Experimental Investigation of Producing Brown's Gas using a Metal-Plate Electrolyzer for Diesel Vehicle Applications

Abstract

The infusion of Brown's gas into a diesel engine can create faster, more efficient combustion, which can be produced by using an alkaline electrolyzer. Brown's gas consists of 2/3 hydrogen and 1/3 oxygen by volume, as produced by the alkaline electrolyzer. An experimental test setup for verifying the performance of an electrolyzer provided by Stan Warner was established in the Battery Laboratory at the University of California Davis. A series of tests were performed under constant current and pulse current modes. Analyses of the data indicate that the electrolyzer functions reliably at all current values. The Brown's gas production rate is found to be linear with current and in agreement with Faraday's Law. The electrolyzer meets the developer's claim of 3–4 L min−1 of Brown's gas production at 60 A. The calculated energy efficiency of the electrolyzer based on the high heating value of hydrogen is 61 % at 70 A. The performance and pulse characteristics of the Warner electrolyzer indicate it is well suited for use in diesel truck applications.

Full text

DOI: 10.1002/ente.201600222
Experimental Investigation of Producing Browns Gas
using a Metal-Plate Electrolyzer for Diesel Vehicle
Applications&&ok?&&
Yimin Wu,[a, b] Changjun Xie,*[b, c] Andrew Burke,*[c] Hengbing Zhao,[c] Marshall Miller,[c]
and Stan Warner[c]
Introduction
At the present time, the worlds energy demands are met pri-
marily using fossil fuels. This is especially the case for provid-
ing fuels for vehicles using internal combustion engines. Due
to the depletion of fossil fuel sources and increasing oil
prices, alternative approaches to increasing fuel economy
and reducing harmful emissions from internal combustion
engines are being sought. The use of hydrogen as an energy
carrier is one of the options being studied[1–20] to improve
combustion. This approach is included in most governmental
strategic plans for sustainable energy systems.
One of the advantages of hydrogen lies in the variety of
ways that it can be produced as well as the long-term viabili-
ty of many renewable energy sources (e.g., from biomass,
wind, solar, seawater, or nuclear power). The variety of ways
to use hydrogen as a fuel (e.g., internal combustion engines,
gas turbines, fuel cells) with virtually zero harmful emissions
and high efficiency makes hydrogen a very attractive fuel.[1]
Stan Warner and his colleagues have been developing
water electrolyzers and testing them in trucks powered by
diesel engines for a decade as a means of improving the fuel
economy and reducing the emissions from trucks. An electro-
lyzer of this sort was delivered to the Battery Laboratory at
the Institute of Transportation Studies, University of Califor-
nia, Davis for evaluation. This study describes in detail the
testing of Warner electrolyzer and the data analysis. The fol-
lowing sections review the literature on the use of hydrogen
and Browns gas in diesel engines as well as the recent litera-
ture on electrolyzers; an analytical description of the opera-
tion of the electrolyzer cells and a derivation of the equa-
tions useful in the data analysis are presented to follow.
Hydrogen as a fuel in internal combustion engines
The most promising application of hydrogen in transporta-
tion is in hydrogen fuel cell (H2FC) vehicles.[2–4] However,
H2FC vehicles cannot be used widely at the present time be-
cause of limitations of infrastructure and the high cost of the
fuel cell. In addition to use in fuel cells, hydrogen can be
used as a fuel for internal combustion engines. The use of hy-
drogen as a fuel in internal combustion engines has been
studied by a number of research groups worldwide.[5–12]
These studies have indicated that hydrogen alone can be
used with good efficiency in spark ignition engines and as
a mixture with diesel fuel in compression ignition engines. It
is this latter application that is of primary interest in this
study, particularly in application to dual-fuel type diesel en-
gines in which hydrogen is used to enhance the diesel-fuel
combustion.[10–13] These studies supply hydrogen to the intake
air, and the diesel fuel is injected directly into the cylinder,
[a] Dr. Y. M. Wu
Hubei Key Laboratory of Advanced Technology for Automotive Components,
Automobile Engineering Institute
Wuhan University of Technology
430070 Wuhan (PR China)
[b] Dr. Y. M. Wu, Prof. C. J. Xie
School of Automation
Wuhan University of Technology
430070 Wuhan (PR China)
E-mail: jackxie@whut.edu.cn
[c] Prof. C. J. Xie, Prof. A. Burke, Dr. H. B. Zhao, Dr. M. Miller, S. Warner
Institute of Transportation Studies
University of California
Davis, CA 95616 (USA)
E-mail: afburke@ucdavis.edu
The infusion of Browns gas into a diesel engine can create
faster, more efficient combustion, which can be produced by
using an alkaline electrolyzer. Browns gas consists of 2/3 hy-
drogen and 1/3 oxygen by volume, as produced by the alka-
line electrolyzer. An experimental test setup for verifying the
performance of an electrolyzer provided by Stan Warner was
established in the Battery Laboratory at the University of
California Davis. A series of tests were performed under
constant current and pulse current modes. Analyses of the
data indicate that the electrolyzer functions reliably at all
current values. The Browns gas production rate is found to
be linear with current and in agreement with Faradays Law.
The electrolyzer meets the developers claim of 3–4 L min1
of Browns gas production at 60 A. The calculated energy ef-
ficiency of the electrolyzer based on the high heating value
of hydrogen is 61% at 70 A. The performance and pulse
characteristics of the Warner electrolyzer indicate it is well
suited for use in diesel truck applications.
Energy Technol. 2016,4, 1 – 10  2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 1
These are not the final page numbers! ÞÞ
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
10 10
11 11
12 12
13 13
14 14
15 15
16 16
17 17
18 18
19 19
20 20
21 21
22 22
23 23
24 24
25 25
26 26
27 27
28 28
29 29
30 30
31 31
32 32
33 33
34 34
35 35
36 36
37 37
38 38
39 39
40 40
41 41
42 42
43 43
44 44
45 45
46 46
47 47
48 48
49 49
50 50
51 51
52 52
53 53
54 54
55 55
56 56
57 57

