Journal of Physics: Conference
Series
PAPER • OPEN ACCESS You may also like
- Effect of Different Cooling Methods on
The DEPLOY! Project: Development of a Thermal Performance of Single Loop
Pulsating Heat Pipe
Deployable Pulsating Heat Pipe experiment on a N Iwata, F Bozzoli, L Cattani et al.
- Progress of cryogenic pulsating heat pipes
parabolic flight at UW-Madison
Luis Diego Fonseca, Mason Mok, John
Pfotenhauer et al.
To cite this article: A. Billi et al 2024 J. Phys.: Conf. Ser. 2685 012070
- The design of an extended length helium
pulsating heat pipe experiment
L Kossel, J Pfotenhauer and F Miller
View the article online for updates and enhancements.
This content was downloaded from IP address 203.110.242.30 on 08/04/2024 at 03:37
40th UIT International Heat Transfer Conference (UIT 2023) IOP Publishing
Journal of Physics: Conference Series 2685 (2024) 012070 doi:10.1088/1742-6596/2685/1/012070
The DEPLOY! Project: Development of a
Deployable Pulsating Heat Pipe experiment on a
parabolic flight
A. Billi1,*, R. Perna2, S. Picchi2, V. Rosellini5, E. Saltmarsh3, M. Bocelli4, N. Ricci2,
N. Miché3, F. Bozzoli4, M. Mameli2, S.Filippeschi2
1
Department of industrial engineering, University of Pisa, Pisa, Italy
2
Department of Energy, Systems Land and Construction Engineering, University of
Pisa, Pisa, Italy
3
School of architecture, technology and engineering, University of Brighton,
Brighton, UK
4
Department of Engineering and Architecture, University of Parma, Parma, Italy
5
Department of information engineering, University of Pisa, Pisa, Italy
*
[email protected] Abstract. DEPLOY! Project aims at analysing the behaviour of Deployable Pulsating Heat Pipe
(PHP) shaped as a helicoidal torsional spring in the adiabatic section on board a Parabolic Flight
platform. A PHP is a passive thermal control device where the heat is efficiently transported by
means of the self-sustained oscillatory fluid motion driven by the phase change phenomena. The
microgravity environment allows to eliminate the buoyancy force contribution in the liquid phase
momentum. Consequently, it is possible to isolate the contribution of the pressure drop caused
by the 3D arrangement and infer on their effect on the PHP performance. As a result, a proper
design based on the previous considerations would increase the flexibility of the PHP for use in
space applications without significant reductions in efficiency. The goal of DEPLOY! is to
demonstrate the functionality of a Deployable Pulsating Heat Pipe in various unfolding
configurations by analysing its thermal-hydraulic response throughout a Parabolic Flight. The
presented Deployable PHP is composed of an aluminium tube (inner/outer diameters 1.6mm/2.6
mm) and filled with HFE-7000. It is heated at the evaporator using a flat heater and cooled at the
condenser with a water-cooled cold plate. T-type thermocouples are used to measure the wall
temperature in several locations, while two pressure transducers (one located in the evaporator
section and another in the condenser section of the same pipe) measure the local fluid pressure.
Additionally, an IR Camera will be used to observe a section of the pipe for further analysis of
the flow frequency. The device operation will be tested on ground and 0-g at different heat loads
(24W, 34W), in multiple static positions corresponding to different opening angles (0°, 45°, 90°,
135°, 180°) and during its dynamic opening from 0° to 180°, thanks to a remotely controllable
motion system.
Keywords: Deployable Pulsating Heat Pipe, Parabolic flight, Microgravity, infrared
thermographic acquisitions, Inverse heat conduction problem
1 Introduction
In space missions, the persistent threat of irreparable damage presents an ongoing challenge.
