the heat transfer characteristics of molten waste plastics
THE HEAT TRANSFER CHARACTERISTICS OF MOLTEN WASTE PLASTICS PYROLYSIS IN A VERTICAL FALLING FILM REACTOR ZECHEN JIN*, LIJIE YIN*, DEZHEN CHEN*, YUYAN HU*, YUANJIE JIA* *Thermal and Environmental Engineering Institute, Tongji University, 200092, Shanghai, China
SUMMARY: The heat transfer characteristics of molten waste plastics pyrolysis in a vertical falling film reactor is studied by experiment and numerical simulation, and the heat transfer coefficients are determined and compared. The contours of temperature distribution and the thickness of the liquid film in the reactor are predicted, and the influence of heating temperatures on the heat transfer coefficients is discussed.
1. INTRODUCTION In China, about 25 million tons of municipal plastic waste (MPW) was generated in 2015 (Lv et al., 2005; http://data.stats.gov), but only 25%-30% of MPW were energy recycled, chemical recycled and mechanical recycled (Wu et al., 2010; www.irinbank.com). As one of the ways of energy and chemical recycled, pyrolysis is attracting more and more attention from international scholars. The pyrolysis of plastics is an endothermic process, which is divided into several stages including: melting (from solid to liquid), volatile evaporating (from liquid to gas) and coke formation, requiring quantities of heat. While the molten plastics is usually characterized with high viscosity and low thermal conductivity, so how to enhance the heat transfer during pyrolysis is very important. The liquid film heat transfer is an effective method for heat transfer enhancement (Nikhin et al., 2013), can be used to deal with high viscosity and low thermal conductivity materials (La et al., 2006; López et al., 2011), and has been widely applied to various industrial production (Khaled, 2015; Mehrjouei et al., 2012). Volume of fluid (VOF) model is a surface tracking method based on a fixed Eulerian mesh, can be used to simulate the gas-liquid two-phase flow with big free phase interface (Yan et al., 2010; Ho et al., 2011)，and is fit to deal with high viscosity and low thermal conductivity materials. Guo simulated the falling film model of the gas-liquid two-phase in a evaporative condenser, and analyzed the effect of wall heating flux on the temperature of the liquid film (Guo, 2009). Liu simulated the evaporation process of water in a filler-type saturator, and discussed the influence of inlet velocity and feeding rate on the heating and mass transfer coefficients (Liu, 2012). Though VOF model has been used to simulated gas-liquid two-phase flow in falling film reactors, the research mainly based on Newtonian fluid, and there is little research on high viscosity molten fluid. In this work, the second stage of plastics pyrolysis was experimented in a vertical falling film reactor (VFFR) and numerical simulated by VOF model, the temperature distribution in the reactor was obtained, and the heat transfer coefficient was determinted and compared. Proceedings Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium/ 2 - 6 October 2017 S. Margherita di Pula, Cagliari, Italy / © 2017 by CISA Publisher, Italy
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
2. EXPERIMENTAL METHODS The pyrolysis plant is shown in Fig.1. It is composed of two parts: the pyrolysis system, the heating and control systems. The pyrolysis system consists of a melting tank and a falling film plate. The vertical falling film plate is 0.05×1 m. A feed inlet and a nitrogen inlet are located at the top of the melting tank. There are thermocouples mounted to monitor temperatures of heating wall and liquid film, which can be displayed and recorded by an electronic panel. A glass window in front of the falling film plate is used to observe the liquid film. The pyrolysis volatile outlet is at the top of the falling film plate. Another nitrogen inlet is at the bottom of the falling film plate to purge the glass surface.
