Accepted Manuscript

Enhancing a vertical earth-to-air heat exchanger system using tubular phase

change material

Zhengxuan Liu, Pengcheng Sun, Shuisheng Li, Zhun (Jerry) Yu, Mohamed El
Mankibi, Letizia Roccamena, Tingting Yang, Guoqiang Zhang

PII: S0959-6526(19)32623-X

DOI: 10.1016/j.jclepro.2019.117763

Article Number: 117763

Reference: JCLP 117763

To appear in: Journal of Cleaner Production

Received Date: 22 May 2019

Accepted Date: 23 July 2019

Please cite this article as: Zhengxuan Liu, Pengcheng Sun, Shuisheng Li, Zhun (Jerry) Yu,
Mohamed El Mankibi, Letizia Roccamena, Tingting Yang, Guoqiang Zhang, Enhancing a vertical
earth-to-air heat exchanger system using tubular phase change material, Journal of Cleaner
Production (2019), doi: 10.1016/j.jclepro.2019.117763

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form.
Please note that during the production process errors may be discovered which could affect the
content, and all legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

1 Enhancing a vertical earth-to-air heat exchanger

2 system using tubular phase change material
3
4 Zhengxuan Liua,b, Pengcheng Sunc, Shuisheng Lid, Zhun (Jerry) Yua,b*, Mohamed El Mankibie, Letizia
5 Roccamenae, Tingting Yangf, Guoqiang Zhanga,b
6
7 aCollege of Civil Engineering, National Center for International Research Collaboration in Building
8 Safety and Environment, Hunan University, Changsha, Hunan 410082, China
9 bCollaborative Innovation Center of Building Energy Conservation & Environmental Control, Hunan
10 412007, China
11 cChina State Construction Engineering Corporation Technical Center, Beijing 100029, China
12 dChina Construction Fifth Engineering Division Corporation Limited, Changsha, Hunan 410004, China
13 eENTPE-University of Lyon, 3 rue Maurice Audin, Vaulx en Velin 69120, France
14 fCollege of Civil Engineering and Mechanics, Xiangtan University, Xiangtan, Hunan 411105, China

15

16 *Corresponding author. Tel.: +86 88821040

17 E-mail address: zhunyu@hnu.edu.cn;

18

19 Highlights:
20  A vertical EAHE system coupled with tubular PCM components was proposed.
21  Different experiments for the proposed system were conducted.
22  A numerical model of the proposed system was developed and verified.
23  Influences of the tubular PCM component and different parameters were
24 analyzed.
25  The proposed system’s static payback period was calculated.

26

27 Abstract:
28 In this study, a new vertical earth-to-air heat exchanger (VEAHE) system
29 coupled with tubular phase change material (PCM) components was proposed. The
30 proposed system has smaller occupied areas and higher geothermal energy use

1
 ACCEPTED MANUSCRIPT

1 efficiency than traditional EAHE systems. Moreover, the tubular PCM component
2 enables the system’s air temperature at the outlet to be more stable, thereby enhancing
3 the system’s thermal performance. In particular, the tubular components’ simple
4 geometry, easy fabrication, convenient assembling-disassembling and low-cost
5 facilitate their usage in practical applications. To explore the system’s thermal
6 performance, an experimental set-up was established and its numerical model was
7 developed. Then the model was verified through a comparison between the system’s
8 simulated and monitored air temperatures at the outlet. The verification produced
9 acceptable results with the maximum absolute relative error of 1.33%. Based on this
10 model, the influences of the tubular PCM component, tube depth, PCM conductivity
11 and container length on the proposed system’s thermal performance were investigated.
12 Results indicated that the tubular PCM components can effectively reduce the
13 VEAHE system’s air temperature peak and fluctuation at the outlet, and increase its
14 average cooling capacity. For the proposed system, as the tube depths increase from
15 12 to 24 m, the air temperature peak and fluctuation at the outlet decrease from
16 25.74ºC and 3.59ºC to 21.01ºC and 0.62ºC, respectively. Different thermal
17 conductivity of PCM has almost same influences on the system’s air temperature at
18 the outlet as the container diameter is 50 mm. Such influences, however, become
19 apparent and require to be considered as the container diameter increases to 150 mm.
20 The air temperature fluctuation at the outlet decreases as the container lengths
21 increase, and 12 m can be considered as an appropriate length for the proposed system.
22 Additionally, the system's static payback period was calculated as 18.03 years.
23

24 Keywords: Vertical earth-to-air heat exchanger; Geothermal energy; Phase change

25 material; Tubular component; Static payback period

26

27

28

2
 ACCEPTED MANUSCRIPT

Glossary
Symbols Subscripts
𝐶𝑝 Specific heat capacity, J/(kg K) 𝑖 Node location
T Temperature, ºC a Air
𝑅 Thermal resistance, ºC/W s Soil
𝑚 Mass flow rate, kg/s t Tube
𝑇 Average temperature, ºC I Insulation
𝑟1 Fixed radius, m 𝑔 Outermost soil
𝑟𝑃 Outermost radius of PCM, m 𝑗 Vertical direction node location
𝑆 Heat transfer area, m2 P Phase change material
ℎ Heat transfer coefficient, W/(m2·ºC) k Number of selected data samples
𝑉 Volume, m3 m Number of soil column
𝑟𝑖 Inner radius of tube, m
𝑑 Tube wall thickness, m Superscripts
∆𝑧 Height of divided volume, m t Time
𝐶𝑡 Total cost, $ 𝑡-∆𝑡 Previous time of the time t
𝐶𝑎 Cost of a conventional air-conditioner, $
𝑆𝑦 Annual income produced of energy-saving, $ Abbreviations
VEAHE Vertical earth-to-air heat
exchanger
Greek letters PCM Phase change material
𝜌 Density, m3/kg EAHE Earth-to-air heat exchanger
𝜑 Heating or cooling rate, ºC/s GSHP Ground source heat pump
𝜆 Thermal conductivity, W/(m·K) SHC Specific heat capacity
SPP Static payback period

2 1. Introduction
3 Geothermal energy, as a cleaner and nearly emission-free alternative to fossil
4 fuels (Balbay, 2010; Esen et al., 2017), has been widely used to provide
5 cooling/heating for various energy-saving systems applied in buildings (Esen and
6 Yuksel, 2013; Yu et al., 2019; Zhang et al., 2019). Common utilization of geothermal
7 energy includes the ground source heat pump (GSHP) and earth-to-air heat exchanger
8 (EAHE) (Esen et al., 2007a; Esen et al., 2007c; Esen, 2000) . In recent years, the
9 earth-to-air heat exchanger (EAHE) system has been attracting considerable interest
10 with advantages of lower operational costs and smaller environmental impacts
11 (Mehdid et al., 2018; Rouag et al., 2018). The EAHE system could be applied to
3
 ACCEPTED MANUSCRIPT

1 cool/heat the outdoor air before it was sent into a building due to the stable soil
2 temperature at a given depth (Balbay and Esen, 2013), thereby reducing the building
3 energy consumption. For instance, Tittelein et al. (Tittelein et al., 2009) reported that,
4 compared to a traditional ventilation system, the EAHE system could decrease the
5 building energy consumption of 7 and 9 kWh/m2 per year as the tube depths were 0.6
6 m and 2 m, respectively. Ascione et al. (Ascione et al., 2016) made a comparison
7 between an EAHE system and a conventional mechanical ventilation system, and the
8 results showed that the EAHE system could contribute to the building energy
9 reduction of about 29% in winter and 36-46% in summer for the Mediterranean
10 climate. The EAHE system has been proved as an effective cooling and heating
11 technology for reducing building energy consumption. However, their practical
12 limitations, such as large land occupation demands during construction (Soni et al.,
13 2016), produce severe limitations on their successful applications, particularly in
14 densely built areas. Moreover, in most existing studies the depth of the buried tubes
15 was 2-4 m (Singh et al., 2018)) while, at this depth, the soil temperature tended to be
16 affected by soil properties and environmental conditions such as ambient temperatures
17 and rainfall. This could lead to a large variance of outlet air temperature of EAHE
18 systems (Mustafa Omer, 2008), thereby decreasing the systems’ thermal performance.
19 To address the above issue, a vertical EAHE system (i.e., VEAHE) with a
20 U-tube was proposed in (Liu et al., 2019b; Liu et al., 2019c). In this VEHAE system,
21 a small hole (less than 1 m2) was drilled for the bury of the U-tube and thus the land
22 occupation was significantly reduced compared to the traditional EAHE systems.
23 Moreover, the VEAHE system was designed with a depth of more than 15 m where
24 the soil temperature was more stable than that of the shallow soil. Such soil
25 temperatures were also closer to the required cold/heat source temperature of building
26 ventilation systems (Xi et al., 2017; Zhou et al., 2016), leading to an improvement of
27 energy utilization efficiency. The experimental results (Liu et al., 2019b) and
28 numerical analysis (Liu et al., 2019d) have demonstrated the VEAHE system’s
29 effectiveness. However, subject to the outdoor air temperature variations and air

