Parametric study of solar hydrogen production from saline ...Keywords: electrolysis of water, hydrogen production, solar energy conversion. Reference to this paper should be made as

164 Int. J. Environment and Pollution Vol. 8, Nos,l/2. 1997

Parametric study of solar hydrogen production fromsaline water electrolysis

S.M. El-HaggarEngineering Department, The American University in Cairo, PO Box 251l,Cairo, Egypt

M. Khali lAerospace Department, Faculty of Engineering, Cairo University, Giza,Egypt

Abstract: The purpose of this work is to study the electrolysis of water for theproduction of hydrogen. A number of parameters, including salinity, voltage,cunent density and quantity of electricity, were investigated, and their effect onhydrogen production using a modified simple Hoffman electrolysis cell isreported.

Keywords: electrolysis of water, hydrogen production, solar energy conversion.

Reference to this paper should be made as follows: El-Haggar, S.M. andKhalil, M. (1997) 'Parametric study for solar hydrogen production from salinewater electrolysis', Int. J. Environment and Pollution, Vol. 8, Nos. 1/2,pp.164-173.

1 Introduction

In view of the global warming problem the world is facing today, strategies fordeveloping economical solar energy converters have emerged, targeted at significantlyreducing the use of fossil fuels. Considering that fluctuating regenerative energies, suchas wind and solar power, constitute over 20Vo of the total electricity generated, it hasbecome necessary to use any surplus electricity produced by such fluctuations. Oneoption would be to use the surplus energy to produce hydrogen to be used as a fuel;another would be to install large-scale storage devices or battery systems to store theexcess energy.l

The heating value of hydrogen is 141.65 MJ/kg (or 12.65 MJlm3). On basis of mass,this heating value of hydrogen is the highest among known fuels, as shown in Figure 1.Unlike fuels containing carbon (petroleum and coal), hydrogen burns cleanly leaving nosmoke and no ash, which are typical by-products of carbon combustion.

There are several methods for the production of hydrogen using solar energy.Electrolysis, direct thermal, thermochemical, and photoelectrochemical methods are afew common methods for splitting water molecules into hydrogen and oxygen. A briefreview has been published by Ohta and Veziroglu,2 in which these methods are discussedin some detail. In addition to these direct means, these are indirect methods that use solarenergy in generating electrical power, which in turn is used to electrolyse water to evolvehydrogen.3

Copyright @ 1997 Inderscience Enterprises Ltd.

Solar hydrogen production from saline water electrolysis

1601{1.65

14

1n't@

80

60

40

n0

5'l.E 56.1714.1 12.7

t65t9.

-

;r

-llFidrogen %

I Fleating Vdue lt/lJ/kg

15.25 15.75i r 6 r

s P e e e E E E Eg € # E 8 H ; E Ei ; u r i g F 3 5

Figure I Hydrogen content and heating value of fuel.

€ : * *q = s o= c

E

s fornn ljgEf

t hrsOnr;

fir€lE [h€

llass.

r e l

r8 n,o

crE]tre ebriefrss€Insolanr0l\'e

Alkaline water electrolysis is the technology curently in use for large-scale electrolytic

hydrogen production. Low efficiency, low current density and a lack of proper scale-up

piactice arg the primary drawbacks of the present technology' Significant improvements

have been made, making it possible to reach improved cell efficiencies and higher current

densitiesa. Many advanced concepts relating to various aspects of electrolytic hydrogen

production are reported in the literature.s-7

Alkaline water electrolysis uses fresh water with low salt content, and hence

additional treatment and desalination systems add to the cost of hydrogen produced. The

cell for fresh water electrolysis is known as the H2lO2 cell. Hydrogen and oxygen are

produced in the ratio of 2:1.

This paper describes the electrolysis of saline water of varying TDS (total dissolved

solids) content for hydrogen production'

2 Water electrolysis

2.1 Energy of electrolysis of water

In general, hydrogen can be generated by electrolysing saline water or caustic soda. The

electrochemical reaction at the cathode is

2H2O +2e- -----+2OH- +lI2

where e- denotes an electfon. At the anode, oxygen is generated by the reaction

2OH------+H2O+ \O2+2d

Equations 1 and 2 lead to the overall reaction

H2O -> H2+ \O2

The equilibrium potential at the cathode, E", and at the anode, Eu, are represented as

( l )

(2)

(3)

L66 S.M. El-Haggar and M. Khalil

E^ = E2 * RT ln ^otr'o2F a"o^ - a,,

and

E^=EA *! !6oUl:" 'oa v2 2F o6r-

(4)

(5)

respectively, where Eo is the normal electrode potential, F the Faraday constant, R the gasconstant, and athe activity.

