- Email: [email protected]
Investigation on the hydrogen production by methanol steam reforming with engine exhaust heat recovery strategy
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 4 9 5 7 e4 9 6 8
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/he
Investigation on the hydrogen production by methanol steam reforming with engine exhaust heat recovery strategy Cheng-Hsun Liao, Rong-Fang Horng* Department of Mechanical Engineering, Intelligent Vehicle Research Center, Kun Shan University, No.195, Kunda Rd., Yungkang Dist., Tainan City 710, Taiwan, ROC
article info
abstract
Article history:
In this study, a reformer with an engine exhaust heat exchanger was employed to produce
Received 26 November 2015
hydrogen. The purpose of this study is to investigate a methanol reformer with engine
Received in revised form
exhaust heat recovery to produce hydrogen with steam reforming of methanol (SRM). Steam
12 January 2016
reforming is an endothermic reaction requiring additional energy for reaction. Thus, engine
Accepted 20 January 2016
exhaust heat was recovered as an energy source for the reforming to generate hydrogen. The
Available online 15 February 2016
experiments were initiated from basic engine firstly to measure the exhaust temperature and emissions; hence, the engine exhaust heat flow rates were calculated under various
Keywords:
operating conditions. After the basic experiments, a reformer was installed in the engine
Methanol
exhaust system of a 1.4-L four-stroke spark-ignition engine. Additionally, the components of
Steam reforming
the reformate gas were measured to identify the correlation between hydrogen flow rate and
Heat recovery
exhaust heat exchange rate under various operating conditions.
Heat exchanger Spark ignition engine
The results show that when the engine was running between 2000 and 3000 rpm and throttle opening set at 20%, the exhaust temperature and heat flow were enough for SRM from the basic engine tests. In the reforming processes, the reformer heat recovery rate increased with engine speed. However, increasing exhaust mass flow rate resulting in poor heat exchange rate at high engine speed. Furthermore, methanol conversion efficiency increased with reformer heat exchange rate. When the S/C ratio was set to 1.2, and the methanol supply rate was fixed at 15.8 g/min, methanol conversion efficiency approached 93% and the hydrogen production was stably close to 75%. The molar rate of hydrogen was approximately 1.34 mol/min. In addition, hydrogen production per unit exhaust heat was 1.6 mol/MJ. Copyright © 2016, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
Introduction Technological advancements have led to a considerable increase in energy consumption. All over the world, fossil fuels
remain the dominant energy resource, yet they have attracted worldwide attention because of issues regarding environmental protection. Many countries have initiated measures to control greenhouse gas (GHG) emissions, among which carbon dioxide (CO2) and methane (CH4) are the most severe.
* Corresponding author. Fax: þ886 6 2050509. E-mail address: [email protected] (R.-F. Horng). http://dx.doi.org/10.1016/j.ijhydene.2016.01.100 0360-3199/Copyright © 2016, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
4958
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 4 9 5 7 e4 9 6 8
Therefore, many countries have actively invested in developing two primary categories of energy: clean energy, which refers to pollution-free energy that generates zero carbon emissions, and alternative energy, which refers to reducing the hydrocarbon content of fuels in order to reduce carbon emissions. Chiriac et al. stated that hydrogen combustion emits no carbon; thus, it would be used as an attractive alternative fuel. When 2% of the energy percentage of diesel is replaced with hydrogen, CO2 emissions can be reduced by approximately 2.5% [1]. In the past decade, continuing growth in the steel industry has been accompanied by high levels of energy consumption. Slag in furnaces after steelmaking can reach temperatures between 1450 C and 1550 C [2], that is, it remains much waste heat. During the production of cement, the primary heat loss results from flue gas emissions, cooling clinkers, and kilns [3]. In addition, there are remarkably huge amounts of pollution and waste heat from the worldwide vehicle exhaust systems. Hossain reported that approximately 38% of the supplied fuel energy in diesel engine is emitted through the exhaust emissions [4,5]. Shudo et al. used an exhaust gas heat recovery strategy in a homogenous charge compression ignition (HCCI) engine for hydrogen production by dimethyl ether and methanol reforming. The results showed that hydrogen addition could avoid oxidation of dimethyl ether (DME) at low temperature and control the ignition timing of the mixture in cylinder. Moreover, the methanol reforming by HCCI waste heat recovery could achieve a high overall system efficiency [6]. Bueno used a heat recovery reformer to reform glycerol into H2-rich gases, hydrogen, and methane. The best reaction temperature for producing hydrogen is 700 C [7]. Thus, alternative vehicle power sources should be developed to reduce the use of fossil fuels and carbon emissions. To achieve this goal, hydrogen energy plays an essential role. Water is the product of hydrogen under exothermic oxidation reaction, which is clean and does not create environmental pollution. The advantages of hydrogen energy include abundance, high heating value, excellent burning performance, extensive applications, environmental friendliness, and high potential for economic benefits [8e10]. Hydrogen can be obtained through various means such as hydrocarbon fuel reforming and water electrolysis. The popular hydrocarbon fuels for hydrogen production can be divided into two categories, which are methane (gas) and alcohol (liquid). Where, methanol possesses many advantages such as easiness to obtain, convenient transportation and storage, high energy density, and high hydrogenecarbon ratio. Currently, common reforming methods for hydrogen production include partial oxidation reforming (POX), steam reforming (SR), oxidation steam reforming (OSR), autothermal reforming (ATR), coal gasification, plasma reformer, and dry reforming (DR) [11e16]. Worldwide, a considerable number of reforming experiments have been conducted for hydrogen production and introducing hydrogen into engines. Horng et al. used a methanol reformer and set multiple parameters to perform a cold-start experiment for fast hydrogen production. With a heating power of 240 W, heating temperature of 100 C, methanol supply rate at 25 cc/min, and an airflow rate of 70 L/min, the cold start was activated in minimal time; specifically, stable hydrogen production required only 3 min
from a cold start. Hydrogen production began at a catalyst outlet temperature of 100 C, becoming stable at approximately 380 C [17]. Horng et al. examined the transient response of mode shifting between POX and ATR by considering heating temperature, methanol supply rate, steadymode shifting temperature, and oxygen-to-carbon (O2/C) and steam-to-carbon (S/C) molar ratios to evaluate the effect of reformer fast start and increased hydrogen production. When POX was used, the hydrogen concentration and hydrogen flow rate were 40% and 20.5 L/min, respectively; when the mode was shifted to ATR, the corresponding values were 49.12% and 23 L/min [18]. Moreover, Kamarudin et al. designed a 5-kW mobile proton exchange membrane fuel cell and tested it using ATR, water gas shift (WGS), and preferential oxidation (PrOX) systems. The experimental results revealed an ATR hydrogen concentration of 73% and CO concentration of 2% when the O2/C and S/C ratios were 0.25 and 1.3, respectively. Subsequently, introducing WGS into the system decreased the CO to be smaller than 2000 ppm; further introducing PrOX reduced the CO to smaller than 100 ppm [19]. Yoon et al. compared copper-based catalysts and noble metal catalysts in SR and ATR systems under different O2/CH3OH ratios. Multiple start-up and shut-down cycles were applied for testing, and when noble metal catalysts were used in the ATR system, the conversion efficiency of supplied methanol approached 100% [20]. Mohammadi et al. reported that hydrogen has low ignition energy requirement, a wide flammability range, and high burning velocity, which make it a crucial fuel for engines. However, pre-mixing hydrogen externally with intake air causes backfire and engine knocking, particularly at high loads. The experimental results showed that direct injection of hydrogen during the late-compression stroke can prevent backfiring and achieve high thermal efficiency and output power. Further optimization of injection timing under high loads can reduce NOx emissions [21]. Akansu et al. introduced various proportions of hydrogen and methane mixtures into engine to evaluate the relevant exhaust emissions and engine performance. The results indicated that introducing hydrogen and methane increased the engine combustion efficiency and reduced exhaust emissions [22]. Horng et al. introduced plasma-reformed H2-rich gases into motorcycle engine to investigate the vehicle performance at a constant speed and transient driving conditions. At a low speed, the energyconservation improvement rate reached 12.2%, whereas the improvement at a high speed was negligible. Regarding exhaust emissions, the CO and HC emissions of the motorcycle engine after introducing H2-rich gases were similar to those generated by the original engine; however, the NOx emissions exhibited a 56.8% improvement rate [23]. Referencing the aforementioned experimental methods, the present study investigated conducting steam reforming of methanol (SRM) reactions to produce H2-rich syngas. However, SR reactions are endothermic reactions that require a supply of external energy. Exhaust heat recovery was utilized as the SRM energy source. Thus, an additional energy supply was unnecessary and hence energy consumption for reforming can be reduced. In addition, Lu utilized diesel engine exhaust heat in examining the hydrogen production characteristics of SRM. Engine exhaust heat was transferred to the reformer via a heat exchanger installed with fins. The
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 4 9 5 7 e4 9 6 8
hydrogen concentration and hydrogen production flow rate were approximately 72.6% and 17.3 L/min, respectively. The estimation results showed that approximately 92.6% of 49.5% recoverable waste heat was convertible. In other words, 36.7% of the waste heat could be recovered, converted, and stored in the produced fuel [24]. Chein et al. employed Cu/ZnO/Al2O3 catalysts of varying thicknesses to explore the SRM conversion and heat transfer effectiveness. The optimal methanol conversion efficiency and a high reforming temperature were attained when the catalyst thickness was 90% of the radius of the reformer [25]. Goicoechea et al. conducted a thermodynamic analysis of acetic acid SR. At a steam-to-acetic acid ratio of 2:1, pressure of 1 bar, and reaction temperature higher than 700 C, the hydrogen and CO production was improved and the CO/CO2 ratio was greater than 1 [26]. The exhaust heat-reforming method employed in the present study can be widely applied in internal combustion engines as well as various other fields such as industry, kiln, furnace and energy sectors. Especially in the last three sectors, which have stable and high-temperature exhaust heat sources that can be recycled as a stable energy supply for reforming reactions. The produced H2-rich syngas by reforming can be reintroduced into the furnace to improve the burning efficiency of coal or heavy oil, thereby achieving the goals of energy conservation and carbon reduction. The objective of this study is to explore a methanol reformer with engine exhaust heat recovery to produce hydrogen with steam reforming of methanol. The SRM with engine exhaust heat recovery have several advantages. Hydrogen production from SRM with engine exhaust heat, the exhaust waste heat energy could be recovered and the overall system efficiency could be improved. In addition, hydrogen could be used as the fuel for engine, and so that the emissions could also be improved generally.
