Heat Transfer Research 53(9):1–27 (2022)

EXISTENCE OF SUPERCRITICAL “LIQUID-LIKE” STATE

IN SUBCRITICAL REGION, OPTIMAL HEAT TRANSFER
ENHANCEMENT, AND ARGON AS A NONREACTING,
NONCORRODING SC HEAT TRANSFER FLUID
Vish Prasad,1,* Karan Kakroo,1 & Debjyoti Banerjee2
1
Department of Mechanical Engineering, University of North Texas, Denton, Texas 76207, USA
2
J. Mike Walker ’66 Department of Mechanical Engineering and Department of Petroleum
Engineering, Texas A&M University, College Station, TX 77843, USA
*Address all correspondence to: Vish Prasad, Department of Mechanical Engineering, UNT Discovery
Park, University of North Texas, Denton, TX 76207-7102, USA; Tel.: +19409997854; Fax: +19405652666,
E-mail: [email protected]

Original Manuscript Submitted: 2/2/2022; Final Draft Received: 3/8/2022

Supercritical (SC) fluid, which was discovered exactly two hundred years ago, was initially thought to have only one
state with no distinction between the liquid and vapor. It is now considered to have two distinct regions –- “liquid-
like” and “gas-like,” with thermophysical properties exhibiting anomalous behavior near the critical point (CP). This
work highlights that the anomalous behavior, as observed above CP in the Widom region, extends to pressures and
temperatures much below the critical pressure Pc and/or critical temperature Tc. Indeed, the anomalous region starts
within the subcritical liquid state and extends well into the SC region. Motivated by many applications, extensive
research has been devoted to thermal transport by SC water and CO2. However, both of these fluids are reactive and
corrosive, and require high pressures (Pc = 217.8 and 72.8 atm, respectively); SC water also needs high temperature (Tc
= 373.95°C for water). This paper presents, for the first time, argon as an alternative SC fluid for thermal transport,
whose Pc is lower (47.994 atm) and it can work at low temperatures (Tc = –122.46°C). Argon is easily-available,
nonreactive, noncorrosive, and environmentally benign. A thermal transport analysis of fully developed flow of SC Ar
through an isothermally heated duct and its comparison with SC CO2 and water demonstrates that for similar rates
of heat transfer, the increase in cost of using argon is small, in terms of flow work required, whereas the benefits are
substantial with regard to complexity (due to Pc being much lower), operational maintenance, and life of the thermal
systems. It is also revealed that 2 to 3 orders-of-magnitude enhancement in gaseous state heat transfer is possible by
moving from subcritical to high-supercritical pressures; the effect being more pronounced at lower SC temperatures,
even at cryogenic temperatures in contrast to SC CO2 (Tc = 30.98°C) and water.

KEY WORDS: anomalous behavior, argon, cross-over, cryogenic, Frenkel line, gas-like, heat transfer
enhancement, liquid-like, pseudo-critical, subcritical, supercritical, Widom region

1. INTRODUCTION
The existence of a condition at which the interface between liquid and vapor vanishes was discovered exactly 200
years ago, by Cagniard de la Tour (1822). This was later confirmed by Faraday in 1845 who gave it the name of “Cag-
niard de la Tour point.” Subsequently, Andrews (1869) through his experiments at varying pressures and temperatures,
in a sealed glass tube with partially liquefied carbonic acid, observed the existence of a pressure–temperature (P, T)
point where this transformation took place and continued with P and T beyond the critical values. That means there
was no reversal to two phases past this point; Andrews (1969) referred to it as the “critical point.” The first application-

1064-2285/22/$35.00 © 2022 by Begell House, Inc. www.begellhouse.com 1

2 Prasad, Kakroo, & Banerjee

NOMENCLATURE

A area, m2 Greek Symbols

CH isobaric heat capacity, (ρCp), α thermal diffusivity, (k/ρCp),
J/m3 K m2/s
Cp specific heat at constant pressure, β isobaric coefficient of volumetric
J/kg K expansion, K–1
Cs isentropic specific heat, J/kg K ΔP/L pressure loss per unit length,
Cv specific heat at constant volume, (fρum2/2D), Pa/m
J/kg K ΔT bulk mean temperature difference,
D inside diameter of duct, m (Tm,o – Tm,i), K
f friction factor κs isentropic compressibility
g gravitational acceleration, m/s2 κT isothermal compressibility
h convective heat transfer μ dynamic viscosity, Pa s
coefficient, W/m2 K ν kinematic viscosity, (μ/ρ),
he specific enthalpy, J/kg m2/s
k thermal conductivity, W/m K ρ density, kg/m3
L heated length, m
ṁ mass-flow rate, kg/s Abbreviations
Nu Nusselt number (hD/k) AL anomalous liquid
P pressure, Pa BP boiling point
Pr Prandtl number, (ν/α) or (μCp/k) CP critical point
q heat-transfer rate, (ṁCpΔT) or DHT deteriorated heat transfer
(hπDLΔTlm), W EHT enhanced heat transfer
q″ heat flux, (q/A), W/m2 FL Frenkel line
Re Reynolds number, (umD/ν) or GL gas-like
(ρumD/μ) IHT improved heat transfer
T temperature, K LL liquid-like
u axial velocity, m/s NHT normal heat transfer
v specific volume, m3/kg SC supercritical
W/L work required per unit length, SCGL supercritical gas-like
(fπDρum3/8), W/m SCLL supercritical liquid-like
x axial coordinate, m TP triple point

oriented property of supercritical (SC) fluids, high solubility of solid compounds under SC conditions, was debated in
late 1800s (Hannay and Hogarth, 1879) and a convincing experiment on dissolution of solid naphthalene in SC CO2
was reported by Buchner (1906). This led to extensive research on solvent properties of SC fluids for a wide range of
extraction, separation, and fractionalization in mining, chemical, food processing, and pharmaceutical applications, to
name a few (Palmer and Ting, 1995).
Taking advantage of the enhanced solubility at SC conditions, hydrothermal growth of high-quality crystals
involving thermal transport was developed in 1800s, replicating the natural phenomena of processing minerals in
aqueous solutions beneath the Earth’s surface (Ely and Baker, 1983; Levelt Sengers, 2000). Since then the process that
involves dissolution, SC flow, species transport, heat transfer, and separation/deposition, has been industrially used
for bulk growth of quartz, sapphire, and many other large crystals. Indeed, the first computational model to simulate
hydrothermal growth (at ~ 150 MPa, 340–360°C) was developed by the first author, Prasad and his co-workers (Chen
et al., 1999a,b) which was later extended to ammonothermal growth of gallium nitride crystals (400–500 MPa, 500–
600°C) (Chen et al., 2003, 2006a,b, 2011; Pendurti et al., 2006), a material most commonly used for LED lights today.

Heat Transfer Research

Existence of Supercritical “Liquid-Like” State in Subcritical Region 3

As reported by Pioro and Pioro (1997), Pioro and Duffey (2003), and others, the first work on thermal
transport under SC conditions started as early as in 1930s and first notable paper on SC heat transfer was pub-
lished by Schmidt et al. (1946). During an investigation of cooling of turbine blades, they observed a significant
increase in free convection heat transfer near the critical state (Schmidt et al., 1946; Schmidt, 1960). They took
advantage of this phenomenon to build a single-phase thermosyphon with the intermediate fluid working at near
critical point. This was the beginning of research on heat transfer under SC conditions and was later followed
by Deissler and Cleveland (1954), Schmidt (1959), and Shitsman (1963), to name a few, in the late fifties and
early sixties. Extensive surveys of research on heat transfer and hydraulic resistance of water, carbon dioxide,
helium, and other fluids at supercritical and near-critical pressures have been presented by Cheng and Schul-
enberg (2001), Pioro and Duffy (2003), Pioro (2019), Cabeza et al. (2017) and more recently, by Bodkha and
Maheshwari (2021).
In the late fifties, it was also envisioned that the use of supercritical fluids may eliminate the challenges with
respect to phase change from liquid-to-vapor in thermal power systems, and hence, the critical heat flux and dryout.
This led to research on the use of SC water in nuclear reactors in the US, UK, France, and former USSR. However,
no significant progress was made, particularly with the emergence of light water reactors (LWRs), until 1990s (Pioro,
2016). With the maturing of LWR technology and renewed interest in higher capacity and safety, significant research
has been (and is being) pursued in recent years on the use of SC water in power-plant steam generators, Generation IV
nuclear reactors, and so on (Pioro, 2016).
Since the critical pressure Pc and temperature Tc of water are relatively high, 22.064 MPa (217.8 atm) and 647.096
K (373.95°C), respectively, there has been tremendous interest in fluids other than water that can be used at lower
SC temperatures and pressures. Supercritical carbon dioxide with Pc = 7.377 MPa (72.8 atm) and Tc = 304.128 K
(30.98°C) has emerged as the next on the list of SC fluids with wide range of potential applications, ranging from
power systems, refrigeration, renewable energy conversion, waste treatment, and carbon capture and storage (CCS)
to pharmaceutical, food, cosmetic, chemical, polymer/plastic, wood, and textile industries (Palmer and Ting, 1995;
Pioro and Duffy, 2007; Anderson, 2012; Pioro, 2014; Knez et al., 2014; Zhang et al., 2020). However, both of these
fluids require complex thermal transport systems primarily because of high critical pressures, and also, temperature
in the case of SC water.
Helium (Pioro and Duffy, 2007) is another fluid that has been extensively investigated because of its low critical
pressure and temperature [Pc = 0.2275 MPa (2.24 atm) and Tc = −267.95°C (5.195 K)] and high thermal diffusivity (~
25 times higher than SC CO2 at 10 MPa, 400 K, and ~ 10 times higher than water at 30 MPa, 900 K). However, helium
is not available easily on the Earth. Moreover, He atoms can easily migrate through the material defects and grain
boundaries of the walls (Schilling and Kiritani, 1982), which makes it unsuitable for most thermal transport systems.
Our discussions here will not include helium.
In addition to high critical pressures of SC CO2 and water and high critical temperature of SC water, there are
many other challenges associated with them such as chemical reaction, corrosion, and system strength/complexity
(see Section 5). In this paper, we present, for the first time†, a nontoxic, nonreacting, noncorroding fluid, argon [Pc =
4.863 MPa (47.994) atm and Tc = –122.463°C (150.687 K)] that can offer many advantages over SC water, CO2, and
other fluids, with enormous potential of applications in thermal transport, including heat exchangers. Argon can also
offer much larger range of working SC pressure and temperature.
We also present discussions on the supercritical state not being of a single state but having two regimes – “liquid-
like” (SCLL) and “gas-like,” (SCGL) as identified in recent years. Anomalous behavior of thermophysical properties
of Ar, very similar to that observed in the case of other SC fluids, is presented and the pressures and temperatures at
which crossover from “liquid-like” to “gas-like” behavior in the supercritical region are analyzed for SC Ar, CO2,
and water. Furthermore, we demonstrate that the anomalous behavior not only exists in the SC region, but extends
well into the subcritical region, i.e., below the critical pressure and/or temperature. This leads to identification of “SC
liquid-like” state dividing the liquid region in two parts. In the case of subcritical liquid water, this region is relatively
much larger which can pose many challenges with the use of water near the critical point.