1-Control panel, 2-Nitrogen inlet, 3-Feedstock inlet, 4-Melting tank, 5-Glass window, 6Thermocouple, 7-Falling film plate, 8-Heating tube, 9-Volatile outlet Figure 1. VFFR for plastics pyrolysis The plastics used is a mixture of polyethylene(PE), polypropylene(PP) and polystyrene(PS) with blending ratio of 24:9:8 according to the proportion of fresh waste plastics (Mustafa, 1993; Finch, 1992), produced by the Daqing Refining and Chemical Company (Daqing, China). The proximate analyses are shown in Tab.1. The heat required for endothermic reactions in the process of pyrolysis is measured by Differential scanning calorimetry (DSC) as shown in Fig.2 (Cafiero et al., 2015; Frederick et al., 1975), which includes the chemical reaction heat of the feedstock, the latent heat of vaporization, and the sensible heat in the main range of weightless temperature. We proposed a term of apparent heat transfer coefficient ( happ ) (Wang, 2015) to comprehend the influence of chemical reaction, fluid flow and volatile evaporation on the pyrolysis process. The happ of the molten mixed-plastics pyrolysis in the VFFR is computed by:
ΔH = happ A(Tw − Ts )
(1)
Where ΔH is the enthalpy of molten plastics pyrolysis, A is the area of the molten plastics flowing through the heating wall, Tw is the temperature of heating wall and Ts is the temperatures of liquid film. Tw and Ts are obtained by online monitoring. Since the experiments are carried out at constant temperatures (823,848,873,898K), the happ is an average value computed for 20 minutes in the process of pyrolysis. Table 1. Proximate analyses of the feedstock
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017 Feedsto
Proximate analysis（wt./%）
ck
M
V
FC
A
PE
0.38
91.59
0.19
7.84
PP
0.13
99.72
0.11
0.04
PS
0.46
96.82
1.42
1.30
mixed-plastics
2
o
570.1 C o
356.8 C
0
Heat Flow (W/g)
1
Δ=541.4±5.9J/g
-1 o
530.1 C
-2 0
200
400 600 o Temperature( C)
800
Figure 2. DSC heat flow curve obtained for molten mixed-plastics 3. SIMULATION METHODS Considering that there are little solid residue remained, the products include two kinds of volatile evaporating (non-condensable gas and oil vapor). The chemical reaction equation is: Molten mixed-plastics → oil vapor + non-condensable gas ΔH = 541.4 kJ.kg-1 The conservation equations are shown in Tab.2. Table 2. Conservation equations of molten plastics pyrolysis in the falling film reactor Volume fraction equation ∂ρlα l + ∇ρlα l u = Sl ∂t
Momentum equation
ρ l :Density of liquid phase; α l :Volume fractions of liquid; :Velocity of average mass; Sl :Source of mass transfer from u molten plastics pyrolysis
ρ :Density of fluid; µ : Viscosity of fluid; C : Surface tension κ :Curvature; ∇γ :Normal vector of any point
T ∂ρ u + ∇ ⋅ ( ρ uu ) = −∇p + ∇ ⋅ [ µ (∇u + ∇u )] + ρ g + Cκ∇ γ coefficient; ∂t
Energy equation ∂ρE + ∇ ⋅ [u ( ρE + p)] = ∇ ⋅ [k∇T ] + ωhvW ∂t
Species transport equation
r ∂ρYi + ∇（ ⋅ ρ uYi）= ∇ ⋅ [ ρ Di∇Yi ] + SYi ∂t
E :Energy of average mass; T :Temperature of average mass; k :Effective thermal conductivity of fluid; ω :Pyrolysis rate; hv :Change of enthalpy during pyrolysis; W :Molecular weight of molten plastics
Yi :Mass fraction of two kinds of volatiles; Di :Diffusion coefficient of two kinds of volatiles; SY :Averaged source term of two kinds of volatiles i
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
ρ g :Density of gas phase; ρ i :Densities of two kinds of volatiles; µl :Viscosity of liquid phase; µ g :Viscosity of gas phase; µi :Viscosities of two kinds of volatiles; kl :Effective thermal conductivity of liquid phase; k g :Effective thermal conductivity of gas phase; ki :Effective thermal conductivity of two kinds of volatiles
Constitutive relations ρ =αl ρl + (1 − αl ) ρ g , ρ g = ∑Yi ρi i
µ =αl µl + (1 − αl ) µg , µg = ∑Yi µi i
k =αl kl + (1 − αl ) kg , kg = ∑Yi ki i
The simulation is carried out on the Ansys Fluent 15.0. The structure scheme and grid are shown in Fig.3. The time step size is 10-4 s. The pyrolysis of the molten mixed-plastics is a heterogeneous reaction, and the reaction rate is defined by user defined function (UDF). The parameters used in the simulation are based on the experiments, and shown in Tab.3.