4
 ACCEPTED MANUSCRIPT

1 velocity increases, the VEAHE system could still have a large air temperature
2 fluctuation at the outlet. For instance, such temperature fluctuations were 2.6ºC and
3 3.8ºC in summer for the system’s air velocity of 1 and 2 m/s, respectively (Liu et al.,
4 2019d), which is difficult to satisfy the requirements of the indoor human comfort. In
5 this view, a VEAHE system coupled with macro-encapsulated phase change material
6 (PCM) was proposed in (Liu et al., 2019a). In this system, three identical
7 stainless-steel annular PCM containers were installed from the outlet to the depth of
8 3.6 m in the left leg of U-tube. The PCM could store cooling/heating energy and use it
9 at a later time, and thus reduce the fluctuation of air temperatures at the outlet (Zhou
10 et al., 2019). The results showed that, for the air velocity of 1 m/s, the system’s air
11 temperature fluctuation and peak at the outlet were decreased by 0.8ºC and 0.5ºC,
12 respectively. When the air velocity became 2 m/s, the decreasing of the air
13 temperature fluctuation and peak at the outlet became 1.1ºC and 1.0ºC, respectively.
14 The integration of annular PCM components into VEAHE systems has been
15 demonstrated with more stable air temperatures at the outlet. However, its relatively
16 complicated structure could add manufacture and installation difficulties as well as
17 expenses, thereby significantly increasing the system’s initial investment. For
18 example, the economic analysis in (Liu et al., 2019a) indicated that the capital
19 investment of the annular PCM components with a length of 3.6 m was $160.0 and
20 the system’s static payback period was calculated as 20.8 years. Note that the
21 manufacture expenses of annular PCM components will be dramatically increased
22 with the increase of component lengths. Such a large initial investment would prevent
23 the proposed system from being adopted and generalized in practical applications.
24 Indeed, the economic feasibility has been treated as an important criterion in choosing
25 an energy efficient technology/system (Esen et al., 2006, 2007b). Therefore, it is
26 highly desirable that an appropriate form of PCM macro-encapsulation can be
27 designed and integrated into the VEAHE systems, with the goal of reducing
28 manufacture expenses and improving the system’s thermal performance. Indeed,
29 different forms of PCM macro-encapsulation have been proposed and applied in the

5
 ACCEPTED MANUSCRIPT

1 building domain, such as rectangular, tubular and spherical. The application of these
2 forms has been reviewed and compared in (Dadollahi and Mehrpooya, 2017; Esen
3 and Ayhan, 1996; Esen et al., 1998; Liu et al., 2018; Zhou Y., 2019). Among these
4 forms, tubular PCM components, due to their simple geometry, easy fabrication,
5 convenient assembling-disassembling and low-cost, have been widely used in solar
6 energy systems (Arunkumar and Kabeel, 2017; Englmair et al., 2019), energy storage
7 tanks (Zhang et al., 2014), and heat exchangers (Dubovsky et al., 2011). In this view,
8 a new VEAHE system coupled with tubular PCM components was proposed in this
9 paper. Specifically, a tubular PCM component was installed from the outlet to the
10 bottom in the left-leg of the U-tube. PCM inside the tubular container can absorb and
11 release the latent heat when its phase change process occurs (Zhou et al., 2018),
12 which can reduce the fluctuation in the system’s air temperature at the outlet as well
13 as its peak.
14 This paper aims at presenting an experimental and numerical study conducted on
15 a new VEAHE system coupled with tubular PCM components. First, its experimental
16 set-up was introduced. Then, the system’s numerical model was developed using the
17 MATLAB/Simulink platform, and the developed model was verified by the
18 monitored data. Based on the verified model, the influences of the tubular PCM
19 component, tube depth, PCM conductivity and container length on the system
20 performance were investigated. Finally, the system’s static payback period was also
21 calculated.
22

23 2. Experimental set-up
24 The proposed system’s schematic diagram and its detailed dimensions are
25 indicated in Fig.1. A stainless steel U-tube with an outer diameter of 219 mm was
26 buried in a drilled hole. A bypass was designed at the U-tube’s bottom for the
27 drainage of condensate water. The left leg of the U-tube (from the outlet to the depth
28 of 7.5 m) was insulated with polyurethane. A tubular PCM component was set in the
29 center of the tube and its container was made up of stainless-steel with a length of
6
 ACCEPTED MANUSCRIPT

1 15.5. The selected PCM was paraffin due to its low cost, high chemical stability, no
2 pollution for the environment etc. (Liu et al., 2018). The site pictures are shown in Fig.
3 2.
4

5
6 Fig.1. The schematic diagram and detailed dimensions of the proposed system

7
 ACCEPTED MANUSCRIPT

1
2 Fig. 2. Site pictures: (a) Drilling hole and burying U-tube; (b) Tubular PCM component
3 The location of the experimental set-up was Changsha (Latitude/Longitude:
4 N28º12'/E112º59'), China. The air and PCM temperature were tested, from October 8
5 to 9, 2017 by using PT100 temperature sensors with a permissible error ±0.15°C. An
6 Agilent 34972A was utilized for experimental data measurement and collection. The
7 sensors of from IT-1 to IT-14 set in the center of the tube are labeled as air
8 temperatures in the tube, as shown in Fig.3. The sensors of TPT-1 to TPT-6 are
9 labeled as PCM temperatures located in the center of the tubular container.

8
 ACCEPTED MANUSCRIPT

1
2 Fig.3. Arrangement of temperature sensors
3

4 3. Numerical model
5 A numerical model was developed to examine the influences of the tubular PCM
6 component. The model was segmented into three parts based on the structure of
7 proposed system. Specifically, the first part (i.e., Part I) is from the inlet to the depth
8 of 15.5 m on the U-tube’s right leg. The second part (i.e., Part II) and third part (i.e.,
9 part III) are from the depth of 15.5 to 7.5 m and 7.5 m to the outlet on the U-tube’s
10 left leg, respectively. In this study, the model equations of each divided part were
11 built and then solved in the MATLAB/Simulink environment. Through the
12 MATLAB/Simulink platform, a block for each divided part was developed using the
13 S-Function. Then these blocks were linked by Simulink platform to predict the
14 temperature variation of the mediums with the operation time.
9
 ACCEPTED MANUSCRIPT

2 3.1. Assumptions of the model

3 The following assumptions can be made for the development of proposed
4 system’s model:
5  Air is incompressible and the influence of its humidity change on the heat transfer
6 can be ignored.
7  The heat conduction between two adjacent layers along the vertical direction can
8 be ignored.
9  The thermal interference between the U-tube’s right and left leg can be ignored
10 due to the minor influences between the neighbor legs as their interval distance is
11 more than two times of the tube diameter (Shojaee and Malek, 2017).
12  The PCM is considered as homogeneous and isotropic.
13  The influence of the container is ignored because of its small wall thickness and
14 high thermal conductivity.
15

16 3.2. Model equations

17 The model of each divided part is regarded as a cylindrical coordinate system
18 (Niu et al., 2015b). Accordingly, various mediums, including air, PCM, insulation,
19 tube and soil, in the vertical direction are segmented into n layers for Part I, II and III.
20 The soil along its radial direction was segmented into 𝑚 columns. The PCM layer
21 was segmented into (N+1) columns along the radial direction. The heat transfer
22 governing equations of these different mediums for each divided unit can be
23 established based on energy balance as follows (Bahrar et al., 2018):

