Content uploaded by Robin Deb
Author content
All content in this area was uploaded by Robin Deb on Jan 18, 2024
Content may be subject to copyright.
1
Development of Self-Heating Concrete using Low-Temperature Phase
1
Change Materials: Multi-scale and In-situ Real-Time Evaluation of
2
Snow-Melting and Freeze-thaw Performance
3
Robin Deb1, Nishant Shrestha2, Kham Phan2, Mohamed Cissao2, Parsa Namakiaraghi1, Yousif Alqenai1, Sharaniaya
4
Visvalingam1, Angela Mutua1, and Yaghoob (Amir) Farnam3
5
1 Ph.D. Candidate, Department of Civil, Architectural, and Environmental Engineering, Drexel University, 3141
6
Chestnut Street, Philadelphia, PA-19104, Email: rd673@drexel.edu.
7
2 BSc./MSc. Student, Department of Civil, Architectural, and Environmental Engineering, Drexel University, 3141
8
Chestnut Street, Philadelphia, PA-19104, USA, Email: ns3343@drexel.edu.
9
3 BSc./MSc. Student, Department of Civil, Architectural, and Environmental Engineering, Drexel University, 3141
10
Chestnut Street, Philadelphia, PA-19104, USA, Email: kvp36@drexel.edu.
11
4 BSc./MSc. Student, Department of Civil, Architectural, and Environmental Engineering, Drexel University, 3141
12
Chestnut Street, Philadelphia, PA-19104, USA, Email: mmc444@drexel.edu.
13
5 Ph.D. Candidate, Department of Civil, Architectural, and Environmental Engineering, Drexel University, 3141
14
Chestnut Street, Philadelphia, PA-19104, USA, Email: pn363@drexel.edu.
15
6 Ph.D. Candidate, Department of Civil, Architectural, and Environmental Engineering, Drexel University, 3141
16
Chestnut Street, Philadelphia, PA-19104, USA, Email: yma34@drexel.edu.
17
7 Ph.D. Candidate, Department of Civil, Architectural, and Environmental Engineering, Drexel University, 3141
18
Chestnut Street, Philadelphia, PA-19104, USA, Email: sv672@drexel.edu.
19
8 Ph.D. Candidate, Department of Civil, Architectural, and Environmental Engineering, Drexel University, 3141
20
Chestnut Street, Philadelphia, PA-19104, USA, Email: awm54@drexel.edu.
21
9Associate Professor, Department of Civil, Architectural, and Environmental Engineering, Drexel University, 3141
22
Chestnut St, Philadelphia, PA-19104, USA, Email: yf338@drexel.edu.
23
24
Abstract
25
This work examined the performance of self-heating concrete under laboratory thermal conditions and
26
outdoor real-time conditions during the fall and winter seasons. Snow-melting and freeze-thaw
27
performance of low-temperature phase change materials (PCM) incorporated self-heating concrete slabs
28
in various scales were evaluated. PCM exhibited high enthalpy of fusion (ΔHf ≈ 170-180 J/g), long-term
29
thermal stability, and desirable supercooling. The experimental program includes (i) optimization of
30
concrete mix designs for maximum PCM incorporation, (ii) characterization of thermal properties of
31
PCM-mortar specimens using longitudinal guarded comparative calorimetry (LGCC), and (iii) large-
32
scale PCM concrete slabs in outdoor conditions to evaluate the real-time thermal performance against
33
freeze-thaw events and snow-melting efficiency. Two different approaches were used to incorporate PCM
34
in concrete: (i) submersion of liquid PCM in porous lightweight aggregates (PCM-LWA) and (ii) micro-
35
encapsulated PCM (MPCM). Both PCM-LWA and MPCM concrete not only exhibit promising snow-
36
2
melting capabilities, but also lowered the number of freeze-thaw cycles during cold seasons. PCM-LWA
37
concrete performed better in decreasing the number of freeze-thaw (F-T) cycles due to the undercooling
38
phenomenon created by the LWA pore network confinement pressure, allowing gradual latent heat
39
release; the undercooling phenomenon in PCM-LWA results in phase transformation in a wider low-
40
temperature range (i.e., 3.94 oC to -13.04 oC). Therefore, the PCM-LWA concrete was effective in melting
41
snow within a wider range of low temperatures. MPCM concrete was found to provide a rapid melting
42
capability during a snowfall event due to its ‘one-shot’ heat release phenomenon. Both LWA-PCM and
43
MPCM concrete slabs demonstrate promising heat response and snow melting capability.
44
Keywords: freeze-thaw, lightweight aggregate, phase change materials, self-heating concrete, snow-
45
melting, undercooling.
46
Practical Applications
47
Snowfall and freeze-thaw cycles occur frequently during winter seasons in North America regions with
48
cold climate, resulting in snow accumulation on concrete roads and flatworks as well as concrete freeze-
49
thaw damage. In this paper, a ‘self-heating’ concrete was developed via incorporation of low-temperature
50
phase change material (PCM) and its promising snow removal and freeze-thaw improvements were
51
validated. The self-heating concrete can be used to construct pavements, driveway, bridge decks, and
52
any other types of flatworks. When the ambient temperature falls to ~0 oC, PCM will release desirable
53
amounts of heat energy (ΔHf =170-180 J per g of PCM added) by changing its phase from liquid to solid.
54
As a result, the accumulated snow and ice melts at a gradual pace. In addition, heat release from the
55
incorporated PCM lowers the number of freeze-thaw cycles improving freeze-thaw performance of
56
concrete made elements in cold regions, which in turn, improves concrete durability and service life by
57
minimizing the susceptibility to freeze-thaw scaling and spalling.
58
59
3
Introduction
60
Previous studies have shown the potential use of PCM in the construction industry for thermoregulation
61
of buildings; thereby, tackling rising costs of energy and reducing the impact of global warming (Rathod
62
and Banerjee 2013; Marani and Nehdi 2019; Asadi et al. 2022). In addition, several studies have
63
conducted numerical and experimental analyses at a multi-scale level to establish the possibility of
64
incorporating PCM in concrete pavements for deicing applications and curtailing F-T cycles(Bentz and
65
Turpin 2007; Sakulich and Bentz 2012; L. Liston et al. 2014b; Farnam et al. 2016b; 2017; Urgessa et al.
66
2019; Balapour, Mutua, and Farnam 2021; W. Li et al. 2019; Esmaeeli et al. 2018). The use of PCM for
67
low-temperature applications stems from the fact that (1) conventional snow removal methods (i.e., snow
68
plowing, or deicing salts) are not only energy intensive, but can also accelerate damage to concrete and
69
(2) it can help to lower F-T damage by lowering the numbers of freeze-thaw cycles(Smith et al. 2019).
70
There are mainly two major sources of damage in concrete in cold environments where PCM concrete
71
can be beneficial to alleviate both: (1) physical damage due to freeze-thaw exposure, and (2) chemical
72
damage due to deicing salt exposure. Physical freeze-thaw damage can be observed when concrete
73
reaches critical degree of saturation and experiences freezing and thawing cycles, therefore, physical
74
expansion due to ice formation generate internal pressure and causes micro-cracking formation and
75
growth(Y. Li et al. 2022; Farnam et al. 2014a; Valenza II and Scherer 2007; Scherer 1999).Chemical
76
damage can be observed when chloride-based salts, such as sodium chloride (NaCl), calcium chloride
77
(CaCl2), and magnesium chloride (MgCl2) are used (Farnam et al. 2015; Liu et al. 2021). These deicing
78
salts were found deleterious during freezing and thawing; they can react with by-products of the cement
79
hydration process (Althoey and Farnam 2019; Scrivener 2004), and form expansive compounds, such as
80
calcium oxychloride (CAOXY), and magnesium oxychloride (MGOXY), among other compounds
81
during freezing and thawing. These reactions happen when the concentration of the deicing salts,
82
diffusing through the capillary pores of the concrete, reaches a critical degree of saturation, resulting in
83
4
damage to concrete microstructure, scaling, spalling of concrete, and reinforcement corrosion (Farnam et
84
al. 2015; Smith et al. 2019).
85
The use of PCM in concrete for deicing and snow-melting applications can help lower the use of deicing
86
salts and increase the freeze-thaw service life of concrete pavement, bridges, and flatworks (Sakulich and
87
Bentz 2012; W. Li et al. 2019; Sharifi and Sakulich 2015; Balapour, Mutua, and Farnam 2021; Farnam
88
et al. 2017; 2016b; L. C. Liston et al. 2016). PCM comes in many different forms (i.e., organic, inorganic,
89
and eutectic) and can be tailored to suit specific applications. A unique characteristic of PCM is that it
90
can release/absorb large amounts of heat energy during its phase transition near a targeted phase change
91
temperature range for effective thermoregulation (Rathod and Banerjee 2013; Ling and Poon 2013;
92
Marani and Nehdi 2019; Nazir et al. 2019). For instance, organic PCM, such as paraffin (PCM6P),
93
exhibits ΔHf = 170-180 J/g between 3-6 oC (Microtek 2019). Paraffin-based PCM show minimal
94
supercooling effect, long-term thermal stability, non-reactive and non-corrosive to cementitious products;
95
these properties deem it desirable for incorporation in concrete for deicing applications (L. C. Liston et
96
al. 2016; Farnam et al. 2016a). Most recently, Farnam et al. (Farnam et al. 2017) performed an
97
experimental study using two types of PCM incorporated large-scale specimens in lab conditions: (i)
98
PCM infused in lightweight aggregates (LWA) concrete slab (PCM-LWA) and (ii) concrete slabs with
99
embedded tubes containing PCM (PCM-PIPE). The authors concluded that incorporated PCM in concrete
100
slabs exhibits a large release of latent heat during temperature decrease associated with PCM
101
solidification. Moreover, the heat from PCM-LWA concrete slab was gradual compared to the rapid heat
102
release over a narrow range of temperatures from the concrete slab with PCM-PIPE (Farnam et al. 2017).
