See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/332248053

Influence of hydrogen addition to pipeline natural gas on the combustion

performance of a cooktop burner

Article  in  International Journal of Hydrogen Energy · April 2019

DOI: 10.1016/j.ijhydene.2019.03.100

CITATIONS READS

5 483

3 authors, including:

Yan Zhao Vincent Mcdonell

University of California, Irvine University of California, Irvine
9 PUBLICATIONS   14 CITATIONS    236 PUBLICATIONS   2,791 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Evaluación de las características de la combustión del metano con adición de productos de electrólisis del agua en condiciones de mezcla pobre View project

Fuel Flexibility-Computational Fluid Dynamics-Hydrogen production and use View project

All content following this page was uploaded by Yan Zhao on 05 June 2019.

The user has requested enhancement of the downloaded file.

International Journal of Hydrogen Energy

Influence of Hydrogen Addition to Pipeline Natural Gas on the Combustion Performance of a

Cooktop Burner

Yan Zhao, Vincent McDonell*, Scott Samuelsen

UCI Combustion Laboratory, University of California, Irvine, CA 92697-3550

ABSTRACT
Displacing pipeline natural gas with renewable hydrogen is a promising way to reduce the emission of carbon dioxide, which is a
major greenhouse gas. However, due to significantly differing characteristics of hydrogen and natural gas, such as flame speed, adiabatic
flame temperature and stability limits, the combustion performance of hydrogen/natural gas mixture differs from pure natural gas. From
the perspective of residential end users, a key question is: how much hydrogen can be injected into the pipeline natural gas without
influencing the performance of the residential burners? A representative cooktop burner is selected to study the influence of hydrogen
addition on the combustion and cooking performance. Flashback limits, ignition time, flame characteristics, cooking performance,
combustion noise, burner temperature, and various emissions (NO, NO2, N2O, CO, unburned hydrocarbon (UHC), NH3) are evaluated
for different levels of hydrogen addition. According to the experimental results, the combustion performance of the cooktop burner is
not significantly affected with up to about 15% hydrogen addition by volume, which shows the feasibility of utilizing hydrogen on
existing cooking appliances without any modification. The experiment methodologies and results in this study will serve as a reference
for future test and emission regulation standards on domestic burners.
Keywords: Renewable hydrogen; natural gas, cooktop burner; flashback; emissions

INTRODUCTION
To combat climate change by reducing the greenhouse gas emissions, renewable and sustainable fuels are needed to replace
relatively carbon intensive fossil fuels. Hydrogen is considered as a promising energy carrier because of the high energy density on mass
base, and its carbon-free property [1].Being the lightest element, hydrogen can be used as a competitive energy carrier and it can be
shipped in tanks or transported in pipelines to end users. The hydrogen heating value on a mass base is about 2.5 times that of methane,
which makes liquified hydrogen a promising energy carrier. With the development of hydrogen production and storage technologies,
compressed hydrogen can be used as a very competitive energy storage strategy to serve the areas that electricity grid can’t reach.
Hydrogen can be generated from different sources, including fossil fuels, renewable fuels, or electrolysis of water. Conventional
hydrogen production methods using fossil fuels include 1) steam reformation of natural gas 2) partial oxidation and 3) gasification [2].
However, to reduce carbon intensity of energy use, renewable methods for generating hydrogen are of interest, which also help address
energy security concerns for regions that are highly dependent on energy imports [3]. Renewable hydrogen can be generated from the
methods using renewable feedstocks [4], wind power [5], solar energy [6], nuclear energy [7] or water electrolysis [8]. Water electrolysis
uses the grid electricity which has been termed as grid power to gas (P2G) technology [9][10]. Renewable hydrogen production
technology development is also being driven by developments in hydrogen fuel cell technology [11][12][13] or hydrogen internal
combustion engine [14][15] for transportation and low emission fuel flexible gas turbine technology [16][17].
With the development of hydrogen generation and hydrogen end use technologies, injecting renewable hydrogen into the existing
natural gas pipeline becomes a potentially attractive step in displacing carbon from fossil fuels [18][19]. Currently, the majority of
pipeline gas is consumed by combustion devices, either to generate electricity in power plants, or to provide heat in industrial,
commercial and residential applications. Due to the significantly varying combustion properties between hydrogen and natural gas (e.g.,
wide flammability range and high flame speed of hydrogen), it is not practical to replace the natural gas with hydrogen directly.
Evaluation of fuel interchangeability is needed for different combustion applications to test the tolerance of the existing combustion
devices for hydrogen. Gas turbines manufactures have already started to develop fuel flexible turbine applications [20][21]. Compared
to evaluation and development of hydrogen for use in heavy duty gas turbines, evaluation of hydrogen tolerance in burners for light
industry, commercial, and residential applications have received little attention. Yet, these combustion applications consume a large
amount of pipeline natural gas in the US. In 2017, power generation used 34.1% of the pipeline natural gas, and industrial, commercial
and residential applications consume 29.3%, 11.7% and 16.3% of the natural gas separately in the US [22].
Interestingly, the use of hydrogen as the energy source in regional pipelines is not new, as it is one of the major species in coal gas
which was used extensively in appliances decades ago in Europe and the United States. Even today, coal gas is still used as the residential

*
Corresponding author: Tel: (949)-824-5950 x11121; fax: (949)-824-7423. E-mail: [email protected]

© 2019 Elsevier Ltd. All rights reserved 1

International Journal of Hydrogen Energy

gas in part of China and Southeast Asia. Coal gas was usually called town gas in Europe, especially in the UK [23], but was given several
names in the US, such as water gas, water coal gas or carbureted water gas [24][25]. Coal gas is a mixture of hydrogen, carbon monoxide,
methane and carbon dioxide. The hydrogen percentage in the coal gas can be very high, up to 50% or even 80% [26], and the hydrogen
percentage is highly dependent on the manufacturing method. Since combustion devices once could operate on hydrogen-rich fuel
existed before, the question arises: Why can’t current appliance burners be modified to match those from one hundred years ago? The
reason is that appliance technology has evolved to meet certain regulations that were not in place decades ago, thus current combustion
devices need to be evaluated, especially in the context of advanced lean premixed low emission combustion technology development.
During the coal gas era, diffusion combustion technology was widely applied to prevent flashback and increase the stability of the
flames. Due to the lack of fuel/air mixing before combustion, diffusion combustion has a tendency of producing higher unburned
hydrocarbon and carbon monoxide emissions [27]. The carbon monoxide emission from the combustion process and carbon monoxide
in the coal gas itself is a great threat to human health and residential safety [28][29]. With the discovery of natural gas reserves around
the world and rapid development of extraction technology, pipeline gas supplies have largely switched from coal gas to natural gas even
without detailed understanding of the combustion phenomena in those devices.
Nowadays, appliance emission regulations are driving combustion technology advancement, and these devices must now bear more
advanced combustion science. Emission regulations are being adopted and frequently updated for technologies associated with gas
turbines for power generation, transportation applications, and industrial burners. Due to advances in combustion systems of modern
gas turbines and application of selective catalytic reduction (SCR) technology, CO and NOX emissions from power generation have
decreased dramatically down to single-digit level [27]. However, combustion technology advancement has not been as substantial for
other combustion devices. For example, a cooktop burner is a self-aspirating combustion device, which uses partially premixed
combustion technology. This type of burner is widely used in residential appliances (oven, grill, room furnace, water heaters, etc.),
commercial applications (restaurant kitchen burner, central furnace, etc.), and light industrial combustion devices (glass industry burner,
surface treatment burner, etc.). Yet emission regulations for such devices are limited primarily to the context of safety. In 2017, the
natural gas consumption in residential, commercial and industrial applications added up to 57.3% of the total natural gas consumption
in the US. As a result, emission control of these combustion devices represents a significant opportunity to improve air quality. California
is one of the most diligent regions in the world at promoting emission regulations, since start of the battle against Los Angeles smog in
the middle of the 20th century. Since the 1960s, several environment related agencies were founded. These agencies include California
Air Resources Board (CARB) and South Coast Air Quality Management District (SCAQMD), and they contribute in applying and
updating emission regulations [30]. However, emission regulations on low heating load combustion devices, like residential appliances,
are still few in number.
As a first step in exploring the interchangeability of hydrogen with natural gas, the objective of this experimental study is to
investigate the upper limit of hydrogen percentage in fuel mixtures that are applied to cooktop burners, which will serve as a reference
for hydrogen injection limit into the pipeline natural gas. Multiple combustion performances are evaluated, including ignition time,
flame characteristics, cooking performance, combustion noise, burner temperature, multiple emission species (NO, NO2, N2O, CO,
UHC, NH3). The results contribute to a needed residential burner fuel interchangeability test methods and results database and will
provide guidance for future emission regulations for residential burners.