{
𝑇𝑒𝑚𝑝𝑒𝑟𝑎𝑡𝑢𝑟𝑒
}{
𝐶𝑜𝑛𝑑𝑢𝑐𝑡𝑖𝑜𝑛 /𝐶𝑜𝑛𝑣𝑒𝑐𝑡𝑖𝑣𝑒
}{
𝐴𝑑𝑣𝑒𝑐𝑡𝑖𝑜𝑛
24 𝑐ℎ𝑎𝑛𝑔𝑒 𝑟𝑎𝑡𝑒 =
𝑖𝑛 𝑢𝑛𝑖𝑡
𝑡𝑟𝑎𝑛𝑠𝑓𝑒𝑟𝑠
𝑖𝑛 𝑢𝑛𝑖𝑡
+ 𝑡𝑟𝑎𝑛𝑠𝑓𝑒𝑟𝑠
𝑖𝑛 𝑢𝑛𝑖𝑡 } (1)

25 The temperature change rate (i.e., storage of energy) of the divided unit can be
26 written as:

{
𝑇𝑒𝑚𝑝𝑒𝑟𝑎𝑡𝑢𝑟𝑒
27
} 𝑑𝑇
𝑐ℎ𝑎𝑛𝑔𝑒 𝑟𝑎𝑡𝑒 = 𝜌 × 𝐶𝑝 × 𝑉 × 𝑑𝑡
𝑖𝑛 𝑢𝑛𝑖𝑡
(2)

10
 ACCEPTED MANUSCRIPT

1 where 𝜌, 𝐶𝑝, 𝑉 and 𝑇 is the density, specific heat capacity (SHC), volume and

2 temperature of the medium, respectively; t is time.

3 The conduction transfers between the adjacent mediums can be written as:

{
𝐶𝑜𝑛𝑑𝑢𝑐𝑡𝑖𝑜𝑛/𝐶𝑜𝑛𝑣𝑒𝑐𝑡𝑖𝑣𝑒
4 𝑡𝑟𝑎𝑛𝑠𝑓𝑒𝑟𝑠
𝑖𝑛 𝑢𝑛𝑖𝑡
= }
𝑇𝑖 ― 1 ― 2𝑇𝑖 + 𝑇𝑖 + 1
𝑅
(3)

5 where 𝑇𝑖 ― 1, 𝑇𝑖 and 𝑇𝑖 + 1 is the node temperature at node i-1, i, and i+1, 𝑅 is the

6 thermal resistance between adjacent nodes.

7 The advection transfer concerns the flowing air, which represents the energy
8 transported into an air node i from an adjacent air node i-1 along the flowing air
9 direction, as follows:

{
𝐴𝑑𝑣𝑒𝑐𝑡𝑖𝑜𝑛
10
}
𝑡𝑟𝑎𝑛𝑠𝑓𝑒𝑟𝑠 = 𝑚 × 𝐶𝑝 × (𝑇𝑎,𝑖 ― 1 ― 𝑇𝑎,𝑖)
𝑖𝑛 𝑢𝑛𝑖𝑡
(4)

11 where 𝑚 the mass flow rate, 𝑇𝑎,𝑖 ― 1 and 𝑇𝑎,𝑖 is the air temperature at node i and i-1,

12 respectively.
13 The mean temperature between two adjacent time (e.g., time t-Δt and t) was
14 introduced as each divided unit temperature (i.e., 𝑇 = (𝑇𝑡 + 𝑇𝑡 ― ∆𝑡)/2, where 𝑇𝑡 and
15 𝑇𝑡 ― ∆𝑡 are the temperature of each divided unit at any time t and t-Δt (Bahrar et al.,
16 2018; Stathopoulos et al., 2016)). The finite difference method is used to estimate the
17 temperature derivative terms and the final form of the mathematical equations can be
18 written a matrix formulation for each divided unit. The following section conducts the
19 analysis for different mediums of each part.
20

21 3.2.1. Model equations of Part I

22 Part I has the same structure with the VEAHE without PCM studied in (Liu et al.,
23 2019d). Thus they have the same mathematical equations, which can refer to (Liu et
24 al., 2019d).

25 3.2.2. Model equations of Part II

26 In Part II model, the PCM model was segmented into (N+1) layers. For its
11
 ACCEPTED MANUSCRIPT

1 (N+1)-th layer, a unidirectional heat conduction from it to the N-th layer was assumed
2 to simplify the calculation without considering the heat exchange from another
3 internal layer (Roccamena, 2017). To decrease the influence of this assumption, a
4 fixed radius of r1 (1 mm) was assigned for the (N+1)-th layer. The PCM from r1 to
5 outermost (𝑟𝑃) was segmented into N layers with the identical thickness. The nodal

6 representation in Part II is illustrated in Fig.4.

7
8 Fig. 4. The nodal representation in Part II.
9
10 1) Air model in Part II
11 The heat transfer for the flowing air in Part II includes the energy delivered from
12 the (j-1)-th to j-th layer, and the heat convection of the flowing air with both the tube
13 and PCM 1. For each air unit (e.g., j-th layer) at t-Δt:
2(𝑇𝑃𝑎,𝑗 ― 𝑇𝑡𝑃𝑎,𝑗
― ∆𝑡
) (𝑇𝑃,1,𝑗 ― 𝑇𝑃𝑎,𝑗) (𝑇𝑡,𝑗 ― 𝑇𝑃𝑎,𝑗)
14 𝜌𝑎𝐶𝑎𝑉𝑃,𝑎 × 𝑑𝑡 = 𝐶𝑎𝑚 × (𝑇𝑃𝑎,𝑗 ― 1 ― 𝑇𝑃𝑎,𝑗) + 𝑅𝑃,1,𝑗 + 𝑅𝑡,𝑗 (5)

15 The Eq. (5) can be written as:

16 𝑇𝑡𝑃𝑎,𝑗
― ∆𝑡
[ 𝐶𝑎𝑚𝑑𝑡 𝑑𝑡 𝑑𝑡
] 𝐶𝑎𝑚𝑑𝑡
= 1 + 2 × 𝜌𝑎𝐶𝑎𝑉𝑃,𝑎 + 2 × 𝜌𝑎𝐶𝑎𝑉𝑃,𝑎 × 𝑅𝑡,𝑗 + 2 × 𝜌𝑎𝐶𝑎𝑉𝑃,𝑎 × 𝑅𝑃,1,1 × 𝑇𝑃𝑎,𝑗 ― 2 × 𝜌𝑎𝐶𝑎𝑉𝑃,𝑎 × 𝑇𝑃𝑎,𝑗 ― 1

𝑑𝑡 𝑑𝑡
17 ― 2 × 𝜌𝑎𝐶𝑎𝑉𝑃,𝑎 × 𝑅𝑡,𝑗 × 𝑇𝑡,𝑗 ― 2 × 𝜌𝑎𝐶𝑎𝑉𝑃,𝑎 × 𝑅𝑃,1,1 × 𝑇𝑃,1,𝑗 (6)

18 where 𝐶𝑎 , 𝜌𝑎, and 𝑚 are the SHC, density and mass flow of the flowing air,

19 respectively; 𝑇𝑃𝑎,𝑗 and 𝑇𝑃𝑎,𝑗 ― 1 are the mean air temperature between t-Δt and t;

20 𝑇𝑡𝑃𝑎,𝑗
― ∆𝑡
is the air temperature in Part II at t-Δt; 𝑇𝑃,1,𝑗 is the mean PCM 1 temperature

21 between t-Δt and t; 𝑉𝑃,𝑎 is the divided air volume in Part II; 𝑅𝑡,𝑗 and 𝑅𝑃,1,𝑗 are the
12
 ACCEPTED MANUSCRIPT

1 thermal resistance between tube wall and air, and between air and PCM 1,
2 respectively:
1
1 1 𝑟𝑖 + 2 × 𝑑
3 𝑅𝑡,𝑗 = ℎ𝑡𝑆𝑡 + 2𝜋 × 𝜆𝑡 × ∆𝑧 ln 𝑟𝑖 (7)
1 1 𝑟𝑃
4 𝑅𝑃,1,𝑗 = ℎ𝑃𝑆𝑃 + 2𝜋 × 𝜆𝑝𝑐𝑚 × ∆𝑧ln (𝑟𝑃 ― 𝑟1) (8)
𝑟𝑃 ― 2×𝑁 ×1