103
Esmaeeli et al. did a further numerical analysis to explain this observation; the authors concluded that the
104
incorporation of PCM has a direct effect on its phase-changing properties, which can be explained using
105
the Gibbs-Thomson theory (Esmaeeli et al. 2018; Scalfi, Coasne, and Rotenberg 2021). Organics or ionic
106
fluids absorbed in small pores (rpores< 100 nm) can experience alterations in their phase change behavior
107
due to the high magnitude of confining pressure and exhibit supercooling/undercooling effect (Scalfi,
108
5
Coasne, and Rotenberg 2021; Wang et al. 2014). This was a critical issue that needed to be addressed
109
before evaluating the PCM-incorporated concrete slabs at outdoor conditions during winter seasons.
110
Deb et al. did a comprehensive study to evaluate the thermal properties of PCM in different incorporation
111
techniques (PCM-filled LWA and microcapsules). Pore-size characterization of LWA indicated that the
112
majority of pores (> 92%) are macropores (rpores > 17.3 nm). Using the Gibbs-Thomson’s theory, it was
113
concluded that the PCM with a melting temperature of 4.28 oC incorporated in LWA would exhibit phase
114
change between -5 oC and 4.28 oC due to pore confinement pressure. They further studied the
115
microcapsules of PCM (MPCM) with size range between 15-30 µm; authors found MPCM may exhibit
116
larger heat release within a narrow temperature range near ~4.28 oC because MPCM does not experience
117
considerable pore confinement pressure. Fig. 1 graphically illustrates the effect of pore pressure on
118
depressing the freezing temperature of PCM using Eq. (1) (Deb et al. 2024).
119
Tm-pores(r)=Tm-bulk-2Tm-bulk
γLs
⁄×vL
∆HLS
⁄×r
(1)
120
Fig. 1: The effect of pore size on the melting temperature of liquid PCM with melting temperature of 4.28 ℃ by
121
using the Gibbs-Thomson equation (Deb et al. 2024).
122
6
In summary, several researchers have reported that PCM incorporation in construction materials,
123
specifically for deicing applications, can be a viable alternative pathway to curtail the use of deicing salts
124
and integrate self-heating properties in concrete (Sakulich and Bentz 2012; Farnam et al. 2016a; 2017;
125
Balapour, Mutua, and Farnam 2021; Yeon and Kim 2018; W. Li et al. 2019; Sharifi and Sakulich 2015;
126
Urgessa et al. 2019). However, there is a lack of outdoor real-time evidence to validate self-heating
127
property in PCM concrete for deicing applications. This study aimed to address this issue by devising an
128
experimental plan to investigate the thermal response of PCM mortar and concrete composites in large-
129
scale concrete slabs in outdoor environmental conditions. Snow-melting capability and F-T resilience
130
were evaluated, and the real-time data was correlated to experimental observations made in controlled
131
laboratory-scale thermal and calorimetry experiments.
132
Experimental Program
133
Table 1 summarizes the experimental program and specimen preparation details developed for this study.
134
The following sections contain all the schematic information, material properties, specimens’ preparation
135
steps, and the discussion of results.
136
Table 1: Experimental and specimen preparation details.
137
Test Name and Standard
Specimen Types, Dimensions, and Curing
Age
Description
Determination of Heat Flow
of Solids Using Guarded-
Comparative-Longitudinal
Heat Flow Technique,
ASTM E1225 (ASTM 2014;
Farnam et al. 2014)
• 25.4 mm × 25.4 mm × 50.8 mm (1" × 1" ×
2").
• PCM-LWA (Field), PCM-LWA (Lab),
MPCM (Field), and MPCM (Lab).
Lab Specimens were cured for 28 days in
double-sealed plastic bags.
To measure heat flow to
characterize the thermal
properties of PCM mortar
specimens including heat
release during PCM
solidification.
Measurement of Freeze-
thaw Resistivity and Snow-
melting efficiency under
real-time conditions
• 762 mm × 762 mm × 203 mm (30" × 30" ×
8").
• PCM-LWA (Field Slab), MPCM (Field
Slab), and Reference (Field Slab).
• Field slab specimens were cast on
November17th, 2021 and sealed cured for
14 days at ambient conditions.
To investigate the efficiency
of PCM for pavement
construction and to evaluate
the snow-melting
performance and freeze-thaw
resilience of PCM concrete
during Fall and Winter
climates.
7
Materials
138
Type I Ordinary Portland Cement (OPC) with a Blaine fineness of 400.5 m2/kg, compliant with ASTM
139
C150 (ASTM Committee 2007), was used for the specimen preparation in this study; Table 2 outlines
140
the chemical and Bogue composition of the Type I OPC. PCM incorporation was done via two methods:
141
PCM-LWA (PCM infused in the porous structure of LWA) and MPCM. Liquid paraffin oil PCM6P (n-
142
tetradecane, C14H30, > 99 % purity) was used with enthalpy of fusion of ΔHf = 170-180 J/g during its
143
melting temperature of 4.28 oC, specific density/gravity = 0.76 and vapor pressure < 0.01 mm Hg
144
(Microtek 2019). An expanded shale oven dried LWA, manufactured by Arcosa®, was used with a
145
specific gravity of 1.55; 72-hour PCM absorption capacity, utilizing submersion followed by centrifuge
146
method (i.e., 1700~1800 revolutions per minute) (Miller et al. 2014), in-room temperature conditions
147
(i.e., 23 ± 2 oC), was measured at 14.55 ± 0.56 % (ASTM C128-15 2015). The LWA absorption was
148
achieved through soaking/submerging LWA in PCM for 72 hr. Centrifuge was only used to achieve
149
Saturated Surface Dry (SSD) condition. Previous studies have proved the effectiveness of using
150
submersion followed by centrifugal method to achieve high absorption capacity and saturated surface
151
dried state(Miller, Barrett, et al. 2014; Farnam et al. 2017). Although vacuum absorption helps to obtain
152
higher absorption capacity, it is not practical for large scale field applications. Soaking/submersion is
153
currently a common practice being implemented by ready mix concrete producers to produce lightweight
154
concrete structures. For MPCM, the liquid PCM encapsulated by melamine-formaldehyde resin;
155
combined specific density of 0.89 was achieved. Particle size distribution of microcapsules range between
156
15-30 µm, and approximately ~79.6 % of liquid PCM in its core. The MPCM powder was acquired
157
commercially from Microtek Laboratories® (Microtek 2019).
158
159
160
8
Table 2: Chemical composition, Bogue composition, and relevant properties of Type I OPC used in this study.
161
Item
Percentage by mass (%)
Properties
Silicon Dioxide (SiO2)
19.22
Air Content (%)
9.93
Aluminum Oxide (Al2O3)
5.92
Blaine (m2/kg)
400.5
Ferric Oxide (Fe2O3)
2.73
% Pass 325 Mesh
96.62
Calcium Oxide (CaO)
62.59
14-day C 1038 Expansion %
0.006
Magnesium Oxide (MgO)
2.76
% Autoclave Expansion
0.09
Loss on Ignition (LOI)
2.70
Sulfur Trioxide (SO3)
4.15
Compressive Strength
Sodium Oxide (Na2O)
0.34
1-Day (psi)
2872
Potassium Oxide (K2O)
1.04
3-Day (psi)
4033
Total Alkali
1.02
7-Day (psi)
4620
Insoluble Residue
0.24
28-Day (psi)
5613
Limestone
<=5.0
L color
59.14
Time of Set (minutes)
C3S (Bogue)
53.30
Initial
117
C2S (Bogue)
14.90
C3A (Bogue)
11.10
C4AF (Bogue)
8.30
162
Optimization of Concrete Mix Designs
163
The priority of the optimization was to incorporate the maximum amount of PCM in concrete in order to
164
achieve the highest amount of heat release during low-temperature conditions of snowfall and severe
165
freeze-thaw exposure conditions. Secondly, inclusion of microcapsules and lightweight aggregates
166
required an increase in cement paste content and superplasticizers to achieve desirable fresh properties of
167
concrete after incorporating PCM via two methods: (i) liquid PCM infused in LWA (PCM-LWA), and
168
(ii) MPCM. Three different types of concrete mixtures were designed: (i) Reference concrete (ACI-318
169
concrete mix design for pavement construction), (ii) PCM-LWA concrete (concrete incorporated with
170
PCM-LWA), and (iii) MPCM concrete (concrete incorporated with MPCM as a partial replacement of
171
fine aggregates). To optimize workability, specifically for PCM-LWA and MPCM concrete mixes, small
172
batches of concrete were prepared (≈2-3 kg) for the flow table test. Incremental ±5 ml adjustments were
173
made to admixtures dosage until ±100 % flow was achieved; nominal values for acceptable flowability
174
range between 0 to 150 % (ASTM C230 2010). For large-scale concrete preparation procedure, the
175
MPCM powder was first blended with the cement in dry condition. In a concrete rotary drum mixer, fine
176
9
aggregates were then blended with a portion of water to allow dry fine aggregates to reach saturation.