EXPERIMENTAL METHODOLOGY
In this study, natural gas and hydrogen mixtures are created using a critical flow orifice-based gas mixing device and are regulated
down to 2000 Pa (8 inches of water) at the point of connection to the cookstove. The 2000 Pa level is consistent with the installation
requirements for the cooktop burner as operated on natural gas. Procedurally, this also mimics the situation that would arise should a
natural gas appliance be subjected to changes on the incoming fuel composition that may be imposed on the gas pipeline. The percentage
of hydrogen in the fuel mixture is increased gradually from 0% to the maximum that can be attained without encountering a significant
operability issue (e.g., flashback or overtemp).

Burner Configuration
Before selecting the cooktop burner device for the present study, consideration was given to the types of burner configuration that
could be found among current devices. More than twenty cooktop burners on the market were examined. Most of them have very similar
or even identical burner geometries which is represented by the device used in the present study. The burner geometry is shown in Fig.1,
which has the heating load of 9100 Btu/hr. All of the investigated cooktop burners use “self-aspirating” technology. As shown, fuel is
injected into the burner head while entraining a small amount of surrounding air. Part of the mixture flows out of the ignition ports and
arrive at the ignitor through the ignition tube. After ignition, the flame propagates to the burner head and a full flame will be established.

© 2019 Elsevier Ltd. All rights reserved 2

International Journal of Hydrogen Energy

(a) burner schematic (b) device tested

Fig.1 Cooktop Burner Configuration
The self-aspirating burner utilizes partially premixed combustion technology, which means part of the air (primary air) is mixed
with fuel before combustion, and the rest of the air is supplied to the flame as secondary air in the combustion process. The primary air
to fuel ratio is established such that the mixture is above the rich flammability limit for safety purpose. The burner configuration studied
has a measured fuel concentration of around 18%, which means that the mixture exceeds the rich flammability limit of natural gas and
therefore combustion cannot occur within the burner head. However, hydrogen has a flammability range of 4% - 75%, this means that
if the burner operated on hydrogen-rich fuel, the hydrogen percentage within the burner head can fall within the flammability range very
easily. Hence the risk of the flame flashback into the burner head for hydrogen is much higher than natural gas. As a result of the widely
different combustion characteristics of hydrogen and natural gas, flashback is expected to occur at some point as the hydrogen level
increases. In the present study, this flashback limit amount was considered as an important limiting factor to be determined.

Experiment Setup
Natural gas and hydrogen mixtures are established using the control panel of the mixing station. Flame images are recorded by a
Nikon D90 digital camera. Ignition behavior is recorded by high speed camera (Nikon J1), which can take up to 1400 frames per second.
An infrared imaging camera (FLIR-T640) is used to analyze the burner surface temperature. The combustion noise and temperature are
recorded by Extech-407762 sound level meter and thermocouples, respectively. Emissions are measured using a Horiba MEXA-
QCL1400, a Horiba PG350, and a Horiba FMA220. Species measured include NO, NO2, N2O, CO, UHC, NH3. The output signals of
these devices are sent to National Instrument CompactRIO, and then processed on the computer by LabView. The experiment device
diagram is shown in Fig.2.

Fig.2 Experiment Setup Diagram

The three Horiba emission analyzers (PG350, FMA220, QCL) are synchronized under the testing condition. Hence, the readings of
different analyzers present the emission level at the same sampling location simultaneously. The exhaust goes through a water drop-out

© 2019 Elsevier Ltd. All rights reserved 3

International Journal of Hydrogen Energy

system before it is supplied to the analyzers. Therefore, the exhaust is cooled down to the room temperature and the emission readings
are on a dry basis.
The details of the measurement uncertainty of the emission analyzers are as follows:
• Flame Ionization Magneto-Pneumatic Analyzer 220 (FMA220): UHC measurement range: 100 ppm. Accuracy: ±1% of the
maximum range setting. Zero and span drift: ±1% of the maximum range setting.
• Horiba Quantum-Cascade Laser Analyzer 1400 (QCL1400): NO: 0-100 ppm, NO2: 0-50 ppm, N2O: 0-100 ppm. NH3: 0-50
ppm. Zero drift: ±1% of the maximum range setting. Span drift: ±2% of the maximum range setting.
• Horiba Portable Gas Analyzer 350 (PG350): NO: 0-250 ppm, NOX: 0-250 ppm; CO: 0-500 ppm; CO2: 0-10 vol%; O2: 0-25
vol%. Zero drift: ±1% of the maximum range setting. Drift: ±1% of the maximum range setting.

Gas Property
The composition of the baseline pipeline natural gas used in this study is shown in Table 1 -. The higher and lower heating values at
21.1 °C, 1 atm are 37.7 MJ/m3 (1012.5 Btu/scf), and 34.0 MJ/m3 (912.6 Btu/scf), respectively. The methane percentage of the natural
gas used is 95.8%, which is higher than the US average natural gas supply [31].

Table 1 - Natural gas contents

Molecular
Content Formula Molecular Fraction
methane CH4 95.8%
ethane C2H6 1.4%
propane C3H8 0.4%
iso-butane C4H10 0.05%
n-butane C4H10 0.05%
iso-pentane C5H12 0.025%
n-pentane C5H12 0.025%
C6 C6H14 0.017%
C7 C7H16 0.017%
C8 C8H18 0.016%
carbon dioxide CO2 1.9%
oxygen O2 0%
nitrogen N2 0.3%

To assist in the interpretation of the results obtained, key properties parameters, especially those related to combustion, are listed
in Table 2 -. The values are from reference [27] at room condition (25 °C, 1 atm), except as noted.

Table 2 - Fuel Property Comparison

Methane Hydrogen
Density kg/m3 0.648 0.0813
Viscosity 10-5 Pas 1.11 0.89
Laminar Flame Speed m/s 0.4 2.1
𝜙 0.53 0.14
Low Flammability[32] vol % 5 4
𝜙 1.6 2.54
High Flammability[32] vol % 15 75
Ignition Energy 10-5 J 33 2
MJ/m3 34.0 10.2
Low Heating Value[33] MJ/kg 49.9 120.1
MJ/m3 37.8 12.5
High Heating Value[33] MJ/kg 55.5 142.1
Adiabatic Flame Temperature K 2226 2318
Wobbe Index MJ/m3 51.9 48.5

