Nuclear Engineering and Design 340 (2018) 9–16

Contents lists available at ScienceDirect

Nuclear Engineering and Design

journal homepage: www.elsevier.com/locate/nucengdes

A miniature integrated nuclear reactor design with gravity independent T

autonomous circulation
⁎
Qin Zhoua, Yan Xiab,c, , Guoqing Liub,c, Xiaoping Ouyangd
a
Institute of Nuclear and New Energy Technology of Tsinghua University, the Key Laboratory of Advanced Reactor Engineering and Safety, Ministry of Education, Beijing
100084, China
b
Beijing Institute of Spacecraft Environment Engineering, Beijing 100094, China
c
Science and Technology on Reliability and Environmental Engineering Laboratory, Beijing 100094, China
d
State Key Laboratory of Intense Pulsed Radiation Simulation and Effect, and Radiation Detection Research Center, Northwest Institute of Nuclear Technology, Xi’an
710024, China

A R T I C LE I N FO A B S T R A C T

Keywords: A miniature integrated nuclear reactor design with gravity independent autonomous circulation (ACMIR) was
Autonomous circulation newly proposed. The reactor core, energy transfer system of Stirling and linear electric motors are integrated in
Integrated nuclear reactor the reactor pressure vessel to achieve high power density and autonomous circulation capability. The coolant
Stirling cycle circulation is autonomously driven by gas expansion at heat end and compression at cool end thus is in-
Independence on gravity
dependent on gravity. Twelve sets of rotary drum controller and reflector are used to regulate reactivity outside
Inherent safety
the reactor pressure vessel. The physics and thermodynamic properties, as well as the safety performances are
analyzed. According to these analysis, inherent safety characteristics are obtained, and the reactor is capable to
shut down and remove residual heat passively without any external intervention, in accident conditions such as
loss of external power supply, overtemperature/overpressure of the reactor, impact when rocket launching or
landing, stagnation of the displacer or piston, et al. The integration, less pipes design, gravity independent
autonomous circulation features make it a good candidate for space flight propulsion, the Moon and Mars base
power supply, the deep-sea or other tilt and swing situation applications.

1. Introduction
As a kind of clean energy which provides large scale stable electric
power, nuclear energy is one of the most important means to alleviate
the contradiction between energy needs and environment. The key is-
sues to the application of nuclear energy is its safety. With the devel-
opment of nuclear technology, most of Generation III and Generation IV
nuclear power systems have inherently safe characteristics, as with key
assurance techniques such as full power or partial power natural cir-
culation, passive residual heat removal system, et al. (Abram and Ion,
2008; Rowinski et al., 2015).
Man is exploring the universe deeper while the solar power gets less
available and adequate, especially for missions beyond Mars (Oman,
2003; Summerer and Stephenson, 2011; Woo and Lee, 2014; Nuclear
Reactors to Power Space Exploration, 2017). Even for Moon surface
bases’ usage; solar power faces the shortage problem as one moon night
lasts for a half earth month. ESA and NASA have been conducting re-
searches on nuclear power sources (NPS) in space for several years
(Oman, 2003; Summerer and Stephenson, 2011; Woo and Lee, 2014;
Nuclear Reactors to Power Space Exploration, 2017; Bennett, 2008;
Cassady, 2008). Thus, space usage friendly nuclear electric power of
scores or hundreds of kilo watt is a critical requirement.
Space NPS includes radioisotope power sources (RPSs) and space
nuclear fission reactors. RPSs convert decay heat of radioisotopes into
electricity. They are relatively small and satisfy mainly under 1 kilo
watts supply need. Nuclear fission reactors are relatively large, and are
more economic for space power demands of Moon/Mars bases and deep
solar system explorer propulsion (Oman, 2003; Nuclear Reactors to
Power Space Exploration, 2017).
Space nuclear fission reactors are principally similar to terrestrial
reactors, but have to make structurally and functionally space adaptive
modifications in all parts like reactor cores, cooling and heat transfer
system, control system, energy conversion system, and heat rejection
system. Even though, as shown in Fig. 1-1, typical terrestrial pressure
water reactor nuclear power plant and space nuclear fission reactor
power system are still the same in taking two functional units coupling

⁎
Corresponding author at: Beijing Institute of Spacecraft Environment Engineering, Beijing 100094, China.
E-mail address: xiayan_cast@163.com (Y. Xia).