177
Afterward, blended MPCM and cement powder were added to the drum mixer followed by addition of
178
chemical admixtures and remainder of water. The entire blend was blended for 90 seconds, rested for 60
179
seconds and then blended for another 90 seconds before the concrete was cast in-place. For PCM-LWA,
180
it was simply treated as fine aggregates replacement and the same concrete procedure was followed as
181
described earlier. Furthermore, minimal leakage of PCM was observed during the mixing process; the
182
porous network of LWA keeps the liquid PCM via capillary suction in PCM-LWA while the capsule
183
shell prevents PCM leakage in MPCM. Afterward, several iterations of slump tests (ASTM
184
C143/C143M 2015) and air content tests (ASTM C231/C231M 2008) were performed by preparing
185
concrete batches (≈10-15 kg) until desired fresh properties were attained. Table 3 shows the final concrete
186
mix proportions, and fresh properties measured (i.e., unit weight, slump, and air content) of Reference,
187
PCM-LWA, and MPCM concrete mixes. Moreover, PCM-LWA and MPCM concrete mixes
188
demonstrated higher slump values after mixing due to the addition of high-dosage poly-carboxylate-based
189
high-range water reducing admixture. However, MPCM concrete showed a lower air content value (i.e.,
190
2.9 %); this can be attributed to the high-volume fraction of fine particles in the mix design, which hinders
191
entrained air bubble formation (Mehta and Monteiro 2014). In addition, the compressive strength values
192
after 28 days were 36.26 MPa, 14.16 MPa, and 40.23 MPa for Reference, MPCM, and PCM-LWA
193
concrete samples, respectively. Results show that the strength gain of PCM-LWA concrete was sufficient
194
to meet the strength criteria for paving applications (i.e., concrete strength requirement is 20.7 MPa after
195
28 days per Federal Highway Administration) (Zimmerman et al. 2016) . While MPCM concrete
196
compressive strength was lower by 60.98 % with respect to Reference samples and does not satisfy the
197
strength required, the PCM-LWA concrete showed very promising strength gain meeting the required
198
value (i.e., 10.95 % higher than Reference at 28 days, 94.34 % higher than Federal Highway
199
Administration standard). This indicated that it is the PCM incorporation method, not the quantity,
200
affecting strength.
201
10
Table 3: Concrete mix designs for Reference Slab, PCM-LWA Slab, and MPCM Slab.
202
Material
Specific
Gravity
Volume Fraction (0 to 1)
Reference
(kg/m3)
PCM-
LWA
(kg/m3)
MPCM
(kg/m3)
Reference
PCM-
LWA
MPCM
Cement (Type-I)
Water
CA (SSD)
FA (SSD)
LWA (OD)
PCM Oil
MPCM
Air Content
3
1
2.61
2.63
1.55
0.76
0.89
-
0.147
0.185
0.412
0.194
-
-
-
0.05
0.21
0.265
-
-
0.475
-
-
0.05
0.21
0.265
-
0.32
-
-
0.16
0.05
441
185
1075
510
-
-
-
-
631
265
-
-
736
110
-
-
631
265
-
838
-
-
139
-
Total
-
1
1
1
2211
1742
1873
w/c (by mass)
-
0.42
0.42
0.42
-
-
-
Fresh Unit Weight
(kg/m3)
2388
1791
1954
Air Content (%)
5.0
4.9
2.9
Slump (mm)
152.4
190.5
215.9
Air Entraining
Admixture, (ml/100
kg of cement)
230
230
230
Water-reducing
admixtures (ml/100
kg of cement)
150
(Low
range)
350
(High
range)
400
(High
range)
*CA (SSD) – Coarse Aggregate in Saturated Surface Dried condition
*FA (SSD) - Fine Aggregate in Saturated Surface Dried condition
*LWA (OD) – Lightweight Aggregate in Oven Dried Condition
203
Measuring Heat Flow of PCM Mortar Composites Using Longitudinal Guarded Comparative
204
Calorimeter
205
Measurement of heat flow using a longitudinal guarded comparative calorimeter (LGCC) was used to
206
evaluate the heat response of PCM incorporated in mortar or concrete specimens. This method is also
207
known as the cut-bar or divided-bar method (Farnam et al. 2016; Technique 2020a; Brütting et al. 2016).
208
25.4 mm × 25.4 mm × 50.8 mm (1" × 2" × 2") bar was prepared from 25.4 mm × 25.4 mm × 300 mm (1"
209
× 1" × 11.81") samples using a wet saw. The cut-bar specimen (i.e., PCM-LWA or MPCM) was stacked
210
between two reference bars of known thermal properties (i.e., pyroceram (Technique 2020b)), and placed
211
on a cold plate to apply temperature change to allow axial heat flow (Farnam et al. 2015; 2016b; Balapour,
212
Mutua, and Farnam 2021; Brütting et al. 2016).
213
11
Fig. 2(a) shows the test setup, and Fig. 2(b) shows the temperature profile gradient implemented for this
214
investigation (i.e., 4 oC/hour ramp rate). Type-T thermocouples [Copper(+)/Constantan(-)] (Don Dowell
215
2010) were placed at seven locations to measure the temperature changes under quasi-steady state
216
conditions (Technique 2020b).
217
218
Fig. 2: (a) Schematic diagram of the longitudinal guarded comparative calorimetry test setup (dotted lines represent
219
the placement of Type-T thermocouples for temperature measurement during axial heat flow); (b) Temperature
220
profile applied by the cold plate during the investigation.
221
Design and Preparation of Large-Scale Concrete Slabs for Outdoor Investigation
222
Three concrete slabs were designed for construction at an outdoor location in the City of Philadelphia,
223
PA, United States: (i) Reference slab, (ii) PCM-LWA slab, and (iii) MPCM slab. Before the concrete mix
224
commenced, wooden frameworks were prepared and installed into the ground. In addition, Styrofoam
225
thermal insulator panels were added at four sides to allow one-dimensional heat transfer (Farnam et al.
226
2017; Esmaeeli et al. 2018). The concrete slabs were exposed to the ground soil at the bottom side to
227
allow the incorporated PCM to recharge from the energy exchange from the bottom surface or solar
228
energy from the top surface method during ambient temperature rises, specifically during daytime (Lee
229
et al. 2010). Dimensions of each of the concrete slabs were 762 mm (length) × 762 mm (width) × 203
230
12
mm (height) (30" × 30" × 8"). These testing conditions and setup were designed to simulate one
231
directional heat flow (i.e., bottom to top) during sub-zero weather conditions of large-scale concrete
232
elements. Styrofoam panels on the sides prevent heat loss in x-x direction and allows the incorporated
233
PCM to direct the heat release towards the surface. Conducting this investigation will facilitate a clear
234
understanding of the thermal performance of ‘self-heating’ concrete during snowfall events.
235
To measure the temperature changes at different locations of the concrete slabs, five Type T
236
thermocouples were placed in each slab mold at a spacing of 50 mm (2"). Likewise, one Type T
237
thermocouple was exposed to the ambient air to measure temperature changes of the climate. The stripped
238
ends of the thermocouple sensors were coated with a protective layer to prevent chemical interactions
239
with the alkaline environment of concrete. A backup set of five additional thermocouples was installed
240
into each concrete slab at a spacing of 25.4 mm from the main set of thermocouples. Finally, the
241
thermocouples were connected to a data acquisition system (i.e, Datataker 85 (CAS Dataloggers)); the
242
system was configured to record one data point (i.e., temperature in Celsius) over sixteen channels every
243
three minutes. Fig. 3 illustrates the dimensions of the outdoor slabs, the A-A cross section of the concrete
244
slabs, the spacing of thermocouples (50 mm), and the depth of soil (75 mm) on which the concrete slabs
245
had been placed on.
246
13
247
Fig. 3: Dimensions of formwork for (a) Concrete Slab, (b) Section A-A illustrating the sensor placements and slab
248
depth in the soil, and boundary insulation of molds to allow one-dimensional heat flow, and (c) sectioned illustration
249
of the concrete slab for temperature sensors.
250
251
Fig. 4: Pictures of the experimental setup of the outdoor investigation during (a) daytime and (b) night-time.