Figure 3. Structure scheme and grid Table 3. Physical parameters used for simulations Significance Temperature of heating wall Velocity of molten mixed-plastisc
Significance
Simulation data
823, 848, 873,
Dynamic viscosity of oil vapor
0.000007kg ⋅ m
Dynamic viscosity of molten mixed-
0.048kg ⋅ m
898K -1 0.5m ⋅ s
Temperature of molten mixed-
plastics Dynamic viscosity of non-condensable
523K
plastics Density of oil vapor
8.66kg ⋅ m
Density of molten mixed-plastics
891kg ⋅ m
Density of non-condensable gas
1.62kg ⋅ m
Specific heat of oil vapor Specific heat of molten mixedplastics Specific heat of non-condensable gas
Simulation data
K
-1
K
-1
K
-1
gas
-3
-3
-3
-1
0.0178W ⋅ m
Conductivity of molten mixed-plastics
7.1W ⋅ m
Conductivity of non-condensable gas
-2
⋅ s-1
⋅ s-1
0.000008kg ⋅ m
Conductivity of oil vapor
-1
-2
-1
⋅ s-1
⋅ K-1
⋅ K-1(Wang et al.,
2013) -2 -1 0.03W ⋅ m ⋅ K
2282.7J ⋅ kg
-1
⋅
Conductivity of heating wall
16.27W ⋅ m
4316.7J ⋅ kg
-1
⋅
Grid number
101304
4076.1J ⋅ kg
-1
⋅
-2
⋅ K-1
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
4. RESULTS AND DISCUSSIONS
4.1 Temperature distribution The liquid film is heated when flow down along with heating wall, and it decomposes into volatiles when temperature reaches to the pyrolysis temperature. As shown in Fig.4(a), the liquid film temperature at bottom of the reactor is higher than that at top. Fig.4(b) shows the temperature distribution in horizontal direction at different heights. The temperature decreases dramatically from the heating surface to the center of reactor at each height.
T=823K y=0.025m z=0.01 z=0.39 z=0.59 z=0.79 z=0.99
700
Temperature (K)
800
600
500
0.000
0.005
0.010
0.015
x (m)
(a)
(b)
Figure 4. Contours of temperature distribution in the reactor(a) and temperature distribution in horizontal direction(b) 4.2 Thickness of the liquid film
1.0 823K
0.6
Height (m)
0.8
0.4 0.2 0.0 4.0
4.5 5.0 5.5 6.0 Thickness of the liquid film (mm)
6.5
Figure 5. Thickness of the liquid film in longitudinal direction The thickness of the liquid film decreases along with the flow direction for pyrolysis as shown in fig.5. Besides, it can be seen that the interface of the liquid film features a wavy structure, is not smooth enough, maybe due to the existence of shear stress resulted from the relative velocities between the gas phase and the liquid phase. 4.3 Effect of heating temperature The heating temperature does not only affect the heating rate of liquid film, but also affects
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
the stability of the fluid film during pyrolysis process, so it is a key factor that affects pyrolysis products component. Fig.6 shows the averaged fluid temperatures in longitudinal direction under different heating temperatures. From the inlet, the heating rate slowly decreases when the liquid flow down along the surface, indicating the liquid need to absorb more heat to participate in chemical reaction so that the reduction of sensible heat. It is found that the distribution of temperature is fluctuant, and the range of fluctuation increases with the raising pyrolysis temperature. Because there are more volatile produced at higher temperature, leading to the interface of temperature featuring not a smooth structure. Besides, the averaged temperature of the fluid increases with higher temperature of the heating wall.
1.0
0.6
823K 848K
Height (m)
0.8
0.4
873K 898K
0.2 0.0 500
550
600 Temperature (K)
650
700
Figure 6. Averaged fluid temperature in longitudinal direction 4.4 Heat transfer coefficient The heat apparent heat be seen that temperatures.
transfer coefficient is determined according to Equation (1). Fig.7 shows the transfer coefficient calculated from experiments (he) and simulations (hs). It can the heat transfer coefficients decrease with the increasing of the heating The result is in accordance with the result from Wang (Wang, 2015). She
calculated the happ of plastics pyrolysis in rotary kiln, and found that the happ of molten plastics decrease with the increasing of temperature then become constant. The he and hs have a similar trend with a deviation of 24.7% to 34.4%, from 3160.8 W.m-2.K-1 of 823K to 2297.7 W.m-2.K-1 of 898K and from 4254.6 W.m-2.K-1 of 823K to 3051.2 W.m-2.K-1 of 898K. The difference between he and hs may be caused by the flow of liquid film on the falling film plate, for the flow will become worse in the later stage of the experiment. 4200
hs
-2.