© 2019 Elsevier Ltd. All rights reserved 4

International Journal of Hydrogen Energy

While the physically small hydrogen molecules increase the risk of leakage from the hydrogen container, the low density of
hydrogen causes it to dissipate quickly rather than accumulating in the area close to the leakage area or storage facility. While this helps
decrease the possibility of ignition occurrence after leakage, the very wide flammability limits of hydrogen/air mixtures can pose a
concern regarding leakage and flammable gas buildup. The laminar flame speed of hydrogen is 5 times that of methane, this causes
difficulty in burner design to achieve desired hydrogen flame stabilization distances and the wide flammability range of hydrogen
exacerbates this problem. Currently, many existing burners on the market utilize technology that allows controlled premixing of the fuel
and air, including the cooktop burner investigated in this study. As the hydrogen percentage increases in the fuel, the burner head shown
in Fig.1 will be filled with more hydrogen which increases the flashback tendency. Moreover, hydrogen is more reactive than methane,
and the ignition energy of hydrogen is 20 μJ, which is much lower than 330 μJ of methane. This property of hydrogen makes it easier
for hydrogen-rich fuel ignition. These disparate properties result in uncertainty in terms of predicting the combustion behavior, further
motivating the present study.
The heating values of methane and hydrogen are also shown in Table 2 -. Heating value is an important parameter to evaluate the
ability of a fuel as an energy carrier. The difference between lower heating value (LHV) and higher heating value (HHV) is the latent
heat of water in the combustion exhaust. In most combustion devices, only the LHV is practically available due to complexity of
condensing and managing the water in the exhaust. Although the energy density of hydrogen is higher than methane on mass base, the
volumetric heating value of methane is more than 3 times that of hydrogen. This implies that the same volume flowrate of the fuel will
produce much less heat for hydrogen compared to natural gas. However, in practical combustion applications, the heating value is not
the main parameter affecting heat input. The Wobbe Index is the factor commonly used to evaluate fuel interchangeability in terms of
heat release rate [34] in most practical appliances. If two fuels have the same Wobbe Index, the heat output will be constant if the inlet
pressure is held constant for the combustion device.
𝐻𝑖𝑔ℎ𝑒𝑟 𝐻𝑒𝑎𝑡𝑖𝑛𝑔 𝑉𝑎𝑙𝑢𝑒
𝑊𝑜𝑏𝑏𝑒 𝐼𝑛𝑑𝑒𝑥 =
√𝐹𝑢𝑒𝑙 𝑆𝑝𝑒𝑐𝑖𝑓𝑖𝑐 𝐺𝑟𝑎𝑣𝑖𝑡𝑦
(1)
While the Wobbe Index is widely applied today, the concept of using heating value over square root of fuel density as a fuel
interchangeability parameter originates from a 1920s American Gas Association (AGA) study. In this study, thousands of experiments
on fuel interchangeability were conducted, and a “C-index” of change in performance of appliances was established as a critical
parameter, which is the predecessor of the widely used Wobbe Index today [35]. To illustrate the reason why Wobbe Index can be used
to predict interchangeability of heat release rate, the Bernoulli equation can be used as a starting point, and a few assumptions made
which are applicable to typical appliance operation. By assuming the fuel flow within the pipeline feeding the device and the combustion
device itself is at steady state, and the flow is incompressible (constant fluid density) and inviscid, the Bernoulli equation is reduced to
equation 2. In this equation, p1 is the natural gas pressure in household pipeline, which is reduced to a low level around 2000 Pa in order
to reduce any possible leakage within the house and for overall safety. For most appliances, combustion usually takes place at open-air
condition, therefore, p2 is the atmospheric pressure.
1 1
𝑝1 + 𝜌𝑓 𝑉12 = 𝑝2 + 𝜌𝑓 𝑉22
2 2
(2)

This can be rearranged to solve for the volumetric flow of the gas existing the fuel port:

2∆𝑝
𝑉𝑓̇ = 𝐴2 √ 𝐴2 2
(3)
𝜌𝑓 (1−(𝐴 ) )
1

As shown, the volume flow rate of the fuel is a function of the fuel density, fuel flow areas, and pressure drop from the household
pipeline to the combustion device. However, due to the relatively small orifice area for the combustion device compared to the gas feed
cross section, (A2/A1)2 can usually be ignored in equation 3, which leads to equation 4.
2∆𝑝
𝑉𝑓̇ = 𝐴2 √
𝜌
(4)
𝑓
Then, the heat output of a combustion device can be presented as
2∆𝑝 𝐻𝐻𝑉
𝑞̇ = 𝑉𝑓̇ 𝐻𝐻𝑉 = √
𝜌𝑎𝑖𝑟 √𝑠𝑔𝑓
(5)
Based on the assumption of constant density and neglecting viscous effect, the heat output of the combustion device is only a
function of the heating value over the square root of the fuel specific density.
𝐻𝐻𝑉
𝑊𝑜𝑏𝑏𝑒 𝐼𝑛𝑑𝑒𝑥 = ~𝑞̇ (6)
√𝑠𝑔𝑓
The reason Wobbe Index works well for predicting the heat output of different combustion devices is the validity of the assumptions
in practice. The gas flow velocity in the household Pipeline is much slower than the sound velocity; therefore, it can be assumed that
the gas is incompressible in the pipeline. When the gas flows through the orifice of a combustion device, the gas pressure drop can

© 2019 Elsevier Ltd. All rights reserved 5

International Journal of Hydrogen Energy

accelerate the flow, which will result in friction loss. However, the gauge pressure in the pipeline is only 2000 Pa, which is less than 2%
of the atmosphere pressure. The pressure loss is not significant.
As shown in Table 2 -, although the heating values of hydrogen and methane are significantly different, they have similar Wobbe
Index values. This observation leads to the consideration that part of the pipeline natural gas can be replaced with hydrogen without
influencing the heat output in current combustion devices. In other words, a 9000 Btu/hr burner will retain a nearly 9000 Btu/hr rating
when used with natural gas/hydrogen mixtures. To analyze the hydrogen/natural gas mixture limits for the cooktop burner, the fuel
mixture properties need to be understood. The heating value and Wobbe Index of natural gas and hydrogen mixtures (assuming hydrogen
takes x% of the volume in the fuel mixture) are given by equation 7 and 8.

𝐻𝐻𝑉𝑀𝑖𝑥 = (1 − 𝑥%)𝐻𝐻𝑉𝑁𝐺 + 𝑥%𝐻𝐻𝑉𝐻2 (7)

𝐻𝐻𝑉𝑀𝑖𝑥
𝑊𝐼𝑀𝑖𝑥 = (8)
(1−𝑥%)𝜌𝑁𝐺 +𝑥%𝜌𝐻
√ 2
𝜌𝐴𝑖𝑟

The heating value and Wobbe index of different fuel mixtures are shown in Fig.3. The solid lines represent the fuel mixture property,
and the linear dashed lines are the heating value and Wobbe Index regulations of the natural gas supplier for UCI Combustion Laboratory
(UCICL). Every gas company has their own gas quality regulation, and the heating value and Wobbe Index are usually the most important
factors. Based on the heating value regulation, up to 3.2% pipeline natural gas can be replaced by hydrogen. However, up to 13.8% can
be added if the Wobbe Index is the only factor considered.

45 54

40 52

Wobbe Index (MJ/m3)

50
Heating Value (MJ/m3)

35

30 48

25 Heating Value 46
HHV-Low Limit
20 HHV-Up Limit 44
Wobbe Index
15 WI-Low Limit
42
WI-Low Limit
10 40
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
H2 Percentage

Fig.3 Heating value and Wobbe Index variation of hydrogen/natural gas mixtures

In Fig.3, the heating load shows a linear trend, but the Wobbe Index has an inflection point at around 81% hydrogen addition. Also,
100% hydrogen has identical Wobbe Index as 35% hydrogen/65% natural gas mixture. However, the fuel properties of these two fuel
classes, for example, flammability, flame speed and flame temperature will be very different, and the combustion performance will differ
significantly. This implies that, although Wobbe Index predicts the heating load, it can’t be used as the only fuel interchangeability factor
for combustion devices. In this study, the test conditions purposefully go beyond the current regulations. The hydrogen is increased all
the way from 0% to the point where an operability issue occurs.

RESULTS AND DISCUSSION

To compare the combustion performance of cooktop burners operating on different amounts of hydrogen added to natural gas,
ignition performance, flame characteristics, cooking performance, combustion noise, burner temperature and emissions are examined.