252
14
The outdoor area was cleaned of vegetation, and wooden frameworks, thermocouples, Styrofoam, and
253
the data acquisition system had been set up before the construction process. Four timelapse cameras,
254
capable of capturing one frame per minute, were positioned at four different angles to record any evidence
255
of snow-melting during the Fall 2021 and Winter 2021-2022 seasons. Fig. 4(a) and Fig. 4(b) show the
256
experimental setup during daylight and night-time.
257
Results and Discussion
258
In the following sub-sections, detailed discussions of the analysis of the thermal responses of PCM mortar
259
composites, freeze-thaw performance, and snow melting capability of the PCM concrete slabs compared
260
to the Reference concrete slab have been outlined and discussed.
261
Thermal Response of PCM mortar and concrete composites
262
MPCM and PCM-LWA mortar specimens exhibit exothermic heat release during the cooling process; on
263
the contrary, the mortar specimens show endothermic heat absorption during the heating cycle, as shown
264
in Fig. 5(a) and Fig. 5(b). The heat signatures are associated with the phase transition temperature of the
265
PCM (~3-6 oC). For MPCM, Tonset, (i.e., onset temperature of phase change), was 5.68 oC and for PCM-
266
LWA, Tonset was 3.94 oC. In comparison to MPCM, the onset temperature for PCM-LWA was depressed
267
by 1.74 oC; this phenomenon can be attributed towards the pore confinement effect (Esmaeeli et al. 2018;
268
Farnam et al. 2017; Scalfi, Coasne, and Rotenberg 2021). Similarly, the Tend (i.e., end temperature of
269
phase change), for the exothermic heat release of MPCM and PCM-LWA was -11.95 oC and -13.04 oC
270
respectively. Pore confinement pressure allows PCM-LWA to demonstrate desirable supercooling which
271
can be beneficial for deicing and snow-melting at temperature below – 10 oC, Therefore, PCM-LWA
272
concrete was able to exhibit exothermic heat release between 4.28 oC to -5 oC whereas MPCM concrete
273
exhibited heat release over a narrow temperature range (i.e., ~ 4.28 oC) [Deb et al., under review]. To
274
determine the heat evolution, heat flow as a function of time was plotted, as shown in Fig. 6. The area
275
under the graph (qB-qT) can be integrated to determine the heat evolution during phase transition. Heat
276
15
release during PCM solidification was 149.24 J/g of PCM (i.e., 9.42 J/g of PCM-LWA mortar) and 175.89
277
J/g of PCM (i.e., 13.05 J/of MPCM mortar) for PCM-LWA and MPCM mortar specimens, respectively.
278
Although the PCM-LWA mortar specimen heat release value during PCM solidification is 17.86 % lower
279
compared to MPCM mortar, this phenomenon can be attributed to the pore confinement effect created by
280
various pore structures in LWA. As PCM is incorporated in a porous network of LWA, variable capillary
281
pressures of PCM liquid associated with variable pore sizes cause supercooling/undercooling effect
282
(Farnam et al. 2017; Esmaeeli et al. 2018; Deb et al. 2024). As a result, PCM in LWA pores demonstrate
283
gradual heat release over a colder temperature range below -10 oC (i.e., 3.94 oC to -13.04 oC within ~7.20
284
hours) in comparison to MPCM (i.e., 5.68 oC to -11.95 oC within ~4.41 hours). Overall, in accordance
285
with the confinement effect of LWA pores and experimental evidence of heat evolution at temperatures
286
lower than 0 oC, PCM-LWA concrete was more suited for deicing applications at sub-zero temperatures
287
due to its gradual heat release within wider range of temperature. On the contrary, MPCM concrete
288
demonstrated larger heat release at a narrow temperature range close to the phase change transition
289
temperature of PCM (4.28 oC). The finding shows MPCM concrete demonstrates rapid ‘one-shot’ heat
290
release phenomenon with faster snow-melting rate in a narrow temperature range while PCM-LWA
291
concrete demonstrates a gradual heat release over a wide temperature range.
292
293
16
Fig. 5: Heat flow and enthalpy of fusion as a function of temperature change (i.e., heating, and cooling cycles)
294
through the (a) MPCM specimen, and (b) PCM-LWA specimen.
295
296
Fig. 6: Heat Flow and enthalpy of fusion and temperature as a function of time (i.e., heating, and cooling cycles)
297
through the (a) MPCM specimen and (b) PCM-LWA specimen.
298
Evaluation of Freeze-thaw Performance of Outdoor Concrete Slabs
299
Fig. 7 shows the climate profiles and temperature changes from December 2021 to March 2022 in the
300
outdoor experimental field located in the city of Philadelphia, PA. The temperature data from all the
301
thermocouples were calibrated and maintained an accuracy of ±1 oC (Don Dowell 2010) and to remove
302
noise from the temperature data, a Savitzky-Golay smoothing filter with 25 points of window was used
303
(Press and Teukolsky 1990). To classify the freeze-thaw events, all the temperature drops below 0 oC
304
were identified by blue-shaded regions. The number of freeze-thaw events were found to be 1 during
305
December 2021, 18 during January 2022, 9 during February 2022, and 4 during March 2022. The
306
thermocouple sensors were assigned labels A to E based on their depth from the surface of concrete slab
307
(see Fig. 3(c)). To identify F-T cycles of each concrete section of the slabs, the temperature data was
308
superimposed with the air temperature data from each month on the same axes.
309
310
17
311
Fig. 7: Air temperature profiles in the outdoor experimental field located in the city of Philadelphia, PA during the
312
period of investigation: (a) December 2021, (b) January 2022, (c) February 2022, (d) March 2022 (this data was
313
collected using an air temperature sensor connected to a data acquisition system).
314
Fig. 8 shows the bar charts that juxtapose the number of freeze-thaw events in each concrete section of
315
the respective concrete slabs. The dotted line at each bar chart represents the number of F-T cycles
316
detected by the air temperature sensor; it should be noted that the air temperature sensor does not measure
317
other environmental parameters such as wind or solar radiation effects on concrete slab surface
318
temperature. Table 4 shows the percentage reduction of F-T cycles in each section of Reference, PCM-
319
LWA, and MPCM concrete slabs. The data was normalized in comparison to Section E (i.e., the top
320
section near the slab surface) of the Reference concrete slab. Overall, the subsequent sections of the
321
Reference concrete experienced the highest number of F-T cycles between December 2021 and March
322
2022; meanwhile, PCM-LWA and MPCM respective concrete sections exhibited substantial reductions
323
in F-T cycles, as shown in Fig. 8. PCM-LWA demonstrated the highest reduction of F-T cycles in each
324
18
subsequent section compared to the MPCM slab. It appears the gradual heat release during the F-T event
325
for PCM-LWA slab over an extensive range of temperature due to pore confinement phenomenon
326
(Esmaeeli et al. 2018; Farnam et al. 2017) resulted in a higher efficiency in terms of F-T reduction
327
compared to MPCM slab. In general, both PCM-LWA and MPCM concrete slabs exhibited desirable
328
exothermic heat release during freezing events and lowering F-T cycles; thereby, indicating the potential
329
use of PCM in concrete pavements to reduce F-T cycles and to improve the concrete durability in cold
330
environments (Bentz and Turpin 2007; L. Liston et al. 2014b; Farnam et al. 2016; 2017).
331
Table 4: Percentage Reduction of F-T cycles in each respective sections of concrete slabs (normalized with
332
respective to section E of Reference slab).
333
Percentage Reduction
A (%)
B (%)
C (%)
D (%)
E (%)
Reference
41.3
23.9
15.2
6.5
0
PCM-LWA
89.1
76.1
54.3
41.3
4.3
MPCM
56.5
41.3
30.4
15.2
2.2
19
334
Fig. 8: Bar chart representing the number of freeze-thaw cycles by each respective section of the concrete slabs as
335
defined in Fig. 3(c) of (a) December 2021, (b) January 2022, (c) February 2022, and (d) March 2022.
336
Evaluation of the Snow-Melting Performance of the Outdoor Concrete Slabs
337
Table 5 lists all the major events of snowfall during the investigation period (December 2021-March
338
2022) in Philadelphia, PA. The data was collected from the National Centers for Environmental
339
Information database (https://www.noaa.gov/). Five major snowfall events were evaluated to determine
340
the snow-melting performance of concrete slabs listed in Table 6 with their associated time-lapse videos.
341
Fig. 9 shows the first significant snowfall event (Event 1) starting January 7th, 2022; snow depth was 2.9
342
inches in Philadelphia County and the ambient temperature during the snowfall was 1.51 ±1.76 oC. The
343
snow shower started at 1.10 AM on January 7th, 2022, and continued for approximately ~8 hours (snowfall
344
duration has been highlighted by the blue box on the plot in Fig. 9(b)). To discuss the thermal response
345
20
of the concrete slabs, the temperature data from section D (shown in Fig. 3(c)) was obtained and the
346
average temperature differences between Reference, PCM-LWA, and MPCM slabs and air were
347
calculated as Dref-air, DPCM-LWA-air and DMPCM-air, respectively, as shown in Fig. 13.