-1
Apparent heat transfer cofficient (W m K )
.
he 3600
3000
2400
1800 820
840
860 Temperaure (K)
880
900
Figure 7. Apparent heat transfer coefficient of experiments and simulations
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
5. CONCLUSIONS The main conclusions as follows: 1. The temperature in horizontal direction decreases dramatically from the heating surface to the center of the reactor, and the averaged temperature in longitudinal direction increases from the top to the bottom of the reactor at the pyrolysis temperature of 823K. 2. The thickness of the liquid film decreases along with the flow direction for pyrolysis. The interface of the temperature features a wavy structure, and the averaged fluid temperature increases with the increasing of pyrolysis temperature. 3. The apparent heat transfer coefficient computed from experiments and simulations have a same downward trend with the increasing of pyrolysis temperature as well as a deviation of 24.7% to 34.4% ACKNOWLEDGEMENT The work is financed by the National Natural Science Foundation of China (Grant No. 51503154) and the Shanghai Municipal Science and Technology Commission Fund for improving the economy in the Yangtze River Delta region (Grant No. 12195811100).
REFERENCES Cafiero L., Fabbri D., Trinca E., Tuffi R., Ciprioti S.V. (2015) Thermal and spectroscopic (TG/DSC–FTIR) characterization of mixed plastics for materials and energy recovery under pyrolytic conditions. Journal of Thermal Analysis and Calorimetry, vol. 121, 1-9. Finch. C. A. (1992). Plastics Recycling: Products and Process (ed. R. J. Ehrig). Hanser Publisher, vol. 136. Frederick W.J., Mentzer C.C. (1975). Determination of heats of volatilization for polymers by differential scanning calorimetry. J. Appl. Polym. Sci., vol. 19, 1799–804. Guo C.Q. (2009). Numerical simulation of heat and mass transfer in plate evaporative condenser. Journal of South China University of Technology (Natural Science Edition), vol. 37, 53-57. Ho C.D., Chang H., Chen H.J., Chang C.L., Li H.H., Chang Y.Y. (2011). CFD simulation of the two-phase flow for a falling film microreactor. International Journal of Heat and Mass Transfer, vol. 54, 3740–3748. http://data.stats.gov. Khaled A.R.A. (2015). Investigation of heat transfer enhancement inside developing gravitydriven films along a vertical permeable plate subject to nonuniform suction, Adv. Mech. Eng, 7 (7), 1-13 . La D., Li M.D., Qin W., Li W.J., Li Y.D. (2006). Study on heat transfer characteristics of falling film evaporation process. Applied Energy Technology, vol. 5, 6–10. Liu G.B. (2012). Measuring of gas-liquid temperature and investigation of heat and mass transfer of gas-liquid flow in saturator of hat cycle. Shanghai Jiao Tong University. López A., Marco de.I., Caballero B.M., Laresgoiti M.F., Adrados A. (2011). Influence of time and temperature on pyrolysis of plastic wastes in a semi-batch reactor. Chemical Engineering Journal, vol. 173, 62–71. Lv C.Y., Yang X.H., Yang B.G. (2005). Study on separation and utilization technology of
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
municipal plastic waste. Environmental Engineering, vol. 23, 45-47. Mehrjouei M., Müller S., Moler D. (2012). Removal of fuel oxygenates from water using advanced oxidation technologies by means of falling film reactor. Chem. Eng. J., vol. 211-212, 353-359 . Mustafa N. (1993). Plastics waste management: disposal, recycling and reuse. Environmental science and pollution control series: 1. Nikhin M., Issam M. (2013). Investigation of eddy diffusivity and heat transfer coefficient for freefalling turbulent liquid films subjected to sensible heating. Int. J. Heat Mass Transf, vol. 64, 647660. Wang H. (2015). Heat transfer characteristics of MSW and its typical components at different pyrolysis stages. Tongji University. Wang C., Ma X.B., Wang H., Chen D.Z. (2013) Study on Effective Thermal Conductivity Coefficient of Plastic Wastes Pyrolysis Process. Materials Review, vol. 27, 108-111. Wu C.F., Williams P.T. (2010). Pyrolysis-gasification of plastics, mixed plastics and real-world plastic waste with and without Ni-Mg-Al catalyst. Fuel, vol. 89, 3022-3032. www.irinbank.com. Yan K., Che D.F. (2010). A coupled model for simulation of the gas–liquid two-phase flow with complex flow patterns. International Journal of Multiphase Flow 36 (2010) 333–348.