Ignition Performance
American National Standard Institute-Z21.1 (ANSI-Z21.1)[36] regulates that eligible cooktop burners should ignite within 4
seconds from turning on the spark to stable flame establishment. In this study, the 4 cooktop burners (each with same heating rate

© 2019 Elsevier Ltd. All rights reserved 6

International Journal of Hydrogen Energy

specification) are tested. Each burner test is repeated 5 times under both cold and hot ignition conditions. Cold condition means the
ignition test is conducted when the cooktop burner starts at room temperature. Hot conditions represent a re-ignition test: the burner is
operated for 5 minutes, and is then turned off and then turned on again. The re-ignition time is recorded as the hot ignition time. The
different operating conditions for the ignition time are shown in Fig.4. In this figure, each error bar reflects the standard deviation of 20
repeated tests.
1.6

1.4

1.2
Ignition Time (s)

0.8

0.6

0.4

0.2

0
0%-Cold 0%-Hot 5%-Cold 5%-Hot 10%-Cold 10%-Hot 15%-Cold 15%-Hot 20%-Cold 20%-Hot
H2 Percentage

Fig.4 Cooktop burner ignition time with H2 addition

As shown, the ignition time under hot conditions is shorter than that under cold conditions. Fig.4 demonstrates that ignition time
decreases a little by increasing hydrogen percentage in the fuel. However, due to the unsteadiness of the ignition process, the variance
of the ignition time is relatively large, which is shown by the 95% confidential interval error bars. Under cold ignition condition, 5%
hydrogen addition in the fuel decreases the ignition time. However, under hot ignition condition, the ignition time difference between
pure natural gas and 5% hydrogen addition is not significant, which is shown by the confidential interval overlap of these two conditions.
The reason why this phenomenon appears is that the ignition time decrease caused by hydrogen addition is overshadowed by the high
burner temperature.
From 5% to 15% H2 addition, no significant effect upon ignition is observed. However, with 20% or more hydrogen addition an
intermittent flashback starts to occur. Fig.5 shows the flashback process during ignition. After the control knob was turned to “Lite”
position, the flame propagates from the igniter to the burner ports. However, before the flame is fully established at the burner ports,
flashback occurs. The flame propagates through the burner ports and the ignition ports into the burner head. The fuel/air mixture in the
burner head is within the flammable limit, therefore the mixture ignites immediately and flame propagates out of the burner through the
bottom aeration port with a loud noise. At last, the flame extinguishes.
Based on these results, it is concluded that more than about 15% hydrogen added to natural gas will lead to operability issues on
this type of cooktop burner.

0s 0.14s 0.28s

© 2019 Elsevier Ltd. All rights reserved 7

International Journal of Hydrogen Energy

0.42s 0.56s 0.70s

Fig.5 Flashback procedure of a cooktop burner

Flame Characteristics and Flashback

The flame appearance with different hydrogen concentrations is shown in Fig.6. In this test, the hydrogen limit indicated by the
ignition process is superseded in order to assess flame structure at hydrogen levels well above that which leads to flashback. This is
accomplished by igniting the burner with pure natural gas and then adjusting the hydrogen content. As shown, the cooktop burner flame
exhibits a light blue color. Low percentages of hydrogen addition do not have much influence on the flame appearance. In the methane
combustion process, hundreds of reactions occur. In the present flame, the blueish color is attributed to the chemiluminescence of CH*
[27], which is an intermediate radical in hydrocarbon flames. Also noted is an obvious reddish flame tail when hydrogen percentage is
above 60%. Fundamental studies show that the flame color of hydrogen should be invisible or light blue under reduced light condition.
Therefore, this reddish color phenomenon of hydrogen rich fuel is inconsistent with the expected hydrogen flame radiation spectrum. In
the present study, fine particulate filters are used which reduced the reddish appearance compared to unfiltered hydrogen. As a result, it
is believed that this observation is due to radiation from contaminants in the hydrogen associated with the steel cylinders used. The
hydrogen purity is 99.999%, so it is unlikely that trace gas species are causing this phenomenon.

Fig.6 Cooktop burner flames characteristics with H2 addition

As the hydrogen percentage increases, the flame shows increasing unsteadiness. When the hydrogen percentage increases above
75%, the flame flashes back into the burner head before the 80% hydrogen target is reached. Hence the burner port velocity and
quenching characteristics are no longer able to prevent the flame from entering the burner head. It should be noted that this experiment
was run under idle conditions, meaning no pot or pan is placed above the flame on the burner top.
Under actual pot heating conditions, an obvious interaction between the flame tip and the bottom surface of the cooking pot is
observed, which adds uncertainty and unsteadiness to the flame. As shown in Error! Reference source not found., when heating the
pot, flashback happens at 55% hydrogen addition, and the burner was destroyed within half a minute. Hence the behavior observed

© 2019 Elsevier Ltd. All rights reserved 8

International Journal of Hydrogen Energy

under idle conditions are significantly different from that when actually heating a pot. This observation leads to consideration of how
tests are done should a standard method be needed.
Regarding the observed flashback behavior, it is noteworthy that Jones [37] also conducted the hydrogen addition research in 2018.
A cooktop burner used in the UK was investigated. However, the only factor considered in Jones’s study was flashback safety, compared
to the larger spectrum of combustion performance tests conducted in the current study. Jones successfully injected 34.7% hydrogen into
the cooktop burner, but the experiment details were not given; therefore, it is not clear whether that was the ignition limit, normal
operation limit, cooking limit or the lowest limit of these three. From his analysis, Jones concluded that “It is thus proposed that up to
30 mol% of the natural gas supply may be replaced in the UK with guaranteed safety and reliability for the domestic end-user, without
any modification of the appliance infrastructure.” However, Jones’s conclusion is in conflict with the current results which suggest a
limit of less than 20% hydrogen addition. It is noted that the cooktop burner used in the present study is also available on the UK market.
Therefore, studies should be conducted on more burners and multiple combustion performance factors should be considered, before
drawing the final conclusion of how much hydrogen should be injected into the pipeline.

Fig.7 Flashback in the water boiling process

Cooking Efficiency
In this section, results associated with the actual heat transfer to a cooking pot are presented. The standard 1 gallon cooking pot
recommended by ANSI, is utilized for this test. The pot is filled with water to 80% of the total 1 gallon (3.785 L) capacity.
Due to the flashback limit of 55% hydrogen during pot heating, the limit for hydrogen addition is 50% for these tests. Two main
cooking factors are considered: cooking time and cooking efficiency. Cooking time is the time period of the water temperature rising
from the room temperature to the boiling point. There are many ways to define cooking efficiency. Oberascher [38] boiled potatoes and
eggs to establish the efficiency of cooking. The efficiency was divided into several grades based on the properties of the egg yolk: very
liquid, liquid, firm, frim and dry, etc. However, using this method to evaluate cooking efficiency is subjective and egg dependent.
Therefore, to increase the repeatability of cooking efficiency measurement, the cooking efficiency is defined as the heat absorbed by the
water over the heat released from the fuel[39]. In fuel interchangeability study for cooktop burners, an increase in efficiency is desirable,
but the case of cooking time is different. A situation resulting in longer cooking time is usually not desirable, and a situation that can
dramatically reduce cooking time is also not ideal. This is because shorter cooking time means the cooking appliance is not working at
its rated heating load. Therefore, the ideal situation of fuel interchangeability is increasing the cooking efficiency while keeping the
cooking time constant.

𝑐𝑝 ∙𝑚𝑙,𝑤𝑎𝑡𝑒𝑟 ∙Δ𝑇+Δ𝑚∙ℎ𝑔,𝑤𝑎𝑡𝑒𝑟
𝜂𝑐𝑜𝑜𝑘𝑖𝑛𝑔 =
𝑚𝑓𝑢𝑒𝑙 ∙𝐿𝐻𝑉
(9)

Cooking time and efficiency variation with hydrogen percentage are shown in Fig.8. Each data point in the figure is the average of
4 individual cooktop burners (each with same heat rate specification). As shown, both cooking time and cooking efficiency increase
slightly with hydrogen addition. From 0% to 50% hydrogen addition, cooking time increases by 9% and efficiency increases by 4%.
Many factors influence cooking time and cooking efficiency. For example, the adiabatic flame temperature of hydrogen is higher than
that of natural gas for a given fuel/air ratio, but the heat input using hydrogen-rich fuel is lower than natural gas due to the slightly lower
Wobbe Index of hydrogen. These two effects might compensate each other and result in small variation of cooking time and efficiency.
Moreover, the flame characteristic change might also influence heat transfer by radiation and convection. The small variation of
hydrogen-rich fuel cooking performance bodes well for adding hydrogen to natural gas as a means to displace carbon.