348
349
Fig. 9: Jan 7-11 snowfall evaluation: (a) Image showing the snowfall accumulation on top of the outdoor slabs,(b)
350
temperature profiles of Air, Reference, PCM-LWA, and MPCM at slab Location D as defined in Fig. 3(c) during
351
the five-day period (blue shaded region indicates the duration of the snowfall), and (c), zoomed in temperature
352
profile plots of the during the snowfall.
353
21
Table 5: Snowfall data during the period of investigation (collected from the National Centers for Environmental
354
Information database).
355
Date
Average Ambient
Temperature
during snowfall
event (oC)
Snowfall
depth
(inches)
Ability of
Reference Slab
in Snow Melting
Ability of PCM-
LWA Slab in
Snow Melting
Ability of
MPCM Slab
in Snow
Melting
January 7th,
2022
1.51 ± 1.76
2.9
No Melting
Partial Snow
Melting
Partial Snow
Melting
January
16th,2022
1.05 ± 0.87
0.7
Not Applicable
Not Applicable
Not Applicable
January
28th, 2022
4.42 ± 3.02
1.7
No Melting
Complete Snow
Melting
Partial Snow
Melting
January
29th, 2022
2.15 ± 4.63
5.8
No Melting
No Melting
No Melting
February
4th, 2022
-2.03 ± 1.66
<0.1
Not Applicable
Not Applicable
Not Applicable
February
7th, 2022
-2.84 ± 2.01
<0.1
Not Applicable
Not Applicable
Not Applicable
February
13th, 2022
3.54 ± 1.75
0.7
No Melting
Complete Snow
Melting
Complete Snow
Melting
February
24th, 2022
-1.27 ± 1.02
<0.1
Not Applicable
Not Applicable
Not Applicable
356
Table 6: Video Links of the PCM concrete slabs melting the snow.
357
Event
Snowfall Date
Time-Lapse Video Link
1
January 7th, 2022
https://www.youtube.com/watch?v=og3104SJ7Q8
2
January 16th, 2022
https://www.youtube.com/watch?v=9sge9lIdwKA
3 and 4
January 28th and 29th, 2022
https://www.youtube.com/watch?v=Jp6WwIKOpCg
5
February 13th, 2022
https://www.youtube.com/watch?v=cNMak79BLiM
358
In accordance with the video evidence listed on Table 6, no snow melting was observed for the first 24
359
hours. Air temperature shown in Fig. 9(c) during that period was indicative evidence of the
360
ineffectiveness of PCM concrete slabs; the average air temperature after the snowfall (Tavg(air)) was 0.29
361
oC (Tmax(air) = 1.73 oC, Tmin(air) = -1.26 oC). In addition, after sunset, the air temperature gradually went
362
22
down to below -5 oC. Although the PCM incorporated concrete slabs maintained a positive temperature
363
difference in comparison to air for 8 hours during the snowfall (i.e., Fig. 13), evidence suggested that the
364
heat release was not sufficient to melt the snow completely due to significant heat transfer to the
365
environment before the snowfall event. In addition, the time lapse cameras, set to record the changes on
366
the top surface, could not detect the melting on the bottom layer of the snow. Therefore, the remaining
367
snow accumulated on the slabs turned to ice and remained unchanged until ~11 AM January 9th, 2022. In
368
the end, the snow melted on all concrete slabs due to a rise in environmental temperature.
369
The second major snowfall event (Event 2) started at 6:10 PM on January 16th, 2022 (snow depth: 0.7
370
inches, air temperature: 1.05 ± 0.87 oC). Fig. 10 shows the summary of the January 16th snowfall events
371
and its associated temperature profiles; specifically, this event was notable as the weather forecast
372
predicted the air temperature to increase above ~0 oC after the snowfall. It would have been beneficial as
373
the rise of the environmental temperature by a few degrees Celsius after snowfall would help to melt the
374
remainder of snow from the top surface and manifests the efficiency of PCM in melting snow. However,
375
a successful snowfall was disrupted by the sudden rainfall. As a result, any kind of snow-melting evidence
376
capture was not viable as the rain washed away the snow accumulated on the surface of the slabs.
377
23
378
Fig. 10: Jan 16-20 snowfalls evaluation: (a) Images showing the snowfall accumulation on top of the outdoor
379
slabs and the rainfall shortly afterward, (b) temperature profiles of Air, Reference, PCM-LWA, and MPCM at slab
380
Location D as defined in Fig. 3(c) during the five-day period (blue shaded region indicates the duration of the
381
snowfall), and (c) zoomed in temperature profile plots of the during the snowfall.
382
24
The third and four significant snowfall events occurred between January 28th and January 29th, 2022. Fig.
383
11 summarizes the snowfall events and its associated temperature profiles. The first snowfall (Event 3)
384
started at ~9 A.M and continued for six hours; the total snowfall depth was reported to be 1.7 inches and
385
the air temperature during that period was 4.42 ± 3.02 oC. The overview camera video evidence available
386
in Table 6 and Fig. 11(a) show that both PCM-LWA and MPCM slabs started to melt the accumulated
387
snow while no snow-melting was observed for reference slab. The whole melting process took
388
approximately 4~5 hours from the start of the snowfall, and the snow melting rate was calculated to be
389
0.34~0.43 inches/hour for PCM-LWA slab (i.e., snowmelt rate = (Depth of snowfall)/ (Duration of
390
snowmelt on the respective video). The snow-melting was found to be more evident on PCM-LWA slab
391
compared with MPCM slab where partial melting was observed. After a brief intermission, the snowfall
392
resumed after ~6 P.M (Event 4) on the same day and continued for an additional ~10 hours (snow depth:
393
5.8 inches, air temperature: 2.15 ± 4.63 oC). Within a few minutes, the air temperature dropped below ~0
394
oC and the snow accumulated on the slabs turned into ice. In this instance, no melting was observed by
395
the time lapse cameras during the snowfall. A logical reasoning behind it can be the ‘one-shot’
396
effectiveness of the PCM heat release; as the incorporated PCM in both PCM-LWA and MPCM slabs
397
were not able to recharge itself due to low ambient temperature (Taverage(air)= -2.83 oC, Tmax(air) = +5.64 oC,
398
Tmin(air) = -5.44 oC) during the brief intermission, the accumulated snow and ice did not melt away for the
399
next ~96 hours. It should be noted that PCM needs to recharge itself before the consequent
400
snowfall/freeze-thaw cycle during heating; otherwise, the incorporated PCM will not be able to
401
demonstrate self-heating mechanism. PCM recharge can be achieved when the ambient temperature
402
increases above melting temperature of PCM to store thermal energy through endothermic melting
403
process. Finally, when the environmental temperature increased, all the snow and ice melted an equivalent
404
rate on all the slabs (refer to the video listed on Table 6).
405
406
407
25
408
Fig. 11: Jan 27-31 snowfalls evaluation (a) Images showing the snowfall accumulation on top of the outdoor slabs
409
and the snow-melting events, (b) temperature profiles of Air, Reference, PCM-LWA, and MPCM at slab Location
410
D as defined in Fig. 3(c) during the five-day period (blue shaded region indicates the duration of the snowfall), and
411
(c) zoomed in temperature profile plots of the during the snowfall.
412
26
The fifth major snowfall (Event 5) event occurred on February 13th, 2022; the snowfall started at midnight
413
and continued for ~10 hours until noon (snow depth: 0.7 inches, air temperature: 3.54 ± 1.75 oC). Fig.
414
12(a) shows the image at the end of the snowfall event, Fig. 12(b) and Fig. 12(c) show the temperature
415
profiles of air, Reference, PCM-LWA, and MPCM. During this event, the MPCM and PCM-LWA slabs
416
melted the snow at an average snow melting rate of ~0.07 inches/hour. The video evidence listed in Table
417
6, as shown in Fig. 12(a) and show the melting events on both PCM slabs, whereas the snow remained
418
unchanged on the reference slab during that time period. MPCM slab released all latent heat energy over
419
a narrow temperature range, and therefore, it melted the total accumulated snow as soon as the snowfall
420
ended. Alternatively, the PCM-LWA slab gradually released the latent heat energy from phase transition,
421
due to the pore confinement effect (Scalfi, Coasne, and Rotenberg 2021; Farnam et al. 2017; Esmaeeli et
422
al. 2018); the snow melted away at an abated pace compared to the rate of melting on the MPCM slab.
423
The logical reasoning behind PCM slabs’ effectiveness in this instance is due to the elevated air
424
temperature during that period. Ambient temperature remained consistent above 0 oC (Taverage(air)= 2.60
425
oC, Tmax(air) = 13.3 oC, Tmin(air) = -3.44 oC), which allowed minimal heat loss to the environment due to low
426
temperature difference between the PCM slabs and ambient air. Hence, the PCM slabs can direct the heat
427
release towards the purpose of melting the snow on the surfaces of the slabs.
428
27
429
Fig. 12: Feb 12-16 snowfall evaluation (a) Images showing the snowfall accumulation on top of the outdoor slabs
430
and the snow-melting events, (b) temperature profiles of Air, Reference, PCM-LWA, and MPCM at slab Location
431
D as defined in Fig. 3(c) during the five-day period (blue shaded region indicates the duration of the snowfall), and
432
(c) zoomed in temperature profile plots of the during the snowfall.