Addressing risks in the harsh deep space environment involves two primary strategies: minimizing the
likelihood of risk occurrence and implementing effective risk mitigation measures. Retractable
radiators play a pivotal role in reducing risks associated with micrometeoroid impacts, not only
alleviating space constraints at launch but also minimizing the vulnerable surface area when inactive
[1]. Passively deployable radiative panels, incorporating shape memory alloys for unfolding dynamics
control, are currently under investigation as an innovative technological solution [2]. The deployable
Pulsating Heat Pipe (PHP) represents a noteworthy advancement in thermal management after over
two decades of research [3]. This experiment focuses on exploring the thermal response of deployable
Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd 1
40th UIT International Heat Transfer Conference (UIT 2023) IOP Publishing
Journal of Physics: Conference Series 2685 (2024) 012070 doi:10.1088/1742-6596/2685/1/012070
PHP in microgravity conditions. Limited scientific literature addresses various flexible PHP devices,
such as the flexible flat plate PHP, examined in different configurations, including a polycarbonate
copper multi-layer structure [4] and a polypropylene semi-flat double sheet [5]. The latter, tested
during the ESA 77th parabolic flight campaign, demonstrated promise. Another flexible PHP,
developed in Japan with stainless steel flexible micro pipes, yielded valuable outcomes on
microgravity platforms like the REXUS ballistic rocket and parabolic flight plane [6] [7]. The device
under study, tested in three unfolding configurations (0°, 90°, and 180°) and three heating power
levels (14W, 24W, and 34W) during the 79th ESA Parabolic flight campaign, showed encouraging,
albeit inconclusive, results. Notably, the PHP configuration had minimal impact on working
conditions when the evaporator temperature exceeded 50 °C. The current experiment aims to gather
crucial information on device operation, particularly under more demanding conditions with higher
heating power inputs. Key novelties and upgrades from previous experiences will be explored.
i) investigation of intermediate unfolding angle configurations (45 and 135°) to produce clearer
indication on the effect of the device configuration on its working conditions;
ii) implementation of a remotely controlled actuation subsystem to analyze the performances of
the device during its unfolding dynamic (from closed/0°, to open/180°) in microgravity
conditions;
iii) inclusion of an IR Camera which will record the temperature variation on the outside of portions
of multiple channels close to the evaporator section. Such observation outcomes will be useful
for future analyses.
The test campaign will take place in November 2023 and will comprehend static and dynamic conditions
testing.
2 Experiments
The experimental setup, shown in Figure 1, comprises a containment box and a control rack, designed
to satisfy the platform interfaces requirements and safety guidelines. The containment box houses the
Deployable PHP (i.e., the test cell) supported by a structural framework. Additionally, the containment
box accommodates components of the actuation and thermal subsystems, as well as an IR Camera. The
control rack serves as the central control centre for the experiment. It incorporates power supplies, the
cooling, data acquisition and actuation management subsystems. These elements provide the necessary
control and measurement capabilities to tune the experiment and collect relevant data.
2.1 Test cell
Experiment
al box
Control
IR rack
camera
Control
panel
Cooling
loop
Stepper
motor PHP
Figure 1 Experimental set up: on the right the control rack, on the left the experimental box.
2
40th UIT International Heat Transfer Conference (UIT 2023) IOP Publishing
Journal of Physics: Conference Series 2685 (2024) 012070 doi:10.1088/1742-6596/2685/1/012070
The Deployable Pulsating Heat Pipe under study, depicted in Figure 2, features a capillary tube (1.6/2.6
mm I.D./O.D.) made of aluminum 6063. It is bent into a closed-loop serpentine with 11 U-turns in the
evaporation section, resulting in 22 channels. The adiabatic, evaporator, and condenser sections are
715mm, 10mm, and 75mm long, respectively. Shaped as a torsional spring in the adiabatic section, the
foldable PHP has a coil diameter (d) of 65mm, N = 3.5 coils per connection, totaling Lcoil = 715mm
Figure 2 Deployable pulsating heat pipe (folded configuration). The heater is in red, the cold plate is
in blue.
(90% of the channel length). The PHP is partially filled with HFE-7000 at a volumetric filling ratio of
70%. The evaporator section, as shown in Figure 2, is situated between two aluminum heat spreaders
fastened to the tubes with screws. A flat, 500mm long heater (Joule Heater Hotspring®/Maxi - type
WRP) connected to the spreaders provides heat input. A cold plate extracts heat from the condenser
section and maintains a constant temperature, connected to a Liquid Loop System (LLS). The actuation
subsystem (explained in section 2.2.2) deploys the condenser section while the evaporator remains fixed
to the support structure.
2.2 Experimental setup
The experimental apparatus comprises multiple subsystems situated in two primary support structures.
