Advertisement

JMST Advances

, Volume 1, Issue 1–2, pp 181–190 | Cite as

A study on the application of metal–air battery to large size uninterruptible power supply with a hybrid system

  • Bonhyun Gu
  • Sung Hyun Yoon
  • Sung Kwan Park
  • Suyoung Byun
  • Suk Won ChaEmail author
Letter
  • 176 Downloads

Abstract

Despite high technology in a battery, the most popular energy storage system for uninterruptible power supply (UPS) these days is a lead–acid battery. Although lead–acid battery has high reliability and low cost, many disadvantages like frequent maintenance work, bulk size, extra expenses and limit of development. Metal–air batteries are developing technology in the battery system and enlargement and commercialization of that are not common yet. But they have extremely high energy density and power density. This study shows a realistic possibility of a metal–air battery adapted to UPS with performance requirement. Cell model, stack design, and hybrid system are developed for a new type of UPS. Especially, electrolyte isolation and hybrid system is a principal specification for better performance and feasibility. Evaluation of suggested UPS system in accord with required specification shows technical feasibility.

Keywords

Hybrid system Metal–air battery Stack modeling Uninterruptible power supply Zinc–air battery 

1 Introduction

Metal–air batteries, due to their high theoretical energy density receive tremendous attention and dedication to replace the rechargeable battery (a.k.a secondary battery) [1, 2]. However, it is still far from commercialization to replace the already existing rechargeable batteries. Meanwhile, commercial UPSs are always stored on activated state so that it is maintained at its fully charged state. This facilitates self-discharge and decreases the battery replacement period. In this study, Metal–air battery integrated UPS system has been developed. The eco-friendly and high performance of the metal–air battery is incorporated to overcome the current low-efficient UPS system and to provide a more reliable system [3].

2 Theory

The uninterruptible power supply is a device that provides emergency power to the load when the input power source from the fundamental power supply is failed by accident. Hence, the purpose of UPS is to provide electricity to the load until the input power source is completely recovered.

2.1 Original UPS

Original UPS for the target system is a type of online UPS. Online UPS has advantages like short switching time, stable output, high control efficiency despite high cost. The batteries which construct the energy storage in UPS are only lead–acid batteries. All batteries are connected as series in-line on discharging state for always stand by. The specification of the batteries is shown in Fig. 1 and Table 1 below and a little different depends on branches.
Fig. 1

Lead–acid battery being used in UPS

Table 1

Battery specification

 

Voltage

Capacity

Type

Discharging property

Alternating current

240 V

1500 Ah

Lead–acid batteries (cell voltage: 2 V, life cycle 15 years)

1 h rate: 1.67 V

1000 Ah

10 h rate: 1.8 V

Direct current

110 V

1500 Ah

1 h rate volt range

2200 Ah

91.85–110 V

2500 Ah

200.4–240 V

The emergency situation from unexpected power failure is 1 time for several years at most. The minimum supplying time is 2 h in any time as the required specification.

2.2 Zinc–air battery

Metal–air battery is divided into aqueous and non-aqueous depending on the composed electrolyte. The electrolyte is arranged by which metal is used as an electrode. Research for finding appropriate metal to the electrode is done first.

Property of batteries is in Table 2. The best metal–air battery for UPS system is Zn–air battery according to research like reactivity test with air or electrolyte, cell test for the output voltage. Reactant after discharging is easy to disposal.
Table 2

Property depends on metal species

System

Aqueous

Non-aqueous

Fe–air

Zn–air

Al–air

Mg–air

Na–O2

K–O2

Li–O2

Metal price ($/kg)

0.40

1.85

2.75

2.75

1.7

20

68

Energy density (Wh/kg)

763

1086

2796

2840

1106

935

3458

Voltage (V)

~ 1.0

~ 1.4

1.1–1.4

1.2–1.4

~ 2.2

~ 2.4

~ 2.6

Stability with air

Stable

Stable

Stable

Stable

Unstable

Unstable

Unstable

Etc

Low voltage

Simple disposal

Unstable with electrolyte

Additional structure

   

2.3 Hybrid system

Among different types of a battery system, a hybrid battery system is a type of system which has two distinct advantages or roles that is in serial/parallel connection to provide the energy source. In this study, the metal–air battery is stored separately with its electrolyte and the required activation time is compensated by designing a hybrid system with a small lithium-ion battery and a zinc–air battery connected in parallel. The small lithium-ion battery used as a secondary battery has better performance in contrast with a lead–acid battery in charging/discharging and footprint size. A schematic diagram of the hybrid system appears in Fig. 2.
Fig. 2

Hybrid system

2.4 Dynamics

Cell modeling is constructed from the results of cell output test. Stack in UPS is modeled with parameters from the cell modeling as 1 dimension. Parameters are designed with principal dynamics like air flow, heating by a reaction, electrolyte injection, etc. Modeling and simulation is done several times because these dynamics is based on stack design.
  • Airflow by fans in a stack

Transition to turbulence corresponding to a static pressure of fans is calculated from the specification of fans and Moody chart. Although Z-type channel flow is better than U-type, this narrow channel has a limit from structure [4]. Every gap between cells is very narrow about 2 mm. An experiment is conducted showing liquid flows through the narrow gap.
  • Thermal dynamics

Convection in this model is divided to natural convection (1) and forced convection (2) by fans.
$$Q = U \times A \times \Delta$$
(1)
$$Q = C_{p} \times \rho Av \times \Delta T$$
(2)
Heat capacity (Cp) is calculated with a principal component aluminum case, electrolyte, zinc, polydimethylsiloxane (PDMS), activated carbon, polyvinyl chloride (PVC), etc. Natural convection is calculated with vertical heat coefficient and facing up heat transfer coefficient from the aluminum case. It is assumed that the entire energy from zinc and reaction consists of electric energy and thermal energy.
  • Electrolyte injection

Isolation of electrolyte during deactivation time is the essential part of this system for longer life cycle by holding back self-discharging and easy maintenance. But the method to execute injection to the stack is difficult. Injection should be done without electric power because UPS is on after power failure. The method gravity and small trigger used and quickly assisting reaction is suggested.

