You are on page 1of 8

Applied Thermal Engineering 45-46 (2012) 1e8

Contents lists available at SciVerse ScienceDirect

Applied Thermal Engineering


journal homepage: www.elsevier.com/locate/apthermeng

Liquid butane as an alternative fuel for diesel oil burners


Alejandro Sez a, Alex Flores-Maradiaga b, Mario Toledo a, *
a
b

Department of Mechanical Engineering, Universidad Tcnica Federico Santa Maria, Av. Espaa 1680, Valparaiso, Chile
cole de Technologie Suprieure, 1100 rue Notre-Dame Ouest, Montral, Qubec, Canada

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 9 January 2011
Accepted 9 April 2012
Available online 17 April 2012

The experimental tests performed to study the combustion process of liquid butane employing a diesel
oil burner are presented. For these tests, a dual pumping and injection system was designed to operate
with pressures varying from 0.8 to 2.0 MPa. Five distinctive cases were tested for each fuel, obtaining
a complete characterization of the combustion processes in comparable conditions. Flame geometries,
temperatures and the main chemical products of the combustion process were recorded experimentally
for both liquid butane and diesel oil. It was observed that liquid butane ames present elongated conical
shapes, with low radiation cyan color at the base position, followed by a higher radiation zone in the core
and ame front positions. Also, the temperatures and NOx concentrations of liquid butane ames are
lower than those of diesel oil ames. In general, it is feasible to modify the combustion technology of
diesel oil burners to use liquid butane as an alternative fuel.
2012 Elsevier Ltd. All rights reserved.

Keywords:
Liquid butane
Diesel oil
Dual injection
Burner
Combustion

1. Introduction
In the last two decades, with the worldwide energy crisis, the
price of diesel oil has sharply increased, imposing a huge economic
burden to produce power at a reasonable cost for residential needs,
commercial facilities and industrial production. This situation
demands a constant search for alternatives that ensure a highly
efcient power generation at the lowest possible cost. Thus, it is
vital to analyze the feasibility of modifying the technology
currently used for combustion processes to introduce alternative
fuels that substitute diesel oil.
Diesel and fuel oil are the most consumed fuels in the industrial
facilities and thermal power plants, which represents a major
problem due to the permanent price increments and high speculation in oil markets. Additionally, as pointed out by Martyr and
Plint [1], the hazardous pollution caused by handling and
consumption of diesel oil has contributed to the rapid environmental degradation, which turns unfeasible the use of these fuels in
the nearby future. Therefore, an opportunity to introduce alternative fuels has motivated detailed studies of biofuels, fuel blends,
polluting exhaust emissions and comparative analysis of fuel prices.
Generally speaking, biofuels seem to be attractive substitutes to
petroleum-based fuels, mostly because the majority of biofuels are
obtained from renewable sources and the greenhouse effect gases
are reduced. Nonetheless, experimental studies, such as the ones
* Corresponding author. Tel.: 56 32 2654162; fax: 56 32 2797472.
E-mail address: mario.toledo@usm.cl (M. Toledo).
1359-4311/$ e see front matter 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.applthermaleng.2012.04.024

