Thermal analysis of multifunctional structural battery for satellite applications

Thermal analysis of multifunctional structural battery for satellite applications

Accepted Manuscript Thermal analysis of multifunctional structural battery for satellite applications Yang Wang, Chaoyi Peng, Weihua Zhang PII: S1359...

1MB Sizes 1 Downloads 41 Views

Accepted Manuscript Thermal analysis of multifunctional structural battery for satellite applications Yang Wang, Chaoyi Peng, Weihua Zhang PII:

S1359-4311(14)01183-1

DOI:

10.1016/j.applthermaleng.2014.12.054

Reference:

ATE 6248

To appear in:

Applied Thermal Engineering

Received Date: 3 September 2014 Revised Date:

26 November 2014

Accepted Date: 22 December 2014

Please cite this article as: Y. Wang, C. Peng, W. Zhang, Thermal analysis of multifunctional structural battery for satellite applications, Applied Thermal Engineering (2015), doi: 10.1016/ j.applthermaleng.2014.12.054. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

RI PT

Thermal analysis of multifunctional structural battery for satellite applications

Yang Wang*, Chaoyi Peng, Weihua Zhang

SC

College of Aerospace Science and Engineering, National University of Defense Technology,

M AN U

Changsha 410073, Hunan Province, PR China

*Corresponding author

AC C

EP

TE D

Email: [email protected]

1

ACCEPTED MANUSCRIPT

Abstract

RI PT

Multifunctional structures combine multiple functionalities in a single structural part in order to increase performance while limiting mass and volume. Conventional satellite batteries are designed separately as a stand-alone subsystem then added to satellite host structure in the last assembly, bringing assistant

SC

containers and connectors etc. that often cause undesirable mass loading effects and consume valuable

M AN U

space. Power subsystem and then the whole satellite can benefit from the introduction of multi-functionality as a means of improving overall system efficiency. This paper presents the investigation of a new multifunctional structural battery (MFSB) consisting of energy storage, energy supply, and load bearing ability in a single composite structural panel for satellite applications. Thermal

TE D

vacuum testing (TVT) was carried out and transient thermal analysis was performed, and the results demonstrated that the current material formula and workmanship are viable in the space. When mounted

EP

within satellites as an equipment-panel, the designed MFSB panel can keep itself retaining the required

AC C

operational temperature without supplementary thermal control approaches.

Keywords

Multifunctional structural battery (MFSB), satellite, thermal vacuum testing (TVT), transient thermal analysis

2

ACCEPTED MANUSCRIPT

1.

Introduction

RI PT

Size and weight play an important constraint in space industrial. However, traditional satellites are designed and fabricated subsystem by subsystem and then assembled finally, introducing separate

packages and additional connectors etc. that are structural redundant from the whole system point of view

SC

and therefore taking up considerable mass and volume that would otherwise be assigned to payloads, thus

M AN U

new technologies and methodologies are explored to eliminate this redundancy and one of them is called Multifunctional Structure (MFS) mainly developed in the US at the end of the 1990’s [1]. Multifunction is an innovative concept that involves the development of new materials and devices and is capable of performing multiple functions within a single structure or system. MFS incorporates functionalities that

TE D

exist independently in the past such as sensing, actuation, energy storage and the like to the primary structure of a satellite [2-3]. Multifunctional Structural Battery (MFSB), as a kind of MFS, is the

EP

integration of power storage and load bearing functionalities into one structure. Embedded into the satellite host structure, the once stand-alone batteries’ purely structural packaging components and other

AC C

parasitic elements like connectors become unnecessary and thus can be eliminated. Meanwhile, as the embedded batteries are capable of carrying structural loads, then part of the primary structure may also be removed. Therefore not only the space within the satellite is saved, but also, the total mass and volume of the satellite can probably be reduced. During the past decades, many research efforts about MFSB have been undertaken, involving

3

ACCEPTED MANUSCRIPT

conception, design, and fabrication as well as performance characterization. Neudecker et al. [4, 5] from

RI PT

ITN Energy Systems (a United States company) introduced LiBaCore concept in which thin-film solid electrolyte lithium batteries were deposited onto the available surface inside honeycomb core and then this core was fabricated into a composite panel which thus doubled as a power storage devices while

SC

adding a small amount of mass and virtually no volume. Besides providing specific power of greater than

M AN U

200 Wh kg-1, the battery core in the tests showed its capability to withstand the typical thermal and mechanical rigors experienced by a structural panel. However, significant space qualification and validation testing was still required. In sequent studies [6, 7], thin film lithium-ion battery components were coated onto single carbon fiber filament to form a novel, multifunctional thin-film structural battery,

TE D

namely Power-fiber, which can be embedded into a composite system. Tests showed that a 10 cm×10 cm patch consisting of 1000 coated fibers could deliver 9 W at 3V and 3A while supplying 0.1 Wh of energy.