†
Kalsia et al. (2017) have mentioned the future use of SC argon for cooling of high-temperature superconducting cables but their paper is only
focused on statistical correlations for SC properties of Ar from (Pc, Tc) to (Pc + 10 bar, Tc + 10 K); no heat transfer data and analysis are presented.

Volume 53, Issue 9, 2022

4 Prasad, Kakroo, & Banerjee

A heat transfer and pressure loss analysis of fully developed flow of SC Ar through an isothermally heated duct
is presented and compared with the predictions for SC CO2 and water. These data reveal that the heat transfer rate is
substantially enhanced as the fluid moves from its regular (subcritical) gaseous state to supercritical “gas-like” state.
This effect becomes much stronger as the temperature is decreased, with maximum benefits at cryogenic temperatures.
Finally, the advantages of using SC argon as a heat transfer/exchanger fluid is discussed and it is shown that the cost
of these benefits is minimal in terms of pressure loss and work required.

2. THERMOPHYSICAL PROPERTIES OF SC ARGON AND ANOMALOUS BEHAVIOR

Thermophysical properties of argon – density ρ, specific heat Cp, dynamic viscosity μ, and thermal conductivity k, ob-
tained using the NIST Chemistry WebBook (2021), are presented in Fig. 1a–1d for P = 4.863 (Pc), 5.0, 6.0, 10.0, 20.0,
30.0 MPa and 120 < T < 250 K (Tc = 150.687 K). Also, presented are the derived properties, thermal diffusivity α and
Prandtl number Pr in Fig. 2a and 2b, respectively. Note that the isobaric line of Pc for some of the properties such as Cp,
k, α, and Pr show discontinuity at and around critical temperature Tc, since the values cannot be precisely defined by the
NIST WebBook (2021). In Fig. 1a, 1c and 1d, the density, dynamic viscosity, and thermal conductivity, respectively,

FIG. 1: Thermophysical properties of argon at and above critical conditions (Pc = 4.863 MPa, 47.994 atm, and Tc = –122.463°C,
150.687 K): (a) density ρ, (b) specific heat Cp, (c) dynamic viscosity μ, and (d) thermal conductivity k

Heat Transfer Research

Existence of Supercritical “Liquid-Like” State in Subcritical Region 5

FIG. 2: (a) Thermal diffusivity α and (b) Prandtl number Pr of argon at and above critical conditions (Pc = 4.863 MPa and Tc =
150.687 K)

show substantial drops in their values (almost vertical at CP). On the other hand, the specific heat Cp in Fig. 1b and Pr
in Fig. 2b first increase with temperature, achieve peak values, and then decrease to much lower values, the peaks being
highest at Pc and Tc. As the pressure P increases beyond the critical point, this behavior of sharp changes is weakened,
and at high SC pressures they vanish, leading to almost monotonic trends. The anomalous behavior in these plots, e.g.,
sharp changes in most thermophysical properties at and beyond the critical point is very similar to that observed in the
case of water and CO2 (Pioro, 2014), and many other fluids, and is widely reported in the literature.
To amplify the monotonic trend of thermophysical properties at high pressures and temperatures, the density and
specific heat in the range of 200 < T < 700 K are presented in Fig. 3a and 3b; note that 700 K is the upper limit of
temperature for which the properties can be obtained from the NIST WebBook (2021). However, this monotonic be-
havior is true even for P > 40 MPa at T > 200 K, and for P > 50 MPa at T >160 K (not shown here). Evidently, all other

FIG. 3: (a) Density and (b) specific heat for argon at P = 4.0, 4.5, 4.863 (Pc), 5.0, 6.0, 10.0, 20.0 and 30.0 MPa for temperature T
much beyond the critical temperature Tc. Note the monotonically decreasing trends with temperature in the range of 200 < T < 700
K, and nearly straight-line behavior in 600 < T < 700 K. Similar trends are observed in the case of other thermophysical properties.

Volume 53, Issue 9, 2022

6 Prasad, Kakroo, & Banerjee

properties show similar trends at high SC pressures and temperatures (Figs. 1 and 2). Note that although in Fig. 3b,
the 30 MPa curve shows lower values of Cp than that for 20 MPa when T < 250 K, overall trends still remain similar.
To characterize the anomalous behavior of SC fluids near (Pc, Tc), a pseudo-critical point was initially described
as the temperature at which the isobaric specific heat Cp had a peak at Pc. Later, the (P, T) line representing the pres-
sure and temperature with respect to peak values of Cp under supercritical conditions was called the pseudocritical line
(Imre et al., 2012). Subsequently, the region in which various fluid properties vary sharply and the fluid behavior is
observed to be anomalous was identified as the pseudocritical region that needed to be treated differently. Mostly, the
discussions on pseudocritical region have focused on pressure and temperature beyond the critical point.
Investigating a liquid–liquid critical point, Xu et al. (2005) observed a correlation between the dynamic crossover
and the locus of specific heat maxima, and called it “Widom line.” They believed that locations of maxima of different
SC properties may not lie far from each other, and one line may be sufficient to determine the crossover from liquid-
like (SCLL) to gas-like (SCGL) behavior. This contradicted the initial assumption that complete transformation to SC
state occurs uniquely at (Pc, Tc), and was challenged first in 2006 by Gorelli et al. Later, Simeoni et al. (2010) observed
a sharp transition on crossing the Widom line which demonstrated the division of SC region in two parts. Interestingly,
it was found that even in the case of simplest Lennard–Jones fluid the lines for maxima of various thermophysical
properties may lie away from each other (Brazhkin et al., 2011). Indeed, as shown by many other models, the lines of
maxima diverged significantly in a P–T plane (Fomin et al., 2014), as shown in Fig. 4.
Brazhkin et al. (2011) also demonstrated that Widom lines for various properties merge into a single line, repre-
sented by the line of critical isochore/Cp,max at T < 1.1Tc and P < 1.5Pc and become smeared at T ≅ 2.5Tc and P ≅ 10Pc
(Simeoni, 2010). They hypothesized that at sufficiently high pressures (P/Pc > 20), dynamic, not thermodynamic,
characteristics will need to be considered to separate the liquid-like and gas-like SC states. Note that the lines cor-
responding to supercritical maxima for various thermophysical properties, for argon, are in good agreement with the
computer simulation data obtained for Lennard–Jones fluids by Brazhkin et al. (2011).
Figure 4 presents the Widom lines for SC Ar (real Ar rather than the van der Waals Ar) together with that for SC
water and CO2, based on the correlations presented by Imre et al. (2019) in their Table 3 and verified here with the
values obtained from the NIST WebBook (2021). Note that in Fig. 4, the data for Ar end at 20 MPa and that for water
and CO2 at 50 MPa. Interestingly, in Fig. 1b, the 20 MPa line for Cp shows a peak around 190 K (with less than 2%
variation from the values at 180 and 200 K) which becomes further weaker at 30 MPa (~ 6.16Pc) and continues to di-
minish with the pressure; becomes nonexistent at 50 MPa, as seen from the NIST WebBook (2021). Similar behavior
is exhibited by CO2 and water beyond 50 and 500 MPa, respectively. This means that beyond these pressures, the Wi-

FIG. 4: Widom lines/region for argon, CO2 and water following Imre et al. (2019) and data from NIST WebBook (2021); Frenkel
line for Ar based on data from Gorelli et al. (2013), Frenkel point from Bolmatov et al. for Ar (2015) and CO2 (2014)

Heat Transfer Research

Existence of Supercritical “Liquid-Like” State in Subcritical Region 7

dom region may not be able to characterize the anomalous behavior, which may require more information at (much)
higher values of P and T beyond the critical point, as hypothesized by Brazhkin et al. (2012).
The dynamic boundary between the SC “liquid-like” and “gas-like” behavior was recently identified by Brazh-
kin et al. (2012) and Brazhkin and Trachenko (2012), who named it the “Frenkel line.” At the same time, Gorelli et
al. (2013) reported a study on the dynamics and thermodynamics beyond the critical point and used Ar as a fluid to
demarcate the liquid-like and gas-like behavior in the range of T/Tc and P/Pc up to 5 and 50, respectively. Bolmatov et
al. (2015) observed structural crossovers in a deeply supercritical sample of Ar through diffraction measurements and
discovered the thermodynamic boundary of Frenkel line (FL) on the P–T diagram. They also explained the existence
of crossover using the phonon theory of liquids. Structural evolution of SC CO2 across the Frenkel line was reported
by Bolmatov et al. (2014) and its thermodynamic properties with respect to the Widom and Frenkel lines were pre-
sented by Fomin et al. (2015). In addition, dynamical/structural crossover in the case of SC water was reported by
Cockrell et al. (2020).
Some of the data for the Frenkel line from Gorelli et al. (2013) and Bolmatov et al. (2015) for SC Ar and
from Bolmatov et al. (2014) for SC CO2 are also plotted in Fig. 4. While the data of Gorelli et al. (2013) show a
clear trend and their line falls between the lines of maxima (Widom lines) for isentropic compressibility κs and
isentropic specific heat Cs, the data of Bolmatov et al. (2015) do not present any trend. This is not surprising since
Bolmatov et al.’s (2015) data have been taken from the dashed line in their Fig. 1 that is only for guide. The ex-
perimental data were presented only for P > 400 MPa. In the case of SC CO2, we could pick only two data points
from Bolmatov et al. (2015), i.e., at P = 10 and 40 MPa. The first point is within the criterion of Brazhkin et al.
(2012), one Widom line for P < 1.5Pc where the Frenkel line should coincide with the Widom lines altogether. On
the other hand, similar to Gorelli et al.’s (2013) data the second point (at 40 MPa) does lie between the Widom
lines of κs and Cs.
Although the concept of crossover from liquid-like to gas-like SC region by the above authors have been accepted
by several researchers in the field, the work by Bryk et al. (2017) “critically undermine the definition of the FL and its
significance for any relevant partition of the supercritical phase.” Bryk et al.’s (2017) assertion was later challenged
by Brazhkin et al. in a comment published in 2018. Obviously, the issue is not settled and more work is needed be-
fore a universally acceptable theory on crossover from “liquid-like” to “gas-like” SC region emerges. However, from
the thermal transport point of view it is evident that transport phenomena in the Widom region would be extremely
complex.
Before moving further, it may be helpful that we summarize the key observations from Fig. 4:
• The divergence among Widom lines is largest for SC CO2 and lowest for SC water, among the three fluids
considered here.
• The lines for isentropic specific heat Cs and isothermal compressibility κT diverge the most, to much higher
temperatures beyond Tc. Do these two properties influence thermo-fluid-species transport in any significant
manner? If not, then the Widom regions may be considered much smaller for heat and mass transfer.
• Maxima of several properties move to the right in Fig. 4, i.e., as the SC pressure increases the temperatures
for the peaks also increase. On the other hand, for some other properties, e.g., Cv, ρ, and Cp in the case of SC
CO2, the behavior may be just reverse. Again, how does this affect the transport behavior?
• In the case of SC Ar, all lines except for Cs and κT lie within Tc < T < 200 K. Therefore, the assumption of
SCGL state at T > 200 K should be valid, as also shown by Gorelli et al. (2013). That means, SC Ar is in the
SCGL state for all pressures, P > Pc and T > 200 K. Similar behavior may be valid for CO2 at T > 400 K and
for water at T > 750 K; these threshold temperatures have been used for heat transfer calculations in Section
4.
• Argon, therefore, exists in supercritical gas-like state for much larger range of pressure and temperature,
beginning with the cryogenic temperatures, unlike CO2 and water (Fig. 4).