which can bring significant improvements in emissions and
fuel efficiency.
Using hydrogen as an additive to enhance the convention-
al diesel engine performance has been investigated[14–22] by
several researchers, and the outcomes are promising. Some
experiments[14–16] indicated that the addition of a small
amount of H2had a minimal effect on the cylinder pressure
and the combustion process; the addition of a relatively
large amount of H2substantially increased the peak cylinder
pressure, and the peak heat release rate at high loads was
not practical, considering both the safety and mechanical du-
rability. Moreover, providing a large amount of H2onboard
a vehicle is difficult and expensive.
However, additional interesting research shows that an on-
board hydrogen–oxygen generator, which produces a H2/O2
mixture through the electrolysis of water, has significant po-
tential to enhance the conventional diesel engine per-
formance. Baris works[17] were performed under constant
speed with varying load and amount of H2/O2mixture added.
The test results show that by using 4.84, 6.06, and 6.12%
total diesel equivalents of H2/O2mixtures, the brake thermal
efficiency values increased from 32.0 to 34.6%, 32.9 to
35.8%, and 34.7 to 36.3 % at 19, 22, and 28 kW, respectively.
These settings resulted in respective fuel savings of 15.07,
15.16, and 14.96%. The emissions of hydrocarbons (HC),
CO2, and CO decreased, whereas the NOxemissions in-
creased. Other researchers[18–22] found that the H2/O2mixture
can result both in a reduction in the concentration of emis-
sion pollutant constituents and an enhancement in engine ef-
ficiency. The H2/O2mixture gas is referred to as Browns
gas,[18–20] HHO gas,[18, 19] or HRG gas.[20] &&Please define
HHO and HRG acronyms here.&&
The infusion of Browns gas into a diesel internal combus-
tion engine is known to create faster, more-efficient diesel
combustion, which will create more power/better mileage
and fewer pollutants. It is also known that the infusion of hy-
drogen alone does not produce the same benefits. Both hy-
drogen and oxygen are needed to produce the beneficial ef-
fects. &&Reference needed?&& Browns gas may also
contain radicals of hydrogen and oxygen, which will influ-
ence the combustion process.
For years, the problem has been how to produce Browns
gas for injection into the diesel engine. Compressing Browns
gas, as an option, could be problematic and surely dangerous.
The preferred option is the onboard production of the gas, as
needed and then infusion directly into the engine air intake
system. The development of an onboard electrolyzer con-
nected to the vehicle alternator system is a convenient solu-
tion for vehicle applications.
Water electrolysis technology and cell analysis for onboard pro-
duction of hydrogen
Literature review
Browns gas is a H2/O2mixture gas from the electrolysis of
water. Zeng[23] reviewed the current state of technology of
hydrogen production using water electrolysis as an energy-
storage mechanism for generating electricity from wind and
solar in remote areas. Zhang[24] evaluated and analyzed the
performance characteristics of a phosphoric acid–water elec-
trolyzer (PAWE) for hydrogen production and developed
some optimal design strategies of a PAWE system. Some
new electrolytes and technology have been presented and
studied for improving the performance of water electroly-
sis.[25–30]
Carmo[31] compared alkaline electrolysis with PEM elec-
trolysis and solid-oxide electrolysis (SOEC), describing that
alkaline electrolysis is a well-established technology with ad-
vantages of non-noble catalysts and relatively low cost. The
above comparison illustrates that alkaline water electrolysis
is currently mature with reasonable efficiency relative to the
other emergent water electrolysis technologies.
Diguez[32] tested the performance of a commercial alka-
line water electrolyzer (HySTAT from Hydrogenics) de-
signed for a rated hydrogen production of 1 Nm3H2h1at an
overall power consumption of 4.90 kWh (Nm3)1H2.
ukic
´[33] constructed and tested an alkaline electrolyzer with
50 50 2 mm3Ni metal-foam electrodes, A 50 50 0.4 mm3
Zirphon membrane; and 25% alkaline (KOH) solution elec-
trolyte, followed by calculating the energy efficiency (based
on hydrogen’s high heating value), which was above 55%.
Mazloomi[34] reviewed the factors that affect the electrical ef-
ficiency of electrolysis, including the electrolyte quality, tem-
perature, pressure, electrical resistance of the electrolyte,
electrode material, separator material, and applied voltage
waveform. In summary, alkaline water electrolysis is an at-
tractive method for hydrogen production, offering the ad-
vantage of simplicity.
Cell analysis for the production of Brown’s gas by water electrol-
ysis
The decomposition of water into hydrogen and oxygen can
be achieved by passing a direct electric current (DC) be-
tween two electrodes separated by an aqueous electrolyte
with good ionic conductivity. The overall reaction[35] for split-
ting water is
H2Oð1Þþelectrical energy !H2ðgÞþ1
2O2ðgÞð1Þ
In an alkaline electrolyzer the electrolyte is usually KOH
or NaOH, where the positive ions K+or Na+and hydroxide
ion OHform the ionic current between the electrodes. The
anodic and cathodic reactions[35–39] at the electrodes are
Anode 2 OH1aqðÞ!
1
2O2gðÞþH2OlðÞþ2eð2Þ
Cathode :2H2OlðÞþ2e!H2gðÞþ2OH1aqðÞ ð3Þ
Overall :H2OlðÞ!H2gðÞþ
1
2O2gðÞ ð4Þ
Energy Technol. 2016,4, 1 – 10  2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 2
These are not the final page numbers! ÞÞ
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
10 10
11 11
12 12
13 13
14 14
15 15
16 16
17 17
18 18
19 19
20 20
21 21
22 22
23 23
24 24
25 25
26 26
27 27
28 28
29 29
30 30
31 31
32 32
33 33
34 34
35 35
36 36
37 37
38 38
39 39
40 40
41 41
42 42
43 43
44 44
45 45
46 46
47 47
48 48
49 49
50 50
51 51
52 52
53 53
54 54
55 55
56 56
57 57