At25 "C, E!r, and E$, are -0.828 V and +0.401 V, respectively. If the activities ofwater in the electrolyte, hydrogen and oxygen are all taken as unity (the pressures ofhydrogen and oxygen are 1 atm), Equations 4 and 5 (at25'C) can be written as:

E" = -O. 828 - 0.059 log ao"-

and

E" = 0.40t - 0.059logao"- Q)

respectively. Thus Eu - 8", the lowest voltage needed to drive reactions I and 2 (that is,the theoreticaf decomposition voltage of water at 25 oC) is evaluated to be 1.299 Y,irrespective of pH value of the electrolyte. A plot of Eu and E. against pH is shown inFieure 2.- pot€nr ia l (V)

r.229 Potcrural ( V )

0.401

I A

-0.828

Figure 2 Hydrogen and oxygen electrode potentials against pH of electrolyte solution at25 oC.

The smallest amount of electricity required to produce one mole of hydrogen and half amole of oxygen from one mole of water is 236.96 kJ, whereas the amount of heatgenerated by the combustion of one mole of hydrogen is 285.58 H at 25 oC. So theenergy efficiency ofelectrolytic hydrogen generation, 4gr, is given by

heat of combustion of hydrogen17", =

electrical energy required to produce one mole of hydrogen

x100= 285'58 x 100=120.5Vo236.96

(6)

Solar hydrogen productionfrom saline warcr electrolysis 167

This means that 2g5.5g - 236.96 = 4g.62 kJ must be absorbed from the surroundings ofelectrolvsers if water is electroryse a witn | .zlg v at lllg- innirr* eiectricat energy of185.58 kJ, that is 1.481_V, to a ware, "t""t otyr",

?r?: "Cwould generate hydrogen andoxygen isothermary. The values 1.229v uno r.+arv u." "uri"J th" ."u"..ible orequilibrium voltage and the thermoneutrat uott ge at 25 oC,respectively.

These voltagesvary with temperature as shown in Figure 3.

1 . 7

cellpotentialfl/olts) r.l

t ,0

Figure 3 Idealized operation condition for water electrolysers.

Three areas can be identified in Figure 3:1 Hydrogen generation is impossible;

2 Hydrogen is evolved as erectrical efficiency greater than r00vo:3 Hydrogen is made with production of waste heat.These areas show, however, merely the theoretical limitation of water electrolysis, and nohydrogen is evolved at the reversible vottage lecause the ** oi"i""oorysis is actuallynull and the activity of water in the elect"rolyte is not exactry unity. To promote thedecomposition of water, an excess voltage must be applied.

2.2 Electrolytes

In general, saline water (or an aqueous solution of soda) is used as the electrolyte forwater electrolysis. The electrolyte must have high purity. 'cmo.ro",

Ld sulfates must beexcluded completely because of their corrosive-action on tn"-"r""troa"s, especiallyanodes. Though the conductivity of the electroryte and the energy efriciency of theprocess increase as the temperature increases, present-day water electrorysis cells areusually operated at 50-80 'c, in order to ,"du"e the consumption or "i".t otyser material.As the water is erectrolysed, *ut"-uf *uter must be added. Any non-volatileimpurities present in the added water remairias contaminants in the alkaline solution and,in order to preserve the cell operating characteristics, water of high purity must be added.Therefore, high quality waterpurification plants are necessary.

Electicrty otd tonate hydrog:rLWest€ heat solved

Heat gnd electicrBured to mqt a lq/dpgEtl

Hydrojrnrmpodble

SEneiauon

168 S.M. Et-Haggar and M. Khalit

2.3 Electrode materials

Because iron has a superior corrosion resistivity to alkaline solutions and a low hydrogenovervoltage, it is recommended for use as the cathode material in water electrolysis. Thehydrogen overvoltage depends on the treatment of the electrode surface, and variousmethods of surface treatment have been proposed.In general, electrodes of nickel-plated iron sheet or wire mesh, the surface of which isactivated and/or expanded in various ways, e.g. by incorporation of surfur compounds orzinc compounds, which are removed on commissioning by the action of the carsfir- sn.ro

are good for both cathodes and anodes.