presented by Yoon and Lee [2] and Crookes et al. [3], have proven
that some biofuels, under certain conditions, may retard the ignition phase and reduce the combustion performance, compromising
any cost-effective power generation. In the case of biodiesel blends
from vegetable oils, Agarwal et al. [4] explain that, without much
chemical processing, usually these blends present high viscosities
and fairly low volatility, causing high droplet size due to poor fuel
atomization and jet-mixing. Hence, operational performance and
durability of combustion technology are severely reduced.
Also, as claried by Atsbury [5], the high soot formation from the
combustion of some biodiesel blends and the ammable/explosive
nature of biogases, bioethanols and hydrogen-based fuels may
become safety hazards and environmental drawbacks to their
complete approval for commercial thermal machines. Nonetheless,
with the appropriate treatment and manipulation, biofuels still are
good alternatives which mostly improve in the emission reduction
of carbon monoxide (CO), unburned hydrocarbons (HC) and
particulate material (PM) as demonstrated in the studies by Sayin
and Canakci [6] and Rodrigues de Souza et al. [7].
From an economic point of view, an interesting result presented
by Demirbas [8] indicates that, even though biodiesel blends have
lower economic performance than diesel oil (in terms of plant
capacity, process technology, raw material cost and chemical costs),
they are strong competitors of natural gas and diesel oil for urban
transportation and industries. Demirbas also remarks that biofuels
provide the prospect of new economic opportunities by creating
jobs for people in rural areas of developing countries, which usually
are highly dependent in oil consumption. However, Chang and Su

A. Sez et al. / Applied Thermal Engineering 45-46 (2012) 1e8

[9] clearly explain that the great paradox of the conventional biofuel production and consumption, in favorable periods of high oil
prices, is the shortage of basic foods used as raw material (e.g., corn
and soybean) that are fundamental part of the food security of
humans and livestock. For that reason, they stress the need for
a careful choice of the raw material to be used for biofuel
production and suggest the selection of inedible algae-based
feedstock.
Other researchers have focused on low sulfur gaseous hydrocarbons, such as n-butane, propane and natural gas, which generally are liqueed to be injected in dual-fuel technology. Studies by
Huang and Sung [10], Huang et al. [11], Lee et al. [12] and Yamasaki
and Iida [13] have identied n-butane as one of the main substitute
fuels for diesel oil, mostly due to its similar thermo-physical characteristics (important factor for dual-fuel systems) coupled with
the high combustion efciency and low emissions achieved by the
two-stage autoignition process (i.e., low heat release stage, LTR, and
high heat release stage, HTR).
In fact, n-butane (C4H10) is the fuel with lowest carbon number
of the parafn family found to interchange well in combustion
technology with higher hydrocarbons (such as diesel oil), and has
proven to reduce signicantly the CO and PM concentrations in
exhaust gases [12]. Additionally, the small carbon number of
n-butane is indicative of the fast chemical kinetics of its combustion
process which aids the speed-up for autoignition and improves
combustion performance. However, this last characteristic has
challenged designers and combustion engineers, who need to
control properly the ignition timing and combustion duration at
very high temperatures and injection velocities [13].
Liquid n-butane (hereafter referred to as liquid butane) has
inspired different inventions over the past 60 years (e.g., Schylander [14], Pillard [15] and Kaufman [16]) and is becoming the
mainstream of substitute fuels in oil combustion technology [17,18].
As Fleisch et al. [18] point out, just until recent years gas-derived
fuels could not compete with conventional petroleum-based
liquid fuels because gas-to-liquid conversion has been and still is
an expensive process. But now, a signicant cost reduction from
improved technologies and economies of scale has raised the
competitiveness of liqueed gas fuels for commercial applications.
Either for small-scale indoor purposes (such as the case presented
by Ghosn et al. [19]) or large-scale industrial purposes (such as
metal piece heating discussed by Wang et al. [20]), commercial
developers and users can benet from the quick gasication and
ignition of liquid butane for a more efcient combustion.
Consequently, the main objective of this study is to design and
test of a dual system for liquid butane and diesel oil injection in
a Joannes diesel oil burner, aiming for the optimization of this
combustion technology to facilitate the introduction of liquid
butane as the alternative fuel. For that purpose, rstly, the fuel
injection, jet-mixing and ignition systems are fully studied, and
a new system of liquid butane injection complementary to that of
diesel oil is proposed. Secondly, the experimental setup is implemented on a horizontal isothermal vessel, which includes instrumentation for mass ux, pressure and temperature measurements,
ue-gas analysis, PLC automation, solenoid valves, water pumping,
fuel injection and security systems. Finally, diesel oil and liquid
butane combustion are investigated, respectively.
This research article presents the nal results obtained with the
new dual injection system. In Section 2 the general set-up of the
testing bench is detailed. Then, the simplied radiation model used
for thermal performance assessment is presented. The experimental results for both fuels are discussed in Section 3, emphasizing on the main features and benets of liquid butane
combustion. Finally, in Section 4 the concluding remarks are
presented.