© 2019 Elsevier Ltd. All rights reserved 9

International Journal of Hydrogen Energy

45 60%
Cooking Time
40 50%
Cooking Efficiency
Cooking Time (min)

Cooking Efficiency
35 40%

30 30%

25 20%

20 10%

15 0%
0% 10% 20% 30% 40% 50%
H2 Percentage
Fig.8 Cooking time and efficiency of different hydrogen/natural gas mixtures

Combustion Noise and Burner Temperature

Combustion noise is measured by Extech-407762 at the burner operating condition. The background laboratory sound pressure
level is 56 dB. When the burner is operating on pure natural gas, the sound level increases slightly to 56.8 dB. As shown in Fig.9, the
combustion noise intensity doesn’t change significantly for cases with hydrogen addition up to 40%. At hydrogen addition levels above
40%, however, an obvious “buzz” noise of the flame is evident which is associated with the flame dynamics at higher hydrogen
percentages.

58.5

58
Sound Level (Decibel)

57.5

57

56.5

56

55.5

55
Room 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50%
H2 Percentage
Fig.9 Combustion noise level of different percentage of hydrogen addition

Due to the difference between the hydrogen flame velocity and the adiabatic flame temperature, hydrogen addition might also
influence the flame stabilization distance at the burner ports which can, in turn, influence the burner temperature. The temperature
distribution image of the cooktop burner is shown in Fig.10. The highest temperature of the burner is the area near the injection ports.
Therefore, this area is chosen as the representative area for burner temperature measurement. Fig.10 (b) shows that the hydrogen addition
increases the burner temperature steadily. At pure natural gas operating condition, the burner temperature is around 223 C, and it
increases to 248 C at 75% hydrogen addition. Burner life cycles may be reduced by temperature increase. However, the burner is very
robust and it might take hundreds of hours of continuous running to find the influence of this temperature increase. If the real cooking
procedure was 1-hour level operating, the influence of this temperature increase might be negligible.

© 2019 Elsevier Ltd. All rights reserved 10

International Journal of Hydrogen Energy

(a) FLIR image of the cooktop burner

270
Cooktop Burner Temperature

260
250
240
(°C)

230
220
210
200
190
0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75%
H2 Percentage

(b) cooktop burner temperature

Fig.10 Cooktop burner temperature of different hydrogen/natural gas mixtures

Emissions
In this section, the emissions performance of the cookstove burners are assessed as a function of increasing hydrogen content. As
context for the emissions levels measured, Table 3 - summarizes several emission regulations put forth by various agencies. However,
the regulations from the 1980s are generally applied as opposed to being device specific. For example, Rule 407 requires that any single
device shouldn’t emit more than 2000 ppm CO into the environment in 15 consecutive minutes. Rules 407, 409 and 474 were first
adopted in the 1970s and then they were amended in the 1980s. Although those regulations are still valid, they have not been updated
or amended for more than three decades. In contrast, regulations developed for specific combustion devices, like gas turbines or internal
combustion engines, continue to be updated and will be stricter in the future. However, recent regulations for residential burners are
much less developed. In 2014, CARB issued emission regulation for commercial ovens used in the food industry or restaurants which
can be used as a reference for residential burners. For CO emissions, this regulation states that emissions should be less than 800 ppm
(@3% O2). NOX emissions, under this regulation, are temperature dependent. If the oven is operating at 500 °F the emission should be
lower than 40 ppm (@3% O2), but the emissions can be up to 60 ppm (@3% O 2) when the oven operates at higher temperatures. In
California, water heating consumes almost half of the total natural gas used for residential purposes [40]. Considering the high natural
gas consumption rate of water heaters, San Diego County Air Pollution Control District (SDCAPCD) released the Rule 69.5.1 to regulate
the NOX emission for water heaters in 2015, and this rule has been in effect since July 1st, 2016. Rule 69.5.1 has been adopted by
numerous counties in California. For residential appliance manufactures, American National Standard Institute updates their regulations
every few years, and these regulations are being adopted worldwide. However, these regulations only address CO emissions due to
safety concern caused by CO poisoning.

Table 3 - California emission regulation references for residential appliance

Regulation title Agency Time Major content
Rule 407[41] SCAQMD 1982 CO shouldn’t exceed 2,000 ppm by volume measured on a
dry basis, averaged over 15 consecutive minutes
Rule 409[42] SCAQMD 1981 Combustion contaminants shouldn’t exceed 0.23 gram per
cubic meter of gas calculated to 12 % CO2.
Rule 474[43] SCAQMD 1981 NOX: 125, 225, 300 ppm (@3% O2) depending on the load.

© 2019 Elsevier Ltd. All rights reserved 11

International Journal of Hydrogen Energy

… … … …
Rule 1153.1[44] CARB 2014 Commercial ovens: CO:800 ppm (@3% O2)
NOX:40 ppm (@3% O2)-500 °F-60 ppm (@3% O2)
Rule 69.5.1[45] SDCAPCD 2016 Water heaters: 40 ng of nitrogen oxides (NO2) per joule; 10
ng of nitrogen oxides (NO2) per joule.
ANSI Z21.10.1-2014[46] American National 2014- CO: water heaters less than 800 ppm
ANSI Z21.86-2016[47] Standard Institute& 2016 CO: room furnaces less than 200 ppm
ANSI Z21.1-2016[36] Canadian Standards CO: cooking appliances less than 800 ppm
ANSI Z21.11.2 -2016[48] Association CO: unvented room heaters less than 200 ppm

The emission regulations in Table 3 - vary significantly. Some regulation variations in this table include using 3% O 2, 12% CO2,
calorific emission index, or no basis at all. In cooktop burner emission measurements, a challenge is associated with the entrainment of
surrounding air which inevitably dilutes the sample. Sometimes, the exhaust is diluted to reach the emission measurement range of the
analyzer, or when the exhaust is at very high temperature, it needs to be cooled down by dilution air. In addition, it is not possible to
measure the open-air flame without any dilution air going into the exhaust, which is the same case in the present study. This phenomenon
emphasizes the importance of sample correction to proper reference levels. However, in the context of fuel interchangeability, it is
unclear which correction basis is best.
In this study, all three emission correction methodologies are used to evaluate the emission level of the cooktop burner. The oxygen
correction method that is widely used in industry[49] is shown in equation 10. Equation 11 is usually used in residential burners or light
industry burners [50] for 12% CO2 correction. This correction method is also called “air-free” method, because 12% is the CO2
percentage in the exhaust (dry base) when methane is burned in air at stoichiometric condition. The calorific correction is shown in
Equation 12 [51], which is also considered as a reliable emission correction method due to its independence on the exhaust species. In
many respects, the representation provided by Equation 12 should be the most robust, because it is independent of both O2 and CO2 in
the clarification. However, it requires more information to be known in addition to adjust the measured emissions levels which makes
it inherently more difficult to implement.
20.9−3
[𝑋]3% 𝑂2 ,𝑝𝑝𝑚 = [𝑋]𝑎𝑏𝑠,𝑝𝑝𝑚
20.9−[𝑂2 ]𝑎𝑏𝑠,%
(10)

12
[𝑋]12% 𝐶𝑂2 ,𝑝𝑝𝑚 = [𝑋]𝑎𝑏𝑠,𝑝𝑝𝑚
[𝐶𝑂
(11)
2 ]𝑎𝑏𝑠,% −[𝐶𝑂2 ]𝑎𝑖𝑟,%

0.1[𝑋]𝑎𝑏𝑠,𝑝𝑝𝑚 𝑚𝑜𝑙 𝐶𝑂2

[𝑋]𝑛𝑔/𝐽 =
[𝐶𝑂
𝑀𝑋,𝑔/𝑚𝑜𝑙 (12)
2 ]𝑎𝑏𝑠,% −[𝐶𝑂2 ]𝑎𝑖𝑟,% 𝑀𝐽 𝐹𝑢𝑒𝑙

Emission collection methodology for the cooktop burner is demonstrated in Fig.11. A hood is utilized to fully cover the cooking
range top while conducting water boiling experiment. Emissions are then collected through the vent at the top of the hood. To ensure
the normal operation of the cooktop burner, the hood location is set as 12.7 cm (5 inches) above the cooking range top surface, which is
recommended by ANSI Z21.1[36]. At this height, enough secondary air is provided to the flame to avoid impacting the flame while not
resulting in extraordinarily high sample dilution. During the experiment, the emission data is taken every second for 15 consecutive
minutes.