Experimental instruments and methods Instruments In this study, the instruments involved five systems, namely reforming, engine, fuel supply, data acquisition and measurement and analysis, and product gas cooling systems. Fig. 1 shows a diagram of the instruments, and Table 1 shows the specifications of the methanol reformer used in this study. The main body of the reforming system comprised a reformer and external heat exchanger, and an exhaust diverter and control valve were installed at the front of the reformer to control the high temperature exhaust stream entering the reformer. The exhaust heat recovery control strategy for regulating the reforming temperature is also displayed (Fig. 2). The reforming operating temperature was controlled between 240 C and 300 C for better reforming performance and preventing the damage of reformer catalyst. Thus, the exhaust flow rates introduced into the reformer were determined by the target temperature of the reformer; and the excessive exhaust gases leave engine cylinder were conducted into another path. The excess exhaust heat was directly emitted into the atmosphere. The engine test system primarily comprised an engine dynamometer and gasoline engine.
4959
Table 2 shows the engine specifications. The fuel supply system comprises reformer and engine components. The reformer component is the methanol-water solution supply pump (YANGTECH, EMP-3706ACH-U), which was installed with an electronic mass spectrometer for monitoring the mass flow rate of the methanol-water solution. The engine component contained a fuel mass flow meter. The extraction measurement and analysis system comprised a temperature measurement system, exhaust emission analyzer (HORIBA, MEXA-554JA), and gas chromatograph (Agilent 6850 GC). Temperature was measured using K type thermocouples, which were installed at the engine exhaust stream, heat exchanger inlet, and reformer main body to facilitate calculating the thermal flow rate and amount of heat exchange as well as for monitoring and controlling the reformer temperature.
Methods and procedure The SRM reaction conducted in the current study was an endothermic reaction that required an additional energy supply to react. Thus, waste heat recovered from engine exhaust was employed as the SRM energy source. First, the basic engine was tested initially to determine the appropriate engine operation conditions for the reforming applications. The engine used in the experiment was a 1328-cc four-stroke spark-ignition engine. The engine was tested at a rotation speed of 2000e4000 rpm with the throttle opening range set between 20% and 50%. The reformer operating temperature was controlled between 240 C and 300 C. The exhaust heat flow rate varied with the engine operation conditions. Thus, the exhaust diverter and exhaust control valve opening of diverting tube were regulated to control the internal temperature of the reformer, and to prevent the catalyst from sintering at high temperatures. Regarding the experimental parameters, the S/C ratios were set at 1.2 and 1.4 respectively and the methanol supply flow rate was set at 15.8 g/min to investigate the hydrogen production of the reformer with exhaust heat recovery.
Calculations Molar flow rate of engine exhaust emissions This section describes the calculations for the molar flow rate of the exhaust gases according to the measured exhaust emission concentrations through chemical equilibrium equations under various engine operation conditions. Therefore, the exhaust heat flow rate was obtained by the measured exhaust temperature and calculated exhaust molar flow rate. After the exhaust heat flow rate was determined under various operation conditions, the reforming effect of heat recovery would be evaluated. The actual molar flow rates of all exhaust gases were determined by Eq. (1). xC8 H15 þ zðO2 þ 3:76N2 Þ/aCO þ bCO2 þ dHC þ eO2 þ fN2 þ gH2 O (1) where x, z, and g are equilibrium coefficients to be determined; a, b, d, e, and f are the exhaust concentrations measured using an emission analyzer by dry base analysis.
4960
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 4 9 5 7 e4 9 6 8
Fig. 1 e Schematic of the experimental setup.
Table 1 e The reformer specifications. Reformed fuel Catalyst Operation temperature (oC) Catalyst mass (g) Steam to carbon ratio Max. flow rate of H2 production (m3/h)
Methanol-water solution CuOeZnO/Al2O3 220e320 700 1.2:1, 1.4:1 2
Determination of exhaust flow rate entering the reformer During the experiments, an adequate additional flow rate of N2 was introduced into the diverting inlet of the reformer for calculating the flow rate of exhaust gas into the reformer. Equation (2) was used to calculate the exhaust gas molar flow rate entering the reformer. Subsequently, the waste heat flow rate was calculated. ðaCO þ bCO2 þ dHC þ eO2 þ fN2 Þ þ f 0 N2 þ gH2 O/ a1 CO þ b1 CO2 þ d1 HC þ e1 O2 þ f1 N2 þ g1 H2 O (2) The left parenthesis in Eq. (2) is the exhaust gas components introduced into the reformer pipeline, and f'N2 is the additional N2 introduced. The concentrations of the exhaust gas components are known, and aef represent the measured concentrations (Vol%). The right parenthesis in Eq. (2) refers to the components and concentrations of exhaust gases, which were analyzed using the emission analyzer after introducing additional N2 to the exhaust stream at the reformer outlet;
a1ef1 represent the measured concentrations (Vol%). The molar flow rates of CO, CO2, HC, O2, and H2O at the reformer inlet and outlet are identical (assuming that no more chemical reactions occur during the heat exchange process while the exhaust gases pass through the reformer) and the molar flow rate of the introduced N2 is known. However, the molar rates of gases entering the reformer are unknown. The total molar flow rate of the exhaust gas is obtained as substituting Eqs. (3)e(5) into Eq. (6); and the flow rates of N2 at the reformer outlet and other exhaust components can be determined by substituting the Eqs. (3)e(5) into Eqs. (7) and (8), respectively. n_N f ¼P 2 n_ex
(3)
n_1N2 n_ex þ n_addN2
(4)
n_1N2 ¼ n_N2 þ n_addN2
(5)
f1 ¼ P
X
ð1 f Þ n_ex ¼ n_addN2 f1 f
ð1 f Þ n_addN2 n_1N2 ¼ f1 f1 f a1 n_i ¼ n_1N2 f1 i ¼ CO, CO2, H2O, N2, O2, HC (measured exhaust species)
(6)
(7) (8)
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 4 9 5 7 e4 9 6 8
4961
Fig. 2 e Schematic of the exhaust heat recovery control strategy.
Determination of exhaust heat flow rate
Definition of calculation parameters
To calculate the exhaust heat flow rate, the molar flow rates and enthalpy values of all exhaust products must be determined first. The temperatures of the products differed according to the engine operation conditions. Thus, Eq. (9) was used to calculate the enthalpy values of the products at various temperatures. Equation (10) was used to calculate the exhaust heat flow rate of the exhaust gases. Table 4 shows the calculation results.
The theoretical steam reforming of methanol reaction is shown in Eq. (11a), and the actual reaction is shown in Eq. (11b). The aforementioned calculations can be used to derive a series of results. Such as, hydrogen production includes hydrogen mass percent (HMP), which represents the ratio of hydrogen mass to gasoline and hydrogen masses fed into engine, as shown in Eq. (12); and hydrogen energy percent (HEP), which represents the ratio of hydrogen heating value to gasoline and hydrogen heating values fed into engine, as shown in Eq. (13). Engine exhaust heat recovery efficiency was calculated according to Eq. (14), which is the ratio of engine exhaust heat flow rate fed into reformer to the total engine exhaust heat flow rate. With the hydrogen production and methanol consumed in reforming, molecules of hydrogen production and reformed methanol per unit exhaust heat flow rate fed into reformer were calculated by Eqs. (15) and (16), respectively. Thermal efficiency of exhaust heat was calculated by Eq. (17), which is the produced hydrogen and carbon monoxide heating values in reforming per unit exhaust heat flow rate fed into reformer. Additionally, the thermal efficiency of reforming was determined by Eq. (18). Finally, the reformer heat recycled efficiency was calculated by Eq. (19), which describes the percentage of exhaust heat flow rate entering the reformer in the process of exhaust heat exchange. Further, the reformer heat exchange efficiency was determined by Eq. (20), which represents the enthalpy change rate in the reforming to the exhaust heat flow rate fed into
ZT ðDhi ÞT ¼
Cp i dT
(9)
298:15
where, i ¼ CO, CO2, H2O, N2, O2, HC (measured exhaust species), T is the exhaust temperature. Q_ ¼ n_CO ðDhCO ÞT þ n_H2 O DhH2 O T þ n_CO2 DhCO2 T þ n_O2 DhO2 T þ n_HC ðDhHC ÞT þ n_N2 DhN2 T
(10)
Table 2 e The engine specifications. Engine manufacturer Displacement volume (c.c.) Valve train Bore Stroke (mm mm) Compression ratio Compression pressure (kg/cm2/rpm)
Honda 1328 DOHC 75.0 75.2 9.5:1 12/250
4962
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 4 9 5 7 e4 9 6 8
ðn_MeOH Þconsump ðn_MeOH ÞUEH ¼ Q_ in
Table 3 e Measurement uncertainties. o
Temperature: ±2.6 C Fuel flow rate: ±2.22% CO2: ±0.51% CO: ±0.51% HC: ±0.50% H2: ±1.01%
ðn_ HHVÞH2 þ ðn_ HHVÞCO prod ðhth Þex ¼ Q_ in
ðhth ÞRe
CH3 OHðlÞ þ H2 OðlÞ /CO2ðgÞ þ 3H2 OðgÞ
(11a)
xCH3 OHðlÞ þ yH2 OðlÞ /aCOðgÞ þ bCO2ðgÞ þ dH2ðgÞ þ eH2 OðgÞ
m_ H2 m_ gasoline þ m_ H2
(12)
hrecycled ¼
hheatrecycled
reformer
(18)
Q_ in Q_ out reformer ¼ Q_ in
(19)
reformer
P
hheatexchange
reformer
¼
ðn_ DhÞprod Q_ in
P
ðn_ DhÞreact
Re
(20)
reformer
reformer
P ¼
P ðn_ DhÞprod ðn_ DhÞreact Re Q_ in Q_ out
(21)
reformer
(13)
Measurement error analysis
reformer
(14)
Q_ ex
hheatusage
ðn_ HHVÞH2 ðn_ HHVÞgasoline þ ðn_ HHVÞH2 Q_ in
ðn_ HHVÞH2 þ ðn_ HHVÞCO prod ¼ ðn_ HHVÞMeOH react
(11b)
where, x, y, and e are the unknown coefficients to be determined, and a, b and d are the known fractions analyzed by the gas chromatography (GC).
