Article

Optimisation of Heat Exchanger Performance Using Modified

Gyroid-Based TPMS Structures
Martin Beer * and Radim Rybár

Faculty of Mining, Ecology, Process Technologies and Geotechnology, Institute of Earth Sources, Technical
University of Kosice, Letna 9, 04200 Kosice, Slovakia; [Link]@[Link]
* Correspondence: [Link]@[Link]

Abstract: Triply periodic minimal surfaces (TPMS) represent an innovative approach to the design
of heat exchangers, enabling the optimisation of thermal and hydraulic performance. This study
presents a comparative analysis of three geometric TPMS configurations: sheet gyroid, skeletal gyroid,
and the newly proposed combined gyroid geometry. Using numerical analysis based on simulations
of fluid flow and heat transfer, key parameters such as the heat transfer coefficient, Nusselt number,
friction factor, Chilton–Colburn j-factor, and pressure drop were evaluated. The results demonstrated
that the combined gyroid geometry achieves the highest heat transfer efficiency, exhibiting significant
improvements in the Nusselt number and heat transfer coefficient across the entire flow range.
Simultaneously, it maintains low pressure losses, making it well suited for applications demanding
high thermal performance with minimal energy losses. This study highlights the potential of TPMS
geometries for optimising heat exchanger design and opens new paths for their implementation in
industrial systems.

Keywords: triply periodic minimal surface; gyroid; numerical analysis; heat transfer; friction factor;
pressure drop

1. Introduction
Heat transfer and the efficiency of thermal exchange processes represent key challenges
Citation: Beer, M.; Rybár, R.
in contemporary engineering research, with their optimisation being critical across a wide
Optimisation of Heat Exchanger
range of industrial and technological applications. From the energy sector and cooling
Performance Using Modified
systems to advanced microelectronic devices, the ever-increasing demands for energy
Gyroid-Based TPMS Structures.
efficiency and thermal performance impose unprecedented pressure on the development of
Processes 2024, 12, 2943. https://
[Link]/10.3390/pr12122943
innovative materials and design strategies. Over the past decades, triply periodic minimal
surfaces (TPMS) have emerged as a highly promising tool for enhancing thermal exchange
Academic Editor: Udo Fritsching and optimising fluid dynamics. These complex geometric structures, characterised by
Received: 29 November 2024
minimal mean curvature and spatially periodic arrangements, offer unique properties that
Revised: 17 December 2024 significantly improve thermal conductivity and heat flow distribution [1–5].
Accepted: 21 December 2024 Scientific research unequivocally highlights the potential of TPMS in thermal man-
Published: 23 December 2024 agement contexts. The work of Schoen et al. [6] clearly demonstrated that TPMS-derived
structures, such as the gyroid, Schwarz-P, and diamond, exhibit remarkably favourable
heat transfer characteristics compared to conventional designs. These geometric config-
urations enable increased surface area, improved turbulent flow, and optimised thermal
Copyright: © 2024 by the authors. resistance, directly contributing to overall thermal efficiency enhancement [7–10]. Further-
Licensee MDPI, Basel, Switzerland.
more, advances in manufacturing technologies, such as 3D printing and micromachining,
This article is an open access article
allow for the precise and complex fabrication of TPMS structures with high accuracy and
distributed under the terms and
repeatability [11–14]. This progress opens new horizons for implementing these geometries
conditions of the Creative Commons
in various heat exchangers, cooling systems, and thermoelectric devices.
Attribution (CC BY) license (https://
Despite significant advancements, several critical research questions remain unre-
[Link]/licenses/by/
4.0/).
solved. Specifically, in-depth analyses of the influence of TPMS topological parameters on

Processes 2024, 12, 2943. [Link] [Link]

Processes 2024, 12, 2943 2 of 21

heat transfer, quantitative modelling of thermal flows within these structures, and optimisa-
tion of their manufacturing processes are required. One promising path for expanding the
utility of TPMS involves modifying the porous structure through recursive function-based
geometry creation or employing Boolean operations to combine different types of TPMS
geometries. Alternatively, linking multiple configurations of a specific TPMS type using
varying input parameters defining the fundamental shape offers additional potential.
Kerme et al. [15], in an experimental study, investigated three gyroid structures with
differing cell sizes and porosities. The analysis of the thermal and hydraulic properties re-
vealed that the Gyroid-A structure (15 mm/0.75) achieved outstanding thermal parameters.
Compared to Gyroid-C, it exhibited a 51.5% higher Nusselt number, 45.4% lower thermal
resistance, and 75.9% reduced total entropy. Smaller cell sizes and lower porosity enabled
a more uniform temperature distribution and more efficient heat transfer. Moradamand
et al. [16] studied flow and heat transfer in heat exchangers with Schwarz-P and gyroid
structures of varying cell numbers using numerical and experimental methods. Their analy-
sis spanned laminar (Re = 10–100) and turbulent (Re = 1850–5500) flow regimes. Numerical
results demonstrated that a gyroid structure with 64 cells achieved a 24% improvement
in performance criteria and a 40% increase in the Nusselt number compared to standard
tubular heat exchangers, while Schwarz-P showed no significant improvement. Prototypes
were produced via 3D printing.
A study by Pulin et al. [17] examined the efficiency enhancement of heat exchangers
through the introduction of turbulent lattice structures based on gyroid TPMS geometries
fabricated with additive manufacturing. Numerical heat transfer modelling demonstrated
that lattice structures intensify heat exchange by generating vortex flow structures and
increasing the heat transfer surface area. Specifically, the results indicated up to a 3.03-fold
increase in the convection-to-conduction ratio. Comparative analysis showed that the
designed turbulators were more effective than conventional flow turbulators, albeit with
higher pressure drops. This research highlights the potential of additive manufacturing
technologies in optimising heat exchangers. Kus et al. [18] presented a novel heat ex-
changer concept with a TPMS-based core, specifically a gyroid lattice produced using laser
powder bed fusion technology. Numerical analyses conducted under various operating
conditions aimed to determine the optimal unit cell size. Compared to a 30% larger plate
heat exchanger, the gyroid exchanger achieved an average of 10.5% higher transfer units
(NTU) and 5% greater thermal efficiency. This study emphasises the potential of advanced
geometric structures and additive manufacturing technologies in heat exchanger design.
Yeranee et al. [19] provided a comprehensive perspective on TPMS utilisation in thermal
engineering and the geometric factors affecting heat exchange processes in devices em-
ploying TPMS structures. Their analysis revealed that changes in design variables, such
as porosity, wall thickness, and cell size, significantly impact flow characteristics and heat
transfer performance.
In recent years, additive manufacturing technologies have significantly advanced,
enabling the precise fabrication of complex structures such as triply periodic minimal
surfaces (TPMS). Materials used in 3D printing, such as metal powders (e.g., aluminium,
copper, or stainless steel) processed via selective laser melting (SLM) or stereolithography
(SLA), exhibit high thermal conductivity. For instance, copper achieves thermal conduc-
tivity values of up to 390 W/(m·K), making it highly suitable for thermal management
applications. Additionally, modern manufacturing methods allow for minimising surface
roughness. Through post-processing techniques such as electrolytic polishing, surfaces
can achieve average roughness (Ra) values below 1 µm. Such smooth surfaces reduce
pressure losses and enhance heat transfer efficiency between the working fluid and the heat
exchanger structure.
Specifically, for TPMS structures, manufacturing processes are optimised to preserve
intricate geometric details while ensuring mechanical strength and resistance to deforma-
tion. Significant advancements have also been made in the development of thermoplastic
filaments infused with metal particles, combining easier printing capabilities with adequate
Processes 2024, 12, 2943 3 of 21

thermo-mechanical properties. These technological innovations confirm the suitability

of 3D-printed materials for high-performance heat transfer applications that meet the
demanding requirements of modern thermal management systems.
The above literature review highlights a promising direction in TPMS research, focus-
ing on modifying fundamental parameters, such as porosity, unit cell number, and wall
thickness. The present study concentrates on the numerical analysis and comparative eval-
uation of three TPMS structures concerning fundamental hydraulic and thermal-technical
characteristics. The compared structures are derived from the gyroid, specifically the
skeletal gyroid, sheet gyroid, and a geometry proposed by authors that combines these
essential configurations. The essential impetus for developing this novel approach to
creating a TPMS-based heat exchange element was the desire to eliminate the negative
Processes 2025, 13, x FOR PEER REVIEW 4 of 22
effects of porous TPMS structures on the flowing medium, particularly in terms of reducing
friction factor and pressure drop, while simultaneously enhancing or at least maintaining
the thermal-technical properties of the modified geometry.
a lower density of larger pores. The interconnected pores in the resultant geometry form
channels
2. Materialsthatand
facilitate
Methodsfluid flow through the structure.
The foundation
The basic sheetofgyroid geometry
numerical is lies
analysis characterised by uniform
in a trio of geometric flow,derived
bodies predominantly
from the
laminar in nature, and by the separation of flow into two non-intersecting
principles of TPMS (see Figure 1). In the initial step, a gyroid surface was generated groups of based
chan-
nels.
on In contrast,
Equation the skeletal
(1) [20], gyroid as
which served exhibits
the basisa higher
for allmaterial volume
subsequent and increased
geometric tur-
operations
performed duringinthe
bulence, evident thecreation of the three-dimensional
larger cross-sectional bodies.
channels. The The structure
hybrid fundamental unit cell
designed by
coordinates in the XY,aXZ,
the authors leverages and YZ planes
combination of thesewere π × π to
features × minimise
π. A detailed description
pressure loss andoffric-
the
process for creating
tion factor the baseefficient
while ensuring geometry is provided
heat exchangeinbetween
[21]. the gyroid material and the
flowing medium.
sin x cos y + sin y cos z + sin z cos x = 0 (1)

