Chemical kinetics method for evaluating the thermal stability of Organic Rankine Cycle working fluids

Chemical kinetics method for evaluating the thermal stability of Organic Rankine Cycle working fluids

Applied Thermal Engineering 100 (2016) 708–713 Contents lists available at ScienceDirect Applied Thermal Engineering j o u r n a l h o m e p a g e :...

489KB Sizes 0 Downloads 62 Views

Applied Thermal Engineering 100 (2016) 708–713

Contents lists available at ScienceDirect

Applied Thermal Engineering j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / a p t h e r m e n g

Research Paper

Chemical kinetics method for evaluating the thermal stability of Organic Rankine Cycle working fluids Xiaoye Dai a, Lin Shi a,*, Qingsong An b, Weizhong Qian c a Key Laboratory of Thermal Science and Power Engineering of Ministry of Education of China, Department of Thermal Engineering, Tsinghua University, Beijing, China b Key Laboratory of Efficient Utilization of Low and Medium Grade Energy, MOE, School of Mechanical Engineering, Tianjin University, Tianjin, China c Beijing Key Laboratory of Green Chemical Reaction Engineering and Technology, Department of Chemical Engineering, Tsinghua University, Beijing, China

H I G H L I G H T S

• • • •

G R A P H I C A L

A B S T R A C T

A chemical kinetics method for evaluating the thermal stability of ORC working fluids. An experimental system to obtain the apparent kinetic parameters. An apparent chemical kinetics model to predict the thermal stability. The n-pentane pre-exponential factor was 7.5275 × 1012 s−1 and the activation energy was 227.57 kJ/mol.

A R T I C L E

I N F O

Article history: Received 9 December 2015 Accepted 25 February 2016 Available online 2 March 2016 Key word: Organic Rankine Cycle (ORC) Thermal stability Apparent chemical kinetics model n-pentane

A B S T R A C T

Organic Rankine Cycle (ORC) systems are widely used to generate electricity with industrial waste heat and renewable energy. Transcritical ORCs with high temperature heat sources are more attractive than subcritical ORCs due to their lower exergy losses, higher thermal efficiencies and higher work outputs. The working fluid thermal stability is the primary consideration in the working fluid selection due to decomposition at high temperatures. This paper presents a chemical kinetics method for evaluating the thermal stability of ORC working fluids. A chemical kinetics experimental system was built with n-pentane as the test working fluid. The influences of pressure and temperature were analyzed experimentally and theoretically. An apparent chemical kinetics model was established to predict the thermal stability with the apparent kinetic parameters of n-pentane measured experimentally. This apparent chemical kinetics model gives significant guidance for the working fluid selection and ORC system design. © 2016 Elsevier Ltd. All rights reserved.

* Corresponding author. Tel.: +86 10 62787613; fax: +86 10 62787613. E-mail address: [email protected] (L. Shi). http://dx.doi.org/10.1016/j.applthermaleng.2016.02.091 1359-4311/© 2016 Elsevier Ltd. All rights reserved.

X.Y. Dai et al./Applied Thermal Engineering 100 (2016) 708–713

1. Introduction The Organic Rankine Cycle (ORC) is a promising technology for industrial waste heat recovery and renewable energy utilization. Subcritical ORCs have been widely used in the geothermal power, solar power, biomass power and industrial waste heat recovery systems [1–4]. Subcritical ORC heat source temperatures are usually less than 150 °C, which limits their thermal efficiencies and leads to high costs. Transcritical ORCs, which use high temperature heat sources (150–350 °C), are more efficient and less expensive. The literature suggests that transcritical ORCs have much better efficiencies than subcritical ORCs with the same working fluids and heat source temperatures. Karellas and Schuster [5] calculated the thermal efficiencies of transcritical ORCs with R134a and R245fa to show that the thermal efficiencies were 10–20% higher than subcritical ORCs. Similar results were obtained by Chen et al. [6] for transcritical ORCs with mixture working fluids with 10~30% improvements. Transcritical ORC temperatures match better with the heat sources, so they have lower exergy losses. Higher thermal efficiencies and work outputs are then obtained because of the higher temperature heat sources, and the system costs are reduced by smaller heat exchanger areas. However, some new problems have not been addressed in the studies of transcritical ORCs. The primary problem is the thermal stability of the ORC working fluids at high temperatures. All organics decompose at high temperatures. If decomposition occurs in the ORC system, the thermal efficiency and work output will decrease because of the different working conditions. Decomposition products like non-condensable gases and deposits can seriously damage the components and compromise safety. Thus, thermal stability is the primary consideration for working fluid selection and ORC system design. Studies of working fluid thermal stability have been conducted since the early 1970s [7]. However, the heat source temperatures were usually less than 150 °C and the working fluids were found to be stable in early ORC studies. Working fluid thermal stability has attracted much interest recently because of the higher temperature heat sources. Andersen and Bruno [8] presented a method for rapid screening of the working fluid thermal stability for ORCs. The reaction rate constants of some hydrocarbon working fluids were measured. Ginosar et al. [9] measured the thermochemical decomposition of cyclopentane at 240, 300, and 350 °C at 4.3 MPa in a glass-lined heated tube. Air was found to have a large effect on the decomposition. Pasetti et al. [10] used pressure changes to imply decomposition to identify safe temperatures for cyclopentane, iso-pentane and n-butane. However, previous studies have not considered the relationships between thermal stability and temperature or time. Previous studies only considered special working conditions and cannot be applied to other general conditions. Thus, more thermal stability experimental data and a universal evaluation method are needed. This paper presents a chemical kinetics method for thermal stability to evaluate the ORC working fluids. A chemical kinetics measurement system was used to get the data for an apparent chemical kinetics model. The model could predict the thermal stability at different temperatures and times. N-pentane was used as the test fluid because it was one of the most desirable ORC working fluids.