3. EXISTENCE OF “SUPERCRITICAL LIQUID-LIKE” STATE IN SUBCRITICAL REGION

To examine further the anomalous behavior in the subcritical to supercritical region (near the critical point), we pres-
ent, in Fig. 5a–5c, the specific heat of Ar, CO2, and water for a range of pressure, varying from subcritical to super-

Volume 53, Issue 9, 2022

8 Prasad, Kakroo, & Banerjee

FIG. 5: Specific heat Cp for (a) Ar, (b) CO2, and (c) water in the subcritical liquid to supercritical region, and (d) a typical liquid–
vapor phase change in the subcritical region, water at P = 0.1, 1.0, 3.0, 5.0, and 7.0 MPa

critical values. Figure 5d demonstrates a typical liquid–vapor phase change behavior from pre-boiling to post-boiling
states, presented here for water.‡
As is well known, a step-change in (all) thermophysical properties occurs as the liquid is converted into vapor
at its boiling temperature, which is a strong function of pressure. It is interesting to note that the simple behavior
of small changes in the pre-boiling liquid region, followed by a step-change at the boiling point, and again simple
changes in the vapor region as shown by the 0.1 MPa line, does not continue as the pressure is increased towards
critical pressure Pc. Indeed, the drop in the Cp value at the boiling point Tb in Fig. 5d continuously decreases as the
pressure is increased, for example, from 4.2152 → 2.0784 at 0.1 MPa (close to atmospheric pressure, Tb = 372.76 K)
to 4.1786 → 3.6119 at 3 MPa (Tb = 507 K) to 5.0368 → 4.4380 at 5 MPa (Tb = 536 K) to 5.4025 → 5.3792, almost
zero at 7 MPa (Tb = 559 K). Similar changes occur in other properties with increasing pressure. It is therefore evident
that in a SC thermal transport system where the working fluid transits from subcritical liquid to SC “gas-like” state,
and vice versa, the variation in thermophysical properties must be considered in its analysis, computer simulation,
and design.
‡
Note that most of the studies on thermophysical properties and transport behavior of SC fluids have focused only on pressures and temperatures
at and above the critical point.

Heat Transfer Research

Existence of Supercritical “Liquid-Like” State in Subcritical Region 9

Coming back to Fig. 5a for Ar, it shows a one-step drop in the value of Cp at P = 3.0 and 3.3 MPa. However, start-
ing with 3.5 MPa (< Pc = 4.863 MPa) the isobaric line for Cp shows a behavior very similar to that observed in the SC
region (Fig. 1b), the peaks occurring at much lower pressure and temperature than the critical point (Pc, Tc). Similar
behavior is shown in Fig. 5b and 5c for SC CO2 (P ≅ 5 MPa and higher) and SC water (P ≅ 8 MPa and higher), re-
spectively. Indeed, in the case of water, this phenomenon occurs at relatively much lower pressure, P/Pc = 0.36 against
0.68 for CO2 and 0.72 for Ar.
Indeed, Fig. 5 raises an important question. That is, should not we consider a part of the liquid region, say beyond
3.3 MPa and 137 K and up to (Pc, Tc) in the case of Ar (Fig. 5a), differently than the subcritical liquid region of P < 3.3
MPa and T < 137 K. This part of the SC region may be called “supercritical liquid-like” state where the fluid behavior
is very similar to that in the “liquid-like” SC region (Figs. 1b and 4a).
The question is then can we combine this region with the “liquid-like” SC region? Are the two subregions con-
tinuous? The answer to this question will determine how to treat the liquid in the region near the critical point, on both
sides. Apparently, from thermal transport point-of-view, the behavior does look very similar. As noted earlier, this
phenomenon occurs much earlier in the case of water (P/Pc = 0.36) which is expected to have far more applications in
power industry and other high-temperature and high-pressure systems. Deep understanding of the two liquid-states in
the subcritical region and “liquid-like” state in the supercritical region is, therefore, of paramount importance to SC
thermal transport.
With the above observations, we present a phase diagram (Fig. 6) that depicts the “SC liquid-like” state (right of
PLQ) in the subcritical liquid region and “liquid-like” (left of PGQ) and “gas-like” states in the supercritical region.
Figure 6 follows and extends the works of Pioro (2014, 2016), Gorelli et al. (2013), Bolmatov et al. (2014, 2015),
Bryk et al. (2017), and others, and shows the existence of two states of subcritical liquid. It is interesting to note that
the phase diagram presented in the abstract of Bryk et al. (2017) explicitly, but hypothetically, demarcates the region
of the compressed liquid (P > Pc and T < Tc) and extends the demarcation line to the subcritical liquid state (http://
femtoscopy.org/static/RES_ACT/my_papers/acs.jpclett.7b02176.pdf). These authors have also connected this line in
the subcritical region to the Frenkel line in the SC region. In addition, they have extended the boiling point (P–T) line
in the subcritical region to the SC region by a dotted line without much explanation. From thermal transport perspec-
tive, it is evident that the entire subcritical liquid region cannot be treated as one kind of state. A similar concept of
two states of liquid referred to as “rigid liquid” and “nonrigid liquid” has been presented in Brazhkin et al. (2012) and
Proctor (2021).

FIG. 6: Modified phase-diagram depicting anomalous region –“SC liquid-like” state in the subcritical liquid region (right of line
PLQ) and “liquid-like” and “gas-like” states (separated by line AGQ) in the supercritical region, extending the work of Pioro (2014,
2016), Gorelli et al. (2013), Bolmatov et al. (2014, 2015), and Bryk et al. (2017)

Volume 53, Issue 9, 2022

10 Prasad, Kakroo, & Banerjee

Another issue that needs to be resolved is at what pressure and temperature, indeed much beyond (Pc, Tc), the
line separating solid/liquid states merges with the line demarcating the liquid/“SC liquid-like” state and then with the
line separating the states of SC “liquid-like”/SC “gas-like” states (Fig. 6). From diffraction measurements in SC Ar at
high SC pressures and temperatures in a diamond anvil cell, Bolmatov et al. (2015), have shown that the SC liquid-
like region contracts as the pressure and temperature increase beyond 400 MPa and 325 K, and ultimately lead to a
very narrow gap between the solid/liquid line and the Frenkel line beyond 1 GPa and 400 K; see Fig. 1 in Bolmatov
et al. (2015) (https://www.nature.com/articles/srep15850). Even though for SC thermal transport such high pressures
may not be practical, it is important that a thorough understanding of states and crossover from solid-to-liquid-to-SC
“liquid-like” (SCLL in subcritical and supercritical regions) and then -to-SC “gas-like” (SCGL) states in the SC region
is developed.
Also remains unresolved is what happens to the transport behavior at P/Pc > 20 when the dynamic, not thermody-
namic, characteristics are expected to separate the SCLL and SCGL behavior (Fomin et al., 2014), particularly since
the transport properties such as ρ, Cp, μ, and k exhibit monotonic behaviors at such pressures. Moreover, since the
SCLL/SCGL line (Fig. 4) seems to lie before the lines for isentropic heat capacity and isothermal compressibility, do
these properties (Cs and κT) have any effect on heat and mass transfer beyond the crossover line from “liquid-like” to
“gas-like” states. If not, then this crossover line may be conveniently used as a line beyond which the transport quanti-
ties can be obtained from the relations developed for subcritical vapor (gas). Answers to these questions are beyond
the scope of this paper.
Using the above findings/concepts of crossovers in the subcritical region from “pure liquid” to “SC liquid-like”
state and in the supercritical region from “liquid-like” to “gas-like” state, it may be possible to explain the com-
plex phenomena of thermal transport in near-critical and supercritical regions which involve improved heat transfer
(IHT), deteriorated heat transfer (DHT), and normal heat transfer (NHT); the three thermal transport regimes that
have been identified by many investigators (Pioro, 2014, 2016, 2019; Anderson, 2012). NHT exists away from the
Widom region and the Frenkel line where the transport behavior is very similar to that under the subcritical condi-
tions and existing flow, heat, and mass transport theories/correlations can be supposedly employed. DHT occurs in
a region where temperature for any P > Pc increases beyond the crossover point temperature and fluid converts from
“liquid-like” to “gas-like” state resulting in huge changes in transport properties, particularly density (Fig. 1a); this
may cause the heat transfer to decrease. DHT is easy to observe in the case of flow over an isoflux surface where the
high heat flux can cause the fluid at the surface to change from “liquid-like” to “gas-like” while in other regions the
fluid still remains “liquid-like.” IHT is then simply the increase from DHT to NHT when the “gas-like” fluid leaves
and the “liquid-like” fluid occupies that volume. On the other hand, enhanced heat transfer (EHT) is the increase in
energy transport from the subcritical region to the supercritical region even though under certain conditions DHT
may occur in between.

4. THERMAL TRANSPORT
As noted in Section 1, research on heat transfer in supercritical fluids started as early as in 1930s and was followed
by extensive studies on thermal-hydraulics of supercritical water in nuclear systems. Detailed reviews of these stud-
ies have been presented by Chen and Schulenberg (2001), Pioro and and Duffey (2003, 2007), Pioro et al. (2004),
Pioro (2016, 2019), Cabeza et al. (2017), and very recently, by Bodkha and Maheshwari (2021). Since the horizontal
and vertical tubes are basic configurations of power systems and heat exchangers, a significant amount of theoretical,
numerical, and experimental research has been devoted to these two geometries, including buoyancy effects that are
more prominent in the case of vertical tubes. Chen and Schulenberg had presented an excellent review of these studies
until 2001 and analyzed the heat transfer correlations available in the literature at that time. Recent reviews have been
authored by Pioro (2019), Cabeza et al. (2017), and Shitsi et al. (2018). From these reviews, it is evident that even
under purely forced convection conditions in a horizontal tube, no convection correlations are valid for a wide range
of pressure and temperature, particularly in the pseudo-critical (Widom) region, primarily because of the anomalous
behavior of thermophysical properties. However, as noted by Mokry et al. (2011), Dittus–Boelter type equations can
be valid far away from the critical point where SC fluid behaves like a conventional single-phase fluid, SC “gas-like”
state, although this needs further validation.

Heat Transfer Research

Existence of Supercritical “Liquid-Like” State in Subcritical Region 11

In this section, we analyze the thermal transport of argon and compare its heat transfer characteristics with those
of CO2 and water in order to demonstrate the feasibility of SC Ar as a working fluid for a wide range of subcritical to
supercritical applications. For this analysis, we consider a fully developed flow and forced-convection heat transfer
in an isothermally heated duct, and restrict our calculations only to the gaseous and supercritical “gas-like” states.
Indeed, following the discussions in Section 2 and observation by Mokry et al. (2011) we select SC pressures and
temperatures outside the Widom region, particularly where the thermophysical properties of the selected fluids, SC Ar,
CO2, and water exhibit monotonic behavior (Figs. 3 and 4). This facilitates a direct comparison without any anomalous
effects.