Figure 1. A visual representation of the creation process for the combined gyroid TPMS structure,
Figure [Link]
derived A visual representation
the sheet of thewith
gyroid structure creation
a sizeprocess forthe
of 2π and theskeletal
combined gyroid
gyroid TPMSwith
structure structure,
a size
derived
of π. from the sheet gyroid structure with a size of 2π and the skeletal gyroid structure with a
size of π.
In the subsequent step towards achieving the final shape, volumetric bodies were
derived from the two-dimensional gyroid surface. These included the sheet gyroid, com-
Numerical simulations were conducted using Solidworks Flow Simulation 2018 soft-
posed of a unit cell with coordinates 2π × 2π × 2π, a wall thickness of 0.5 mm, and
ware, employing a steady-state study with the k-ε viscosity model.
overall dimensions of 20 × 20 × 20 mm, and the skeletal gyroid, based on a unit cell of
π×π The
× preference
π, filling a for the k-εofmodel
volume 20 × 20for ×solving
20 [Link] These
numerical simulations
volumetric bodiespresented
were sub-in
this article replicated
sequently was basedalong
on several assumptions.
the x-axis to createThethefirst
finalisgeometries
the robustness
withand universality
dimensions of
of ×
20 this
20model
× 80 mm across a broader
(width range
× height of Reynolds numbers, as well as a wider range of
× length).
The turbulent
possible heat exchange surface areainfor
manifestations thetheflowing
sheet gyroid
medium. is 19,045.54 mm2assumption
The second , for the combined
is the
gyroid it isreduction
10,412.72in 2
m computational
, and for the skeletal gyroid 2
dramatic demands and,it therefore,
is 9068.49 the
mmrequired
. The compactness
time com-
values,
pared toi.e.,
the the
use surface
of the SSTarea
k-ωtomodel.
volume ratio, are
However, weasacknowledge
follows: for that
the the
sheetk-εgyroid, it is
model may
overestimate heat transfer in the low Reynolds number regime due to its inherent assump-
tions regarding turbulent viscosity. For this range of Reynolds numbers, the SST k-ω
model could be more accurate, as it better accounts for the flow properties in viscous sub-
layers, thus providing more precise results for thermal-hydraulic simulations under low-
Processes 2024, 12, 2943 4 of 21

0.595 mm2 /mm3 ; for the combined gyroid, it is 0.325 mm2 /mm3 ; and for the skeletal
gyroid, it is 0.283 mm2 /mm3 .
The geometry proposed by the authors, designed to mitigate the negative effects of
increased pressure loss and friction factor, utilises a combination of these two structures.
Boolean operations were employed to achieve the final form, whereby the skeletal gyroid
volume was subtracted from the sheet gyroid volume. This resulted in a hybrid geometry
combining two structural types: one with a high density of small pores and the other with
a lower density of larger pores. The interconnected pores in the resultant geometry form
channels that facilitate fluid flow through the structure.
The basic sheet gyroid geometry is characterised by uniform flow, predominantly
laminar in nature, and by the separation of flow into two non-intersecting groups of
channels. In contrast, the skeletal gyroid exhibits a higher material volume and increased
turbulence, evident in the larger cross-sectional channels. The hybrid structure designed
by the authors leverages a combination of these features to minimise pressure loss and
friction factor while ensuring efficient heat exchange between the gyroid material and the
flowing medium.
Numerical simulations were conducted using Solidworks Flow Simulation 2018 soft-
ware, employing a steady-state study with the k-ε viscosity model.
The preference for the k-ε model for solving the numerical simulations presented in
this article was based on several assumptions. The first is the robustness and universality of
this model across a broader range of Reynolds numbers, as well as a wider range of possible
turbulent manifestations in the flowing medium. The second assumption is the dramatic
reduction in computational demands and, therefore, the required time compared to the use
of the SST k-ω model. However, we acknowledge that the k-ε model may overestimate
heat transfer in the low Reynolds number regime due to its inherent assumptions regarding
turbulent viscosity. For this range of Reynolds numbers, the SST k-ω model could be more
accurate, as it better accounts for the flow properties in viscous sublayers, thus providing
more precise results for thermal-hydraulic simulations under low-turbulence conditions.
Nevertheless, we believe that the assessed boundary conditions and related characteristics
allow for the use of the k-ε model. This study primarily focuses on comparing geometric
configurations and their thermal characteristics, with the model selection justified by its
widespread acceptance in numerical analyses of TPMS geometric structures in the existing
literature [16,22–24].
The presented numerical study utilised commercially available software that provides
a balanced set of tools for CAD modelling, physical simulations, and subsequent result
integration into the additive manufacturing process. During the numerical analysis, no
user-defined functions or custom code modifications to the solution methods were applied.
The solution framework was based on the finite volume method, with convergence criteria
set to 10−6 . In the domain of fluid flow, fundamental mathematical approaches were
employed, grounded in the system of Navier–Stokes equations, which encapsulate the
conservation laws of mass, momentum, and energy [24].

∂ρ ∂(ρui )
+ = 0, (2)
∂t ∂xi

∂(ρui ) ∂
∂p ∂  
+ ρui u j + = τij + τijR + Si , (3)
∂t ∂x j ∂xi ∂x j
∂ρH ∂ρui H ∂     ∂p ∂u
+ = u j τij + τijR + qi + − τijR i + ρε + Si ui + Q H , (4)
∂t ∂xi ∂xi ∂t ∂x j
where u is the fluid velocity, ρ is the density, Si represents body forces distributed through
the medium (e.g., gravity or rotational forces), H = h+ u2 /2 is the total enthalpy, h is the
specific enthalpy, QH is the heat source per unit volume, τ ij is the viscous stress tensor, τ ij R
is the Reynolds stress tensor, and qi represents the diffusive heat flux. The indices denote
summation over the three coordinate directions. Heat transfer in solids and fluids, along
 where u is the fluid velocity, ρ is the density, Si represents body forces distributed through
the medium (e.g., gravity or rotational forces), H = h+ u2/2 is the total enthalpy, h is the
specific enthalpy, QH is the heat source per unit volume, τij is the viscous stress tensor, τijR
Processes 2024, 12, 2943 5 of 21
is the Reynolds stress tensor, and qi represents the diffusive heat flux. The indices denote
summation over the three coordinate directions. Heat transfer in solids and fluids, along
withthe
with theenergy
energyexchange
exchangebetween
betweenthemthem(conjugate
(conjugateheatheattransfer),
transfer),isisaafundamental
fundamentaland
and
implicit component of Solidworks Flow Simulation 2018 software. The
implicit component of Solidworks Flow Simulation 2018 software. The phenomenon of phenomenon of
heatconduction
heat conductionininsolid
solidmedia
mediaisisdescribed
describedby
bythethefollowing
followingequation
equation[24]:
[24]:

= 𝜆 +𝑄 , (5)
 
∂ρe ∂ ∂T
= λi + QH , (5)
∂t ∂xi ∂xi
where e represents specific internal energy, λ is the thermal conductivity tensor, and QH
is the especific
where represents
ratespecific
of heatinternal
release energy, λ is the thermal
(or absorption) per unitconductivity and QHde-
tensor,detailed
volume. A more is
the specific rate of heat release (or absorption) per unit volume.
scription can be found in the technical documentation provided [25].A more detailed description
can beThefound in theof
purpose technical documentation
the numerical analysis provided [25].
was to compare the three geometries in the
The purpose of the numerical analysis was to compare the three geometries in the
context of the heat exchange process. This was simulated using a plate with a thickness of
context of the heat exchange process. This was simulated using a plate with a thickness of
0.5 mm, which complemented the TPMS structures. A constant temperature of 100 °C was
0.5 mm, which complemented the TPMS structures. A constant temperature of 100 ◦ C was
appliedtotothe
applied thesurface
surfaceofofthe
theplate
plate(see
(seeFigure
Figure2).
2).

[Link]
Figure Thevisualisation
visualisationofof
thethe
simulation boundary
simulation conditions,
boundary placement,
conditions, and and
placement, dimensions of theof
dimensions
computational domain.
the computational domain.

Channels 80 mm in length were created before and after the TPMS structure to stabilise
Channels 80 mm in length were created before and after the TPMS structure to stabi-
and fully develop the flow. Air was used as the heat transfer medium, and five scenarios
lise and fully develop the flow. Air was used as the heat transfer medium, and five sce-
were considered, distinguished by different volumetric flow rates of 0.4, 0.8, 1.2, 1.6, and
narios
2.0 wereThe
m3 /h. considered,
compareddistinguished by different
geometries were modelledvolumetric flow rates
or represented of 0.4,
in the 0.8, 1.2,
numerical
1.6, and 2.0 m 3/h. The compared geometries were modelled or represented in the numer-
simulations using an aluminium alloy, specifically aluminium 6061, which has a density of
ical kg/m
2700 3 , a thermal
simulations usingconductivity
an aluminium alloy,
of 152 W/(mspecifically
·K), and aaluminium 6061,
heat capacity which
of 897 J/(kghas
·K).a
The outer walls of the channels were defined as adiabatic with a no-slip condition.
In this study, the adiabatic wall condition was intentionally chosen to isolate the effect of
heat transfer within the designed TPMS structures and to prevent interactions with lateral
boundaries. The primary objective was to evaluate the internal heat transfer and flow char-
acteristics of the TPMS structures without external influence. Adiabatic walls ensure that
no heat exchange occurs at the lateral boundaries, allowing us to focus exclusively on the
thermal-hydraulic performance of the TPMS structures. The use of symmetric or periodic
boundary conditions could introduce unrealistic assumptions about flow development,
particularly for geometries such as the combined gyroid, where flow and thermal fields are
significantly heterogeneous and exhibit a high degree of mixing.
Based on these boundary conditions, a three-dimensional model was created and
adapted into a computational mesh prior to the numerical simulation. Solidworks Flow
Simulations 2018 software employs a unique approach to mesh generation based on an
immersed body mesh method. This technique generates a mesh independently of the
model geometry, allowing mesh cells to intersect the boundary between solid and fluid
regions. This facilitates the implementation of a Cartesian grid consisting of cuboidal
(or rectangular) cells aligned with Cartesian coordinates and adjacent to the external
 regions. This facilitates the implementation of a Cartesian grid consisting of cuboidal (or
rectangular) cells aligned with Cartesian coordinates and adjacent to the external bound-
aries of the computational domain. The mesh cells are classified as solid, fluid, or partial
Processes 2024, 12, 2943 (those intersecting the boundary between solid and fluid regions). For partial cells, 6 of 21
a two-
stage wall function is applied, combining two methods depending on the thickness of the
boundary layer. This approach enables the resolution of the Navier–Stokes equations
boundaries
within volumesofdefined
the computational
by the [Link]. The mesh cells are classified as solid, fluid, or
partial (those intersecting the boundary between solid and fluid regions). For partial cells,
aFrom the perspective
two-stage wall function of validating
is applied, the numerical
combining two methods analysis results,
depending a key
on the step was the
thickness
grid of
independence
the boundary [Link],
Thiswhich examined
approach enables thetheresolution
relationship
of the between the number
Navier–Stokes equations of gen-
within volumes defined by the mesh.
erated mesh elements and the monitored control parameter’s value. In this study, the tem-
From the perspective of validating the numerical analysis results, a key step was
perature of the heat transfer medium at the outlet of the computational domain was ob-
the grid independence study, which examined the relationship between the number of
served over the
generated surface
mesh defined
elements as monitored
and the the pressure outlet,
control and the value.
parameter’s area-weighted average
In this study, the was
evaluated for a ofvolumetric
temperature flow medium
the heat transfer rate of at 1.2themoutlet
3 /h. The results
of the of the grid
computational independence
domain was
observed over the surface defined as the pressure outlet, and the area-weighted
study are presented in Figure 3. Six variants of3 input parameters influencing the final cell average
was evaluated for a volumetric flow rate of 1.2 m /h. The results of the grid independence
size and
studycellaredistribution
presented in were
Figureutilised in the mesh
3. Six variants of input generation
parameters process. Thethe
influencing study
finaldemon-
strated
cell that convergence
size and of thewere
cell distribution values was in
utilised achieved
the meshfrom the fourth
generation variant
process. of the mesh
The study
demonstrated
generation, whichthat was convergence of the valuescomprehensive
used for subsequent was achieved from the fourthanalyses,
numerical variant ofthereby
the mesh generation, which was used for subsequent comprehensive numerical analyses,
reducing the computational time required for simulations.
thereby reducing the computational time required for simulations.