dC A = −kC A dt

709

(1)

where CA is the reactant concentration, t is the time and k is the reaction rate constant which is only related to the reaction temperature. Equation 1 can also be written as

− ln

CA = − ln C = − ln (1 − x ) = kt C0

(2)

where C0 is the original reactant concentration, T is the reaction time, CA is the concentration at time t, C is the concentration ratio and x is the reactant decomposition concentration. Equation 2 shows that lnC is related linear with t because k is constant at a given temperature. Thus, the reaction rate constant, k, can be calculated from lnC-t curves if the reactant concentrations are measured at different times at a given temperature. The reaction rate constant can then be used to predict the reactant decomposition concentration at different times at the same temperature. The decomposition products may also decompose and the reaction mechanism becomes more complex when the reactant is mixed with products. The product influence is ignored in this paper because the reactant decomposition rate is very slow in the measured temperature range. The assumption that the decomposition is a first order reaction is accurate enough for stability studies. The Arrhenius Equation is used to describe the reaction ratio at different temperatures.

k = Ae − Ea

(3)

RT

where k is the reaction rate constant, A is the pre-exponential factor, Ea is the activation energy, R is the gas constant and T is the temperature. The pre-exponential factor, A, and the activation energy, Ea, are normally assumed to be constants in the Arrhenius Equation, and they are independent of any environmental condition. The Arrhenius Equation is a classical equation in chemical kinetics and can be used in most practical environmental conditions. The Arrhenius Equation can also be written as

ln k = −

Ea 1 + ln A R T

(4)

Equation 4 shows that lnk is linearly related with 1/T because the pre-exponential factor, A, and activation energy, Ea, are constant for given reactant. Thus, the pre-exponential factor and the activation energy can be calculated from the lnk-1/T curve if the reaction rate constants are measured at different temperatures. The apparent chemical kinetics model can then be obtained by combining Equations 2 and 4.

ln (1− x ) = − Ate − Ea

RT

(5)

This apparent chemical kinetics model can be used for all the ORC working fluids because the derivational process in theory is suitable for any organic. The model is accurate enough when the experimental temperature is not very high (to ensure the condition of the Arrhenius Equation) and the decomposition concentration is not very big (to ensure the condition of first order reactions). Equation 5 shows that the variables in the model are the reactant decomposition concentration, time and temperature. Thus, the reactant decomposition concentration at any time and temperature conditions can be calculated once the pre-exponential factor and the activation energy are measured.

2. Theory 3. Experiment Thermochemical decomposition of a simple substance is a unimolecular reaction in terms of the chemical kinetics. Unimolecular reactions are usually first order reactions except in some extreme conditions. The reaction rate of a first order reaction is proportional to the reactant concentration as

3.1. Chemicals N-pentane was chosen as the working fluid because it is one of the most promising ORC working fluids. The n-pentane samples were