5 where 𝑟𝑃 and 𝜆𝑝𝑐𝑚 are the radius and thermal conductivity of the PCM,
6 respectively; 𝑆𝑃 and ℎ𝑡 are the heat exchange area and coefficient of between air
7 and PCM, respectively; 𝑆𝑡 and ℎ𝑃 are the heat exchange area and coefficient of
8 between air and tube, respectively; and ℎ𝑃 are related to 𝑟𝑖, 𝑟𝑃 and 𝑣𝑚 (Niu et al.,
𝑟𝑖2 × 𝑣
9 2015a), and 𝑣𝑝 is the mean air velocity in Part II, 𝑣𝑚 = (𝑟2 ― 𝑟2 ) .
𝑖 𝑃

10 `
11 2) PCM model
12
13 For PCM 1:
14 The heat transfer for PCM 1 in Part II includes the heat convection between
15 PCM 1 and the flowing air, and the heat conduction between PCM 1 and 2. The
16 temperature of PCM 1 for each divided layer (e.g., j-th) at t-Δt is counted as:

17 𝑇𝑡𝑃,1,𝑗
― ∆𝑡
( 𝑑𝑡 1 𝑑𝑡 1
)
= 1 + 2 × 𝜌𝑃𝐶𝑃,1,𝑗𝑉𝑃,1,𝑗 × 𝑅𝑃,2,𝑗 + 2 × 𝜌𝑃𝐶𝑃,1,𝑗𝑉𝑃,1,𝑗 × 𝑅𝑃,1,𝑗 × 𝑇𝑃,1,𝑗 ― 2 × 𝜌𝑃𝐶𝑃,1,𝑗𝑉𝑃,1,𝑗 × 𝑅𝑃,2,𝑗
𝑑𝑡 1

𝑑𝑡 1
18 × 𝑇𝑃,2,𝑗 ― 2 × 𝜌𝑃𝐶𝑃,1,𝑗𝑉𝑃,1,𝑗 × 𝑅𝑃,1,𝑗 × 𝑇𝑃𝑎,𝑗 (9)

19 where 𝑇𝑃,2,𝑗 is the mean temperature of PCM 2 between t-Δt and t; 𝑉𝑃,1,𝑗, 𝐶𝑃,1,𝑗 and

20 𝜌𝑃 is the divided volume, SHC, density of the PCM 1, respectively; and the 𝐶𝑃,1,𝑗
21 varies with PCM temperature and can be obtained by DSC testing (Bahrar et al.,
22 2018).
23
24 For PCM k:
25 The heat transfer for PCM k in Part II includes the heat conduction between
26 PCM k and PCM (k-1), and between PCM k and PCM (k+1). The temperature of
27 PCM k for each divided layer (e.g., j-th) at t-Δt is counted as:

28 𝑇𝑡𝑃,𝑘,𝑗
― ∆𝑡
( 𝑑𝑡 1 𝑑𝑡 1
)
= 1 + 2 × 𝜌𝑃𝐶𝑃,𝑘,𝑗𝑉𝑃,𝑘,𝑗 × 𝑅𝑃,𝑘,𝑗 + 2 × 𝜌𝑃𝐶𝑃,𝑘,𝑗𝑉𝑃,𝑘,𝑗 × 𝑅𝑃,𝑘 + 1,𝑗 × 𝑇𝑃,𝑘,𝑗 ― 2 × 𝜌𝑃𝐶𝑃,𝑘,𝑗𝑉𝑃,𝑘,𝑗 × 𝑅𝑃,𝑘,𝑗
𝑑𝑡 1

𝑑𝑡 1
29 × 𝑇𝑃,𝑘 ― 1,𝑗 ― 2 × 𝜌𝑃𝐶𝑃,𝑘,𝑗𝑉𝑃,𝑘,𝑗 × 𝑅𝑃,𝑘 + 1,𝑗 × 𝑇𝑃,𝑘 + 1,𝑗 (10)
13
 ACCEPTED MANUSCRIPT

1 where 𝑇𝑃,𝑘 ― 1,𝑗 , 𝑇𝑃,𝑘,𝑗 and 𝑇𝑃,𝑘 + 1,𝑗 are the mean temperature of PCM (k-1), k, and

2 (k+1) between t-Δt and t; 𝐶𝑃,𝑘,𝑗 and 𝑉𝑃,𝑘,𝑗 is the SHC and divided volume of the

3 PCM k, respectively.
4
5 For PCM N:
6 The heat exchange for PCM N in Part II includes the heat conduction between
7 PCM N and (N-1), and between PCM N and (N+1). The temperature of PCM N for
8 each divided layer (e.g., j-th) at t-Δt is counted as:

9 𝑇𝑡𝑃,𝑁,𝑗
― ∆𝑡
( 𝑑𝑡 1 𝑑𝑡 1
)
= 1 + 2 × 𝜌𝑃𝐶𝑃,𝑁,𝑗𝑉𝑃,𝑁,𝑗 × 𝑅𝑃,𝑁,𝑗 + 2 × 𝜌𝑃𝐶𝑃,𝑁,𝑗𝑉𝑃,𝑁,𝑗 × 𝑅𝑃,𝑁 + 1,𝑗 × 𝑇𝑃,𝑁,𝑗 ― 2 × 𝜌𝑃𝐶𝑃,𝑁,𝑗𝑉𝑃,𝑁,𝑗
𝑑𝑡

1 𝑑𝑡 1
10 × 𝑅𝑃,𝑁,𝑗 × 𝑇𝑃,𝑁 ― 1,𝑗 ― 2 × 𝜌𝑃𝐶𝑃,𝑁,𝑗𝑉𝑃,𝑁,𝑗 × 𝑅𝑃,𝑁 + 1,𝑗 × 𝑇𝑃,𝑁 + 1,𝑗 (11)

11 where 𝑇𝑃,𝑁 ― 1,𝑗 , 𝑇𝑃,𝑁,𝑗 and 𝑇𝑃,𝑁 + 1,𝑗 are the mean temperature of the PCM (N-1),

12 N and (N+1) between t-Δt and t; 𝐶𝑃,𝑁,𝑗 and 𝑉𝑃,𝑁,𝑗 are the SHC and divided volume

13 of the PCM N, respectively.

14
15 For PCM (N+1):
16 The heat exchange for PCM (N+1) in Part II includes the heat conduction
17 between PCM (N+1) and N. The temperature of PCM (N+1) for each divided layer
18 (e.g., j-th) at t-Δt is counted as:

19 𝑇𝑡𝑃,𝑁
― ∆𝑡
(
+ 1,𝑗 = 1 +
𝑑𝑡
2 × 𝜌𝑃𝐶𝑃,𝑁 + 1,𝑗𝑉𝑃,𝑁 + 1,𝑗
×
1
𝑅𝑃,𝑁 + 1,𝑗 ) × 𝑇𝑃,𝑁 + 1,𝑗 ―
𝑑𝑡
2 × 𝜌𝑃𝐶𝑃,𝑁 + 1,𝑗𝑉𝑃,𝑁 + 1,𝑗
1
20 × 𝑅𝑃,𝑁 + 1,𝑗 × 𝑇𝑃,𝑁,𝑗 (12)

21 where 𝐶𝑃,𝑁 + 1,𝑗 and 𝑉𝑃,𝑁 + 1,𝑗 are the SHC and divided volume of the PCM (N+1),

22 respectively.
23 3) Tube wall model
24 The heat transfer for the tube wall in Part II consists of the heat convection
25 between the tube wall and flowing air, and the heat conduction between soil 1 and the
26 tube wall. Thus, the temperature of the tube wall for each divided layer (e.g., j-th) at
27 t-Δt is counted as:

28 ( 𝑑𝑡 1 𝑑𝑡 1
) 𝑑𝑡
𝑇𝑡𝑡,𝑗― ∆𝑡 = 1 + 2 × 𝜌𝑡𝐶𝑡𝑉𝑡,𝑗 × 𝑅𝑡,𝑗 + 2 × 𝜌𝑡𝐶𝑡𝑉𝑡,𝑗 × 𝑅𝑠,1,𝑗 × 𝑇𝑡,𝑗 ― 2 × 𝜌𝑡𝐶𝑡𝑉𝑡,𝑗 × 𝑅𝑡,𝑗 × 𝑇𝑎,𝑗 ―
1