433
28
Fig. 13 shows the ΔT (Tslab – Tair) of the five snowfall events. During the first snowfall event (Event 1),
434
average ref-air, LWA-air and MPCM-air, was +1.27 oC, +2.56 oC, and +1.35 oC, respectively. These
435
differences show that the PCM concrete slabs maintained a higher temperature difference with air in
436
comparison to the reference slab. However, the snowfall depth was 5.8 inches (i.e., heavy snow), and a
437
significant heat loss might had occurred due to low ambient air temperature before the event i.e., below
438
0 oC. Therefore, the heat released by PCM was not sufficient to melt the snow in a complete manner (refer
439
to the video on Table 6). As mentioned above, the test conditions and setup had a direct correlation
440
between the thermal performance of the ‘self-heating’ concrete slabs; placement of low thermally
441
conductive Styrofoam panels allowed uni-directional heat flow (i.e., bottom to top). As a result, the
442
temperature fluxes and deicing mechanisms can be assumed accurate.
443
During the second snowfall event (Event 2), average ref-air, LWA-air and MPCM-air was -1.08 oC, +0.21 oC,
444
and -1.29 oC, respectively. While the PCM-LWA maintained a positive difference, the reference and the
445
MPCM slab showed negative temperature difference during the snowfall. This can be attributed towards
446
the sudden increase of air temperature during the snowfall which resulted in rain. As a result, no snow
447
melting was observed during that period due to sudden disruption by the rainfall (refer to the video on
448
Table 6).
449
The third and fourth snowfall events (Event 3 and Event 4) were successful events for understanding the
450
effectiveness of PCM slabs for snow melting. For instance, during the third event (i.e., snowfall depth
451
1.7 inches), average ref-air, LWA-air and MPCM-air was -0.59 oC, +2.99 oC, and +1.66 oC, respectively; while
452
the reference slab maintained a negative temperature difference, both PCM-LWA and MPCM slabs
453
maintained a positive temperature difference during that period. Therefore, video evidence showed both
454
PCM slabs were melting the snow. Specifically, the PCM-LWA slab was able to melt the snow in
455
complete manner whereas the MPCM slab melted the snow partially. On the contrary, during the fourth
456
snowfall event, average ref-air, LWA-air and MPCM-air was -3.08 oC, -1.51 oC, and -2.76 oC, respectively; all
457
29
the slabs maintained negative temperature differences during the snowfall (i.e., snowfall depth 5.8
458
inches). Since there was a brief break between the third and fourth snowfall events, the incorporated PCM
459
was not able to recharge itself from the environment as the air temperature remained below 0 oC during
460
that period. As a result, the snow remained unchanged on the surfaces of the slabs for the next ~96 hours.
461
In the end, the snow melted away due to a rise in environmental temperature (refer to the video on Table
462
6).
463
The fifth snowfall event (Event 5) (i.e., snowfall depth 0.7 inches) was another significant addition to the
464
success of the PCM slabs. Both MPCM and PCM-LWA slabs exhibited complete snow melting during
465
that six-hour period. Average ref-air, LWA-air and MPCM-air was +0.41 oC, +1.94 oC, and +0.56 oC,
466
respectively. Although the ΔT (Tslab – Tair) for the MPCM slab was lower in comparison to that of PCM-
467
LWA slab, MPCM slab exhibited higher rate of melting. This observation is in clear coherence with the
468
experimental evidence of the LGCC results shown in Fig. 6; MPCM releases the latent heat energy over
469
a smaller temperature range which allows rapid heat release, whereas the PCM-LWA exhibits gradual
470
heat release over a larger temperature range (refer to the video on Table 6).
471
472
Fig. 13: ΔT (Tslab-Tair) of all snowfall events.
473
30
In conclusion, the five snowfall events act as different scenarios in which the PCM incorporated concrete
474
slabs demonstrated their effectiveness. Based on the observations discussed here, factors that were found
475
to affect the snow-melting efficiency are as follows:
476
(1) Ambient temperature before the snowfall event: Evidence suggests that if the temperature before the
477
snowfall event remains near 0 oC, PCM loses its effectiveness over time. As a result, the PCM
478
incorporated slabs will not be able to melt the accumulated snow. Incidentally, if the snowfall occurs as
479
soon as the temperature decreases from ~5 oC, PCM incorporated slabs are effective for melting the snow.
480
During the successful melting events (i.e., Event 2 and Event 5), ambient temperatures during that period
481
were 4.42 oC and 3.54 oC. Therefore, PCM heat transfer was more directed towards the top surface and
482
melted the accumulated snow rather than loss of heat to the environment.
483
(2) Duration and the depth of snowfall: Two successful melting events were observed where the snowfall
484
depth was 0.7 inches and 1.7 inches (i.e., light snow); both snowfall duration was between 6 to 10 hours.
485
On the other hand, during the two unsuccessful events, snow depth was 2.9 inches and 5.8 inches (i.e.,
486
heavy snow); in addition, the snowfall duration at both instances was larger than 10 hours. It becomes
487
difficult to conclude the effectiveness of PCM in longer durations as higher amount of snow and durations
488
can prevent visual observations. Quantification of molten snow in future research may be more beneficial
489
to evaluate longer snow events with higher amount of snow.
490
(3) Wind speed: Wind movement is a critical factor before and during the snowfall event. Mathematical
491
modeling using heat transfer equations reported in (Kim, Ban, and Park 2020; Jürges 1924) shows thermal
492
evolution between concrete surface and air is dependent on the wind speed. While this paper did not study
493
wind speed, its effect on the thermal response of PCM incorporated concrete may be considered for future
494
large-scale investigation.
495
(4) Temperature changes before snowfall: To allow PCM to recharge itself via an endothermic phase
496
transition from solid to liquid, ambient temperature needs to remain above ~5 oC for enough time before
497
31
snowfall event. Therefore, the climactic conditions and ambient temperature variations can significantly
498
influence the effectiveness of the PCM slabs.
499
(5) Influence of PCM incorporation mechanisms on thermal performance: PCM-LWA concrete was
500
found to provide superior capabilities in freeze-thaw, snow-melting and deicing, and mechanical
501
performances in comparison to MPCM concrete. MPCM concrete showed considerable strength
502
reduction and low thermal response due to one-shot heat release which was only effective in a narrow
503
range of temperature. Previous studies have shown the microcapsules form poor interfacial transition
504
zones with cement paste (Qiu et al. 2017), low intrinsic strength of encapsulating polymeric material
505
(Zhang et al. 2020), and loss of liquid PCM due to shell breakage during the mixing process (Azimi
506
Yancheshme et al. 2020). Therefore, the limitations and the thermal performance of MPCM concrete in
507
outdoor conditions deem it unsuitable for infrastructure applications.
508
Summary and Conclusions
509
This paper evaluated the thermal responses of mortar cementitious composites and large-scale concrete
510
slabs containing PCM; it further correlated the real-time outdoor data made in outdoor experiments with
511
controlled thermal and calorimetry data obtained in the lab environment. Two different PCM
512
incorporation techniques were used: PCM infused in LWA pores and MPCM (as a partial replacement of
513
fine aggregates). PCMs selected for deicing applications exhibit large enthalpy of fusion (ΔHf ≈ 170-180
514
J per gram of PCM) near 3-6 oC, have long-term thermal stability, and demonstrate desirable
515
supercooling.
516
For a multi-scale thermal evaluation, heat flow and energy exchange were assessed using LGCC on PCM-
517
LWA and MPCM mortar specimens during cooling and heating cycles. Both PCM-LWA and MPCM lab
518
mortar specimens exhibited exothermic heat release during LGCC experiment between 3.94 oC and -
519
13.04 oC and between 5.68 oC and -11.95 oC, respectively. Heat evolution analysis indicated during PCM
520
solidification indicated that enthalpy of fusion for PCM-LWA and MPCM mortar composites were
521
32
149.24 J/ (g of PCM) (within timespan of ~ 7.20 hours) and 175.89 J/(g of PCM) (within timespan of ~
522
4.41 hours), respectively. Due to pore confinement effect, PCM-LWA mortar specimen exhibited 17.86
523
% lower heat release in comparison to MPCM; on the contrary, PCM-LWA mortar specimen exhibited
524
gradual heat release over an extended period of time and colder temperature ranges which makes PCM-
525
LWA a more desirable candidate for deicing applications and F-T resilience.