EP

Lyman et al. [8] from Boundless (another US corporation) developed structural bi-cell in which, electrode, either anode or cathode, was shared between unit-cells. Then the bi-cell was strengthened and corrugated

AC C

fabricated to work as the sandwich core. Using this type bi-cell, Aglietti et al. [9] proposed ten kinds of candidate MFSBs (they call it multifunctional power structures (MFPS)) for satellite applications. After parameter optimization [10] and experimental determination [11] of the dynamic behaviors, a favorite multifunctional prototype design that had highest first resonance frequency while lowest material density and thus largest energy density was identified. As custom-built cells are always expensive, especially for

4

ACCEPTED MANUSCRIPT

small production runs, commercial off-the-shelf polymer lithium ion (PLI) cells were took into utilization

RI PT

[12-15], where brick-shape PLI cells were embedded into the honeycomb core of a sandwich panel, and issues such as feasibility assessment, design, manufacture, experiment and numerical simulation were

addressed. What’s more, to make the best of multiple-functionality, multifunctional solar array (MFSA)

SC

was further developed and explored, and attention was mainly paid to strategies for thermal control

M AN U

problems [16, 17]. However, expansions of the embedded commercial PLI cells resulting from pressure difference between the inner electrolytes and outer vacuum orbit environment, which would therefore result in cell disable, were not paid so much attention to. Liu et al. [18], adopting a quite different approach, developed a new structural battery that had tunable mechanical performance by redesign and

TE D

strengthening each component of the battery. Aside from spacecraft, in many other structural fields, such as aircraft [19], unmanned underwater vehicles [20, 22], ground vehicles [23] and even domestic

EP

buildings [24], integration of power storage and load bearing functionalities was proposed and developed, intending to enhance the efficiency of related components in terms of performance to mass, volume and

AC C

cost ratio.

Since heat generated and released from the embedded batteries in the MFSB during charge/discharge operation should be dissipated in case of overheating which could lead to possible dysfunction or destruction, this paper mainly studied thermal behaviors of the MFSB. Transient thermal analysis model was set up with the help of thermal vacuum testing (TVT), and then effects posed by surface emissivities,

5

ACCEPTED MANUSCRIPT

discharge-rates, initial and ambient temperatures on thermal behaviors were investigated. It is found that

RI PT

the current material constituents and process methods are viable for space environment. And if served as the satellite internal equipment-panel, the designed MFSB panel can maintain itself within the required operational temperature ranges without additional thermal control means. These simulation results

SC

demonstrate the feasibility of the designed MFSB as a potential candidate for load bearing and power

2.

Prototype design of MFSB

2.1 Configuration

M AN U

supplying in satellite applications and therefore for other spacecraft applications.

The embedded cells selected in this work are of the gel polymer lithium-ion (GPLI) type that has high

TE D

energy density and high working voltage, and were designed and manufactured in National Univ. of Defense Technology (NUDT). Unlike the widely used space cells such as Cd-Ni battery and H2-Ni battery

EP

etc. which have liquid electrolytes and are deposited in cylindrical metal casings and therefore are rather heavy, the GPLI cells have quasi-solid electrolytes and hence good load bearing capability, use aluminum

AC C

laminate film as initial package material and therefore may be configured with great flexibility. Moreover, their feasibility of being applied in space missions was assessed in JAXA [25-26], involving behaviors in space environment, performance-degradation mechanisms and effects of operations on the cycle-life, and the promising results implied that the GPLI cells could power well for satellites applications. Since carbon-fiber-reinforced composite structures are prevalently used in aerospace industry due to

6

ACCEPTED MANUSCRIPT

carbon/epoxy fabric was chosen as the MFSB face sheets.