https://doi.org/10.1016/j.nucengdes.2018.09.013
Received 5 July 2018; Received in revised form 3 September 2018; Accepted 14 September 2018
Available online 28 September 2018
0029-5493/ © 2018 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/BY-NC-ND/4.0/).
Q. Zhou et al.
Fig. 1. -1. Typical pressure water reactor nuclear power plant and space nuclear fission reactor power system (RPV: Reactor Pressure Vessel, SG: Steam Generator,
TG: Turbo-Generator).
Fig. 1. -2. Typical space nuclear power system and autonomous circulation micro integrated nuclear reactor (ACMIR) in this article.
structure, miniature integrated reactors are barely designed and dis-
cussed.
The cooling choices of space reactor mainly include liquid metal
cooling, gas cooling, liquid metal heat pipe, et al. Liquid metal heat
pipes have high security performance via avoiding single point failure,
but technique for transferring hundreds of kilo watt or larger power
from a small reactor core via heat pipes is immature (Summerer and
Stephenson, 2011; Woo and Lee, 2014). Liquid metal cooling and gas
cooling have no large differences to those of terrestrial reactors, and
thus have higher technological maturity. But terrestrial coolant natural
circulation is generally driven by gravity and coolant density difference,
which is no longer valid in space environment Thus, pumps or draught
fans have to be used in space reactors for driving gas or metal coolant,
and bring risk of core melting caused by pump or draught fans failure
and loss of cooling capability (Poston, 2001; Powell et al., 2003; El-
Genk and Tournier, 2004; Na and Upadhyaya, 2006; Na, 2006; Na and
Upadhyaya, 2006; Na and Upadhyaya, 2006; King and El-Genk, 2006;
Maise et al., 2006, 2006.; Na and Upadhyaya, 2007; King and El-Genk,
2007, 2007.; King and El-Genk, 2007; Schriener and El-Genk, 2008;
Poston, 2008; King and El-Genk, 2009; Craft and King, 2009).
The energy conversion choices of space reactor mainly include
thermocouple, thermion, Stirling cycle, and close Brayton cycle, et al.
Thermocouple conversion methods are most widely used at present for
its composing none moving components. However, thermocouples have
poor radiation resistance and low energy conversion efficiency mostly
under 10%, – extremely high hot end temperature beyond 1200 K is
necessary to promote beyond 10% conversion efficiency but brings
challenges to high-temperature tolerant materials and structures. Thus,
ESA and NASA have been vigorously promoting other energy conver-
sion methods as Stirling cycle and close Brayton cycle. Compared to
10
Nuclear Engineering and Design 340 (2018) 9–16
thermocouples, they have much higher energy conversion efficiency,
but also have larger and more complex mechanical structures, more
moving components, especially for close Brayton cycle (Summerer and
Stephenson, 2011; Woo and Lee, 2014).
This article presents a work on mainly space use dedicated minia-
ture integrated nuclear reactor design with gravity independent au-
tonomous circulation (ACMIR for short) and its preliminary reactor
physics and thermal hydraulic analysis. The design achieves inherently
safety and realizes space autonomous circulation independent on
gravity. It also reduces mass and avoided complex and frail mechanical
structures of energy conversion units. It provides higher energy con-
version efficiency at relatively lower hot end temperature. It contains
less pipes and is more likely to survive impact during launching.
The main idea for achieving both key features of gravity in-
dependent coolant autonomous circulation and simplified energy con-
version system is to integrate the reactor core and energy conversion
system together in the reactor pressure vessel (RPV), i.e., integrate two
function units into one, as shown in Fig. 1-2.
The design is a dynamic reactor with ingenious and novel reactor
structures. The reactor core continually heats the gas coolant making it
expand and push moving parts to move. Moving parts does work to
electric machinery, let coolant cool and compress contributing extra
moving forces. The coolant is autonomously driven by gas expansion at
heat end and compression at cool end thus is independent on gravity.
The coolant in RPV takes two roles, on the one hand is reactor core
cooling and transferring heat to energy conversion system, on the other
hand is working medium of conversion system and doing work to
electric machinery.
Q. Zhou et al.
Fig. 2. -1. The structure of ACMIR system.
2. Design description
The ACMIR system is designed to provide about 200 k We for more
than 5 years of full power operation and is cooled by helium gas. The
reactor is composed of several components, as RPV, the reactor internal
components, the reactor core, the displacer, the piston, the linear
generator, rotary drums, reflectors, heat pipes, and the shield struc-
tures, et al., as shown in Fig. 2-1.
The components used for converting fission energy into electric
power, constructing coolant flow channel are integrated in RPV,