14
 ACCEPTED MANUSCRIPT

𝑑𝑡 1
1 2 × 𝜌𝑡𝐶𝑡𝑉𝑡,𝑗 × 𝑅𝑠,1,𝑗 × 𝑇𝑠,1,𝑗 (13)

2 where 𝐶𝑡 and 𝜌𝑡 is the SHC and density of the tube material, respectively;

4 4) Soil model
5 The heat transfer for the soil i (𝑖 ≥ 2) in Part II includes the heat conduction
6 between the soil (i-1) and i, and between soil i and (i+1). Thus, the temperature of the
7 soil i for each divided layer (e.g., j-th layer) at t-Δt can be counted as:

8 𝑇𝑡𝑠,𝑖,𝑗
― ∆𝑡
( 𝑑𝑡 1 𝑑𝑡 1
) 𝑑𝑡
= 1 + 2 × 𝜌𝑠𝐶𝑠𝑉𝑠,𝑖,𝑗 × 𝑅𝑠,𝑖,𝑗 + 2 × 𝜌𝑠𝐶𝑠𝑉𝑠,𝑖,𝑗 × 𝑅𝑠,𝑖 + 1,𝑗 × 𝑇𝑠,𝑖,𝑗 ― 2 × 𝜌𝑠𝐶𝑠𝑉𝑠,𝑖,𝑗 ×
1
𝑅𝑠,𝑖,𝑗 × 𝑇𝑠,𝑖 ― 1,𝑗

𝑑𝑡 1
9 ― 2 × 𝜌𝑠𝐶𝑠𝑉𝑠,𝑖,𝑗 × 𝑅𝑠,𝑖 + 1,𝑗 × 𝑇𝑠,𝑖 + 1,𝑗 (14)

10 where 𝑇𝑠,𝑖,𝑗 , 𝑇𝑠,𝑖 ― 1,𝑗 and 𝑇𝑠,𝑖 + 1,𝑗 are the mean temperature of soil i, (i-1) and (i+1)

11 between t-Δt and t, respectively; and for i=1, 𝑇𝑠,𝑖 ― 1,𝑗 = 𝑇𝑡,𝑗 ; and for i=m, 𝑇𝑠,𝑖 + 1,𝑗 =

12 𝑇𝑔,𝑗, and 𝑇𝑔,𝑗 is the outermost soil temperature.

13 3.2.3. Model equations of Part III

14 The nodal representation in Part III are illustrated in Fig.5.

15
16 Fig. 5. The nodal representation in Part III.
17

18 1) Air and PCM model

19 The air and PCM model in Part III have almost identical equations comparing
20 with the Part II, except for the initial air and PCM temperatures based on the tube
15
 ACCEPTED MANUSCRIPT

1 depths.
2

3 2) Tube wall model

4 The heat transfer for the tube wall in Part III consists of the heat convection
5 between the tube wall and air, and the heat conduction between the tube wall and
6 insulation. Hence, the temperature of tube wall for each divided layer (e.g., the j-th) at
7 t-Δt is counted as:

8 (
𝑇𝑡𝑡,𝑗― ∆𝑡 = 1 +
𝑑𝑡
×
1
+
𝑑𝑡
×
1
2 × 𝜌𝑡𝐶𝑡𝑉𝑡,𝑗 𝑅𝑡,𝑗 2 × 𝜌𝑡𝐶𝑡𝑉𝑡,𝑗 𝑅𝐼,𝑗
× 𝑇𝑡,𝑗 ―
𝑑𝑡
) ×
1
2 × 𝜌𝑡𝐶𝑡𝑉𝑡,𝑗 𝑅𝑡,𝑗
× 𝑇𝑃𝑎,𝑗 ―

𝑑𝑡 1
9 2 × 𝜌𝑡𝐶𝑡𝑉𝑡,𝑗 × 𝑅𝐼,𝑗 × 𝑇𝐼,𝑗 (15)

10 where 𝑇𝐼,𝑗 is the mean temperature of the insulation between t-Δt and t.

11 3) Insulation model
12 The heat transfer of the insulation in Part III consists of the heat conduction
13 between the insulation and tube wall, and heat conduction between soil 1 and the
14 insulation. Thus, the temperature of the insulation for each divided layer (e.g., the j-th)
15 at t-Δt is counted as:

16 ( 𝑑𝑡 1 𝑑𝑡 1
) 𝑑𝑡 1 𝑑𝑡
𝑇𝑡𝐼,𝑗― ∆𝑡 = 1 + 2 × 𝜌𝐼𝐶𝐼𝑉𝐼 × 𝑅𝑃𝐼,𝑗 ― 2 × 𝜌𝐼𝐶𝐼𝑉𝐼 × 𝑅𝑃𝑠,1,𝑗 × 𝑇𝐼,𝑗 ― 2 × 𝜌𝐼𝐶𝐼𝑉𝐼 × 𝑅𝐼,𝑗 × 𝑇𝑡,𝑗 ― 2 × 𝜌𝐼𝐶𝐼𝑉𝐼

1
17 × 𝑅𝑃𝑠,1,𝑗 × 𝑇𝑃𝑠,1,𝑗 (16)

18 where 𝑇𝑃𝑠,1,𝑗 is the mean temperature of soil 1 between t-Δt and t; 𝑉𝐼, 𝐶𝐼 and 𝜌𝐼

19 is the divided volume, SHC and density of the insulation, respectively.

20

21 4) Soil model
22 The soil model of Part III has the same heat transfer equations as Part II.
23

24 3.3. SHC treatment

25 The SHC of selected PCM was obtained using the method of DSC
26 measurement, with a cooling and heating rate of 1°C per minute (Kong et al., 2013)
27 and Eq.(21) (Kheradmand et al., 2016; Kheradmand et al., 2015).

16
 ACCEPTED MANUSCRIPT

𝐷(𝑇)
1 𝐶𝑃(𝑇) = 𝜑 (17)

2 where 𝐶𝑃(𝑇) is the PCM’s SHC at the tested temperature of T ºC, J/(g·ºC). 𝐷(𝑇) is

3 the corresponding data of the DSC testing curve at tested temperature of T ºC,
4 mW/mg. 𝜑 is the testing rate of heating/cooling, ºC/s. The PCM’s SHC is calculated
5 and illustrated in Fig.6.

6
7 Fig.6. The curves of SHC under heating/cooling testing condition

8 In the development of system model, the curve values of the SHC with the
9 change in PCM temperatures were introduced by using an interpolation method based
10 on the comparison of PCM temperature values between consecutive time steps (e.g.,
11 at t-Δt and t) (Bahrar et al., 2018; Stathopoulos et al., 2016).
12

17
 ACCEPTED MANUSCRIPT

1 4. Model verification

2
3 Fig.7. The comparison of the simulated and experimental air temperatures at the outlet

4 The numerical model was verified against the experimental results. Fig.7
5 indicates the variation of the air temperature at the outlet under the simulated and
6 monitored conditions. It can be seen that the verification produced acceptable results.
7 In particular, for the system’s air temperature at the outlet, the maximum difference
8 between the monitored and simulated results is 0.28ºC, and the corresponding
9 maximum absolute relative error is 1.33%. In addition, the mean outlet air
10 temperature difference between simulated and measured results is less than 0.07°C.
11 The results show that the model is accurate to explore the influences of the tubular
12 PCM component.
13

14 5. Results and discussion

15 5.1. Comparison of the system between with and without PCM

16 Based on different mass flow rates, the air temperatures at the outlet of the
17 VEAHE system with and without PCM were compared to analyze the influences of
18 the tubular PCM component. For demonstration purposes, a typical outdoor weather
19 condition of Changsha in summer (i.e., the outdoor temperature ranges from 25.28ºC
20 to 40.04ºC on August 18th, 2017) was used as the air temperature at the inlet of the
18
 ACCEPTED MANUSCRIPT

1 developed model, as illustrated in Fig.8.