2.4 Diaphragms

In order to maintain separation between the hydrogen generated at the cathode and theoxygen evolved at the adjacent anode, a diaphiagm is inserted between the twoelectrodes' The diaphragm must be stable in caustic alkaline solution and must minimizediffusion of the product gases without impeding ionic conductivity. To prevent thepassage of gas bubbles, the diaphragm must consisi of small pores, the capillary pressureof which is greater than the maximum differential pressure apptied across the cell. Thediaphragm must be wgtted by the electrolyte, as otherwise the gases w1r collectpreferentially in pores, leading to an increase in resistance and eventually to passage ofgas through the diaphragm. A woven asbestos cloth, an artificial fibre cloth, or a rubbercloth on a metallic net is usually employed as a diaphragm material.3

2.5 Modes of water electrolysis

commercial electrolysers are generally classified into two types, unipolar and bipolar.The voltage input required per celr is generally between 1.9 a; | .2 y , andthe gases aregenerated at current densities between 100 and 200 mAlcm2.

2.5.1 Unipolar cellsAs shown in Figure 4, this type of electrolyser has the cathode and anode separated by adiaphragm and both sides of each electrode have the same polarity.

ccllrrt4: = 2Y X nrJcr dcrlL

Figure 4 Unipolar cell construction.

hogems- Tberious

tich isds orcna "

d thett*'o

nizeI thesureThellectp o fbber

Nar"are

Solar hydrogen production from saline water electrolysis r69

The electrolyser has a number of electrodes connected in parallel, and alternateelectrodes, usually the cathodes, are surrounded by a diaphragm. This is called a tank-rype or parallel-type electrolyser, operating at high current and low voltage, which resultsin an awkward electrical rectification problem. The individual cells are usually connectedin series to raise the voltage of the system. Tank-type units are, however, wasteful offloor space. Nevertheless, a current efficiency of 100Vo is easily attained. This type ofelectrolyser has a life of longer than 25 years and is maintenance free for 10 years.Moreover, relatively few parts are required and individual cells may be isolated for repairor replacement by short-circuiting the two adjacent cells.

2.5.2 Bipolar cellsThis type of electrolyser contains individual electrodes separated by insulators. One sideofthe electrode serves as the cathode ofone cell, and the other side serves as the anode ofthe adjacent cell. That is, the electrolyser is constructed with alternate layers ofelectrodesand diaphragm.

.As it resembles a filter press in type, it is called a filter-press-type or a series-typeelectrolyser; a typical construction is shown in Figure 5.

Cell voltage = 2Y X number of pairs ol electode

Figure 5 Bipolar cell construction.

It is usually desirable to circulate electrolyte material throughout the cells, therebyseparating the gas and the electrolyte in a separating drum mounted on top of theelectrolyser. The circulation of the electrolyte is maintained by the gas lift of thegenerated hydrogen and oxygen.

This type of electrolyser provides for pairs of 30 to several hundreds of electrodes inone electrolyser and, as individual cell voltages are additive within an electrolyser, 60-1200 volts are required. Thus, this type ofelectrolyser operates at high voltages, resultingin a lower cost of rectifier; but, because of the bypass or leakage currents in theircommon electrolyte, it has a current efficiency of about 95Vo and the bipolar constructionis usually more expensive than the unipolar system for a given rate of hydrogenproduction. Breakdowns in bipolar cells are rare, but when they occur, recommissioningmay take a long time. In Figure 6, typical voltage-+urrent graphs for bipolar and unipolarcells are shown.

r70 S.M. El-Haggar and M. Khalil

Uni-polar cell

Volts ?'1Per z.o

l .s

1 .8

1 . 7

1 .6

t . 5r00 150 200 250

Curent D ensity (mAlcnZ )

Figure 6 Performance characteristics of a unipolar cell and a bipolar cell using uncatalysedelectrodes at 80 "C.

3 Experimental

The apparatus for the electrolysis of saline water, prepared in the solar laboratory, isshown in Figure 7. It is the basic Hoffman arrangement (single cell, bipolar type) withtwo vertical gas collecting tubes. Gases produced at the cathode and anode are collected

by the displacement of saline water. Stainless steel electrodes (150 x 50x 1.0 mm

thickness) are used and held in the upper end of the cell. The solutions are prepared using

deionized water plus NaCl in the appropriate ratio.

The energy input to the cell is a direct current obtained from a solar cell at 24 volts

and about 10 amps. The voltage control circuit is constructed as shown in Figure 8, to

obtain different output voltage from the supply.

Air F

c.fr-*ntn

C*

Dt?lrTr

Figure 7 Experimental set-uP.

