- Email: [email protected]
Selection and performance assessment of Phase Change Materials for heating, ventilation and air-conditioning applications
Energy Conversion and Management 89 (2015) 260–269
Contents lists available at ScienceDirect
Energy Conversion and Management journal homepage: www.elsevier.com/locate/enconman
Selection and performance assessment of Phase Change Materials for heating, ventilation and air-conditioning applications Monisha Rastogi, Aditya Chauhan, Rahul Vaish ⇑, Anil Kishan School of Engineering, Indian Institute of Technology Mandi, Mandi 175 001, India
a r t i c l e
i n f o
Article history: Received 2 July 2014 Accepted 29 September 2014
Keywords: Phase Change Materials Heating, ventilation and air-conditioning Ashby approach Materials selection
a b s t r a c t The rapid commercialization of Phase Change Materials (PCMs) for heating, ventilation and air-conditioning (HVAC) applications, has paved way for effective utilization of ambient thermal fluctuations. However, given a long list of contemporary candidates, it is crucial to select the best material to obtain maximum efficiency for any given application. This article attempts to extend Multiple Criteria Decision Making (MCDM) approach for ranking and selecting PCMs for domestic HVAC application. Firstly, Ashby approach has been employed for determining two novel figure of merits (FOM) to grade PCMs performance. The FOMs thus obtained were subjected to Pareto Optimality test. The graded materials were ranked using Technique for Order Preference by Similarity to Ideal Solution (TOPSIS). The relative weights for the different attributes were calculated using Shannon’s entropy method. In order to justify the rankings obtained, the top materials were subjected to a standard simulation study to evaluate their relative performance using PCMExpress with the aim of maintaining human comfort temperature. It was observed that the results obtained by simulation are in good agreement with those obtained using MCDM approach. The candidates with the best ranks showed significant improvement in ameliorating the temperature conditions. Thus it can be concluded that integration of MCDM approach for PCMs selection would prove to an economical and swift alternative technique for ranking and screening of materials. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Phase Change Materials (PCMs) have been the subject of rigorous investigation in the past few decades [1–8]. These materials have been explored for thermal energy storage (TES) [9–11], heating, ventilation and air-conditioning (HVAC) [12–14], temperature regulation for industrial application [15,16], electricity generation [17] and even soilless crop growth [18]. Recent years have seen significant improvement in terms of material’s chemistry [11,19], performance and engineering [15,20]. Techniques like microencapsulation [20] and macro-encapsulation [8], have enabled high efficiency versatile operation. This has been followed by rapid commercialization of the technology and is now available for industrial and domestic use at large [21]. These advancements can be credited to the growing awareness towards global climate change and the need to push forward with cleaner and cheaper technology. PCM as the name implies is a category of various organic or inorganic compounds exhibiting a change of phase within the required operating temperature range [1,8]. As the materials ⇑ Corresponding author. Tel.: +91 1905 237921; fax: +91 1905 237945. E-mail address: [email protected] (R. Vaish). http://dx.doi.org/10.1016/j.enconman.2014.09.077 0196-8904/Ó 2014 Elsevier Ltd. All rights reserved.
undergo phase transformation, they absorb or release a large amount of thermal energy. Since the energy flow is associated with a change in the physical state of the material, the entirety of the heat exchange is approximately isothermal in nature. Thus, if a suitable material can be fabricated within the required temperature range, a large portion of the thermal energy can be stored in a recoverable manner using PCMs. This concept has been widely employed for thermal energy harvesting and storage for solar energy conversion and space heating applications [1,5,9,10,12]. PCMs also offer an additional advantage of self-sustained and automatic operation. Once a system is installed and operational little or no supervision is required to ensure its working. This coupled with the benefits of solid-state operation, long life, negligible carbon footprints and low operating costs makes PCMs a strong contender for various TES and HVAC applications. In HVAC applications, PCMs can be integrated into buildings in three ways: (a) PCMs directly used in form of sheets, wallboards etc. [14,22], (b) PCMs integrated into building structure such as facade cement [23] and (c) PCMs used in separate heat and cold devices [24]. The first two systems are passive systems which need no regulation to release or absorb heat. However the third system requires active components such as fan, pumps and a control system [25]. This proposed work primarily focuses on the first two
M. Rastogi et al. / Energy Conversion and Management 89 (2015) 260–269
systems and compares the various commercially available PCMs to meet the end. The bone of contention lies in selecting the PCMs that marks the strongest candidature in the building applications. Literature is full of excellent reviews articles regarding various grades of pure and commercialized PCMs and their possible applications [6,7,10,21,26]. Reports have also been made proposing various unconventional materials as possible candidates for PCMs applications [19]. However, a direct comparison detailing relative performance of a large number of PCMs for application specific purposes is not yet reported. The primary reason behind the absence of such a study stems from the lack of suitable and complete thermo-physical data for potential candidates. Additionally, it becomes a tedious and cumbersome task to theoretically or experimentally verify the performance of suitable PCMs as the list of potential candidates can run into hundreds or thousands. A similar problem is faced by designers and engineers when selecting optimum material for a particular application. Under such circumstances one is often forced to rely on the availability of experimental data, expert judgment and experience. This can however lead to sub-optimal selection which can adversely affect the vested economic interests of the customers regarding the operation of such installations. One possible solution to such problems is the use of Multiple Criteria Decision Making (MCDM) approach for selecting the best alternative [27]. The suitability of PCMs and their performance is directly dependent on various thermo-physical properties. Since there are multiple criteria associated with each candidate, such a selection problem is classified under MCDM. MCDM is further subdivided into two separate branches of Multiple Objective Decision Making (MODM) [28] and Multiple Attribute Decision Making (MADM) techniques [29]. MODM approach makes use of various Figures of Merit (FOMs) to numerically identify the relative performance of participating candidates. The FOMs are derived using functional relationship between the various properties of participating contenders, such that maximization of each FOM leads to enhanced fulfilment of a desired objective. Examples of such objectives are: maximum strength per unit weight or minimum cost per unit volume. On the other hand, MADM approach makes use of predefined mathematical models to rank the alternatives based directly on the measure of their associated attributes. A functional relationship is not required to be established between the various properties. A MCDM approach can effectively utilize both MODM and MADM techniques to give a true indicator of the relative performance/ranking of the participating alternatives. Through this study the authors have attempted to screen and rank various PCMs for domestic HVAC applications using MCDM approach. Two novel FOMs have been proposed for grading various PCMs based on their heat extraction ability and response time. The conflicting nature of the FOM based performance has been resolved by employing Technique for Order Preference by Similarity to Ideal Solution (TOPSIS), a MADM approach. The materials have also been tested for Pareto Optimality and two separate fronts were generated thereof. All these techniques together have been proposed as an efficient selection tool for initial ranking and screening requirements for various PCM applications. To the best of our knowledge, no similar study has been made till date.
2. Materials and methods 2.1. Materials A large number of materials have been explored and documented as PCM for various applications. Excellent reviews are available in the literature discussing various PCMs and their
261
possible applications [10,21]. The number of documented PCMs, including pure compounds and commercialized products, presently exceed over a thousand. However, for the case study discussed in this report, technologically important materials are listed in Table 1. Indexed in Table 1 are potential PCM materials available for regulating temperature in HVAC applications for human comfort. The important thermo-physical properties considered for this study are phase change temperature (°C), mass density (kg m3), latent heat of fusion (kJ kg1), specific heat capacity at constant pressure (kJ kg1 K1) and thermal conductivity (W m1 K1). Since the study in question aims at selecting the best material for regulating temperature within human comfort limits, phase change temperature becomes the single most important screening criterion. Hence, our initial selection of materials was limited to those exhibiting phase change in the temperature range of 17–25 °C. Density, thermal conductivity and latent heat of fusion are the primary factors that determine the performance of a PCM. The higher the density, easier it is to store a larger amount of material in a small volume. Similarly, higher the latent heat of fusion better will be the thermal stability provided by the use of respective PCM. Additional parameter like specific heat capacity, setting and melting enthalpy helps to determine the performance of the PCM in the sensible heating/cooling zone. Hence these parameters have been selected to help in the evaluation process of the candidate materials. 2.2. Methods 2.2.1. Ashby approach and FOMs A popular MODM tool widely used by the scientific community for various screening and selection problems is the Ashby approach. This technique was first proposed by Ashby [30]. The underlying principle dictates that the performance (P) of any engineered system can be determined as a function of its functional (F), geometric (G) and materials (M) parameters. This statement can be mathematically represented as:
P ¼ f ðF; G; MÞ
ð1Þ
Here, f denotes ‘function of’. However, each of the aforementioned parameters operates independently of the rest and their collective output determines the overall performance. Hence, Eq. (1) can be rewritten as:
P ¼ f ðFÞ f ðGÞ f ðMÞ
ð2Þ
Since, the aim of this study is to comparatively rank the PCM for generalized operation; we will only concern ourselves with the materials parameters. The first step towards implementation of the Ashby approach is to determine the screening parameters. In our case, this has been limited to identification of suitable PCMs which are able to operate in the temperature range of 17–25 °C. A list of such potential candidates is listed in Table 1. The second step involves determination of suitable FOMs. Since the primary objective of PCMs is to store maximum amount of thermal energy in a minimum amount of space, the first FOM can be derived as:
Q ¼ mL ¼qv L
ð3Þ
Here, Q represents the total heat extracted during the phase change process. While the symbols m, q, v, and L denote mass, density, volume and latent heat respectively. Eq. (3) can be modified to obtain the first FOM by isolating the materials parameters to the right hand side of the equation. Thus, Eq. (3) can be rewritten as:
FOM1 ¼
Q
v
¼qL
ð4Þ
262
M. Rastogi et al. / Energy Conversion and Management 89 (2015) 260–269
Table 1 List of selected Phase Change Material and their thermo-physical properties. Compound
Phase change temp. (C)
Density (kg/m3)
Heat of fusion (kJ/kg)
Specific heat capacity (kJ/ kg-K)
Thermal conductivity (W/m-K)
FOM1 (heat extracted per unit volume) ⁄ 106 q ⁄ L
FOM2 (response time) ⁄ 106 k/q ⁄ Cp
Rank
First front
Second front
Refs.