SUMMARY: The heat transfer characteristics of molten waste plastics pyrolysis in a vertical falling film reactor is studied by experiment and numerical simulation, and the heat transfer coefficients are determined and compared. The contours of temperature distribution and the thickness of the liquid film in the reactor are predicted, and the influence of heating temperatures on the heat transfer coefficients is discussed.
1. INTRODUCTION In China, about 25 million tons of municipal plastic waste (MPW) was generated in 2015 (Lv et al., 2005; http://data.stats.gov), but only 25%-30% of MPW were energy recycled, chemical recycled and mechanical recycled (Wu et al., 2010; www.irinbank.com). As one of the ways of energy and chemical recycled, pyrolysis is attracting more and more attention from international scholars. The pyrolysis of plastics is an endothermic process, which is divided into several stages including: melting (from solid to liquid), volatile evaporating (from liquid to gas) and coke formation, requiring quantities of heat. While the molten plastics is usually characterized with high viscosity and low thermal conductivity, so how to enhance the heat transfer during pyrolysis is very important. The liquid film heat transfer is an effective method for heat transfer enhancement (Nikhin et al., 2013), can be used to deal with high viscosity and low thermal conductivity materials (La et al., 2006; López et al., 2011), and has been widely applied to various industrial production (Khaled, 2015; Mehrjouei et al., 2012). Volume of fluid (VOF) model is a surface tracking method based on a fixed Eulerian mesh, can be used to simulate the gas-liquid two-phase flow with big free phase interface (Yan et al., 2010; Ho et al., 2011)，and is fit to deal with high viscosity and low thermal conductivity materials. Guo simulated the falling film model of the gas-liquid two-phase in a evaporative condenser, and analyzed the effect of wall heating flux on the temperature of the liquid film (Guo, 2009). Liu simulated the evaporation process of water in a filler-type saturator, and discussed the influence of inlet velocity and feeding rate on the heating and mass transfer coefficients (Liu, 2012). Though VOF model has been used to simulated gas-liquid two-phase flow in falling film reactors, the research mainly based on Newtonian fluid, and there is little research on high viscosity molten fluid. In this work, the second stage of plastics pyrolysis was experimented in a vertical falling film reactor (VFFR) and numerical simulated by VOF model, the temperature distribution in the reactor was obtained, and the heat transfer coefficient was determinted and compared. Proceedings Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium/ 2 - 6 October 2017 S. Margherita di Pula, Cagliari, Italy / © 2017 by CISA Publisher, Italy
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
2. EXPERIMENTAL METHODS The pyrolysis plant is shown in Fig.1. It is composed of two parts: the pyrolysis system, the heating and control systems. The pyrolysis system consists of a melting tank and a falling film plate. The vertical falling film plate is 0.05×1 m. A feed inlet and a nitrogen inlet are located at the top of the melting tank. There are thermocouples mounted to monitor temperatures of heating wall and liquid film, which can be displayed and recorded by an electronic panel. A glass window in front of the falling film plate is used to observe the liquid film. The pyrolysis volatile outlet is at the top of the falling film plate. Another nitrogen inlet is at the bottom of the falling film plate to purge the glass surface.