. As shown in Figure 1, the experimental setup consists of a control rack on the right and an experimental
box on the left. The control rack houses the cooling loop (blue), automation and heating subsystem
power supplies (small shelves on the left), a control panel with laptops (light blue), a monitoring screen
(black), and control switches. The experimental box includes the Deployable PHP supported by an
aluminium structure (light grey), the stepper motor with torque transmission, and an IR camera (blue).
Figure 3 Experiment assembly - block description.
The main novelties of this experiment reside in the data acquisition subsystem composition and the
actuation subsystem, that are described in more detail in the next sections. The block scheme in Figure
3
40th UIT International Heat Transfer Conference (UIT 2023) IOP Publishing
Journal of Physics: Conference Series 2685 (2024) 012070 doi:10.1088/1742-6596/2685/1/012070
3 helps visualizing the connections between within and between subsystems. In red and blue the PHP
and the thermal subsystem, the actuation subsystem in yellow, the data acquisition subsystem in purple.
2.2.1 Data acquisition subsystem
The system is composed by two main data paths. The first is the one connected to the IR Camera, the
second one is related to the sensors connected to the PHP.
The IR Camera (Flir 7000) is in front of the evaporator section of the PHP, at 254 mm of distance.
The device captures images at a 25Hz frequency, focusing on five pipes in direct contact with the
evaporator section (60 mm height). Subsequently, thermal imaging data undergoes processing to extract
a rectangular pixel section (approximately 20 pixels wide and 40 pixels high), representing the central
image of a selected pipe. The acquired data is then filtered through the Altair acquisition program.
Boundary conditions recorded during the campaign are integrated into Altair. A thermocouple (TC)
strategically placed on the PHP in the observed area serves as a reference for IR Camera readings.
Further processing occurs in a MATLAB script, transforming records into n-dimensional matrices, with
'n' corresponding to a single frame. Finally, the n-dimensional matrix undergoes processing using the
Inverse Heat Conduction Problem (IHCP) algorithm. This algorithm estimates heat flow exchanged by
the device and provides a reading on the pulsation frequency of the filling fluid, starting from a known
temperature [8]. During the test campaigns, real-time sensor data are displayed and monitored to ensure
the safe and optimal performance of the experimental setup. All collected data are stored for further
analysis in multiple external storages. Pressure and temperature readings are plotted, and the results
obtained from ground testing and parabolic flight campaign are compared to evaluate the influence of
altered gravity on the PHP working conditions.
Temperature and pressure readings are performed during both the ground tests and the three parabolic
flight whole duration. A total of 15 T-type thermocouples (±1 °C) are used to measure temperature.
These thermocouples are positioned on the surfaces of 5 selected pipes: 5 at the evaporator section, 5 at
the condenser section, and 5 at the adiabatic section. Additionally, 2 pressure transducers (Keller® PAA-
33X, ± 0.5 kPa) are placed at the evaporator and condenser extremities of the same loop. To collect the
data, a cRIO data acquisition system is utilized. All the collected data is transmitted to a computer
equipped with data visualization software, allowing for real-time display and storage of the data.
Following the completion of the parabolic flight campaign, the collected data is analysed and compared
with the results obtained during the ground tests. Both the pressure transducers and the thermocouples
readings are collected with a 1000Hz frequency.
2.2.2 Actuation subsystem
This subsystem is composed by a Nema 34 stepper motor (5.9 Nm torque), a transmission, a controller,
and their power supplies. The motor changes the PHP configuration by opening it and closing it by 45°
steps from a closed configuration (0°) to an open configuration (180°). In Figure 4 the PHP is depicted
in all the tested configurations. The stepper motor is connected to the PHP condenser section support
shaft with a pulley + belt system with a gear ratio of 1 to 4. The motor is controlled by the driver board,
which redirects power coming from a PSH 6066-A power supply and controls from an Arduino MEGA
to the motor. The Arduino MEGA is powered and controlled by the data acquisition laptop, but its
primary input source is a switch control panel located on the control rack next to said laptop. Safety
measures are implemented to prevent damage to the experiment caused by accidental extra rotations of
the condenser section via limit switches located at the end of the rotation span. The condenser weight is
of 5.5 Kg and the maximum torque required to hold its position is of approximately 15N*m, well below
the top torque transmission from the actuation of over 20 N*m. To avoid complications, no rotation is
4
40th UIT International Heat Transfer Conference (UIT 2023) IOP Publishing
Journal of Physics: Conference Series 2685 (2024) 012070 doi:10.1088/1742-6596/2685/1/012070
planned to happen during the hyper gravity phase of the parabolic flight, and the rotation speed is kept
at lowest possible value of π/10 rad/s at the PHP shaft.