From Equation (4), the produced Browns gas consists of 2/3
hydrogen and 1/3 oxygen by volume.
The production rate of Browns gas can be calculated by
using Faradays law and the ideal gas relation as follows:
Icell ¼nFmH2cellðÞ¼2;mBRG cellðÞ¼1:5mH2cellðÞ ð5Þ
in which Iis the current (A) through the cell, Fis Faradays
constant (0.965 105), nis the number of electrons per mole
of H2,mH2is the rate of hydrogen production (moles H2/sec),
and mBRG is the rate of Browns gas production (molgas s1). It
follows from Equation (5) that:
mH2mol s1

¼0:518 105Ið6Þ
Using the ideal gas law, the volume production of hydro-
gen is given by VH2m3s1

¼ðRgasT=PÞmH2, in which Rgas
is the gas constant equal to 8.314 &&units?&&,Tand P
are the temperature and pressure of the gas. At room tem-
perature and pressure
VH2m3s1

¼2:375 102mH2;VH2Lmin
1

¼1:425 103mH2
ð7Þ
Substituting from Equation (6) for mH2
VH2Lmin
1

¼0:738 102Ið8Þ
The volume of the Browns gas production is 3/2 times the
hydrogen production. Hence the volume rate of Browns gas
production for a particular electrolyzer is given by:
VBRG Lmin
1

=electrolyzer ¼1:11 102NcIcell ð9Þ
in which Ncis the number of cells in the electrolyzer.
The calculation of the cell voltage is not simple, as shown
in Ref. [35–37], involving the determination of the open-cir-
cuit voltage and the cell resistance in the electrolyzer. The
open-circuit voltage can be determined from thermodynam-
ics,[32] and the reversible cell voltage V0,rev is given from
Gibbs Free Energy (DG) for the overall reaction and the
thermo-neutral voltage V0,thn from the total energy change
DH, where V0;rev ¼V0;rev
nF and V0;thn ¼V0;thn
nF . For room tempera-
ture and pressure V0;rev ¼1:23V and V0;thn ¼1:48 V.
For the electrolyzer, the appropriate open-circuit voltage
is V0;thn ¼1:48 V. If it is assumed that the cell resistance Ris
a constant, the cell voltage can be expressed as:
Vcell ¼1:48 þIcell Rð10Þ
The efficiency hof an electrolyzer can be expressed as the
ratio of the heating value of the hydrogen produced to the
electricity needed to produce that hydrogen from the electro-
lyzer.[33,40, 41] Hence the efficiency can be written as :
h¼mH2mol s1