710

X.Y. Dai et al./Applied Thermal Engineering 100 (2016) 708–713

obtained from commercial suppliers with purity higher than 99.87%. The purity was confirmed by gas chromatography (GC) to be not less than 99.9%. High pressure nitrogen was obtained from commercial suppliers and was used to clean the experimental equipment. 3.2. Test system The thermochemical decomposition of a working fluid is an irreversible chain scission reaction. Thus, a sustained reaction in a reactor is equivalent to a batch reaction in a cycle. A tubular reactor was then chosen for the experimental system to avoid the inaccuracies and safety problems of long experimentals. The chemical kinetics experimental system is shown in Fig. 1. The inside volume of the reactor was 200 mL. The reactor was made of 316L stainless steel, which had favorable mechanical and corrosion properties at high temperatures. Also, stainless steel has little effect on the decomposition reaction. The pipes and valves in contact with the working fluid were also made of stainless steel. The reactor was placed in a customized heating furnace. The furnace could maintain the temperatures stable and the temperature differences inside the reactor were less than 3 °C. Thermocouples and pressure sensors were installed in the reactor to measure the temperature and pressure changes. A Proportion Integration Differentiation (PID) controller was used to adjust the heating power of the furnace. An injector and a corresponding interface were used to fill the working fluids into the reactor. The reactor was also connected to a nitrogen cylinder for system purging. Online sampling was used to improve the experimental accuracy. A sampling system was designed to connect the reactor and the GC. The whole sampling system was placed in a holding furnace to avoid errors due to the condensation of the working fluids. A calibrated needle valve (V-10) and a reducing valve (RV) were used to control the working fluid mass flow using the same valve opening and pressure difference. A ten-way valve was used to simultaneously shunt the flow into the different detectors in the GC. The high pressure nitrogen was used to purge the system for 30 min before the experiment. At the beginning of each experiment, all the valves were confirmed to be closed. Then, the vacuum pump evacuated the air inside the system when valves V-3 and V-7 were open. This procedure was continued for more than 30 min until the system pressure was less than 10 Pa. This procedure was very important because the oxygen in the air had a big effect on the decomposition reaction. The n-pentane filling mass was calculated by the commercial property software REFPROP 9.0. The density of n-pentane at the given experimental temperature and pressure can be calculated by the REFPROP 9.0. The filling mass was the density of n-pentane times the reactor volume.

Fig. 1. Schematic layout of the experimental system.

The injector was cooled to avoid gasification of n-pentane during operation. Then, the injector was connected to the reactor interface and n-pentane flowed into the reactor. All these steps were done quickly and carefully. After the filling procedure, all the valves were closed and the heating furnace and holding furnace were started. The time when the temperature reached the set point was recorded. The time to reach the test temperature was less than 0.5 h. This heat period was short enough when compared to the experimental period and had little effect on the decomposition experiments. To start the sampling procedure, valve V-10 was opened to the set opening calibration. Then valve RV was opened to control the pressure to the set point. The ten-way valve was opened after the fluid flowed steadily for 30 s and the samples were analyzed by the GC. After each sampling procedure, the sampling system was purified by the vacuum pump to remove the residual sample from the last procedure. The sample mass was very small compared to the filled mass (less than 1%), so the sampling procedure should have little influence on the decomposition reaction. After the experiment, valves V-2, V-3, AV, V-10 and RV were opened. High pressure nitrogen was used to purge all the system for 30 min to remove the residual n-pentane. The reactor was removed and cleaned with ethanol and acetone until the solvent was colorless. Then, the reactor was heated to 110 °C to remove the residual solvent. 3.3. Analytical methods The reactant decomposition rate was measured by a commercial gas chromatograph (GC) with a flame ionization detector (FID) and a thermal conductivity detector (TCD). The organic products were analyzed by FID with a PLOT Al 2 O 3 capillary column (50 m × 0.53 mm × 0.025 mm). Inorganic products like hydrogen were analyzed by the TCD with the TDX-01 column (2 m × 3 mm). The pressure before the GC column was 0.06 MPa. The GC column temperature was 60 °C and the split–splitless GC injector inlet temperature was 150 °C. The FID temperature was 250 °C and the TCD was 200 °C. Nitrogen was used as the carrier and makeup gas. The reactant concentration ratio was calculated from the GC peak area percent. The results were mainly from the FID GC peaks since the hydrogen production was very small. 4. Results and discussion A blank test at a low temperature was used as a reference to investigate the effect of high temperatures on n-pentane. The temperature was 100 °C and the pressure was 1 MPa. The results are shown in Fig. 2. The reactant concentration was almost constant and the biggest change was only 0.03%, which could be considered as the measured uncertainty of the GC. Thus, n-pentane had a good thermal stability at low temperatures. The pressure effect on the decomposition in the apparent chemical kinetics model was considered to be negligible. This assumption was verified by first measuring the reactant decomposition at the same temperature but different pressures. The temperature was 340 °C and the test period was 24 h. All the decomposition would be repeated for 5 times to get more accurate results. The results are shown in Fig. 3. The decomposition had occurred at 340 °C with an obvious change in the reactant concentration. The experimental pressure was increased from 0.4 MPa to 3.4 MPa (higher than the critical pressure of n-pentane), but the reactant concentration was almost constant as the pressure rose. Thus, the results showed that the pressure did not affect the reactant decomposition rates. This phenomenon can be explained by chemical kinetics theory that the decomposition of n-pentane is an irreversible chain scission

X.Y. Dai et al./Applied Thermal Engineering 100 (2016) 708–713

711

Fig. 2. The blank test at 100 °C and 1 MPa.