Figure 4 PHP unfolding positions. from top left to bottom: 0°-closed, 45°, 90°, 135°. 180°-open
Figure 5 Actuation subsystem particular: from the left stepper motor, motor support,
transmission (pulleys and timing belt), PHP shaft.
2.3 Experimental procedures
The experiment consists of two different test campaigns: a parabolic flight test and a ground test. The
first is the main focus of the investigation, allowing an observation of the working conditions of the
device in altered gravity, while the ladder is performed before it to set up a baseline for the tested values
and establish the most functional test matrix to be designed for the PFC.
Ground tests serve as a baseline for subsequent Parabolic Flight Campaign tests. The objective is to
establish a consistent comparison framework and all activities performed during the flight will be
simulated.
The Parabolic Flight Campaign (PFC) is scheduled to take place in November 2023 aboard the
Airbus A310 Zero-G aircraft of Novespace. The campaign is designed to be conducted over a span of
three flight days, with each day consisting of 30 microgravity intervals lasting for 20 seconds each,
interspersed with hyper-gravity intervals of 20 seconds at 1.8g. During a Zero-G flight, the parabolic
maneuver is divided into three stages. Firstly, there is the pull-up stage, where the pilot lifts the nose of
the A310 Zero G airbus to an angle of 45 degrees. This action subjects the experimental apparatus to
5
40th UIT International Heat Transfer Conference (UIT 2023) IOP Publishing
Journal of Physics: Conference Series 2685 (2024) 012070 doi:10.1088/1742-6596/2685/1/012070
hyper gravity, a pull that is 1.8 times than Earth's gravity. The pull-up, or "nose-up," lasts for
approximately 20 seconds. Next is the parabola stage. As the aircraft ascends, the pilots reduce engine
speed, and the aircraft follows a ballistic trajectory. Weightlessness begins when the aircraft enters a
parabola, during which it experiences free fall for about 20 seconds. Finally, there is the pull-out stage.
The nose of the plane is tilted downward by 45 degrees, and the pilots gradually level off the aircraft by
increasing engine speed. Like the pull-up stage, the experimental apparatus is subjected to hyper gravity,
1.8 times stronger than Earth's gravity. After 20 seconds, the A310 Zero G returns to a horizontal
trajectory. This maneuver is performed 30 times for each flight, in groups of 5 parabolas each. After
performing 5 maneuvers, the aircraft enters a phase of 5 to 10 minutes of steady horizontal flight before
proceeding to the next stage.
The PFC test sessions follow a predefined procedure to ensure compliance with safety and technical
requirements in the environment. All experiments, including the assembly for this experiment on the
plane, are powered up after take-off and must be shut down before the landing maneuver begins. The
first set of 5 parabolic flight maneuvers begins 20 to 30 minutes after take-off, which aligns with the
expected preparation time for the experiment to become fully operational before the injection of the first
maneuver. Ground tests are conducted to verify this expectation. If the system takes longer than the
allocated interval, the first 1 or 2 parabolic maneuvers of the first set may not be used for data collection,
but 3 to 4 parabolic maneuvers will still be available for testing. Additionally, the parabolic flight
consists of six groups of maneuvers, and the experiment is planned to be tested in five different
configurations, reserving one group of parabolic maneuvers for a backup attempt.
On the first and second day, the tests resemble the static tests conducted during ground. The system
is powered up, and the PHP is in its initial position (closed at 0°). When operational conditions are
achieved, the test session begins, and data from the first set of 5 parabolic maneuvers are collected. At
the end of this batch, the actuation subsystem adjusts the PHP angle from 0° to 45°. It is expected that
the system will be minimally affected by this change, and the PHP should reach new steady-state
conditions in a few seconds. The second set of parabolic maneuvers is used to test the device in the
second configuration. This procedure repeats until the end of the fifth batch. The last parabolic session
can serve as a backup in case any of the previous measurements were not optimally recorded. At the end
of the final batch, the PHP is returned to its closed starting position, and the system is shut down before
landing. This process is repeated similarly on the first and second flight days. The difference between
the two flight tests lies in the thermal conditions to which the PHP is subjected. On the first day, the
evaporator input power is 34W, resulting in a heater temperature of 110°C, while the cold plate
temperature remains at 23°C throughout the test. The thermal conditions for the second day are expected
to be different, with specific values determined after analysing data from the ground test campaign.