2. Experimental setup and methodology


2.1. Isothermal reactor and instrumentation
For this study, a horizontal vessel of 550 mm in diameter and
1650 mm long, with a 180 mm diameter exhaust pipe, was adapted
to operate as an isothermal reactor. Its furnace interior, which has
high alumina refractory coating, is surrounded by a cooling water
circuit which maintains a stable temperature prole, aided by
a two-step heat exchanger that removes heat transported by the
cooling water. Additionally, the AZ14 Joannes diesel oil burner,
designed to generate a maximum power of 180 kW, was installed to
test the dual-fuel system.
The schematic design of the dual injection system is presented
in Fig. 1. Diesel oil is stored in a 45 L tank and is injected with
a Suntec fuel pump at 0.8, 0.9 and 1.0 MPa. Liquid butane is fed
from a low-pressure cylinder (0.25 MPa) into a variable ux fuel
pump that rises the fuel pressure between 1.0 and 2.0 MPa. These
high-pressure lines are connected to a liquid accumulator and
pressure reduction valves that adjust the nal injection pressure.
The fuel ow is controlled primarily by automated solenoid valves
which obey to a sequence program launched by a PLC and monitored from a remote computer. Whenever a change in time or
operating sequence is needed, the program is modied, recompiled
and reloaded into the PLC.
Diesel oil and liquid butane ow rates are measured by a rotameter calibrated in situ, and a digital volumetric ow-meter is used
in the cooling water circuit as well. Also, to adapt the power output
to the reactors capacity, the original burner nozzle of 3 GPH was
replaced for a 1.75 GPH nozzle. This is a major change in the burners
system based on the need of a low power range control. For this
experiments, a Danfoss pressure-swirl nozzle (of 2.94 kg/h, HR
hollow cone, 60 degrees spray) was employed to obtain the desired
1.75 GPH fuel injection. A schematic diagram of this nozzle, with its
components and dimensions, is presented in Fig. 2.
Five type R thermocouples of 0.75 mm in diameter and
platinumerhodium coating (13% rhodium) were used to scan
the furnace wall temperatures along the combustion chamber.
Also, a type K mobile thermocouple (1.5 mm in diameter and
chromeealumel coating) was installed to measure the ame
temperature at ve positions along the centerline and ue-gas
temperatures at 1500 mm from the exhaust pipe entrance. The
ame and wall temperatures at the burners exit (rst position,
0 m) were obtained with a LumaTech infrared pyrometer
(measuring range: 30 to 750  C) and two ue-gas meters (TESTO
350-S and Bosch BEA 250-EU) were used in order to assess and
control combustion parameters in real time. A smoke meter, Bosch
EFAW 68A, was used to measure soot emissions.
For the ame geometry characterization and monitoring,
a Sony HDR-FX1000 professional camcorder was employed for
multiple imaging and footage recording. The captured images were
post-processed with the Adobe Premier Pro toolkit (Photoshop
CS5), using the furnaces dimensions as reference for the sizing of
the ames, obtaining the integral scales of length and radius at the
ames base, core and front positions discussed in Section 3.
The operation sequence of this experimental hybrid system was
dened as follows:
 Air pre-cleaning of the combustion chamber is set to 10 s to
drag any remaining unburned hydrocarbons from previous
tests. In case of an emergency, the sweeping procedure
continues for 3 min;
 Turning-on the ignition spark is set for 13 s;
 Preheating the reactor and nozzle with a diesel oil ame is
done for 35 min;

A. Sez et al. / Applied Thermal Engineering 45-46 (2012) 1e8

Fig. 1. Design of the hybrid system for dual injection of liquid butane and diesel oil.