DHJmol
1

HHV
VI ð11Þ
where DHJmol
1

HHV ¼284 103Jmol
1

H2:&&Ok with
DHin Eq. 11?&&
Experimental setup and testing
The experimental setup for testing the Warner electrolyzer is
shown in Scheme 1. The scheme shows the Warner H2electrolyz-
er, the Bitrode battery tester, gas flow meter, and the data ac-
quisition device. The electrolyzer was tested much like a battery
in that the current was specified, controlled by the Bitrode, and
the response of the test device was measured. In the case of the
electrolyzer, this meant that the voltage and Browns gas pro-
duced were the measured quantities.
The Warner electrolyzer unit consisted of the two banks shown
in Scheme 1as the left bank and the right bank, and the banks
could be tested separately or in parallel. There are 5 cells con-
nected in series for each bank. Hence there are 10 cells in the
unit and the current per cell is the total current divided by 2. The
operating voltage of the unit was in the range 10–15 V similar to
the lead-acid battery used in truck applications. The technical
specifications of the Warner electrolyzer are as follows: (i) the
electrodes were metal plates constructed of 360 stainless steel;
(ii) the plates were 4 inches  17.5 inches (1 inch =2.54 cm) and
the active area is 452 cm2per plate.
Measurement of the Browns gas produced is not straightfor-
ward, because the composition of the gas is not known for cer-
tain, and its temperature and pressure are not known accurately.
Hence it was desired to measure directly the volume of gas pro-
duced. This measurement was performed using a Gilian Gilibra-
tion2 Calibration System (Sensidyne) on loan from the Hydrogen
Lab at UC Davis. This device measures the rate of gas produced
by tracking the motion of a soap film bubble between two
known points in a cell. Repeated measurements were made and
Scheme 1. Schematic of the experimental setup. The warner H2electrolyzer is
made up of parallel left and right banks. The banks were tested separately
and in parallel.
Energy Technol. 2016,4, 1 – 10  2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 3
These are not the final page numbers! ÞÞ
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
10 10
11 11
12 12
13 13
14 14
15 15
16 16
17 17
18 18
19 19
20 20
21 21
22 22
23 23
24 24
25 25
26 26
27 27
28 28
29 29
30 30
31 31
32 32
33 33
34 34
35 35
36 36
37 37
38 38
39 39
40 40
41 41
42 42
43 43
44 44
45 45
46 46
47 47
48 48
49 49
50 50
51 51
52 52
53 53
54 54
55 55
56 56
57 57

the results were averaged. A photograph of the test setup is
shown Scheme 2.
Utilized to enhance the performance and fuel economy of
a diesel internal combustion engine, the amount of Browns gas
would be varied depending on the engine load and engine speed.
Hence it is of interest to know the limits of the Browns gas pro-
duction and the associated voltage and the pulse characteristics
of the electrolyzer. To meet these requirements, the electrolyzer
was tested as follows:
(1) The Browns gas produced (Lmin1) and voltage were mea-
sured as a function of input current. The stability of opera-
tion at each test condition was noted.
(2) The dynamic performance of producing Browns gas in
a pulse current testing mode was measured for various
pulse/current conditions.
(3) Detailed data analysis was performed to determine how well
the electrolyzer performance could be described in terms of
well-established laws of gas chemistry and electrochemistry,
such as the ideal gas law and Faradays law.
Results and Discussion
A series of tests was conducted to investigate the per-
formance of the Warner electrolyzer operated in the constant
current and variable current modes to achieve the test re-
quirements (described above).&&ok?&& The response
voltage was monitored at intervals of 100 ms, and the flow
rate of Browns gas was measured at intervals of 10 s during
the intermediate stage of every test and averaged (of six
measurements).
Constant-current testing
When the K1 and K2 switches are closed (Scheme 1), the left
bank and the right bank are in parallel and the current is
split between the two banks. Figures 1–3 show the voltage re-
sponse curves when the electrolyzer is fed with constant cur-
rent values of 40, 50, and 60 A, respectively. As presented in
Figures 1 and 2, for current values of 40 and 50 A, the re-
sponse voltage curves show relatively flat behavior for
10 min. However, when the test time is extended to 1 h as
shown in Figure 3for a current of 60 A, the slope of the volt-
age curve becomes apparent and it is clear the voltage de-
creases slowly with time. A careful study of Figure 1and
Figure 2 reveals that the slopes of the voltage curves for 30
and 40 A are consistent with the longer test shown in
Figure 1. Voltage response curves of both banks for the electrolyzer constant
electrolyzer feed current of 40 A.
Scheme 2. Photograph of the experimental setup including the Warner H2
electrolyzer, the gas flow meter, the data acquisition device, and the DC
power supply.
Figure 2. Voltage response curves of both banks for the electrolyzer constant
electrolyzer feed current of 50 A.
Energy Technol. 2016,4, 1 – 10  2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 4
These are not the final page numbers! ÞÞ
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
10 10
11 11
12 12
13 13
14 14
15 15
16 16
17 17
18 18
19 19
20 20
21 21
22 22
23 23
24 24
25 25
26 26
27 27
28 28
29 29
30 30
31 31
32 32
33 33
34 34
35 35
36 36
37 37
38 38
39 39
40 40
41 41
42 42
43 43
44 44
45 45
46 46
47 47
48 48
49 49
50 50
51 51
52 52
53 53
54 54
55 55
56 56
57 57