HEP ¼
(17)
reformer
reformer; and the reformer heat usage efficiency refers to the relationship between the enthalpy change in the reforming and the exhaust heat exchange ðQ_ in Q_ out Þrefornmer in the process, as shown in Eq. (21).
HMP ¼
(16)
reformer
The total measurement error (dk) consists of bias error (b) and precision error (εk) such as that [27], dk þ b ¼ ε k
n_H2 n_H2 UEH ¼ _ Q in
prod
(15)
reformer
The precision error is repeated N times to take measurements to determine. The each series measurement which the precision index is usually expressed as follows: S Sx ¼ pffiffiffiffi N
Table 4 e The exhaust emissions under various engine operating conditions. Engine speed (rpm)
THa (%) CO CO2
2000 2500 3000 3500 4000 2000 2500 3000 3500 4000 2000 2500 3000 3500 4000 2000 2500 3000 3500 4000 a
TH: throttle opening.
HC
O2
N2 H2O(g)
Vol% 20
30
40
50
0.53 0.84 1.27 0.80 1.27 4.27 0.51 0.61 0.52 3.20 4.75 3.03 3.11 0.71 3.75 4.79 2.28 2.85 2.90 4.30
12.58 12.52 12.66 7.86 12.66 11.76 11.82 8.74 8.94 7.92 11.16 9.08 7.50 10.42 9.06 11.04 9.88 6.85 9.10 8.70
0.0084 0.0052 0.0062 0.0006 0.0008 0.0084 0.0164 0.0020 0.0006 0.0004 0.0170 0.0034 0.0056 0.0014 0.0058 0.0086 0.0014 0.0036 0.0044 0.0064
1.38 1.00 3.98 8.40 6.96 0.02 2.46 6.94 6.76 5.48 0.00 4.38 6.62 4.84 3.82 0.04 3.56 4.46 4.90 3.40
85.50 85.63 82.09 82.94 79.11 83.93 85.21 83.71 83.78 83.39 84.07 83.51 82.76 84.03 83.36 84.12 84.28 85.84 83.10 83.59
12.28 12.52 13.06 8.12 13.06 15.01 11.56 8.76 8.87 10.42 14.89 11.35 9.94 10.43 12.00 14.83 11.40 9.09 11.24 12.18
where, S is the standard deviation in N repeated measurements. The bias error considered as the systematic error remains constant in testing process. The measurement uncertainty be 95% confidence, according to the following: i12 h URSS ¼ B2 þ ðtSx Þ2 where U is the measurement uncertainty, and was calculated by the root sum square (RSS) method, B is the bias limit, and t is set equal to 2 for large samples (N > 30). According to a series analysis, the measurement uncertainties were estimated and as revealed in Table 3.
Results and discussion Basic engine tests The exhaust temperature and exhaust emission concentrations varied with the operation conditions (i.e., throttle opening and engine speed). Thus, the exhaust heat flow rate varied with the exhaust components, concentrations and total exhaust gas flow rate. To evaluate the exhaust heat for reforming and producing H2-rich gases, the temperature was
4963
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 4 9 5 7 e4 9 6 8
measured at various locations along the exhaust pipe under various engine operation conditions. A basic engine test was conducted with the engine running at various rotation speeds between 2000 and 4000 rpm with the throttle opening ranging between 20% and 50%. The parameters measured were the engine fuel consumption rate, exhaust temperature, and exhaust emission concentrations. Table 4 shows the exhaust components and concentrations. All component concentrations combined with N2 flow rate were substituted into Eq. (1) to calculate the exhaust molar flow rate, and Eqs. (9) and (10) were used to calculate the exhaust heat flow rates under various operation conditions. As shown in Fig. 3, the exhaust temperature increased with the engine speed, ranging from 524 C to 736 C. However, because the operational ceiling of the reformer temperature was 320 C, such exhaust temperatures are sufficient for SRM for engine running at 2000e3000 rpm with the throttle opening set at 20%. Table 5 shows the basic engine test results obtained under these selected conditions. The experimental measurements indicated that the exhaust temperature was between 529 C and 592 C and the exhaust heat accounted for 28%e31% of the input gasoline heating value. In this study, the basic engine test was conducted with the throttle opening set at 20%, 30%, 40%, and 50% with the engine speeds between 2000 and 4000 rpm. In experimental processes, the constant speed mode of dynamometer controller was used. Under each engine speed, the engine output (torque or power) was determined by the throttle opening. However, in the reforming experiments, the engine speed was set between 2000 and 3000 rpm and the throttle opening was fixed at 20%, because the exhaust temperatures and waste heat are sufficient for methanol reforming reactions according to the results of basic engine test.
ratio was 1.2 and 1.4, the hydrogen produced concentration is stable, approaching 75%. The hydrogen concentration does not vary with the reforming temperature or engine speed, whereas the molar flow rate of the hydrogen produced increases with the reforming temperature but decreases as the engine speed increases. This is because when the reforming temperature increases, the exhaust gas energy for reforming increases, thereby increasing the likelihood of methanol being converted into hydrogen. However, the total gas flow rate from the exhaust pipe increases with engine speed. Therefore, the opening of the exhaust control valve mounted in the reformer pipe must be reduced to regulate the temperature of reformer. Thus the velocity of the exhaust gas into the reformer increases and the exhaust stream cannot contact the wall of heat exchanger uniformly, which weakens the heat exchange effect and lowering the methanol conversion efficiency and thereby reducing the molar flow rate of hydrogen when the engine speed increases.
Hydrogen production characteristics Hydrogen produced by reforming could be widely applied in fuel cells, engines, industrial sectors, boilers, power mechanics, and various other devices and fields. The produced H2-rich syngas in this study demonstrates potential for future use in engines. Equations (11) and (12) were used to calculate the HMP (Fig. 5) and HEP (Fig. 6) for the investigated engine. Both figures depict similar tendencies. HMP and HEP increase with the reforming temperature because the hydrogen production increases with the reforming temperature. HMP and HEP decrease with increasing engine speed because hydrogen production cannot be improved and the engine consumes more fuel as engine speed increases. The HMP ranges between 1.37% and 3.63% and HEP ranges between 4.4% and 11.08%.
Reforming test SR is an endothermic reaction, and engine exhaust pipe emits a substantial amount of heat flow that can be utilized as energy for reforming reactions to generate H2-rich syngas. Thus, a steam reforming of methanol for hydrogen production test is conducted with engine speed set at 2000e3000 rpm and the throttle opening set at 20%. Fig. 4(a) shows that when the S/C
Exhaust heat recovering and reforming Equations (2)e(10) and (14) and (15) are employed to identify the relationship between the exhaust heat recovery and hydrogen production. Fig. 7 depicts the hydrogen molar flow rate per unit exhaust heat is 0.62e1.6 mol/MJ. When the reforming temperature increases, the molar flow
Table 5 e Exhaust waste heat and engine input heating value under various engine speed. (TH: 20%). Engine speed (rpm) Tex Q_ ex Q_ gasoline o
2000
2500
3000
Fig. 3 e Exhaust temperature and exhaust heat flow rate under various engine speeds and throttle openings.
C
kW
kW
529 534 537 537 560 564 567 570 583 586 589 592
12.93 13.10 12.97 12.97 15.89 15.66 16.06 15.96 18.81 18.98 19.17 19.43
45.75 46.11 45.39 45.39 53.99 52.60 53.67 52.99 60.08 60.44 60.08 61.16
4964
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 4 9 5 7 e4 9 6 8
Fig. 5 e Effect of reforming temperature on hydrogen mass percentage.
Fig. 4 e The hydrogen concentration and molar flow rate under various engine speeds and reforming temperatures.
rate per unit exhaust heat increases. The hydrogen production is close to a stable value when the temperature reaches around 260e280 C. However, the increase in engine speed and S/C reduces the molar flow rate of the hydrogen production. This is because high S/C ratios represent an increase of water in the reformed fuel. Liquid water absorbs latent heat for vaporization and lowering the temperature, which inhibits reforming reactions. Thus, the produced hydrogen molar flow rate declines even under the same engine speed and heat input. Fig. 8 shows the molar flow rate of methanol that can be reformed from each unit of exhaust heat. The tendency of molar flow rate of reformed methanol in Fig. 8 resembles that in Fig. 7. The term is 0.2e0.57 mol/MJ. Fig. 9 depicts the effect of the engine speed on the thermal efficiency of exhaust heat. As shown in Eq. (16), the thermal efficiency of the exhaust heat is the ratio of heating value of the H2-sysgas generated to the exhaust heat entering the reformer. The thermal efficiency of exhaust heat decreases with increasing engine speed, but increases with the reforming temperature. The thermal efficiency of exhaust heat is between 17.7% and 47.5%.
Fig. 6 e Effect of reforming temperature on hydrogen energy percentage.
Methanol conversion efficiency The methanol conversion efficiency affects the production of hydrogen. As shown in Fig. 10, the methanol conversion efficiency increases with the reforming temperature, and the hydrogen production flow rate increases with the methanol conversion efficiency. However, the methanol conversion efficiency decreases when the S/C ratio and engine speed increase. This phenomenon is caused by an increase of latent heat due to the amount of water addition and decrease in the heat exchange rate of reformer. The hydrogen production molar flow rate is approximately 0.58e1.34 mol/min. Fig. 11 demonstrates the thermal efficiency of reforming. It reveals a tendency similar to that in Fig. 10. The thermal efficiency of reforming increases with the reformer heat usage efficiency and reforming temperature, but decreases with the engine speed and S/C ratio. The thermal efficiency of reforming is 46.5%e104.2%, where the value over 100% should be due to the gain of hydrogen from water by the exhaust heat recovery.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 4 9 5 7 e4 9 6 8
Fig. 7 e Hydrogen production per unit waste heat under various operating conditions.
4965
Fig. 9 e The relationship of thermal efficiency of exhaust heat and engine speed.