4.1 Methodology
We obtain the friction factor f for laminar and turbulent fully developed flows in a smooth pipe from the following
equation:

f = 64/Re, Re < 2000 (for laminar flow) (1)

and Petukhov correlation,

f = (0.79 ln (Re)-1.64)–2. 3000 < Re < 5 ×106 (2)

The pressure drop along the tube is given by

L
ΔP = f (ρum2/2) (3)
D

where um is the mean velocity. The pumping power W, to drive the flow through a pipe, is then

W = (πD2um/4) ΔP = f (πDLρum3)/8 (4)

Both the pressure loss and work required will be presented per unit length, i.e., ΔP/L and W/L.
The heat transfer rate in laminar and turbulent regimes is obtained using

q = h(πDL) ΔTlm (5)

Volume 53, Issue 9, 2022

12 Prasad, Kakroo, & Banerjee

where the log-mean temperature difference is

ΔTlm = (ΔTo – ΔTi)/ln(ΔTo/ΔTi), with ΔTi = (Ts – Tm,i) and ΔTo = (Ts – Tm,0) (6)

The average heat transfer coefficient h can be obtained using the Nusselt number Nu correlations, where Nu =
hD/k. In the laminar regime, Nu is given by

Nu = hD/k = 3.66, Re < 2000 (7)

The Dittus–Boelter and Gnielinski equations are the two most widely used correlations for the turbulent Nusselt
number. Here, we will use the Gnielinski correlation:

 f
  ( Re − 1000)Pr  0.5 ≤ Pr ≤ 2000 
8
Nu =   (8)
 f (
0.5 2
1 + 12.7   Pr 3 − 1
 8
)
 3 × 10 < Re < 5 × 10 
3 6

that is based on the friction factor f [Eq. (2)].

In a recently published paper (Bamido et al., 2020), we have shown that there is very little difference between the
predictions by the Dittus–Boelter and Gnielinski correlations and either one can be used with confidence, particularly
when the purpose is to demonstrate the heat transfer behavior and trends. Indeed, for SC CO2 with 101 ≤ Re ≤ 104,
these correlations predict convective heat transfer very close to the numerical predictions using Ansys/Fluent R19.1
with k–ω SST turbulent model. The simulation was carried out for D = 1, 2, 5, and 10 mm with Tm,i = 700 K and heat
dissipation from the outer surface of the horizontal duct by natural convection, to ambient at 300 K. The computer
model has been further extended to simulate heat transfer in a duct with SC Ar as the working fluid, for a similar range
of Re and thermal boundary conditions (Bamido et al., 2022).
The predictions of the Nusselt number purely based on Re and Pr and those using Eqs. (1)–(8) are presented in
Fig. 7 for 5 × 103 ≤ Re ≤ 105 and 0.7 ≤ Pr ≤ 1.1. The Nusselt number behavior in this figure is consistent with what is
expected, i.e., the rate of increase in Nu increases as Re goes up and the Prandtl number has a small effect on Nu at low
values of Re but the effect becomes more pronounced at higher Reynolds numbers. Figure 7 also presents the selected
values of Nu obtained using the Gnielinski correlation for SC Ar, CO2, and water for the selected cases in Tables 1–3
with Tf = (Tm,i + Tm,o)/2.

4.2 Details of the Cases Considered

Tables 1–3 present the thermophysical properties ρ, μ, Cp, k, and Pr with pressure P for Ar, CO2, and water at Tf = (Ts
+ Tm,i)/2. The selected pressures for heat transfer calculations are as follows:

Heat Transfer Research

Existence of Supercritical “Liquid-Like” State in Subcritical Region 13

FIG. 7: Nusselt number vs. Reynolds number using the Gnielinski correlations for turbulent fully developed flow and thermal
conditions; symbols represent values from of the selected cases in Tables 1–3 using Tf = (Tm,i + Tm,o)/2

i. Table 1a for Ar (Tm,i = 200 K, Ts = 400 K, and Tf = 300 K): 0.1 (close to ambient pressure), 1, 2, 3, 4, 5, 10,
20, 30, 40, and 50 MPa (subzero to subcritical to supercritical);
ii. Table 1b for Ar and Table 2a for CO2 (400, 600, and 500 K): 0.1, 1, 2, 3, 4, 5, 10, 20, 30, 40, and 50 MPa (T
>> Tc for CO2);
iii. Table 1c for Ar and Table 2b for CO2 (600, 800, and 700 K): 10, 20, 30, 40, and 50 MPa (P > Pc and T >> Tc
for CO2);
iv. Table 1d for Ar, Table 2c for CO2, and Table 3 for water (800, 1000, and 900 K): 30, 40, and 50 MPa (P > Pc
and T >> Tc for water).

Tables 1–3 also present the values of the Reynolds number Re, mass flow rate ṁ, friction factor f, average Nusselt
number Nu, bulk mean temperature difference ΔT = (Tm,o – Tm,i), heat transfer rate per unit length q/L, pressure drop
per unit length ΔP/L, and the work required per unit length W/L for fully developed flow through a 10-mm diameter,
1-m-long tube maintained at a constant surface temperature Ts. For the case in Table 1a, fluid enters the tube at 200 K
with um = 1 m/s and Ts = 400 K, and heated fluid leaves the tube at Tm,o. Table 1a also indicates the state of the fluid in
both subcritical and SC regimes; with this, other cases in Tables 1–3 are self-explanatory.
Note that Tf for property evaluation in Tables 1–3 has been taken as (Tm,i + Ts)/2 rather than (Tm,i + Tm,o)/2, which
is usually recommended to be used for heat transfer calculations. This is primarily to demonstrate the true effect of
pressure variation on flow and heat transfer and not lose track by changing both pressure and fluid temperature Tf
simultaneously. An iterative procedure to obtain flow and heat transfer quantities with Tf = (Tm,i + Tm,o)/2 shows an er-
ror of 5% or less in most cases; in a few cases it may rise to 10%. However, in Fig. 7 the Nusselt number denoted by
symbols are all based on Tf = (Tm,i + Tm,o)/2.
Furthermore, the NIST WebBook provides Ar properties only up to 700 K. Since all of the thermophysical proper-
ties of SC Ar show almost straight-line behavior near 700 K – very similar to that shown in Fig. 3a and 3b for ρ and
Cp, respectively, it may be reasonable to extrapolate the property values at temperatures beyond 700 K, up to 900 K
for the first level of analysis here. This can facilitate comparison with SC water.

4.3 Optimal Enhancement in Heat Transfer from Subcritical to Supercritical Gaseous State
Although Figs. 1 and 2 demonstrate the property variation for Ar with pressure very well, the third column in Table 1a
and 1b is quite revealing that the density ρ increases almost by a factor of 10 when the pressure increases from 0.1 to

Volume 53, Issue 9, 2022

 TABLE 1: Thermophysical properties, heat transfer rate, and pressure loss/work required for fully developed flow of argon (Ar) through an isothermally

14
heated circular pipe at various pressures under subcritical and supercritical conditions (D = 10 mm, L = 1.0 m, um = 1 m/s, Tf = (Tm,i + Ts)/2)
(a) Tm,i = 200 K, Ts = 400 K (Tf = 300 K)
P, State ρ, kg/m3 μ, 10–5 Cp, J/ k, Pr ReD, ṁ, 10–3 f, 10–2 NuD ΔT, K q/L, ΔP/L, W/L,
MPa Pa·s kg·K W/m·K 104 kg/s W/m Pa/m W/m
0.1 Gas 1.603 2.27 521.52 0.0177 0.668 0.0707 0.126 7.97 3.66 191 12.5 6.38 0.00050
1 Gas 16.11 2.29 532.35 0.0180 0.676 0.705 1.26 3.48 22.0 169 113 28.0 0.00220
2 Gas 32.41 2.31 544.65 0.0184 0.684 1.40 2.54 2.87 38.5 160 221 46.5 0.00365
3 Gas 48.88 2.33 557.19 0.0188 0.691 2.10 3.84 2.58 52.9 154 328 63.2 0.00496
4 Gas 65.50 2.36 569.95 0.0192 0.699 2.78 5.14 2.41 66.1 149 436 78.9 0.00619
5 SCGL 82.28 2.39 582.87 0.0197 0.706 3.45 6.46 2.29 78.6 145 546 94.0 0.00738
10 SCGL 167.60 2.56 648.15 0.0223 0.745 6.55 13.2 1.97 133 133 1,130 165 0.0130
20 SCGL 335.84 3.03 758.24 0.0279 0.825 11.1 26.4 1.76 215 122 2,440 296 0.0232
30 SCGL 482.58 3.62 813.34 0.0339 0.870 13.3 37.9 1.69 257 118 3,620 409 0.0321
40 SCGL 600.10 4.24 826.34 0.0399 0.878 14.2 47.1 1.67 271 116 4,530 502 0.0394
50 SCGL 692.78 4.84 821.19 0.0456 0.872 14.3 54.4 1.67 273 116 5,200 578 0.0454
(b) Tm,i = 400 K, Ts = 600 K (Tf = 500 K)
0.1 Gas 0.961 3.40 520.69 0.0265 0.667 0.0283 0.0754 12.6 3.66 200 7.85 6.05 0.00048
1 Gas 9.593 3.41 523.88 0.0268 0.668 0.281 0.753 4.66 9.13 171 67.6 22.3 0.00175
2 Gas 19.15 3.42 527.39 0.0270 0.669 0.559 1.50 3.73 18.0 171 135 35.7 0.00280
3 Gas 28.671 3.44 530.87 0.0272 0.669 0.834 2.25 3.31 25.2 167 200 47.5 0.00373
4 Gas 38.153 3.45 534.30 0.0275 0.670 1.11 3.00 3.06 31.6 164 262 58.4 0.00458
5 SCGL 47.592 3.46 537.69 0.0278 0.670 1.37 3.74 2.89 37.5 161 323 68.7 0.00539
10 SCGL 94.094 3.54 553.95 0.0292 0.671 2.66 7.39 2.44 62.5 151 617 115 0.00899
20 SCGL 182.91 3.73 582.41 0.0322 0.673 4.91 14.4 2.10 100 141 1,180 192 0.0151

Prasad, Kakroo, & Banerjee

30 SCGL 265.14 3.95 604.94 0.0352 0.679 6.72 20.8 1.96 129 135 1,700 260 0.0204
Heat Transfer Research

40 SCGL 340.23 4.19 621.85 0.0380 0.685 8.13 26.7 1.88 150 132 2,190 320 0.0251
50 SCGL 408.27 4.44 634.09 0.0407 0.691 9.19 32.0 1.83 166 130 2,630 374 0.0294
(c) Tm,i = 600 K, Ts = 800 K (Tf = 700 K)
10 SCGL 66.802 4.44 535.70 0.0360 0.662 1.50 5.24 2.82 40.0 160 449 94.1 0.00739
20 SCGL 129.80 4.56 548.95 0.0382 0.656 2.85 10.2 2.39 65.0 150 841 155 0.0122
Volume 53, Issue 9, 2022

Existence of Supercritical “Liquid-Like” State in Subcritical Region

TABLE 1: (continued)
30 SCGL 188.91 4.69 560.17 0.0404 0.651 4.03 14.8 2.20 84.4 145 1200 208 0.0163
40 SCGL 244.21 4.84 569.55 0.0425 0.648 5.05 19.2 2.09 100 141 1540 255 0.0200
50 SCGL 295.84 4.99 577.33 0.0446 0.646 5.93 23.2 2.02 113 139 1860 298 0.0234
(d) Tm,i = 800 K, Ts = 1000 K (Tf = 900 K)
30 SCGL 109.6 5.48 566.10 0.0462 0.672 2.00 8.60 2.62 50.2 155 756 143 0.0112
40 SCGL 144.4 5.63 576.42 0.0482 0.673 2.57 11.3 2.46 60.9 151 988 177 0.0139
50 SCGL 179.5 5.78 579.81 0.0503 0.667 3.11 14.1 2.34 70.2 148 1,210 210 0.0165