namely the reactor internal components, the reactor core, the displacer,
the piston, the linear generator. The components used for rejection of
waste heat, control of reactor core reactivity, and shield of radiation are
arranged outside the RPV, namely the rotary drums and reflectors, the
heat pipes, the shield structures.
The lower and narrower part of RPV contains reactor internal
components (RIC), the reactor core, the displacer, and the piston. The
upper part of RPV is wider, and contains the linear motor/generator.
The reactor internal components are fixed to the side wall of the RPV,
and divide lower RPV into two concentric spaces—the coolant flows
upwards in the outer space and downwards in the inner space through
the core.
11
Nuclear Engineering and Design 340 (2018) 9–16
The displacer and piston are allowed to move up and down in the
internal components enclosure. A rod passing through the center of
piston connects displacer and a series of elastic structures in the top part
of the RPV. The piston connects to the moving coils linear generator
and another series of elastic structures. The space between piston and
displacer is the compression chamber, and the space under the displacer
and containing the reactor core is the expansion chamber. The elastic
segments are specially designed leaf springs, and assure the displacer
and piston reciprocating. They are made of austenitic stainless steel SA
316. There are some other choices as elastic segments, such as gas
springs.
The fixed coils of linear generator are fasten to the upper part of
RPV, and moving coils of linear generator fix to the top of piston and
move along with the piston. By this way, mechanical energy is trans-
ferred to electrical energy via linear generator.
The reactor core is formed by 276 fuel elements in diameter of
14 mm. The enrichment of these uranium nitride fuel is 67.5%. The
diameter and height of reactor core are both 25 cm. Outside the reactor
core, there is a molybdenum rhenium alloy (Mo-14%Re) layer of
thickness of 2 mm.
The 12 rotary drum controllers and reflectors are placed in a cir-
cular arrangement outside the lower part of the RPV and of the reactor
core height. One half of each rotary drum edge is gadolinium absorber,
with the rest of the drum is the same reflector material of graphite as
the container cylinder is composed of.
The rotary drums are mechanically stable facing the reactor core
with the side of gadolinium absorbers. Turning each drum away from
its stable phase is controlled by one electric motor separately to reg-
ulate the reactivity of the reactor core. If the electric motors of rotary
drums lose their power, the drums would turn back to face the core
with gadolinium, and shutdown the reactor.
At the same time, rotary drums are able to partially move out of
reactor core active range. There is a compressed spring at bottom of
each rotary drum, which can push this rotary drum to move upwards
once its balance force gone. There is a cylinder containment to the top
of each rotary drum, in which its electric motor occupies. The cylinder
containments moves vertically along with the drum, but the cylinder
containment itself can't rotate. On top of each cylinder containment, a
pressure gas chamber is set, and all 12 gas chambers open to 1 thin
secondary gas chamber circularly set on their top. This secondary gas
chamber is connected to the RPV inner gas space through a safety valve
and an outer safety relief containment (not shown in the Figure)
through another valve. At normal working conditions, high pressure gas
in gas chambers counteracts the bottom spring force and keeps the
cylinder containments and rotary drums in positions. If the reactor gets
over pressure, the RPV safety valve goes open, and the secondary gas
chamber is relatively small and would get over pressure immediately,
then valve connecting the safety relief containment would open too to
discharge RPV pressure. At this time, gas of all drum gas chambers goes
off to the safety relief containment as well and losses pressure. The
springs lose its balance force and press the containments along with
drums into gas chamber. The reflectors moves along upwards out of the
core range for a gas chamber height and shutdown the reactor. In ad-
dition, each spring is partially restricted by a movable pin at working
temperature. If the reactor gets over temperature, the pins would re-
tract to make the springs release, which partly pushing reflectors out of
working region to lower down the reactivity. Thus, ACMIR has two
independent passive shutdown means.
Outside the lower part of RPV, above the reactor core height range,
dozens of heat pipes cling to the side wall of RPV, making that part of
RPV to be the cold end of the reactor. The part of heat pipes contacting
to RPV is the hot end of heat pipes, and the cold end of heat pipes
discharges the waste heat via panel radiator (not shown in the Figure)
of an area about 150 square meters.
Q. Zhou et al. Nuclear Engineering and Design 340 (2018) 9–16