Figs. 12e15 depict the analysis of the exhaust heat recycled, heat exchange rate, and thermal efficiency of reforming, from which the effect of the heat exchange on reforming can be identified. Fig. 12 shows the relationship between the reformer heat exchange rate and exhaust heat recycled. The reformer heat exchange rate and exhaust heat recycled decrease with lifting engine speed but increase with rising reforming temperature. Fig. 13 shows the relationship between reformer heat recycled and heat exchange rate, which are derived from Eqs. (18) and (19). In the figure, the reformer heat exchange decreases as the engine speed and S/C ratio increase, whereas the reformer heat recycled increases with engine speed. In the exhaust heat recovery process for reforming, it is difficult to insulate the reformer system completely to prevent the heat loss, hence the reformer heat
exchange rate decreases as the exhaust mass flow rate increases. Fig. 14 depicts the relationship between the thermal efficiency of reforming and heat recovery rate. The thermal efficiency of reforming decreases with the increase in the reformer heat recycled rate because of lower heat use rate and heat dissipation. Further in Fig. 15, it demonstrates that the methanol conversion efficiency increases with the reformer heat usage rate. That is, when the available thermal energy for reforming increases, the methanol conversion efficiency is improved. Fig. 16(a) and (b) illustrate the engine exhaust mass flow rate and methanol conversion efficiency with reformer heat usage efficiency under S/C ratio of 1.2 and 1.4 respectively. The figures show that the engine exhaust mass flow rate is inversely developed with the reformer heat usage efficiency; whereas the methanol conversion efficiency increases with reformer heat usage efficiency. High reformer heat usage rate representing a high reaction enthalpy is generated in the reforming process, thereby high methanol
Fig. 8 e Reformed methanol per unit waste heat under various operating conditions.
Fig. 10 e The relationship of hydrogen molar flow rate and methanol conversion efficiency under various operating conditions.
The characteristics of exhaust heat recovery and heat exchange
4966
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 4 9 5 7 e4 9 6 8
Fig. 11 e The relationship of thermal efficiency of reforming and methanol conversion efficiency under various operating conditions.
Fig. 14 e The effect of reformer heat recycled on thermal efficiency of reforming.
conversion efficiency is achieved. The engine exhaust mass flow rate increases at high engine speed. However, it lowers the heat exchange in the reforming reactions. In Fig. 16(a) and (b), the reformer heat usage efficiency at the identical engine speed differs clearly because of the different S/C ratios.
Conclusions
Fig. 12 e The relationship of exhaust heat recycled with reformer heat exchange rate.
Fig. 13 e The relationship of reformer heat recycled with reformer heat exchange rate.
The methanol steam reforming system with exhaust heat recycling system was investigated in this study. The basic engine was tested initially ahead of a series of experiments. The results showed that, under various engine operation conditions, the exhaust temperature and emissions differ according to the throttle opening and engine speed. Thus, the exhaust heat flow rate varies with the exhaust components and concentrations. When the engine speed was set between 2000 and 3000 rpm and throttle opening at 20%, the measured exhaust temperature between 529 C and 592 C was enough for methanol reforming. Under the operation conditions, the
Fig. 15 e Reformer heat usage rate on methanol conversion efficiency under various operating conditions.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 4 9 5 7 e4 9 6 8
exhaust heat accounted for 28%e31% of the input gasoline heating value. In the reforming processes with exhaust recycling, when S/ C was set at 1.2 and 1.4, the produced hydrogen concentration was stable, approaching 75%. The hydrogen molar flow rate increased with the reforming temperature but decreased with the engine speed. The produced hydrogen molar flow rate was approximately 0.58e1.34 mol/min. When the parameters for the engine application were evaluated, the results show that the HMP was 1.37%e3.63%; HEP was 4.4%e11.08%. Further considering the exhaust heat effectiveness on reforming, the hydrogen production per unit exhaust heat was 0.62e1.6 mol/MJ, and the reformed methanol per unit exhaust heat was 0.2e0.57 mol/MJ. In addition, the thermal efficiency of exhaust heat was 17.7%e 47.5%, and the thermal efficiency of reforming was 46.5%e 104.2%. Overall, reformer heat recovery rate increased with engine speed, whereas exhaust heat recovery decreased as engine speed increased. At high engine speed, exhaust mass flow rate increased, resulting in poor heat exchange effect. Finally, methanol conversion efficiency increased with reformer heat exchange rate.
Fig. 16 e Engine exhaust mass flow rate and methanol conversion efficiency with reformer heat usage rate.
4967
Acknowledgments This study was supported and subsidized by the Ministry of Science and Technology (MOST 103-2221-E-168-022) and the NSC-EPA Air Pollution Prevention Project (NSC102-EPA-F-007002) of Taiwan. The authors hereby express their appreciation to MOST, NSC, and EPA of Taiwan.
references
[1] Chiriac R, Racovitza A, Podevin P, Descombes G. On the possibility to reduce CO2 emissions of heat engines fuelled partially with hydrogen produced by waste heat recovery. Int J Hydrogen Energy 2015;40:15856e63. [2] Zhang H, Wang H, Zhu X, Qiu YJ, Li K, Chen R, et al. A review of waste heat recovery technologies towards molten slag in steel industry. Appl Energy 2013;112:956e66. [3] Karellas S, Leontaritis AD, Panousis G, Bellos E, Kakaras E. Energetic and exergetic analysis of waste heat recovery systems in the cement industry. Energy 2013;58:147e56. [4] Hossain SN, Bari S. Waste heat recovery from the exhaust of a diesel generator using Rankine Cycle. Energy Convers Manag 2013;75:141e51. [5] Bari S, Hossain SN. Waste heat recovery from a diesel engine using shell and tube heat exchanger. Appl Therm Eng 2013;61:355e63. [6] Shudo T, Shima Y, Fuji T. Production of dimethyl ether and hydrogen by methanol reforming for an HCCI engine system with waste heat recovery e continuous control of fuel ignitability and utilization of exhaust gas heat. Int J Hydrogen Energy 2009;34:7638e47. [7] Bueno AV, de Oliveira MLM. Glycerol steam reforming in a bench scale continuous flow heat recovery reactor. Int J Hydrogen Energy 2013;38:13991e4001. [8] Das LM, Gulati R. A comparative evaluation of the performance characteristics of a spark ignition engine using hydrogen and compressed natural gas as alternative fuels. Int J Hydrogen Energy 2000;25:783e93. [9] Li H, Karim GA. A prediction study of a spark ignition supercharged hydrogen engine. Int J Hydrogen Energy 2005;30:1491e9. [10] Al-Baghdadi MARS, Al-Janabi HAS. Exhaust emissions from an SI engine operating on gaseous fuel mixtures containing hydrogen. Int J Hydrogen Energy 2003;44:3143e50. [11] Park B, Kwon S. Compact design of oxidative steam reforming of methanol assisted by blending hydrogen peroxide. Int J Hydrogen Energy 2015;40:12697e704. [12] Toyir J, Piscina PR, Homs N. Ga-promoted copper-based catalysts highly selective for methanol steam reforming to hydrogen; relation with the hydrogenation of CO2 to methanol. Int J Hydrogen Energy 2015;40:11261e6. [13] Chang CC, Hsu CC, Chang CT, Chen YP, Liaw BJ, Chen YZ. Effect of noble metal on oxidative steam reforming of methanol over CuO/ZnO/Al2O3 catalysts. Int J Hydrogen Energy 2012;37:11176e84. [14] Paasikallio V, Azhari A, Kihlman J, Simell P, Lehtonen J. Oxidative steam reforming of pyrolysis oil aqueous fraction with zirconia pre-conversion catalyst. Int J Hydrogen Energy 2015;40:12088e96. [15] Lee KY, Huang YJ. Low CO generation on tunable oxygen vacancies of non-precious metallic Cu/ZnO catalysts for partial oxidation of methanol reaction. Appl Catal B Environ 2014;150e151:506e14.
4968
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 4 9 5 7 e4 9 6 8
[16] Corigliano O, Fragiacomo P. Technical analysis of hydrogenrich stream generation through CO2 reforming of biogas by using numerical modeling. Fuel 2015;158:538e48. [17] Horng RF, Chen CR, Wu TS, Chan CH. Cold start response of a small methanol reformer by partial oxidation reforming of hydrogen for fuel cell. Appl Therm Eng 2006;26:1115e24. [18] Horng RF, Chou HM, Lee CH, Tsai HT. Characteristics of hydrogen produced by partial oxidation and auto-thermal reforming in a small methanol reformer. J Power Sources 2006;163:119e24. [19] Kamarudin SK, Daud WRW, Som AM, Takriff MS, Mohammad AW, Loke YK. Design of a fuel processor unit for PEM fuel cell via shortcut design method. Chem Eng J 2004;104:7e17. [20] Yoon HC, Erickson PA, Kim HM. Lowering the O2/CH3OH ratio in autothermal reforming of methanol by using a reduced copper-based catalyst. Int J Hydrogen Energy 2008;33:6619e26. [21] Mohammadi A, Shioji M, Nakaib Y, Ishikura W, Tabo E. Performance and combustion characteristics of a direct injection SI hydrogen engine. Int J Hydrogen Energy 2007;32:296e304. [22] Akansu SO, Kahraman N, Ceper B. Experimental study on a spark ignition engine fuelled by methaneehydrogen mixtures. Int J Hydrogen Energy 2007;32:4279e84. [23] Horng RF, Wen CS, Liauh CT, Chao Y, Huang CT. Driving characteristics of a motorcycle fuelled with hydrogen-rich gas produced by an onboard plasma reformer. Int J Hydrogen Energy 2008;33:7619e29. [24] Lu LC, Lu JH. Energy analysis for hydrogen generation with the waste heat of internal combustion engine. Adv Mater Res 2012;512e515:1492e8. [25] Chein RY, Chen LC, Chen YC, Chung JN. Heat transfer effects on the methanol-steam reforming with partially filled catalyst layers. Int J Hydrogen Energy 2009;34:5398e408.
[26] Goicoechea S, Ehrich H, Arias PL, Kockmann N. Thermodynamic analysis of acetic acid steam reforming for hydrogen production. J Power Sources 2015;279:312e22. [27] Abernethy RB, Benedict RP, Dowdell RB. ASME measurement uncertainty. Trans ASME J Fluids Eng 1985;107:161e4.