2
3 Fig.8. The variation of outdoor air temperatures in this study

4
5 Fig.9. The air temperatures at the outlet of the proposed system and pure VEAHE system
6 The influences of the tubular PCM component on the VEAHE system’s thermal
7 performance were examined based on different mass flow rates (i.e., 0.024, 0.047,
8 0.094 and 0.141 m/s). The air temperatures at the outlet of the proposed system and
9 the pure VEAHE system (i.e., without PCM) are shown in Fig.9. As can be seen, the
10 proposed system has a lower air temperature at the outlet than the pure VEAHE
11 system under the different mass flow rates. Specifically, for the proposed system, its
12 peak temperatures are 22.35ºC, 23.44ºC, 25.13ºC and 26.44ºC under the mass flow
13 rates of 0.024, 0.047, 0.094 and 0.141 m/s, respectively. Compared to the pure

19
 ACCEPTED MANUSCRIPT

1 VEAHE system, the corresponding drops in peak temperatures are 0.74ºC, 0.96ºC,
2 1.11ºC and 1.12ºC. Meanwhile, it also can be seen that the proposed system has a
3 smaller temperature fluctuation than the pure VEAHE system. Specifically, as the
4 mass flow rates are 0.024, 0.047, 0.094 and 0.141 m/s, the temperature fluctuations
5 are 1.26ºC, 2.13ºC, 3.39ºC and 4.35ºC for the proposed system, and 1.93ºC, 2.95ºC,
6 4.30ºC and 5.25ºC for the pure VEAHE system, respectively. Compared to the
7 influence of the annular PCM component on the VEAHE system in (Liu et al., 2019a),
8 the tubular PCM component has a larger drop in the air temperature fluctuation and
9 peak at the outlet. For example, as the mass flow rate is 0.047 kg/s, the air
10 temperature fluctuation drop at the outlet for the system with the tubular and annular
11 PCM component is 0.82ºC and 0.80°C, respectively. And the corresponding outlet air
12 peak temperature drop is 0.96ºC and 0.54°C. Note that the outdoor temperature ranges
13 in this study (25.28ºC to 40.04ºC) is larger than that in (Liu et al., 2019a) (24.32ºC to
14 38.17ºC). These results show that the tubular PCM component integrated into the
15 VEAHE system can effectively cut down the air temperature’s peak and fluctuation at
16 the outlet by the PCM’s energy release and storage. Moreover, the mean cooling
17 capacity of the proposed system during its operation is 208.25 W, 389.11 W, 684.70
18 W and 814.02 W under the mass flow rates of 0.024, 0.047, 0.094 and 0.141 m/s,
19 respectively. The corresponding cooling capacity for the pure VEAHE system is
20 201.51 W, 370.62 W, 644.12 W and 764.90 W. This result indicates that the tubular
21 PCM component can increase the corresponding mean cooling capacity of the
22 VEAHE system by 2.89%, 4.53%, 5.67% and 5.58%, respectively.

20
 ACCEPTED MANUSCRIPT

1 5.2. The influences of different tube depths

2
3 Fig.10. The influences of the tube depths on air temperature at the outlet
4 Four tube depths (i.e., 12, 16, 20 and 24 m) were considered to investigate its
5 influences on the proposed system’s air temperature at the outlet. The air
6 temperatures at the outlet under different tube depths are shown in Fig.10. As can be
7 seen, the air temperature’s peak and fluctuation at the outlet reduce with the increase
8 of the tube depths. Specifically, as the tube depths are 12, 16, 20 and 24 m, the peak
9 temperatures are 25.74ºC, 23.20ºC, 21.79ºC and 21.01ºC, respectively. The
10 corresponding air temperature fluctuations at the outlet are 3.59ºC, 1.98ºC, 1.11ºC
11 and 0.62ºC. This result can be explained by the fact that a larger tube depth enables
12 more heat exchange between air and deep soil. Therefore, a more stable and lower
13 air temperature at the outlet can be acquired by increasing the tube depths, thereby
14 improving the proposed system’s performance. However, a larger tube depth will
15 also result in a higher initial investment. Meanwhile, the difficulty in the system
16 construction will be considerably increased, particularly for drilling the hole and
17 burying the U-tube. From Fig. 10, it also can be seen that the air temperature’s peak
18 and fluctuations decrease at the outlet tends to reduce with the increase of tube
19 depths. For example, the air temperature’s peak and fluctuation at the outlet
20 decreases by 2.54°C and 1.61°C with the increase of tube depth from 12 to 16 m,
21 respectively. While the corresponding temperature decreases by 0.78°C and 0.49°C

21
 ACCEPTED MANUSCRIPT

1 with the increase of tube depth from 16 to 20 m, respectively. In this view, there are
2 technical and economic trade-offs associated with the tube depth in the design stage
3 in practical applications. Additionally, it should be noted that, as the system’s air
4 temperatures at the outlet will reduce with the increasing of the tube depth, a
5 lower/higher phase change temperature of the PCM will be needed when a
6 larger/smaller tube depth is adopted. For example, when 12 and 24 m are selected,
7 the outlet air temperatures are 22.15-25.74ºC and 20.38-21.00ºC respectively, based
8 on the model prediction. Thus, the corresponding phase change temperature of the
9 PCM could be selected as 22.10-25.80ºC and 20.30-21.00ºC.
10

11 5.3. The influences of PCM thermal conductivity

12
13 Fig.11(a). The influences of PCM thermal conductivity for 50 mm container diameter

22
 ACCEPTED MANUSCRIPT

1
2 Fig.11(b). The influences of PCM thermal conductivity for 150 mm container diameter
3
4 Different values of PCM thermal conductivity (i.e., 0.5, 1, 2, 4 and 8 W/(m K))
5 were utilized to analyze their influences on the system’s air temperature at the outlet,
6 based on the container diameters of 50 and 150 mm. The air temperatures at the outlet
7 under different PCM thermal conductivity are shown in Fig.11. As shown in Fig.11(a),
8 when the container diameter is 50 mm, PCM thermal conductivity has almost the
9 same influences on the system’s air temperature at the outlet. This indicates that its
10 increasing is not an effective means for improving the system’s performance as the
11 container diameter is relatively small. As shown in Fig.11(b), as the container
12 diameter increases to 150 mm, this influence become apparent and requires to be
13 considered. The main reason is that the PCM thermal resistances are related to PCM
14 conductivity, lengths and diameters. For a small container diameter, the PCM thermal
15 resistance is kept at a low level, and thus its increase has a small influence on the
16 system’s air temperature at the outlet. When the container diameter increases to 150
17 mm, its thermal resistance increases thereby enlarging the influences of PCM thermal
18 conductivity. However, such influences will be weakened with the increase of the
19 PCM thermal conductivity. Specifically, the air temperature fluctuation at the outlet
20 ranges from 0.75ºC to 0.61ºC and 0.61ºC to 0.56ºC when the thermal conductivity
21 increases from 0.5 to 2 and 2 to 8 W/(m K), respectively. Similar studies are found in
22 a VEAHE system (Liu et al., 2019d) and a VEAHE system integrated with annular
23
 ACCEPTED MANUSCRIPT

1 PCM components (Liu et al., 2019a). In these two studies, the results also indicate
2 that the thermal conductivity has a negligible effect on the system performance when
3 the tube wall thickness and PCM thickness are relatively small, respectively.
4 Therefore, the simple selection of PCM with high thermal conductivity is not an
5 efficient means of improving the system performance in practical applications.