Figure 3. Grid independence study.

Figure 3. Grid independence study.
The sheet gyroid, skeletal gyroid, and combined gyroid TPMS structures proposed
by the authors were compared based on numerical parameters. The comparison utilised
both two-dimensional visualisations of parameters, such as flow velocity and temper-
ature distribution in different parts of the model, as well as three-dimensional visuali-
sations of flow trajectories. The hydraulic parameters were evaluated in terms of the
Reynolds number, friction factor, and pressure drop. The Reynolds number is defined as
follows [25,26].
ρuDh
Re = , (6)
µε P
where ρ is density, Dh is hydraulic diameter, u is superficial fluid velocity, µ is dynamic
viscosity, and εP is porosity. Porosity is described as the proportion of the void volume
to the total bulk volume of the structure. The hydraulic diameter (Dh ) can be determined
using various formulas linking the channel’s cross-sectional area and its wetted perimeter.
However, for porous structures, such as the gyroid structure discussed here, it is more
Processes 2024, 12, 2943 7 of 21

effective to calculate Dh using the void volume (Vvoid ) and the specific surface area (As ).
The Dh parameter can then be defined as follows [27].

4Vvoid
Dh = (7)
As

The last monitored parameter is the Fanning friction factor (f ), which characterises the
dimensionless pressure loss or flow resistance in the channels of the gyroid structure and
which is defined according to Equation (4) [9].

2∆pε2P Dh
f = , (8)
Lρu2

where ∆p is the pressure loss; specifically, it refers to the pressure difference of the flowing
medium between the inlet and outlet of the model. Here, εP denotes the porosity, Dh
represents the hydraulic diameter, L is the length of the TPMS structure, ρ indicates the
density of the medium, and u is the superficial fluid velocity.
The thermal-technical properties of the evaluated TPMS structures were assessed by
monitoring the temperature of the flowing medium at the inlet and outlet surfaces. These
values were subsequently used to calculate the heat transfer coefficient, the Nusselt number,
and the Chilton–Colburn j-factor. The convective heat transfer coefficient, which represents
the rate of heat transfer via convection relative to conduction within the fluid, is defined as
follows [3]:
ρU A′ C ( Tout − Tin )
h= , (9)
A′′ ∆TLMTD
where ρ is the density, U is the velocity of the medium flow through the inlet surface,
A′ is the channel cross-sectional area, C represents the specific heat capacity, Tout is the
temperature of the medium at the outlet of the model, Tin is the temperature of the medium
at the inlet of the model, A′′ is the surface area with a defined temperature of 100 ◦ C, and
∆TLMTD is the logarithmic mean temperature difference, calculated as follows:

Tout − Tin
∆TLMTD =  , (10)
ln TTh−−TTout
in
h

The Nusselt number, which represents the ratio of convective to conductive heat
transfer and thus indicates the overall efficiency of convective heat transfer relative to
conduction within the flowing medium, is determined using the following equation [28]:

hDh
Nu = , (11)
λ
where h is the convective heat transfer coefficient, Dh is the hydraulic diameter, and λ is the
thermal conductivity of the medium. Finally, the efficiency of the heat exchange process
was evaluated using the Chilton–Colburn j-factor, as defined by Equation (12):

Nu
j= 1
, (12)
RePr 3
where Nu is the Nusselt number, Re is the Reynolds number, and Pr is the Prandtl number
determined according to [29].
The selected results of the presented numerical analysis were validated against the re-
sults calculated using the ε-NTU method, where the heat transfer coefficient was compared
based on the following equations:

ε = 1 − e− NTU , (13)
 The selected results of the presented numerical analysis were validated against the
results calculated using the ɛ-NTU method, where the heat transfer coefficient was com-
pared based on the following equations:
𝜀 =1−𝑒 , (13)
Processes 2024, 12, 2943 8 of 21

𝑁𝑇𝑈 = , (14)

where ε is the effectiveness of the heatNTU hAprocess,

exchange s NTU is the number of transfer
= (14)
Cmin
units, and Cmin is the minimum heat capacity rate, equal to the product of the mass flow
where
rate andε is
thethe effectiveness
specific of the heat exchange process, NTU is the number of transfer
heat capacity.
units, and Cmin is the minimum heat capacity rate, equal to the product of the mass flow
rate
3. and the specific heat capacity.
Results
The core concept of this study is the development of a geometry through the combi-
3. Results
nation of core
The existing TPMS
concept of structures to the
this study is mitigate the negative
development effects of through
of a geometry porous TPMS geom-
the combina-
etries
tion ofon the hydraulic
existing characteristics
TPMS structures of thethe
to mitigate flowing medium
negative effects while maintaining
of porous the per-
TPMS geometries
on the hydraulic
formance metricscharacteristics of the flowing
of the heat exchange [Link] whilesheet
In the basic maintaining
gyroid the performance
geometry, a ho-
metrics of the
mogeneous heat exchange
distribution process.
of pores In the basic
and resulting sheet is
channels gyroid geometry,
evident. a homogeneous
In contrast, the skeletal
distribution
gyroid of pores and
is characterised by resulting
a massivechannels is evident.
flow channel (relativeIn to
contrast, the skeletal
the overall geometry) gyroid
and is
a
characterised by a massive flow channel (relative to the overall geometry) and a substan-
substantial volume of solid material forming the three-dimensional structure. This dis-
tial volume of solid material forming the three-dimensional structure. This distinction is
tinction is visible in Figure 4, which illustrates the velocity maps of the heat transfer me-
visible in Figure 4, which illustrates the velocity maps of the heat transfer medium as it
dium
flows as it flowsthe
through through
porousthe porous structures
structures of the compared
of the compared geometries. geometries.
The sectionsThearesections
taken
are taken transversely at distances of 0, 11.5, 23, 34.5, 46, 57.5, 69, and 80
transversely at distances of 0, 11.5, 23, 34.5, 46, 57.5, 69, and 80 mm along the flow axis, mm along the
flow axis, starting from the edge of the model for a volumetric
starting from the edge of the model for a volumetric flow rate of 1.2 m /h. flow 3
rate of 1.2 m 3/h.

Figure Three-dimensionalvisualisation
Figure 4. Three-dimensional visualisationofofvelocity
velocity maps
maps in flow
in flow cross-sections
cross-sections at a at a volumetric
volumetric flow
flow rate
rate of 3 /h.
1.2ofm1.2 m3/h.

The sheet
The sheet gyroid
gyroid geometry
geometry displays
displays aa homogeneous velocity distribution
homogeneous velocity distribution without
without
significant variations in the different section views or throughout the entire
significant variations in the different section views or throughout the entire geometry geometry
length. This uniformity results from the steady flow in the porous structure, which does
length. This uniformity results from the steady flow in the porous structure, which does
not induce notable changes in turbulent kinetic energy values. In stark contrast, the skeletal
not induce
gyroid notableexhibits
geometry changesa in turbulent
markedly kinetic energyvelocity
heterogeneous values. distribution
In stark contrast,
in thethe skel-
section
etal
[Link]
The geometry exhibits
heat transfer a markedly
medium heterogeneous
flows through velocity distribution
a single channel, with velocityinincreases
the sec-
tion views. The heat transfer medium flows through a single channel, with velocity
corresponding to changes in the channel’s cross-sectional area. The highest velocities occur in-
creases corresponding
in regions to changes
where the channel in theconsistent
narrows, channel’swith
cross-sectional [Link].
the continuity The highest veloc-
The proposed
ities occur in regionscombined
where thegeometry
channeldemonstrates a mix ofwith
narrows, consistent the the
characteristics
continuity described.
equation.
Certain regions show higher velocity values, while others display more
The proposed combined geometry demonstrates a mix of the characteristics uniform flow and
de-
lower velocities, leading to enhanced mixing of the flowing medium. This behaviour
scribed. Certain regions show higher velocity values, while others display more uniform arises
from a combination of regions with larger channels and areas with lower porosity. The
design effectively utilises these contrasting features to achieve improved mixing and flow
uniformity, while leveraging the strengths of both larger and smaller channel regions.
This phenomenon is particularly evident when examining the cross-sectional view
of the channels at the outlet of the TPMS structure. At this point, where the gyroid
structure is no longer present in the channel cross-section, it influences the flow trajectories
only indirectly. In Figure 5, regions with maximum flow velocity and the distribution of
streamlines are clearly visible, indicating the formation of turbulent zones associated with
local maxima. These zones are concentrated in areas adjacent to regions with lower porosity
or larger channel cross-sections.
 the channels at the outlet of the TPMS structure. At this point, where the gyroid structure
is no longer present in the channel cross-section, it influences the flow trajectories only
indirectly. In Figure 5, regions with maximum flow velocity and the distribution of
streamlines are clearly visible, indicating the formation of turbulent zones associated with
Processes 2024, 12, 2943 local maxima. These zones are concentrated in areas adjacent to regions with lower9 of po-
21
rosity or larger channel cross-sections.