RI PT

their advantages of having high specific strength and stiffness and strong design flexibility, laminated

Polymethacrylimid (PMI), a kind of rigid and closed-cell foam of 75 kg m-3 density (from Evonik Degussa, Germany), having good workability and space adaptability [27], which can simplify the

SC

sandwich composite structure fabrication considerably and foremost offer enough structural resistance

M AN U

against battery expansion caused by pressure difference between the inside and outside of the GPLI battery, was chosen as the main part of the MFSB core. 2.2 Configuration

The proposed multifunctional structure-battery configuration is shown in Fig. 1 (a). The GPLI cell is

TE D

embedded in a sandwich structure and used as a part of the sandwich core. Besides, cables that connect cells in series or parallel ways as well as power control module (PCM) and other equipment that belong to

EP

satellite power subsystem are also inserted between the sandwich structure’s face sheets. From a structural point of view, the main part of the MFSB core can be honeycomb, foam, or truss,

AC C

determined by highest strength-to-density or stiffness-to-density ratio. As for the embedded lithium ion cells: shape can be manufactured as prismatic or strip according to the inner space of the sandwich structure that could be made use of (shown in Fig. 1 (b)), location can be distributed in a regular array or just near the power consumer to provide localized and autonomous power in remote locations, thereby eliminating heavy power transmission cables to and from centralized power sources to save space and

7

ACCEPTED MANUSCRIPT

weight (shown in Fig. 1 (c)), configuration can be designed as flat panel style or hetero-type conforming

RI PT

to the shape of satellite host structure, thus promising highest integration between structures and batteries (shown in Fig. 1 (d)). All the above huge flexibilities and various choices make this MFSB concept rather appealing and advantageous.

SC

Fig. 1. Schematic of multifunctional structural battery.

M AN U

The prototype MFSB was designed and fabricated as a sandwich beam shape in order that it’s also convenient to study the mechanical and electrical properties, as was done in Ref. [28]. In the MFSB, a GPLI cell was framed in the geometric center within PMI foam channels and then bonded between a flat laminate on one side, and a conformal laminate on the other side. Besides, aero cables were selected for

TE D

power bussing. The dimensions of the designed MFSB were 300×80×20 mm (length× width× thickness), see Fig. 2 (a). Both upper and lower face sheet thickness was 2 mm, consisting ten layers of carbon fiber

EP

plain weave, each of 0.2 mm thickness, and the foam thickness was 16 mm. The selected GPLI cell was designed and fabricated in NUDT, with dimensions of 105×67×6.6 mm and capacity of 6.5 Ah.

AC C

Fig. 2. (a) Planform and section view of MFSB (thermal couples included and cables hided); (b) Layered-construction of the GPLI cell in the MFSB

3.

Experiment and simulation

3.1 Thermal vacuum testing To employ the GPLI batteries in space applications in such a multifunctional manner, particular

8

ACCEPTED MANUSCRIPT

operational conditions and environments, such as vacuum, out-gassing from the polymer components,

spacecraft orbit etc. should be taken into consideration.

RI PT

safety concerns, long cycle-life requirements, and charge/discharge intervals limited strictly by the

Thermal vacuum testing is the most comprehensive risk mitigation component of the integrated

SC

spacecraft environmental test program, verifying satellite/component performance in a simulated space

M AN U

environment with temperature extremes beyond which the satellite/component is expected to experience on orbit. And generally the fundamental purpose of thermal vacuum testing is to understand the satellite and component’s performances through environmental extremes and thus increase mission assurance through a test-like-you-fly environmental testing program.

TE D

As is the integration of mechanical and electrical subsystems, the MFSB deserves aborative and strict tests to demonstrate that the integrated satellite batteries and walls can withstand the anticipated thermal

EP

environment in the vacuum of space. And also the tests should qualify the MFSB’s functional performances in the simulated space environment and obtain thermal data to confirm whether the thermal

AC C

design and workmanship are reasonable and if there is a need that thermal control approaches be applied. Thermal data is also required for establishing and correlating with the MFSB thermal analysis model, which is important for thermal management design and also useful when the conception are modified and optimized. Meanwhile, concerning the possibility that power failure emerges resulting from GPLI battery expansion caused by inner and outer pressure difference because of removal of the metal packaging and

9

ACCEPTED MANUSCRIPT

insufficient mechanical support offered by the ambient PMI foams, electric motioning of MFSB is pretty

RI PT

necessary during the thermal vacuum testing and can be equivalently checked by cycling of power charge and discharge.

Thermal vacuum testing was carried out for MFSB in Space Center Lab in NUDT. Before the

SC

implementation of the TVT, the MFSB was electrically checked out under a regimen of constant-current,

M AN U

constant-voltage (CC-CV) charge mode with a taper voltage of 4.2±0.2V and charge rate of 0.2C. When decreased to less than 0.01C, charging was ceased and the MFSB was laid aside for half an hour. Then constant-current (CC) discharge mode was applied with a discharge rate of 0.2C and a cutoff voltage of 2.75±0.2V. The charge current was calculated here by using the nominal cell capacity. Next, the MFSB

TE D

was charged again until full capacity was reached and as thermal vacuum test was going on, discharge was performed at 0.5C for 30 min when low-temperature stage came. While in the high-temperature