15
 16
TABLE 2: Thermophysical properties, heat transfer rate, and pressure loss/work required for fully developed flow of carbon dioxide (CO2) through
an isothermally heated circular pipe at various pressures under subcritical and supercritical conditions (D = 10 mm, L = 1.0 m, um = 1 m/s, Tf = (Tm,i
+ Ts)/2)
(a) Tm,i = 400 K, Ts = 600 K (Tf = 500 K)
P, State ρ, kg/ μ, 10–5 Cp, J/ k, Pr ReD, ṁ, 10–3 f, 10–2 NuD ΔT, K q/L, ΔP/L, W/L,
MPa m3 Pa·s kg·K W/m·K 104 kg/s W/m Pa/m W/m
0.1 Gas 1.0594 2.40 1015.4 0.0335 0.728 0.0441 0.0832 9.95 3.66 198 16.7 5.27 0.00041
1 Gas 10.664 2.41 1027.3 0.0337 0.733 0.443 0.837 4.01 15.2 169 145 21.4 0.00168
2 Gas 21.482 2.41 1040.9 0.0340 0.738 0.890 1.69 3.25 27.9 163 287 34.9 0.00274
3 Gas 32.451 2.42 1054.9 0.0343 0.744 1.34 2.55 2.90 38.8 158 424 47.1 0.00370
4 Gas 43.566 2.43 1069.2 0.0347 0.750 1.79 3.42 2.69 48.9 153 561 58.6 0.00460
5 Gas 54.826 2.45 1084.0 0.0351 0.756 2.24 4.30 2.54 58.6 150 699 69.6 0.00547
10 SCGL 113.07 2.54 1162.4 0.0373 0.792 4.45 8.88 2.15 102 137 1,420 122 0.00955
20 SCGL 235.24 2.90 1322.8 0.0434 0.883 8.12 18.5 1.88 175 125 3,050 221 0.0174
30 SCGL 352.51 3.44 1440.9 0.0511 0.971 10.2 27.7 1.79 223 118 4,720 315 0.0248
40 SCGL 452.94 4.07 1498.1 0.0591 1.03 11.1 35.6 1.76 247 115 6,150 398 0.0313
50 SCGL 534.42 4.70 1513.6 0.0668 1.06 11.4 42.0 1.75 256 114 7,250 468 0.0367
(b) Tm,i = 600 K, Ts = 800 K (Tf = 700 K)
10 SCGL 75.49 3.24 1177.7 0.0516 0.739 2.33 5.93 2.52 59.6 150 1,050 95 0.00745
20 SCGL 149.27 3.39 1225.3 0.0547 0.759 4.41 11.7 2.16 99.1 139 1,990 161 0.0126
30 SCGL 219.39 3.60 1265.7 0.0583 0.782 6.09 17.2 2.00 130 133 2,890 220 0.0173
40 SCGL 284.29 3.87 1297.5 0.0622 0.806 7.36 22.3 1.92 153 129 3,730 273 0.0214
50 SCGL 343.27 4.16 1321.5 0.0662 0.830 8.26 26.9 1.87 171 126 4,500 322 0.0253

Prasad, Kakroo, & Banerjee

(c) Ts = Tm,i = 800 K, 1000 K (Tf = 900 K)
Heat Transfer Research

30 SCGL 165.44 4.09 1275.2 0.0700 0.746 4.04 13.0 2.20 91.8 141 2,330 182 0.0143
40 SCGL 214.66 4.25 1292.6 0.0725 0.757 5.06 16.9 2.09 110 137 2,980 224 0.0176
50 SCGL 260.55 4.42 1307.0 0.0752 0.768 5.89 20.5 2.02 125 134 3,580 263 0.0206
Volume 53, Issue 9, 2022

Existence of Supercritical “Liquid-Like” State in Subcritical Region

TABLE 3: Thermophysical properties, heat transfer rate, and pressure loss/work required for fully developed flow of water through an isothermally
heated circular pipe at various pressures under supercritical conditions (D = 10 mm, L = 1.0 m, um = 1 m/s, Tf = (Tm,i + Ts)/2)
Tm,i = 800 K, Ts = 1000 K (Tf = 900 K)
P, State ρ, kg/ μ, ×10–5
Cp, J/ k, W/m Pr ReD, ṁ, ×10–3 f, ×10–2 NuD ΔT, K q/L, ΔP/L, W/L
MPa m3 Pa-s kg K K ×104 kg/s W/m Pa/m W/m
30 SCGL 82.84 3.62 3040.8 0.1137 0.967 2.29 6.50 2.53 68.0 141 2,796 105 0.0082
40 SCGL 116.01 3.77 3403.1 0.1304 0.985 3.08 9.11 2.35 86.7 136 4,225 136 0.0107
50 SCGL 152.03 3.97 3786.9 0.1501 1.000 3.83 11.9 2.23 104 132 5,985 169 0.0133

17
18 Prasad, Kakroo, & Banerjee

1.0 and then from 1 to 10 MPa. This trend continues as the pressure is increased further, > 4 times increase between
10 and 50 MPa, although with deaviation from ideal gas. It is evident that the pressure has a substantial impact on
all variables from Re to W/L even though the rate of increase with pressure in the SC regime goes down as the fluid
temperature Tf increases, see the data for ρ in Table 1a–1d at P = 30, 40, and 50 MPa.
Not that large, but also significant is the change in the thermal conductivity; k increase of 46% between 5 and 50
MPa (SC states). On the other hand, the increases in dynamic viscosity and specific heat are only 28% and 18%, re-
spectively, and subsequent change in Pr is only 3%. Obviously, all of these properties affect heat transfer in a complex
manner, the effect of Prandtl number being the minimal.
It is known that the heat transfer rate increases significantly from subcritical to supercritical conditions. How-
ever, a simple exercise here will demonstrate the order-of-magnitude of this increase and elucidate the effectiveness
of high-pressure thermal systems. For example, the Nusselt number Nu and heat transfer rate q/L increase by a factor
of ~ 10 and ~ 40, respectively, for Case b (Table 1b), when the pressure is increased from 0.1 MPa (near ambient)
to 5 MPa (slightly above Pc), with very little increase in the work required W/L ~ 0.005 W/m, and some reduction in
the temperature rise. These increases remain as pronounced in the SC region: Nu goes up from 37.5 to 166 and q/L
increases from 323 W to 2.63 kW as P increases from 5 to 50 MPa with work required changing by only 0.03 W/m
(Table 1b). The corresponding change in the heat flux is from 193 W/m2 at 0.1 MPa to ~ 10 kW/m2 at 5 MPa and ~
20 kW/m2 at 10 MPa.
Interestingly, these effects become further pronounced as (Tm,i, Ts) are decreased from (400 K, 600 K) to (200 K,
400 K). For example, Nu goes up from 78.6 to 273 and q/L increases from 546 W to 5.2 kW as P increases from 5 to
50 MPa (Table 1a). The corresponding change in heat flux is from 397 W/m2 at 0.1 MPa to 17.4 kW/m2 at 5 MPa and
~ 36 kW/m2 at 10 MPa.
Presented also are the data in SC “gas-like” regime for P = 10, 20, 30, 40, and 50 MPa with (Tm,i, Ts) as 600 K, 800
K in Table 1c and for P = 30, 40, and 50 MPa with (Tm,i, Ts) as 800 K, 1000 K in Table 1d that can facilitate compari-
son with SC CO2 and SC water since their critical pressures are 7.377 and 22.064 MPa, respectively. Table 1c and 1d
exhibits similar behavior with respect to changes in the transport properties with SC pressure as in Table 1a and 1b.
Furthermore, a comparison of parts a to d of Table 1 reveals that the above influences of pressure in both subcritical
and SC regions are far more pronounced at systems operating at lower temperatures, such as near-ambient and cryo-
genic temperatures, since density goes up with the temperature going down.
Although the primary contribution to increase in heat transfer comes from the increase in density with pressure,
it is not unexpected that the effect of pressure and transition from subcritical to SC gaseous state becomes much
stronger at lower temperatures. In the case of Ar, this implies moving toward the cryogenic temperatures since its Tc
(–122.463°C) is below zero. Indeed, this behavior is further amplified by the increase in Cp as the pressure increases
and temperature decreases, particularly in the SC regime. To examine the combined effect, one may consider (ρCp) as
one parameter whose value changes from 171 to 569 kJ/m3 K as Tf goes down from 700 K to 300 K at 50 MPa; the
corresponding change at 5 MPa is only from 17 to 48 kJ/m3 K, but still about three times larger.
The above trend of significant increases in ρ, μ, Cp, k, and subsequently, in the mass flow rate ṁ, Reynolds number
Re, Nusselt number Nu, heat transfer rate per unit length q/L, pressure drop per unit length ΔP/L, and work required
per unit length W/L (small increase) is not restricted to Ar only but is also true for CO2 (Table 2) as long as the fluid
is under gaseous state from subcritical to supercritical state. Although not explicitly shown here for wide range of
parameters except that in Table 3, water in vapor/gaseous state behaves in a similar manner, as clearly indicated by the
80 MPa line in Fig. 5c; one can also examine this from the data on the NIST website.
The above enhancements in Nu and q are highly instructive and indicate that in a given system, heat transfer can
be tremendously enhanced by just increasing the base pressure of the system; subsequent increase in pressure loss/
pumping power will be small. It also reveals that this enhancement is more pronounced at lower temperatures, which
is great for cryogenic systems.
We do realize that very high pressures such as 30 MPa (~ 300 atm) or higher may not be reasonable in most appli-
cations currently. However, SC water-based nuclear and other energy systems have no options than to operate above
22 MPa, possibly at 30–40 MPa to take advantage of the SC pressures and temperatures. Therefore, as the materials
and sealing technologies develop and mature for SC water-based power plants, there would be immediate transfer of
these materials and technologies to other fields of energy, heat transfer, and heat exchange systems.

Heat Transfer Research

Existence of Supercritical “Liquid-Like” State in Subcritical Region 19

4.4 Heat Transfer by SC Ar, CO2, and Water

Table 2 shows that the changes in thermophysical properties and flow/heat transfer characteristics in the case of CO2 are
all similar to those observed for Ar, both in subcritical and SC regimes. A comparison with Table 1 also reveals that the
heat transfer rate q/L at any P and T for similar flow and thermal conditions is higher for CO2 than for Ar, the effect being
more pronounced at higher SC pressures, e.g., almost twice at subcritical pressures to 2.8 times at 50 MPa. This is primar-
ily because the density, thermal conductivity, and the Prandtl number are all higher for CO2 at any P and T; however, the
strongest effect is caused by the specific heat, being double or more. On the other hand, the dynamic viscosity for Ar is
generally higher than that for CO2 at all conditions – subcritical and supercritical. Table 2a–2c also supports the previous
conclusion that the increase in heat transfer rate with pressure increases as the working temperature goes down. However,
SC CO2 cannot be used for systems operating at subzero temperatures since its Tc is 30.98°C. Obviously, Ar with Tc of
–122.463°C is far more suitable fluid for SC cryogenic systems; Ar also has the lower critical pressure advantage.
Furthermore, a comparison between Table 2c for CO2 and Table 3 for water is quite revealing. It shows that the
use of SC water at 30 MPa has only a small advantage in terms of heat transfer, 20% higher for water. However, as the
SC pressure is increased the benefit of using water becomes much more pronounced, increasing to 67% difference at
50 MPa. Interestingly, water has only Cp advantage, 2 to 3 times more than that for CO2, but not the density benefit,
which may be 40–50% lower for water. Note also that k and Pr are higher for water and change significantly with SC
pressure. Needless to mention that SC water is generally far more effective in transferring heat than SC Ar.