6 5.4. The influences of container lengths

7
8 Fig.12. The influences of the container lengths
9 Six container lengths (i.e., 4, 6, 8, 10, 12 and 14 m) were considered to
10 investigate the influences of container lengths on the system’s air temperature at the
11 outlet. The air temperatures at the outlet under different container lengths are shown
12 in Fig.12. As can be seen, the air temperature’s peak and fluctuation at the outlet
13 decrease with the increase of the container lengths. In particular, as the container
14 lengths are 4, 6, 8, 10, 12 and 14 m, the air temperature fluctuations at the outlet are
15 2.73ºC, 2.64ºC, 2.54ºC, 2.47ºC, 2.37ºC and 2.28ºC, respectively. And the
16 corresponding peak temperatures are 24.09ºC, 23.96ºC, 23.81ºC, 23.73ºC, 23.64ºC
17 and 23.62ºC. The main reason is that a larger container length can contain more PCM
18 and thus provide more energy to adjust the air temperature variation in the tube,
19 thereby leading to a smaller air temperature fluctuation at the outlet. It also can be
20 seen that the air temperature peak decrease tends to reduce with the rise of the
21 container lengths. In particular, a peak temperature drop of 0.02ºC occurs when the

24
 ACCEPTED MANUSCRIPT

1 container length increases from 12 to 14 m. Meanwhile, from Fig.12, it can be seen

2 that the average air temperatures at the outlet are 22.48ºC, 22.42ºC, 22.34ºC, 22.31ºC,
3 22.30ºC and 22.33ºC for the container lengths of 4, 6, 8, 10, 12 and 14 m, respectively.
4 This result shows that the average air temperature at the outlet is the lowest as the
5 container length is 12 m. Based on above analysis, 12 m can be considered as an
6 appropriate container length for the proposed system. Similar results can be found in
7 (Liu et al., 2019a; Liu et al., 2019d). Thus, obtaining a more stable air temperature at
8 the outlet by simply increasing the PCM length may not be an effective method in
9 practical applications, especially when the system’s capital costs are considered.
10

11 5.5. Economic comparison between tubular and annular components

12 To evaluate the proposed system’s economic feasibility, the initial investment of
13 the tubular component and the system’s static payback period (SPP) were calculated
14 and compared with those of the system integrated with annular components in (Liu et
15 al., 2019a). As an important economic analysis method for evaluating a new system,
16 SPP can be calculated as a ratio of the total initial investment of the proposed system
17 to its yearly cost-saving from the energy savings by using the following equation (Mi
18 et al., 2016) :
𝐶𝑡 ― 𝐶𝑎
19 SPP = 𝑆𝑦
(18)

20 where 𝐶𝑡 is the total cost of the VEAHE system (i.e., $1313.9 (Liu et al., 2019b))

21 and the tubular PCM component (including both the container and PCM, with a total
22 of $99.3, which is much lower than the cost of the annular PCM component with the
23 same amount of PCM (i.e., $160.0) (Liu et al., 2019a)); 𝐶𝑎 is the cost of a
24 conventional air-conditioner, which can be assumed as $300 (Liu et al., 2019b); 𝑆𝑦 is

25 the annual income produced from the energy-saving comparing with a traditional air
26 conditioning with a COP of 2.7 (Huang and Mauerhofer, 2016).
27 According to the calculation method in (Liu et al., 2019b), the system’s annual
28 energy conservation for cooling can be calculated as 514.66 kWh as the air velocity
29 was set at 2 m/s. Currently the electric rate is 0.12$/kWh in Changsha. Thus, the 𝑆𝑦
25
 ACCEPTED MANUSCRIPT

1 of the proposed system was $61.76. Based on above analysis, the system’s SPP can be
2 calculated as 18.03 years, which is lower than the VEAHE system coupled with
3 annular PCM components (i.e., 20.8 years (Liu et al., 2019a)).
4

5 6. Conclusions and future work

6 This study proposes a new VEAHE system coupled with the tubular PCM
7 component. Comparing with the traditional EAHE system, its main advantage
8 includes smaller land occupation during construction, deeper buried tube depth for
9 more efficient utilization of geothermal energy, more stable air temperatures at the
10 outlet for improving the indoor thermal comfort and higher cooling capacity for
11 reducing energy consumption. To evaluate the system’s thermal feasibility, different
12 experiments were conducted and a numerical model was developed. Then the model
13 was verified through a comparison between the system’s simulated and monitored air
14 temperatures at the outlet. The verification produced acceptable results with the
15 maximum absolute relative error of 1.33%. The verified model was used to study the
16 influences of the tubular PCM component based on different air velocities. Also, the
17 influences of tube depths, PCM conductivity and container lengths were analyzed. In
18 addition, the system's static payback period was also conducted and compared with
19 the system coupled with annular PCM components.
20 The tubular PCM component integrated into the VEAHE system can decrease its
21 outlet air temperature’s peak of 0.74ºC, 0.96ºC, 1.11ºC and 1.12ºC, and the
22 corresponding temperature fluctuation of 0.67ºC, 0.82ºC, 0.91ºC and 0.90ºC as the
23 mass flow rates are 0.024, 0.047, 0.094 and 0.141 m/s, respectively. Meanwhile, the
24 tubular PCM component can increase the corresponding mean cooling capacity of the
25 VEAHE system by 2.89 %, 4.53%, 5.67% and 5.58%, respectively.
26 The air temperature’s peak and fluctuation at the outlet reduce with the
27 increasing of tube depths. In view of the fact that a larger tube depth will result in a
28 higher initial investment and construction difficulty, there are technical and economic
29 trade-offs associated with the tube depth in practical applications.
26
 ACCEPTED MANUSCRIPT

1 This study also revealed that different PCM thermal conductivity has almost the
2 same influences on the system’s air temperature at the outlet as the container diameter
3 is 50 mm, thus the simple selection of PCM with high thermal conductivity is not an
4 efficient means when the container diameter is relatively small. However, when the
5 container diameter increases to 150 mm, such influences become apparent and require
6 to be considered.
7 The air temperature peak and fluctuation at the outlet reduce with the increase of
8 container lengths, however, the decrease in air temperature peaks tends to reduce with
9 the increasing of container lengths. In this study, 12 m can be considered as an
10 appropriate container length for the proposed system.
11 The proposed system’s static payback period could be calculated as 18.03 years,
12 which is shorter than the VEAHE system with annular PCM components.
13 These results demonstrate the proposed system’s feasibility and effectiveness. In
14 our future work, the developed model will be coupled with a building model to
15 explore the influences of the proposed system on building energy consumption in
16 different areas. In addition, the combination between the proposed system and other
17 energy-saving technologies/systems (e.g., night ventilation and PCM walls) will be
18 explored to achieve the system's intermittent operation as well as better thermal
19 performance.
20

21 Acknowledgements
22 The authors gratefully acknowledge the support from the China Construction
23 Fifth Engineering Division Corp., Ltd. (Project No. 201991370055).

24

25 References
26 Arunkumar, T., Kabeel, A.E., 2017. Effect of phase change material on concentric circular tubular
27 solar still-Integration meets enhancement. Desalination 414, 46-50.
28 Ascione, F., D'Agostino, D., Marino, C., Minichiello, F., 2016. Earth-to-air heat exchanger for NZEB
29 in Mediterranean climate. Renewable Energy 99, 553-563.
27
 ACCEPTED MANUSCRIPT

1 Bahrar, M., Djamai, Z.I., El Mankibi, M., Si Larbi, A., Salvia, M., 2018. Numerical and experimental
2 study on the use of microencapsulated phase change materials (PCMs) in textile reinforced concrete
3 panels for energy storage. Sustainable Cities and Society 41, 455-468.
4 Balbay, A., Esen, M., 2013. Temperature distributions in pavement and bridge slabs heated by using
5 vertical ground-source heat pump systems. Acta Scientiarum Technology 35(4), 677-685.
6 Balbay, A., Esen, M., 2010. Experimental Investigation of Using Ground Source Heat Pump System
7 for Snow Melting on Pavements and Bridge Decks. Scientific Research and Essays, 5(24), 3955-3966.
8 Dadollahi, M., Mehrpooya, M., 2017. Modeling and investigation of high temperature phase change
9 materials (PCM) in different storage tank configurations. Journal of Cleaner Production 161, 831-839.
10 Dubovsky, V., Ziskind, G., Letan, R., 2011. Analytical model of a PCM-air heat exchanger. Applied
11 Thermal Engineering 31(16), 3453-3462.
12 Englmair, G., Moser, C., Schranzhofer, H., Fan, J., Furbo, S., 2019. A solar combi-system utilizing
13 stable supercooling of sodium acetate trihydrate for heat storage: Numerical performance investigation.
14 Applied Energy 242, 1108-1120.
15 Esen, H., Esen, M., Ozsolak, O., 2017. Modelling and experimental performance analysis of
16 solar-assisted ground source heat pump system. Journal of Experimental & Theoretical Artificial
17 Intelligence 29(1), 1-17.
18 Esen, H., Inalli, M., Esen, M., 2006. Technoeconomic appraisal of a ground source heat pump system
19 for a heating season in eastern Turkey. Energy Conversion and Management 47(9), 1281-1297.
20 Esen, H., Inalli, M., Esen, M., 2007a. Numerical and experimental analysis of a horizontal
21 ground-coupled heat pump system. Building and Environment 42(3), 1126-1134.
22 Esen, H., Inalli, M., Esen, M., 2007b. A techno-economic comparison of ground-coupled and
23 air-coupled heat pump system for space cooling. Building and Environment 42(5), 1955-1965.
24 Esen, H., Inalli, M., Esen, M., Pihtili, K., 2007c. Energy and exergy analysis of a ground-coupled heat
25 pump system with two horizontal ground heat exchangers. Building and Environment 42(10),
26 3606-3615.
27 Esen, M., 2000. Thermal performance of a solar-aided latent heat store used for space heating by heat
28 pump. Solar Energy 69(1), 15-25.
29 Esen, M., Ayhan, T., 1996. Development of a model compatible with solar assisted cylindrical energy
30 storage tank and variation of stored energy with time for different phase change materials. Energy
31 Conversion and Management 37(12), 1775-1785.
32 Esen, M., Durmuş, A., Durmuş, A., 1998. Geometric design of solar-aided latent heat store depending
33 on various parameters and phase change materials. Solar Energy 62(1), 19-28.
34 Esen, M., Yuksel, T., 2013. Experimental evaluation of using various renewable energy sources for
35 heating a greenhouse. Energy and Buildings 65, 340-351.
36 Huang, B., Mauerhofer, V., 2016. Life cycle sustainability assessment of ground source heat pump
37 in Shanghai, China. Journal of Cleaner Production 119, 207-214.
38 Kheradmand, M., Azenha, M., de Aguiar, J.L.B., Castro-Gomes, J., 2016. Experimental and numerical
39 studies of hybrid PCM embedded in plastering mortar for enhanced thermal behaviour of buildings.
40 Energy 94, 250-261.
41 Kheradmand, M., Castro-Gomes, J., Azenha, M., Silva, P.D., de Aguiar, J.L.B., Zoorob, S.E., 2015.
42 Assessing the feasibility of impregnating phase change materials in lightweight aggregate for
43 development of thermal energy storage systems. Construction and Building Materials 89, 48-59.
44 Kong, X., Lu, S., Huang, J., Cai, Z., Wei, S., 2013. Experimental research on the use of phase change