EP

phase of each round, charging was applied at 0.2C until 4.2V. After the thermal vacuum test, capacity verification was carried out again under the same regime as was done before. Note that thermal vacuum

AC C

test must start with the MFSB power off while the vacuum chamber was pulled down to a hard vacuum, and again the power must be off when the air was being let back in. And the MFSB was prepared with thermocouples that were located in the geometry center of MFSB’s upper and lower surfaces to measure the temperature evolution on the surface during the test run and then to compare with the electro-thermal model prediction. According to space standard [29], pressure was set to 1.3×10-3 Pa, the highest and

10

ACCEPTED MANUSCRIPT

lowest temperatures were set to 313±3k and 263±3K, respectively, and a total number of 3 cycles of

RI PT

experiments were carried out. 3.2 Numerical simulation

In this section, finite element analysis model of the thermal behavior of MFSB was firstly set up with

SC

‘help’ of the thermal vacuum testing data, then performance of the MFSB in typical satellite orbits was

M AN U

simulated using the established model with necessary modification. In order to obtain a precise simulation of the thermal behavior of the GPLI cell, generally, the geometry, configuration, physical, chemical and electrochemical properties should be delineated as accurately as possible in the model. However, an unacceptable amount of calculation time would be encountered if the model was too complicated.

TE D

Therefore, it’s rational to take the layered-structure of the embedded GPLI cells as orthotropic homogeneous materials, which had been proved to be practical in Ref. [30].

EP

The GPLI cell embedded in this MFSB is comprised of pouch case, electrode plates, electrolytes and separators, formulating a layered-construction (shown in Fig. 2 (b)). In the GPLI cell, the quasi-solid gel

AC C

electrolytes present limited and neglectable mobility, so the contribution of convection to heat transfer inside the battery could be disused. And since the GPLI battery is an opaque system, the inside radiative heat transfer is naturally insignificant and thus the conductive heat transfer is the main mechanism within the GPLI battery, so it is in the MFSB. The transient three dimensional conductive heat transfer equation is as follows:

11

ACCEPTED MANUSCRIPT

ρc

∂T ∂ ∂T ∂ ∂T ∂ ∂T & = (k x ) + (k y ) + (k z )+Φ ∂t ∂x ∂x ∂y ∂y ∂z ∂z

(1)

RI PT

& are the density, heat capacity, thermal conductivity and heat generation rate where ρ , c , k and Φ per unit volume, respectively. For the layered GPLI cell, k x equates to k y , so do the laminated CFRP

& exists. face sheets. And only in the GPLI cell, Φ

SC

The product value of heat capacity is calculated using the volume and heat capacity of each component

M AN U

as follows:

ρ cbat =

∑ ρcv ∑v

i i i

i

(2)

i i

where ci and vi are heat capacity and volume of a specific component, respectively, and cbat

TE D

represents the average or equivalent heat capacity of the battery.

The average density of the GPLI battery can be calculated using equation:

ρ=

mtot vtot

(3)

EP

where mtot , vtot denote the total mass and total volume of the embedded battery, respectively.

AC C

Referring to the equivalent heat conductivity of the layer-structured battery, the following models can be used, respectively:

kz =

L pos k pos

kx = k y =

+

Ltot Lneg kneg

+

Lsep k sep

k pos ⋅ Apos + kneg ⋅ Aneg + ksep ⋅ Asep Atot

12

(4)

(5)

ACCEPTED MANUSCRIPT

Eq. (4) is for elements connected in series while Eq. (5) for parallel connection. Heat conductivity in

RI PT

different directions include: x direction, k x , y direction, k y , and z direction, kz . Note that the coordinate system adopted is shown in Fig. 2 (b). L denotes the thickness of each specific component while A is the cross section area of each component that perpendicular to the heat transfer direction, and

SC

footnotes pos, neg, sep, represent the anode, cathode and separator layer in the battery, respectively. And

M AN U

Ltot is the combination of thickness of each component, while Atot is the summation of cross section area of each component.

Using the above equations and cell information provided by the manufacture, physical properties of the MFSB was calculated and presented in Table 1.

TE D

Table 1 Physical properties of MFSB used for transient thermal analysis. The local heat generation rate during charge/discharge cycle in the GPLI cell is estimated based on the

EP

various losses in the cell [30]. And heat generation equation developed by Bernardi et al. [31] and studied

AC C

in Ref. [30] is used in this study:

& = I [( E − U ) − T dE0 ] Φ 0 V dT

(6)

where I , V , E0 and U denote the total current of the battery, the total volume of the GPLI cell, the open-circuit potential and the working voltage, respectively. dE0 / dT expresses the temperature dependence of the equilibrium voltage and was estimated using experimental data measured in the thermal vacuum tests in this paper.