Fig. 4. N-pentane concentrations at 340 °C.

reaction. Irreversible reactions are only a function of the temperature, so the pressure had no effect on the chain scission. Decomposition may include some reversible reactions like dehydrogenation and subsequent decomposition which were closely related to the pressure, but these reversible reactions were slow in this temperature range and cannot influence the total decomposition rate. Thus, the experimental results showed that the pressure did not influence the decomposition and that the model assumption was appropriate. For this reason, the pressure was kept at 1 MPa for the other chemical kinetics tests. The uniform low pressure allowed convenient control of the mass flow rate and avoided leaks and safety problems that can occur at high pressures. The measured reactant concentration changed with time at 340 °C are shown in Fig. 4. The results show that the concentration decreased over time by almost 1% at 70 h. The reactant concentration curve is converted to an lnC-t curve in Fig. 5. The data points were almost linear, which is consistent with the apparent chemical kinetics model. The determination coefficient (R2) of the lnC-t curve was 0.9824. The slope of the lnC-t curve (-k) was calculated using a least squares method with the reaction rate constant of n-pentane at 340 °C found to be (3.2998 ± 0.1971) × 10−8 s−1. The reaction rate constants at different temperatures were then measured to obtain the lnk-1/T curve for n-pentane. The reactant concentrations at different times at 330 °C and 350 °C are shown

in Figs. 6 and 7. The results showed that temperature had a big influence on the decomposition reaction with the reactant decomposition rate increasing when the temperature was increased 10 °C. In the same 70 h experimental period, the

Fig. 3. N-pentane concentrations at different pressures (340 °C, 24 h).

Fig. 6. N-pentane concentrations at 330 °C.

Fig. 5. N-pentane lnC-t curve at 340 °C.

712

X.Y. Dai et al./Applied Thermal Engineering 100 (2016) 708–713

Fig. 9. N-pentane lnC-t curve at 350 °C.

Fig. 7. N-pentane concentrations at 350 °C.

The experimental results from the study of Andersen and Bruno [8] at 315 °C were used to verify the model. They gave the n-pentane

reaction rate constant at 315 °C as 4.70 × 10−9 s−1 and the calculation result of the apparent chemical kinetics model was 4.62 × 10−9 s−1. This agreement is very good considering that the reaction rate constant is very sensitive to the experimental measurement. The n-pentane reactant decomposition concentration ratio at any temperature and time can be calculated by this apparent chemical kinetics model. For example, the maximum allowable composition change in engineering practice is assumed to be 1%. The model predicts that the 1% composition change at 280 °C will not occur until 478 days. Since the cycle reaction period is at least one order of magnitude larger than the sustained reaction period [9], the actual time is nearly 13.1 years for a composition change of 1% at 280 °C. This period is approximately the service life of an ORC system. Thus, 280 °C can be considered to be a safe temperature for n-pentane in ORCs. The model gives an actual time of nearly 2.3 years for a 1% composition change at 300 °C. This time is far shorter than the service life of the ORC system. Thus, 300 °C would be excessive for n-pentane in ORCs. This analysis shows that this apparent chemical kinetics model is much more useful for working fluid selection and ORC system design than scattered experimental data. And the influence of decomposition in ORC systems is very important for evaluating the thermal stability. Further work is needed to investigate the influence of decomposition in ORC systems.

Fig. 8. N-pentane lnC-t curve at 330 °C.