4.5 Conditions for Substitution of Higher CP Fluid by a Lower CP Fluid

As noted earlier, CO2 has higher capacity to transfer heat at all P and T in comparison with Ar. However, it is pos-
sible to achieve the same rate of heat transfer q/L and bulk mean temperature difference (Tm,o – Tm,i) using Ar. Table 4a
presents a direct comparison at all pressures in Table 2a for CO2; the second and third columns in Table 4 are based on
q/L and (Tm,o – Tm,i) from Table 2a. To achieve identical q/L and (Tm,o – Tm,i), the only variable that needs to be changed
is the mass-flow rate which can be obtained from q = ṁCpΔT. All other quantities, e.g., um, ReD, f, NuD, ΔP/L, and W/L
can then be obtained using equations in Section 4.1. For example, um increases from 1 m/s for CO2 to 2.180 m/s at 1
MPa, 2.322 m/s at 5 MPa, and 3.125 m/s at 50 MPa for Ar. As a result, the work required also increases, e.g., 0.049,
0.107, and 0.675 W/m at P = 5, 10, and 50 MPa. As one can notice, this increase is small; not a significant cost to use
Ar given its benefits (see Section 5).
It is also interesting to examine whether q/L and (Tm,o – Tm,i) for SC CO2 can be achieved by Ar at a lower SC
pressure. Table 4b presents an example where CO2 at 10 MPa is compared with Ar at 5 MPa. For a mean flow veloc-
ity of 5.136 m/s (0.0192 kg/s) and W/L of 0.491 W/m, Ar can transfer the same amount of heat, 1.42 kW with 137 K
temperature rise, as CO2 can with 1 m/s (0.00888 kg/s) and 0.0096 W/m work.
Table 5 presents a comparison of Ar and CO2 with water under SC conditions, as in Table 3. Again, it is feasible to
achieve a similar rate of heat transfer and temperature difference between the outlet and the inlet, except that it would
require much larger flow velocity and mass-flow rate. For example, um will increase from 1 m/s to 1.194 m/s for CO2
and 3.879 m/s for Ar at 30 MPa and to 1.691 m/s for CO2 and 5.779 m/s for Ar at 50 MPa. However, this analysis
does demonstrate that there exists a choice between the lower pumping power (CO2 and water) and lower system’s
complexity and maintenance, and longer life (Ar).

5. ADVANTAGES WITH THE USE OF SC ARGON AS A HEAT TRANSFER FLUID

The major advantages of using supercritical argon as a heat transfer working fluid (Banerjee et al., 2022) are given as
follows.

5.1 Pressure and Temperature Advantage

SC Ar provides much larger range of pressure and temperature where it can be used for thermal transport in com-
parison with SC water and SC CO2 even if the Widom region with anomalous behavior is avoided. The exception

Volume 53, Issue 9, 2022

 20
TABLE 4: Comparison between Ar and CO2: Heat transfer rate and pressure loss/work required for fully developed flow through a circular pipe under
subcritical and supercritical conditions
CO2 (from Table 1b) Argon (equivalent) Tm,i = 400 K, Ts = 600 K (Tf = 500 K) Additional work w.r.
P, MPa q/L, W/m ΔT, K P, MPa ṁ, 10 kg/s u , m/s Re , 10 f, 10
–3 4 –2
Nu ΔP/L, Pa/m W/L, W/m to CO2, W/L, W/m
m D D

0.1 16.71 198 0.1 0.162 2.150 0.061 8.53 3.66 19.0 0.0032 0.0027
1 145.4 169 1 1.641 2.180 0.613 3.63 19.5 82.7 0.0142 0.013
2 286.6 163 2 3.328 2.214 1.24 2.97 34.5 139 0.0242 0.021
3 424.2 158 3 5.062 2.249 1.88 2.66 47.7 193 0.0340 0.030
4 561.1 153 4 6.844 2.285 2.53 2.47 60.0 246 0.0441 0.039
5 698.7 150 5 8.677 2.322 3.19 2.33 71.8 299 0.0545 0.049
10 1418 137 10 18.63 2.521 6.70 1.96 127 587 0.116 0.107
20 3046 124 20 41.94 2.921 14.3 1.67 231 1,300 0.299 0.282
30 4722 118 30 65.91 3.167 21.3 1.54 317 2,050 0.510 0.485
40 6149 115 40 85.66 3.207 26.1 1.48 376 2,590 0.653 0.622
50 7252 114 50 100.1 3.125 28.7 1.46 408 2,900 0.712 0.675
(a) Ar at 5 MPa vs. CO2 at 10 MPa
10 1,420 137 5 19.19 5.136 7.06 1.94 132 1,220 0.491 0.481

Prasad, Kakroo, & Banerjee

Heat Transfer Research
Volume 53, Issue 9, 2022

Existence of Supercritical “Liquid-Like” State in Subcritical Region

TABLE 5: Comparison of Ar and CO2 with water: heat transfer rate and pressure loss/work required for fully developed flow through a circular pipe
at various pressures under supercritical conditions
Water (Table 3) Tm,i = 800 K, Ts = 1000 K (Tf = 900 K) Additional work
(a) Argon (equivalent) w.r. to water,
W/L, W/m
P, MPa q/L, W/m ΔT, K ṁ, 10–3 kg/s um, m/s ReD, 104 f, 10–2 NuD ΔP/L, Pa/m W/L, W/m
30 2,796 141 34.930 4.06 8.12 1.88 148 1699 0.542 0.534
40 4,225 136 53.765 4.74 12.17 1.72 203 2804 1.044 1.033
50 5,985 132 77.946 5.53 17.19 1.61 265 4419 1.919 1.906
(b) CO2 (equivalent)
30 2,796 141 15.507 1.194 4.83 2.11 105 249 0.0233 0.0151
40 4,225 136 23.976 1.423 7.19 1.93 145 420 0.0469 0.0362
50 5,985 132 34.579 1.691 9.96 1.80 189 670 0.0890 0.0757

21
22 Prasad, Kakroo, & Banerjee

may be the applications where large changes in thermophysical properties near CP are desirable and subzero critical
temperature is not required.

5.2 Subzero Temperature Advantage

Lower working temperature gives SC Ar an added advantage of its use at ambient and cryogenic temperatures. Note
that there are many other fluids whose critical temperature and/or pressure are low such as helium (5.19 K, 0.227
MPa), hydrogen (33.20 K, 1.30 MPa), neon (44.40 K, 2.76 MPa), nitrogen (126.2 K, 3.39 MPa), fluorine (144.30
K, 5.22 MPa), oxygen (154.6 K, 5.05 MPa), methane (CH4, 190.8 K, 4.64 MPa), and krypton (209.3 K, 5.50
MPa). However, most of these fluids are expensive, environmentally nonbenign, and/or chemically not as inert as
argon.

5.3 System Complexity Advantage

In addition, the critical pressure and temperature of argon being much lower can minimize the system complexity in
comparison with the SC water, SC CO2, and others when used for heat transfer/exchange.

5.4 Avoidance of Piston Effect

When the conditions in a system are such that the supercritical fluid transits from SC conditions to subcritical condi-
tions, and vice versa, a large change in density/volume occurs, causing the piston effect (Boukari et al., 1990; Zappoli,
1998; Beysens et al., 2021). Indeed, the use of Ar may avoid the need to operate a system under near-critical and sub-
critical conditions even in the case of near-ambient temperature as is the case with SC CO2-based systems.

5.5 Nonreacting and Noncorroding

Owing to the potential of use of SC water in next generation nuclear reactors, renewable energy generation systems,
efficient treatment of organic waste, and so on, a large volume of corrosion studies have been conducted to examine
the effect of SC water on various metals and alloys. Tang et al. (2016) have presented an extensive summary of these
results (based on 120 published articles from 1994 to 2014) on iron-based alloys, Ni-based alloys, titanium-based al-
loys, ceramics, and pure metals such as zirconium, niobium, tantalum, and molybdenum. Their conclusion is that “no
single material can withstand all kinds of SCWO (SC water oxidation) conditions.” Indeed, the corrosion of materials
under harsh conditions of SC temperature and pressure and strong corrosive anions is considered a major obstacle to
large-scale applications and commercialization of SC water.
Similarly, the interaction of SC CO2 with common metal alloys has been studied extensively, due to its applica-
tion as a coolant in nuclear power plants, solar energy systems, carbon capture and storage (CCS), and others. Many
steel alloys oxidize and are subject to increased carbon content in the formation of insoluble chromium carbides at the
grain boundaries when exposed to SC CO2 (Costa et al., 2017). Indeed, localized corrosion of carbon steel has been
observed in the dynamic SC CO2-saturated aqueous phase (Wei et al., 2019), although at low temperatures it is pos-
sible that corrosion may not occur with pure SC CO2; however, water content is a key parameter. It has been observed
that temperature, water, and pollutants may play a major role in increasing the uniform or localized corrosion rates
with SC CO2 (Costa et al., 2017). A study with zinc and wet SC CO2 has also exhibited the growth of ZnCO3 and the
presence of ZnO on the zinc surface (Wei et al., 2019). A comprehensive review of SC CO2 corrosion is presented by
Sarrade et al. (2017), Cuia et al. (2019), and Kaleva et al. (2020) and it seems that under most conditions SC CO2 can
be corrosive to metals, alloys, and many other materials. However, more studies may be needed to find materials and
processing conditions which will be noncorrosive to SC CO2.
Argon, on the other hand, is exceedingly stable and chemically inert. The Ar atoms neither combine with one
another nor chemically combine with the atoms of other elements. The only exception is that Ar atoms can be trapped
mechanically in the cavities among molecules of other substances, as in ice crystals or organic compounds such as
hydroquinone, known as Ar clathrates. Evidently, SC Ar can be a wonderful fluid that would not react with or corrode

Heat Transfer Research

Existence of Supercritical “Liquid-Like” State in Subcritical Region 23

the walls of thermal systems, pipelines, and heat exchangers. Consequently, the system life can be greatly enhanced
and maintenance requirement can be significantly reduced if Ar is adopted as a heat transfer fluid.

5.6 Nonadsorbing
As demonstrated by Phadungbut et al. (2015), the absorbed phase of SCAr on a surface is confined to 1 to 2 layers and
the absolute adsorbed density approaches a constant at high pressures. This may not be true with other supercritical
fluids.

5.7 Nontoxic
It is also well known that argon is nontoxic unlike many other fluids such as ammonia, hydrocarbons, halides, and
refrigerants.