Nomenclature cp: specific heat capacity, (kJ kg1 K1) HEP: hydrogen energy percentage HHV: higher heating value, (kJ kg1) HMP: hydrogen mass percentage Dh: change of enthalpy, (kJ) _ mass flow rate, (kg sec1) m: _ molar flow rate, (mole min1) n: _ heat flow rate, (kW) Q: Greek letters h: efficiency, (%) hheatexchange: heat exchange efficiency, (%) hheatrecyced: heat recycled efficiency, (%) hheatusage: heat usage efficiency, (%) hth: thermal efficiency, (%) Subscripts addN2: added nitrogen consump: consumption ex: engine exhaust prod: product Re: reforming React: reactant Recycled: engine exhaust heat recovery UEH: unit exhaust heat
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/he
Investigation on the hydrogen production by methanol steam reforming with engine exhaust heat recovery strategy Cheng-Hsun Liao, Rong-Fang Horng* Department of Mechanical Engineering, Intelligent Vehicle Research Center, Kun Shan University, No.195, Kunda Rd., Yungkang Dist., Tainan City 710, Taiwan, ROC
article info
abstract
Article history:
In this study, a reformer with an engine exhaust heat exchanger was employed to produce
Received 26 November 2015
hydrogen. The purpose of this study is to investigate a methanol reformer with engine
Received in revised form
exhaust heat recovery to produce hydrogen with steam reforming of methanol (SRM). Steam
12 January 2016
reforming is an endothermic reaction requiring additional energy for reaction. Thus, engine
Accepted 20 January 2016
exhaust heat was recovered as an energy source for the reforming to generate hydrogen. The
Available online 15 February 2016
experiments were initiated from basic engine firstly to measure the exhaust temperature and emissions; hence, the engine exhaust heat flow rates were calculated under various
Keywords:
operating conditions. After the basic experiments, a reformer was installed in the engine
Methanol
exhaust system of a 1.4-L four-stroke spark-ignition engine. Additionally, the components of
Steam reforming
the reformate gas were measured to identify the correlation between hydrogen flow rate and
Heat recovery
exhaust heat exchange rate under various operating conditions.
Heat exchanger Spark ignition engine
The results show that when the engine was running between 2000 and 3000 rpm and throttle opening set at 20%, the exhaust temperature and heat flow were enough for SRM from the basic engine tests. In the reforming processes, the reformer heat recovery rate increased with engine speed. However, increasing exhaust mass flow rate resulting in poor heat exchange rate at high engine speed. Furthermore, methanol conversion efficiency increased with reformer heat exchange rate. When the S/C ratio was set to 1.2, and the methanol supply rate was fixed at 15.8 g/min, methanol conversion efficiency approached 93% and the hydrogen production was stably close to 75%. The molar rate of hydrogen was approximately 1.34 mol/min. In addition, hydrogen production per unit exhaust heat was 1.6 mol/MJ. Copyright © 2016, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
Introduction Technological advancements have led to a considerable increase in energy consumption. All over the world, fossil fuels
remain the dominant energy resource, yet they have attracted worldwide attention because of issues regarding environmental protection. Many countries have initiated measures to control greenhouse gas (GHG) emissions, among which carbon dioxide (CO2) and methane (CH4) are the most severe.
* Corresponding author. Fax: þ886 6 2050509. E-mail address: [email protected] (R.-F. Horng). http://dx.doi.org/10.1016/j.ijhydene.2016.01.100 0360-3199/Copyright © 2016, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
4958
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 4 9 5 7 e4 9 6 8
Therefore, many countries have actively invested in developing two primary categories of energy: clean energy, which refers to pollution-free energy that generates zero carbon emissions, and alternative energy, which refers to reducing the hydrocarbon content of fuels in order to reduce carbon emissions. Chiriac et al. stated that hydrogen combustion emits no carbon; thus, it would be used as an attractive alternative fuel. When 2% of the energy percentage of diesel is replaced with hydrogen, CO2 emissions can be reduced by approximately 2.5% [1]. In the past decade, continuing growth in the steel industry has been accompanied by high levels of energy consumption. Slag in furnaces after steelmaking can reach temperatures between 1450 C and 1550 C [2], that is, it remains much waste heat. During the production of cement, the primary heat loss results from flue gas emissions, cooling clinkers, and kilns [3]. In addition, there are remarkably huge amounts of pollution and waste heat from the worldwide vehicle exhaust systems. Hossain reported that approximately 38% of the supplied fuel energy in diesel engine is emitted through the exhaust emissions [4,5]. Shudo et al. used an exhaust gas heat recovery strategy in a homogenous charge compression ignition (HCCI) engine for hydrogen production by dimethyl ether and methanol reforming. The results showed that hydrogen addition could avoid oxidation of dimethyl ether (DME) at low temperature and control the ignition timing of the mixture in cylinder. Moreover, the methanol reforming by HCCI waste heat recovery could achieve a high overall system efficiency [6]. Bueno used a heat recovery reformer to reform glycerol into H2-rich gases, hydrogen, and methane. The best reaction temperature for producing hydrogen is 700 C [7]. Thus, alternative vehicle power sources should be developed to reduce the use of fossil fuels and carbon emissions. To achieve this goal, hydrogen energy plays an essential role. Water is the product of hydrogen under exothermic oxidation reaction, which is clean and does not create environmental pollution. The advantages of hydrogen energy include abundance, high heating value, excellent burning performance, extensive applications, environmental friendliness, and high potential for economic benefits [8e10]. Hydrogen can be obtained through various means such as hydrocarbon fuel reforming and water electrolysis. The popular hydrocarbon fuels for hydrogen production can be divided into two categories, which are methane (gas) and alcohol (liquid). Where, methanol possesses many advantages such as easiness to obtain, convenient transportation and storage, high energy density, and high hydrogenecarbon ratio. Currently, common reforming methods for hydrogen production include partial oxidation reforming (POX), steam reforming (SR), oxidation steam reforming (OSR), autothermal reforming (ATR), coal gasification, plasma reformer, and dry reforming (DR) [11e16]. Worldwide, a considerable number of reforming experiments have been conducted for hydrogen production and introducing hydrogen into engines. Horng et al. used a methanol reformer and set multiple parameters to perform a cold-start experiment for fast hydrogen production. With a heating power of 240 W, heating temperature of 100 C, methanol supply rate at 25 cc/min, and an airflow rate of 70 L/min, the cold start was activated in minimal time; specifically, stable hydrogen production required only 3 min
from a cold start. Hydrogen production began at a catalyst outlet temperature of 100 C, becoming stable at approximately 380 C [17]. Horng et al. examined the transient response of mode shifting between POX and ATR by considering heating temperature, methanol supply rate, steadymode shifting temperature, and oxygen-to-carbon (O2/C) and steam-to-carbon (S/C) molar ratios to evaluate the effect of reformer fast start and increased hydrogen production. When POX was used, the hydrogen concentration and hydrogen flow rate were 40% and 20.5 L/min, respectively; when the mode was shifted to ATR, the corresponding values were 49.12% and 23 L/min [18]. Moreover, Kamarudin et al. designed a 5-kW mobile proton exchange membrane fuel cell and tested it using ATR, water gas shift (WGS), and preferential oxidation (PrOX) systems. The experimental results revealed an ATR hydrogen concentration of 73% and CO concentration of 2% when the O2/C and S/C ratios were 0.25 and 1.3, respectively. Subsequently, introducing WGS into the system decreased the CO to be smaller than 2000 ppm; further introducing PrOX reduced the CO to smaller than 100 ppm [19]. Yoon et al. compared copper-based catalysts and noble metal catalysts in SR and ATR systems under different O2/CH3OH ratios. Multiple start-up and shut-down cycles were applied for testing, and when noble metal catalysts were used in the ATR system, the conversion efficiency of supplied methanol approached 100% [20]. Mohammadi et al. reported that hydrogen has low ignition energy requirement, a wide flammability range, and high burning velocity, which make it a crucial fuel for engines. However, pre-mixing hydrogen externally with intake air causes backfire and engine knocking, particularly at high loads. The experimental results showed that direct injection of hydrogen during the late-compression stroke can prevent backfiring and achieve high thermal efficiency and output power. Further optimization of injection timing under high loads can reduce NOx emissions [21]. Akansu et al. introduced various proportions of hydrogen and methane mixtures into engine to evaluate the relevant exhaust emissions and engine performance. The results indicated that introducing hydrogen and methane increased the engine combustion efficiency and reduced exhaust emissions [22]. Horng et al. introduced plasma-reformed H2-rich gases into motorcycle engine to investigate the vehicle performance at a constant speed and transient driving conditions. At a low speed, the energyconservation improvement rate reached 12.2%, whereas the improvement at a high speed was negligible. Regarding exhaust emissions, the CO and HC emissions of the motorcycle engine after introducing H2-rich gases were similar to those generated by the original engine; however, the NOx emissions exhibited a 56.8% improvement rate [23]. Referencing the aforementioned experimental methods, the present study investigated conducting steam reforming of methanol (SRM) reactions to produce H2-rich syngas. However, SR reactions are endothermic reactions that require a supply of external energy. Exhaust heat recovery was utilized as the SRM energy source. Thus, an additional energy supply was unnecessary and hence energy consumption for reforming can be reduced. In addition, Lu utilized diesel engine exhaust heat in examining the hydrogen production characteristics of SRM. Engine exhaust heat was transferred to the reformer via a heat exchanger installed with fins. The
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 4 9 5 7 e4 9 6 8
hydrogen concentration and hydrogen production flow rate were approximately 72.6% and 17.3 L/min, respectively. The estimation results showed that approximately 92.6% of 49.5% recoverable waste heat was convertible. In other words, 36.7% of the waste heat could be recovered, converted, and stored in the produced fuel [24]. Chein et al. employed Cu/ZnO/Al2O3 catalysts of varying thicknesses to explore the SRM conversion and heat transfer effectiveness. The optimal methanol conversion efficiency and a high reforming temperature were attained when the catalyst thickness was 90% of the radius of the reformer [25]. Goicoechea et al. conducted a thermodynamic analysis of acetic acid SR. At a steam-to-acetic acid ratio of 2:1, pressure of 1 bar, and reaction temperature higher than 700 C, the hydrogen and CO production was improved and the CO/CO2 ratio was greater than 1 [26]. The exhaust heat-reforming method employed in the present study can be widely applied in internal combustion engines as well as various other fields such as industry, kiln, furnace and energy sectors. Especially in the last three sectors, which have stable and high-temperature exhaust heat sources that can be recycled as a stable energy supply for reforming reactions. The produced H2-rich syngas by reforming can be reintroduced into the furnace to improve the burning efficiency of coal or heavy oil, thereby achieving the goals of energy conservation and carbon reduction. The objective of this study is to explore a methanol reformer with engine exhaust heat recovery to produce hydrogen with steam reforming of methanol. The SRM with engine exhaust heat recovery have several advantages. Hydrogen production from SRM with engine exhaust heat, the exhaust waste heat energy could be recovered and the overall system efficiency could be improved. In addition, hydrogen could be used as the fuel for engine, and so that the emissions could also be improved generally.