KF4H2O RUBITHERM GmbH, Phase Change Materials, SP 24 E RUBITHERM GmbH, Phase Change Materials, SP 26 E RUBITHERM GmbH, Phase Change Materials, SP 25 E2 Micronal PCM 26 Gibsbauplatte, BASF RUBITHERM GmbH, Phase Change Materials, SP 21 E PlusICE PCM, (Hydrated salts) PCM SOLUTIONS, S25 LebastLehmplatte PCM 23(generisch), LehmOrange PlusICE PCM, (Hydrated salts) PCM SOLUTIONS, S23 savENRG PCM-HS24P PlusICE PCM, (Hydrated salts) PCM SOLUTIONS, S21 PUR–PCM-Coating 40%, BASF Polyurethanes GmbH Micronal PCM 23 Gibsbauplatte, BASF PLUsICE PCM, (Hydrated salts) PCM SOLUTIONS, S17 PlusICE PCM, (Hydrated salts) PCM SOLUTIONS, S19 CoolZONE23, Armstrong savENRG PCM-HS22P ThermalCORE 23 °C/ 73 °F, National Gypsum, USA Weber.murclima 23, St. Gobain-weber RUBITHERM GmbH, Phase Change Materials, RT 25HC RUBITHERM GmbH, Phase Change Materials, PX 27 PlusICE PCM, (organic) PCM SOLUTIONS Range, A25H RUBITHERM GmbH, Phase Change Materials, RT 22HC RUBITHERM GmbH, Phase Change Materials, RT 21HC RUBITHERM GmbH, Phase Change Materials, RT 27 PlusICE PCM, (organic) PCM SOLUTIONS Range, A22H RUBITHERM GmbH, Phase Change Materials, RT 21 RUBITHERM GmbH,
18.5 24–35
1455 1500
246 190
1.62 2
0.5 0.6
357.93 285
0.2121 0.2
1 2
1 0
n.a 1
[50] [51]
25–27
1500
190
2
0.6
285
0.2
3
0
0
[51]
24–26
1500
180
2
0.6
270
0.2
4
0
0
[51]
25–27
770
385
1.6
0.20
296.45
0.15909
5
0
1
[47]
21–23
1500
160
2
0.6
240
0.2
6
0
0
[51]
25
1530
180
2.2
0.54
275.4
0.16042
7
0
0
[52]
22–23
1300
170
1.6
0.45
221
0.21634
8
1
n.a
[47]
23
1530
175
2.2
0.54
267.75
0.16042
9
0
0
[52]
24 21
1820 1530
185 170
2.26 2.2
0.5–1.09 0.54
336.7 260.1
0.12161 0.16042
10 11
0 0
1 0
[53] [52]
22–24
970
365
2
0.19
354.05
0.09793
12
0
1
[47]
22–24
770
343
1.8
0.20
264.11
0.141414
13
0
0
[47]
17
1525
160
1.9
0.43
244
0.14840
14
0
0
[52]
19
1520
160
1.9
0.43
243.2
0.14889
15
0
0
[52]
21–22 23 22–24
770 1540 770
342 185 342
2 3.05 2.2
0.2 0.5–1.09 0.2
263.34 284.9 263.34
0.12987013 0.10631 0.11806
16 17 18
0 0 0
0 0 0
[47] [47]
22–24
950
170
2.32
0.38
161.5
0.17241
19
0
0
[47]
22–26
880
230
2
0.2
202.4
0.11363
20
0
0
[51]
25–28
650
102
1.6
0.2
66.3
0.19230
21
0
0
[51]
25
810
226
2.15
0.18
183.06
0.10335
22
0
0
[52]
20–23
760
200
2
0.2
152
0.1315
23
0
0
[51]
20–23
880
190
2
0.2
167.2
0.1136
24
0
0
[51]
25–28
880
179
2
0.2
157.52
0.113636
25
0
0
[51]
22
820
216
2.85
0.18
177.12
0.07702
26
0
0
[52]
18–23
880
150
2
0.2
132
0.113636
27
0
0
[51]
21–25
880
150
2
0.2
132
0.113636
28
0
0
[51]
263
M. Rastogi et al. / Energy Conversion and Management 89 (2015) 260–269 Table 1 (continued) Compound
Phase Change Materials, RT 24 RUBITHERM GmbH, Phase Change Materials, RT 25 PlusICE PCM, (organic) PCM SOLUTIONS Range, A25 PlusICE PCM, (organic) PCM SOLUTIONS Range, A22 PlusICE PCM, (organic) PCM SOLUTIONS Range, A23 PlusICE PCM, (organic) PCM SOLUTIONS Range, A24 PCM-Akustikputz 23(RAL-Werte), SchreffGmbh& Co. KG RUBITHERM GmbH, Phase Change Materials, PX 25
Phase change temp. (C)
Density (kg/m3)
Heat of fusion (kJ/kg)
Specific heat capacity (kJ/ kg-K)
Thermal conductivity (W/m-K)
FOM1 (heat extracted per unit volume) ⁄ 106 q ⁄ L
FOM2 (response time) ⁄ 106 k/q ⁄ Cp
Rank
First front
Second front
Refs.
22–26
880
148
2
0.2
130.24
0.113636
29
0
0
[51]
25
785
150
2.26
0.18
117.75
0.101459
30
0
0
[52]
22
785
145
2.22
0.18
113.82
0.10328
31
0
0
[52]
23
785
145
2.22
0.18
113.82
0.10328
32
0
0
[52]
24
790
145
2.22
0.18
114.55
0.10263428
33
0
0
[52]
21–22
400
196
1.7
0.08
78.4
0.11764
34
0
0
[47]
22–25
650
96
1.6
0.1
62.4
0.09615
35
0
0
[51]
Eq. (4) represents the amount of heat extracted per unit volume. This gives a direct measurement of the amount of thermal energy storage density of any PCMs per unit volume. Hence, it can be employed as a ready reference for comparative analysis. Additionally, another important factor that determines the performance of PCMs with respect to space heating/cooling applications is its thermal inertia. A lower thermal inertia implies a better response time and lower temperature fluctuations for the required application. Thermal diffusivity of any material is generally regarded as a direct measurement of its thermal inertia. Physically, thermal diffusivity represents the ratio of a body’s ability to conduct heat to its ability to store it. Mathematically, it is represented as:
FOM2 ¼ a ¼ k=ðq cp Þ
ð5Þ
Here, a denotes the thermal diffusivity of the material. The symbols k and Cp represent thermal conductivity and specific heat capacity of the material respectively. Mathematically, thermal diffusivity (a) is the ratio of the time derivative of temperature to its curvature [31], or @T ¼ ar2 T. Where, left hand side of the expression denotes @t partial derivative of the temperature with respect to time. It represents the rate at which the temperature concavity is smoothed out in a bulk material system. Thus, for a material possessing a higher value of thermal diffusivity, the rate of energy transfer within the system would be more. In terms of PCM it implies that a larger mass would be involved for creating the required thermal balance, instead of localised phase transition. This would ensure a better temperature regulation and faster response for systems employing macroencapsulation or other form of bulk morphological installations. These two FOMs have been used for the ranking purposes. However, the nature of the two FOMs is conflicting, in the sense that maximization of one will lead to minimization of another. Thus, two additional steps of Pareto Optimality test and TOPSIS have been implemented to obtain the final ranking. 2.2.2. Pareto optimal solution This approach was first proposed by famous neo-classical Italian engineer and economist Vilfredo Pareto [32,33]. According to
the theory developed by him, under the constraints of conflicting objectives, the solution space is inhabited by primarily three types of outcomes. The first category belongs to the dominated solution space which consists of weak or inferior solutions. The second category is that of extreme performers, which consists of solutions which maximize one objective but remain uncompensated in other respects. The third is the non-dominated solution which consists of optimized values of all the conflicting attributes. This is known as Pareto optimal solution and is represented by a physical boundary known as the Pareto front. We have derived the Pareto front for our two conflicting FOMs. The results were obtained by running the materials data (FOM) through a computer code to look for nondominated solutions in the solution space. However, for practical purposes the process can be represented as first selecting one attribute at a time. It is then required to find the highest value available from the solution space of that attribute. Now the corresponding values of other attributes are multiplied to get a product. The process is then repeated for the second highest value. If the second product is more than the first, the first one is eliminated and the process is repeated until only non-dominant solutions are left. This set of remaining solutions is now referred to as the Pareto optimal solution. For our study, the first front was limited to two materials; therefore, a second front was calculated by eliminating the first two from the solution space. A logical representation of the Pareto Optimality test for both the fronts is given in Table 1. 2.2.3. Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) TOPSIS is a popular and efficient MADM technique used for solving many ranking and selection problems. The initial form of TOPSIS was first drafted by Yoon and Hwang [29]. However, over the years TOPSIS has been altered and modified a number of times to suit specific selection requirements [34–42]. Our own previous works have demonstrated the effectiveness of the technique and its good general agreement with practical observations [35,38– 40,42]. TOPSIS implies that a solution space can be interpreted as ‘n’ dimensional hyper plane where the number of dimensions is determined by the relevant number of attributes considered for
264
M. Rastogi et al. / Energy Conversion and Management 89 (2015) 260–269
the study. The position of the competing alternatives can then be represented on this plane by the value of their attributes. Two ideal solutions are then determined by considering the best and the worst possible outcomes respectively from the solution space. The best solution is then determined by considering the largest available ratio of the distance of the alternative from negative ideal solution to the cumulative distance from both the solutions. This is known as TOPSIS rank index and the alternatives are then ranked in the decreasing order of this index. Since the Pareto Optimality test does not give a numerical index of comparison, TOPSIS has been employed to rank the PCMs employing the two FOMs as attributes. Shannon’s entropy was used to calculated the relative influence of the attributes on the decision making process [43]. The relative entropy measurement of attribute data has a direct bearing on the outcome of any decision making process [43–45]. Attributes which have a higher distribution of data between the two extremes of the solution space are said to possess a relatively higher entropy measure. The increased randomness of the data directly implies that the attribute concerned will have a higher impact on the decision making algorithm. Thus, entropy measures can be used to predict the relative importance or weight of the attributes for any MADM process. We have previously used Shannon’s entropy to determine the weights of the attributes in many material selection problems [32,35]. For the present study, it was observed that measured entropy, and subsequently obtained weights, did not differ by a large amount (<5%). Hence, equal (unity) weights were assigned to the attributes for the ranking procedure. The TOPSIS index and subsequently obtained ranks are given in Table 1 for easy reference. 2.2.4. Simulation In the proposed work, the top ranked PCMs were selected for a simulation study using open source software PCMs Express (developed by German company Valentin Energy Software, in collaboration with the Fraunhofer Institute for Solar Energy (ISE) in Freiburg and partners from industry, a planning and simulation program for buildings that are integrated with PCMs). PCM express is based on a graphical user interface (GUI) based approach. The workflow is basically performed using a series of tab sheets and hence each tab sheet must have the required input parameters so that simulation can be executed. The nodal parameters and boundary conditions are defined accordingly and then the simulation is started. After defining the parameters, one can store them so as to reload them, for further applications [46]. It employs finite difference mathematical model and enthalpy method for simulation. It is important to note that melting and setting enthalpy, density and thermal conductivity plays the crucial role in determining the performance of PCMs. With the aim of maintaining the human comfort temperature (21–26 °C), the simulation were executed for a span of a year (8670 h). A conventional system of brick masonry walls, concrete cement roof and ceiling was used in conjunction with a thermoactive system (installed with PCMs wallboards in three walls) for simultaneous comparison. The thickness of each wallboard is considered to be 15 mm. The other boundary conditions being the night ventilation, window in outer wall (occupying 40% area of the total wall) and equivalent radiator output of 50 W/m2. Besides both the system have average volume of 125 m3 and were normally packed (i.e. the furniture and the number of people were optimum and comparable to the volume of room). The details of the construction materials and the respective properties of the simulated conventional domain have been listed in Table 3. The simulation was additionally used to compare the performance of the installed PCMs system with a conventional one. The physical illustration of the node–edge model is represented by Fig. 1. The