The third test session, on the final day of PFC, focuses on analysing the behaviour of the PHP in a
dynamic situation. The pre-test procedures are the same as on previous days, with the system expected
to reach nominal conditions within the same timeframe. The dynamic action is described in Figure 6.
At the start of the first viable microgravity interval, the automation subsystem unfolds the device by
rotating the condenser section 180° around the coil axis, transitioning the PHP from its initial position
(open at 0°) to its open position (180°). This deployment occurs between 2 and 4 seconds after the
injection and lasts for 10 seconds. Once the final position is reached, the PHP remains still until the end
of the hyper gravity interval (approximately 30 seconds). During the subsequent steady horizontal flight
interval, the condenser is returned to its starting position by the actuation subsystem. It may take several
minutes for the PHP to return to its initial nominal conditions, making the following parabola unsuitable
for data collection. This test is expected to be repeated during the first, third, and fifth parabolas of each
batch. This test matrix allows the collection of three relevant samples in six different thermal
configurations. For each batch, thermal conditions such as heating input and condenser temperature can
be adjusted, creating a diverse set of data points. The initial thermal conditions match those of the first
flight day, with subsequent conditions determined based on ground test results.
6
40th UIT International Heat Transfer Conference (UIT 2023) IOP Publishing
Journal of Physics: Conference Series 2685 (2024) 012070 doi:10.1088/1742-6596/2685/1/012070
0 28 32 42 53 75 85
1,8 g 0g 1,8 g 1g
” ” ” ” ” ” ”
Figure 6 PFC day 3: dynamic test configuration.
3 Data analysis and expected results
The device's thermal resistance, without a working fluid, is determined to be 3 K/W based on prior
measurements [6]. This resistance arises from heat conduction through the tubes and fluid flow
resistance during operation. With a working fluid, the effective resistance is influenced by lower
resistance from internal motions. Computation of thermal resistance involves mean temperatures at the
evaporator and condenser and power input. The combined standard uncertainty considers diverse
parameters, ensuring a comprehensive evaluation. The pressure drop model assumes incompressible
fluid, no oscillation, full power transfer to the PHP, and 1D heat transfer. Slug motion is described by
the momentum equation, considering gravity, hydrostatic pressure, and shear forces. The simplified
equation includes contributions from gravity and shear forces, defining the pressure drop.
𝑓𝑠
Δ𝑃 = 𝜌𝑔⃗ cos(𝜃) + 𝑙(1 − 𝛼) ( 𝜌 ⃗⃗⃗⃗)
𝑣
2𝑑𝑖 𝑙 𝑙
To determine the fluid velocity, a wavelet analysis will be conducted [5] employing a complex Morlet
wavelet. The rationale behind this selection is the resemblance between the wavelet shape and the signal
profile [6]; a strong correlation between the two shapes leads to improved quality outcomes. Using the
Morlet wavelet as the mother wavelet, the transformation equation, with scale factor s and time shift
factor u, takes the following form.
+∞
1 1 𝑡−𝑢 (𝑡 − 𝑢)2
𝑊𝑇(𝑢, 𝑠) = ∫ 𝑓(𝑡) 𝜋4 exp [𝑖𝜔0 ( )] exp [− ] 𝑑𝑡
−∞ √𝑠 𝑠 2𝑠2
Once the dominant frequency is determined, it is possible to assume that this frequency leads to the
mean velocity of the individual liquid slug. It is then straightforward to calculate 𝑣⃗ as 𝑣⃗ = 𝑙 ⋅ 𝑓𝑧 .