 Dual injection and combustion of diesel oil and liquid butane is


done for 5 min;
 Combustion of liquid butane, waiting 25 min for stabilization
between each case, and collecting relevant data is set for
30 min; and,
 The isothermal reactor is cooled for 60 min before shutting
down the system.

axisymmetric cylindrical ame concentrically aligned within the


cylindrical reactor vessel. Gunn and Horton [21] remarked that for
cylindrical furnace industrial boilers the dominant heat production
mechanism is thermal radiation. Thus, equation (1) is used to
calculate the mean heat transfer due to thermal radiation of the
ame q_ f , which corresponds to the net energy absorbed by the
reactors inner surface at each position, as presented by engel and
Ghajar [22] for a simplied ame radiation model.

2.2. Thermal performance calculations


Thermal assessment of the combustion process is done
employing a simple heat radiation model, by assuming a steady

q_ f

1=3 f



4
Af s Tf4  Tw


1=aw  1 rf =rw

(1)

Here, the constant of StefaneBoltzmann (s) is set to 5.67 


108 W/m2K4, and the mean ame and wall temperatures (Tf and
Tw, respectively) are averaged over 30 min of data collected every
10 s at each of the ve control positions. Also, a constant ame
emissivity (3 f) of 0.8 for diesel oil ames and 0.5 for liquid butane
ames, and a constant wall spectral absorption coefcient (aw) of
0.8, are adopted for all the cases. This model considers a unity form
factor Ff /w 1, and the ratio of the ames longitudinal surface
area (Af 2prfl) to the furnaces internal surface area (Aw2prwl) is
reduced to the ratio of ame to wall radius (rf/rw), taking the same
ame length scale (l) in each case.
Secondly, the net rate of heat transported by water q_ H2 O that
circulates in the cooling circuit of the reactors shell is calculated by
the general relationship expressed in equation (2).

_ H2 O cpH2 O DTH2 O
q_ H2 O m

Fig. 2. Pressure-swirl nozzle components and dimensions. Source: Danfoss HR/SR oil
nozzle general datasheet.

(2)

_ H2 O and the
Here, the mean ow rate of cooling water m
temperature difference between the inlet and outlet DTH2 O DTo 
DTi are averaged over 30 min for each case. The water specic heat
coefcient at constant pressure is set to 4184 kJ/kg  C.
Lastly, equation (3) represents the thermal absorption efciency
(h), calculated from the ratio of net heat transfer of cooling water to
the mean ame radiation at each position.

A. Sez et al. / Applied Thermal Engineering 45-46 (2012) 1e8

q_ H2 O
q_ f

(3)

3. Experimental results and discussion


The thermo-physical properties of both fuels are presented in
Table 1. In general, the following assessment compares the
combustion performance based on the ames heat release, ue-gas
emissions and the adaptability of the dual injection system to
a diesel oil burner. The experimental tests are aimed to obtain results
from pre-dened input parameters (i.e., fuel injection pressure and
air excess) that yield useful output information for the analysis.
Under these operating conditions, ve distinctive cases for each fuel
were performed (see Table 2) in order to trace a characteristic
behavior for their particular combustion processes. For this purpose,
measurements of ame and wall temperatures, ame imaging and
ue-gas analysis were achieved. Note that, except for the cases 3A
and 1B with 1.0 MPa, the injection pressure ranges are different for
both fuels due to technological limitations of the testing bench. The
main difculty encountered was the high instability of liquid butane
for high-pressure variable ow-rate pumping which directly
affected the fuel injection pressure. No doubt this system can be
improved to operate with a lower pressure range for liquid butane,
but an adequate testing pump must be designed to handle such low
ow rates in order to obtain a continuous fuel jet.
In the rst three cases for both fuels the inuence of changing
the injection pressure was observed, which permitted modications in the fuel atomization; then, the last two cases were aimed to
observe the oxidizer inuence (air excess) in extreme operating
conditions. Each of these cases was observed during 30 min,
allowing the temperature and fuel ow rates to reach a quasi-steady level before collecting data. The experiments were performed within a temperature range of 16e20  C and a humidity
range of 70e80%. At the starting point, all measuring equipment
have been adjusted and double checked to minimize errors in the
experimental results.