results of the simulation have been used to justify the ranking obtained by using the MCDM approach and have been discussed in the following section. Though the commercialized PCMs are tailor made and ready to use, however utter attention has to be paid to the installation situation of the PCMs in the passive building envelope construction. In order to increase the efficiency of PCMs envelopes, the wall that is exposed to outside air, must have PCMs installed towards the outer side(i.e. exposed to the outer temperature) while the rest two interior walls can have PCMs installed towards inner side (i.e. exposed to the room temperature). 3. Results and discussion The performance grade of various PCMs according to the proposed Figures of Merit, the outcome of Pareto Optimality test and the final rankings received using TOPSIS approach are listed in Table 1. Even though KF4H2O is a pure compound and has limited commercial applicability, it has been used as a standard reference to relatively compare the performance of commercial products. In order to evaluate the relative importance of individual steps involved, it becomes crucial to look the results drawn from each step independently. Beginning from FOM1, the top five materials having the highest heat extraction density are: KF4H2O; PURPCM-Coating 40%, BASF Polyurethanes GmbH; savENRG PCMHS24P; Micronal PCM 26 Gibsbauplatte BASF and RUBITHERM GmbH Phase Change Materials SP 24 E. These materials have the largest heat storage capacity per unit volume and thus will provide for a better temperature regulation. Contrastingly though, when the performance is considered with regards to FOM2, the best materials obtained are: LebastLehmplatte PCM 23 (generisch), LehmOrange; KF.4H2O; RUBITHERM GmbH, Phase Change Materials, SP 24 E; RUBITHERM GmbH, Phase Change Materials, SP 26 E and RUBITHERM GmbH, Phase Change Materials, SP 25 E2. Since FOM2 signifies thermal inertia, these materials will react to phase change in a swift manner and thus, the thermal fluctuations will be minimized by their use. This knowledge implies that the two FOMs derived for materials selection are conflicting in nature, that is, one can increase at the cost of other. For a commercial system, a best trade-off between conflicting objectives is required to ensure enhanced productivity and desirable output. Hence it becomes necessary to determine suitable materials with optimum value of FOMs. To this effect, the selected materials were subjected to a Pareto Optimality test using the two FOMs as selection criteria. Two separate logical fronts were generated and the same have been represented in Table 1. It was observed that the first front yielded the duo of KF4H2O and LebastLehmplatte PCM 23 (generisch), LehmOrange as the best materials. While the second front gives RUBITHERM GmbH, Phase Change Materials, SP 24 E; Micronal PCM 26 Gibsbauplatte, BASF; savENRG PCM-HS24P and PURPCM-Coating 40%, BASF Polyurethanes GmbH as the best PCM. However, Pareto front is only limited to providing with a qualitative evaluation as it is not possible to obtain a numerical output required for quantitative analysis. Thus, TOPSIS approach has been utilized for producing a final ranking of the candidate materials using the two FOMs as attributes. Shannon’s entropy was used for calculating the relative weights to be used for the ranking procedure. The value of entropy obtained for FOM1 and FOM2 differed by less than 5% (E1 = 0.9764, E2 = 0.9740, where E1 and E2 are entropy for FOM1 and FOM2 respectively). Hence, equal (unity) weights were assigned to each category. Fig. 2 gives a graphical representation of the TOPSIS solution space along with the position of first and second Pareto fronts as obtained for our calculations. The ranks thus obtained have been reproduced in Table 1 for ready reference. The final ranking indicates that the following materials would be best suited for a balanced performance in any generalized application: KF4H2O; RUBITHERM GmbH, Phase Change Materials,
265
M. Rastogi et al. / Energy Conversion and Management 89 (2015) 260–269 Table 2 Comparison of the conventional and thermo-active systems. S. no
Number of hours within comfort temperature zone in conventional system (per 8760 h)
PCMs
1 2 3 4 5
5002 5002 5002 5002 5002
RUBITHERM GmbH, Phase Change Materials, RUBITHERM GmbH, Phase Change Materials, RUBITHERM GmbH, Phase Change Materials, Micronal PCM 26 Gibsbauplatte, BASF RUBITHERM GmbH, Phase Change Materials,
SP 24 E SP 26 E SP 25 E2 SP 21 E
Ranking
Number of hours within comfort temperature zone in thermo-active system (per 8760 h)
Percentage improvement by use of PCM (%)
2 3 4 5 6
6614 6491 6342 6272 6088
18.40 17.00 15.30 14.50 12.40
Table 3 Details of the construction materials and properties of the simulated conventional room. S. no
Wall/ceiling/floor
Construction materials
Thickness (mm)
Thermal conductivity (W/mK)
1.
External wall solid brick
2
Ceiling/floor reinforced concrete
3.
Internal brick solid masonry
Gypsum interior plaster Full brick masonry Gypsum interior plaster Concrete W/C Gypsum interior plaster Full brick masonry
15 300 15 250 15 115
0.2 0.15 0.2 1.6 0.2 0.6
Fig. 1. Node–edge model for the simulated building.
SP 24 E; RUBITHERM GmbH, Phase Change Materials, SP 26 E; RUBITHERM GmbH, Phase Change Materials, SP 25 E2 and Micronal PCM 26 Gibsbauplatte, BASF. These materials possess the best numerical trade-off between their concerned thermo-physical properties and thus are expected to give good performance for temperature regulation within the human comfort zone. In order to further justify the obtained rankings, the top PCM candidates were simulated for standard environmental conditions and their performance was evaluated. During simulation the results are drawn with the help of commercial tool PCM Express [47] with a finite element mathematical model and enthalpy method, which makes it possible to account all the enthalpies of the PCMs during the simulation process. Due to the presence of two phases (solid and liquid) simultaneously and moving
phase change interface, it becomes difficult to determine the amount of heat absorbed or released at the interface boundary. Thus, to obtain an exact solution of phase change problems is difficult. Hence it is required that the energy equations should be written separately for both the phases. Additionally, their respective temperatures should also be accounted for, at the interface region. This requires determination of the exact location of the interface region, which is difficult to achieve by means of finite difference methods. Enthalpy method is an approach that overcomes these shortcomings by utilizing enthalpy form of the energy equation, which is applicable for both the phases. Thus, the need to obtain separate equations for both phases is eliminated. The enthalpy form of the energy equation is given in the following expression [48,49].
266
M. Rastogi et al. / Energy Conversion and Management 89 (2015) 260–269
Fig. 2. A two dimensional graphical representation of the TOPSIS solution space. The Phase Change Materials are indicated by their respective TOPSIS ranks. The figure also contains information about the position of positive and negative ideal solutions and the first and second Pareto fronts indicated by circle and diamond shapes respectively.
rðk rTÞ ¼ qð@HðTÞ=@t Þ
ð6Þ
where r represents del operator; therefore, rT represents the temperature gradient (vector). ‘H’ signifies enthalpy and ‘t’ represents time. The expression ð@HðTÞ=@t Þ therefore represents the enthalpy change with respect to time. In building simulation, heat transfer through building elements (walls, roofs, etc.) is generally solved by employing one-dimensional analysis. This can be further illustrated by exemplifying the physical illustration of the phenomenon in Fig. 3. Considering the solidification of a liquid, this is at an initial uniform temperature of T0 (higher than melting temperature) and having boundaries 0 6 x 6 B. If the boundary temperature at x = 0 for time t > 0 is kept at temperature T1, lower than the melting temperature, and assuming no temperature gradient at the boundary x = B. Then Eq. (6) and the related boundary conditions can be represented as:
q
@H @2T ¼ k 2 ; in 0 < x < B and t > 0; @t @x
ð7Þ
Eq. (7) in PCM EXPRESS is approximated with implicit finite difference method as:
qcp Dx
nþ1 nþ1 nþ1 ðT nþ1 T ni Þ kðT nþ1 Þ kðT iþ1 T i Þ i i1 T i ¼ þ Dt Dx Dx
ð8Þ
Here Cp is the specific heat capacity, Dt is time step and Dx = B/ M where M is the number of parts that the region 0 6 x 6 B is divided into; ‘i’ denotes spatial discretization and ‘n’ is used for discretization of time steps. Eq. (8) can now be solved by the algorithm upon utilization of the data which is retrieved from the library, in the form of enthalpy temperature values. Thus, enthalpy–temperature function is obtained. The enthalpy variation with respect to temperature is
Fig. 3. Physical illustration of enthalpy form of the energy equation and boundary conditions.
determined experimentally (such as by a 3 layer calorimeter) in terms of setting and melting enthalpy, as a function of applied temperature. These enthalpies are used as input data by the user to model the PCM within the software. The sudden formation of peak determines the specific temperature range in which the particular PCM would undergo phase change. It also helps in selecting the operational temperature range in which a PCM candidate would function efficiently. Fig. 4(a) further illustrates enthalpy variation of the room per hour for Micronal PCM 26 Gibsbauplatte, BASF, for duration of 24 h. Fig. 4(b) gives the enthalpy as a function of temperature, for Micronal PCM 26 Gibsbauplatte, BASF. The peak formation at 23.5 °C and at 22.5 °C while determining melting and setting enthalpy respectively helps in determining the working temperature range for the chosen PCM candidate. Similar graphs can be obtained for other PCMs understudy, which are not shown here. Eq. (8) and enthalpy–temperature function is then generated for each node of the modelled PCM. The node temperatures are updated after each iteration along with node enthalpies and are used to develop a variable Cp for PCM. This is done by using the value of Cp given by Eq. (9). By this algorithm the correct enthalpy is used for each time step thus the correct Cp for nth iteration is obtained as:
Cp ¼
Hnþ1 Hni i T nþ1 T ni i
ð9Þ
Using the aforementioned methodology, the control volume was subjected to atmospheric conditions prevalent in Berlin (Germany), and the time span of the simulation was extended to one year. PCM is used in the constructions of the floor, ceiling and the walls. T0 describe the impact of PCM the phase change and the regeneration has to be calculated accurately. Therefore all layers of the constructions are simulated as a node in the mathematical node–edge-model. Non-PCM layers are defined with a constant capacity. PCM are defined by the temperature dependent enthalpy to describe the phase change. Each room is represented by one node for the inside air. The heat flow between the nodes results from the thermal conductivity and is represented by an edge connecting two layers. The outside layer is connected over an edge to the environment (heat exchange and irradiation). Irradiation through the windows is distributed to the air nodes and the inner layer of the constructions. The air node is connected to the technical building services as heating, cooling and ventilation. Moreover each layer can be defined as an active layer, e.g. a plaster layer including plastic mats.
M. Rastogi et al. / Energy Conversion and Management 89 (2015) 260–269
267
Fig. 4. (a) Enthalpy variation of the room plotted with respect to time for a period of 24 h, for Micronal PCM 26 Gibsbauplatte, BASF; (b) the experimentally determined temperature–enthalpy curve, for Micronal PCM 26 Gibsbauplatte, BASF.
Active layers are supported with water for heating or cooling purposes. For all services adequate control strategies are implemented [38]. Constant thermal capacity was defined for all non-PCM elements, while PCM are defined by the temperature dependent enthalpy, i.e. setting and melting enthalpy, to describe the phase change. The subset of the results is used to draw the inferences, which were mainly based on comparison of the conventional system and thermo-active system. Fig. 5 exhibits the distribution of room temperature by plotting the relationship between frequency (%) and room air temperature for conventional system against RUBITHERM GmbH, Phase Change Materials, SP 24 E, which has scored second rank in Table 1. Similar figures can be plotted for other PCMs understudy and are not shown here. These figures pave the way for the comparison between conventional system and PCM candidate’s performance. Frequency (%) in the figures represents the recurrence of the particular room air temperature, which is hence a good medium, for calculating the number of hours in the human comfort zone. A higher frequency of the PCM in comfort zone represents a better temperature regulation. The second ranked candidate, RUBITHERM GmbH, Phase Change Materials, SP 24 E, has a frequency of 75.5% within the human comfort zone which indicates that, 75.5% of the total time, the room air temperature (of the object under consideration) would vary between 21 °C and 26 °C, while the conventional system has total frequency of 57.1%. The frequency % in the conventional system can be improved up to 18.4% using the suitable high ranked PCMs. Fig. 6 displays the comparison between RUBITHERM GmbH, Phase Change Materials SP 24 E; RUBITHERM GmbH, Phase Change Materials, SP 26 E; RUBITHERM GmbH, Phase Change Materials, SP 25
E2; Micronal PCM 26 Gibsbauplatte, BASF and RUBITHERM GmbH, Phase Change Materials, SP 21 E against the day which were having the best PCM effect. The day with greatest PCM effect means the day (out of 365 days under consideration) when the PCM performed the best so as to regulate the room air temperature. A significant temperature control, up to 7 °C can be achieved. Fig. 6 highlights and compares the relative performance of the various PCMs understudy, but it holds no good when it comes to decide the overall ranking of the PCMs. Hence, selection and ranking tools like TOPSIS and Ashby approach are required to verify and compliment the results obtained from numerical simulation studies. Tangible observations, as drawn from simulation, are presented in Table 2. Both the commercial tool and analytical tools were at par in determining the performance of the PCM candidates. RUBITHERM GmbH, Phase Change Materials, SP 24E demonstrated a significant improvement over conventional system (18.40%) in maintaining comfort zone The simulation results justify the ranks obtained with the MCDM technique. The results obtained using simulation are in good agreement with those obtained by using MCDM approach. It is noteworthy to mention that MCDM approach is computationally far less taxing than simulation or analytical techniques. Hence, it can be safely concluded that incorporation of MCDM approach as a necessary material selection step during any design or engineering procedure provides an economical advantage. Additionally, this approach allows for swift and easy comparison of various competing materials for any application specific requirements. Thus, it is proposed to be a handy screening and selection tool and warrants wide spread application into the field of PCM based systems.