Experimental instruments and methods Instruments In this study, the instruments involved five systems, namely reforming, engine, fuel supply, data acquisition and measurement and analysis, and product gas cooling systems. Fig. 1 shows a diagram of the instruments, and Table 1 shows the specifications of the methanol reformer used in this study. The main body of the reforming system comprised a reformer and external heat exchanger, and an exhaust diverter and control valve were installed at the front of the reformer to control the high temperature exhaust stream entering the reformer. The exhaust heat recovery control strategy for regulating the reforming temperature is also displayed (Fig. 2). The reforming operating temperature was controlled between 240 C and 300 C for better reforming performance and preventing the damage of reformer catalyst. Thus, the exhaust flow rates introduced into the reformer were determined by the target temperature of the reformer; and the excessive exhaust gases leave engine cylinder were conducted into another path. The excess exhaust heat was directly emitted into the atmosphere. The engine test system primarily comprised an engine dynamometer and gasoline engine.
4959
Table 2 shows the engine specifications. The fuel supply system comprises reformer and engine components. The reformer component is the methanol-water solution supply pump (YANGTECH, EMP-3706ACH-U), which was installed with an electronic mass spectrometer for monitoring the mass flow rate of the methanol-water solution. The engine component contained a fuel mass flow meter. The extraction measurement and analysis system comprised a temperature measurement system, exhaust emission analyzer (HORIBA, MEXA-554JA), and gas chromatograph (Agilent 6850 GC). Temperature was measured using K type thermocouples, which were installed at the engine exhaust stream, heat exchanger inlet, and reformer main body to facilitate calculating the thermal flow rate and amount of heat exchange as well as for monitoring and controlling the reformer temperature.
Methods and procedure The SRM reaction conducted in the current study was an endothermic reaction that required an additional energy supply to react. Thus, waste heat recovered from engine exhaust was employed as the SRM energy source. First, the basic engine was tested initially to determine the appropriate engine operation conditions for the reforming applications. The engine used in the experiment was a 1328-cc four-stroke spark-ignition engine. The engine was tested at a rotation speed of 2000e4000 rpm with the throttle opening range set between 20% and 50%. The reformer operating temperature was controlled between 240 C and 300 C. The exhaust heat flow rate varied with the engine operation conditions. Thus, the exhaust diverter and exhaust control valve opening of diverting tube were regulated to control the internal temperature of the reformer, and to prevent the catalyst from sintering at high temperatures. Regarding the experimental parameters, the S/C ratios were set at 1.2 and 1.4 respectively and the methanol supply flow rate was set at 15.8 g/min to investigate the hydrogen production of the reformer with exhaust heat recovery.
Calculations Molar flow rate of engine exhaust emissions This section describes the calculations for the molar flow rate of the exhaust gases according to the measured exhaust emission concentrations through chemical equilibrium equations under various engine operation conditions. Therefore, the exhaust heat flow rate was obtained by the measured exhaust temperature and calculated exhaust molar flow rate. After the exhaust heat flow rate was determined under various operation conditions, the reforming effect of heat recovery would be evaluated. The actual molar flow rates of all exhaust gases were determined by Eq. (1). xC8 H15 þ zðO2 þ 3:76N2 Þ/aCO þ bCO2 þ dHC þ eO2 þ fN2 þ gH2 O (1) where x, z, and g are equilibrium coefficients to be determined; a, b, d, e, and f are the exhaust concentrations measured using an emission analyzer by dry base analysis.
4960
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 4 9 5 7 e4 9 6 8
Fig. 1 e Schematic of the experimental setup.
Table 1 e The reformer specifications. Reformed fuel Catalyst Operation temperature (oC) Catalyst mass (g) Steam to carbon ratio Max. flow rate of H2 production (m3/h)
Methanol-water solution CuOeZnO/Al2O3 220e320 700 1.2:1, 1.4:1 2
Determination of exhaust flow rate entering the reformer During the experiments, an adequate additional flow rate of N2 was introduced into the diverting inlet of the reformer for calculating the flow rate of exhaust gas into the reformer. Equation (2) was used to calculate the exhaust gas molar flow rate entering the reformer. Subsequently, the waste heat flow rate was calculated. ðaCO þ bCO2 þ dHC þ eO2 þ fN2 Þ þ f 0 N2 þ gH2 O/ a1 CO þ b1 CO2 þ d1 HC þ e1 O2 þ f1 N2 þ g1 H2 O (2) The left parenthesis in Eq. (2) is the exhaust gas components introduced into the reformer pipeline, and f'N2 is the additional N2 introduced. The concentrations of the exhaust gas components are known, and aef represent the measured concentrations (Vol%). The right parenthesis in Eq. (2) refers to the components and concentrations of exhaust gases, which were analyzed using the emission analyzer after introducing additional N2 to the exhaust stream at the reformer outlet;
a1ef1 represent the measured concentrations (Vol%). The molar flow rates of CO, CO2, HC, O2, and H2O at the reformer inlet and outlet are identical (assuming that no more chemical reactions occur during the heat exchange process while the exhaust gases pass through the reformer) and the molar flow rate of the introduced N2 is known. However, the molar rates of gases entering the reformer are unknown. The total molar flow rate of the exhaust gas is obtained as substituting Eqs. (3)e(5) into Eq. (6); and the flow rates of N2 at the reformer outlet and other exhaust components can be determined by substituting the Eqs. (3)e(5) into Eqs. (7) and (8), respectively. n_N f ¼P 2 n_ex
(3)
n_1N2 n_ex þ n_addN2
(4)
n_1N2 ¼ n_N2 þ n_addN2
(5)
f1 ¼ P
X
ð1 f Þ n_ex ¼ n_addN2 f1 f
ð1 f Þ n_addN2 n_1N2 ¼ f1 f1 f a1 n_i ¼ n_1N2 f1 i ¼ CO, CO2, H2O, N2, O2, HC (measured exhaust species)
(6)
(7) (8)
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 4 9 5 7 e4 9 6 8
4961
Fig. 2 e Schematic of the exhaust heat recovery control strategy.
Determination of exhaust heat flow rate
Definition of calculation parameters
To calculate the exhaust heat flow rate, the molar flow rates and enthalpy values of all exhaust products must be determined first. The temperatures of the products differed according to the engine operation conditions. Thus, Eq. (9) was used to calculate the enthalpy values of the products at various temperatures. Equation (10) was used to calculate the exhaust heat flow rate of the exhaust gases. Table 4 shows the calculation results.
The theoretical steam reforming of methanol reaction is shown in Eq. (11a), and the actual reaction is shown in Eq. (11b). The aforementioned calculations can be used to derive a series of results. Such as, hydrogen production includes hydrogen mass percent (HMP), which represents the ratio of hydrogen mass to gasoline and hydrogen masses fed into engine, as shown in Eq. (12); and hydrogen energy percent (HEP), which represents the ratio of hydrogen heating value to gasoline and hydrogen heating values fed into engine, as shown in Eq. (13). Engine exhaust heat recovery efficiency was calculated according to Eq. (14), which is the ratio of engine exhaust heat flow rate fed into reformer to the total engine exhaust heat flow rate. With the hydrogen production and methanol consumed in reforming, molecules of hydrogen production and reformed methanol per unit exhaust heat flow rate fed into reformer were calculated by Eqs. (15) and (16), respectively. Thermal efficiency of exhaust heat was calculated by Eq. (17), which is the produced hydrogen and carbon monoxide heating values in reforming per unit exhaust heat flow rate fed into reformer. Additionally, the thermal efficiency of reforming was determined by Eq. (18). Finally, the reformer heat recycled efficiency was calculated by Eq. (19), which describes the percentage of exhaust heat flow rate entering the reformer in the process of exhaust heat exchange. Further, the reformer heat exchange efficiency was determined by Eq. (20), which represents the enthalpy change rate in the reforming to the exhaust heat flow rate fed into
ZT ðDhi ÞT ¼
Cp i dT
(9)
298:15
where, i ¼ CO, CO2, H2O, N2, O2, HC (measured exhaust species), T is the exhaust temperature. Q_ ¼ n_CO ðDhCO ÞT þ n_H2 O DhH2 O T þ n_CO2 DhCO2 T þ n_O2 DhO2 T þ n_HC ðDhHC ÞT þ n_N2 DhN2 T
(10)
Table 2 e The engine specifications. Engine manufacturer Displacement volume (c.c.) Valve train Bore Stroke (mm mm) Compression ratio Compression pressure (kg/cm2/rpm)
Honda 1328 DOHC 75.0 75.2 9.5:1 12/250
4962
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 4 9 5 7 e4 9 6 8
ðn_MeOH Þconsump ðn_MeOH ÞUEH ¼ Q_ in
Table 3 e Measurement uncertainties. o
Temperature: ±2.6 C Fuel flow rate: ±2.22% CO2: ±0.51% CO: ±0.51% HC: ±0.50% H2: ±1.01%
ðn_ HHVÞH2 þ ðn_ HHVÞCO prod ðhth Þex ¼ Q_ in
ðhth ÞRe
CH3 OHðlÞ þ H2 OðlÞ /CO2ðgÞ þ 3H2 OðgÞ
(11a)
xCH3 OHðlÞ þ yH2 OðlÞ /aCOðgÞ þ bCO2ðgÞ þ dH2ðgÞ þ eH2 OðgÞ
m_ H2 m_ gasoline þ m_ H2
(12)
hrecycled ¼
hheatrecycled
reformer
(18)
Q_ in Q_ out reformer ¼ Q_ in
(19)
reformer
P
hheatexchange
reformer
¼
ðn_ DhÞprod Q_ in
P
ðn_ DhÞreact
Re
(20)
reformer
reformer
P ¼
P ðn_ DhÞprod ðn_ DhÞreact Re Q_ in Q_ out
(21)
reformer
(13)
Measurement error analysis
reformer
(14)
Q_ ex
hheatusage
ðn_ HHVÞH2 ðn_ HHVÞgasoline þ ðn_ HHVÞH2 Q_ in
ðn_ HHVÞH2 þ ðn_ HHVÞCO prod ¼ ðn_ HHVÞMeOH react
(11b)
where, x, y, and e are the unknown coefficients to be determined, and a, b and d are the known fractions analyzed by the gas chromatography (GC).