Fig. 10. The lnk-1/T curve of n-pentane.

concentration decreased by about 0.4% at 330 °C, 1% at 340 °C and 1.5% at 350 °C. Thus, the temperature has a strong effect on the decomposition reaction. The reactant concentration curves at 330 and 350 °C was converted to the lnC-t curves shown in Figs. 8 and 9. The data points were again almost linear distribution as in the 340 °C curve. The R2 of the lnC-t curve was 0.9668 at 330 °C and 0.9705 at 350 °C. The slope of the lnC-t curves were again calculated to get the reaction rate constant of n-pentane of (1.4246 ± 0.1176) × 10−8 s−1 at 330 °C and (6.1062 ± 0.4745) × 10−8 s−1 at 350 °C. The lnk-1/T curve is then shown in Fig. 10 using the reaction rate constants at different temperatures. The data points are almost linear, which is consistent with the apparent chemical kinetics model. The R2 of the lnk-1/T curves was 0.9874. The slope (-Ea/R) and intercept (lnA) of the lnk-1/T curves were then calculated using a least squares method. The n-pentane pre-exponential factor was found to be 7.5275 × 1012 s−1 and the n-pentane activation energy was 227.57 kJ/mol. The complete apparent chemical kinetics model of n-pentane is then:

ln (1 − x ) = − Ate − Ea

RT

, A = 7.5275 × 1012 s −1, Ea = 227.57 kJ mol

(6)

X.Y. Dai et al./Applied Thermal Engineering 100 (2016) 708–713

5. Conclusions This paper presents a chemical kinetics study method for evaluating the thermal stability of ORC working fluids using an apparent chemical kinetics model. An experimental system was designed to obtain the kinetic parameters including the pre-exponential factor and the activation energy. N-pentane was chosen as the working fluid. However, this method is not only good for n-pentane but also for other working fluids. The experimental results showed that pressure did not affect the decomposition. This conclusion confirms the chemical kinetics theory and indirectly verifies the model accuracy. The reaction rate constants of n-pentane were measured at 330 °C, 340 °C and 350 °C to obtain the apparent chemical kinetics parameters of n-pentane. The thermal stability of ORC working fluids can be described at any temperature and time by the apparent chemical kinetics model. This model gives important guidance for working fluid selection and ORC system design. And the decomposition influence to the ORC systems will be a potential research direction. Acknowledgment This work was supported by the State Key Program of the National Natural Science Foundation of China (Grant No. 51236004), and the Science Fund for Creative Research Group (No. 51321002). Nomenclature CA C0 C t

Reactant concentration (mol/L) Original reactant concentration (mol/L) Concentration ratio (%) Reaction time (s)

k T x A Ea R

713

Reaction rate constant (s−1) Reaction temperature (K) Reactant decomposition concentration (%) Pre-exponential factor (s−1) Activation energy (kJ/mol) Gas constant (kJ/mol·K)

References [1] Q. Liu, A.J. Shen, Y.Y. Duan, Parametric optimization and performance analyses of geothermal organic Rankine cycles using R600a/R601a mixtures as working fluids, Appl. Energy 148 (2015) 410–420. [2] B. Twomey, P.A. Jacobs, H. Gurgenci, Dynamic performance estimation of small-scale solar cogeneration with an organic Rankine cycle using a scroll expander, Appl. Therm. Eng. 51 (1) (2013) 1307–1316. [3] M. Uris, J.I. Linares, E. Arenas, Size optimization of a biomass-fired cogeneration plant CHP/CCHP (Combined heat and power/Combined heat, cooling and power) based on Organic Rankine Cycle for a district network in Spain, Energy 88 (2015) 935–945. [4] N.F.T. Ozdil, M.R. Segmen, A. Tantekin, Thermodynamic analysis of an Organic Rankine Cycle (ORC) based on industrial data, Appl. Therm. Eng. 91 (2015) 43–52. [5] S. Karellas, A. Schuster, Supercritical fluid parameters in organic Rankine cycle applications, Int. J. Thermodyn. 11 ((3) (2008) 101–108. [6] H. Chen, D.Y. Goswami, M.M. Rahman, E.K. Stefanakos, A supercritical Rankine cycle using zeotropic mixture working fluids for the conversion of low-grade heat into power, Energy 36 (1) (2011) 549–555. [7] H.M. Curran, Use of Organic Working Fluids in Rankine Engines, Hittman Associates, Inc., Columbia, MD, USA, 1979, pp. 2–3. [8] W.C. Andersen, T.J. Bruno, Rapid screening of fluids for chemical stability in organic Rankine cycle applications, Ind. Eng. Chem. Res. 44 (15) (2005) 5560–5566. [9] D.M. Ginosar, L.M. Petkovic, D.P. Guillen, Thermal stability of cyclopentane as an organic Rankine cycle working fluid, Energy Fuels 25 (9) (2011) 4138–4144. [10] M. Pasetti, C.M. Invernizzi, P. Iora, Thermal stability of working fluids for organic Rankine cycles: an improved survey method and experimental results for cyclopentane, isopentane and n-butane, Appl. Therm. Eng. 73 (1) (2014) 764–774.