Figure 3. By using linear fitting for the steady voltage–cur-
rent curves listed in Figures 1–3, the following fitting func-
tions result: y40 =0.928E5*x+11.561; y50 =
1.025E5*x+11.972; y60 =1.182 E5*x +12.341. It was
determined that all of the slope values were very small, and
the slope values increase with increased applied current.
It is expected that the temperature of the electrolyzer will
slowly increase with time due to resistive heating. This in-
crease in temperature is the likely reason[34, 39, 40] for the slow
decrease in the voltage shown in Figure 3. Additionally, the
temperature increases faster with the increase of the applied
current. Unfortunately, it was not possible to sample the in-
ternal temperature of the Warner electrolyzer due to its
closed structure. The performance of the electrolyzer will be
analyzed in the next section using pulse current data.
Pulse current testing
As the response voltages at the different currents are depen-
dent on the temperature, the water electrolyzer was operated
at 50 A for 30 min prior to beginning the pulse testing. The
pulse test consisted of a series of pulses: 0–40, 0–50, 0–60, 0–
70, 0–80, 0–90, 0–100, and 0 A. Each pulse lasted for 60 s.
Figure 4shows the test current curve with both banks con-
nected. The voltage response curve of the electrolyzer to the
pulse currents is shown in Figure 5. Magnified views of the
voltage from 11.2 to 14.0 V, and from 0 to 2 V (open-circuit
voltage at I=0) are shown in Figure 6and Figure 7, respec-
tively. The data indicate that the Warner electrolyzer has
good dynamic performance that is predictable over a wide
range of current values up to 100 A. In fact, it is shown
below that the Browns gas volume rate was nearly 6 Lmin1
at 100 A current, &&Which figure should be referenced
here? Please check as they appeared to be referenced out of
order.&& and the gas production rate increased linearly
with current during the entire pulse test sequence.
Figure 3. Voltage response curves of both banks for the electrolyzer constant
electrolyzer feed current of 60 A. Figure 4. Pulse current curve for both banks. The pulse test consisted of
a series of pulses: 0–40, 0–50, 0–60, 0–70, 0–80, 0–90, 0–100, and 0 A. Each
pulse lasted for 60 s.
Figure 5. Response voltage of the electrolyzer with pulse current.
Figure 6. Magnified view of the response voltage from 11.2 to 14.0 V.
Energy Technol. 2016,4, 1 – 10  2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 5
These are not the final page numbers! ÞÞ
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
10 10
11 11
12 12
13 13
14 14
15 15
16 16
17 17
18 18
19 19
20 20
21 21
22 22
23 23
24 24
25 25
26 26
27 27
28 28
29 29
30 30
31 31
32 32
33 33
34 34
35 35
36 36
37 37
38 38
39 39
40 40
41 41
42 42
43 43
44 44
45 45
46 46
47 47
48 48
49 49
50 50
51 51
52 52
53 53
54 54
55 55
56 56
57 57

Comparison of the left and right bank performances
The Warner electrolyzer consists of the left bank and the
right bank. It was of interest to test each bank separately to
determine their consistency. Each bank was tested at one-
half the current values shown in Figure 4for both banks.
However, in this test, the current pulses were for 240 s and
the rest period between pulses was 60 s. Figure 8shows the
current curve for each bank. The voltage response curves of
the left bank and right bank with the pulse current are
shown in Figure 9. The enlarged views of the voltage re-
sponses from 10 to 14 V, and from 0 to 2 Vare shown in Fig-
ure 10and Figure 11, respectively, which show a voltage dif-
ference between them of less than 0.15 V (the error is
1.3%). Hence, the performances of the left and right banks
are highly consistent.
Characteristics of the Warner electrolyzer
The pulse test results will be used to determine the per-
formance characteristics of the Warner electrolyzer and
whether the results are consistent with the relationships de-
rived above using Faradays law and the ideal gas equation.
The pulse characteristics of the electrolyzer for a range of
currents (cell currents between 20–50 A) have been shown in
Figure 5–Figure 11. The cell voltages vary systematically with
current and in all cases the voltage rises rapidly when the
current is applied. For the purposes of cell characterization,
the nearly steady values achieved for each current step are
Figure 7. Magnified view of the response voltage from 0 to 2 V.
Figure 8. Pulse current curve for each bank. To determine their consistency,
each bank was tested at one-half the currents shown in Figure 4for both
banks. However, the current pulses lasted for 240 s and the rest period be-
tween pulses was 60 s.
Figure 9. Voltage response vs. pulse current. The blue curve represents the
left bank voltage and the red curve represents right bank voltage.
Figure 10. Magnified view of response voltage from 10 to 14 V. The blue
curve represents left bank voltage and the red curve represents right bank
voltage.
Energy Technol. 2016,4, 1 – 10  2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 6
These are not the final page numbers! ÞÞ
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
10 10
11 11
12 12
13 13
14 14
15 15
16 16
17 17
18 18
19 19
20 20
21 21
22 22
23 23
24 24
25 25
26 26
27 27
28 28
29 29
30 30
31 31
32 32
33 33
34 34
35 35
36 36
37 37
38 38
39 39
40 40
41 41
42 42
43 43
44 44
45 45
46 46
47 47
48 48
49 49
50 50
51 51
52 52
53 53
54 54
55 55
56 56
57 57

of primary interest. Each bank consists of five cells in series
and two banks in parallel. Hence the average current
through each cell is Itotal/2, and the average voltage is Vtotal/2.
The average cell voltage as a function of cell current is
shown in Figure 12. Note that the curve of Vcell versus Icell is
nearly linear except between 20–30 A so it can be approxi-
mated as V=A+BI, where, as discussed above, Equa-
tion (10), Ais related to the open-circuit voltage of the cell,
and Bis related to its resistance. A curve fit from the range
30–50 A yields indicating a resistance of 0.017 Wper cell.
Vcell ¼1:95 þ0:017 Icell ð12Þ
Figure 7indicates a cell open-circuit voltage of 1.3–1.4 V,
which is significantly less than the 1.95 V value in Equa-
tion (12). This higher value is likely due to the overpotential
at the stainless-steel plate needed to generate the limiting ex-
change current.[42]
The measured dependence of the Browns gas production
rate on the total current to the electrolyzer is shown in
Figure 13. There is considerable scatter in the Browns gas
data, but the curve of the average values is almost linear
with current, as predicted by Equation (9) above. The pre-
dicted Browns gas production rate (L min1) is the follow-
ing:
VBRG Lmin
1