HEP ¼
(17)
reformer
reformer; and the reformer heat usage efficiency refers to the relationship between the enthalpy change in the reforming and the exhaust heat exchange ðQ_ in Q_ out Þrefornmer in the process, as shown in Eq. (21).
HMP ¼
(16)
reformer
The total measurement error (dk) consists of bias error (b) and precision error (εk) such as that [27], dk þ b ¼ ε k
n_H2 n_H2 UEH ¼ _ Q in
prod
(15)
reformer
The precision error is repeated N times to take measurements to determine. The each series measurement which the precision index is usually expressed as follows: S Sx ¼ pffiffiffiffi N
Table 4 e The exhaust emissions under various engine operating conditions. Engine speed (rpm)
THa (%) CO CO2
2000 2500 3000 3500 4000 2000 2500 3000 3500 4000 2000 2500 3000 3500 4000 2000 2500 3000 3500 4000 a
TH: throttle opening.
HC
O2
N2 H2O(g)
Vol% 20
30
40
50
0.53 0.84 1.27 0.80 1.27 4.27 0.51 0.61 0.52 3.20 4.75 3.03 3.11 0.71 3.75 4.79 2.28 2.85 2.90 4.30
12.58 12.52 12.66 7.86 12.66 11.76 11.82 8.74 8.94 7.92 11.16 9.08 7.50 10.42 9.06 11.04 9.88 6.85 9.10 8.70
0.0084 0.0052 0.0062 0.0006 0.0008 0.0084 0.0164 0.0020 0.0006 0.0004 0.0170 0.0034 0.0056 0.0014 0.0058 0.0086 0.0014 0.0036 0.0044 0.0064
1.38 1.00 3.98 8.40 6.96 0.02 2.46 6.94 6.76 5.48 0.00 4.38 6.62 4.84 3.82 0.04 3.56 4.46 4.90 3.40
85.50 85.63 82.09 82.94 79.11 83.93 85.21 83.71 83.78 83.39 84.07 83.51 82.76 84.03 83.36 84.12 84.28 85.84 83.10 83.59
12.28 12.52 13.06 8.12 13.06 15.01 11.56 8.76 8.87 10.42 14.89 11.35 9.94 10.43 12.00 14.83 11.40 9.09 11.24 12.18
where, S is the standard deviation in N repeated measurements. The bias error considered as the systematic error remains constant in testing process. The measurement uncertainty be 95% confidence, according to the following: i12 h URSS ¼ B2 þ ðtSx Þ2 where U is the measurement uncertainty, and was calculated by the root sum square (RSS) method, B is the bias limit, and t is set equal to 2 for large samples (N > 30). According to a series analysis, the measurement uncertainties were estimated and as revealed in Table 3.
Results and discussion Basic engine tests The exhaust temperature and exhaust emission concentrations varied with the operation conditions (i.e., throttle opening and engine speed). Thus, the exhaust heat flow rate varied with the exhaust components, concentrations and total exhaust gas flow rate. To evaluate the exhaust heat for reforming and producing H2-rich gases, the temperature was
4963
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 4 9 5 7 e4 9 6 8
measured at various locations along the exhaust pipe under various engine operation conditions. A basic engine test was conducted with the engine running at various rotation speeds between 2000 and 4000 rpm with the throttle opening ranging between 20% and 50%. The parameters measured were the engine fuel consumption rate, exhaust temperature, and exhaust emission concentrations. Table 4 shows the exhaust components and concentrations. All component concentrations combined with N2 flow rate were substituted into Eq. (1) to calculate the exhaust molar flow rate, and Eqs. (9) and (10) were used to calculate the exhaust heat flow rates under various operation conditions. As shown in Fig. 3, the exhaust temperature increased with the engine speed, ranging from 524 C to 736 C. However, because the operational ceiling of the reformer temperature was 320 C, such exhaust temperatures are sufficient for SRM for engine running at 2000e3000 rpm with the throttle opening set at 20%. Table 5 shows the basic engine test results obtained under these selected conditions. The experimental measurements indicated that the exhaust temperature was between 529 C and 592 C and the exhaust heat accounted for 28%e31% of the input gasoline heating value. In this study, the basic engine test was conducted with the throttle opening set at 20%, 30%, 40%, and 50% with the engine speeds between 2000 and 4000 rpm. In experimental processes, the constant speed mode of dynamometer controller was used. Under each engine speed, the engine output (torque or power) was determined by the throttle opening. However, in the reforming experiments, the engine speed was set between 2000 and 3000 rpm and the throttle opening was fixed at 20%, because the exhaust temperatures and waste heat are sufficient for methanol reforming reactions according to the results of basic engine test.
ratio was 1.2 and 1.4, the hydrogen produced concentration is stable, approaching 75%. The hydrogen concentration does not vary with the reforming temperature or engine speed, whereas the molar flow rate of the hydrogen produced increases with the reforming temperature but decreases as the engine speed increases. This is because when the reforming temperature increases, the exhaust gas energy for reforming increases, thereby increasing the likelihood of methanol being converted into hydrogen. However, the total gas flow rate from the exhaust pipe increases with engine speed. Therefore, the opening of the exhaust control valve mounted in the reformer pipe must be reduced to regulate the temperature of reformer. Thus the velocity of the exhaust gas into the reformer increases and the exhaust stream cannot contact the wall of heat exchanger uniformly, which weakens the heat exchange effect and lowering the methanol conversion efficiency and thereby reducing the molar flow rate of hydrogen when the engine speed increases.
Hydrogen production characteristics Hydrogen produced by reforming could be widely applied in fuel cells, engines, industrial sectors, boilers, power mechanics, and various other devices and fields. The produced H2-rich syngas in this study demonstrates potential for future use in engines. Equations (11) and (12) were used to calculate the HMP (Fig. 5) and HEP (Fig. 6) for the investigated engine. Both figures depict similar tendencies. HMP and HEP increase with the reforming temperature because the hydrogen production increases with the reforming temperature. HMP and HEP decrease with increasing engine speed because hydrogen production cannot be improved and the engine consumes more fuel as engine speed increases. The HMP ranges between 1.37% and 3.63% and HEP ranges between 4.4% and 11.08%.
Reforming test SR is an endothermic reaction, and engine exhaust pipe emits a substantial amount of heat flow that can be utilized as energy for reforming reactions to generate H2-rich syngas. Thus, a steam reforming of methanol for hydrogen production test is conducted with engine speed set at 2000e3000 rpm and the throttle opening set at 20%. Fig. 4(a) shows that when the S/C
Exhaust heat recovering and reforming Equations (2)e(10) and (14) and (15) are employed to identify the relationship between the exhaust heat recovery and hydrogen production. Fig. 7 depicts the hydrogen molar flow rate per unit exhaust heat is 0.62e1.6 mol/MJ. When the reforming temperature increases, the molar flow
Table 5 e Exhaust waste heat and engine input heating value under various engine speed. (TH: 20%). Engine speed (rpm) Tex Q_ ex Q_ gasoline o
2000
2500
3000
Fig. 3 e Exhaust temperature and exhaust heat flow rate under various engine speeds and throttle openings.
C
kW
kW
529 534 537 537 560 564 567 570 583 586 589 592
12.93 13.10 12.97 12.97 15.89 15.66 16.06 15.96 18.81 18.98 19.17 19.43
45.75 46.11 45.39 45.39 53.99 52.60 53.67 52.99 60.08 60.44 60.08 61.16
4964
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 4 9 5 7 e4 9 6 8
Fig. 5 e Effect of reforming temperature on hydrogen mass percentage.
Fig. 4 e The hydrogen concentration and molar flow rate under various engine speeds and reforming temperatures.
rate per unit exhaust heat increases. The hydrogen production is close to a stable value when the temperature reaches around 260e280 C. However, the increase in engine speed and S/C reduces the molar flow rate of the hydrogen production. This is because high S/C ratios represent an increase of water in the reformed fuel. Liquid water absorbs latent heat for vaporization and lowering the temperature, which inhibits reforming reactions. Thus, the produced hydrogen molar flow rate declines even under the same engine speed and heat input. Fig. 8 shows the molar flow rate of methanol that can be reformed from each unit of exhaust heat. The tendency of molar flow rate of reformed methanol in Fig. 8 resembles that in Fig. 7. The term is 0.2e0.57 mol/MJ. Fig. 9 depicts the effect of the engine speed on the thermal efficiency of exhaust heat. As shown in Eq. (16), the thermal efficiency of the exhaust heat is the ratio of heating value of the H2-sysgas generated to the exhaust heat entering the reformer. The thermal efficiency of exhaust heat decreases with increasing engine speed, but increases with the reforming temperature. The thermal efficiency of exhaust heat is between 17.7% and 47.5%.
Fig. 6 e Effect of reforming temperature on hydrogen energy percentage.
Methanol conversion efficiency The methanol conversion efficiency affects the production of hydrogen. As shown in Fig. 10, the methanol conversion efficiency increases with the reforming temperature, and the hydrogen production flow rate increases with the methanol conversion efficiency. However, the methanol conversion efficiency decreases when the S/C ratio and engine speed increase. This phenomenon is caused by an increase of latent heat due to the amount of water addition and decrease in the heat exchange rate of reformer. The hydrogen production molar flow rate is approximately 0.58e1.34 mol/min. Fig. 11 demonstrates the thermal efficiency of reforming. It reveals a tendency similar to that in Fig. 10. The thermal efficiency of reforming increases with the reformer heat usage efficiency and reforming temperature, but decreases with the engine speed and S/C ratio. The thermal efficiency of reforming is 46.5%e104.2%, where the value over 100% should be due to the gain of hydrogen from water by the exhaust heat recovery.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 4 9 5 7 e4 9 6 8
Fig. 7 e Hydrogen production per unit waste heat under various operating conditions.
4965
Fig. 9 e The relationship of thermal efficiency of exhaust heat and engine speed.