Figure 5. Velocity
Velocity map and streamline visualisation in the transverse cross-section of the channel at
the outlet of the heat transfer medium
medium from
from the
the TPMS
TPMSstructure
structureat
ataavolumetric
volumetricflow
flowrate 1.2mm33/h.
rateofof1.2 /h.

A comprehensive
A comprehensive overview
overview of of the
the flow
flow characteristics
characteristics isis provided
provided in
in Figure
Figure 6,6, which
which
depicts a longitudinal section through the model at half its width. This visualisation
depicts a longitudinal section through the model at half its width. This visualisation also
also highlights flow velocity and streamlines. From this perspective, the change in flow
highlights
characteristicsflowforvelocity and streamlines.
the combined geometry From this perspective,
is even more [Link] change
Uniform in streamline
flow char-
acteristics for the combined geometry is even more apparent. Uniform streamline
distribution and the absence of vortex regions, compared to the skeletal gyroid, are evident. distri-
bution and the absence of vortex regions, compared to the skeletal
When comparing the combined geometry to the basic sheet gyroid, there is no noticeablegyroid, are evident.
When
decrease comparing the combined
in flow velocity geometry
or distortion of flowto the basic sheet
trajectories gyroid,
in the there
vertical is no noticeable
direction. This can
Processes 2025, 13, x FOR PEER REVIEW
be attributed to the openness of the lower part of the combined geometry to the
decrease in flow velocity or distortion of flow trajectories in the vertical direction. This heat 10 ofcan
transfer22
medium flow and the lack of a uniform TPMS structure in the form of
be attributed to the openness of the lower part of the combined geometry to the heat trans- interconnected
channels,
fer medium as flow
seen and
in the
thebasic
lacksheet gyroid geometry.
of a uniform TPMS structure in the form of interconnected
channels, as seen in the basic sheet gyroid geometry.

Figure 6.
Figure Visualisation of
6. Visualisation of flow
flow velocity
velocity and
and streamlines
streamlines in
in the
the longitudinal
longitudinal section
section at
at half
half the
the width
width
3
of the geometry at a volumetric flow rate of 1.2 m /h.
of the geometry at a volumetric flow rate of 1.2 m3/h.
When examining the flow trajectories and temperature distribution throughout the
entireWhen
volumeexamining the flow trajectories
of the evaluated geometry (seeandFigures
temperature
7 and distribution
8)—excluding throughout the
the inlet and
entire volume of the disadvantage
outlet channels—the evaluated geometry (see flow
of uniform Figures 7 and
in the 8)—excluding
sheet the inlet
gyroid geometry and
becomes
outlet channels—the
distinctly apparent. Indisadvantage
this case, theof uniform flow
segregation in the sheet
of channels, gyroid
or heat geometry
transfer mediumbecomes
flow,
into two groups
distinctly is evident.
apparent. In one
In this case, thegroup, a moreofpronounced
segregation channels, orthermal flux occurs
heat transfer medium from the
flow,
TPMS
into body
two to the
groups is heat transfer
evident. medium.
In one group, This
a moresegregation arises
pronounced from the
thermal fluxfundamental
occurs from
design of the gyroid geometry, which, based on the basic 2π × 2π × 2π unit cell,
the TPMS body to the heat transfer medium. This segregation arises from the fundamental divides the
design of the gyroid geometry, which, based on the basic 2π × 2π × 2π unit cell, divides
the resulting space into two curved surfaces forming walls of a disconnected channel net-
work. These channels are not in identical contact with the surface, transferring thermal
flux to the TPMS structure. Consequently, this phenomenon leads to a gradual and slow
 the TPMS body to the heat transfer medium. This segregation arises from the fundamental
design of the gyroid geometry, which, based on the basic 2π × 2π × 2π unit cell, divides
the resulting space into two curved surfaces forming walls of a disconnected channel net-
work. These channels are not in identical contact with the surface, transferring thermal
Processes 2024, 12, 2943
flux to the TPMS structure. Consequently, this phenomenon leads to a gradual and10slow of 21

equalisation of temperatures within the channels, dependent on the flow trajectory length
or the length of the heat exchange element.
resulting space into two curved surfaces forming walls of a disconnected channel network.
TheseInchannels
contrast,are
thenot
combined geometry
in identical contact demonstrates a relatively
with the surface, uniform
transferring temperature
thermal flux to the
distribution across the entire cross-section and within individual channels. The
TPMS structure. Consequently, this phenomenon leads to a gradual and slow equalisation modifica-
tions made ensure
of temperatures interconnectivity
within the channels, between
dependent theonchannels,
the flow resulting
trajectory in enhanced
length or themixing
length
and homogenisation
of the of the temperature throughout the flow volume of the geometry.
heat exchange element.

Processes 2025, 13, x FOR PEER REVIEW 11 of 22

Figure 7. Visualisation
Visualisation ofof flow
flowtrajectories
trajectoriesand
andtemperature
temperaturevaluesvaluesofofthe
theheat
heattransfer
transfermedium
mediuminina
a cross-sectional view 3
cross-sectional view at at a volumetric
a volumetric flow
flow rate
rate of of
1.21.2
mm 3 /h,/h, excluding
excluding the the depiction
depiction of the
of the TPMSTPMS
ge-
geometry.
ometry.

Figure 8. Visualisation of turbulent dissipation at a volumetric flow rate of 1.2 m3 /h, excluding the
Figure 8. Visualisation of turbulent dissipation at a volumetric flow rate of 1.2 m3/h, excluding the
depiction of the TPMS geometry.
depiction of the TPMS geometry.
In contrast, the combined geometry demonstrates a relatively uniform temperature
The results
distribution of the
across thenumerical analysis were
entire cross-section processed
and within not onlychannels.
individual through theThepresented
modifica-
visualisations but also using Equations (6) to (12), enabling the quantification of mixing
tions made ensure interconnectivity between the channels, resulting in enhanced the re-
and homogenisation
sulting parameters into of the
twotemperature throughout
categories: hydraulic the flow volume
characteristics, which of address
the geometry.
the qual-
The
itative results ofofthe
properties thenumerical analysis
fluid flow, were processed
and thermal not only
characteristics, through
which the presented
quantify the heat
exchange process between the heat transfer medium and the TPMS [Link] resulting
visualisations but also using Equations (6) to (12), enabling the quantification of
parameters into two categories: hydraulic characteristics, which address the qualitative
The heat transfer coefficient is a key parameter in the design and optimisation of heat
properties of the fluid flow, and thermal characteristics, which quantify the heat exchange
exchangers, as it directly
process between the heat determines the efficiency
transfer medium and the ofTPMSheatstructure.
transfer between two environ-
ments. Accurate
The determination
heat transfer andisoptimisation
coefficient a key parameterof thisincoefficient
the design significantly influence
and optimisation of
the overall performance and efficiency of the device. Precise knowledge
heat exchangers, as it directly determines the efficiency of heat transfer between two of the heat trans-
fer coefficient allows
environments. for calculations,
Accurate determinationsuchandas determining
optimisationthe required
of this surface
coefficient area of the
significantly
exchanger to achieve the desired heat transfer rate. Figure 9 (left) depicts this parameter
as a function of volumetric flow rate, ranging from 0.4 to 2.0 m3/h.
The results show that the combined gyroid geometry achieves the highest heat trans-
fer coefficient values, reaching 362.17 W/(m2·K) at a flow rate of 2.0 m3/h, demonstrating
 Processes 2024, 12, 2943
Processes 2025, 13, x FOR PEER REVIEW 12 of1122of 21

influence the overall performance and efficiency of the device. Precise knowledge of the
for heat
optimising
transferheat exchange
coefficient in thefor
allows system. In the context
calculations, such asofdetermining
heat exchanger applications,
the required surface
thearea
combined gyroid geometry is the most suitable for scenarios requiring maximum
of the exchanger to achieve the desired heat transfer rate. Figure 9 (left) depicts heatthis
transfer, such as acompact
parameter functionheat exchangersflow
of volumetric withrate,
highranging
thermalfrom 0.4 to 2.0 m3 /h.
loads.

Figure
Figure 9. Graphical
9. Graphical representation
representation ofofthe
theheat
heattransfer
transfercoefficient
coefficient for the
the applied
appliedvolumetric
volumetricflow rates
flow
rates of the heat transfer medium (left) and the graphical depiction of the Nusselt number as a func- of
of the heat transfer medium (left) and the graphical depiction of the Nusselt number as a function
tionthe achieved
of the Reynolds
achieved number
Reynolds (right).
number (right).