Table 2
Tested cases for diesel oil and liquid butane.
Fuel

Case

Pressure
(MPa)

Flow rate
(L/min)

Remarks

Diesel oil
Diesel oil
Diesel oil

1A
2A
3A

0.8
0.9
1.0

0.124
0.126
0.138

Diesel oil

4A

0.9

0.141

Diesel oil

5A

0.9

0.160

Liquid butane
Liquid butane
Liquid butane

1B
2B
3B

1.0
1.5
2.0

0.137
0.154
0.179

Liquid butane

4B

2.0

0.134

Liquid butane

5B

2.0

0.188

Constant air on
blower scale to observe
injection pressure
inuence for diesel
oil (D.O.)
More air at constant
pressure for D.O.
Few air at constant
pressure for D.O.
Constant air on blower
scale to observe
injection pressure
inuence for liquid
butane (L.B.)
More air at constant
pressure for L.B.
Few air at constant
pressure for L.B.

From Fig. 3b, it is evident that geometries of liquid butane


ames are different from diesel oil ames in size and shape. Under
the setting conditions listed in Table 2, liquid butane ames present
larger sizes and more elongated conical shapes with respect to

3.1. Flame geometries and general characteristics


The diameter and length scales of diesel oil and liquid butane
ames are shown in Fig. 3a and b, respectively. It is clearly observed
that for diesel oil combustion, with constant air input and variable
injection pressure, case 3A produces a large ame due to higher fuel
consumption. In contrast, keeping a constant injection pressure
with variable air input, case 4A produces a shorter ame due to an
increase of air excess, and case 5A produces the largest ame with
signicantly higher fuel consumption. In general, diesel oil ames
display a turbulent yellow-orange color indicative of high thermal
and infrared radiation levels. Hence, it was noticed that as the
fuel injection pressure raised the ame size increased and the
yellow-orange colour became more intense, related to larger fuel
droplets and more soot formation.

Table 1
Thermo-physical characteristics of diesel oil and liquid butane.
Property

Diesel oil

Liquid butane

Density (at 20  C), kg/m3


Viscosity (at 20  C), mm2/s
Molecular weight, kg/kmol
Higher heat value, kJ/kg
Lower heat value, kJ/kg
Vaporization latent heat, kJ/kg
Specic heat (at 772 K), kJ/kg K
Thermal conductivity, W/mK

840
5.5  106
142.28
48,020
44,602
277
3.26
0.15

585
2.42  104
58.12
49,546
45,742
386
2.4
0.014

Fig. 3. Lengths and radius of diesel oil (a) and liquid butane (b) ames.