5.8 Nonpolluting
Since argon can be extracted from air and released into the air, there is no question of pollution unlike many other
fluids as well as CO2 obtained from nonambient sources.

5.9 Easy Availability

Argon, the third most abundant gas besides nitrogen and oxygen, is 0.934% of the atmosphere. It can be easily
produced by air liquefaction as part of the process to extract O2 and N2. Note that the boiling temperatures of liquid
oxygen, argon, and nitrogen are 90.2, 87.3, and 77.3 K, respectively; that means it can be evaporated from liquefied
air after N2 and before O2.

5.10 Marginally Higher Operational Cost

The operating cost in terms of pressure loss and work required in forced convection systems would be marginally
increased (Section 4) in comparison with other SC fluids such as CO2. However, almost no cleaning will be required
and the life of the system would be greatly enhanced since Ar is nonreacting and noncorroding.

6. CONCLUDING REMARKS
This research highlights many features of the transition from subcritical liquid to supercritical fluid, the anomalous
behavior during this transition, and enhancement in heat transfer as pressure is increased to the critical point and then
to higher SC pressures. It also presents, for the first time, supercritical argon as a heat transfer fluid that can offer many
advantages over other SC fluids.

6.1 Existence of Supercritical “Liquid-Like” State in Subcritical Region

• It is now well established that SC fluids have two states – “liquid-like” and “gas-like,” and their thermophysi-
cal properties exhibit anomalous behavior in the near-critical region beyond the critical point (Pc, Tc), now
being referred to as the “Widom region.” We have convincingly re-emphasized that the anomalous behavior,
as observed above the critical point, extends to pressures and temperatures much below (Pc, Tc), and a “SC
liquid-like” state exists within the subcritical region where sharp changes in property occur. The “SC liquid-
like” state in the subcritical region and the “liquid-like” state in the SC region may be treated continuous;
however, the property variations may be large.
• The liquid state therefore needs to be divided into two regions: “regular liquid” as we know, and “SC liquid-
like,” where changes are anomalous. The pressure and temperature where the “SC liquid-like state” begins

Volume 53, Issue 9, 2022

24 Prasad, Kakroo, & Banerjee

lies between the triple point and the critical point on a P–T phase diagram. This is much more pronounced in
the case of water in comparison to CO2 and Ar.
• In the “SC gas-like” state, where the properties exhibit monotonic trends, the existing convection heat trans-
fer correlations may be valid. However, this needs extensive experimental validation. The thermal transport
calculations and analysis will be simpler if the transition occurs directly from the subcritical “gaseous” state
to the SC “gas-like” state.
• It is important that the “regular liquid”/“SC liquid-like” crossover line in the subcritical region and the SC
“liquid-like”/“gas-like” crossover line (PGQ) in the supercritical region (Fig. 6) be demarcated.
• Demarcating the conditions when fluid in the gaseous state directly moves into the SC “gas-like” state will
also be very helpful in the analysis and design of thermal systems which does not involve liquid phase, e.g.,
the cases considered for heat transfer calculations in Section 4.

6.2 Orders-of-Magnitude Enhancement in Heat Transfer

• The rate of convective heat transfer increases as the pressure increases and the fluid moves from the gaseous
state to the SC “gas-like” state. Indeed, a two-to-three orders-of-magnitude enhancement as compared to the
heat transfer at atmospheric pressure is possible if the SC pressure becomes high.
• The lower the working temperature, the larger is the increase in heat transfer.
• Interestingly, in the SCGL state the dynamic viscosity decreases with temperature but the pressure has an
opposite effect.
• The increase in heat transfer is directly related to increases in density and specific heat with increasing pres-
sure and decreasing temperature; the effects of other properties are smaller but still important.
• The increase in pressure loss and pumping power required to achieve SC conditions and higher rate of heat
transfer are not very significant.
• For given SC working pressure and temperature, SC water will yield higher rates of heat transfer than SC
CO2; SC Ar will have the lowest values among the three. However, the critical pressure and temperature for
Ar with 4.863 MPa and –122.463°C are much lower than those of water (22.064 MPa, 373.95°C) and CO2
(7.377 MPa, 30.98°C), which can give Ar significant advantages in many applications.
• By increasing the mass flow rate and work required, argon with lower critical pressure and critical tempera-
ture can match the heat transfer by SC CO2, or even SC water.

6.3 Argon as an Ideal Fluid for Supercritical Heat Transfer

• Argon is a noble gas. It is nonreacting, noncorroding, nondiffusive (through system walls), and nontoxic. It
is also environmentally benign.
• It is easily available (0.934% of air) and can be obtained using the air liquefaction process that is commer-
cially used to produce oxygen and nitrogen. Since it can be evaporated from liquefied air between nitrogen
and oxygen, the cost to produce argon is low.
• Critical pressure and temperature of argon are much lower than those of many other fluids such as ammonia,
carbon dioxide, and water. This can provide a major advantage of being in the SC “gas-like” state for much
larger range of pressure and temperature, even beyond the Widom region.
• It can be easily used as a supercritical fluid at cryogenic temperatures without the requirement of liquid–va-
por phase change.
• Since argon would not require liquid–vapor phase change beyond ~ 200 K, the challenges caused by high
heat flux and dryout would be nonexistent. Also nonexistent would be the piston effect.
• As noted earlier, SC argon can match or exceed the supercritical heat transfer rate of other fluids, such as SC
CO2 and water by simply increasing the mass flow rate and flow work either at the same pressure or even
lower pressures.
• Thermal systems using SC Ar can work at much lower pressures and/or temperatures, and therefore, would
reduce system complexity and maintenance, and have longer life.

Heat Transfer Research

Existence of Supercritical “Liquid-Like” State in Subcritical Region 25

ACKNOWLEDGMENTS
Acknowledgements are due to Surendra Saxena (Professor Emeritus, Florida International University and Uppsala
University) for helpful discussions on supercritical fluids, to Rajiv Mishra (Professor, Materials Science and Engineer-
ing, UNT) for comments on materials with high thermal conductivity and high fatigue-resistance for high pressure
applications, and to Laura Almara for graphics of Fig. 6.

REFERENCES
Anderson, M., Materials, Turbomachinery and Heat Exchangers for Supercritical CO2 Systems, Reactor Concepts RD&D, Project
No. 09-778, US Department of Energy, 2012.
Andrews, T., The Bakerian Lecture: On the Continuity of the Gaseous and Liquid States of Matter, Phil. Trans. Roy. Soc. London,
vol. 159, pp. 575–590, 1869.
Bamido, A., Prasad, V., and Banerjee, D., Comparison of Forced Convection Heat Transfer Properties of Supercritical Carbon
Dioxide and Supercritical Argon in a Circular Horizontal Tube, Under Preparation, 2022
Bamido, A., Pyle, T., Thyagarajan, A., Dhir, V.K., Prasad, V., and Banerjee, D., A Numerical Study of Forced Convective Heat
Transfer Characteristics of Supercritical Fluid in a Horizontal Circular-Pipe, Proc., ASME Summer Heat Transfer Conf., Vir-
tual, 2020.
Beysens, D.A., Garrabos, Y., and Zappoli, B., Thermal Effects in Near-Critical Fluids: Piston Effect and Related Phenomena, Chap-
ter 1, Handbook of Research on Advancements in Supercritical Fluids Applications for Sustainable Energy Systems, Hershey,
PA: IGI Global, pp. 1–31, 2021.
Banerjee, D., Prasad, V., and Mishra, R., Method and System for Supercritical Argon as a Working Fluid, Provisional Patent, Ap-
plication No. 63/275,143, 2022.
Bodkha, K. and Maheshwari, N.K., Heat Transfer in Supercritical Fluids: A Review, J. Nucl. Eng. Rad. Sci., vol. 7, no. 3, pp. 1–19,
2021.
Bolmatov, D., Zav’yalov, D., Gao, M., and Zhernenkov, M., Structural Evolution of Supercritical CO2 across the Frenkel Line, J.
Phys. Chem. Lett., vol. 5, pp. 2785−2790, 2014.
Bolmatov, D., Zhernenkov, M., Zav’yalov, D., Tkachev, S.N., Cunsolo, A., and Cai, Y.Q., The Frenkel Line: A Direct Experimental
Evidence for the New Thermodynamic Boundary, Sci. Rep., vol. 5, pp. 1–10, 2015.
Boukari, H., Briggs, M.E., Shaumeyer, J.N., and Gammon, R.W., Critical Speeding Up Observed, Phys. Rev. Lett., vol. 65, no. 21,
pp. 2654–2657, 1990.
Brazhkin, V.V. and Trachenko, K., What Separates a Liquid from Gas? Phys. Today, vol. 65, p. 68, 2012.
Brazhkin, V.V., Fomin, Y.D., Lyapin, A.G., Ryzhov, V.N., and Trachenko, K., Two Liquid States of Matter: A Dynamic Line on a
Phase Diagram, Phys. Rev. E, vol. 85, pp. 031203/1–12, 2012.
Brazhkin, V.V., Fomin, Y.D., Lyapin, A.G., Ryzhov, V.N., and Tsiok, E.N., Widom Line for the Liquid–Gas Transition in Lennard–
Jones System, J. Phys. Chem. B, vol. 115, no. 48, pp. 14112–14115, 2011.
Brazhkin, V.V., Prescher, C., Fomin, Y.D., Tsiok, E.N., Lyapin, A.G., Ryzhov, V.N., Prakapenka, V.B., Stefanski, J., Trachenko,
K., and Sapelkin, A.V., Comment on Behavior of Supercritical Fluids across the Frenkel Line, J. Phys. Chem. B, vol. 122, pp.
6124–6128, 2018.
Bryk, T., Gorelli, F.A., Mryglod, I., Ruocco, G., Santoro, M., and Scopigno, T., Behavior of Supercritical Fluids across the ‘Frenkel
Line’, J. Phys. Chem. Lett., vol. 8, no. 20, pp. 4995–5001, 2017.
Buchner, E.G., Die beschrankte Mischbarkeit von Flussigkeiten das System Diphenyamin und Kohlensaure, Z. Phys. Chem., vol.
56, p. 257, 1906.
Cabeza, L.F., Gracia, A.D., Fernandez, A.I., and Farid, M.M., Supercritical CO2 as Heat Transfer Fluid: A Review, Appl. Therm.
Eng., vol. 125, pp. 799–810, 2017.
Cagniard de la Tour, C., Exposé de Quelques Résultats Obtenu par l’action Combinée de la Chaleur et de la Compression sur
Certains Liquides, Tels que l’eau, l’alcool, l’éther Sulfurique et l’essence de Pétrole Rectifiée, Ann. Chim. Phys., vol. 21, pp.
127–132, 1822.
Chen, Q.S., Chatterjee, A., Prasad, V., and Larkin, J., A Porous Media-Based Transport Model for Hydrothermal Growth, J. Crystal
Growth, vols. 198/199, pp. 710–715, 1999a.
Chen, Q.-S., Prasad, V., and Chatterjee, A., Modeling of Fluid Flow and Heat Transfer in a Hydrothermal Crystal Growth System:
Use of Fluid-Superposed Porous Layer Theory, ASME J. Heat Transf., vol. 121, pp. 1049–1058, 1999b.
Chen, Q.-S., Jiang, Y., Yan, J., Li, W., and Prasad V., Modeling of Ammonothermal Growth Processes of GaN Crystal in Large-Size
Pressure Systems, Res. Chem. Intermediates, vol. 37, pp. 467–477, 2011.