The results show that the combined gyroid geometry achieves the highest heat transfer
The results presented in Figure 9 (right) fully support the previously stated conclu-
coefficient values, reaching 362.17 W/(m2 ·K) at a flow rate of 2.0 m3 /h, demonstrating
sions regarding
superior the performance
heat transfer of the all
efficiency across examined
examined geometries
flow [Link] heat exchangers.
skeletal The
gyroid exhibits
Nusselt numberheat
intermediate serves as an coefficient
transfer indicator of heat transfer
values, peaking efficiency,
at 295.74 W/(mwith higher
2 ·K) forvalues
the same cor-
flow
responding
rate. While to it
better
performsthermal performance.
significantly betterThe
thancombined
the sheetgyroid
gyroid,achieves theshort
it still falls highestof the
combined
Nusselt number gyroid.
across Conversely,
all Reynolds thenumber
sheet gyroid
ranges,hasdemonstrating
the lowest heat itstransfer
superiorcoefficient
thermal at
2 2.0 m3 /h,
230.55 W/(m
performance. This·K) for a flow
finding aligns rate
withof earlier indicating
conclusions lower
about efficiency
the highest heatin heat transfer
transfer
compared to the other
coefficient values for this geometry. two geometries.
TheAt the opposite
skeletal end of the
gyroid exhibits flow rate Nusselt
intermediate spectrum, defined
number by low
values, volumetric
confirming it as flow
a
rates (minimum 0.4 m3 /h in this comparison), the heat transfer coefficient values are
compromise between performance and other factors, such as pressure drop and manufac-
63.38 W/(m2 ·K) for the sheet gyroid, 149.69 W/(m2 ·K) for the skeletal gyroid, and
turing complexity. 2 In contrast,
188.57 W/(m ·K) for the combined the sheet gyroid
[Link]
This the lowest Nusselt
comparison reveals number
that atvalues,
no flow
consistent with its lower heat transfer coefficient and reduced heat
rate does the heat transfer coefficient of the combined gyroid decrease significantly, transfer efficiency. Ad-
ditionally, the graph indicates that the Nusselt number increases
demonstrating its ability to maintain heat exchange efficiency consistently. with rising Reynolds
numbersThe forobserved
all geometries,
trend of a generally
increasingexpected trendcoefficient
heat transfer where higher with flow
risingrates improve
volumetric flow
heatrate is evident
transfer. for allpronounced
The most [Link]
This indicates that for
is observed higher flow rates gyroid,
the combined lead to further
improved
thermal performance,
supporting highlighting
its use in applications the importance
requiring of selecting an appropriate flow rate
maximum performance.
for optimising heat exchange in the system.
Due to differences in the hydraulic diameter (Dh) among the In the context of heat exchanger
geometries, applications,
a direct com-
the combined gyroid geometry is the most suitable for scenarios requiring maximum heat
parison at identical Reynolds number ranges is not feasible. At the same volumetric flow
transfer, such as compact heat exchangers with high thermal loads.
rate, each geometry achieves a different Reynolds number. But, for instance, at a Reynolds
The results presented in Figure 9 (right) fully support the previously stated conclusions
number of approximately
regarding the performance 250, the sheet
of the gyroid achieves
examined geometries around
in heat25 exchangers.
NTU, the skeletal gy-
The Nusselt
roidnumber
44 NTU, and the combined gyroid about 88 NTU.
serves as an indicator of heat transfer efficiency, with higher values corresponding
toIn terms
better of hydraulic
thermal parameters,
performance. pressure drop
The combined gyroidis aachieves
significant
the metric,
highest reflecting the
Nusselt number
structure’s
across allpermeability
Reynolds number to theranges,
flowing medium anditsthe
demonstrating efficiency
superior thermalof fluid transportThis
performance.
finding
through thealigns
porous with earlier conclusions
material. Figure 10 shows aboutthethepressure
highest drop
heat transfer coefficient
as a function values for
of volumet-
this geometry.
ric flow rate, highlighting its strong dependence on geometry and the expected increase
with higherThe flow
skeletal
[Link] exhibits intermediate
The combined gyroid exhibitsNusselt number
the lowest values,drops
pressure confirming
across it allas a
compromise between performance and other factors, such as pressure drop and manufac-
volumetric flow rates, underscoring its advantage in hydraulic performance. This makes
turing complexity. In contrast, the sheet gyroid shows the lowest Nusselt number values,
it particularly suitable for applications requiring a balance between high thermal perfor-
consistent with its lower heat transfer coefficient and reduced heat transfer efficiency. Ad-
mance (as confirmed
ditionally, the graph by indicates
the highest thatheat
the transfer
Nusselt coefficient and Nusselt
number increases with number) and
rising Reynolds
minimal
numbers pressure
for allloss.
geometries, a generally expected trend where higher flow rates improve
heat transfer. The most pronounced increase is observed for the combined gyroid, further
supporting its use in applications requiring maximum performance.
Processes 2024, 12, 2943 12 of 21

Due to differences in the hydraulic diameter (Dh ) among the geometries, a direct
comparison at identical Reynolds number ranges is not feasible. At the same volumetric
flow rate, each geometry achieves a different Reynolds number. But, for instance, at a
Reynolds number of approximately 250, the sheet gyroid achieves around 25 NTU, the
skeletal gyroid 44 NTU, and the combined gyroid about 88 NTU.
In terms of hydraulic parameters, pressure drop is a significant metric, reflecting
the structure’s permeability to the flowing medium and the efficiency of fluid transport
through the porous material. Figure 10 shows the pressure drop as a function of volumetric
flow rate, highlighting its strong dependence on geometry and the expected increase with
higher flow rates. The combined gyroid exhibits the lowest pressure drops across all
volumetric flow rates, underscoring its advantage in hydraulic performance. This makes it
particularly suitable for applications requiring a balance between high thermal performance
Processes 2025, 13, x FOR PEER REVIEW 13 of 22
(as confirmed by the highest heat transfer coefficient and Nusselt number) and minimal
pressure loss.

100 Sheet gyroid

80 Combined gyroid
Skeletal gyroid
Δp [Pa]

60

40

20

0
0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
q [m3/h]

Figure [Link]
Figure10. Pressuredrop
dropper
perunit
unitlength
lengthanalysis.
analysis.

Conversely, the skeletal gyroid shows the highest pressure drops at the given flow rates,
Conversely, the skeletal gyroid shows the highest pressure drops at the given flow
suggesting that its use may be less favourable for applications sensitive to energy losses
rates, suggesting that its use may be less favourable for applications sensitive to energy
caused by flow resistance. Despite its intermediate thermal performance, the high-pressure
losses
drop caused
could byits
limit flow resistance.
practicality, Despite its
especially intermediate
at higher thermal
flow rates. The performance, the high-
sheet gyroid achieves
pressure drop could limit its practicality, especially at higher flow rates.
pressure drops between those of the combined and skeletal gyroids. This result, combined The sheet gyroid
achieves
with pressure
its lower drops between
heat transfer those
efficiency, of the combined
as indicated and heat
by its lower skeletal gyroids.
transfer This result,
coefficient and
combined
Nusselt with its
number, lower that
suggests heatitstransfer efficiency,
application mightasbeindicated
restrictedby to its lower heat
situations where transfer
heat
transfer
coefficient demands are lessnumber,
and Nusselt critical, but pressure
suggests thatlosses must remain
its application mightwithin acceptabletolimits.
be restricted situa-
At a maximum volumetric flow rate of 2.0 m 3 /h, the pressure drop for the combined
tions where heat transfer demands are less critical, but pressure losses must remain within
gyroid is one-third
acceptable limits. that of the sheet gyroid and nearly one-fourth that of the skeletal gyroid.
A similar trend is observed at the minimum flow rate of 0.4 m 3 /h, where the pressure drop
At a maximum volumetric flow rate of 2.0 m3/h, the pressure drop for the combined
for the combined gyroid is 1.86 Pa, compared to 6.4 Pa for the sheet gyroid and 7.23 Pa
gyroid is one-third that of the sheet gyroid and nearly one-fourth that of the skeletal gy-
for the skeletal gyroid. Overall, the comparison confirms that the combined gyroid offers
roid.
the A similar
best trend is
optimisation ofobserved at theand
heat transfer minimum
pressureflow rateThis
drop. of 0.4 m3/h, where
geometry the for
is ideal pressure
heat
drop for the
exchangers combined
requiring highgyroid is 1.86
efficiency Pa,minimal
with compared to 6.4
energy Pa for
costs the sheet
associated gyroid
with and 7.23
overcoming
Pa for
flow the skeletal gyroid. Overall, the comparison confirms that the combined gyroid of-
resistance.
fers The
the graphs in Figure 11ofillustrate
best optimisation a comparison
heat transfer of thedrop.
and pressure heat transfer coefficient
This geometry values
is ideal for
calculated based requiring
heat exchangers on the outputs of the presented
high efficiency CFDenergy
with minimal simulations with the values
costs associated of
with over-
this coefficient
coming determined using the ε-NTU method, which evaluates the heat transfer
flow resistance.
performance
The graphsthe
of in heat
Figureexchanger element
11 illustrate by employing
a comparison of the the
heatmathematical framework
transfer coefficient values
represented by Equations (13) and (14).
calculated based on the outputs of the presented CFD simulations with the values of this
coefficient determined using the ε-NTU method, which evaluates the heat transfer perfor-
mance of the heat exchanger element by employing the mathematical framework repre-
sented by Equations (13) and (14).
 coming flow resistance.
The graphs in Figure 11 illustrate a comparison of the heat transfer coefficient values
calculated based on the outputs of the presented CFD simulations with the values of this
coefficient determined using the ε-NTU method, which evaluates the heat transfer perfor-
Processes 2024, 12, 2943 mance of the heat exchanger element by employing the mathematical framework repre- 13 of 21
sented by Equations (13) and (14).

11. The
Figure11.
Figure The results
results of
of the
the comparison of the
comparison of the heat
heat transfer
transfercoefficient
coefficientfrom
fromCFD
CFDand theε-NTU
andthe ε-NTU
method
method(left:
(left:sheet
sheet gyroid;
gyroid; centre: combined gyroid;
centre: combined gyroid;right: skeletalgyroid).
right:skeletal gyroid).