=electrolyzer ¼1:11 101Icell ¼0:056Itotal
ð13Þ
The characteristics of the Warner electrolyzer are summar-
ized in Table 1. The volume of the Browns gas produced is
certainly consistent with the claim of 3–4 Lmin1by Warner.
Normalizing the volume rate by the total current yielded an
average value of 0.0573 Lmin1A1, which is in good agree-
ment with the prediction.
The energy efficiency of the electrolyzer can be calculated
using Equation (11), which was also discussed above. Substi-
tuting Equation (6) for mH2, the efficiency hof this electro-
lyzer becomes
helect ¼Nc284 103Jmol
1

H20:518 105Icell mol s1

Vtotal Itotal
ð14Þ
and Itotal =2Icell; thus, helect =0.736·Nc/Vtotal.&&ok?&&
There is one formula between VH2and mH2shown in Equa-
tion (7), and the efficiency hof this electrolyzer can be calcu-
lated based on VH2according to Equations (11) and (7), that
Figure 11. Magnified view of response voltage from 0 to 2 V. The blue curve
represents left bank voltage and the red curve represents right bank voltage.
Figure 12. Averaged current–voltage curve for each cell calculated by using
the test data and the connection topology shown in Scheme 1, that is, each
bank consisted of five cells in series and the two banks are in parallel.
Figure 13. Brown’s gas production rate vs. current. The Brown’s gas produc-
tion rate uses the gas flow to measure which is observed by the gas flow
meter.
Energy Technol. 2016,4, 1 – 10  2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 7
These are not the final page numbers! ÞÞ
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
10 10
11 11
12 12
13 13
14 14
15 15
16 16
17 17
18 18
19 19
20 20
21 21
22 22
23 23
24 24
25 25
26 26
27 27
28 28
29 29
30 30
31 31
32 32
33 33
34 34
35 35
36 36
37 37
38 38
39 39
40 40
41 41
42 42
43 43
44 44
45 45
46 46
47 47
48 48
49 49
50 50
51 51
52 52
53 53
54 54
55 55
56 56
57 57

is, helect ¼284103mH2
VtotalItotal ¼VH2Lmin
1
ðÞ
284
1:425
ðÞ
VtotalItotal . Browns gas consists of
2/3 hydrogen and 1/3 oxygen by volume, that is,
VBRG Lmin
1