3. Operational principles Table 1-1

Nomenclature.
When temperature difference exists between hot end and cold end A Area (m2)
of the reactor, the displacer and piston will move with a certain phase f Frequency (Hz)
angle passively. Their movements cooperate with gas coolant flux, form M Mass of working fluid (mol)
an autonomous circulation, and transfer thermal energy into mechan- P Power (W)
p Pressure (Pa)
ical energy. s Moving element stroke (m)
The operation process of the reactor is similar to Stirling cycle, t Time (s)
while conventional Stirling generators are external combustion engines, T Temperature (K)
while in this design the hot source of ACMIR is inside RPV with Striling V Volume (m3)
Vsw Swept volume (m3)
structures, but this special structure doesn't affect the principle of
Stirling cycle. Greek symbol
The output electric power and moving stroke of displacer and piston ω Pulsation (rad/s)
α Swept volume phase angle (rad)
is determined by reactor core thermal power, which is monitored by αdp Displacer-piston phase angle (rad)
rotary control drums outside RPV. η Efficiency
Unidirectional channels are set at particular regions on barrel wall κ Swept volume ratio
of reactor internal components and the displacer. While the displacer τ Expansion-compression chamber temperature ratio
and piston are moving with a certain phase angle, the helium coolant Subscripts
are circulating autonomously driven by temperature difference between C Compression chamber
CC Compression chamber minimum (volume)
the hot end and cold end of reactor. The coolant circulation flow
d Displacer
pathway is shown by arrows in Fig. 2-1. dc Compression side of displacer
de Expansion side of displacer
4. Modeling results E Expansion chamber
EE Expansion chamber minimum (volume)
h Heater
4.1. Physics properties
i Indicated mechanical (power)
k Cooler
The physics analysis described in this article based on results ob- mean Average
tained via Monte Carlo N-Particle version 5 (MCNP 5). MCNP 5 is p Piston
R Regenerator
distributed in pre-compiled form, and MCNP library ENDL851 is se-
r Ideal gas constant
lected. Each results are achieved via 1000 cycles calculation, 5000
particles per cycle, and the first 50 cycles are abandoned.
The obtained Keff results were used as a basis to adjust the enrich- Table 4.1-1
ment of fuels, size and distribution of fuel elements, thickness of the Dimensions of base case ACMIR with a 25 cm active core diameter.
reflector, and size of rotary drum controller. The dimensions of base
Parameter Value
case ACMIR with a 25 cm active core diameter is shown in Table 4.1-1.
There are two separate reactivity control mechanisms in ACMIR, a Active core
one is rotation of drum controllers to adjust absorbers positions, and the Active core diameter 25 cm
other is ejection of drum controllers to remove reflectors. In normal Active core height 25 cm
Active core length/width ratio 1
working conditions, curvilinear Gadolinium controllers’ orientation
Coolant He
turns with the rotary drums to monitor the reactivity. In accident b Uranium nitride fuel
conditions, all 12 controllers passively turn facing the reactor core and Enrichment 67.8%
shutdown the reactor. In some serious accident situations, some or all Fuel assembly diameter 1.4 cm
reflector along with the rotary drums would be ejected out of working Fuel assembly pitch 1.6 cm
Fuel assembly amount 276
area, and shutdown the reactor. c RPV and IC
The quantitative analysis on reactivity monitoring efficiency of Mo-Re alloy inner/outer diameter 25/25.4 cm
above mentioned two mechanisms were simulated with MCNP 5. Displacer inner/outer diameter 26/30 cm
Dimensions in Table 4.1-1 are used as reactor structure parameters. Internal components inner/outer diameter 30/32 cm
RPV inner/outer diameter 34/40 cm
In normal conditions, 12 drums are kept intact in working positions,
d C reflector and Ga controller
setting 0, 4, 8, 12 Gadolinium controllers facing towards the core, the Graphite reflector inner/outer diameter 40/80 cm
reactivity results are shown in Table 4.1-2 and Fig. 4. We could see from Rotary drum controller amount 12
the results, 4 controllers facing the reactor core is a proper working Gadolinium controller thickness 6 cm
condition and is marked in Fig. 4-1.
In case of drums rotation mechanisms being out of control, leaving a
worst case that all controllers turn back to the core, reactivity mon- Table 4.1-2
itoring efficiency by ejection of the reflectors/drums are shown in Reactivity monitoring efficiency of drum controllers’ rotation mechanism.
Table 4.1-3 and Fig. 4-2. Ejection of more than 2 reflectors is enough to Number of Gadolinium Number of reflector Keff
shut down the reactor, requiring no more interventions. controllers towards the reactor drums in intact
core condition
The ACMIR has a negative temperature coefficient. Setting a work
condition as marked in Table 4.1-2, i.e., all drums are intact in posi- 0 12 1.00919 ± 0.00128
tions, while 4 controllers turn towards and 8 turn back to the reactor 4 12 1.00046 ± 0.00122
core, change tendency of Keff with reflector temperature and reactivity 8 12 0.99190 ± 0.00145
temperature coefficient ρ were simulated and are shown in Table 4.1-4. 12 12 0.98777 ± 0.00142