The
Thedifferences
differences inin the
the heat
heat transfer coefficient,obtained
transfer coefficient, obtainedfrom
fromCFD
CFDanalysis
analysisand
andthe
the
ε-NTU method, vary depending on the type of TPMS geometry. For the sheet gyroid,
ε-NTU method, vary depending on the type of TPMS geometry. For the sheet gyroid, the the
largest differences were observed, with h-values derived from the ε-NTU method being
largest differences were observed, with h-values derived from the ε-NTU method being
consistently higher than the CFD results. At a flow rate of q = 2.0 m33 /h, the difference
consistently higher than2 the CFD results. At a flow rate of q = 2.0 m /h, the difference
reaches up to 50 W/(m ·K), which can be attributed to the simplified assumptions of
the ε-NTU method, which do not account for detailed flow behaviour and boundary
effects. In the case of the combined gyroid, these differences are more moderate, with the
ε-NTU method still providing higher values; however, the discrepancies do not exceed
40 W/(m2 ·K). This suggests that the more complex structure of the combined gyroid
facilitates efficient heat transfer, which is better represented in CFD simulations. For the
skeletal gyroid, the differences between the two methods are the smallest; the maximum
difference at higher flow rates is approximately 30 W/(m2 ·K). The accuracy of the results
obtained from CFD analysis, compared to the ε-NTU method, can be assessed based on the
differences between these two approaches for each TPMS geometry. Given that the ε-NTU
method is a theoretical approach that operates under simplified assumptions of uniform
heat distribution and ideal flow, it can serve as a reference point for verifying CFD results.
For the sheet gyroid, CFD values were consistently lower than the epsilon-NTU results,
with discrepancies increasing at higher flow rates. This difference may indicate that CFD
analysis more realistically accounts for the presence of local resistances, turbulent flow, and
thermal losses, which are neglected in the ε-NTU method. In the case of the combined
gyroid, the discrepancy between the two methods is smaller, indicating good accuracy of
the CFD results, with minor deviations likely related to the structural properties of the
geometry that support efficient heat transfer. Here, the differences reach approximately
10–15% at higher flow rates, confirming that CFD provides reliable results with respect to
physical effects. For the skeletal gyroid, the CFD results are closest to the values obtained
from the ε-NTU method, with discrepancies in the range of 5–10%. This agreement indicates
that the skeletal gyroid geometry has homogeneous thermal and flow properties, which
are well captured in CFD analysis. The smaller deviation confirms the high accuracy of
CFD simulations for this geometry.
Based on these findings, it can be concluded that CFD analysis provides more realistic
and accurate results, as it considers complex physical phenomena, such as turbulent flow,
pressure losses, and boundary effects, which the ε-NTU method, in its simplified approach,
does not capture. However, the differences between the methods highlight that the accuracy
of CFD results can be influenced by the specific geometry and its ability to ensure uniform
flow and heat transfer.
The comparison of the three geometries in terms of their potential applicability as
heat exchangers, shown in Figure 12, provides a comprehensive perspective on heat
transfer dynamics, temperature distribution uniformity, and the effectiveness of eliminating
thermally stagnant zones. Each geometry exhibits unique characteristics that define its
suitability for various applications.
Processes 2025, 13, x FOR PEER REVIEW 15 of 22
Processes 2024, 12, 2943 14 of 21

FigureFigure 12. Visualisation

12. Visualisation of temperature
of temperature maps
maps in in individual
individual cross-sections
cross-sections for thefor the compared
compared geome-geometries
at a volumetric flow rate of 1.2 m3 /h.
tries at a volumetric flow rate of 1.2 m /h.
3
Processes 2024, 12, 2943 15 of 21

The sheet gyroid geometry is distinguished by the uniform distribution of the flowing
medium across the cross-sectional area, particularly in the middle and final sections of the
system. By the midpoint of the heat exchanger, the medium’s temperature approaches
equilibrium, with temperature differences across cross-sections gradually diminishing.
This indicates high heat transfer efficiency and the absence of significant thermal gradients,
which minimise the risk of thermally stressed regions. However, relatively low thermal
gradients are observed at the system’s inlet, suggesting a slower onset of medium heating.
Consequently, this geometry is optimal for applications requiring uniform heating of the
medium without significant temperature fluctuations, especially when the time to reach
thermal equilibrium is not critical.
In contrast, the skeletal gyroid geometry demonstrates intense thermal gradients in
the initial sections, highlighting efficient heat transfer at the start of the system. The heat
exchange surfaces interact rapidly with the medium, promoting faster heating. However,
this effect is accompanied by substantial local temperature variations, particularly in the
middle sections of the heat exchanger, where stagnant zones with uneven heating may
form. While the medium achieves high temperatures by the system’s end, the distribution
is less uniform compared to the sheet gyroid. This geometry is better suited for applications
where rapid heat transfer over short distances is crucial, but is less ideal for scenarios
requiring uniform temperature distribution. The combined gyroid geometry exhibits the
most complex behaviour, significantly influenced by flow dynamics. It combines the fea-
tures of both the sheet and skeletal gyroids, achieving a balance between rapid heating and
temperature uniformity. Early sections show moderate thermal gradients, indicative of
effective initial heat transfer. In the middle and final sections, the interconnected channel
network facilitates mixing and homogenisation, reducing stagnant zones and enhancing
uniformity. This makes the combined geometry highly versatile and suitable for appli-
cations demanding a balance between efficient heat transfer, rapid heating, and uniform
temperature distribution.
In the initial stages of the system, intense thermal gradients form due to the interac-
tion of the medium with the uneven geometric profile. In the middle section of the heat
exchanger, significant temperature differences are observed, with the flow dynamics pro-
moting mixing and eliminating stagnant zones. At the system’s end, the medium reaches
high temperatures with a relatively uniform distribution. However, local temperature
extremes suggest that geometry may facilitate the formation of thermally stressed regions
in areas of increased turbulence.
A notable feature is the openness of the porous structure at the bottom, where the
flowing medium directly contacts the surface, delivering heat flux. This design ensures
the maximum extraction of thermal energy from this surface and the effective transfer of
the thermal gradient to other parts of the structure. This geometry is ideal for applications
prioritising enhanced mixing, dynamic heat transfer, and the minimisation of stagnant
zones, although it may require a longer path to achieve thermal equilibrium.
From a broader perspective, sheet gyroid geometry is most suitable for applications
requiring uniform heating and high temperature field homogeneity. The skeletal gyroid
geometry is preferred for scenarios demanding rapid heat transfer over short distances,
albeit with reduced temperature homogeneity. The combined gyroid geometry is best
suited for processes requiring intensive medium mixing, delivering high heat exchange
efficiency over longer distances.
Another parameter examined was the friction factor, a dimensionless measure char-
acterising fluid flow through porous materials, including triply periodic minimal surface
(TPMS) structures. This factor quantifies pressure losses caused by friction due to interac-
tions between the fluid and the surface of the structure during flow. It is expressed as the
ratio of wall shear stress to the dynamic pressure of the flow, with its value dependent on
the flow regime (laminar or turbulent) and the geometry of the structure.
For TPMS structures, which are characterised by periodic and highly porous designs,
the friction factor provides insights into hydraulic efficiency and flow distribution within
 The combined gyroid achieves intermediate friction factor values, lower than the
sheet gyroid but higher than the skeletal gyroid. This geometry balances the properties of
both structures, offering a compromise between hydraulic losses and effective heat trans-
fer. It is suitable for applications where a balance between pressure losses and heat ex-
Processes 2024, 12, 2943 16 of 21
changer efficiency is needed.
The skeletal gyroid has the lowest friction factor values, approaching 1 at high Reyn-
oldssystem.
the numbers. Thisstructures
These geometry possess
generates the least
complex pressure losses,
morphologies thatmaking it an
allow for ideal choice
efficient fluid
for applications prioritising reduced energy costs associated with fluid flow.
and heat transfer, but can also generate pressure losses. A high friction factor indicates However, its
lower surface
significant area may
pressure limitoften
losses, its capability
resultingfor effective
from heat
intense transfer
friction orcompared
turbulent to the other
flow. Con-
versely, a low
geometries. Invalue suggests
summary, the reduced
choice ofpressure
geometry losses, implying
depends a moreapplication
on specific efficient flow with
require-
minimal resistance.
ments. While the sheet gyroid offers high heat transfer potential at the cost of higher-pres-
sure Figure
losses, 13
the(left) illustrates
skeletal gyroidthe frictionlow
provides factor analysis
hydraulic as a but
losses function
reducesof the
heatReynolds
transfer
number. The graph shows a decline in friction factor with increasing Reynolds
efficiency. The combined gyroid presents a balanced option that is suitable for applica- numbers, a
typical trend where pressure losses per unit of dynamic pressure decrease as the flow tran-
tions needing efficient heat transfer with moderate pressure losses.
sitions from laminar to turbulent regimes. This shift results in changes in the interactions
between the fluid and the structure’s surface.

8 0.45
Sheet gyroid Sheet gyroid
Combined gyroid 0.4 Combined gyroid
7
Skeletal gyroid Skeletal gyroid
0.35
6
0.3

5 0.25
f

j
0.2
4
0.15
3
0.1

2 0.05
0 500 1000 1500 0 500 1000 1500
Re Re

Figure 13. Friction factor (left) and Chilton–Colburn j-factor analysis (right).

Figure [Link]
The Friction factorexhibits
gyroid (left) andthe
Chilton–Colburn j-factor
highest friction factoranalysis
values(right).
across the entire Reynolds
number range, exceeding a value of 7 at low Reynolds numbers. Its high surface area and
The structure
intricate final parameter used toimpact
significantly quantify the flow,
fluid suitability of the
leading TPMS
to the structures
largest evaluated
pressure losses.
in this losses
These comparison was
limit its the Chilton–Colburn
applicability in scenariosj-factor.
whereThis is a dimensionless
minimising quantity
energy costs uti-
for fluid
flow is a priority. However, its extensive surface area may be advantageous for
lised to characterise heat transfer in a flowing medium, integrating the properties of flow, processes
requiring
geometry,intensive
and heat heat transfer.
exchange [30]. For triply periodic minimal surface (TPMS) structures,
The combined gyroid achieves
which are frequently employed in modern intermediate
heat friction factorthe
exchangers, values, lower than the
Chilton–Colburn sheet
j-factor
gyroid but higher than the skeletal gyroid. This geometry balances the properties of both
provides valuable insights into the efficiency of heat transfer relative to pressure drop and
structures, offering a compromise between hydraulic losses and effective heat transfer. It
turbulent flow.
is suitable for applications where a balance between pressure losses and heat exchanger
A high
efficiency value of the Chilton–Colburn j-factor indicates that the TPMS structure
is needed.
achieves
The high heatgyroid
skeletal transferhas
efficiency with relatively
the lowest low energy
friction factor values,losses caused by1pressure
approaching at high
[Link].
Reynolds This typically representsgenerates
This geometry an optimal the combination
least pressureoflosses,
surface area and
making it anturbu-
ideal
choice for applications
lence generated by the prioritising reduced energy
structure, facilitating intensecosts
heatassociated
exchange with fluidmedia.
between flow. How-
Such
ever, its lower surface area may limit its capability for effective heat transfer compared to
the other geometries. In summary, the choice of geometry depends on specific application
requirements. While the sheet gyroid offers high heat transfer potential at the cost of
higher-pressure losses, the skeletal gyroid provides low hydraulic losses but reduces heat
transfer efficiency. The combined gyroid presents a balanced option that is suitable for
applications needing efficient heat transfer with moderate pressure losses.
The final parameter used to quantify the suitability of the TPMS structures evaluated
in this comparison was the Chilton–Colburn j-factor. This is a dimensionless quantity
utilised to characterise heat transfer in a flowing medium, integrating the properties of flow,
geometry, and heat exchange [30]. For triply periodic minimal surface (TPMS) structures,
which are frequently employed in modern heat exchangers, the Chilton–Colburn j-factor
provides valuable insights into the efficiency of heat transfer relative to pressure drop and
turbulent flow.
A high value of the Chilton–Colburn j-factor indicates that the TPMS structure achieves
high heat transfer efficiency with relatively low energy losses caused by pressure resistance.
Processes 2024, 12, 2943 17 of 21