© 2019 Elsevier Ltd. All rights reserved 12

International Journal of Hydrogen Energy

Fig.11 Cooktop burner emission collection methodology

The emission results are shown in Fig.12. Results under the different emission corrections are plotted separately and each plot
contains CO, UHC, NO and total NOX (NO+NO2) readings. In each plot, every point is the average of over 1000 data readings. As can
be seen in Fig.12 (a), the cooktop burner NOX emission is 110 ppm (3% O2) and CO is 191 ppm (3% O2) when operating on natural gas.
With the addition of hydrogen, emissions decrease as shown in Fig.12(a). From pure natural gas to 50% hydrogen, NOX decreases
from 110 ppm (3% O2) to 84 ppm (3% O2), and NO drops from 70 ppm (3% O 2) to 52 ppm (3% O2). CO also decreases from 191 ppm
(3% O2) to 165 ppm (3% O2). Although UHC emission has some variance, the general trend is that hydrogen addition reduces the UHC
emission from 42 ppm (3% O2) to 11 ppm (3% O2). With the increase of hydrogen percentage in the fuel, CO and UHC emissions in the
exhaust can be reduced as the carbon percentage goes down. In situations with added hydrogen, the higher adiabatic flame temperature
contributes to the oxidation rate of the fuel in the combustion process. According to the Zeldovich mechanism [27], higher reaction
temperature usually increases the NOX emission; however, this study shows a trend of deceasing NOX emission with more hydrogen
addition, which is against the known theory. This effect might be caused by the operating condition. With a situation of 50% hydrogen
addition, although hydrogen adiabatic flame temperature is 4% higher than the methane adiabatic flame temperature, the heating load
of hydrogen/natural gas mixture is around 12% lower than natural gas. NOX emissions are reduced because of the steeper heat release
drop which overcomes the adiabatic flame temperature increase. It is not ideal to use the properties of the adiabatic flame temperature
for open-air combustion analysis, because it is significantly different from the ideal adiabatic situations.
Another observation is that NO and NO2 emission levels are very similar. For high temperature combustion (e.g., in gas turbines)
[49], the formation rate of NO is much faster than NO2. However, the NO and NO2 emissions of the cooktop burner are very similar,
which is caused by the relatively low temperature open-air flame property. Because of the significant heat loss caused by heating the
surrounding air and radiating to the surroundings, the NO formation rate is not that different from the NO 2 formation rate. Moreover,
the large amount of oxygen in the surrounding air is supplied to the flame, which helps speed up the oxidation of NO to NO 2.
Finally, based on the MEXA 1400-QCL measurements, little to no NH3 or N2O emissions are observed (measurements are within
the measurement accuracy). Usually, N2O can result from lean combustion conditions, especially when the equivalence ratio is lower
than 0.8 [27]. However, the fuel/air mixture equivalence ratio within the burner head is between 1.5 and 2, which makes the N2O
formation negligible.
In terms of the presentation basis for the emissions, as can be seen in Fig.12 (b), the emissions with added hydrogen content are
higher when 12% CO2 is used as the basis. The CO2 correction equation has the CO2 concentration in the denominator, which will
become lower with the increase of hydrogen in the fuel mixture. Therefore, NOX and CO increase as hydrogen addition increases.
Compared to the oxygen correction, the CO2 correction method might not be “fair” for hydrogen-rich fuel emission evaluation.
The calorific correction shows a similar trend as the oxygen corrected results. The UHC emission drops sharply at high amounts of
added hydrogen. However, the variation of other emission species is smaller compared to oxygen correction. The noteworthy point is
the difference of NO and NOX emission based on calorific correction is larger than that based on O2 and CO2 corrections. This is because
NOX calorific correction converts NO to NO2 on a mass basis.

© 2019 Elsevier Ltd. All rights reserved 13

International Journal of Hydrogen Energy

150 400

NO NOx CO UHC 350

125

CO/UHC (ppm @3% O2)

NO/NOX (ppm @3% O2)
300
100
250

75 200

150
50
100
25
50

0 0
-5% 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55%
H2 Percentage
(a) 3% O2 correction

150 400

NO NOx CO UHC 350

CO/UHC (ppm @12% CO2)

NO/NOX (ppm @12% CO2)

125
300
100
250

75 200

150
50
100
25
50

0 0
-5% 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55%
H2 Percentage
(b) 12% CO2 correction

90 200

NO NOx CO UHC
75
160
NO/NOX/CO (ng/J)

100*UHC (nmol/J)

60
120
45
80
30

40
15

0 0
-5% 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55%
H2 Percentage

© 2019 Elsevier Ltd. All rights reserved 14

International Journal of Hydrogen Energy

(c) Calorific correction

Fig.12 Emissions under different corrections

CONCLUSIONS
The influence of hydrogen addition into the natural gas pipeline on the combustion performance of a representative cooktop burner
is evaluated experimentally. Ignition time, flame characteristics, cooking performance, combustion noise, burner temperature and
emissions are evaluated. It is noted that the general emission levels of a cooktop burner are much higher than the single-digit emission
gas turbines for electricity generation. Although the natural gas consumption in power generation is twice the amount of natural gas
consumed in the residential buildings, if similar self-aspirating burners used in commercial and light industrial applications are included,
the NOX and CO emissions from these simple geometry burners far exceed the emissions from power generation process. Ironically,
regulators continue to drive the gas turbine industry to further reduce already low emissions, while appliance burners are emitting much
more emissions into the atmosphere. It is expected that by conducting both fuel property analysis and experiments, this study can help
establish a protocol for testing domestic burner combustion performance. The experiment results will are also intended to give an insight
for policy makers to consider hydrogen injection into pipeline natural gas and future emission regulations on residential burners.
The key observations from this study are:
• The higher reactivity of hydrogen, hydrogen-rich fuel shortens the ignition time but also leads to flashback limitations. The cooktop
burner presented in this study has a flashback limit of 20% addition.
• Under steady-state operating condition, the hydrogen percentage in the fuel can be increased up to 75%. However, when a cooking
pot is placed over the burner, the flashback limit drops back to 55% due to the interaction between the flame and the bottom of the
cooking pot. Under steady operating condition, no significant difference in cooking time or efficiency with up to 50% hydrogen
addition is observed, which is encouraging regarding use of hydrogen in cooking appliances.
• With 75% hydrogen addition, the burner temperature increases by 25C compared to pure natural gas. While this may not be
significant, it may affect the durability of the burner.
• Combustion noise increases with hydrogen addition, especially when the hydrogen percentage in the fuel is higher than 40%.
• The NOX and CO emissions of the cooktop burner are over 100 ppm (@3% O 2), NO and NO2 produced by the cooktop burner are
roughly at the same level, thus NO2 is a significant portion of the overall NOX. The NH3 and N2O emissions of the cooktop burner
are negligible, even during transient operation. Oxygen, carbon dioxide, and calorific corrections are compared with each other.
Oxygen correction and calorific correction have similar trend presenting emissions, however, carbon dioxide correction makes the
emission levels look higher for high hydrogen percentage fuel.