The density and dimensions of fuels, structure material, reflectors, and

Doppler Effect were considered. The result reveals that when the tem-
perature of reflector increases more than 100 K, the reactor would shut
down automatically.

12
Q. Zhou et al. Nuclear Engineering and Design 340 (2018) 9–16

4.2. Thermodynamic properties

The thermodynamic properties of ACMIR is similar to free piston

Stirling Engine (FPSE). The reactor core is fixed in expansion chamber,
and performs as the heater of FPSE. The reactor core thermal power is
exactly the heating power from heater end, which is equal to the me-
chanical power of linear generator, plus heat radiation power from
cooler end. The thermodynamic analysis results were used as a basis to
determine the structure parameters, such as size of piston, displacer,
and displacer rod, the stroke of piston, and displacer, phase angle of
piston and displacer, temperature of expansion chamber, and com-
pression chamber, and operating frequency (Rogdakis et al., 2004;
Boucher et al., 2007; Formosa and Despesse, 2010; Formosa, 2011;
Karabulut, 2011; Formosa and Frechette, 2013; Zare and Tavakolpour-
Saleh, 2016; Mou, 2016).
For FPSE thermodynamic analysis, expansion and compression
Fig. 4. -1. Reactivity monitoring efficiency of drum controllers’ rotation me-
chanism.
swept volumes (VswE and VswC) as well as their phase angle (α) are
important parameters. In classical Stirling engines’ theory, they are
deduced from mechanical kinematics. Here aforementioned structure
Table 4.1-3 parameters as in Table 4.1-1 were set as control/input parameters of the
Reactivity monitoring efficiency of drum controllers’ ejection mechanism. calculation. Nomenclature of used parameters are shown in Table 1-1.
Number of gadolinium Number of reflector Keff In the deduction, Ade, Adc, Ap, Sd, Sp, and αdp are all control para-
controller turning towards the drum in intact meters, then VswE, VswC, and α are deduced consulting the Schmidt
reactor core condition
analysis (Formosa, 2011; Karabulut, 2011).
0 0 1.00919 ± 0.00128 VswE = 2Ade Sd (1)
0 2 0.99994 ± 0.00139
0 4 0.98963 ± 0.00109
(Ap Spcos(αdp) + Adc Sd )2
0 6 0.98155 ± 0.00130 VswC = 2
0 8 0.97415 ± 0.00135 + ((Ap Spsin(αdp))2 (2)
0 10 0.96594 ± 0.00137
0 12 0.96259 ± 0.00149
Ap Spsin(αdp)
α = π + arctan ⎛⎜ ⎞
⎟
⎝ −A dc Sd + Ap Sp cos(αdp ) ⎠ (3)
The total swept volume denoted as Vsw is deduced from VswE, VswC,
and α. The swept volume ratio is κ = VswC/VswE as usually defined.
Thus, the total swept volume is:

Vsw = VswE 1 + κ 2 + 2κ cosα (4)

Substituting VswE, VswC, and α, we could establish:

(Ap Spcos(αdp) + Adc Sd )2

κ = Ade Sd/
+ ((Ap Spsin(αdp))2 (5)

1
Vsw = 2Ade Sd × 1 + κ 2−2κ
((Ap Spsin(αdp ))2
1+
(Ap Spcos(αdp) −Adc Sd )2 (6)
The Schmidt analysis allows expressing the indicated mechanical
power as a function of chamber temperatures and other structure
Fig. 4. -2. Reactivity monitoring efficiency of drum controllers’ ejection me-
parameters.
chanism. According to the analysis results, the instantaneous pressure is:

1−β 2
p = pmean
Table 4.1-4 1 + β cos(ωt −θ) (7)
Change tendency of Keff with reflector temperature and reactivity temperature
coefficient. where in:
Number of control drums turning towards the reactor Reflector condition:Intact 2τκ cos(α ) + τ 2 + κ 2
core:4 β=
2ν + κ + τ (8)
Temperature increase Keff Reactivity temperature
coefficient κ sin(α )
tan(θ) =
0K 1.00919 ± 0.00128 κ cos(α ) + τ (9)
50 K 1.00080 ± 0.00132 −16.6 pcm/°C
100 K 0.99849 ± 0.00128 −10.6 pcm/°C τ = TC / TE (10)
150 K 0.99833 ± 0.00136 −7.2 pcm/°C
−5.5 pcm/°C VCC VR TC V
200 K 0.99815 ± 0.00105 ν= + + HC τ
250 K 0.99783 ± 0.00141 −4.5 pcm/°C VswE VswE TR VswE (11)
300 K 0.99763 ± 0.00114 −3.8 pcm/°C
κ = VswC /VswE (12)

13
Q. Zhou et al. Nuclear Engineering and Design 340 (2018) 9–16

Table 4.2-2 safety performance of ACMIR in some accident conditions.

Control parameters of the heat exchange of ACMIR.
Symbol Nomenclature Value 4.3.1. Security design considerations
To consolidate inherent security of ACMIR, the authors considered a
Ade Expansion side of displacer area 0.07 m2 series of security issues when designing the reactor structure.
Adc Compression side of displacer area 0.068 m2
Ap Piston area 0.068 m2
Sd Displacer stroke 0.1 m
4.3.1.1. Gravity independent autonomous circulation design. As shown in
Sp Piston stroke 0.05 m Fig. 2-1, the reactor core and energy conversion system are integrated
αdp Displacer-piston phase angle 1.57 rad in RPV, and the coolant in RPV is also the working medium of
TE Expansion chamber temperature 1073 K conversion system and doing work to electric machinery.
TC Compression chamber temperature 473 K
By this integration design, the motive force for coolant circulation
f Operating frequency 30 Hz
VEE Expansion chamber minimum volume 0.004 m3 doesn't come from the coolant density difference and gravity, but from
VCC Compression chamber minimum volume 0.001 m3 temperature difference induced gas expansion at the hot end and
VR Regenerator volume 0.0005 m3 compression at the cold end of the reactor. This character brings the
TR Regenerator temperature 798 K
reactor several advantages. Firstly, autonomous circulation in-
M Mass of working fluid 44 mol
dependent on gravity enables the reactor to operate in non-gravity
environments. Secondly, full power autonomous circulation can be
And pmean can be given by the following expressions: achieved in no need of a relatively large coolant channel height, which
provides possibility for relatively higher power density. Thirdly, the
Mr 1 reactor is not necessary to be vertical when operating, that the reactor is
pmean =
s 1−β 2 (13) able to operate in tilt or swing conditions.

1 T 1
=2 E τ 4.3.1.2. Highly integrated modular structure design. Highly integrated
s VswE 2ν + κ + τ (14) modular structures were designed in ACMIR to enhance the security.
In which M is the total mass of working fluid and r is the ideal gas The reactor core, energy transfer system, the linear motor, and coolant
constant. circulation medium are all arranged in the RPV; electric power is output
The indicated mechanical power is expressed as: through nuclear grade electrical penetration on RPV wall. Thus, no
coolant pipe is connected to RPV, and there would be no loss of coolant
ω VswE κ (τ −1)sin(α )
Pi = pmean (1− 1−β 2 ) accident (LOCA). Design of less pipes hired or penetrating the RPV also
2π 2β τ 2 + κ 2 + 2κτ cos(α ) (15) earns the reactor more chance to survive the impact during launching
According to the above analysis and reactor core properties, the or any other extreme conditions.
dimensions and moving parameters of the heat exchange part of ACMIR In the other hand, all reactivity control systems are assigned outside
are confirmed and shown in Table 4.2-2. RPV. Therefore, no mechanical structure breakthroughs the RPV, which
Important parameters for thermodynamic analysis such as expan- enhances the security of pressure boundary.
sion and compression swept volumes as well as their phase angle,
average pressure, instantaneous pressure, and indicated mechanical 4.3.1.3. Double passive reactivity control system design. As mentioned
power are deduced from aforementioned equations and control para- before, two separate reactivity control systems are used in the reactor,
meters. The deduced parameters of the heat exchange part of ACMIR and both of them are passive safe.
are shown in Table 4.2-3. The derived mechanical power is 203 kW, and One reactivity control system is rotation mechanism of 12 control
the maximum instantaneous pressure of the coolant gas is 20 MPa. drums, which are driven by 12 separate electromotor. When the elec-
As the thermal power of the reactor is 500 kW, the heat rejection tromotor losses electricity, the rotary control drums would turn to face
power should be 300 kW. The temperature of hot end and cold end of the reactor core, and shut down the reactor. The core reactivity with
heat pipes is set 443 K and 433 K, respectively. And the panel radiator different number of control drums turning to face the reactor core is
of an area about 150 square meters is enough to remove 300 kW waster shown in Table 4.1-2 and Fig. 4-1, and approved the effectiveness of
heat with these parameters. this control system.
The other reactivity control system is the rejection mechanism of 12
drums/reflectors. Each reflector is pressed by a top set gas chamber
4.3. Safety discussion
against a bottom set compressed spring to keep in position. If the re-
actor is going overpressure, a gas relief process is triggered and the
Controlling reactivity, preventing radioactive materials from es-
springs loss their rival forces and push the reflectors to move upwards
caping and removing decay heat are basic safety functions of the re-
and go out of core active region, then the reactor would be shut down.
actor. With full power autonomous circulation system independent of
If the reactor is overtemperature, a pin set will release the reflector
gravity, ACMIR has passive safety performance in all kinds of en-
springs, and partly push reflectors out of working region to lower down
vironments such as space, Moon, Mars surface, deep-sea, and tilt and
the reactivity. The core reactivity change with number of reflectors
swing situation. This section discusses main security design and the
moving out of core active region is shown in Table 4.1-3 and Fig. 4-2,
which suggests even a few reflectors move upwards, the reactor would
Table 4.2-3
Deduced parameters of the heat exchange of ACMIR.
be shut down.
One reactivity control system is independent from the other. The
Symbol Nomenclature Value Keff calculation results suggest that either reactivity control system is
VswE Expansion chamber swept volume 0.0140 m3 capable to shutdown the reactor all alone, no matter what the status of
VswC Compression chamber swept volume 0.0152 m3 the other one is.
α Swept volume phase angle 2.68 rad
Vsw Swept volume 0.0068 m3 4.3.1.4. Passive residual heat removal design. ACMIR has a totally
pmean Average pressure of working fluid 13.5 MPa
passive residual heat removal ability. Its heat rejection heat pipes or
p Instantaneous pressure of working fluid 19.9 MPa
Pi Deduced mechanical power 203 kW energy transfer system is capable to remove full residual heat power all
along.

