis.

Problems of Atomic Science and Technology.

Series: Plasma Physics (28), No 6 (142). 2022, pp.

111-113

[2] Glazunov G.P. Method of Generating Ultra Pure

Hydrogen. Patent of Ukraine № 86884. Bulletin No

10, 2009

[3] G.P. Glazunov, A.L. Konotopskiy, D.M.

Vinogradov, M.N. Bondarenko, S.M. Maznichenko,

O.P. Svynarenko. Investigation of some factors

influence on the work of extra pure hydrogen

generator. Problems of Atomic Science and

Technology. Series “Plasma Physics”. No 1 (107),

2017, pp. 109-111.

[4] G.P. Glazunov, D.I. Baron, V.E. Moiseenko,

M.N. Bondarenko, A.L. Konotopskiy, A.V. Lozin,

A.I. Lyssoivan, T. Wauters and I.E. Garkusha.

Characterization of wall conditions in Uragan-2M

stellarator using stainless steel thermal desorption

probe. Fusion Engineering and Design, Vol.137,

2018, pp.196-201.

[5] G.P. Glazunov, A.A. Andreev, D.I. Baron, V.M.

Shulaev, V.A. Stolbovoy, V.Ya. Chernyshenko. The

influence of the method of applying vacuum-arc

TiN coatings on their outgassing in a vacuum at

high temperatures. Physical engineering of the

surface. Vol. 7, No. 4, 2009, pp. 341-346.

[6] G.P.Glazunov, E.D.Volkov, D.I.Baron. Study of

low hydrogen flows into high-vacuum systems. Int.

J. Hydrogen Energy, Vol.24, 1999, pp.829-831.

[7] G.P.Glazunov. Hydrogen permeation through

palladium after hydrogen-thermal treatment and

exposure to a plasma of glow discharge in

hydrogen, helium and argon, Int. J. Hydrogen

Energy. Vol. 22, No. 2/3. 1997. pp. 263-268.

Contribution of Individual Authors to the

Creation of a Scientific Article:

Gennadiy Glazunov: Conceprtualization,

Metodology, Investigation, Writing&Editing.

Aleksey Konotopskiy: Metodology, Investigation.

Dmitriy Yelisieiev: Technical support for

conducting experiment. Igor Garkusha: Project

administration, data analysis, editing. Sergey

Maznichenko: Technical support for the experiment.

No additional funding was received for conducting

this study.

The authors have no conflicts of interest to declare

that are relevant to the content of this article.

This article is published under the terms of the

Creative Commons Attribution License 4.0

https://creativecommons.org/licenses/by/4.0/deed.en

_US

Sources of Funding for Research Presented in a

Scientific Article or Scientific Article Itself

Conflict of Interest

Creative Commons Attribution License 4.0

(Attribution 4.0 International, CC BY 4.0)

International Journal on Applied Physics and Engineering

DOI: 10.37394/232030.2023.2.17

Gennadiy Glazunov, Aleksey Konotopskiy,

Dmitriy Elisieiev, Igor Garkusha, Sergey Maznichenko

E-ISSN: 2945-0489

198

Volume 2, 2023