1-Control panel, 2-Nitrogen inlet, 3-Feedstock inlet, 4-Melting tank, 5-Glass window, 6Thermocouple, 7-Falling film plate, 8-Heating tube, 9-Volatile outlet Figure 1. VFFR for plastics pyrolysis The plastics used is a mixture of polyethylene(PE), polypropylene(PP) and polystyrene(PS) with blending ratio of 24:9:8 according to the proportion of fresh waste plastics (Mustafa, 1993; Finch, 1992), produced by the Daqing Refining and Chemical Company (Daqing, China). The proximate analyses are shown in Tab.1. The heat required for endothermic reactions in the process of pyrolysis is measured by Differential scanning calorimetry (DSC) as shown in Fig.2 (Cafiero et al., 2015; Frederick et al., 1975), which includes the chemical reaction heat of the feedstock, the latent heat of vaporization, and the sensible heat in the main range of weightless temperature. We proposed a term of apparent heat transfer coefficient ( happ ) (Wang, 2015) to comprehend the influence of chemical reaction, fluid flow and volatile evaporation on the pyrolysis process. The happ of the molten mixed-plastics pyrolysis in the VFFR is computed by:
ΔH = happ A(Tw − Ts )
(1)
Where ΔH is the enthalpy of molten plastics pyrolysis, A is the area of the molten plastics flowing through the heating wall, Tw is the temperature of heating wall and Ts is the temperatures of liquid film. Tw and Ts are obtained by online monitoring. Since the experiments are carried out at constant temperatures (823,848,873,898K), the happ is an average value computed for 20 minutes in the process of pyrolysis. Table 1. Proximate analyses of the feedstock
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017 Feedsto
Proximate analysis（wt./%）
ck
M
V
FC
A
PE
0.38
91.59
0.19
7.84
PP
0.13
99.72
0.11
0.04
PS
0.46
96.82
1.42
1.30
mixed-plastics
2
o
570.1 C o
356.8 C
0
Heat Flow (W/g)
1
Δ=541.4±5.9J/g
-1 o
530.1 C
-2 0
200
400 600 o Temperature( C)
800
Figure 2. DSC heat flow curve obtained for molten mixed-plastics 3. SIMULATION METHODS Considering that there are little solid residue remained, the products include two kinds of volatile evaporating (non-condensable gas and oil vapor). The chemical reaction equation is: Molten mixed-plastics → oil vapor + non-condensable gas ΔH = 541.4 kJ.kg-1 The conservation equations are shown in Tab.2. Table 2. Conservation equations of molten plastics pyrolysis in the falling film reactor Volume fraction equation ∂ρlα l + ∇ρlα l u = Sl ∂t
Momentum equation
ρ l :Density of liquid phase; α l :Volume fractions of liquid; :Velocity of average mass; Sl :Source of mass transfer from u molten plastics pyrolysis
ρ :Density of fluid; µ : Viscosity of fluid; C : Surface tension κ :Curvature; ∇γ :Normal vector of any point
T ∂ρ u + ∇ ⋅ ( ρ uu ) = −∇p + ∇ ⋅ [ µ (∇u + ∇u )] + ρ g + Cκ∇ γ coefficient; ∂t
Energy equation ∂ρE + ∇ ⋅ [u ( ρE + p)] = ∇ ⋅ [k∇T ] + ωhvW ∂t
Species transport equation
r ∂ρYi + ∇（ ⋅ ρ uYi）= ∇ ⋅ [ ρ Di∇Yi ] + SYi ∂t
E :Energy of average mass; T :Temperature of average mass; k :Effective thermal conductivity of fluid; ω :Pyrolysis rate; hv :Change of enthalpy during pyrolysis; W :Molecular weight of molten plastics
Yi :Mass fraction of two kinds of volatiles; Di :Diffusion coefficient of two kinds of volatiles; SY :Averaged source term of two kinds of volatiles i
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
ρ g :Density of gas phase; ρ i :Densities of two kinds of volatiles; µl :Viscosity of liquid phase; µ g :Viscosity of gas phase; µi :Viscosities of two kinds of volatiles; kl :Effective thermal conductivity of liquid phase; k g :Effective thermal conductivity of gas phase; ki :Effective thermal conductivity of two kinds of volatiles
Constitutive relations ρ =αl ρl + (1 − αl ) ρ g , ρ g = ∑Yi ρi i
µ =αl µl + (1 − αl ) µg , µg = ∑Yi µi i
k =αl kl + (1 − αl ) kg , kg = ∑Yi ki i
The simulation is carried out on the Ansys Fluent 15.0. The structure scheme and grid are shown in Fig.3. The time step size is 10-4 s. The pyrolysis of the molten mixed-plastics is a heterogeneous reaction, and the reaction rate is defined by user defined function (UDF). The parameters used in the simulation are based on the experiments, and shown in Tab.3.