Fig. 5. Distribution of room temperature for conventional system against RUBITHERM GmbH, Phase Change Materials, SP 24 E.
268
M. Rastogi et al. / Energy Conversion and Management 89 (2015) 260–269
Fig. 6. The comparative analysis of top most PCMs against conventional system, on the basis of day with greatest PCM effect.
4. Conclusion Through the proposed work, the authors have attempted to screen and rank various commercial Phase Change Materials for heating, ventilation and air-conditioning application. A Multiple Criteria Decision Making approach was used for this purpose. Suitable materials were first shortlisted based on the phase change temperature (within the range of 17–25 °C). This was followed by selection of two suitable figure of merits to assess the relative performance of different PCMs based on their heat extraction capacity and thermal inertia. The materials were finally subjected to Pareto Optimality test and TOPSIS was employed to produce the final ranking required. In order to validate the rankings obtained using the TOPSIS approach, the top ranked materials were subjected to finite element based analysis. The commercial tool PCM Express establishes a relationship between the frequency of human comfort temperature range and the room air temperature, which in turn also exhibits the effectiveness of the PCM candidate. During the simulation, the crucial thermo-physical properties affected the performance of PCMs thereby justifying the selection of parameters used in analytical tools, so as to find out the rank of PCMs. It was also concluded that the efficiency of the PCMs can be improved by wisely considering their installing situation. Thus, it can be concluded that the MCDM approach can be utilized for effective screening and ranking of PCM for various industrial and domestic applications. Acknowledgements One of the authors (Rahul Vaish) acknowledges support from the Indian National Science Academy (INSA), New Delhi, India, through a Grant by the Department of Science and Technology (DST), New Delhi, under INSPIRE faculty award-2011 (ENG-01). Aditya Chauhan would like to acknowledge the invaluable expert discussion provided by Satyanarayan Patel. References [1] Humphries WR, Griggs EI. A design handbook for phase change thermal control and energy storage devices. NASA STI/recon technical report N, vol. 78; 1977. p. 15434. [2] Xiao M, Feng B, Gong K. Preparation and performance of shape stabilized phase change thermal storage materials with high thermal conductivity. Energy Convers Manage 2002;43:103–8.
[3] Sarı A. Form-stable paraffin/high density polyethylene composites as solid– liquid phase change material for thermal energy storage: preparation and thermal properties. Energy Convers Manage 2004;45:2033–42. [4] Lyeo H-K, Cahill DG, Lee B-S, Abelson JR, Kwon M-H, Kim K-B, et al. Thermal conductivity of phase-change material Ge2Sb2Te5. Appl Phys Lett 2006;89:151904. [5] Sabbah R, Kizilel R, Selman J, Al-Hallaj S. Active (air-cooled) vs. passive (phase change material) thermal management of high power lithium-ion packs: limitation of temperature rise and uniformity of temperature distribution. J Power Sources 2008;182:630–8. [6] Pasupathy A, Velraj R, Seeniraj R. Phase change material-based building architecture for thermal management in residential and commercial establishments. Renew Sustain Energy Rev 2008;12:39–64. [7] Shukla A, Buddhi D, Sawhney R. Solar water heaters with phase change material thermal energy storage medium: a review. Renew Sustain Energy Rev 2009;13:2119–25. [8] Alkan C, Sarı A, Karaipekli A, Uzun O. Preparation, characterization, and thermal properties of microencapsulated phase change material for thermal energy storage. Sol Energy Mater Sol Cells 2009;93:143–7. [9] Colclough S, Griffiths P, Gschwander S. Thermal energy storage and the passive house standard, 2009. [10] Zalba B, Marı´n JM, Cabeza LF, Mehling H. Review on thermal energy storage with phase change: materials, heat transfer analysis and applications. Appl Therm Eng 2003;23:251–83. [11] Silakhori M, Metselaar HSC, Mahlia TMI, Fauzi H, Baradaran S, Naghavi MS. Palmitic acid/polypyrrole composites as form-stable phase change materials for thermal energy storage. Energy Convers Manage 2014;80:491–7. [12] Turnpenny J, Etheridge D, Reay D. Novel ventilation cooling system for reducing air conditioning in buildings. Part I: testing and theoretical modelling. Appl Therm Eng 2000;20:1019–37. [13] Parameshwaran R, Harikrishnan S, Kalaiselvam S. Energy efficient PCM-based variable air volume air conditioning system for modern buildings. Energy Build 2010;42:1353–60. [14] Sun X, Zhang Q, Medina MA, Lee KO. Energy and economic analysis of a building enclosure outfitted with a phase change material board (PCMB). Energy Convers Manage 2014;83:73–8. [15] Cheralathan M, Velraj R, Renganarayanan S. Performance analysis on industrial refrigeration system integrated with encapsulated PCM-based cool thermal energy storage system. Int J Energy Res 2007;31:1398–413. [16] Chaabane M, Mhiri H, Bournot P. Thermal performance of an integrated collector storage solar water heater (ICSSWH) with phase change materials (PCM). Energy Convers Manage 2014;78:897–903. [17] Jing L, Gang P, Jie J. Optimization of low temperature solar thermal electric generation with Organic Rankine Cycle in different areas. Appl Energy 2010;87:3355–65. [18] Beyhan B, Paksoy H, Dasßgan Y. Root zone temperature control with thermal energy storage in phase change materials for soilless greenhouse applications. Energy Convers Manage 2013;74:446–53. [19] Bo H, Gustafsson EM, Setterwall F. Tetradecane and hexadecane binary mixtures as phase change materials (PCMs) for cool storage in district cooling systems. Energy 1999;24:1015–28. [20] Krupa I, Nógellová Z, Špitalsky´ Z, Janigová I, Boh B, Sumiga B, et al. Phase change materials based on high-density polyethylene filled with microencapsulated paraffin wax. Energy Convers Manage 2014;87:400–9. [21] Farid MM, Khudhair AM, Razack SAK, Al-Hallaj S. A review on phase change energy storage: materials and applications. Energy Convers Manage 2004;45:1597–615.
M. Rastogi et al. / Energy Conversion and Management 89 (2015) 260–269 [22] Zwanzig SD, Lian Y, Brehob EG. Numerical simulation of phase change material composite wallboard in a multi-layered building envelope. Energy Convers Manage 2013;69:27–40. [23] Chan A. Energy and environmental performance of building façades integrated with phase change material in subtropical Hong Kong. Energy Build 2011;43:2947–55. [24] Schrader MD. Bulk heat or cold storage device for thermal energy storage compounds. Google Patents; 1986. [25] http://www.bine.info/fileadmin/content/Publikationen. [26] Rathod MK, Kanzaria HV. A methodological concept for phase change material selection based on multiple criteria decision analysis with and without fuzzy environment. Mater Des 2011;32:3578–85. [27] Zeleny M, Cochrane JL. Multiple criteria decision making. New York: McGrawHill; 1982. [28] Napier TL. Multiple objective decision making. Integrated watershed management: perspectives and problems; 2012. p. 20. [29] Yoon KP, Hwang C-L. Multiple attribute decision making: an introduction. Sage Publications; 1995. [30] Ashby M. Multi-objective optimization in material design and selection. Acta Mater 2000;48:359–69. [31] Venkanna B. Fundamentals of heat and mass transfer. PHI Learning Pvt. Ltd.; 2010. [32] Chauhan A, Vaish R. Pareto optimal microwave dielectric materials. Adv Sci, Eng Med 2013;5:149–55. [33] Zitzler E, Laumanns M, Thiele L. SPEA2: improving the strength Pareto evolutionary algorithm. Eidgenössische Technische Hochschule Zürich (ETH), Institut für Technische Informatik und Kommunikationsnetze (TIK); 2001. [34] Chen T-Y, Tsao C-Y. The interval-valued fuzzy TOPSIS method and experimental analysis. Fuzzy Sets Syst 2008;159:1410–28. [35] Chauhan A, Vaish R. Magnetic material selection using multiple attribute decision making approach. Mater Des 2012;36:1–5. [36] Chauhan A, Vaish R. A comparative study on decision making methods with interval data. J Comput Eng 2014;2014.
269
[37] Chauhan A, Vaish R. Materials selection decision making with incomplete information. Adv Sci Focus 2013;1:340–5. [38] Chauhan A, Vaish R. A comparative study on material selection for microelectromechanical systems. Mater Des 2012;41:177–81. [39] Chauhan A, Vaish R. Hard coating material selection using multi-criteria decision making. Mater Des 2013;44:240–5. [40] Chauhan A, Vaish R. Material selection for piezoelectric devices. Adv Sci, Eng Med 2013;5:715–9. [41] Chauhan A, Vaish R. An assessment of bulk metallic glasses for microelectromechanical system devices. Int J Appl Glass Sci 2013;4:231–41. [42] Chauhan A, Vaish R, Bowen C. Piezoelectric material selection for ultrasonic transducer and actuator applications. Proc Inst Mech Eng, Part L: J Mater Des Appl 2013. 1464420713493591. [43] Lotfi FH, Fallahnejad R. Imprecise Shannon’s entropy and multi attribute decision making. Entropy 2010;12:53–62. [44] Singh VP. The entropy theory as a tool for modelling and decision-making in environmental and water resources. Water Sa-Pretoria 2000;26:1–12. [45] Yeh CH. A problem-based selection of multi-attribute decision-making methods. Int Trans Oper Res 2002;9:169–81. [46] Gatzka B, Valentin G. Pcmexpress – planning and simulation programme for the use of phase change materials (PCM) in buildings: demonstrating its use in residential and office buildings in Ireland. http://www.valentin-software.com/ sites/default/files/icses2011_pcmexpress.pdf. [47] http://www.valentin.de/index_en_page=pcm_express. [48] Cross VVaM. Accurate solutions of moving boundary problems using the enthalpy method. Int J Heat Mass Transfer 1981;24:545–56. [49] Ozisik MN. Heat conduction. 2nd ed. New York: John Wiley & Sons, Inc.; 1993. [50] Sharma A, Tyagi V, Chen C, Buddhi D. Review on thermal energy storage with phase change materials and applications. Renew Sustain Energy Rev 2009;13:318–45. [51] http://www.rubitherm.de/. [52] http://www.pcmproducts.net/. [53] http://www.rgees.com/.