This typically represents an optimal combination of surface area and turbulence generated
by the structure, facilitating intense heat exchange between media. Such structures are
preferred in applications requiring high thermal efficiency while maintaining a compact
design. Conversely, a low Chilton–Colburn j-factor signals less effective heat transfer. This
may result from a suboptimal TPMS geometry that fails to induce sufficient turbulence or
create an adequately large heat exchange surface. Consequently, this could lead to reduced
system performance, higher operational costs, and diminished overall structural efficiency.
The graph in Figure 13 (right) shows that Chilton–Colburn j-factor values decrease
with increasing Reynolds number across all geometries, which is expected behaviour
since, at higher values, turbulent flow stabilises and the relative increase in heat transfer
diminishes. However, this decline varies between geometries, highlighting their differing
capabilities to facilitate intense heat transfer.
The combined gyroid exhibits the highest Chilton–Colburn j-factor values across the
entire Reynolds number range, indicating that it is the most effective in terms of heat
transfer. This geometry likely benefits from an advantageous combination of surface area
and fluid flow dynamics, enabling intense heat exchange. Consequently, the combined
gyroid is suitable for applications where high heat transfer efficiency is essential, such as in
compact heat exchangers or high-performance cooling systems.
The skeletal gyroid achieves intermediate Chilton–Colburn j-factor values, which are
lower than those of the combined gyroid but higher than those of the sheet gyroid. This
geometry offers a balance between heat transfer efficiency and pressure drop, still providing
sufficient performance for engineering applications where both factors are important.
The sheet gyroid has the lowest Chilton–Colburn j-factor values across the entire
Reynolds number range, indicating its reduced ability to transfer heat. This may result
from less effective surface interaction with the fluid or unfavourable flow characteristics.
While this geometry holds the potential for processes requiring large surface areas, its lower
Chilton–Colburn j-factor may limit its applicability in heat transfer-critical applications.
In conclusion, the selection of TPMS geometry should be based on the specific require-
ments of the application. The combined gyroid offers the best performance for efficient
heat transfer, while the skeletal gyroid may be suitable for applications with compromise
requirements. Conversely, the sheet gyroid appears to be a less suitable option for processes
where heat transfer performance is crucial, although it may be attractive in applications
with other priorities.
The relationship between the friction factor and the Chilton–Colburn j-factor for the
evaluated TPMS geometries—sheet gyroid, combined gyroid, and skeletal gyroid—reveals
critical trade-offs between pressure losses and heat transfer. These factors are key perfor-
mance indicators for structures in applications such as heat exchangers, where assessing their
interdependence enables efficient design optimisation.
The graphs indicate that both factors decrease with increasing Reynolds numbers,
suggesting that at higher Reynolds numbers, turbulent flow stabilises, leading to reduced
relative pressure losses and heat transfer. However, this decrease is not uniform across the
different geometries, reflecting their varying performance characteristics.
The sheet gyroid exhibits the highest friction factor values, corresponding to the largest
pressure losses, while its Chilton–Colburn j-factor is the lowest among the evaluated ge-
ometries. This imbalance highlights the inefficiency of this geometry for heat transfer, as the
high-pressure losses are not offset by sufficient heat transfer intensity. Such characteristics
limit the sheet gyroid’s applicability in scenarios where low energy consumption is a key
requirement while ensuring adequate heat transfer. However, its extensive surface area
may make it advantageous for processes focusing on other objectives, such as chemical
reactions or filtration.
The combined gyroid achieves the highest Chilton–Colburn j-factor values, indicating
the most efficient heat transfer, while maintaining moderate friction factor values. This
balance between pressure losses and heat transfer makes the combined gyroid universally
suitable for applications requiring intense heat exchange at acceptable pressure losses.
Processes 2024, 12, 2943 18 of 21

Its combination of large surface area and efficient flow dynamics makes it ideal for high-
performance heat exchangers, where energy efficiency and high output are essential.
The skeletal gyroid has the lowest friction factor values, reflecting the smallest pressure
losses, while its Chilton–Colburn j-factor is moderate. This geometry offers a compromise
solution, minimising pressure losses while maintaining an acceptable level of heat transfer
efficiency. These characteristics make it ideal for applications where pressure losses are the
primary limiting factor, such as systems with low differential pressure or those prioritising
energy savings.
The relationship between these factors highlights the necessity of trade-offs when
optimising TPMS geometries. While a high Chilton–Colburn j-factor is desirable for efficient
heat transfer, it is accompanied by an increased friction factor, leading to higher pressure
losses. In applications where reducing pressure losses is a priority, geometries with a
low friction factor are preferred, even though this may result in limited heat transfer
performance. Conversely, geometries with a high Chilton–Colburn j-factor, such as the
combined gyroid, are suitable for applications requiring maximum thermal performance.
Therefore, the choice of geometry must consider the specific requirements of the application,
whether it involves minimising energy costs or maximising thermal efficiency.
Despite the numerical analysis results indicating the advantages of the proposed
combined gyroid geometry for systems with specific heat exchange parameters, the ultimate
determinant is the manufacturability of the structure using available technologies. In the
current state of technological development, the only viable method is the use of additive
manufacturing techniques. These include fused filament fabrication/fused deposition
modelling (FFF/FDM) with thermally conductive filaments [31], or more suitable methods
such as SLA (laser beam curing resin printing) [32], SLS (laser sintering of powdered
plastics) [33], or powder bed fusion, including selective laser melting (SLM), direct metal
laser sintering (DMLS), and electron beam melting (EBM) [34,35].
The sheet gyroid is characterised by prominent thin walls and regular openings, which
pose a challenge for additive manufacturing, especially when using methods such as
selective laser melting (SLM) or stereolithography (SLA). Thin walls can be fragile during
printing, requiring precise control of process parameters, such as laser power, melting
temperature, and support structures during production. Additionally, the combination
of thin walls and a high surface area increases the risk of deformation, particularly when
using metals. Despite these challenges, the production of this structure via 3D printing is
technically demanding but advantageous in applications requiring high thermal efficiency.
The skeletal gyroid features a more robust structure with thicker elements, reducing
the risk of deformation and improving its suitability for 3D printing. This geometry is
easier to process, as it does not require the same high level of printing precision as the
sheet gyroid. Furthermore, its design allows for better thermal treatment management and
minimises internal stresses, which are essential for metal printing.
The combined gyroid, which merges elements of the sheet and skeletal gyroids,
presents moderate demand for 3D printing. Its complex geometry requires advanced
printing technologies capable of accurately reproducing structural details. Producing this
structure may necessitate the use of support materials or additional post-processing steps,
such as removing supports and performing thermal treatment.
Overall, the feasibility of manufacturing these structures via 3D printing depends on
the chosen material and technology. Metal materials are suitable for all three geometries
but require strict control of printing parameters and often involve complex post-processing.
Polymeric materials may be easier to process; however, their lower thermal conductivity
and strength could limit their applicability in high-performance applications.

4. Conclusions
The final assessment of the presented study highlights the significant advantages of
employing triply periodic minimal surface (TPMS) structures, particularly the combined
gyroid geometry, for applications in heat exchangers. The study utilised advanced nu-
Processes 2024, 12, 2943 19 of 21

merical analysis based on flow and heat transfer simulations performed using the finite
volume method within Solidworks Flow Simulation software. Parameters such as the
Nusselt number, Chilton–Colburn j-factor, Fanning friction factor, and pressure losses were
evaluated concerning Reynolds numbers and volumetric flow rates to achieve a detailed
comparison of the three geometric configurations.
The combined gyroid demonstrated the highest heat transfer efficiency, achieving
superior values for the Nusselt number and Chilton–Colburn j-factor, while maintaining
acceptable levels of the Fanning friction factor and pressure losses. This balance between
pressure losses and thermal performance makes the combined gyroid highly suitable for
compact and energy-efficient heat exchangers.
In contrast, the sheet gyroid geometry, although offering uniform flow distribution
and a high surface area, exhibited high pressure losses and lower heat transfer efficiency.
These characteristics limit its applicability in scenarios requiring low energy consumption
for flow and high thermal performance. Conversely, the skeletal gyroid achieved low
pressure losses, but at the expense of thermal performance, making it a viable option for
applications with low-pressure systems or those prioritising minimal energy consumption
for fluid pumping.
A significant contribution of this study is the introduction of the combined gyroid
geometry, which synergistically combines the benefits of the sheet and skeletal gyroid
geometries while mitigating their drawbacks. This geometry demonstrates the ability to
ensure intense mixing of the flowing medium, leading to a more uniform heat distribution
and the elimination of temperature stagnation zones. Additionally, it significantly reduces
pressure losses compared to other geometric alternatives. Such a combination of properties
enables broad applicability, ranging from industrial cooling systems to high-performance
heat exchangers in the energy sector.
This study provides a valuable framework for future research into TPMS structures. It
highlights opportunities for further optimisation of geometric parameters, such as porosity,
wall thickness, and cell size, to enhance their performance further. Moreover, it emphasises
the potential of advanced manufacturing technologies, such as 3D printing, to precisely
fabricate complex geometries with high repeatability.
The study’s conclusions underscore the importance of innovative design strategies in
addressing the challenges of modern thermal management. They position the combined gy-
roid as an optimal choice for applications requiring a balance between hydraulic efficiency
and thermal performance.