28
 ACCEPTED MANUSCRIPT

1 materials in perforated brick rooms for cooling storage. Energy and Buildings 62, 597-604.
2 Liu, Z., Yu, Z., Yang, T., El Mankibi, M., Roccamena, L., Sun, Y., Sun, P., Li, S., Zhang, G., 2019a.
3 Experimental and numerical study of a vertical earth-to-air heat exchanger system integrated with
4 annular phase change material. Energy Conversion and Management 186, 433-449.
5 Liu, Z., Yu, Z., Yang, T., Li, S., El Mankibi, M., Roccamena, L., Qin, D., Zhang, G., 2019b.
6 Experimental investigation of a vertical earth-to-air heat exchanger system. Energy Conversion and
7 Management 183, 241-251.
8 Liu, Z., Yu, Z., Yang, T., Li, S., Mankibi, M.E., Roccamena, L., Qin, D., Zhang, G., 2019c. Designing
9 and evaluating a new earth-to-air heat exchanger system in hot summer and cold winter areas. Energy
10 Procedia 158, 6087-6092.
11 Liu, Z., Yu, Z., Yang, T., Qin, D., Li, S., Zhang, G., Haghighat, F., Joybari, M.M., 2018. A review on
12 macro-encapsulated phase change material for building envelope applications. Building and
13 Environment 144, 281-294.
14 Liu, Z., Yu, Z., Yang, T., Roccamena, L., Sun, P., Li, S., Zhang, G., El Mankibi, M., 2019d. Numerical
15 modeling and parametric study of a vertical earth-to-air heat exchanger system. Energy 172, 220-231.
16 Mehdid, C.-E., Benchabane, A., Rouag, A., Moummi, N., Melhegueg, M.-A., Moummi, A., Benabdi,
17 M.-L., Brima, A., 2018. Thermal design of Earth-to-air heat exchanger. Part II a new transient
18 semi-analytical model and experimental validation for estimating air temperature. Journal of Cleaner
19 Production 198, 1536-1544.
20 Mi, X., Liu, R., Cui, H., Memon, S.A., Xing, F., Lo, Y., 2016. Energy and economic analysis of
21 building integrated with PCM in different cities of China. Applied Energy 175, 324-336.
22 Mustafa Omer, A., 2008. Ground-source heat pumps systems and applications. Renewable and
23 Sustainable Energy Reviews 12(2), 344-371.
24 Niu, F., Yu, Y., Yu, D., Li, H., 2015a. Heat and mass transfer performance analysis and cooling
25 capacity prediction of earth to air heat exchanger. Applied Energy 137, 211-221.
26 Niu, F., Yu, Y., Yu, D., Li, H., 2015b. Investigation on soil thermal saturation and recovery of an earth
27 to air heat exchanger under different operation strategies. Applied Thermal Engineering 77, 90-100.
28 Roccamena, L., 2017. Optimization of an innovative thermal energy storage technology at low
29 temperatures when coupled to multi-source energy architectures, ENTPE. Université de Lyon, pp.
30 90-94.
31 Rouag, A., Benchabane, A., Mehdid, C.-E., 2018. Thermal design of Earth-to-Air Heat Exchanger. Part
32 I a new transient semi-analytical model for determining soil temperature. Journal of Cleaner Production
33 182, 538-544.
34 Shojaee, S.M.N., Malek, K., 2017. Earth-to-air heat exchangers cooling evaluation for different
35 climates of Iran. Sustainable Energy Technologies and Assessments 23, 111-120.
36 Singh, R., Sawhney, R.L., Lazarus, I.J., Kishore, V.V.N., 2018. Recent advancements in earth air
37 tunnel heat exchanger (EATHE) system for indoor thermal comfort application: A review. Renewable
38 and Sustainable Energy Reviews 82, 2162-2185.
39 Soni, S.K., Pandey, M., Bartaria, V.N., 2016. Hybrid ground coupled heat exchanger systems for space
40 heating/cooling applications: A review. Renewable and Sustainable Energy Reviews 60, 724-738.
41 Stathopoulos, N., El Mankibi, M., Issoglio, R., Michel, P., Haghighat, F., 2016. Air–PCM heat
42 exchanger for peak load management: Experimental and simulation. Solar Energy 132, 453-466.
43 Tittelein, P., Achard, G., Wurtz, E., 2009. Modelling earth-to-air heat exchanger behaviour with the
44 convolutive response factors method. Applied Energy 86(9), 1683-1691.

29
 ACCEPTED MANUSCRIPT

1 Xi, J., Li, Y., Liu, M., Wang, R.Z., 2017. Study on the thermal effect of the ground heat exchanger of
2 GSHP in the eastern China area. Energy 141, 56-65.
3 Yu, L., Li, Y.P., Huang, G.H., 2019. Planning municipal-scale mixed energy system for stimulating
4 renewable energy under multiple uncertainties - The City of Qingdao in Shandong Province, China.
5 Energy 166, 1120-1133.
6 Zhang, H.L., Baeyens, J., Degrève, J., Cáceres, G., Segal, R., Pitié, F., 2014. Latent heat storage with
7 tubular-encapsulated phase change materials (PCMs). Energy 76, 66-72.
8 Zhang, X., Zeng, R., Mu, K., Liu, X., Sun, X., Li, H., 2019. Exergetic and exergoeconomic evaluation
9 of co-firing biomass gas with natural gas in CCHP system integrated with ground source heat pump.
10 Energy Conversion and Management 180, 622-640.
11 Zhou, S., Cui, W., Tao, J., Peng, Q., 2016. Study on ground temperature response of multilayer
12 stratums under operation of ground-source heat pump. Applied Thermal Engineering 101, 173-182.
13 Zhou, Y., Yu, C.W.F., Zhang, G., 2018. Study on heat-transfer mechanism of wallboards containing
14 active phase change material and parameter optimization with ventilation. Applied Thermal
15 Engineering 144, 1091-1108.
16 Zhou, Y., Zheng, S., Zhang, G., 2019. Study on the energy performance enhancement of a new PCMs
17 integrated hybrid system with the active cooling and hybrid ventilations. Energy 179, 111-128.
18 Zhou Y., Z.S., Zhang G., 2019. Artificial neural network based multivariable optimization of a hybrid
19 system integrated with phase change materials, active cooling and hybrid ventilations. Energy
20 Conversion and Management.
21

30