Thermal imaging data is employed to extract a specific rectangular pixel section, measuring 5 pixels
in width and 3 pixels in height, representing the central image of a chosen pipe. The records are filtered
using the Altair acquisition programme, integrating campaign-recorded boundary conditions and
referencing a thermocouple connected to a PHP point observed in the IR camera window. Processed
through a MATLAB script, records transform into n-dimensional matrices (n for a single frame). This
n-dimensional matrix feeds an Inverse Heat Conduction algorithm, estimating device flow and
outputting the filling fluid's pulsation frequency [8].
4 Conclusions
The DEPLOY! Project aims at the technological advancement of the deployable PHP field. The device
at study is an innovative foldable pulsating heat pipe which has been tested once before in a parabolic
flight in 2022. This experiment is presented with novelties such as an actuation subsystem which allows
7
40th UIT International Heat Transfer Conference (UIT 2023) IOP Publishing
Journal of Physics: Conference Series 2685 (2024) 012070 doi:10.1088/1742-6596/2685/1/012070
to change the configuration of the device during the tests and an IR camera to analyse the behaviour of
the working fluid starting from its surface temperature. This experiment contributes to enhance the
knowledge on the deployable PHP, by investigating its working conditions in new configurations and
at different temperatures. The dynamic deployment test, performed in simulated microgravity, is a
novelty of this project; such test will shed light on an important working life condition that the device
will face. The results of this test are important for a first estimate of the influence of the peculiar dynamic
of this device on its working conditions and will offer insights on future testing configurations worth
investigating for a complete understanding of this correlation. The experiment will be tested during a
parabolic flight campaign in November 2023. The results are expected to enhance the knowledge on this
device and generate further testing opportunities for the future.
Acknowledgements
The present work has been carried out as part of the ESA Academy Experiments Programme, supported
by the European Space Agency (ESA) through the ESA Education Office, the European Low Gravity
Research Association (ELGRA) and the UK Space Agency. The test campaign will take place in
Bordeaux (FR) in the Novespace company facilities. The test cell is developed in and based on
components of the laboratories of University of Pisa Energy, Systems, Territory and Construction
Engineering Department (DESTeC), University of Brighton Advanced Engineering Centre, and
University of Parma Engineering and Architecture Department.
References
[1] NASA, “State-of-the-Art of Small Spacecraft Technology - Thermal Control,” 2022.
[2] A. Bacciotti, F. Bucchi, F. Frendo, M. Mameli, R. Perna, and S. Filippeschi, “On the use of
shape memory alloys for deployable passive heat radiators in space satellites,” IOP Conf. Ser.
Mater. Sci. Eng., vol. 1038, no. 1, p. 012061, 2021, doi: 10.1088/1757-899x/1038/1/012061.
[3] M. Mameli, G. Besagni, P. K. Bansal, and C. N. Markides, “Innovations in pulsating heat
pipes: From origins to future perspectives,” Appl. Therm. Eng., vol. 203, no. October 2021, p.
117921, 2022, doi: 10.1016/j.applthermaleng.2021.117921.
[4] C. Jung, J. Lim, and S. J. Kim, “Fabrication and evaluation of a high-performance flexible
pulsating heat pipe hermetically sealed with metal,” Int. J. Heat Mass Transf., vol. 149, 2020,
doi: 10.1016/j.ijheatmasstransfer.2019.119180.
[5] O. Der, A. A. Alqahtani, M. Marengo, and V. Bertola, “Characterization of polypropylene
pulsating heat stripes: Effects of orientation, heat transfer fluid, and loop geometry,” Appl.
Therm. Eng., vol. 184, no. November 2020, p. 116304, 2021, doi:
10.1016/j.applthermaleng.2020.116304.
[6] R. Perna et al., “Experimental investigation of a coil shaped deployable pulsating heat pipe,”
pp. 1–8, 2023.
[7] S. Filippeschi et al., “Thermal performance analysis of a Pulsating Heat Pipe with a long
adiabatic section tested with different working fluids,” THERMINIC 2022 - 28th Int. Work.
Therm. Investig. ICs Syst. Proc., vol. 72, pp. 22–26, 2022, doi:
10.1109/THERMINIC57263.2022.9950642.
[8] L. Pagliarini et al., “Thermal characterization of a multi-turn pulsating heat pipe in
microgravity conditions: Statistical approach to the local wall-to-fluid heat flux,” Int. J. Heat
Mass Transf., vol. 169, p. 120930, 2021, doi: 10.1016/j.ijheatmasstransfer.2021.120930.