Author Contributions: Conceptualisation, M.B. and R.R.; methodology, M.B.; formal analysis,
M.B.; investigation, M.B.; data curation, M.B.; writing—original draft preparation, M.B.; writing—
review and editing, M.B.; visualisation, M.B.; supervision, R.R.; project administration, R.R.; funding
acquisition, R.R. All authors have read and agreed to the published version of the manuscript.
Funding: This research was funded by KEGA: 056TUKE-4/2024, a platform for the effective creation,
evaluation, and transfer of innovations, and the efficient management of university research outputs
with a practical orientation.
Data Availability Statement: Data will be made available on request.
Conflicts of Interest: The authors declare no conflicts of interest.

References
1. Gado, M.G. Thermal management and heat transfer enhancement of electronic devices using integrative phase change material
(PCM) and triply periodic minimal surface (TPMS) heat sinks. Appl. Therm. Eng. 2025, 258, 124504. [CrossRef]
2. Khadiri, I.E.; Abouelmajd, M.; Zemzami, M.; Hmina, N.; Lagache, M.; Belhouideg, S. Comprehensive analysis of flow and heat
transfer performance in triply periodic minimal surface (TPMS) heat exchangers based on Fischer-Koch S, PMY, FRD, and Gyroid
structures. Int. Commun. Heat Mass Transf. 2024, 156, 107617. [CrossRef]
3. Barakat, A.; Sun, B. Controlling TPMS lattice deformation for enhanced convective heat transfer: A comparative study of
Diamond and Gyroid structures. Int. Commun. Heat Mass Transf. 2024, 154, 107443. [CrossRef]
Processes 2024, 12, 2943 20 of 21

4. Shahid, M.U.; Khan, M.M.; Shahid, M.N. Numerical Investigation of the Heat Transfer Rate and Fluid Flow Characteristics of
Conventional and Triply Periodic Minimal Surface (TPMS)-Based Heat Sinks. Eng. Proc. 2024, 75, 35. [CrossRef]
5. Yeranee, K.; Rao, Y.; Yang, L.; Li, H. Improved Thermal Performance of a Serpentine Cooling Channel by Topology Optimization
Infilled with Triply Periodic Minimal Surfaces. Energies 2022, 15, 8924. [CrossRef]
6. Schoen, R.; Yau, S.T. Compact group actions and the topology of manifolds with non-positive curvature. Topology 1979, 18,
361–380. [CrossRef]
7. Liu, J.; Cheng, D.; Oo, K.; Pan, W.; McCrimmon, T.-L.; Bai, S. Optimization of Triply Periodic Minimal Surface Heat Exchanger to
Achieve Compactness, High Efficiency, and Low-Pressure Drop. Energies 2024, 17, 5141. [CrossRef]
8. Wang, J.; Chen, K.; Zeng, M.; Ma, T.; Wang, Q.; Cheng, Z. Investigation on flow and heat transfer in various channels based on
triply periodic minimal surfaces (TPMS). Energy Convers. Manag. 2023, 283, 116955. [CrossRef]
9. Chen, F.; Jiang, X.; Lu, C.; Wang, Y.; Wen, P.; Shen, Q. Heat transfer efficiency enhancement of gyroid heat exchanger based on
multidimensional gradient structure design. Int. Commun. Heat Mass Transf. 2023, 149, 107127. [CrossRef]
10. Ráž, K.; Chval, Z.; Pereira, M. Lattice Structures—Mechanical Description with Respect to Additive Manufacturing. Materials
2024, 17, 5298. [CrossRef] [PubMed]
11. Overbeck, M.; Heimbs, S.; Kube, J.; Hühne, C. Energy Absorption Properties of 3D-Printed Polymeric Gyroid Structures for an
Aircraft Wing Leading Edge. Aerospace 2024, 11, 801. [CrossRef]
12. Liu, J.; Cheng, D.; Oo, K.; McCrimmon, T.-L.; Bai, S. Design and Additive Manufacturing of TPMS Heat Exchangers. Appl. Sci.
2024, 14, 3970. [CrossRef]
13. Gado, M.G.; Ookawara, S. 3D-printed triply periodic minimal surface (TPMS) structures: Towards potential application of
adsorption-based atmospheric water harvesting. Energy Convers. Manag. 2023, 297, 117729. [CrossRef]
14. Kerme, E.D.; Hajialibabei, M.; Saghir, M.Z.; El-Ketan, O. Experimental investigation of porous gyroid structure: Effect of cell size
and porosity on performance. Therm. Sci. Eng. Prog. 2024, 53, 102728. [CrossRef]
15. Moradmand, M.M.; Sohankar, A. Numerical and experimental investigations on the thermal-hydraulic performance of heat
exchangers with Schwarz-P and gyroid structures. Int. J. Therm. Sci. 2024, 197, 108748. [CrossRef]
16. Pulin, A.; Laptev, M.; Kortikov, N.; Barskov, V.; Roschenko, G.; Alisov, K.; Talabira, I.; Gong, B.; Rassokhin, V.; Popovich, A.; et al.
Numerical Investigation of Heat Transfer Intensification Using Lattice Structures in Heat Exchangers. Energies 2024, 17, 3333.
[CrossRef]
17. Kus, K.; Wójcik, M.; Malecha, Z.; Rogala, Z. Numerical and experimental investigation of the gyroid heat exchanger. Int. J. Heat
Mass Transf. 2024, 231, 125882. [CrossRef]
18. Yeranee, K.; Rao, Y. A Review of Recent Investigations on Flow and Heat Transfer Enhancement in Cooling Channels Embedded
with Triply Periodic Minimal Surfaces (TPMS). Energies 2022, 15, 8994. [CrossRef]
19. Al-Ketan, O.; Abu Al-Rub, R.K. Multifunctional Mechanical Metamaterials Based on Triply Periodic Minimal Surface Lattices.
Adv. Eng. Mater 2019, 21, 1900524. [CrossRef]
20. Beer, M.; Rybár, R. Numerical Study of Fluid Flow in a Gyroid-Shaped Heat Transfer Element. Energies 2024, 17, 2244. [CrossRef]
21. Song, N.; Pu, W.; Qiao, L.; Wu, B.; Liu, R.; Luo, F. Numerical simulation study on the heat transfer and flow characteristics
of fuel/lubricating oil heat exchanger based on triply periodic minimal surface (TPMS). Appl. Therm. Eng. 2024, 257, 124437.
[CrossRef]
22. Naji, K.; Kasidecioglu, S.A.; Ertunç, Ö.; Basol, A.M. Numerical analysis of flow and heat transfer characteristics of gyroid type
structures with triply periodic minimal surfaces. In Proceedings of the 9th CHT-24 ICHMT International Symposium on Advances
in Computational Heat Transfer, Istanbul, Turkiye, 26–30 May 2024.
23. Dam, S.; Khondoker, M.A.H. Design and Numerical Investigation of High-Performance Heat Exchangers Containing Triply
Periodic Minimal Surface Lattice Structures. Eng. Proc. 2024, 76, 76. [CrossRef]
24. Solidworks Flow Simulations Technical Reference, Dassault Systemes. 2017.
25. Padrão, D.; Hancock, D.; Paterson, J.; Schoofs, F.; Tuck, C.; Maskery, I. New structure-performance relationships for surface-based
lattice heat sinks. Appl. Therm. Eng. 2024, 236, 121572. [CrossRef]
26. Reynolds, B.W.; Fee, C.J.; Morison, K.R.; Holland, D.J. Characterisation of Heat Transfer within 3D Printed TPMS Heat Exchangers.
Int. J. Heat Mass Transf. 2023, 212, 124264. [CrossRef]
27. Yan, K.; Deng, H.; Xiao, Y.; Wang, J.; Luo, Y. Thermo-hydraulic performance evaluation through experiment and simulation of
additive manufactured Gyroid-structured heat exchanger. Appl. Therm. Eng. 2024, 241, 122402. [CrossRef]
28. Knödler, P.; Dreissigacker, V. Fluid Dynamic Assessment and Development of Nusselt Correlations for Fischer Koch S Structures.
Energies 2024, 17, 688. [CrossRef]
29. Snijders, L.A.; Koppius, A.M.; Nieuwvelt, C. An experimental determination of the turbulent prandtl number in the inner
boundary layer for air flow over a flat plate. Int. J. Heat Mass Transf. 1983, 26, 425–431. [CrossRef]
30. Wilk, J. Heat/mass transfer analogy in the case of convective fluid flow through minichannels. Int. J. Therm. Sci. 2020, 156, 106467.
[CrossRef]
31. Smith, M.; Kim, P.; Lambert, A.; Walde, M.; Lindahl, J.; Mungale, K.; Bougher, T.; Hassen, A.; Kunc, V. Maximizing the Performance
of a 3D Printed Heat Sink by Accounting for Anisotropic Thermal Conductivity During Filament Deposition. In Proceedings of
the 18th IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems, Las Vegas, NV,
USA, 28 May 2019. [CrossRef]
Processes 2024, 12, 2943 21 of 21

32. Anwajler, B.; Szołomicki, J.; Noszczyk, P.; Baryś, M. The Potential of 3D Printing in Thermal Insulating Composite Materials—
Experimental Determination of the Impact of the Geometry on Thermal Resistance. Materials 2024, 17, 1202. [CrossRef] [PubMed]
33. Gabrieli, R.; Wenger, R.; Mazza, M.; Verné, E.; Baino, F. Design, Stereolithographic 3D Printing, and Characterization of TPMS
Scaffolds. Materials 2024, 17, 654. [CrossRef]
34. Zhao, B.; Zhang, J.; Lian, W. Numerical Modeling of Heat Exchanger Filled with Octahedral Lattice Frame Porous Material.
Aerospace 2022, 9, 238. [CrossRef]
35. Choong, Y.H.; Krishnan, M.; Gupta, M. Recent Advances in the 3D Printing of Pure Copper Functional Structures for Thermal
Management Devices. Technologies 2023, 11, 141. [CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.