Contents lists available at ScienceDirect
Energy Conversion and Management journal homepage: www.elsevier.com/locate/enconman
Selection and performance assessment of Phase Change Materials for heating, ventilation and air-conditioning applications Monisha Rastogi, Aditya Chauhan, Rahul Vaish ⇑, Anil Kishan School of Engineering, Indian Institute of Technology Mandi, Mandi 175 001, India
a r t i c l e
i n f o
Article history: Received 2 July 2014 Accepted 29 September 2014
Keywords: Phase Change Materials Heating, ventilation and air-conditioning Ashby approach Materials selection
a b s t r a c t The rapid commercialization of Phase Change Materials (PCMs) for heating, ventilation and air-conditioning (HVAC) applications, has paved way for effective utilization of ambient thermal fluctuations. However, given a long list of contemporary candidates, it is crucial to select the best material to obtain maximum efficiency for any given application. This article attempts to extend Multiple Criteria Decision Making (MCDM) approach for ranking and selecting PCMs for domestic HVAC application. Firstly, Ashby approach has been employed for determining two novel figure of merits (FOM) to grade PCMs performance. The FOMs thus obtained were subjected to Pareto Optimality test. The graded materials were ranked using Technique for Order Preference by Similarity to Ideal Solution (TOPSIS). The relative weights for the different attributes were calculated using Shannon’s entropy method. In order to justify the rankings obtained, the top materials were subjected to a standard simulation study to evaluate their relative performance using PCMExpress with the aim of maintaining human comfort temperature. It was observed that the results obtained by simulation are in good agreement with those obtained using MCDM approach. The candidates with the best ranks showed significant improvement in ameliorating the temperature conditions. Thus it can be concluded that integration of MCDM approach for PCMs selection would prove to an economical and swift alternative technique for ranking and screening of materials. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Phase Change Materials (PCMs) have been the subject of rigorous investigation in the past few decades [1–8]. These materials have been explored for thermal energy storage (TES) [9–11], heating, ventilation and air-conditioning (HVAC) [12–14], temperature regulation for industrial application [15,16], electricity generation [17] and even soilless crop growth [18]. Recent years have seen significant improvement in terms of material’s chemistry [11,19], performance and engineering [15,20]. Techniques like microencapsulation [20] and macro-encapsulation [8], have enabled high efficiency versatile operation. This has been followed by rapid commercialization of the technology and is now available for industrial and domestic use at large [21]. These advancements can be credited to the growing awareness towards global climate change and the need to push forward with cleaner and cheaper technology. PCM as the name implies is a category of various organic or inorganic compounds exhibiting a change of phase within the required operating temperature range [1,8]. As the materials ⇑ Corresponding author. Tel.: +91 1905 237921; fax: +91 1905 237945. E-mail address: [email protected] (R. Vaish). http://dx.doi.org/10.1016/j.enconman.2014.09.077 0196-8904/Ó 2014 Elsevier Ltd. All rights reserved.
undergo phase transformation, they absorb or release a large amount of thermal energy. Since the energy flow is associated with a change in the physical state of the material, the entirety of the heat exchange is approximately isothermal in nature. Thus, if a suitable material can be fabricated within the required temperature range, a large portion of the thermal energy can be stored in a recoverable manner using PCMs. This concept has been widely employed for thermal energy harvesting and storage for solar energy conversion and space heating applications [1,5,9,10,12]. PCMs also offer an additional advantage of self-sustained and automatic operation. Once a system is installed and operational little or no supervision is required to ensure its working. This coupled with the benefits of solid-state operation, long life, negligible carbon footprints and low operating costs makes PCMs a strong contender for various TES and HVAC applications. In HVAC applications, PCMs can be integrated into buildings in three ways: (a) PCMs directly used in form of sheets, wallboards etc. [14,22], (b) PCMs integrated into building structure such as facade cement [23] and (c) PCMs used in separate heat and cold devices [24]. The first two systems are passive systems which need no regulation to release or absorb heat. However the third system requires active components such as fan, pumps and a control system [25]. This proposed work primarily focuses on the first two
M. Rastogi et al. / Energy Conversion and Management 89 (2015) 260–269
systems and compares the various commercially available PCMs to meet the end. The bone of contention lies in selecting the PCMs that marks the strongest candidature in the building applications. Literature is full of excellent reviews articles regarding various grades of pure and commercialized PCMs and their possible applications [6,7,10,21,26]. Reports have also been made proposing various unconventional materials as possible candidates for PCMs applications [19]. However, a direct comparison detailing relative performance of a large number of PCMs for application specific purposes is not yet reported. The primary reason behind the absence of such a study stems from the lack of suitable and complete thermo-physical data for potential candidates. Additionally, it becomes a tedious and cumbersome task to theoretically or experimentally verify the performance of suitable PCMs as the list of potential candidates can run into hundreds or thousands. A similar problem is faced by designers and engineers when selecting optimum material for a particular application. Under such circumstances one is often forced to rely on the availability of experimental data, expert judgment and experience. This can however lead to sub-optimal selection which can adversely affect the vested economic interests of the customers regarding the operation of such installations. One possible solution to such problems is the use of Multiple Criteria Decision Making (MCDM) approach for selecting the best alternative [27]. The suitability of PCMs and their performance is directly dependent on various thermo-physical properties. Since there are multiple criteria associated with each candidate, such a selection problem is classified under MCDM. MCDM is further subdivided into two separate branches of Multiple Objective Decision Making (MODM) [28] and Multiple Attribute Decision Making (MADM) techniques [29]. MODM approach makes use of various Figures of Merit (FOMs) to numerically identify the relative performance of participating candidates. The FOMs are derived using functional relationship between the various properties of participating contenders, such that maximization of each FOM leads to enhanced fulfilment of a desired objective. Examples of such objectives are: maximum strength per unit weight or minimum cost per unit volume. On the other hand, MADM approach makes use of predefined mathematical models to rank the alternatives based directly on the measure of their associated attributes. A functional relationship is not required to be established between the various properties. A MCDM approach can effectively utilize both MODM and MADM techniques to give a true indicator of the relative performance/ranking of the participating alternatives. Through this study the authors have attempted to screen and rank various PCMs for domestic HVAC applications using MCDM approach. Two novel FOMs have been proposed for grading various PCMs based on their heat extraction ability and response time. The conflicting nature of the FOM based performance has been resolved by employing Technique for Order Preference by Similarity to Ideal Solution (TOPSIS), a MADM approach. The materials have also been tested for Pareto Optimality and two separate fronts were generated thereof. All these techniques together have been proposed as an efficient selection tool for initial ranking and screening requirements for various PCM applications. To the best of our knowledge, no similar study has been made till date.
2. Materials and methods 2.1. Materials A large number of materials have been explored and documented as PCM for various applications. Excellent reviews are available in the literature discussing various PCMs and their
261
possible applications [10,21]. The number of documented PCMs, including pure compounds and commercialized products, presently exceed over a thousand. However, for the case study discussed in this report, technologically important materials are listed in Table 1. Indexed in Table 1 are potential PCM materials available for regulating temperature in HVAC applications for human comfort. The important thermo-physical properties considered for this study are phase change temperature (°C), mass density (kg m3), latent heat of fusion (kJ kg1), specific heat capacity at constant pressure (kJ kg1 K1) and thermal conductivity (W m1 K1). Since the study in question aims at selecting the best material for regulating temperature within human comfort limits, phase change temperature becomes the single most important screening criterion. Hence, our initial selection of materials was limited to those exhibiting phase change in the temperature range of 17–25 °C. Density, thermal conductivity and latent heat of fusion are the primary factors that determine the performance of a PCM. The higher the density, easier it is to store a larger amount of material in a small volume. Similarly, higher the latent heat of fusion better will be the thermal stability provided by the use of respective PCM. Additional parameter like specific heat capacity, setting and melting enthalpy helps to determine the performance of the PCM in the sensible heating/cooling zone. Hence these parameters have been selected to help in the evaluation process of the candidate materials. 2.2. Methods 2.2.1. Ashby approach and FOMs A popular MODM tool widely used by the scientific community for various screening and selection problems is the Ashby approach. This technique was first proposed by Ashby [30]. The underlying principle dictates that the performance (P) of any engineered system can be determined as a function of its functional (F), geometric (G) and materials (M) parameters. This statement can be mathematically represented as:
P ¼ f ðF; G; MÞ
ð1Þ
Here, f denotes ‘function of’. However, each of the aforementioned parameters operates independently of the rest and their collective output determines the overall performance. Hence, Eq. (1) can be rewritten as:
P ¼ f ðFÞ f ðGÞ f ðMÞ
ð2Þ
Since, the aim of this study is to comparatively rank the PCM for generalized operation; we will only concern ourselves with the materials parameters. The first step towards implementation of the Ashby approach is to determine the screening parameters. In our case, this has been limited to identification of suitable PCMs which are able to operate in the temperature range of 17–25 °C. A list of such potential candidates is listed in Table 1. The second step involves determination of suitable FOMs. Since the primary objective of PCMs is to store maximum amount of thermal energy in a minimum amount of space, the first FOM can be derived as:
Q ¼ mL ¼qv L
ð3Þ
Here, Q represents the total heat extracted during the phase change process. While the symbols m, q, v, and L denote mass, density, volume and latent heat respectively. Eq. (3) can be modified to obtain the first FOM by isolating the materials parameters to the right hand side of the equation. Thus, Eq. (3) can be rewritten as:
FOM1 ¼
Q
v
¼qL
ð4Þ
262
M. Rastogi et al. / Energy Conversion and Management 89 (2015) 260–269
Table 1 List of selected Phase Change Material and their thermo-physical properties. Compound
Phase change temp. (C)
Density (kg/m3)
Heat of fusion (kJ/kg)
Specific heat capacity (kJ/ kg-K)
Thermal conductivity (W/m-K)
FOM1 (heat extracted per unit volume) ⁄ 106 q ⁄ L
FOM2 (response time) ⁄ 106 k/q ⁄ Cp
Rank
First front
Second front
Refs.