526
The thermal performance of PCM concrete slabs at an outdoor scale was evaluated. The outdoor slabs
527
were exposed to F-T events from December 2021 to March 2022 in the city of Philadelphia, PA. In
528
comparison to the Reference concrete slab, PCM-LWA and MPCM demonstrated significant reductions
529
in F-T cycles within their depth. Since PCM exhibits heat release near its phase change temperature (i.e.,
530
near 3-6 oC), PCM-LWA and MPCM concrete slabs were able to maintain temperatures above 0 oC within
531
their depth during multiple cooling events. Moreover, the pore confinement phenomenon in LWA
532
allowed the incorporated PCM to achieve desirable supercooling effect; as a result, the PCM-LWA
533
concrete slab-maintained temperatures above ~0 oC for a longer period of time in comparison to the
534
MPCM concrete slab. It was found that the thermal performance of the PCM incorporated concrete slabs
535
is dependent on the ambient temperature during the snowfall period. During the two events where the
536
slabs were able to fully melt the snow, the average air temperatures were 4.42 oC and 3.54 oC which are
537
near melting temperature of PCM used in this study. Regarding the snow-melting efficiency, both PCM-
538
LWA and MPCM concrete slabs achieved 50 % success rate over the five significant snowfall events
539
occurred during the outdoor experiment. Snow-melting rate in some snow events reached a high value of
540
0.43 inches/hour in MPCM concrete, while the PCM-LWA concrete showed a melting rate of 0.34
541
inches/hour. The snow-melting melting rate was found to be dependent on the incorporation technique of
542
PCM, snow-accumulation rate, ambient climate temperature before and during the snowfall, and the
543
ambient temperature change after the snowfall has ended. Overall, during successful snow-melting
544
events, PCM maintained heat release for deicing and snow-melting for ~10 hours.
545
33
This work recommends that PCM can be considered as an ideal candidate for F-T reduction, deicing and
546
snow-melting applications in concrete pavement, sidewalks, bridges, or flatworks. While PCM
547
incorporation either by LWA method or capsulation technique was found promising, further research can
548
be done to characterize the efficacy and long-term thermal stability of PCM concrete slabs.
549
Acknowledgements
550
The authors acknowledge the financial support from Compass Minerals®, United States. The authors
551
would also like to extend their appreciation to MicroTek Laboratories® for providing the materials for
552
research purposes. Any findings, opinions, and conclusions or recommendations expressed in this paper
553
are those of the authors and do not necessarily reflect the views of other affiliations. The experiments
554
reported in this paper were conducted in the Advanced Infrastructure Materials (AIM) Lab at Drexel
555
University. The authors acknowledge the support that has made this laboratory and its operation possible.
556
Data Availability Statement
557
All data, models, and code generated or used during the study appear in the submitted article.
558
References
559
Althoey, Fadi, and Yaghoob Farnam. 2019. “The Effect of Using Supplementary Cementitious Materials
560
on Damage Development Due to the Formation of a Chemical Phase Change in Cementitious
561
Materials Exposed to Sodium Chloride.” Construction and Building Materials 210: 685–95.
562
https://doi.org/https://doi.org/10.1016/j.conbuildmat.2019.03.230.
563
Asadi, Iman, Mohammad Hajmohammadian Baghban, Mohammad Hashemi, Nima Izadyar, and Behrang
564
Sajadi. 2022. “Phase Change Materials Incorporated into Geopolymer Concrete for Enhancing
565
Energy Efficiency and Sustainability of Buildings: A Review.” Case Studies in Construction
566
Materials, e01162.
567
ASTM. 2014. “Standard Test Method for Thermal Conductivity of Solids Using the Guarded-” i: 1–10.
568
https://doi.org/10.1520/E1225-13.2.
569
ASTM C128-15. 2015. “Standard Test Method for Relative Density (Specific Gravity) and Absorption
570
of Fine Aggregate.” Annual Book of ASTM Standards i: 1–5. https://doi.org/10.1520/C0128-15.2.
571
ASTM C143/C143M. 2015. “Standard Test Method for Slump of Hydraulic-Cement Concrete.” Astm
572
C143 04 (1): 1–4. https://doi.org/10.1520/C0143.
573
ASTM C230. 2010. “Standard Specification for Flow Table for Use in Tests of Hydraulic Cement 1.”
574
Annual Book of ASTM Standards, 4–9. https://doi.org/10.1520/C0230.
575
34
ASTM C231/C231M. 2008. “Standard Test Method for Air Content of Freshly Mixed Concrete by the
576
Pressure Method.” Manual on Hydrocarbon Analysis, 6th Edition, 545-545–3.
577
https://doi.org/10.1520/mnl10913m.
578
ASTM Committee. 2007. “Designation: C 150-07 Standard Specification for Portland Cement 1.”
579
www.astm.org.
580
Azimi Yancheshme, Amir, Anahita Allahdini, Khosrow Maghsoudi, Reza Jafari, and Gelareh Momen.
581
2020. “Potential Anti-Icing Applications of Encapsulated Phase Change Material–Embedded
582
Coatings; a Review.” Journal of Energy Storage 31 (April): 101638.
583
https://doi.org/10.1016/j.est.2020.101638.
584
Balapour, Mohammad, Angela W Mutua, and Yaghoob Farnam. 2021. “Evaluating the Thermal
585
Efficiency of Microencapsulated Phase Change Materials for Thermal Energy Storage in
586
Cementitious Composites.” Cement and Concrete Composites 116: 103891.
587
https://doi.org/https://doi.org/10.1016/j.cemconcomp.2020.103891.
588
Bentz, Dale P, and Randy Turpin. 2007. “Potential Applications of Phase Change Materials in Concrete
589
Technology.” Cement and Concrete Composites 29 (7): 527–32.
590
Brütting, M., F. Hemberger, S. Vidi, J. Wachtel, H. Mehling, and H. P. Ebert. 2016. “Determination of
591
Heat Capacity by Means of Longitudinal Guarded Comparative Calorimeter - Correction Methods.”
592
International Journal of Thermal Sciences 100: 423–29.
593
https://doi.org/10.1016/j.ijthermalsci.2015.10.022.
594
CAS. “CAS Dataloggers Data Taker DT 85.” https://dataloggerinc.com/product/dt85-universal-input-
595
data-logger/.
596
Deb, Robin, Jialuo He, Geetika Mishra, and Yaghoob (Amir) Farnam. 2024. “Investigating Temperature
597
Change Rate and Pore Confinement Effect on Thermal Properties of Phase Change Materials for
598
De-Icing and Low-Temperature Applications in Cementitious Composites.” Construction and
599
Building Materials 411: 134237.
600
https://doi.org/https://doi.org/10.1016/j.conbuildmat.2023.134237.
601
Don Dowell, D. L. 2010. “A Critical Look at Type T Thermocouples in Low-Temperature Measurement
602
Applications.” International Journal of Thermophysics 31 (8–9): 1527–32.
603
https://doi.org/10.1007/s10765-010-0780-2.
604
Esmaeeli, Hadi S, Yaghoob Farnam, John E Haddock, Pablo D Zavattieri, and W Jason Weiss. 2018.
605
“Numerical Analysis of the Freeze-Thaw Performance of Cementitious Composites That Contain
606
Phase Change Material (PCM).” Materials & Design 145: 74–87.
607
https://doi.org/https://doi.org/10.1016/j.matdes.2018.02.056.
608
Farnam, Yaghoob, Dale Bentz, Aaron Sakulich, Daniel Flynn, and Jason Weiss. 2014a. “Measuring
609
Freeze and Thaw Damage in Mortars Containing Deicing Salt Using a Low-Temperature
610
Longitudinal Guarded Comparative Calorimeter and Acoustic Emission.” Advances in Civil
611
Engineering Materials 3 (1): 316–37.
612
2014b. “Measuring Freeze and Thaw Damage in Mortars Containing Deicing Salt Using a Low-
613
Temperature Longitudinal Guarded Comparative Calorimeter and Acoustic Emission.” Advances
614
in Civil Engineering Materials 3 (1): 20130095. https://doi.org/10.1520/acem20130095.
615
Farnam, Yaghoob, Sarah Dick, Andrew Wiese, Jeffrey Davis, Dale Bentz, and Jason Weiss. 2015. “The
616
Influence of Calcium Chloride Deicing Salt on Phase Changes and Damage Development in
617
35
Cementitious Materials.” Cement and Concrete Composites 64: 1–15.
618
https://doi.org/https://doi.org/10.1016/j.cemconcomp.2015.09.006.
619
Farnam, Yaghoob, Hadi S. Esmaeeli, Pablo D. Zavattieri, John Haddock, and Jason Weiss. 2017.
620
“Incorporating Phase Change Materials in Concrete Pavement to Melt Snow and Ice.” Cement and
621
Concrete Composites 84: 134–45. https://doi.org/10.1016/j.cemconcomp.2017.09.002.
622
Farnam, Yaghoob, Matthew Krafcik, Leah Liston, Taylor Washington, Kendra Erk, Bernard Tao, and
623
Jason Weiss. 2016a. “Evaluating the Use of Phase Change Materials in Concrete Pavement to Melt
624
Ice and Snow.” Journal of Materials in Civil Engineering 28 (4): 04015161.
625
https://doi.org/10.1061/(asce)mt.1943-5533.0001439.
626
2016b. “Evaluating the Use of Phase Change Materials in Concrete Pavement to Melt Ice and Snow.”
627
Journal of Materials in Civil Engineering 28 (4): 04015161. https://doi.org/10.1061/(asce)mt.1943-
628
5533.0001439.