14
Q. Zhou et al. Nuclear Engineering and Design 340 (2018) 9–16

Table 4.3-1 reactor has a negative temperature coefficient, as shown in Table 4.1-4.
Parameters when removing residual heat. When the temperature of RPV wall and reflectors increases more than
Symbol Nomenclature Value 100 K, the reactor would be shutdown automatically.

Ade Expansion side of displacer area 0.07 m2 4.3.2. Accident conditions discuss
Adc Compression side of displacer area 0.068 m2
Based on above series of security design, safety performance of
Ap Piston area 0.068 m2
Sd Displacer stroke 0.1 m
ACMIR in some accident conditions are discussed in this section.
Sp Piston stroke 0.05 m
αdp Displacer-piston phase angle 1.57 rad 4.3.2.1. Loss of coolant accident (LOCA). As mentioned before, the
TE Expansion chamber temperature 473 K reactor has highly integrated modular structures. All systems used to
TC Compression chamber temperature 423 K
produce energy and transfer into electric power are arranged inside the
f Operating frequency 30 Hz
VEE Expansion chamber minimum volume 0.004 m3 RPV, while all systems used to control reactivity and heat rejection are
VCC Compression chamber minimum volume 0.001 m3 assigned outside RPV. No coolant pipe except for safety relief pipe is
VR Regenerator volume 0.0005 m3 connected to RPV, no mechanical structure breaks through RPV. So
TR Regenerator temperature 448 K
there would be no Loss Of Coolant Accident (LOCA).
M Mass of working fluid 44 mol
VswE Expansion chamber swept volume 0.0140 m3
VswC Compression chamber swept volume 0.0152 m3 4.3.2.2. Loss of external power supply. When the reactor loses external
α Swept volume phase angle 2.68 rad power supply, rotary control drums would lose their reactivity
Vsw Swept volume 0.0068 m3 regulating function. But Gadolinium side of all rotary control drums
pmean Average pressure of working fluid 8.3 MPa
would passively turn to face reactor core side, and shut down the
p Instantaneous pressure of working fluid 10.0 MPa
Pi Deduced mechanical power 16.7 kW reactor.
At the same time, the displacer and piston keep moving with a
smaller amplitude, and the gas coolant continues autonomously circu-
When the reactor shuts down, the residual heat power is about 3% lating. Thus, the linear motor keeps outputting electric power and heat
of the reactor full thermal power (about 15 kW). Through this power pipes continues rejecting waster heat. The coolant and heat rejection
reducing process, the temperature and pressure of the reactor de- systems automatically remove the residual heat and cold down reactor
creases. As long as the temperature of reactor core is still beyond 473 K, in no need of any external intervention, until external power recovers.
the displacer and piston would still be working but with a smaller Moreover, the linear motor generates electric power once the re-
amplitude. At that time, the thermodynamic of reactor would be under actor structures maintain intact and temperature difference exists,
another set of parameters, as listed in Table 4.3-1. which encourages realizing a self-supply circuit to support rotary con-
As the key control parameters, the temperature of hot end and cold trol drums regulation function, and to make the system does not rely on
end of heat pipes is set to 423 K and 413 K, respectively. Under this any external power supply.
setting, the deduced mechanical power of Sterling is 16.7 kW, and the
heat rejection power of heat pipes along with the 150 square meters 4.3.2.3. Reactor overtemperature / overpressure. As mentioned in
panel radiator (not shown in the Table) is 20.9 kW. We could see that Sections 4.1 and 4.3.1.3, if the reactor is overpressure, gas release is
the linear motor and heat rejection system each is able to remove more triggered, the springs would push reflectors upwards out of core active
than 15 kW waster heat. No intervention is required, once the reactor region, the reactor would be shut down. If the reactor is
shuts down, an adequate heat removal mode physically starts. overtemperature, a pin set will release the reflector springs, and
As discussed in the next section, the heat removal power would be partly push reflectors upwards out of core region to inversely
regulated automatically with temperature, and ensure all residual heat feedback the reactivity.
removed passively. Therefore, if the reactivity regulation fails to control the reactor
operating within the design temperature and pressure, to let the reactor
go either overtemperature or overpressure, the rector would shut down
4.3.1.5. Automatic output power regulation design. Working frequency of
passively with no need of any external interventions.
the piston is nearly steady as the elastic coefficient of piston plate spring
and mass of piston are constants. The Linear motor output power is
4.3.2.4. Impact when rocket launching or landing. Gravity independent
determined by piston amplitude, which adjusts with temperature
autonomous circulation character makes ACMIR good choice as space
difference between expansion chamber and compression chamber,
reactors.
and thus reactor core power automatically.
During rocket launching or landing, the reactor would experience
At 100% to 20% thermal power, the operational parameters such as
severe impact condition, thus space use prefers simple structure.
the piston amplitude, expansion chamber temperature, compression
ACMIR is highly integrated and modularized, there is no multiple
chamber temperature, and output electric power are calculated via the
coolant circuit or machinery with complex structures such as routine
equations in Section 4.2. The analysis results are shown in Table 4.3-2,
control rod drive mechanism. Thus, the reactor is excellent in impact
and revealing the automatic output power regulation by the core
resistance and reliability.
thermal power.
Shock resistance analysis for the mechanical structures of the re-
actor is conventional. The results indicate that maximum stress in im-
4.3.1.6. Reactivity temperature feedback. As analyzed in Section 4.1, the pact condition and other accident conditions are all less than allowable
stress.
Table 4.3-2
Automatic output power regulation under different thermal power. 4.3.2.5. Displacer or piston stagnation. Piston cylinder is a mature and
Thermal power 100% 80% 60% 40% 20% reliable technology, thus piston (or displacer) stagnation is an event of
very low probability. If piston (or displacer) clamping stagnates, the
Expansion chamber temperature/K 1073 943 853 733 613 reactor would be shut down intentionally, and displacer (or piston)
Compression chamber temperature/K 473 463 453 443 433
would alone driving gas coolant flow. This coolant flow forms not a
Piston amplitude/mm 50 47 38 26 8
Mechanical power/kW 204 156 107 54 11 holonomic circulation, but is capable to carry the residual heat from
reactor core to heat pipes.