KF4H2O RUBITHERM GmbH, Phase Change Materials, SP 24 E RUBITHERM GmbH, Phase Change Materials, SP 26 E RUBITHERM GmbH, Phase Change Materials, SP 25 E2 Micronal PCM 26 Gibsbauplatte, BASF RUBITHERM GmbH, Phase Change Materials, SP 21 E PlusICE PCM, (Hydrated salts) PCM SOLUTIONS, S25 LebastLehmplatte PCM 23(generisch), LehmOrange PlusICE PCM, (Hydrated salts) PCM SOLUTIONS, S23 savENRG PCM-HS24P PlusICE PCM, (Hydrated salts) PCM SOLUTIONS, S21 PUR–PCM-Coating 40%, BASF Polyurethanes GmbH Micronal PCM 23 Gibsbauplatte, BASF PLUsICE PCM, (Hydrated salts) PCM SOLUTIONS, S17 PlusICE PCM, (Hydrated salts) PCM SOLUTIONS, S19 CoolZONE23, Armstrong savENRG PCM-HS22P ThermalCORE 23 °C/ 73 °F, National Gypsum, USA Weber.murclima 23, St. Gobain-weber RUBITHERM GmbH, Phase Change Materials, RT 25HC RUBITHERM GmbH, Phase Change Materials, PX 27 PlusICE PCM, (organic) PCM SOLUTIONS Range, A25H RUBITHERM GmbH, Phase Change Materials, RT 22HC RUBITHERM GmbH, Phase Change Materials, RT 21HC RUBITHERM GmbH, Phase Change Materials, RT 27 PlusICE PCM, (organic) PCM SOLUTIONS Range, A22H RUBITHERM GmbH, Phase Change Materials, RT 21 RUBITHERM GmbH,
18.5 24–35
1455 1500
246 190
1.62 2
0.5 0.6
357.93 285
0.2121 0.2
1 2
1 0
n.a 1
[50] [51]
25–27
1500
190
2
0.6
285
0.2
3
0
0
[51]
24–26
1500
180
2
0.6
270
0.2
4
0
0
[51]
25–27
770
385
1.6
0.20
296.45
0.15909
5
0
1
[47]
21–23
1500
160
2
0.6
240
0.2
6
0
0
[51]
25
1530
180
2.2
0.54
275.4
0.16042
7
0
0
[52]
22–23
1300
170
1.6
0.45
221
0.21634
8
1
n.a
[47]
23
1530
175
2.2
0.54
267.75
0.16042
9
0
0
[52]
24 21
1820 1530
185 170
2.26 2.2
0.5–1.09 0.54
336.7 260.1
0.12161 0.16042
10 11
0 0
1 0
[53] [52]
22–24
970
365
2
0.19
354.05
0.09793
12
0
1
[47]
22–24
770
343
1.8
0.20
264.11
0.141414
13
0
0
[47]
17
1525
160
1.9
0.43
244
0.14840
14
0
0
[52]
19
1520
160
1.9
0.43
243.2
0.14889
15
0
0
[52]
21–22 23 22–24
770 1540 770
342 185 342
2 3.05 2.2
0.2 0.5–1.09 0.2
263.34 284.9 263.34
0.12987013 0.10631 0.11806
16 17 18
0 0 0
0 0 0
[47] [47]
22–24
950
170
2.32
0.38
161.5
0.17241
19
0
0
[47]
22–26
880
230
2
0.2
202.4
0.11363
20
0
0
[51]
25–28
650
102
1.6
0.2
66.3
0.19230
21
0
0
[51]
25
810
226
2.15
0.18
183.06
0.10335
22
0
0
[52]
20–23
760
200
2
0.2
152
0.1315
23
0
0
[51]
20–23
880
190
2
0.2
167.2
0.1136
24
0
0
[51]
25–28
880
179
2
0.2
157.52
0.113636
25
0
0
[51]
22
820
216
2.85
0.18
177.12
0.07702
26
0
0
[52]
18–23
880
150
2
0.2
132
0.113636
27
0
0
[51]
21–25
880
150
2
0.2
132
0.113636
28
0
0
[51]
263
M. Rastogi et al. / Energy Conversion and Management 89 (2015) 260–269 Table 1 (continued) Compound
Phase Change Materials, RT 24 RUBITHERM GmbH, Phase Change Materials, RT 25 PlusICE PCM, (organic) PCM SOLUTIONS Range, A25 PlusICE PCM, (organic) PCM SOLUTIONS Range, A22 PlusICE PCM, (organic) PCM SOLUTIONS Range, A23 PlusICE PCM, (organic) PCM SOLUTIONS Range, A24 PCM-Akustikputz 23(RAL-Werte), SchreffGmbh& Co. KG RUBITHERM GmbH, Phase Change Materials, PX 25
Phase change temp. (C)
Density (kg/m3)
Heat of fusion (kJ/kg)
Specific heat capacity (kJ/ kg-K)
Thermal conductivity (W/m-K)
FOM1 (heat extracted per unit volume) ⁄ 106 q ⁄ L
FOM2 (response time) ⁄ 106 k/q ⁄ Cp
Rank
First front
Second front
Refs.
22–26
880
148
2
0.2
130.24
0.113636
29
0
0
[51]
25
785
150
2.26
0.18
117.75
0.101459
30
0
0
[52]
22
785
145
2.22
0.18
113.82
0.10328
31
0
0
[52]
23
785
145
2.22
0.18
113.82
0.10328
32
0
0
[52]
24
790
145
2.22
0.18
114.55
0.10263428
33
0
0
[52]
21–22
400
196
1.7
0.08
78.4
0.11764
34
0
0
[47]
22–25
650
96
1.6
0.1
62.4
0.09615
35
0
0
[51]
Eq. (4) represents the amount of heat extracted per unit volume. This gives a direct measurement of the amount of thermal energy storage density of any PCMs per unit volume. Hence, it can be employed as a ready reference for comparative analysis. Additionally, another important factor that determines the performance of PCMs with respect to space heating/cooling applications is its thermal inertia. A lower thermal inertia implies a better response time and lower temperature fluctuations for the required application. Thermal diffusivity of any material is generally regarded as a direct measurement of its thermal inertia. Physically, thermal diffusivity represents the ratio of a body’s ability to conduct heat to its ability to store it. Mathematically, it is represented as:
FOM2 ¼ a ¼ k=ðq cp Þ
ð5Þ
Here, a denotes the thermal diffusivity of the material. The symbols k and Cp represent thermal conductivity and specific heat capacity of the material respectively. Mathematically, thermal diffusivity (a) is the ratio of the time derivative of temperature to its curvature [31], or @T ¼ ar2 T. Where, left hand side of the expression denotes @t partial derivative of the temperature with respect to time. It represents the rate at which the temperature concavity is smoothed out in a bulk material system. Thus, for a material possessing a higher value of thermal diffusivity, the rate of energy transfer within the system would be more. In terms of PCM it implies that a larger mass would be involved for creating the required thermal balance, instead of localised phase transition. This would ensure a better temperature regulation and faster response for systems employing macroencapsulation or other form of bulk morphological installations. These two FOMs have been used for the ranking purposes. However, the nature of the two FOMs is conflicting, in the sense that maximization of one will lead to minimization of another. Thus, two additional steps of Pareto Optimality test and TOPSIS have been implemented to obtain the final ranking. 2.2.2. Pareto optimal solution This approach was first proposed by famous neo-classical Italian engineer and economist Vilfredo Pareto [32,33]. According to
the theory developed by him, under the constraints of conflicting objectives, the solution space is inhabited by primarily three types of outcomes. The first category belongs to the dominated solution space which consists of weak or inferior solutions. The second category is that of extreme performers, which consists of solutions which maximize one objective but remain uncompensated in other respects. The third is the non-dominated solution which consists of optimized values of all the conflicting attributes. This is known as Pareto optimal solution and is represented by a physical boundary known as the Pareto front. We have derived the Pareto front for our two conflicting FOMs. The results were obtained by running the materials data (FOM) through a computer code to look for nondominated solutions in the solution space. However, for practical purposes the process can be represented as first selecting one attribute at a time. It is then required to find the highest value available from the solution space of that attribute. Now the corresponding values of other attributes are multiplied to get a product. The process is then repeated for the second highest value. If the second product is more than the first, the first one is eliminated and the process is repeated until only non-dominant solutions are left. This set of remaining solutions is now referred to as the Pareto optimal solution. For our study, the first front was limited to two materials; therefore, a second front was calculated by eliminating the first two from the solution space. A logical representation of the Pareto Optimality test for both the fronts is given in Table 1. 2.2.3. Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) TOPSIS is a popular and efficient MADM technique used for solving many ranking and selection problems. The initial form of TOPSIS was first drafted by Yoon and Hwang [29]. However, over the years TOPSIS has been altered and modified a number of times to suit specific selection requirements [34–42]. Our own previous works have demonstrated the effectiveness of the technique and its good general agreement with practical observations [35,38– 40,42]. TOPSIS implies that a solution space can be interpreted as ‘n’ dimensional hyper plane where the number of dimensions is determined by the relevant number of attributes considered for
264
M. Rastogi et al. / Energy Conversion and Management 89 (2015) 260–269
the study. The position of the competing alternatives can then be represented on this plane by the value of their attributes. Two ideal solutions are then determined by considering the best and the worst possible outcomes respectively from the solution space. The best solution is then determined by considering the largest available ratio of the distance of the alternative from negative ideal solution to the cumulative distance from both the solutions. This is known as TOPSIS rank index and the alternatives are then ranked in the decreasing order of this index. Since the Pareto Optimality test does not give a numerical index of comparison, TOPSIS has been employed to rank the PCMs employing the two FOMs as attributes. Shannon’s entropy was used to calculated the relative influence of the attributes on the decision making process [43]. The relative entropy measurement of attribute data has a direct bearing on the outcome of any decision making process [43–45]. Attributes which have a higher distribution of data between the two extremes of the solution space are said to possess a relatively higher entropy measure. The increased randomness of the data directly implies that the attribute concerned will have a higher impact on the decision making algorithm. Thus, entropy measures can be used to predict the relative importance or weight of the attributes for any MADM process. We have previously used Shannon’s entropy to determine the weights of the attributes in many material selection problems [32,35]. For the present study, it was observed that measured entropy, and subsequently obtained weights, did not differ by a large amount (<5%). Hence, equal (unity) weights were assigned to the attributes for the ranking procedure. The TOPSIS index and subsequently obtained ranks are given in Table 1 for easy reference. 2.2.4. Simulation In the proposed work, the top ranked PCMs were selected for a simulation study using open source software PCMs Express (developed by German company Valentin Energy Software, in collaboration with the Fraunhofer Institute for Solar Energy (ISE) in Freiburg and partners from industry, a planning and simulation program for buildings that are integrated with PCMs). PCM express is based on a graphical user interface (GUI) based approach. The workflow is basically performed using a series of tab sheets and hence each tab sheet must have the required input parameters so that simulation can be executed. The nodal parameters and boundary conditions are defined accordingly and then the simulation is started. After defining the parameters, one can store them so as to reload them, for further applications [46]. It employs finite difference mathematical model and enthalpy method for simulation. It is important to note that melting and setting enthalpy, density and thermal conductivity plays the crucial role in determining the performance of PCMs. With the aim of maintaining the human comfort temperature (21–26 °C), the simulation were executed for a span of a year (8670 h). A conventional system of brick masonry walls, concrete cement roof and ceiling was used in conjunction with a thermoactive system (installed with PCMs wallboards in three walls) for simultaneous comparison. The thickness of each wallboard is considered to be 15 mm. The other boundary conditions being the night ventilation, window in outer wall (occupying 40% area of the total wall) and equivalent radiator output of 50 W/m2. Besides both the system have average volume of 125 m3 and were normally packed (i.e. the furniture and the number of people were optimum and comparable to the volume of room). The details of the construction materials and the respective properties of the simulated conventional domain have been listed in Table 3. The simulation was additionally used to compare the performance of the installed PCMs system with a conventional one. The physical illustration of the node–edge model is represented by Fig. 1. The