¼1:5VH2Lmin
1

; thus Equation (15) is de-
rived:
helect ¼
VBRG Lmin
1

1
1:5
284
1:425

Vtotal Itotal ¼132:86 VBRG Lmin
1

VtotalItotal
ð15Þ
The efficiency of the electrolyzer was calculated using both
Equations (14) and (15) based on the measured Brown’s gas
production rate and calculated H2. The results are shown in
Table 1. The two approaches do not yield the same values,
but the values are in the same range indicating the electro-
lyzer efficiency is approximately 60%.
Conclusions
As one type of onboard hydrogen/oxygen mixture gas,
Browns gas can be produced by using alkaline water elec-
trolysis. A test bench based on the Bitrode battery tester was
established and the experimental tests to investigate the per-
formance of the Warner electrolyzer were performed. The
experimental results showed that the Warner electrolyzer
unit functioned reliably over a wide range of current values
(up to 100 A) and the Browns gas production (L min1A1)
was in excellent agreement with Faradays Law.
The results of the study are summarized below:
*All tests were operated in the constant-current or pulse-
current mode, and the response voltage and the flow rate
of Browns gas were monitored.
*The Warner electrolyzer unit functioned reliably and in
a predictable manner over a wide range of currents (up to
100 A) and exhibited good voltage–current dynamic per-
formance, with highly consistent performances on the left
and right banks.
*The Browns gas production rate on current is almost
linear, which is in excellent agreement with Faradays
Law. Moreover, the unit easily met the claim of 3–
4Lmin
1by Warner.
Acknowledgements
This research was supported by the next STEPS programs of
the Institute of Transportation Studies of the University of
California, Davis, the National Key Basic Research Develop-
ment Plan (973 Plan) (No. 2013CB632505), the National Nat-
ural Science Foundation of China(51477125) and the Science
and technology support program of Hubei Province
(2014BEC074).
Keywords: alkaline electrolyzer ·combustion ·diesel
engines ·electrolysis ·hydrogen
[1] S. Verhelst, T. Wallner, Prog. Energy Combust. Sci. 2009,35, 490 –
527.
[2] H. B. Zhao, A. F. Burke, J. Power Sources 2009,186, 408– 416.
[3] C. J. Xie, X. Y. Xu, P. Bujlo, D. Shen, S. H. Quan, J. Power Sources
2015,279, 487–494.
[4] I. Bartolozzi, F. Rizzi, M. Frey, Appl. Energy 2013,101, 103 –111.
[5] J. M. Gomes Antunes, R. Mikalsen, A. P. Roskilly, Int. J. Hydrogen
Energy 2009,34, 6516–6522.
[6] B. L. Salvi, K. A. Subramanian, Int. J. Hydrogen Energy 2016,41,
5842–5855.
[7] S. Szwaja, K. G. Rogalinski, Int. J. Hydrogen Energy 2009,34, 4413 –
4421.
[8] R. Hari Ganesh, V. Subramanian, V. Balasubramanian, J. M. Malli-
karjuna, A. Ramesh, R. P. Sharma, Renewable Energy 2008,33,
1324–1333.
[9] A. M. de Morais, M. A. Mendes Justino, O. Souza Valente, S. de Mor-
ais Hanriot, J. R. Sodr, Int. J. Hydrogen Energy 2013,38, 6857 –
6864.
[10] T. Miyamoto, H. Hasegawa, M. Mikami, N. Kojima, H. Kabashima,
Y. Urata, Int. J. Hydrogen Energy 2011,36, 13138 – 13149.
[11] N. Saravanan, G. Nagarajan, Appl. Energy 2010,87, 2218–2229.
[12] N. Saravanan, G. Nagarajan, S. Narayanasamy, Renewable Energy
2008,33, 415–421.
[13] D. B. Lata, A. Misra, S. Medhekar, Int. J. Hydrogen Energy 2012,37,
6084–6096.
[14] G. K. Lilik, H. D. Zhang, J. M. Herreros, D. C. Haworth, A. Boeh-
man, Int. J. Hydrogen Energy 2010,35, 4382–4398.
[15] C. Liew, H. Li, J. Nuszkowski, S. Liu, T. Gatts, R. Atkinson, N. Clark,
Int. J. Hydrogen Energy 2010,35, 11357–11365.
[16] C. Liew, H. Li, S. Liu, M. C. Besch, B. Ralston, N. Clark, Y. Huang,
Fuel 2012,91, 155– 163.
[17] S. Bari, M. M. Esmaeil, Fuel 2010,89, 378– 383.
[18] S. A. Musmar, A. A. Rousan, Fuel 2011,90, 3066– 3070.
[19] A. C. Yilmaz, E. Uludamar, K. Aydin, Int. J. Hydrogen Energy 2010,
35, 11366– 11372.
Table 1. Summary table of the investigated characteristics of the Warner electrolyzer.
Current [A] Voltage [V] Brown’s gas flow [L min1] Brown’s gas flow per Amp [L min1A1][a] Efficiency (measured) Efficiency (calculated)
[Eq. (15) Eq. (14)]
40 11.4 2.0 0.05 0.583 0.646
50 11.9 3.0 0.06 0.670 0.618
60 12.3 3.37 0.056 0.607 0.598
70 12.8 4.2 0.060 0.623 0.575
80 13.2 4.8 0.060 0.604 0.557
90 13.6 5.1 0.057 0.554 0.541
100 14.0 5.8 0.058 0.550 0.526
0.0573 (avg)
[a] Total current to the electrolyzer.
Energy Technol. 2016,4, 1 – 10  2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 8
These are not the final page numbers! ÞÞ
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
10 10
11 11
12 12
13 13
14 14
15 15
16 16
17 17
18 18
19 19
20 20
21 21
22 22
23 23
24 24
25 25
26 26
27 27
28 28
29 29
30 30
31 31
32 32
33 33
34 34
35 35
36 36
37 37
38 38
39 39
40 40
41 41
42 42
43 43
44 44
45 45
46 46
47 47
48 48
49 49
50 50
51 51
52 52
53 53
54 54
55 55
56 56
57 57