Figure 3. Structure scheme and grid Table 3. Physical parameters used for simulations Significance Temperature of heating wall Velocity of molten mixed-plastisc
Significance
Simulation data
823, 848, 873,
Dynamic viscosity of oil vapor
0.000007kg ⋅ m
Dynamic viscosity of molten mixed-
0.048kg ⋅ m
898K -1 0.5m ⋅ s
Temperature of molten mixed-
plastics Dynamic viscosity of non-condensable
523K
plastics Density of oil vapor
8.66kg ⋅ m
Density of molten mixed-plastics
891kg ⋅ m
Density of non-condensable gas
1.62kg ⋅ m
Specific heat of oil vapor Specific heat of molten mixedplastics Specific heat of non-condensable gas
Simulation data
K
-1
K
-1
K
-1
gas
-3
-3
-3
-1
0.0178W ⋅ m
Conductivity of molten mixed-plastics
7.1W ⋅ m
Conductivity of non-condensable gas
-2
⋅ s-1
⋅ s-1
0.000008kg ⋅ m
Conductivity of oil vapor
-1
-2
-1
⋅ s-1
⋅ K-1
⋅ K-1(Wang et al.,
2013) -2 -1 0.03W ⋅ m ⋅ K
2282.7J ⋅ kg
-1
⋅
Conductivity of heating wall
16.27W ⋅ m
4316.7J ⋅ kg
-1
⋅
Grid number
101304
4076.1J ⋅ kg
-1
⋅
-2
⋅ K-1
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
4. RESULTS AND DISCUSSIONS
4.1 Temperature distribution The liquid film is heated when flow down along with heating wall, and it decomposes into volatiles when temperature reaches to the pyrolysis temperature. As shown in Fig.4(a), the liquid film temperature at bottom of the reactor is higher than that at top. Fig.4(b) shows the temperature distribution in horizontal direction at different heights. The temperature decreases dramatically from the heating surface to the center of reactor at each height.
T=823K y=0.025m z=0.01 z=0.39 z=0.59 z=0.79 z=0.99
700
Temperature (K)
800
600
500
0.000
0.005
0.010
0.015
x (m)
(a)
(b)
Figure 4. Contours of temperature distribution in the reactor(a) and temperature distribution in horizontal direction(b) 4.2 Thickness of the liquid film
1.0 823K
0.6
Height (m)
0.8
0.4 0.2 0.0 4.0
4.5 5.0 5.5 6.0 Thickness of the liquid film (mm)
6.5
Figure 5. Thickness of the liquid film in longitudinal direction The thickness of the liquid film decreases along with the flow direction for pyrolysis as shown in fig.5. Besides, it can be seen that the interface of the liquid film features a wavy structure, is not smooth enough, maybe due to the existence of shear stress resulted from the relative velocities between the gas phase and the liquid phase. 4.3 Effect of heating temperature The heating temperature does not only affect the heating rate of liquid film, but also affects
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
the stability of the fluid film during pyrolysis process, so it is a key factor that affects pyrolysis products component. Fig.6 shows the averaged fluid temperatures in longitudinal direction under different heating temperatures. From the inlet, the heating rate slowly decreases when the liquid flow down along the surface, indicating the liquid need to absorb more heat to participate in chemical reaction so that the reduction of sensible heat. It is found that the distribution of temperature is fluctuant, and the range of fluctuation increases with the raising pyrolysis temperature. Because there are more volatile produced at higher temperature, leading to the interface of temperature featuring not a smooth structure. Besides, the averaged temperature of the fluid increases with higher temperature of the heating wall.
1.0
0.6
823K 848K
Height (m)
0.8
0.4
873K 898K
0.2 0.0 500
550
600 Temperature (K)
650
700
Figure 6. Averaged fluid temperature in longitudinal direction 4.4 Heat transfer coefficient The heat apparent heat be seen that temperatures.
transfer coefficient is determined according to Equation (1). Fig.7 shows the transfer coefficient calculated from experiments (he) and simulations (hs). It can the heat transfer coefficients decrease with the increasing of the heating The result is in accordance with the result from Wang (Wang, 2015). She
calculated the happ of plastics pyrolysis in rotary kiln, and found that the happ of molten plastics decrease with the increasing of temperature then become constant. The he and hs have a similar trend with a deviation of 24.7% to 34.4%, from 3160.8 W.m-2.K-1 of 823K to 2297.7 W.m-2.K-1 of 898K and from 4254.6 W.m-2.K-1 of 823K to 3051.2 W.m-2.K-1 of 898K. The difference between he and hs may be caused by the flow of liquid film on the falling film plate, for the flow will become worse in the later stage of the experiment. 4200
hs
-2.