results of the simulation have been used to justify the ranking obtained by using the MCDM approach and have been discussed in the following section. Though the commercialized PCMs are tailor made and ready to use, however utter attention has to be paid to the installation situation of the PCMs in the passive building envelope construction. In order to increase the efficiency of PCMs envelopes, the wall that is exposed to outside air, must have PCMs installed towards the outer side(i.e. exposed to the outer temperature) while the rest two interior walls can have PCMs installed towards inner side (i.e. exposed to the room temperature). 3. Results and discussion The performance grade of various PCMs according to the proposed Figures of Merit, the outcome of Pareto Optimality test and the final rankings received using TOPSIS approach are listed in Table 1. Even though KF4H2O is a pure compound and has limited commercial applicability, it has been used as a standard reference to relatively compare the performance of commercial products. In order to evaluate the relative importance of individual steps involved, it becomes crucial to look the results drawn from each step independently. Beginning from FOM1, the top five materials having the highest heat extraction density are: KF4H2O; PURPCM-Coating 40%, BASF Polyurethanes GmbH; savENRG PCMHS24P; Micronal PCM 26 Gibsbauplatte BASF and RUBITHERM GmbH Phase Change Materials SP 24 E. These materials have the largest heat storage capacity per unit volume and thus will provide for a better temperature regulation. Contrastingly though, when the performance is considered with regards to FOM2, the best materials obtained are: LebastLehmplatte PCM 23 (generisch), LehmOrange; KF.4H2O; RUBITHERM GmbH, Phase Change Materials, SP 24 E; RUBITHERM GmbH, Phase Change Materials, SP 26 E and RUBITHERM GmbH, Phase Change Materials, SP 25 E2. Since FOM2 signifies thermal inertia, these materials will react to phase change in a swift manner and thus, the thermal fluctuations will be minimized by their use. This knowledge implies that the two FOMs derived for materials selection are conflicting in nature, that is, one can increase at the cost of other. For a commercial system, a best trade-off between conflicting objectives is required to ensure enhanced productivity and desirable output. Hence it becomes necessary to determine suitable materials with optimum value of FOMs. To this effect, the selected materials were subjected to a Pareto Optimality test using the two FOMs as selection criteria. Two separate logical fronts were generated and the same have been represented in Table 1. It was observed that the first front yielded the duo of KF4H2O and LebastLehmplatte PCM 23 (generisch), LehmOrange as the best materials. While the second front gives RUBITHERM GmbH, Phase Change Materials, SP 24 E; Micronal PCM 26 Gibsbauplatte, BASF; savENRG PCM-HS24P and PURPCM-Coating 40%, BASF Polyurethanes GmbH as the best PCM. However, Pareto front is only limited to providing with a qualitative evaluation as it is not possible to obtain a numerical output required for quantitative analysis. Thus, TOPSIS approach has been utilized for producing a final ranking of the candidate materials using the two FOMs as attributes. Shannon’s entropy was used for calculating the relative weights to be used for the ranking procedure. The value of entropy obtained for FOM1 and FOM2 differed by less than 5% (E1 = 0.9764, E2 = 0.9740, where E1 and E2 are entropy for FOM1 and FOM2 respectively). Hence, equal (unity) weights were assigned to each category. Fig. 2 gives a graphical representation of the TOPSIS solution space along with the position of first and second Pareto fronts as obtained for our calculations. The ranks thus obtained have been reproduced in Table 1 for ready reference. The final ranking indicates that the following materials would be best suited for a balanced performance in any generalized application: KF4H2O; RUBITHERM GmbH, Phase Change Materials,
265
M. Rastogi et al. / Energy Conversion and Management 89 (2015) 260–269 Table 2 Comparison of the conventional and thermo-active systems. S. no
Number of hours within comfort temperature zone in conventional system (per 8760 h)
PCMs
1 2 3 4 5
5002 5002 5002 5002 5002
RUBITHERM GmbH, Phase Change Materials, RUBITHERM GmbH, Phase Change Materials, RUBITHERM GmbH, Phase Change Materials, Micronal PCM 26 Gibsbauplatte, BASF RUBITHERM GmbH, Phase Change Materials,
SP 24 E SP 26 E SP 25 E2 SP 21 E
Ranking
Number of hours within comfort temperature zone in thermo-active system (per 8760 h)
Percentage improvement by use of PCM (%)
2 3 4 5 6
6614 6491 6342 6272 6088
18.40 17.00 15.30 14.50 12.40
Table 3 Details of the construction materials and properties of the simulated conventional room. S. no
Wall/ceiling/floor
Construction materials
Thickness (mm)
Thermal conductivity (W/mK)
1.
External wall solid brick
2
Ceiling/floor reinforced concrete
3.
Internal brick solid masonry
Gypsum interior plaster Full brick masonry Gypsum interior plaster Concrete W/C Gypsum interior plaster Full brick masonry
15 300 15 250 15 115
0.2 0.15 0.2 1.6 0.2 0.6
Fig. 1. Node–edge model for the simulated building.
SP 24 E; RUBITHERM GmbH, Phase Change Materials, SP 26 E; RUBITHERM GmbH, Phase Change Materials, SP 25 E2 and Micronal PCM 26 Gibsbauplatte, BASF. These materials possess the best numerical trade-off between their concerned thermo-physical properties and thus are expected to give good performance for temperature regulation within the human comfort zone. In order to further justify the obtained rankings, the top PCM candidates were simulated for standard environmental conditions and their performance was evaluated. During simulation the results are drawn with the help of commercial tool PCM Express [47] with a finite element mathematical model and enthalpy method, which makes it possible to account all the enthalpies of the PCMs during the simulation process. Due to the presence of two phases (solid and liquid) simultaneously and moving
phase change interface, it becomes difficult to determine the amount of heat absorbed or released at the interface boundary. Thus, to obtain an exact solution of phase change problems is difficult. Hence it is required that the energy equations should be written separately for both the phases. Additionally, their respective temperatures should also be accounted for, at the interface region. This requires determination of the exact location of the interface region, which is difficult to achieve by means of finite difference methods. Enthalpy method is an approach that overcomes these shortcomings by utilizing enthalpy form of the energy equation, which is applicable for both the phases. Thus, the need to obtain separate equations for both phases is eliminated. The enthalpy form of the energy equation is given in the following expression [48,49].
266
M. Rastogi et al. / Energy Conversion and Management 89 (2015) 260–269
Fig. 2. A two dimensional graphical representation of the TOPSIS solution space. The Phase Change Materials are indicated by their respective TOPSIS ranks. The figure also contains information about the position of positive and negative ideal solutions and the first and second Pareto fronts indicated by circle and diamond shapes respectively.
rðk rTÞ ¼ qð@HðTÞ=@t Þ
ð6Þ
where r represents del operator; therefore, rT represents the temperature gradient (vector). ‘H’ signifies enthalpy and ‘t’ represents time. The expression ð@HðTÞ=@t Þ therefore represents the enthalpy change with respect to time. In building simulation, heat transfer through building elements (walls, roofs, etc.) is generally solved by employing one-dimensional analysis. This can be further illustrated by exemplifying the physical illustration of the phenomenon in Fig. 3. Considering the solidification of a liquid, this is at an initial uniform temperature of T0 (higher than melting temperature) and having boundaries 0 6 x 6 B. If the boundary temperature at x = 0 for time t > 0 is kept at temperature T1, lower than the melting temperature, and assuming no temperature gradient at the boundary x = B. Then Eq. (6) and the related boundary conditions can be represented as:
q
@H @2T ¼ k 2 ; in 0 < x < B and t > 0; @t @x
ð7Þ
Eq. (7) in PCM EXPRESS is approximated with implicit finite difference method as:
qcp Dx
nþ1 nþ1 nþ1 ðT nþ1 T ni Þ kðT nþ1 Þ kðT iþ1 T i Þ i i1 T i ¼ þ Dt Dx Dx
ð8Þ
Here Cp is the specific heat capacity, Dt is time step and Dx = B/ M where M is the number of parts that the region 0 6 x 6 B is divided into; ‘i’ denotes spatial discretization and ‘n’ is used for discretization of time steps. Eq. (8) can now be solved by the algorithm upon utilization of the data which is retrieved from the library, in the form of enthalpy temperature values. Thus, enthalpy–temperature function is obtained. The enthalpy variation with respect to temperature is
Fig. 3. Physical illustration of enthalpy form of the energy equation and boundary conditions.
determined experimentally (such as by a 3 layer calorimeter) in terms of setting and melting enthalpy, as a function of applied temperature. These enthalpies are used as input data by the user to model the PCM within the software. The sudden formation of peak determines the specific temperature range in which the particular PCM would undergo phase change. It also helps in selecting the operational temperature range in which a PCM candidate would function efficiently. Fig. 4(a) further illustrates enthalpy variation of the room per hour for Micronal PCM 26 Gibsbauplatte, BASF, for duration of 24 h. Fig. 4(b) gives the enthalpy as a function of temperature, for Micronal PCM 26 Gibsbauplatte, BASF. The peak formation at 23.5 °C and at 22.5 °C while determining melting and setting enthalpy respectively helps in determining the working temperature range for the chosen PCM candidate. Similar graphs can be obtained for other PCMs understudy, which are not shown here. Eq. (8) and enthalpy–temperature function is then generated for each node of the modelled PCM. The node temperatures are updated after each iteration along with node enthalpies and are used to develop a variable Cp for PCM. This is done by using the value of Cp given by Eq. (9). By this algorithm the correct enthalpy is used for each time step thus the correct Cp for nth iteration is obtained as:
Cp ¼
Hnþ1 Hni i T nþ1 T ni i
ð9Þ
Using the aforementioned methodology, the control volume was subjected to atmospheric conditions prevalent in Berlin (Germany), and the time span of the simulation was extended to one year. PCM is used in the constructions of the floor, ceiling and the walls. T0 describe the impact of PCM the phase change and the regeneration has to be calculated accurately. Therefore all layers of the constructions are simulated as a node in the mathematical node–edge-model. Non-PCM layers are defined with a constant capacity. PCM are defined by the temperature dependent enthalpy to describe the phase change. Each room is represented by one node for the inside air. The heat flow between the nodes results from the thermal conductivity and is represented by an edge connecting two layers. The outside layer is connected over an edge to the environment (heat exchange and irradiation). Irradiation through the windows is distributed to the air nodes and the inner layer of the constructions. The air node is connected to the technical building services as heating, cooling and ventilation. Moreover each layer can be defined as an active layer, e.g. a plaster layer including plastic mats.
M. Rastogi et al. / Energy Conversion and Management 89 (2015) 260–269
267
Fig. 4. (a) Enthalpy variation of the room plotted with respect to time for a period of 24 h, for Micronal PCM 26 Gibsbauplatte, BASF; (b) the experimentally determined temperature–enthalpy curve, for Micronal PCM 26 Gibsbauplatte, BASF.
Active layers are supported with water for heating or cooling purposes. For all services adequate control strategies are implemented [38]. Constant thermal capacity was defined for all non-PCM elements, while PCM are defined by the temperature dependent enthalpy, i.e. setting and melting enthalpy, to describe the phase change. The subset of the results is used to draw the inferences, which were mainly based on comparison of the conventional system and thermo-active system. Fig. 5 exhibits the distribution of room temperature by plotting the relationship between frequency (%) and room air temperature for conventional system against RUBITHERM GmbH, Phase Change Materials, SP 24 E, which has scored second rank in Table 1. Similar figures can be plotted for other PCMs understudy and are not shown here. These figures pave the way for the comparison between conventional system and PCM candidate’s performance. Frequency (%) in the figures represents the recurrence of the particular room air temperature, which is hence a good medium, for calculating the number of hours in the human comfort zone. A higher frequency of the PCM in comfort zone represents a better temperature regulation. The second ranked candidate, RUBITHERM GmbH, Phase Change Materials, SP 24 E, has a frequency of 75.5% within the human comfort zone which indicates that, 75.5% of the total time, the room air temperature (of the object under consideration) would vary between 21 °C and 26 °C, while the conventional system has total frequency of 57.1%. The frequency % in the conventional system can be improved up to 18.4% using the suitable high ranked PCMs. Fig. 6 displays the comparison between RUBITHERM GmbH, Phase Change Materials SP 24 E; RUBITHERM GmbH, Phase Change Materials, SP 26 E; RUBITHERM GmbH, Phase Change Materials, SP 25
E2; Micronal PCM 26 Gibsbauplatte, BASF and RUBITHERM GmbH, Phase Change Materials, SP 21 E against the day which were having the best PCM effect. The day with greatest PCM effect means the day (out of 365 days under consideration) when the PCM performed the best so as to regulate the room air temperature. A significant temperature control, up to 7 °C can be achieved. Fig. 6 highlights and compares the relative performance of the various PCMs understudy, but it holds no good when it comes to decide the overall ranking of the PCMs. Hence, selection and ranking tools like TOPSIS and Ashby approach are required to verify and compliment the results obtained from numerical simulation studies. Tangible observations, as drawn from simulation, are presented in Table 2. Both the commercial tool and analytical tools were at par in determining the performance of the PCM candidates. RUBITHERM GmbH, Phase Change Materials, SP 24E demonstrated a significant improvement over conventional system (18.40%) in maintaining comfort zone The simulation results justify the ranks obtained with the MCDM technique. The results obtained using simulation are in good agreement with those obtained by using MCDM approach. It is noteworthy to mention that MCDM approach is computationally far less taxing than simulation or analytical techniques. Hence, it can be safely concluded that incorporation of MCDM approach as a necessary material selection step during any design or engineering procedure provides an economical advantage. Additionally, this approach allows for swift and easy comparison of various competing materials for any application specific requirements. Thus, it is proposed to be a handy screening and selection tool and warrants wide spread application into the field of PCM based systems.