13

ACCEPTED MANUSCRIPT

In the simulation, the designed MFSB was located inside the satellite as shown in Fig. 3 (a). At the

RI PT

boundary, only radiative heat transfer need to be taken into account on the assumptions that no equipment are fixed on the MFSB panel and the four edges are thermal-insulated so as to study thermal behavior of

(b)) and is expressed as follows:

(7)

M AN U

Q = εσ (Tm4 − Ta4 )

SC

MFSB itself and the heat is transferred through radiation on the upper and lower surfaces (shown in Fig. 3

where ε and σ denote the emissivity and the Stefan-Boltzmann constant, respectively. Tm is the temperature of the MFSB, while Ta is the ambient temperature.

Fig. 3. Location of the MFSB panel and corresponding thermal dissipation paths.

TE D

It is worth noting that in different orbits the MFSB works in different charge-discharge modes. When used in GEO vehicles, the MFSB will be discharged in a maximum of 72 min to generate enough power

EP

to meet electrical demands of the bus and mission over a 45-day period on each side of the equinox of a year. At all other times, the GEO satellite is free from the Earth’s shadow, and the solar cells will charge

AC C

the MFSB at an optimum state of charge (SOC). However, when works in LEO crafts, the MFSB will periodically experience about 60 min sunshine and 30 min eclipse, which means that the MFSB will be charged at a short interval of 60 min and then discharged at a very short interval of 30 min. Judging from the curvilinear trend of MFSB temperature variation, only 3 cycles of charge-discharge were necessary and thus retained. A transient thermo-electric model was set up using commercial finite element analysis

14

ACCEPTED MANUSCRIPT

(FEA) software, ANSYS (Ver. 13.0). The MFSB was meshed with SOLID70 elements, and its upper and

RI PT

bottom faces were covered with SURF152 elements for modeling the heat transfer to the environment by radiation. 4.

Results and discussion

SC

Comparison of the electrical capacity of the MFSB before and after the thermal vacuum testing is

M AN U

presented in Fig. 4 (a). And it can be seen that the curves are almost the same, implying that the thermal vacuum environment has little influence on the MFSB electrical capacity. The slight deviation in discharge voltage and capacity at the given discharge rate may be attributed to self-discharge during the capacity verification before and after the thermal vacuum testing. Therefore, the conclusion that the

TE D

designed multi-functional structural battery can survive the simulated space vacuum and heat environment can be drawn through the nearly superposed profiles. And moreover, after seriously

EP

appearance checking of the MFSB, a maximum dimension deformation of only 0.0096% is acquired. The gas components in the vacuum chamber were monitored by Q-mass measurement and found that the

AC C

vacuum chamber maintained a strict vacuum condition from the very beginning to the end of the test, suggesting that no gas was released from the MFSB in the ultrahigh vacuum environment. Thus it’s believed that the current material formula and workmanship are space feasible and the selected GPLI cells are applicable in the current multifunctional utilization manner. Fig. 4. (a) Capacity of MFSB before and after thermal vacuum test (TVT); (b) Surface temperature

15

ACCEPTED MANUSCRIPT

profiles of MFSB in experiment and simulation.

RI PT

Fig. 4 (b) shows surface temperature profiles of the MFSB recorded in the thermal vacuum testing as well as results obtained from the transient thermal analysis model. As cold and hot ambient environments alternated periodically, discharge and charge were performed respectively. It can be clearly seen that the

SC

established model simulates the MFSB surface temperature pretty well. Therefore, it is rational to

can be used in the following analysis.

M AN U

conclude that this model can effectively predict the temperature distribution inside the MFSB and thus

The surface emittance of the MFSB changes, the temperature of the MFSB at inner center location varies accordingly, and the result curves are shown in Fig. 5 (a) (Note: both the initial and ambient

TE D

temperatures were set to be 293K in the simulation). The inner center temperature will rise from 293K to 301.5K when the emittance is 0.86 while from 293K to 300.5K when the emittance equates to 0.98. The

EP

larger the surface emissivity is, the smaller the temperature will relatively ascend. This is because that at this time the MFSB has stronger thermal exchange ability with surrounding satellite walls and heat

AC C

generated inside the MFSB will thus be dissipated swiftly, and consequently, suggesting a quite cheap and probably useful passive approach for thermal dissipation and management of the MFSB in the future actual utilization. By the way, a moderate emissivity of 0.9 will be chosen in the following simulations. Fig. 5. Temperature profiles at center location in MFSB: (a) with different surface emissivity; (b) in GEO; (c) in LEO; (d) at different discharge rate.