-1
Apparent heat transfer cofficient (W m K )
.
he 3600
3000
2400
1800 820
840
860 Temperaure (K)
880
900
Figure 7. Apparent heat transfer coefficient of experiments and simulations
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
5. CONCLUSIONS The main conclusions as follows: 1. The temperature in horizontal direction decreases dramatically from the heating surface to the center of the reactor, and the averaged temperature in longitudinal direction increases from the top to the bottom of the reactor at the pyrolysis temperature of 823K. 2. The thickness of the liquid film decreases along with the flow direction for pyrolysis. The interface of the temperature features a wavy structure, and the averaged fluid temperature increases with the increasing of pyrolysis temperature. 3. The apparent heat transfer coefficient computed from experiments and simulations have a same downward trend with the increasing of pyrolysis temperature as well as a deviation of 24.7% to 34.4% ACKNOWLEDGEMENT The work is financed by the National Natural Science Foundation of China (Grant No. 51503154) and the Shanghai Municipal Science and Technology Commission Fund for improving the economy in the Yangtze River Delta region (Grant No. 12195811100).
REFERENCES Cafiero L., Fabbri D., Trinca E., Tuffi R., Ciprioti S.V. (2015) Thermal and spectroscopic (TG/DSC–FTIR) characterization of mixed plastics for materials and energy recovery under pyrolytic conditions. Journal of Thermal Analysis and Calorimetry, vol. 121, 1-9. Finch. C. A. (1992). Plastics Recycling: Products and Process (ed. R. J. Ehrig). Hanser Publisher, vol. 136. Frederick W.J., Mentzer C.C. (1975). Determination of heats of volatilization for polymers by differential scanning calorimetry. J. Appl. Polym. Sci., vol. 19, 1799–804. Guo C.Q. (2009). Numerical simulation of heat and mass transfer in plate evaporative condenser. Journal of South China University of Technology (Natural Science Edition), vol. 37, 53-57. Ho C.D., Chang H., Chen H.J., Chang C.L., Li H.H., Chang Y.Y. (2011). CFD simulation of the two-phase flow for a falling film microreactor. International Journal of Heat and Mass Transfer, vol. 54, 3740–3748. http://data.stats.gov. Khaled A.R.A. (2015). Investigation of heat transfer enhancement inside developing gravitydriven films along a vertical permeable plate subject to nonuniform suction, Adv. Mech. Eng, 7 (7), 1-13 . La D., Li M.D., Qin W., Li W.J., Li Y.D. (2006). Study on heat transfer characteristics of falling film evaporation process. Applied Energy Technology, vol. 5, 6–10. Liu G.B. (2012). Measuring of gas-liquid temperature and investigation of heat and mass transfer of gas-liquid flow in saturator of hat cycle. Shanghai Jiao Tong University. López A., Marco de.I., Caballero B.M., Laresgoiti M.F., Adrados A. (2011). Influence of time and temperature on pyrolysis of plastic wastes in a semi-batch reactor. Chemical Engineering Journal, vol. 173, 62–71. Lv C.Y., Yang X.H., Yang B.G. (2005). Study on separation and utilization technology of
Sardinia 2017 / Sixteenth International Waste Management and Landfill Symposium / 2 - 6 October 2017
municipal plastic waste. Environmental Engineering, vol. 23, 45-47. Mehrjouei M., Müller S., Moler D. (2012). Removal of fuel oxygenates from water using advanced oxidation technologies by means of falling film reactor. Chem. Eng. J., vol. 211-212, 353-359 . Mustafa N. (1993). Plastics waste management: disposal, recycling and reuse. Environmental science and pollution control series: 1. Nikhin M., Issam M. (2013). Investigation of eddy diffusivity and heat transfer coefficient for freefalling turbulent liquid films subjected to sensible heating. Int. J. Heat Mass Transf, vol. 64, 647660. Wang H. (2015). Heat transfer characteristics of MSW and its typical components at different pyrolysis stages. Tongji University. Wang C., Ma X.B., Wang H., Chen D.Z. (2013) Study on Effective Thermal Conductivity Coefficient of Plastic Wastes Pyrolysis Process. Materials Review, vol. 27, 108-111. Wu C.F., Williams P.T. (2010). Pyrolysis-gasification of plastics, mixed plastics and real-world plastic waste with and without Ni-Mg-Al catalyst. Fuel, vol. 89, 3022-3032. www.irinbank.com. Yan K., Che D.F. (2010). A coupled model for simulation of the gas–liquid two-phase flow with complex flow patterns. International Journal of Multiphase Flow 36 (2010) 333–348.