Major conclusions drawn include:

• Due to the unsteady nature of ignition, flashback limits the amount of hydrogen that can be added. This is exacerbated because the
design for natural gas is not able to accommodate the wide flammability range when hydrogen is added. Therefore, when designing
a burner for hydrogen-rich fuel, larger safety margin needs to be considered for the ignition process.
• Although the volumetric heating value of hydrogen is much lower than natural gas, they have similar Wobbe Index, which makes
it possible to replace part of the pipeline natural gas without great influence on the heating load of combustion devices.
• In general, the results of this study indicate that naturally aspirated natural gas cooktop burners can operate safely and efficiently
with up to 20% hydrogen added to the fuel. Noteworthy is that added hydrogen results in a reduction of pollutant emissions.
• A need exists for standards associated with testing appliances for interchangeability. For example, the basis for presentation of
emission results can influence the conclusions drawn regarding the effect of hydrogen addition. Also, the great variation of the
hydrogen amount that can be added is observed when pots/pans are used on the amount, suggesting consideration of this situation
in any standards that evolve.

NOMENCLATURE
Symbols
𝜙 equivalence ratio
p1 pipeline gas pressure
p2 operating pressure of the combustion device
V1 flow velocity in the pipeline
V2 flow velocity at the utility orifice
A1 pipeline feature area
A2 area of the utility orifice
ρf fuel density
𝑉𝑓̇ fuel flow rate

© 2019 Elsevier Ltd. All rights reserved 15

International Journal of Hydrogen Energy

𝑞̇ heating load of the burner

p pressure drop across the utility orifice
sgf specific density
x% hydrogen percentage in the fuel mixture
cooking cooking efficiency
cP water heat capacity
m4 quarts mass of 4 quarts of water
T temperature increase of water
m mass change of water
he, water latent enthalpy of water
mfuel mass of fuel
Acronyms
NO nitric oxide
NO2 nitrogen dioxide
N2O nitrous oxide
NOX nitrogen oxides
UHC unburned hydrocarbon
Btu British thermal unit
scf standard cubic foot
LHV lower heating value
HHV higher heating value
P2G power to gas

HIGHLIGHTS
• Hydrogen injection testing is conducted on a representative cooktop burner.
• Up to 15% hydrogen added to natural gas has little impact on cooking performance and operability.
• Flashback during ignition is the limiting factor for the amount of hydrogen added.
• The hydrogen injection limit is significantly lower than that proposed in a previous study.

ACKNOWLEDGMENTS
The authors would also like to acknowledge David Morales, Max Venaas for the help on the test stand setup. The financial support
for this project is from the California Energy Commission (Contract PIR‐16‐017).

REFERENCES
[1] Midilli A, Dincer I. Hydrogen as a renewable and sustainable solution in reducing global fossil fuel consumption. International
Journal of Hydrogen Energy. 2008 Aug 1;33(16):4209-22. https://www.sciencedirect.com/science/article/pii/S0360319908005302
[2] Steinberg M, Cheng HC. Modern and prospective technologies for hydrogen production from fossil fuels. International Journal of
Hydrogen Energy. 1989 Jan 1;14(11):797-820. https://www.sciencedirect.com/science/article/pii/0360319989900189
[3] Sheffield JW. Energy security through hydrogen. Assessment of Hydrogen Energy for Sustainable Development 2007 (pp. 1-8).
Springer, Dordrecht. https://link.springer.com/chapter/10.1007/978-1-4020-6442-5_1
[4] Das D, Veziroǧlu TN. Hydrogen production by biological processes: a survey of literature. International journal of hydrogen energy.
2001 Jan 1;26(1):13-28. https://www.sciencedirect.com/science/article/pii/S0360319900000586
[5] Sherif SA, Barbir F, Veziroglu TN. Wind energy and the hydrogen economy—review of the technology. Solar energy. 2005 May
1;78(5):647-60. https://www.sciencedirect.com/science/article/pii/S0038092X05000332
[6] Muradov NZ, Veziroğlu TN. “Green” path from fossil-based to hydrogen economy: an overview of carbon-neutral technologies.
International Journal of Hydrogen Energy. 2008 Dec 1;33(23):6804-39.
https://www.sciencedirect.com/science/article/pii/S036031990801118X
[7] Sorgulu F, Dincer I. Cost evaluation of two potential nuclear power plants for hydrogen production. International Journal of
Hydrogen Energy. 2018 Jun 7;43(23):10522-9. https://www.sciencedirect.com/science/article/pii/S0360319917342714
[8] Götz M, Lefebvre J, Mörs F, Koch AM, Graf F, Bajohr S, Reimert R, Kolb T. Renewable Power-to-Gas: A technological and
economic review. Renewable energy. 2016 Jan 1;85:1371-90.
https://www.sciencedirect.com/science/article/pii/S0960148115301610

© 2019 Elsevier Ltd. All rights reserved 16

International Journal of Hydrogen Energy

[9] Stansberry J, Mejia AH, Zhao L, Brouwer J. Experimental analysis of photovoltaic integration with a proton exchange membrane
electrolysis system for power-to-gas. International Journal of Hydrogen Energy. 2017 Dec 28;42(52):30569-83.
https://www.sciencedirect.com/science/article/pii/S0360319917342763
[10] Colbertaldo P, Agustin SB, Campanari S, Brouwer J. Impact of hydrogen energy storage on California electric power system:
Towards 100% renewable electricity. International Journal of Hydrogen Energy. 2019 Jan 5.
https://www.sciencedirect.com/science/article/pii/S0360319918336449
[11] Winter CJ. Into the hydrogen energy economy—milestones. International Journal of Hydrogen Energy. 2005 Jul 1;30(7):681-5.
https://www.sciencedirect.com/science/article/pii/S0360319904004379
[12] Zhao L, Brouwer J. Dynamic operation and feasibility study of a self-sustainable hydrogen fueling station using renewable energy
sources. International journal of hydrogen energy. 2015 Mar 16;40(10):3822-37.
https://www.sciencedirect.com/science/article/pii/S036031991500083X
[13] Mottaghizadeh P, Santhanam S, Heddrich MP, Friedrich KA, Rinaldi F. Process modeling of a reversible solid oxide cell (r-SOC)
energy storage system utilizing commercially available SOC reactor. Energy Conversion and Management. 2017 Jun 15;142:477-
93. https://www.sciencedirect.com/science/article/pii/S0196890417302121
[14] Karagöz Y, Balcı Ö, Köten H. Investigation of hydrogen usage on combustion characteristics and emissions of a spark ignition
engine. International Journal of Hydrogen Energy. 2019 Feb 8.
https://www.sciencedirect.com/science/article/pii/S0360319919302794
[15] Lee, J., Park, C., Kim, Y., Choi, Y., Bae, J. and Lim, B., 2019. Effect of turbocharger on performance and thermal efficiency of
hydrogen-fueled spark ignition engine. International Journal of Hydrogen Energy, 44(8), pp.4350-4360.
https://www.sciencedirect.com/science/article/pii/S0360319918340813
[16] Funke HW, Beckmann N, Abanteriba S. An overview on dry low NOx micromix combustor development for hydrogen-rich gas
turbine applications. International Journal of Hydrogen Energy. 2019 Feb 14.
https://www.sciencedirect.com/science/article/pii/S0360319919303192
[17] Nam J, Lee Y, Joo S, Yoon Y, Yoh JJ. Numerical analysis of the effect of the hydrogen composition on a partially premixed gas
turbine combustor. International Journal of Hydrogen Energy. 2019 Feb 2.
https://www.sciencedirect.com/science/article/pii/S0360319919301685
[18] Melaina, M.W., Antonia, O. and Penev, M., 2013. Blending hydrogen into natural gas pipeline networks: a review of key issues
(No. NREL/TP-5600-51995). National Renewable Energy Lab.(NREL), Golden, CO (United States).
https://www.nrel.gov/docs/fy13osti/51995.pdf
[19] Deymi-Dashtebayaz M, Ebrahimi-Moghadam A, Pishbin SI, Pourramezan M. Investigating the effect of hydrogen injection on
natural gas thermo-physical properties with various compositions. Energy. 2019 Jan 15;167:235-45.
https://www.sciencedirect.com/science/article/pii/S0360544218321832
[20] Laster WR, Hardes JW, Bartley RH, Gaviria M, inventors; Siemens Energy Inc, assignee. Fuel injector including tandem vanes for
injecting alternate fuels in a gas turbine. United States patent application US 15/540,784. 2018 Jan
25.https://patents.google.com/patent/US20180023812A1/en
[21] INOUE K, MIYAMOTO K, DOMEN S, TAMURA I, KAWAKAMI T, TANIMURA S. Development of Hydrogen and Natural
Gas Co-firing Gas Turbine. Mitsubishi Heavy Industries Technical Review. 2018 Jun;55(2):1.
http://www.mhi.co.jp/technology/review/pdf/e552/e552160.pdf
[22] https://www.eia.gov/dnav/ng/ng_cons_sum_dcu_nus_a.htm
[23] Turnheim B, Geels FW. Regime destabilisation as the flipside of energy transitions: Lessons from the history of the British coal
industry (1913–1997). Energy Policy. 2012 Nov 1;50:35-49.
https://www.sciencedirect.com/science/article/pii/S0301421512003655
[24] Berry WM, Brumbaugh IV, Moulton GF, Shawn GB. Design of atmospheric gas burners. Govt. print. off.; 1921.
https://nvlpubs.nist.gov/nistpubs/nbstechnologic/nbstechnologicpaperT193.pdf
[25] John H. Eiseman, Elmer R. Weaver, and Francis A. Smith. Method for Determining the Most Favorable Design of Gas Burners.
Bureau of Standards Journal of Research, 1932. Vol. 8. https://nvlpubs.nist.gov/nistpubs/jres/8/jresv8n6p669_A2b.pdf
[26] Marchetti C. Hydrogen and energy. Chemical Economy and Engineering Review. 1973 Jan;5(1):7-25.
http://wilsonweb.physics.harvard.edu/HUMANRIGHTS/PHYSICS/ENERGY/MARCHETTI-008.pdf
[27] Turns SR. An introduction to combustion. New York: McGraw-hill; 1996 Jan. http://jmargolin.com/flame/refs/ref12_turns.pdf
[28] Chalke HD, Dewhurst JR, Ward CW. Loss of Sense of Smell in Old People. A Possible Contributory Factor in Accidental Poisoning
from Town Gas. Public health. 1958;72(6):223-30. https://www.cabdirect.org/cabdirect/abstract/19592700016
[29] Kreitman N. The coal gas story. United Kingdom suicide rates, 1960-71. Journal of Epidemiology & Community Health. 1976 Jun
1;30(2):86-93. https://jech.bmj.com/content/30/2/86.short
[30] Rau JG, Wooten DC. Environmental impact analysis handbook. InEnvironmental impact analysis handbook 1980. McGraw-Hill.
https://www.osti.gov/biblio/5329087