16

ACCEPTED MANUSCRIPT

Fig. 5 (b) and Fig. 5 (c) illustrate temperature rise at inner center location in MFSB at different ambient

RI PT

temperatures for GEO and LEO, respectively. Apparently, temperature variations show different

characteristics for different kinds of orbits. However, for a given style of orbit, the temperature fluctuation presents a similar behavior although the magnitudes are different. And the lower the ambient temperature

SC

is, the higher the MFSB temperature will rise for both of the two orbits. This phenomenon implies that it

M AN U

may be an effective thermal compensation when the satellite inner space is a bit cold. For GEO applications, as is shown in Fig. 5 (b), the MFSB temperature rises firstly because heat is released during this discharge stage. Then a sharp decline comes up because of work-cessation of the MFSB, and if the cessation is long enough, the MFSB temperature will achieve a balance by itself to the reference ambient

TE D

temperature without any intervention from satellite thermal control system. Next, as it is switched to charge, heat is again generated and therefore the MFSB gets hot once more. Since the charge lasts longer

EP

than the discharge, the temperature during this period is higher. Besides, a new and high level of equilibrium is attained by the MFSB itself, indicating no need of ‘help’ from thermal control system.

AC C

Lastly, with the ‘sleeping’ of the full-charged embedded GPLI cell, the MFSB becomes cold until the initial temperature. This is a whole cycle consisting of four phases. However, no such obvious cyclic and phasic temperature variation is found in the LEO application. But similarly, the MFSB temperature also rises in the first 60-min discharge phase. Then the temperature continues to grow in the following charge stage but achieves no balance until the second cycle. Next, no matter how many cycles are applied in

17

ACCEPTED MANUSCRIPT

succession, the temperature stays in a steady value. It should be noted that if the satellite inside

RI PT

environment is set to be 313K initially, a temperature rise of approximate 7.5K will appear for both GEO and LEO applications, which may bring the GPLI cell into a severe and challenging working

circumstance, giving a negative influence on cycle life. Note that the satellite ambient temperature will be

SC

typically set as 293K in the forthcoming discussions of discharge rate’s influence on MFSB temperature

M AN U

variation.

Fig. 5 (d) plots temperature profiles at inner center location in MFSB with different discharge rates for GEO applications. From these simulation results it can be understood that with the increase of discharge rate, the MFSB experiences higher temperature rise because more heat is released from the embedded

TE D

GPLI cell, which also can be seen from Eq. (6). When discharging at 2C, the temperature of MFSB will approximately rise by 18.5K, which is a rather severe elevation that may cause disastrous results and

EP

deserves careful consideration, especially when satellite inner space is initially hotter. Besides, it will be the best if the designed MFSB can maintain itself a proper temperature range which means that both high

AC C

efficiency of mechanical and electrical can be achieved simultaneously as no additional thermal control approaches is adopted. Finally, as for the integrated GPLI cell, high rates of discharging will bring irreversible damage to the cycle life, which should be eliminated in the prototype design. If the GPLI cell has to discharge at high rates, e.g. 2C, it’s not suggested that the ambient temperature be higher than 313K.

18

ACCEPTED MANUSCRIPT

5.

Conclusions

RI PT

A new multifunctional structural battery for satellite applications was developed in this paper.

Simulated space environment testing was carried out and the results show that the current material

formula and workmanship are space feasible. Three-dimensional transient electro-thermal model was

SC

established and validated using the experimental data. The electro-thermal model predicts the MFSB

M AN U

temperature magnitudes accurately for the test load cycle and is then used to predict thermal performance at different working orbits and conditions. It is found that the surface emissivity, charge/discharge rates and ambient temperatures are important parameters that affect the thermal behaviors of the MFSB. And if selected and combined properly, a better design of current multiple-functionality conception and

TE D

realization of high thermal-mechanical efficiency will be achieved. When mounted inside a satellite as the Equipment-Panel, the designed MFSB panel can keep itself within the required operational temperature

EP

range without any other thermal control approaches. The current work is also being extended to take safety concerns, long cycle-life requirements, and charge/discharge intervals limited strictly by the

AC C

spacecraft orbit etc. into consideration. This particular type of satellite multiple functionalities structure, through the use of CFRP composites, PMI foams and GPLI cells, is found to present satisfactory thermal behaviors, showing the potential of reducing satellite mass and volume with lower costs if properly designed and applied.