Fig. 5. Distribution of room temperature for conventional system against RUBITHERM GmbH, Phase Change Materials, SP 24 E.
268
M. Rastogi et al. / Energy Conversion and Management 89 (2015) 260–269
Fig. 6. The comparative analysis of top most PCMs against conventional system, on the basis of day with greatest PCM effect.
4. Conclusion Through the proposed work, the authors have attempted to screen and rank various commercial Phase Change Materials for heating, ventilation and air-conditioning application. A Multiple Criteria Decision Making approach was used for this purpose. Suitable materials were first shortlisted based on the phase change temperature (within the range of 17–25 °C). This was followed by selection of two suitable figure of merits to assess the relative performance of different PCMs based on their heat extraction capacity and thermal inertia. The materials were finally subjected to Pareto Optimality test and TOPSIS was employed to produce the final ranking required. In order to validate the rankings obtained using the TOPSIS approach, the top ranked materials were subjected to finite element based analysis. The commercial tool PCM Express establishes a relationship between the frequency of human comfort temperature range and the room air temperature, which in turn also exhibits the effectiveness of the PCM candidate. During the simulation, the crucial thermo-physical properties affected the performance of PCMs thereby justifying the selection of parameters used in analytical tools, so as to find out the rank of PCMs. It was also concluded that the efficiency of the PCMs can be improved by wisely considering their installing situation. Thus, it can be concluded that the MCDM approach can be utilized for effective screening and ranking of PCM for various industrial and domestic applications. Acknowledgements One of the authors (Rahul Vaish) acknowledges support from the Indian National Science Academy (INSA), New Delhi, India, through a Grant by the Department of Science and Technology (DST), New Delhi, under INSPIRE faculty award-2011 (ENG-01). Aditya Chauhan would like to acknowledge the invaluable expert discussion provided by Satyanarayan Patel. References [1] Humphries WR, Griggs EI. A design handbook for phase change thermal control and energy storage devices. NASA STI/recon technical report N, vol. 78; 1977. p. 15434. [2] Xiao M, Feng B, Gong K. Preparation and performance of shape stabilized phase change thermal storage materials with high thermal conductivity. Energy Convers Manage 2002;43:103–8.
[3] Sarı A. Form-stable paraffin/high density polyethylene composites as solid– liquid phase change material for thermal energy storage: preparation and thermal properties. Energy Convers Manage 2004;45:2033–42. [4] Lyeo H-K, Cahill DG, Lee B-S, Abelson JR, Kwon M-H, Kim K-B, et al. Thermal conductivity of phase-change material Ge2Sb2Te5. Appl Phys Lett 2006;89:151904. [5] Sabbah R, Kizilel R, Selman J, Al-Hallaj S. Active (air-cooled) vs. passive (phase change material) thermal management of high power lithium-ion packs: limitation of temperature rise and uniformity of temperature distribution. J Power Sources 2008;182:630–8. [6] Pasupathy A, Velraj R, Seeniraj R. Phase change material-based building architecture for thermal management in residential and commercial establishments. Renew Sustain Energy Rev 2008;12:39–64. [7] Shukla A, Buddhi D, Sawhney R. Solar water heaters with phase change material thermal energy storage medium: a review. Renew Sustain Energy Rev 2009;13:2119–25. [8] Alkan C, Sarı A, Karaipekli A, Uzun O. Preparation, characterization, and thermal properties of microencapsulated phase change material for thermal energy storage. Sol Energy Mater Sol Cells 2009;93:143–7. [9] Colclough S, Griffiths P, Gschwander S. Thermal energy storage and the passive house standard, 2009. [10] Zalba B, Marı´n JM, Cabeza LF, Mehling H. Review on thermal energy storage with phase change: materials, heat transfer analysis and applications. Appl Therm Eng 2003;23:251–83. [11] Silakhori M, Metselaar HSC, Mahlia TMI, Fauzi H, Baradaran S, Naghavi MS. Palmitic acid/polypyrrole composites as form-stable phase change materials for thermal energy storage. Energy Convers Manage 2014;80:491–7. [12] Turnpenny J, Etheridge D, Reay D. Novel ventilation cooling system for reducing air conditioning in buildings. Part I: testing and theoretical modelling. Appl Therm Eng 2000;20:1019–37. [13] Parameshwaran R, Harikrishnan S, Kalaiselvam S. Energy efficient PCM-based variable air volume air conditioning system for modern buildings. Energy Build 2010;42:1353–60. [14] Sun X, Zhang Q, Medina MA, Lee KO. Energy and economic analysis of a building enclosure outfitted with a phase change material board (PCMB). Energy Convers Manage 2014;83:73–8. [15] Cheralathan M, Velraj R, Renganarayanan S. Performance analysis on industrial refrigeration system integrated with encapsulated PCM-based cool thermal energy storage system. Int J Energy Res 2007;31:1398–413. [16] Chaabane M, Mhiri H, Bournot P. Thermal performance of an integrated collector storage solar water heater (ICSSWH) with phase change materials (PCM). Energy Convers Manage 2014;78:897–903. [17] Jing L, Gang P, Jie J. Optimization of low temperature solar thermal electric generation with Organic Rankine Cycle in different areas. Appl Energy 2010;87:3355–65. [18] Beyhan B, Paksoy H, Dasßgan Y. Root zone temperature control with thermal energy storage in phase change materials for soilless greenhouse applications. Energy Convers Manage 2013;74:446–53. [19] Bo H, Gustafsson EM, Setterwall F. Tetradecane and hexadecane binary mixtures as phase change materials (PCMs) for cool storage in district cooling systems. Energy 1999;24:1015–28. [20] Krupa I, Nógellová Z, Špitalsky´ Z, Janigová I, Boh B, Sumiga B, et al. Phase change materials based on high-density polyethylene filled with microencapsulated paraffin wax. Energy Convers Manage 2014;87:400–9. [21] Farid MM, Khudhair AM, Razack SAK, Al-Hallaj S. A review on phase change energy storage: materials and applications. Energy Convers Manage 2004;45:1597–615.
M. Rastogi et al. / Energy Conversion and Management 89 (2015) 260–269 [22] Zwanzig SD, Lian Y, Brehob EG. Numerical simulation of phase change material composite wallboard in a multi-layered building envelope. Energy Convers Manage 2013;69:27–40. [23] Chan A. Energy and environmental performance of building façades integrated with phase change material in subtropical Hong Kong. Energy Build 2011;43:2947–55. [24] Schrader MD. Bulk heat or cold storage device for thermal energy storage compounds. Google Patents; 1986. [25] http://www.bine.info/fileadmin/content/Publikationen. [26] Rathod MK, Kanzaria HV. A methodological concept for phase change material selection based on multiple criteria decision analysis with and without fuzzy environment. Mater Des 2011;32:3578–85. [27] Zeleny M, Cochrane JL. Multiple criteria decision making. New York: McGrawHill; 1982. [28] Napier TL. Multiple objective decision making. Integrated watershed management: perspectives and problems; 2012. p. 20. [29] Yoon KP, Hwang C-L. Multiple attribute decision making: an introduction. Sage Publications; 1995. [30] Ashby M. Multi-objective optimization in material design and selection. Acta Mater 2000;48:359–69. [31] Venkanna B. Fundamentals of heat and mass transfer. PHI Learning Pvt. Ltd.; 2010. [32] Chauhan A, Vaish R. Pareto optimal microwave dielectric materials. Adv Sci, Eng Med 2013;5:149–55. [33] Zitzler E, Laumanns M, Thiele L. SPEA2: improving the strength Pareto evolutionary algorithm. Eidgenössische Technische Hochschule Zürich (ETH), Institut für Technische Informatik und Kommunikationsnetze (TIK); 2001. [34] Chen T-Y, Tsao C-Y. The interval-valued fuzzy TOPSIS method and experimental analysis. Fuzzy Sets Syst 2008;159:1410–28. [35] Chauhan A, Vaish R. Magnetic material selection using multiple attribute decision making approach. Mater Des 2012;36:1–5. [36] Chauhan A, Vaish R. A comparative study on decision making methods with interval data. J Comput Eng 2014;2014.
269
[37] Chauhan A, Vaish R. Materials selection decision making with incomplete information. Adv Sci Focus 2013;1:340–5. [38] Chauhan A, Vaish R. A comparative study on material selection for microelectromechanical systems. Mater Des 2012;41:177–81. [39] Chauhan A, Vaish R. Hard coating material selection using multi-criteria decision making. Mater Des 2013;44:240–5. [40] Chauhan A, Vaish R. Material selection for piezoelectric devices. Adv Sci, Eng Med 2013;5:715–9. [41] Chauhan A, Vaish R. An assessment of bulk metallic glasses for microelectromechanical system devices. Int J Appl Glass Sci 2013;4:231–41. [42] Chauhan A, Vaish R, Bowen C. Piezoelectric material selection for ultrasonic transducer and actuator applications. Proc Inst Mech Eng, Part L: J Mater Des Appl 2013. 1464420713493591. [43] Lotfi FH, Fallahnejad R. Imprecise Shannon’s entropy and multi attribute decision making. Entropy 2010;12:53–62. [44] Singh VP. The entropy theory as a tool for modelling and decision-making in environmental and water resources. Water Sa-Pretoria 2000;26:1–12. [45] Yeh CH. A problem-based selection of multi-attribute decision-making methods. Int Trans Oper Res 2002;9:169–81. [46] Gatzka B, Valentin G. Pcmexpress – planning and simulation programme for the use of phase change materials (PCM) in buildings: demonstrating its use in residential and office buildings in Ireland. http://www.valentin-software.com/ sites/default/files/icses2011_pcmexpress.pdf. [47] http://www.valentin.de/index_en_page=pcm_express. [48] Cross VVaM. Accurate solutions of moving boundary problems using the enthalpy method. Int J Heat Mass Transfer 1981;24:545–56. [49] Ozisik MN. Heat conduction. 2nd ed. New York: John Wiley & Sons, Inc.; 1993. [50] Sharma A, Tyagi V, Chen C, Buddhi D. Review on thermal energy storage with phase change materials and applications. Renew Sustain Energy Rev 2009;13:318–45. [51] http://www.rubitherm.de/. [52] http://www.pcmproducts.net/. [53] http://www.rgees.com/.