Ienrperatrre: SloCXlectrodes : Uncatal5rred

Figt

Fordepandfor24volcurTDpft

rel

wtvo

Ta

\

Solar hydrogen productionfrom saline water electrolysis

Figure I Voltage control circuit.

4 Results and discussions

For the solutions investigated, it was found that the hydrogen production rate wasdependent on the applied voltage and the electrode type. As can be seen from Figures 9and 10 and Table 1, steel electrodes gave a higher production rate than copper electrodesfor the same applied voltage. Over the range of voltages used in these experiments (12-

24Y),hydrogen production increases ohmically (follows Ohm's law) with the increase involtage of electrolyte. At the higher voltage, the initial conductivity no longer affects thecurrent density, and the solution behaves like a resistance, with lower resistance at higherTDS. Hence the increase the salinity gives a proportional increase in hydrogenproductivity (Figure 11). The hydrogen production rate (HPR) and the TDS value arerelated by the equation

HPR = d+bms

where HPR is in litres/tr and TDS is in g/litre. Values of a and b are given, for differentvoltages and electrodes, in Table L

Table I Conelation of hydrogen production with TDS.

Voltage/V Steel electrode Copper electrode

r71

aton'. ispe I uithmllected1.0 rnmed using

2-1 r-oltsme 8. to

1615 1{ 13 12 l l l0 9

TL 494t z 3 l 5 s ?

1 2a 1

0.1 30

0.200

0.0090.035

4.240

0.105

0.018

0.008

172 S.M. El-Haggar and M. Khalil

HPR

A

(,o

o

24Vdtlsle€l El€ctodel

1 2 Volt lsls€l E!6ctr]o.bl

24Vdt lcoFrr Elsctodel

12 Vdtlcopp€r Elocfrdel

Total Dissoh,€d Solid (IDS g/L)

Figure 9 Hydrogen production rate vs TDS at different voltages.

CunentDensity

( mA/cm )

2(D

180

160

140

120

1 @

80

60

4

m0

24\rd19..1Cl clod.l

- ' ' ' t 2 \ r o r t $ . . l c h . n * l

- 2aVdfC4F..ELed.l

-- f 2vof P.pD.. E chd.l

Total Dishrd solid TDS (9/L)

Figure 10 Current density vs TDS at different voltages.

ElectolyteResistance

(Ohm)

2500

2000

1500

10oo

5(D

00 2 0 4 0 6 0 8 0

Total Dissolved Solid TDS (g/L)

Figure 11 Electrolyte resistance vs TDS.

100 120

Solar hydrogen production from saline water electrolysis

Figure 12 shows the variation of pH with TDS of the electrolyte before and after the

electrolysis process using steel electrodes. The measurements show that the pH in the

oxygen compartrnent after electrolysis is the same as before the process. The higher pH

observed in the hydrogen compartment was due to the increased hydrogen ion

concentration.

r73

1 3

12

1 1

1 0

opH-Value -

6

-,.,/,/ 1(x)8020

-----{- Aturd€<foayds

ftytogsr

Cofl?€fiE f

-----A- Attsr€ledlatlis

Oxtgst CorrpartEd

120

T6l Dissohrcd Sdid TDS G/L)

Figure 12 pH vs TDS before and after electrolysis (1 Iitre ofhydrogen production) using steel

electrodes.

References

I Hassmann, K. and Kuhne, H.M. (1993) 'Primary energy sources for hydrogen production', lntJ. Hydrogen Energy, Vol. 18. No. 8. pp.635-640.

2 Ohta, T. and Veziroglu. T.N. (1976) 'Hydrogen production using solar radiation',Int. J.

3A

Hydrogen Energy, Vol. 1, pp.255.

Ohta, T. (1979) Solar Htdrogen Energy' Systems, Pergamon Press.

Abdel-Aal, H,K. and Hussien. I.A. (1993) 'Parametric study for saline water electrolysis. Part

I: hydrogen production'. Int. J. H-tdrogen Energy, Vol. 18, No. 6' pp.485-489.

Winter, C.J. and Nitsch. J. 0988) tudrogen as Energy Carrier, Springer' Berlin.

Dutta, S., Block, D.L. and Pon. R.L. (1990) 'Economic assessment of advanced electrolytic

hydrogen production'. Int. J. Htdrogen Energy, Vol. 15, pp.387-395.

Abdel-Aal, H.K. (1982) 'Porential of storing solar energy in the form of hydrogen for Egypt',

Energy Sources 11. pp.95-103.

)6

D 120