© 2019 Elsevier Ltd. All rights reserved 17

International Journal of Hydrogen Energy

[31] Rao A. Sustainable Energy Conversion for Electricity and Coproducts: Principles, Technologies, and Equipment. John Wiley &
Sons; 2015 Apr 27.
https://books.google.com/books?hl=en&lr=&id=V3JuCAAAQBAJ&oi=fnd&pg=PA230&ots=CnVKbilZQq&sig=Rkfu5Tb004J
ylwQVeaLXNl99tX4#v=onepage&q&f=false
[32] Glassman I, Yetter RA, Glumac NG. Combustion. Academic press; 2014 Dec 2.
https://pennstate.pure.elsevier.com/en/publications/combustion-fifth-edition
[33] Baukal Jr CE, editor. Industrial burners handbook. CrC press; 2003 Oct 29. https://www.taylorfrancis.com/books/9780203488805
[34] Borman GL, Ragland KW. Combustion engineering. New York: McGraw-Hill; 1998 Jan.
http://www.worldcat.org/title/combustion-engineering/oclc/37370580
[35] Weaver ER. Formulas and graphs for representing the interchangeability of fuel gases. Journal of Research of the National Bureau
of Standards. 1951 Mar;46(3):213-45. https://nvlpubs.nist.gov/nistpubs/jres/46/jresv46n3p213_A1b.pdf
[36] ANSI Z21.1-2016/CSA 1.1-2016 - Household cooking gas appliances.
https://webstore.ansi.org/Standards/CSA/ANSIZ212016CSA
[37] Jones DR, Al-Masry WA, Dunnill CW. Hydrogen-enriched natural gas as a domestic fuel: an analysis based on flash-back and
blow-off limits for domestic natural gas appliances within the UK. Sustainable Energy & Fuels. 2018;2(4):710-23.
https://pubs.rsc.org/en/content/articlehtml/2017/se/c7se00598a
[38] Oberascher C, Stamminger R, Pakula C. Energy efficiency in daily food preparation. International Journal of Consumer Studies.
2011 Mar;35(2):201-11. https://onlinelibrary.wiley.com/doi/full/10.1111/j.1470-6431.2010.00963.x
[39] Zhao Y, Choudhury S, McDonell V. Influence of Renewable Gas Addition to Natural Gas on the Combustion Performance of
Cooktop Burners. In ASME 2018 International Mechanical Engineering Congress and Exposition 2018 Nov 9 (pp. V08AT10A012-
V08AT10A012). American Society of Mechanical Engineers.
https://proceedings.asmedigitalcollection.asme.org/proceeding.aspx?articleid=2722161
[40] KEMA, Inc. 2009 California Residential Appliance Saturation Study. California Energy Commission. 2010. Publication number:
CEC‐ 200‐2010‐004‐ES. https://www.energy.ca.gov/2010publications/CEC-200-2010-004/CEC-200-2010-004-ES.PDF
[41] South Coast Air Quality Management District. 1982. Rule 407: Liquid and Gaseous Air Contaminants.
http://www.aqmd.gov/docs/default-source/rule-book/rule-iv/rule-407.pdf
[42] South Coast Air Quality Management District. 1981. Rule 409: Combustion Contaminants. http://www.aqmd.gov/docs/default-
source/rule-book/rule-iv/rule-409.pdf
[43] South Coast Air Quality Management District. 1981. Rule 474: Fuel Burning Equipment: Oxides of Nitrogen.
http://www.aqmd.gov/docs/default-source/rule-book/rule-iv/rule-474.pdf
[44] South Coast Air Quality Management District. 2014. Rule 1153.1: Emissions of Oxides of Nitrogen from Commercial Food Ovens.
http://www.aqmd.gov/docs/default-source/rule-book/reg-xi/rule-1153-1-emissions-of-oxides-of-nitrogen-from-commercial-food-
ovens.pdf
[45] RULE 69.5.1 Natural gas-fired Water Heaters. https://www.arb.ca.gov/drdb/sd/curhtml/r69-5-1.pdf
[46] ANSI Z21.10.1-2014-Gas Water Heaters. https://www.scc.ca/en/standardsdb/standards/27117
[47] ANSI Z21.86-2016/CSA 2.32-2016 - Vented Gas-Fired Space Heating Appliances.
https://webstore.ansi.org/Standards/CSA/ANSIZ21862016CSA32
[48] ANSI Z21.1.2-2016 - Gas-Fired Room Heaters. https://webstore.ansi.org/Standards/CSA/ANSIZ21112016
[49] Lefebvre AH, Ballal DR. Gas turbine combustion: alternative fuels and emissions. CRC press; 2010 Apr 26.
https://www.taylorfrancis.com/books/9781420086058
[50] Jahnke JA. Continuous emission monitoring. John Wiley & Sons; 2000 Apr 3. http://uci.worldcat.org/search?q=isbn%3A0-471-
29227-3
[51] Singer BC, Apte MG, Black DR, Hotchi T, Lucas D, Lunden MM, Mirer AG, Spears M, Sullivan DP. Natural gas variability in
California: Environmental impacts and device performance experimental evaluation of pollutant emissions from residential
appliances. Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States); 2009 Dec 1.
https://www.osti.gov/biblio/980736

© 2019 Elsevier Ltd. All rights reserved 18

View publication stats