A. Sez et al. / Applied Thermal Engineering 45-46 (2012) 1e8

analogous diesel oil ames. In all ve cases tested for liquid butane,
the ames kept the same geometry, characterized by a cyan-blue
colour cone about 20 cm long at the base followed by a turbulent
yellow-orange jet ame. At constant air excess, case 3B has a larger
ame due to the increase of fuel consumption and, with constant
injection pressure, case 4B yields a shorter ame with more air
excess and case 5B gives the largest ame as little air and more fuel
enter the combustion zone.
Also, referring to Table 2, one can verify that any change of the
ame dimensions is tied to its corresponding fuel consumption
rate. At constant air, whichever increase in injection pressure yields
a larger ame and, as a direct consequence, increases the heat
release. Also, keeping a constant injection pressure, an increment of
air excess generates a shorter ame, and vice versa, affecting the
heat release in the same proportion, which will be discussed in
Section 3.2.
The cyan-blue colour cone at the base of the liquid butane
ames indicates that part of the droplet stream evaporates very
quickly, changing to gas phase combustion almost instantaneously
as it enters the combustion zone. This quick evaporation is highly
desirable in order to obtain autoignition and a good combustion
performance. Yet, a small yellow semi-sphere was observed at the
base of the ame, possibly caused by incomplete combustion of
residual olens in the fuel stream. The second portion of the liquid
butane ames behave in a similar way to diesel oil ames by
generating high radiation levels, except that the length is doubled
in most cases. This phenomenon reveals that the combustion of
liquid butane may be incomplete at high injection pressures in
addition to the reactors limited volume, which is an important
factor that should be considered for commercial purposes.
3.2. Thermal assessment of the combustion process
The ame and wall temperature proles for cases with diesel oil
are illustrated in Fig. 4a and b, respectively. The maximum
temperature for diesel oil ames is 1096 K (obtained in case 3A)
and the maximum temperature decay is approximately 350 K at the
ame frontal position. The maximum wall temperature is 748 K for
case 4A, with a maximum temperature decay of 150 K. Most of the
temperature proles show a peak temperature at a relative distance
(x/L) between 0.4 and 0.5 from the injection nozzle. Based on
these experimental results for diesel oil, it is clear that wall
temperatures follow similar trends to ame temperatures, thus,
anticipating that the maximum heat transfer should be expected at
the same relative position (x/L w0.4).
Fig. 5a and b presents the ame and wall temperature proles,
respectively, for liquid butane combustion. The maximum ame
and wall temperatures are 974 K and 668 K, respectively, both
obtained in case 5B. Most of the ame temperature proles show
a peak temperature at the second half of the ame (x/L w0.6),
which is an expected based on the ame geometries. Hence, during
the quasi-steady operation with liquid butane, the temperatures
were better distributed along the reactor compared to those of
diesel oil combustion, mainly due to the elongated liquid butane
ames which cover more heat transfer area of the furnace. Nevertheless, liquid butane ames generate temperatures 120 K cooler
than the diesel oil ames, hence, it is expected to obtain slightly less
heat radiation with liquid butane.
To obtain a more insightful assessment of the thermal performance for both fuels, we can apply equations (1)e(3) presented in
Section 2.2 to compare the heat absorption efciency of the cooling
water with respect to the ames energy release. Fig. 6 presents the
ame radiation generated with both fuels. It is readily veried that
the peak thermal radiation for diesel oil and liquid butane occur at x/
L w0.4 and 0.6, respectively, and their magnitudes are very close.

Fig. 4. Flame (a) and wall (b) temperature proles of diesel oil combustion.

Also, one can observe that, under the same operating conditions
(cases 3A and 1B), the energy release is slightly better distributed for
liquid butane ames.
Fig. 7 depicts the heat absorption efciency at ve positions for
cases with an injection pressure of 1.0 MPa at constant air excess for
diesel oil and liquid butane (cases 3A and 1B, respectively). This are
the only two cases that can be compared directly for the reasons
explained in Section 2, but are sufcient to get a good idea of the
energy transfer phenomenon. For the rst half of the chamber,
Fig. 7 shows higher heat absorption for diesel oil combustion, then,
decaying rapidly in the second half. This behavior was the same for
all diesel oil ames, showing high heat concentration in the region
near the nozzle. In contrast, the absorption efciency of liquid
butane combustion is poor very near the nozzle and improves as
the jet ame develops downstream in the chamber. With this result
one can infer that, in general, heat transfer is better distributed for
liquid butane ames.
3.3. Combustion performance based on ue-gas analysis
A resourceful way to assess the combustion process of this dual
system is to measure the exhaust gases of each fuel. The mean

A. Sez et al. / Applied Thermal Engineering 45-46 (2012) 1e8

Fig. 7. Heat absorption efciency for diesel oil (case 3A) and liquid butane (case 1B).