19

ACCEPTED MANUSCRIPT

Acknowledgements

RI PT

This work has been supported by the Office 503 of Department of Materials in NUDT. The authors would especially like to thank to Prof. D.Z. Li for helpful assistance. References

SC

[1] J. Guerrero, E. Fosness, S. Buckley, Multifunctional structures, Proceedings of the AIAA Space

M AN U

Conference and Exposition, Albuquerque, USA, 2001, P1-P6.

[2] G.S. Aglietti, C.W. Schwingshackl, S.C. Roberts, Multifunctional structure technologies for satellite applications, Shock and Vibration 39 (2007) 381-391.

[3] R.F. Gibson, A review of recent research on mechanics of multifunctional composite materials and

TE D

structures, Composite Structures 92 (2010) 2793-2810.

[4] J. Summers, LiBaCore: power storage in primary structure, Proceedings of the AIAA Space

EP

Technology Conference & Exposition, Albuquerque, USA, 1999, P1-P6. [5] D. Marcelli, J. Summers, B. Neudecker, LiBaCore II: power storage in primary structure,

AC C

Proceedings of the 43rd AIAA/ASME/ASCE Structures, Structural Dynamics, and Materials Conference, Denver, USA, 2002, P1-P8.

[6] M.H. Benson, B.J. Neudecker, Powerfiber for flexible fabric and rigid composite applications, ITN Energy Systems Inc. Report, USA, 2003. [7] B.J. Neudecker, M.H. Benson, B.K. Emerson, Power Fibers: Thin-film batteries on fiber substrates,

20

ACCEPTED MANUSCRIPT

ITN Energy Systems Inc. Report, USA, 2003.

RI PT

[8] J.B. Olson, T.L. Feaver, P.C. Lyman, Structural lithium-ion batteries using dual-functional carbon

fabric composite anodes, Proceedings of the 14th International Conference on Composite Materials, San Diego, USA, 2003, P1-P8.

SC

[9] S.C. Roberts, G.S. Aglietti, Satellite multi-functional power structure-feasibility and mass savings,

M AN U

Proceedings of the IMechE Part G Journal of Aerospace Engineering 222 (2008) 41-51. [10] S.C. Roberts, G.S. Aglietti, Multifunctional power structures for spacecraft, Proceedings of the 57th International Astronautical Congress, Valencia, Spain, 2006, P1-P8.

[11] S.C. Roberts, G.S. Aglietti, Design of multifunctional power structure use plastic lithium-ion battery,

Canada, 2008, P1-P11.

TE D

Proceedings of the AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference, Victoria,

EP

[12] C.W. Schwingshackl, G.S. Aglietti, P.R. Cunningham, The dynamic behavior of multifunctional power structures, Proceedings of the 57th International Astronautical Congress, Valencia, Spain,

AC C

2006, P1-P10.

[13] C.W Schwingshackl, G.S. Aglietti, P.R. Cunningham, Experimental determination of the dynamic behavior of a multifunctional power structure, AIAA Journal 45 (2007) 491-496.

[14] S.C. Roberts, G.S. Aglietti, Structural performance of a multifunctional spacecraft structure based on plastic lithium-ion batteries, Acta Astronaut 67 (2010) 424-439.

21

ACCEPTED MANUSCRIPT

[15] C.W Schwingshackl, G.S. Aglietti, P.R. Cunningham, Parameter optimization of the dynamic

RI PT

behavior of inhomogeneous multifunctional power structures, AIAA Journal 44 (2006) 2286-2294. [16] J.A. Foster, G.S. Aglietti, The thermal environment encountered in space by a multifunctional solar array, Aerospace Science and Technology 14 (2010) 213-219.

M AN U

array, Journal of Aerospace Engineering 25 (2011) 454-462.

SC

[17] J.A. Foster, G.S. Aglietti, Strategies for thermal control of a multifunctional power structure solar

[18] P. Liu, E. Sherman, A. Jacobsen, Design and fabrication of multifunctional structural batteries, Journal of Power Sources 189 (2009) 646-650.

[19] J.P. Thomas, M.T. Keennon, S.M. Qidwai, Multifunctional structure-battery materials for enhanced

TE D

performance in small unmanned air vehicles, Proceedings of ASME International Mechanical Engineering Congress, Washington, USA, 2003, P1-P4.