629
Jürges, W. 1924. “The Heat Transfer at a Flat Wall (Der Wärmeübergang an Einer Ebenen Wand),
630
Beihefte Zum Gesundh.” Ing 1: 19.
631
Kim, Hee Su, Hoki Ban, and Won Jun Park. 2020. “Deicing Concrete Pavements and Roads with Carbon
632
Nanotubes (CNTs) as Heating Elements.” Materials 13 (11). https://doi.org/10.3390/ma13112504.
633
Lee, Seok-Jin, Bong-Chan Kim, Un-Jong Seo, Seung-Ha Lee, and Joo-Ho Lee. 2010. “The Thermal
634
Conductivity Analysis on the Pavement Applying Geothermal Snow Melting System.” In
635
Proceedings of the Korean Geotechical Society Conference, 221–28. Korean Geotechnical Society.
636
Li, Wenting, Chaowei Ling, Zhengwu Jiang, and Qian-qian Yu. 2019. “Evaluation of the Potential Use
637
of Form-Stable Phase Change Materials to Improve the Freeze-Thaw Resistance of Concrete.”
638
Construction and Building Materials 203: 621–32.
639
Li, Yang, Yu Zhou, Ruijun Wang, Yanlong Li, Xiangjing Wu, and Zheng Si. 2022. “Experimental
640
Investigation on the Properties of the Interface between RCC Layers Subjected to Early-Age Frost
641
Damage.” Cement and Concrete Composites 134: 104745.
642
Ling, Tung Chai, and Chi Sun Poon. 2013. “Use of Phase Change Materials for Thermal Energy Storage
643
in Concrete: An Overview.” Construction and Building Materials 46: 55–62.
644
https://doi.org/10.1016/j.conbuildmat.2013.04.031.
645
Liston, Leah C., Yaghoob Farnam, Matthew Krafcik, Jason Weiss, Kendra Erk, and Bernard Y. Tao.
646
2016. “Binary Mixtures of Fatty Acid Methyl Esters as Phase Change Materials for Low
647
Temperature Applications.” Applied Thermal Engineering 96: 501–7.
648
https://doi.org/10.1016/j.applthermaleng.2015.11.007.
649
Liston, Leah, Matthew Krafcik, Yaghoob Farnam, Bernard Tao, Kendra Erk, and Jason Weiss. 2014.
650
“Toward the Use of Phase Change Materials (PCM) in Concrete Pavements: Evaluation of Thermal
651
Properties of PCM.”
652
Liu, Jun, Weizhuo Zhang, Zhenlin Li, Hesong Jin, and Luping Tang. 2021. “Influence of Deicing Salt on
653
the Surface Properties of Concrete Specimens after 20 Years.” Construction and Building Materials
654
295: 123643.
655
Marani, Afshin, and Moncef L Nehdi. 2019. “Integrating Phase Change Materials in Construction
656
Materials: Critical Review.” Construction and Building Materials 217: 36–49.
657
Mehta, P Kumar, and Paulo J M Monteiro. 2014. Concrete: Microstructure, Properties, and Materials.
658
McGraw-Hill Education.
659
36
Microtek. 2019. “PCM 6 Data Sheet.”
660
Miller, A, T Barrett, A Zander, and W Weiss. 2014. “Using a Centrifuge to Determine Moisture Properties
661
of Lightweight Fine Aggregate for Use in Internal Curing.” Advances in Civil Engineering
662
Materials 3 (1): 142–57.
663
Miller, A., R. Spragg, F. C. Antico, W. Ashraf, T. Barrett, A. Behnood, Y. Bu, et al. 2014. “Determining
664
the Moisture Content of Pre-Wetted Lightweight Aggregate: Assessing the Variability of the Paper
665
Towel and Centrifuge Methods.” Proceedings of the 4th International Conference on the Durability
666
of Concrete Structures, ICDCS 2014, no. July: 312–16. https://doi.org/10.5703/1288284315475.
667
Nazir, Hassan, Mariah Batool, Francisco J Bolivar Osorio, Marllory Isaza-Ruiz, Xinhai Xu, K
668
Vignarooban, Patrick Phelan, and Arunachala M Kannan. 2019. “Recent Developments in Phase
669
Change Materials for Energy Storage Applications: A Review.” International Journal of Heat and
670
Mass Transfer 129: 491–523.
671
Press, William H, and Saul A Teukolsky. 1990. “Savitzky‐Golay Smoothing Filters.” Computers in
672
Physics 4 (6): 669–72.
673
Qiu, Zhongzhu, Xiaoli Ma, Peng Li, Xudong Zhao, and Andrew Wright. 2017. “Micro-Encapsulated
674
Phase Change Material (MPCM) Slurries: Characterization and Building Applications.” Renewable
675
and Sustainable Energy Reviews. https://doi.org/10.1016/j.rser.2017.04.001.
676
Rathod, Manish K., and Jyotirmay Banerjee. 2013. “Thermal Stability of Phase Change Materials Used
677
in Latent Heat Energy Storage Systems: A Review.” Renewable and Sustainable Energy Reviews
678
18: 246–58. https://doi.org/10.1016/j.rser.2012.10.022.
679
Sakulich, A. R., and D. P. Bentz. 2012. “Incorporation of Phase Change Materials in Cementitious
680
Systems via Fine Lightweight Aggregate.” Construction and Building Materials 35: 483–90.
681
https://doi.org/10.1016/j.conbuildmat.2012.04.042.
682
Scalfi, Laura, Benoît Coasne, and Benjamin Rotenberg. 2021. “On the Gibbs-Thomson Equation for the
683
Crystallization of Confined Fluids.” Journal of Chemical Physics 154 (11).
684
https://doi.org/10.1063/5.0044330.
685
Scherer, George W. 1999. “Crystallization in Pores.” Cement and Concrete Research 29 (8): 1347–58.
686
Scrivener, Karen L. 2004. “Backscattered Electron Imaging of Cementitious Microstructures:
687
Understanding and Quantification.” Cement and Concrete Composites 26 (8): 935–45.
688
https://doi.org/https://doi.org/10.1016/j.cemconcomp.2004.02.029.
689
Sharifi, Naser P, and Aaron Sakulich. 2015. “Application of Phase Change Materials to Improve the
690
Thermal Performance of Cementitious Material.” Energy and Buildings 103: 83–95.
691
Smith, Scott H., Chunyu Qiao, Prannoy Suraneni, Kimberly E. Kurtis, and William J. Weiss. 2019.
692
“Service-Life of Concrete in Freeze-Thaw Environments: Critical Degree of Saturation and
693
Calcium Oxychloride Formation.” Cement and Concrete Research. Elsevier Ltd.
694
https://doi.org/10.1016/j.cemconres.2019.04.014.
695
Technique, Comparative-longitudinal Heat Flow. 2020a. “Standard Test Method for Thermal
696
Conductivity of Solids Using the Guarded-” i. https://doi.org/10.1520/E1225-20.2.
697
2020b. “Standard Test Method for Thermal Conductivity of Solids Using the Guarded-” i.
698
https://doi.org/10.1520/E1225-20.2.
699
37
Urgessa, Girum, Kyong-Ku Yun, Jaeheum Yeon, and Jung Heum Yeon. 2019. “Thermal Responses of
700
Concrete Slabs Containing Microencapsulated Low-Transition Temperature Phase Change
701
Materials Exposed to Realistic Climate Conditions.” Cement and Concrete Composites 104:
702
103391.
703
Valenza II, John J, and George W Scherer. 2007. “A Review of Salt Scaling: I. Phenomenology.” Cement
704
and Concrete Research 37 (7): 1007–21.
705
Wang, L P, T B Wang, C F Gao, X Lan, and X Z Lan. 2014. “Phase Behavior of Dodecane–Hexadecane
706
Mixtures in Bulk and Confined in SBA-15.” Journal of Thermal Analysis and Calorimetry 116 (1):
707
469–76. https://doi.org/10.1007/s10973-013-3525-1.
708
Yeon, Jung Heum, and Kwan-Kyu Kim. 2018. “Potential Applications of Phase Change Materials to
709
Mitigate Freeze-Thaw Deteriorations in Concrete Pavement.” Construction and Building Materials
710
177: 202–9.
711
Zhang, Baolian, Shanshan Li, Xuening Fei, Hongbin Zhao, and Xuanye Lou. 2020. “Enhanced
712
Mechanical Properties and Thermal Conductivity of Paraffin Microcapsules Shelled by
713
Hydrophobic-Silicon Carbide Modified Melamine-Formaldehyde Resin.” Colloids and Surfaces A:
714
Physicochemical and Engineering Aspects 603: 125219.
715
https://doi.org/https://doi.org/10.1016/j.colsurfa.2020.125219.
716
Zimmerman, K, B Allen, P Ram, G Duncan, Omar Smadi, K Smith, K Manda, and Başak Aldemir Bektaş.
717
2016. “Identification of Effective Next Generation Pavement Performance Measures and Asset
718
Management Methodologies to Support MAP-21 Performance Management Requirements.”
719
720