Volume 53, Issue 9, 2022

26 Prasad, Kakroo, & Banerjee

Chen, Q.-S., Pendurti, S., and Prasad, V., Modeling of Ammonothermal Growth of Gallium Nitride Single Crystals, J. Mater. Sci.,
vol. 41, pp. 1409–1414, 2006a.
Chen, Q.-S., Zhang, H., and Prasad, V., Physics and Modeling of Ammonothermal Growth of Gallium Nitride Single Crystals, in
Focus on Crystal Growth Research, G.V. Karas, Ed., New York: Nova Science Publishers, pp. 151–169, 2006b.
Chen, Q.-S., Prasad, V., and Hu, W.R., Modeling of Ammonothermal Growth of Nitrides, J. Crystal Growth, vol. 258, pp. 181–187,
2003.
Chen, X. and Schulenberg, T., Heat Transfer at Supercritical Pressures – Literature Review and Application to HPWR, Forchun-
gszentrum Karlsruhe GmbH, 2001.
Cockrell, C., Dicks, O.A., Brazhkin, V.V., and Trachenko, K., Pronounced Structural Crossover in Water at Supercritical Pressures,
J. Phys.: Condens. Matter, vol. 32, pp. 385102–385109, 2020.
Costa, G.C.C., Jacobson, N.S., Hunter, G.W., Nakley, L., Radoman-Shaw, B.G., Lukco, D., and Harvey, R.P., Chemical and Mi-
crostructural Changes in Metallic and Ceramic Materials Exposed to Venusian Surface Conditions, NASA Glenn Research
Center, Cleveland, OH, Tech. Rep., 2017.
Cuia, G., Yanga, Z., Liua, J., and Lia, Z., A Comprehensive Review of Metal Corrosion in a Supercritical CO2 Environment, Int. J.
Greenhouse Gas Control, vol. 90, pp. 1–17, 2019.
Deissler, R.G. and Cleveland, O., Heat Transfer and Fluid Friction for Fully Developed Turbulent Flow of Air and Supercritical
Water with Variable Fluid Properties, ASME J. Heat Transf., vol. 76, pp. 73–85, 1954.
Ely, J.F. and Baker, J.K., A Review of Supercritical Fluid Extraction, National Bureau of Standards, US Department of Commerce,
1983.
Faraday, M., On the Liquefaction and Solidification of Bodies Generally Existing as Gases, Phil. Trans. Roy. Soc. London, vol.
135, pp. 155–177, 1845.
Fomin, Y.D., Ryzhov, V.N., Tsiok, E.N., Brazhkin, V.V., and Trachenko, K., Thermodynamics and Widom Lines in Supercritical
Carbon Dioxide, Cond-Mat. Stat.-Mech., pp. 1–6, 2014.
Fomin, Y.D., Ryzhov, V.N., Tsiok, E.N., and Brazhkin, V.V., Thermodynamic Properties of Supercritical Carbon Dioxide: Widom
and Frenkel Lines, Phys. Rev. E, vol. 91, pp. 22111–22116, 2015.
Gorelli, F.A., Bryk, T., Krisch, M., Ruocco, G., Santoro, M., and Scopigno, T., Dynamics and Thermodynamics beyond the Critical
Point, Sci. Rep., vol. 3, pp. 1–5, 2013.
Gorelli, F.A., Santoro, M., Scopigno, T., Krisch, M., and Ruocco, G., Liquid-like Behavior of Supercritical Fluids, Phys. Rev. Lett.,
vol. 97, no. 24, Article ID 245702, 2006.
Hannay, J.B. and Hogarth, J., On the Solubility of Solids in Gases, Proc. Roy. Soc. London, vol. 29, pp. 324–326, 1879.
Imre, A.R., Deiters, U.K., Kraska, T., and Tiselj, I., The Pseudocritical Regions for Supercritical Water, Nucl. Eng. Design, vol.
252, pp. 179–183, 2012.
Imre, A.R., Groniewsky, A., Györke, G., Katona, A., and Velmovszki, D., Anomalous Properties of Some Fluids − with High Rel-
evance in Energy Engineering – In Their Pseudo-Critical (Widom) Region, Periodica Polytech. Chem. Eng., vol. 63, no. 2,
pp. 276–285, 2019.
Kaleva, A., Tassaing, T., Saarimaa, V., Le Bourdon, G., Väisänen, P., Markkula, A., and Levänen, E., Formation of Corrosion Prod-
ucts on Zinc in Wet Supercritical and Subcritical CO2: In Situ Spectroscopic Study, Corrosion Sci., vol. 174, pp. 1–10, 2020.
Kalsia, M., Dondapati, R.S., and Usurumarti, P.R., Statistical Correlations for Thermophysical Properties of Supercritical Argon
(SCAR) Used in Cooling of Futuristic High Temperature Superconducting Cable, Physica C: Superconduct. Appl., vol. 536,
pp. 30–36, 2017.
Knez, Z., Marko, E., Leitgeb, M., Primo, M., Knez, M., and Skerget, M., Industrial Applications of Supercritical Fluids: A Review,
Energy, vol. 77, pp. 235–241, 2014.
Levelt Sengers, J.M.H.L., Supercritical Fluids: Their Properties and Applications, Chapter 1, Supercritical Fluids, E. Kiran, P.G.
Debenedetti, and C.J. Peters, Eds., The Netherlands: Kluwer Academic Publishers, vol. 366, pp. 1–29, 2000.
Mokry, S., Pioro, I., Farah, A., King, K., Gupta, S., Peiman, W., and Kirillov, P., Development of Supercritical Water Heat-Transfer
Correlation for Vertical Bare Tubes, Nucl. Eng. Des., vol. 241, no. 4, pp. 1126–1136, 2011.
NIST, Chemistry WebBook, SRD 69, Department of Commerce, 2021.
Palmer, M.V. and Ting, S.S.T., Applications for Supercritical Fluid Technology in Food Processing, Food Chem., vol. 52, pp.
345–352, 1995.
Pendurti, S., Chen, Q.-S., and Prasad, V., Modeling of Ammonothermal Growth of GaN Single Crystals: The Role of Transport, J.
Crystal Growth, vol. 88, no. 20, Article ID 201908, 2006.
Phadungbut, P., Fan, C., Do, D.D., Nicholson, D., and Tangsathitkulchai, C., Determination of Absolute Adsorption for Argon on
Flat Surfaces under Sub- and Supercritical Conditions, Colloids Surfaces A: Physicochem. Eng. Aspects, vol. 480, pp. 19–27,
2015.

Heat Transfer Research

Existence of Supercritical “Liquid-Like” State in Subcritical Region 27

Pioro, I., Applications of Supercritical Power Engineering: Specifics of Thermophysical Properties and Forced-Convection Heat
Transfer, Supercritical Fluid Technology in Energy and Environmental Application, Amsterdam: Elsevier, pp. 201–233, 2014.
Pioro, I.L. and Duffey, R.B., Heat Transfer and Hydraulic Resistance at Supercritical Pressures in Power Engineering Applica-
tions, New York: ASME Press, 2007.
Pioro, I.L. and Duffey, R.B., Literature Survey of Heat Transfer and Hydraulic Resistance of Water, Carbon Dioxide, Helium and
Other Fluids at Supercritical and Near-Critical Pressures, Chalk River Laboratories, Tech. Rep. AECL -12137, FFC-FCT-409,
2003.
Pioro, I.L., Current Status of Research on Heat Transfer in Forced Convection of Fluids at Supercritical Pressures, Nucl. Eng.
Design, vol. 354, pp. 1–14, 2019.
Pioro, I.L., Ed., Handbook of Generation IV Nuclear Reactors, Sawston, UK: Woodhead Publishing, 2016.
Pioro, I.L., Khartabil, H.F., and Duffey, R.B., Heat Transfer to Supercritical Fluids Flowing in Channels—Empirical Correlations
Survey, Nucl. Eng. Des., vol. 230, no. 1, pp. 69–91, 2004.
Pioro, L.S. and Pioro, I.L., Industrial Two-Phase Thermosyphons, New York: Begell House, 1997.
Proctor, J.E., The Liquid and Supercritical Fluid States of Matter, Boca Raton, FL: CRC Press, 2021.
Sarrade, S., Féron, D., Rouillard, F., Perrin, S., Robin, R., Ruiz, J.C., and Turc, H.A., Overview on Corrosion in Supercritical Flu-
ids, J. Supercrit. Fluids, vol. 120, pp. 335–344, 2017.
Schilling, W. and Kiritani, M., Eds., Diffusion of Helium in Metals, Tokyo: University of Tokyo Press, 1982.
Schmidt, E., Eckert, E., and Grigull, V., Heat Transfer by Liquids near the Critical State, AFF Translation, No. 527, Air Materials
Command, Wright Field, Dayton, OH, USA, 1946.
Schmidt, E., Wärmetransport durch Natürliche Konvection in Stoffen bei Kritischen Zustand, Int. J. Heat Mass Transf., vol. 1, no.
1, pp. 92–101, 1960.
Schmidt, K.R., Thermal Investigations with Heavily Loaded Boiler Heating Surface, Mitt. Ver. Grosselbetr, no. 63, 1959.
Shitsi, E., Debrah, S.K., Agbodemegbe, V.Y., and Ampomah-Amoako, E., Performance of Heat Transfer Correlations Adopted at
Supercritical Pressures: A Review, World J. Eng. Technol., vol. 6, pp. 241–267, 2018.
Shitsman, M.E., Impairment of the Heat Transmission at Supercritical Pressures, High Temp., vol. 1, pp. 237–244, 1963.
Simeoni, G.G., Bryk, T., Gorelli, F.A., Krisch, M., Ruocco, G., Santoro, M., and Scopigno, T., The Widom Line as the Crossover
between Liquid-Like and Gas-Like Behavior in Supercritical Fluids, Nat. Phys., vol. 6, pp. 503–507, 2010.
Tang, X., Wang, S., Qian, L., Ren, M., Sun, P., Li, Y., and Yang, J.Q., Corrosion Properties of Candidate Materials in Supercritical
Water Oxidation Process, J. Adv. Oxid. Technol., vol. 19, no. 1, pp. 141–157, 2016.
Wei, L. and Gao, K., Understanding the General and Localized Corrosion Mechanisms of Cr Containing Steels in Supercritical
CO2-Saturated Aqueous Environments, J. Alloys Compd., vol. 792, pp. 328–340, 2019
Xu, L., Kumar, P., Buldyrev, P., Chen, S.H., Poole, P.H., Sciortino, F., and Stanley, H.E., Relation between the Widom Line
and the Dynamic Crossover in Systems with a Liquid–Liquid Phase Transition, Proc. Nat. Acad. Sci., vol. 102, no. 46, pp.
16558–16562, 2005.
Zappoli, B., Influence of Convection on the Piston Effect, Int. J. Thermophys., vol. 19, no. 3, pp. 803–815, 1998.
Zhang, S., Xu, X., Liu, C., and Dang, C., A Review on Application and Heat Transfer Enhancement of Supercritical CO2 in Low-
Grade Heat Conversion, Appl. Energy, vol. 269, pp. 1–28, 2020.

Volume 53, Issue 9, 2022