EP

[20] S.M. Qidwai, W.R. Pogue, J.P. Thomas, A. Rohatgi, Design and fabrication of multifunctional structure-power composites for marine applications, Proceedings of ASME International Mechanical

AC C

Engineering Congress and Exhibition, Boston, USA, 2008, P1-P9. [21] A. Rohatgi, J.P. Thomas, S.M. Qidwai, W.R. Pogue, Performance characterization of multifunctional structure-battery composites for marine applications, Proceedings of ASME International Mechanical Engineering Congress and Exhibition, Boston, USA, 2008, P1-P9. [22] J.P. Thomas, S.M. Qidwai, W.R. Pogue, G.T. Pham, Multifunctional structure-battery composites for

22

ACCEPTED MANUSCRIPT

marine systems, Journal of Composite Materials 47(1) (2013) 5-26.

RI PT

[23] R.H. Carter, J.F. Snyder, J.T. South, M.J. Hagon, D.C. DeSchepper, E.D. Wetzel, Multifunctional battery and fuel cell composite structures for U.S. Army applications, Proceedings of the 2nd International Conference on Polymer Batteries & Fuel Cells, Abstract #210, ECS.

SC

[24] T. Keller, A.P. Vassilopoulos, B.D. Manshadi, Thermomechanical behavior of multifunctional GFRP

(2010) 470-478.

M AN U

sandwich structures with encapsulated photovoltaic cells, Journal of Composites for Construction 14

[25] X.M. Wang, M. Kato, H. Naito, C. Yamada, G. Segami, K. Kibe, A feasibility study of commercial laminated lithium-ion polymer cells for space applications, Journal of The Electrochemical Society,

TE D

153(1) (2006) A89-A95.

[26] X.M. Wang, C. Yamada, H. Naito, S. Kuwajima, Simulated low-earth-orbit cycle-life testing of

EP

commercial laminated lithium-ion cells in a vacuum, Journal of Power Sources 140 (2005) 129-138. [27] J. Stöcker, P. Parigger, M. Thiel, Structural Development of Equator-S, Proceedings of Conference

AC C

on Spacecraft Structures, Materials & Mechanical Testing, Noordwijk, Netherlands, 1996,P21-P27. [28] Y. Wang, C.Y. Peng, W.H. Zhang, Mechanical and electrical behavior of a novel multifunctional structural battery, Journal of Scientific and Industrial Research 73(3) (2014) 163-167.

[29] GJB 6789: 2009. General specification for lithium-ion rechargeable cells in spacecraft. [30] S.C. Chen, C.C. Wan, Y.Y. Wang, Thermal analysis of lithium-ion batteries, Journal of Power

23

ACCEPTED MANUSCRIPT

Sources 140 (2005) 111-124.

RI PT

[31] D. Bernardi, E. Pawlikowski, J. Newman, A general energy balance for battery system, Journal of

AC C

EP

TE D

M AN U

SC

The Electrochemical Society 132 (1985) 5-12.

24

ACCEPTED MANUSCRIPT

RI PT

Figures and Tables

Figures Fig. 1. Schematic of multifunctional structural battery.

M AN U

Layered-construction of the GPLI cell in the MFSB.

SC

Fig. 2. (a) Planform and section view of MFSB (thermal couples included and cables hided); (b)

Fig. 3. Location of the MFSB panel and corresponding thermal dissipation paths. Fig. 4. (a) Capacity of MFSB before and after thermal vacuum test (TVT); (b) Surface temperature profiles of MFSB in experiment and simulation.

TE D

Fig. 5. Temperature profiles at center location in MFSB: (a) with different surface emissivity; (b) in GEO;

AC C

EP

(c) in LEO; (d) at different discharge rate.

Tables

Table 1 Physical properties of MFSB used for transient thermal analysis.

25

ACCEPTED MANUSCRIPT

Physical properties of MFSB used for transient thermal analysis. Components

Density 3

(kg/m )

RI PT

Table 1

Thermal conductivity

Heat capacity

Thickness

(W/(m K))

(J/(kg K))

(µm)

GPLI cell information provided by the manufacturer 2702

238

903

Cu foil

8933

398

385

Al plastic

1636

0.43

PP separator

658

0.33

Anode coating

2840

3.70

Cathode coating

1671

3.20

15

17

1377

110

1978

20

823

60

1113

50

M AN U

Physical properties derived for the MFSB

SC

Al foil

2369

kx=ky=38.06, kz=1.60

887

-

PMI foam

75

0.03

1200

-

CFRP face sheet

1500

kx=ky=29.66, kz=1.24

840

-

AC C

EP

TE D

GPLI battery

1

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT  A new multifunctional structural battery (MFSB) was developed for satellite applications.  Thermal vacuum testing and transient finite element simulation were performed.  Influences posed by surface emissivity, discharge rate and ambient temperature were studied.

AC C

EP

TE D

M AN U

SC

RI PT

 The proposed MFSB shows great potential for satellite mass and volume reduction.