Figs. 12e15 depict the analysis of the exhaust heat recycled, heat exchange rate, and thermal efficiency of reforming, from which the effect of the heat exchange on reforming can be identified. Fig. 12 shows the relationship between the reformer heat exchange rate and exhaust heat recycled. The reformer heat exchange rate and exhaust heat recycled decrease with lifting engine speed but increase with rising reforming temperature. Fig. 13 shows the relationship between reformer heat recycled and heat exchange rate, which are derived from Eqs. (18) and (19). In the figure, the reformer heat exchange decreases as the engine speed and S/C ratio increase, whereas the reformer heat recycled increases with engine speed. In the exhaust heat recovery process for reforming, it is difficult to insulate the reformer system completely to prevent the heat loss, hence the reformer heat
exchange rate decreases as the exhaust mass flow rate increases. Fig. 14 depicts the relationship between the thermal efficiency of reforming and heat recovery rate. The thermal efficiency of reforming decreases with the increase in the reformer heat recycled rate because of lower heat use rate and heat dissipation. Further in Fig. 15, it demonstrates that the methanol conversion efficiency increases with the reformer heat usage rate. That is, when the available thermal energy for reforming increases, the methanol conversion efficiency is improved. Fig. 16(a) and (b) illustrate the engine exhaust mass flow rate and methanol conversion efficiency with reformer heat usage efficiency under S/C ratio of 1.2 and 1.4 respectively. The figures show that the engine exhaust mass flow rate is inversely developed with the reformer heat usage efficiency; whereas the methanol conversion efficiency increases with reformer heat usage efficiency. High reformer heat usage rate representing a high reaction enthalpy is generated in the reforming process, thereby high methanol
Fig. 8 e Reformed methanol per unit waste heat under various operating conditions.
Fig. 10 e The relationship of hydrogen molar flow rate and methanol conversion efficiency under various operating conditions.
The characteristics of exhaust heat recovery and heat exchange
4966
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 4 9 5 7 e4 9 6 8
Fig. 11 e The relationship of thermal efficiency of reforming and methanol conversion efficiency under various operating conditions.
Fig. 14 e The effect of reformer heat recycled on thermal efficiency of reforming.
conversion efficiency is achieved. The engine exhaust mass flow rate increases at high engine speed. However, it lowers the heat exchange in the reforming reactions. In Fig. 16(a) and (b), the reformer heat usage efficiency at the identical engine speed differs clearly because of the different S/C ratios.
Conclusions
Fig. 12 e The relationship of exhaust heat recycled with reformer heat exchange rate.
Fig. 13 e The relationship of reformer heat recycled with reformer heat exchange rate.
The methanol steam reforming system with exhaust heat recycling system was investigated in this study. The basic engine was tested initially ahead of a series of experiments. The results showed that, under various engine operation conditions, the exhaust temperature and emissions differ according to the throttle opening and engine speed. Thus, the exhaust heat flow rate varies with the exhaust components and concentrations. When the engine speed was set between 2000 and 3000 rpm and throttle opening at 20%, the measured exhaust temperature between 529 C and 592 C was enough for methanol reforming. Under the operation conditions, the
Fig. 15 e Reformer heat usage rate on methanol conversion efficiency under various operating conditions.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 4 9 5 7 e4 9 6 8
exhaust heat accounted for 28%e31% of the input gasoline heating value. In the reforming processes with exhaust recycling, when S/ C was set at 1.2 and 1.4, the produced hydrogen concentration was stable, approaching 75%. The hydrogen molar flow rate increased with the reforming temperature but decreased with the engine speed. The produced hydrogen molar flow rate was approximately 0.58e1.34 mol/min. When the parameters for the engine application were evaluated, the results show that the HMP was 1.37%e3.63%; HEP was 4.4%e11.08%. Further considering the exhaust heat effectiveness on reforming, the hydrogen production per unit exhaust heat was 0.62e1.6 mol/MJ, and the reformed methanol per unit exhaust heat was 0.2e0.57 mol/MJ. In addition, the thermal efficiency of exhaust heat was 17.7%e 47.5%, and the thermal efficiency of reforming was 46.5%e 104.2%. Overall, reformer heat recovery rate increased with engine speed, whereas exhaust heat recovery decreased as engine speed increased. At high engine speed, exhaust mass flow rate increased, resulting in poor heat exchange effect. Finally, methanol conversion efficiency increased with reformer heat exchange rate.
Fig. 16 e Engine exhaust mass flow rate and methanol conversion efficiency with reformer heat usage rate.
4967
Acknowledgments This study was supported and subsidized by the Ministry of Science and Technology (MOST 103-2221-E-168-022) and the NSC-EPA Air Pollution Prevention Project (NSC102-EPA-F-007002) of Taiwan. The authors hereby express their appreciation to MOST, NSC, and EPA of Taiwan.
references
[1] Chiriac R, Racovitza A, Podevin P, Descombes G. On the possibility to reduce CO2 emissions of heat engines fuelled partially with hydrogen produced by waste heat recovery. Int J Hydrogen Energy 2015;40:15856e63. [2] Zhang H, Wang H, Zhu X, Qiu YJ, Li K, Chen R, et al. A review of waste heat recovery technologies towards molten slag in steel industry. Appl Energy 2013;112:956e66. [3] Karellas S, Leontaritis AD, Panousis G, Bellos E, Kakaras E. Energetic and exergetic analysis of waste heat recovery systems in the cement industry. Energy 2013;58:147e56. [4] Hossain SN, Bari S. Waste heat recovery from the exhaust of a diesel generator using Rankine Cycle. Energy Convers Manag 2013;75:141e51. [5] Bari S, Hossain SN. Waste heat recovery from a diesel engine using shell and tube heat exchanger. Appl Therm Eng 2013;61:355e63. [6] Shudo T, Shima Y, Fuji T. Production of dimethyl ether and hydrogen by methanol reforming for an HCCI engine system with waste heat recovery e continuous control of fuel ignitability and utilization of exhaust gas heat. Int J Hydrogen Energy 2009;34:7638e47. [7] Bueno AV, de Oliveira MLM. Glycerol steam reforming in a bench scale continuous flow heat recovery reactor. Int J Hydrogen Energy 2013;38:13991e4001. [8] Das LM, Gulati R. A comparative evaluation of the performance characteristics of a spark ignition engine using hydrogen and compressed natural gas as alternative fuels. Int J Hydrogen Energy 2000;25:783e93. [9] Li H, Karim GA. A prediction study of a spark ignition supercharged hydrogen engine. Int J Hydrogen Energy 2005;30:1491e9. [10] Al-Baghdadi MARS, Al-Janabi HAS. Exhaust emissions from an SI engine operating on gaseous fuel mixtures containing hydrogen. Int J Hydrogen Energy 2003;44:3143e50. [11] Park B, Kwon S. Compact design of oxidative steam reforming of methanol assisted by blending hydrogen peroxide. Int J Hydrogen Energy 2015;40:12697e704. [12] Toyir J, Piscina PR, Homs N. Ga-promoted copper-based catalysts highly selective for methanol steam reforming to hydrogen; relation with the hydrogenation of CO2 to methanol. Int J Hydrogen Energy 2015;40:11261e6. [13] Chang CC, Hsu CC, Chang CT, Chen YP, Liaw BJ, Chen YZ. Effect of noble metal on oxidative steam reforming of methanol over CuO/ZnO/Al2O3 catalysts. Int J Hydrogen Energy 2012;37:11176e84. [14] Paasikallio V, Azhari A, Kihlman J, Simell P, Lehtonen J. Oxidative steam reforming of pyrolysis oil aqueous fraction with zirconia pre-conversion catalyst. Int J Hydrogen Energy 2015;40:12088e96. [15] Lee KY, Huang YJ. Low CO generation on tunable oxygen vacancies of non-precious metallic Cu/ZnO catalysts for partial oxidation of methanol reaction. Appl Catal B Environ 2014;150e151:506e14.
4968
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 4 9 5 7 e4 9 6 8
[16] Corigliano O, Fragiacomo P. Technical analysis of hydrogenrich stream generation through CO2 reforming of biogas by using numerical modeling. Fuel 2015;158:538e48. [17] Horng RF, Chen CR, Wu TS, Chan CH. Cold start response of a small methanol reformer by partial oxidation reforming of hydrogen for fuel cell. Appl Therm Eng 2006;26:1115e24. [18] Horng RF, Chou HM, Lee CH, Tsai HT. Characteristics of hydrogen produced by partial oxidation and auto-thermal reforming in a small methanol reformer. J Power Sources 2006;163:119e24. [19] Kamarudin SK, Daud WRW, Som AM, Takriff MS, Mohammad AW, Loke YK. Design of a fuel processor unit for PEM fuel cell via shortcut design method. Chem Eng J 2004;104:7e17. [20] Yoon HC, Erickson PA, Kim HM. Lowering the O2/CH3OH ratio in autothermal reforming of methanol by using a reduced copper-based catalyst. Int J Hydrogen Energy 2008;33:6619e26. [21] Mohammadi A, Shioji M, Nakaib Y, Ishikura W, Tabo E. Performance and combustion characteristics of a direct injection SI hydrogen engine. Int J Hydrogen Energy 2007;32:296e304. [22] Akansu SO, Kahraman N, Ceper B. Experimental study on a spark ignition engine fuelled by methaneehydrogen mixtures. Int J Hydrogen Energy 2007;32:4279e84. [23] Horng RF, Wen CS, Liauh CT, Chao Y, Huang CT. Driving characteristics of a motorcycle fuelled with hydrogen-rich gas produced by an onboard plasma reformer. Int J Hydrogen Energy 2008;33:7619e29. [24] Lu LC, Lu JH. Energy analysis for hydrogen generation with the waste heat of internal combustion engine. Adv Mater Res 2012;512e515:1492e8. [25] Chein RY, Chen LC, Chen YC, Chung JN. Heat transfer effects on the methanol-steam reforming with partially filled catalyst layers. Int J Hydrogen Energy 2009;34:5398e408.
[26] Goicoechea S, Ehrich H, Arias PL, Kockmann N. Thermodynamic analysis of acetic acid steam reforming for hydrogen production. J Power Sources 2015;279:312e22. [27] Abernethy RB, Benedict RP, Dowdell RB. ASME measurement uncertainty. Trans ASME J Fluids Eng 1985;107:161e4.
Nomenclature cp: specific heat capacity, (kJ kg1 K1) HEP: hydrogen energy percentage HHV: higher heating value, (kJ kg1) HMP: hydrogen mass percentage Dh: change of enthalpy, (kJ) _ mass flow rate, (kg sec1) m: _ molar flow rate, (mole min1) n: _ heat flow rate, (kW) Q: Greek letters h: efficiency, (%) hheatexchange: heat exchange efficiency, (%) hheatrecyced: heat recycled efficiency, (%) hheatusage: heat usage efficiency, (%) hth: thermal efficiency, (%) Subscripts addN2: added nitrogen consump: consumption ex: engine exhaust prod: product Re: reforming React: reactant Recycled: engine exhaust heat recovery UEH: unit exhaust heat