[20] A. Birtas, I. Voicu, C. Petcu, R. Chiriac, N. Apostolescu, Int. J. Hy-
drogen Energy 2011,36, 12007– 12014.
[21] S. F. Wang, C. W. Ji, J. Zhang, B. Zhang, Int. J. Hydrogen Energy
2011,36, 11164– 11173.
[22] F. Christodoulou, A. Megaritis, Int. J. Hydrogen Energy 2014,39,
2692–2702.
[23] K. Zeng, D. K. Zhang, Prog. Energy Combust. Sci. 2010,36, 307 –
326.
[24] H. C. Zhang, S. H. Su, X. H. Chen, G. X. Lin, J. C. Chen, Int. J. Hy-
drogen Energy 2012,37, 18615– 18621.
[25] F. Fiegenbaum, E. M. Martini, M. O. de Souza, M. R. Becker, R. F.
de Souza, J. Power Sources 2013,243, 822–825.
[26] S. Tuomi, A. Santasalo-Aarnio, P. Kanninen, T. Kallio, J. Power Sour-
ces 2013,229, 32– 35.
[27] C. Y. Hung, S. D. Li, C. C. Wang, C. Y. Chen, J. Membr. Sci. 2012,
389, 197– 204.
[28] S. D. Li, C. C. Wang, C. Y. Chen, J. Membr. Sci. 2009,330, 334 – 340.
[29] S. Marini, P. Salvi, P. Nelli, R. Pesenti, M. Villa, M. Berrettoni, G.
Zangari, Y. Kiros, Electrochim. Acta 2012,82, 384 – 391.
[30] F. Rossi, A. Nicolini, Appl. Energy 2012,97, 763– 770.
[31] M. Carmo, D. L. Fritz, J. Mergel, D. Stolten, Int. J. Hydrogen Energy
2013,38, 4901–4934.
[32] P. M. Diguez, A. Ursffla, P. Sanchis, C. Sopena, E. Guelbenzu, L. M.
Ganda, Int. J. Hydrogen Energy 2008,33, 7338–7354.
[33] A. ukic
´, M. Firak, Int. J. Hydrogen Energy 2011,36, 7799– 7806.
[34] S. K. Mazloomi, N. Sulaiman, Renewable Sustainable Energy Rev.
2012,16, 4257–4263.
[35] Ø. Ulleberg, Int. J. Hydrogen Energy 2003,28, 21 –33.
[36] M. Caspersen, J. B. Kirkegaard, Int. J. Hydrogen Energy 2012,37,
7436–7441.
[37] A. Djafour, M. Matoug, H. Bouras, B. Bouchekima, M. S. Aida, B.
Azoui, Int. J. Hydrogen Energy 2011,36, 4117–4124.
[38] M. Z. Shen, N. Bennett, Y. L. Ding, K. Scott, Int. J. Hydrogen Energy
2011,36, 14335– 14341.
[39] P. Millet, S. Grigoriev, Renewable Hydrogen Technologies (Eds.:
L. M. Gandia, G. Arzamendi ), Elsevier, Oxford, 2013, pp. 19 –41.
[40] H. C. Zhang, G. X. Lin, J. C. Chen, Int. J. Hydrogen Energy 2010,35,
10851– 10858.
[41] S. D. Li, C. C. Wang, C. Y. Chen, Electrochim. Acta 2009,54, 3877–
3883.
[42] J. T. Wang, W. W. Wang, C. Wang, Z. Q. Mao, Int. J. Hydrogen
Energy 2012,37, 2069–12073.
Received: March 29, 2016
Revised: May 29, 2016
Published online on &&
&&
, 0000
Energy Technol. 2016,4, 1 – 10  2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 9
These are not the final page numbers! ÞÞ
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
10 10
11 11
12 12
13 13
14 14
15 15
16 16
17 17
18 18
19 19
20 20
21 21
22 22
23 23
24 24
25 25
26 26
27 27
28 28
29 29
30 30
31 31
32 32
33 33
34 34
35 35
36 36
37 37
38 38
39 39
40 40
41 41
42 42
43 43
44 44
45 45
46 46
47 47
48 48
49 49
50 50
51 51
52 52
53 53
54 54
55 55
56 56
57 57

FULL PAPERS
Y. M. Wu, C. J. Xie,* A. Burke,*
H. B. Zhao, M. Miller, S. Warner
&& –&&
Experimental Investigation of
Producing Browns Gas using a Metal-
Plate Electrolyzer for Diesel Vehicle
Applications&&ok?&&
Green chemistry with Brown’s gas: An
experimental test bench is established
in the Battery Laboratory at the Uni-
versity of California-Davis for verify-
ing the performance of an electrolyzer
in diesel engine applications. A series
of tests are performed under constant
current and pulse current modes. The
Browns gas (2/3 hydrogen and 1/3
oxygen by volume) production rate is
found to be linear with current and in
agreement with Faradays Law. The
performance and pulse characteristics
of the Warner electrolyzer indicate it is
well suited for use in the diesel truck
applications.
Improvement of #Diesel Engine Combustion, from Prof. Andrew Burke of @UCDavisCoE @ucdavis @UCDa-
visMagazine SPACE RESERVED FOR IMAGE AND LINK
Share your work on social media! Energy Technology has added Twitter as a means to promote your article.
Twitter is an online microblogging service that enables its users to send and read text-based messages of up to 140
characters, known as “tweets”. Please check the pre-written tweet in the galley proofs for accuracy. Should you or
your institute have a Twitter account, please let us know the appropriate username (i.e., @accountname), and we
will do our best to include this information in the tweet. This tweet will be posted to the journals Twitter account
@EnergyTechnol (follow us!) upon online publication of your article, and we recommended you to repost
(“retweet”) it to alert other researchers about your publication.
Please check that the ORCID identifiers listed below are correct. We encourage all authors to provide an ORCID identifier
for each coauthor. ORCID is a registry that provides researchers with a unique digital identifier. Some funding agencies
recommend or even require the inclusion of ORCID IDs in all published articles, and authors should consult their funding
agency guidelines for details. Registration is easy and free; for further information, see http://orcid.org/.
Dr. Yimin Wu
Prof. Changjun Xie
Prof. Andrew Burke
Dr. Hengbing Zhao
Dr. Marshall Miller
Stan Warner
Energy Technol. 2016,4, 1 – 10  2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 10
These are not the final page numbers! ÞÞ
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
10 10
11 11
12 12
13 13
14 14
15 15
16 16
17 17
18 18
19 19
20 20
21 21
22 22
23 23
24 24
25 25
26 26
27 27
28 28
29 29
30 30
31 31
32 32
33 33
34 34
35 35
36 36
37 37
38 38
39 39
40 40
41 41
42 42
43 43
44 44
45 45
46 46
47 47
48 48
49 49
50 50
51 51
52 52
53 53
54 54
55 55
56 56
57 57