Fig. 5. Flame (a) and wall (b) temperature proles of liquid butane combustion.

ue-gas results in each case are presented in Table 3 and


compared graphically with respect to the air-fuel ratio, lambda,
in Fig. 8.
Clearly, as the air excess increases, oxygen concentrations rise
(Fig. 8a) and carbon dioxide is reduced in a proportional rate
(Fig. 8b). Furthermore, one can observe these concentrations for
both fuels are very close with low air excess but start to separate
over lambda w1.6, showing less CO2 production with liquid
butane at high ue-gas temperatures. Nonetheless, there is relatively more carbon monoxide concentration (Fig. 8c) with liquid
butane ames, which underlines the need to improve the purity of
the n-butane fuel supply that may contain traces of heavier
hydrocarbons.
In addition, concentrations of NOx (Fig. 8d) display the expected
trends; i.e., both fuels yield high NOx levels for rich air-fuel
mixtures and low NOx concentration for poor oxidizing
mixtures. From Table 3, comparing NOx concentrations for both
fuels with 1.0 MPa (cases 3A and 1B), diesel oil combustion yields
a NOx level 6 times higher than liquid butane combustion. Thus, for
higher temperatures and fuel consumption, diesel oil ames
contribute more to the emission of green-house effect gases than
liquid butane ames.
Finally, the soot emissions (Fig. 8e) also rise with increasing
amounts of air excess, which indicates incomplete combustion for
both fuels. However, soot formation for liquid butane combustion it
is consistently lower than diesel oil combustion, which has
a signicant environmental impact for air quality control.

Table 3
Combustion products for diesel oil and liquid butane.

Fig. 6. Flame radiation for diesel oil (case 3A) and liquid butane (case 1B).

Case

Lambda

CO2 (%vol)

CO (ppm)

NOx (ppm)

Temperature (K)

1A
2A
3A
4A
5A
1B
2B
3B
4B
5B

1.37
1.53
1.4
1.78
1.59
1.88
1.4
1.42
2.29
1.56

10.1
11.06
10.9
8.39
9.67
7.5
10.39
11.22
6.3
9.92

17
23
20
10.9
13
13
29
27
7
17

288
310
295
254
273
54
68
68
22
56

771
713
745
823
788
695
700
751
745
773

A. Sez et al. / Applied Thermal Engineering 45-46 (2012) 1e8

Fig. 8. Concentrations of O2 (a), CO2 (b), CO (c), NOx (d) and soot (e) for both diesel oil and liquid butane combustion.

4. Conclusions
Innovation and deployment of the butaneediesel dual injection
system demonstrates it is possible to achieve an effective transformation of commercial diesel oil burners, maintaining good
combustion performance and low pollution levels. Based on the
experimental results, for equivalent operating conditions, the
combustion process of liquid butane yields a similar energy output
with less polluting emissions than the diesel oil combustion. Also,
with liquid butane ames, temperatures are better distributed
throughout the combustion chamber.

Hence, the key factor to obtain a proper interchange from


diesel oil-to-liquid butane in commercial burners is the adequate
control of liquid butanes ow rate to obtain an equivalent energy
output.

Acknowledgements
The authors wish to acknowledge the support of LIPIGAS
S.A. Enterprises and CONICYT-Chile under Fondecyt project
1121188.