15
Q. Zhou et al.
Piston and displacer clamping stagnation at the same time is an
extremely low probability event. Even in this accident, all the residual
heat could be rejected from reactor core to heat pipes via heat con-
duction between gas coolant, reactor internal components, and the RPV
wall.
Therefore, the displacer or piston stagnation, in single or in pairs
would lead to shutdown of the reactor, but wouldn’t cause any further
nuclear safety accidents.
5. Conclusion
A new miniature reactor with integrated reactor core and energy
transfer system is designed. This integrated modular structure with
φ1000 mm × 1600 mm dimensions and 200 kWa electrical power is
named ACMIR (the Autonomous Circulation Miniature Integrated
Reactor), and is different from traditional model of nuclear power
plants and space nuclear power systems. In ACMIR, reactor core and
energy transfer system are integrated in reactor pressure vessel to
achieve high power density, and autonomous circulation capability.
The circulation is not driven by coolant density difference and gravity,
but by structures movements forced via temperature difference between
hot end and cold end.
Based on a series of analysis, this reactor has following advantages.
1) Gravity independent autonomous circulation capability. It makes
the reactor a good choice for space flight propulsion, the Moon and
Mars base power supply, the deep-sea or other tilt and swing si-
tuation applications.
2) Excellent inherent safety characteristic. It is capable to shut down
and remove residual heat passively in all kinds of analyzed accident
conditions with no need of external interventions.
3) High integration and structure simplicity. It makes the reactor
sturdier and more durable. It takes the modular structure and di-
rectly outputs the electric power, therefore is suitable for mass
production and assembly, and adaptable for different power re-
quirements.
4) Novel structure. Structure of this reactor can be extended to con-
struct thermal reactors, fast reactors, even other persistent thermal
sources as radioisotope thermal source. Mechanical power could
also be substantially adjusted with dimensions of the energy transfer
system.
The physics and thermodynamic properties of ACMIR are briefly
presented, and the safety performances are discussed in this article.
According to these analysis, inherent safety characteristics are ob-
tained, and the reactor is capable to shut down and remove residual
heat passively without any external interventions. Detailed physics,
thermodynamic, mechanical properties (including impact resistant
properties), and accident safety analysis will be presented elsewhere.
There are some challenges of ACMIR to be solved in the future work,
such as shutdown margin and control rods efficiency is not abundant,
the scale of energy transfer system is close to the limits of Stirling en-
gines, and the reactor core material faces rapidly changing temperature.
Still preliminarily, this article suggests this new structure of ACMIR has
a decent acceptable work condition and safety and shows good space
application potential.
References
Abram, T., Ion, S., 2008. Generation-IV nuclear power: a review of the state of the
16
Nuclear Engineering and Design 340 (2018) 9–16
science. Energy Policy 36 (12), 4323–4330.
Rowinski, M.K., White, T.J., Zhao, J.Y., 2015. Small and medium sized reactors (SMR): a
review of technology. Renew. Sustain. Energy Rev. 44, 643–656.
Oman, H., 2003. Deep space travel energy sources. IEEE Aerospace Electron. Syst. Mag.
18 (2), 28–35.
Summerer, L. and K. Stephenson, 2011, Nuclear power sources: a key enabling technology
for planetary exploration. Proceedings of the Institution of Mechanical Engineers Part
G-Journal of Aerospace Engineering, 2011. 225(G2): pp. 129-143.
Woo, T.H., Lee, S.H., 2014. Characteristics of nuclear spacecraft in nano-gravity for the
deep space explorer. J. Aerospace Eng. 27 (1), 1–8.
Nuclear Reactors to Power Space Exploration. R&D Magazine, 2017. 59(1): p. 10-12.
Bennett, G.L., 2008. Radar men on the moon: a brief survey of fission surface power
studies. Space Technol. Appl. Int. Forum Staif 2008 (969), 372–379.
Cassady, R.J., et al., 2008. Recent advances in nuclear powered electric propulsion for
space exploration. Energy Conv. Manage. 49 (3), 412–435.
Poston, D.I., 2001. The heatpipe-operated mars exploration reactor (HOMER). Space
Technol. Appl. Int. Forum-2001 552, 797–804.
Powell, J., Maise, G., and Paniagua, J., 2003, SUSEE - Ultra light nuclear space power
using the steam cycle. In Ieee Aerospace Conference Proceedings, Vols 1-8, 2003, p.
491-507.
El-Genk, M.S., Tournier, J.M.P., 2004. “SAIRS” – Scalable amtec integrated reactor space
power system. Progr. Nucl. Energy 45 (1), 25–69.
Na, M.G., and Upadhyaya, B.R., A model predictive controller based on support vector
regression and genetic optimization for an SP-100 space nuclear reactor. In Icinco
2006: Proceedings of the Third International Conference on Informatics in Control,
Automation and Robotics, 2006: p. 136.
Na, M.G., et al., 2006. Design of a model predictive power controller for an SP-100 space
reactor. Nucl. Sci. Eng. 154 (3), 353–366.
Na, M.G., Upadhyaya, B.R., 2006. Application of model predictive control strategy based
on fuzzy identification to an SP-100 space reactor. Ann. Nucl. Energy 33 (17–18),
1467–1478.
Na, M.G., Upadhyaya, B.R., 2006. Model predictive control of an SP-100 space reactor
using support vector regression and genetic optimization. IEEE Trans. Nucl. Sci. 53
(4), 2318–2327.
King, J.C., El-Genk, M.S., 2006. Submersion-subcritical safe space (S-4)reactor. Nucl. Eng.
Des. 236 (17), 1759–1777.
Maise, G., Powell, J., Paniagua, J. SUSEE: a compact, lightweight space nuclear power
system using present water reactor technology. In Space Technology and Applications
International Forum – Staif 2006, 2006. 813: p. 308.
Na, M.G., Upadhyaya, B.R., 2007. Development of a reconfigurable control for an SP-100
space reactor. Nucl. Eng. Technol. 39 (1), 63–74.
King, J.C., El-Genk, M.S., Thermal-hydraulic analyses of the Submersion-Subcritical Safe
Space (S boolean AND 4) reactor. In Space Technology and Applications International
Forum – Staif 2007, 2007. 880: p. 261.
King, J.C., El-Genk, M.S., 2007. Temperature and burnup reactivities and operational
lifetime for the submersion-subcritical, safe space (S boolean AND 4) reactor. Nucl.
Eng. Design 237 (5), 552–564.
Schriener, T.M. El-Genk, M.S., Comparison of reactivity control systems for the submer-
sion subcritical safe space (S boolean AND 4) reactor. In Space Technology and
Applications International Forum Staif 2008, 2008. 969: p. 348–361.
Poston, D.I., et al., 2008. Reference reactor module for the affordable fission surface
power system. Space Technol. Appl. Int. Forum Staif 969, 277–284.
King, J.C., El-Genk, M.S., 2009. Thermal-hydraulic and neutronic analyses of the sub-
mersion-subcritical, safe space (S-4)reactor. Nucl. Eng. Des. 239 (12), 2809–2819.
Craft, A.E., King, J.C., Radiation shielding options for the affordable fission surface power
system. In Space, Propulsion & Energy Sciences International Forum Spesif-2009,
2009. 1103: pp. 466–477.
Rogdakis, E.D., Bormpilas, N.A., Koniakos, I.K., 2004. A thermodynamic study for the
optimization of stable operation of free piston Stirling engines. Energy Conv. Manage.
45 (4), 575–593.
Boucher, J., Lanzetta, F., Nika, P., 2007. Optimization of a dual free piston Stirling en-
gine. Appl. Ther. Eng. 27 (4), 802–811.
Formosa, F., Despesse, G., 2010. Analytical model for stirling cycle machine design.
Energy Conv. Manage. 51 (10), 1855–1863.
Formosa, F., 2011. Coupled thermodynamic-dynamic semi-analytical model of free piston
stirling engines. Energy Conv. Manage. 52 (5), 2098–2109.
Karabulut, H., 2011. Dynamic analysis of a free piston Stirling engine working with closed
and open thermodynamic cycles. Renew. Energy 36 (6), 1704–1709.
Formosa, F., Frechette, L.G., 2013. Scaling laws for free piston Stirling engine design:
Benefits and challenges of miniaturization. Energy 57, 796–808.
Zare, S., Tavakolpour-Saleh, A.R., 2016. Frequency-based design of a free piston stirling
engine using genetic algorithm. Energy 109, 466–480.
Mou, J., et al., 2016. Gas action effect of free piston Stirling engine. Energy Conv.
Manage. 110, 278–286.