A. Sez et al. / Applied Thermal Engineering 45-46 (2012) 1e8

References
[1] A.J. Martyr, M.A. Plint, Fuel and oil storage, supply and treatment, Engine
Testing (2007) 129e143. doi:10.1016/B978-075068439-2/50010-4.
[2] S.H. Yoon, C.S. Lee, Effect of biofuels combustion on the nanoparticle and
emission characteristics of a common-rail DI diesel engine, Fuel 90 (2011)
3071e3077.
[3] R.J. Crookes, F. Kiannejad, M.A. Nazha, Systematic assessment of combustion
characteristics of biofuels and emulsions with water for use as diesel engine
fuels, Energy Conversion and Management 38 (1997) 1785e1795.
[4] D. Agarwal, L. Kumar, A.K. Agarwal, Performance evaluation of a vegetable oil
fuelled compression ignition engine, Renewable Energy 33 (2008) 1147e1156.
[5] G.R. Atsbury, A review of the properties and hazards of some alternative fuels,
Process Safety and Environmental Protection 86 (2008) 397e414.
[6] C. Sayin, M. Canakci, Effects of injection timing on the engine performance and
exhaust emissions of a dual-fuel diesel engine, Energy Conversion and
Management 50 (2009) 203e213.
[7] G. Rodrigues de Souza, A. Moreira dos Santos, S. Lucas, K. Ribeiro, D. Luiz,
Evaluation of the performance of biodiesel from waste vegetable oil in a ame
tube furnace, Applied Thermal Engineering 29 (2009) 2562e2566.
[8] A. Demirbas, Biofuels securing the planets future energy needs, Energy
Conversion and Management 50 (2009) 2239e2249.
[9] T. Chang, H. Su, The substitutive effect of biofuels on fossil fuels in the lower
and higher crude oil price periods, Energy 35 (2010) 2807e2813.
[10] Y. Huang, C.J. Sung, Effects of n-butane addition on reformer gas combustion:
implications for potential of using reformer gas for an engine cold start,
Proceedings of the Combustion Institute 29 (2002) 759e765.
[11] J. Huang, Y. Wang, S. Li, A.P. Roskilly, H. Yu, H. Li, Experimental investigation on the performance and emissions of a diesel engine fuelled with

[12]

[13]

[14]
[15]
[16]
[17]
[18]

[19]

[20]

[21]
[22]

ethanolediesel blends, Applied Thermal Engineering 29 (2009)


2484e2490.
S. Lee, S. Oh, Y. Choi, K. Kang, Performance and emission characteristics of a CI
engine operated with n-butane blended DME fuel, Applied Thermal Engineering 31 (2011) 1929e1935.
Y. Yamasaki, N. Iida, Numerical analysis of autoignition and combustion of nbutane and air mixture in HCCI engine using elementary reactions, JSME
International Journal, Series B 46 (2003) 52e59.
R.L. Schylander, N. J. Hackettstown, Liquid butane or propane vaporizing
burner, US Patent 2,855,987, 1958.
J.C. Pillard, Burners of liquid fuels atomized by expansion of a compressed
auxiliary uid, US Patent 4,002,297, 1977.
R.L. Kaufman, Liquid butane fuel injection for internal combustion engines, US
Patent 5,033,444, 1991.
V.S. Babkin, Filtrational combustion of gases, present state of affairs and
prospects, Pure and Applied Chemistry 65 (1993) 335e344.
T.H. Fleisch, R.A. Sills, M.D. Briscoe, Emergence of the gas-to-liquid industry:
a review of global GTL developments, Journal of Natural Gas Chemistry 11
(2002) 1e14.
M. Ghosn, R. Flouty, N.A. Saliba, Emissions of nitrogen dioxide from butane gas
heaters and stoves indoors, American Journal of Applied Sciences 2 (2005)
707e710.
L. Wang, P. Wu, X. Ni, Combustion of liquid petroleum gas in a uidized bed
furnace with a jetting-mixing distributor, Powder Technology 170 (2006)
86e93.
D. Gunn, R. Horton, Industrial Boilers, Longman Scientic and Technical Press,
New York, 1989.
Y. engel, A. Ghajar, Heat and Mass Transfer, Fundamentals and Applications,
fourth ed. McGraw-Hill, 2010.