3 Research

Basic data for the study is acquired from cell test experiment. Matlab and Simulink serve program in Matlab are used for modeling and simulation in this study.

3.1 Cell modeling

After the experiment for selection of zinc–air battery as power source, principal parameters in cell modeling are discussed [5, 6, 7]. Parameters change during the operating time not depend on the layout of the system should be variable in cell test. In this study, 3 significant variables below are applied to cell test except for other fundamental experiment environments. Cell voltage results are in Fig. 3.
Fig. 3

Cell voltage depends on parameters a current density, b oxygen density, c temperature

  • Operating temperature

  • Oxygen density between the cathode

  • Current density in the electrode

Cell model is modeled by this data and electric dynamics. The numerical approach is performed to model cell voltage, SOC, and instant zinc consumption with feedback. Figure 4 is a flowchart about the numerical cell model.
Fig. 4

Cell model flow chart

3.2 Stack modeling

A simple calculation shows about 200 cells should be stacked up to correspond to the original UPS specification. Zig–zag type and shared air path to every 2 cells are applied as specific parts of zinc–air battery stack for optimal structure [8]. Entire stack in UPS consist of 8 modules of the zinc–air battery. This module system is designed for preventing an overall malfunction from error in a module and smooth operating environments like air supply and stack cooling.

Accurate measurement is arranged by simulation. Also, simulation results are reflected in cell test and re-modeling to optimal sizing. Figure 5 is a stack modeling diagram and Table 3 is about the design parameter of the stack structure.
Fig. 5

Stack modeling diagram

Table 3

The design parameter of the stack structure

Parameter

Requirement

Value

Surface

Current density 28 mA/cm2

Required current 330 A

1.28 m2

Max current load

Boost converter

370 A

Module

Air fan, current 330 A

Area: 0.16 m2

8 modules

Required voltage

Min 0.8 V, max 1.1 V for cell

180–200 V

Stacking

Required current 330 A

170 cells

Mass of zinc

Simulation

170 g per cell

Thickness of zinc

Cell area, a mass of zinc

0.15 mm per cell

Electrolyte

Experiment

2 mm

Cathode

Experiment

0.7 mm

Separator

Experiment

0.3 mm

Thickness of air tank

Cell arrangement, air flow

16 mm

Volume of module

Airflow, fan flow path

Length 2 m, volume 1.44 m2

Maneuvers below can cope with mechanical malfunction possibility in the stack.
  • Modularization

  • Sensor installation in the solenoid valve

  • Serve mechanical switch for the solenoid valve

  • Clearance in the electrolyte chamber

  • Boost converter output control

  • Setting spare electric capacity in total operating time

3.3 Balance of system modeling

A balance of system includes air fan, air flow path, electrolyte injection system, case covering stack in this study. Air fan and flow path are modeled with dynamics as mentioned earlier. The electrolyte injection system is simulated separately to set injection time and develop an algorithm to serve secondary battery.

Figure 6 is about electrolyte injection system using solenoid valve and gravity.
Fig. 6

a

Electrolyte injection system diagram, b electrolyte volume in 1st tank, c electrolyte volume in the 2nd tank

The simulation shows full injection can be done within 40 s. Then serve secondary battery should cover UPS only below 3 min because of fully reacting time for primary battery.

Electrolyte tank and aluminum case for the stack is validated by the 3-D simulator, SOLIDWORKS. The result guarantee within 1% of deformation in load situation.

3.4 Dynamic evaluation for stack

For validation of modeling, dynamic evaluation for the stack is calculated. First, temperature evaluation based on natural convection and forced convection is conducted with fans. Each module covered by aluminum case is simulated with a vertical heat transfer coefficient and facing up heat transfer coefficient.

Forced convection is calculated by a transition to turbulence flow by Moody Chart [9]. Second, the structural dynamic simulation is conducted with the condition where deformation has occurred below 1% of original condition. The module model is simulated as Fig. 7 by the program ‘Solidworks’ under electrolyte and metal is fully charged in the stack.
Fig. 7

Loading simulation of the module

3.5 UPS system modeling

UPS system includes a zinc–air battery, lithium-ion battery, boost converter for stable and high voltage performance. Each installation operates by operating algorithm from standby normal state to restoration. The entire operating algorithm is in Fig. 8 except detailed electrical control logic.
Fig. 8

a

Designed UPS system model, b UPS operating algorithm

Important standard index of this algorithm is battery state of charge (SOC) and output voltage. Basically, checking whether SOC of secondary battery matches SOC minimum margin and whether power failure lasts over 1 min is a trigger point for electrolyte injection. Low SOC and low voltage is another trigger of that. This trigger time is based on reacting time of zinc–air battery to full activation.

4 Result

Simulation under power failure is evaluated because this study is based on modeling. The required performance of the original UPS is shortly that maximum 64 kW power, 2 h of total duration, 133–217 V of voltage. 12% of capacity for spare and temperature is under 80 °C. The simulation results are divided into stack simulation and secondary battery simulation. Mentioned earlier, the operating environment is set by iterate simulation result and existing UPS system, lead–acid battery. Even if the secondary lithium-ion battery is used for full 5 min, this simulation operates for full 120 min. Because real prototype stack is not developed yet, to spare operating time. Figure 9 includes performance results of the stack and secondary lithium-ion battery.
Fig. 9

a

Stack simulation results, b secondary lithium-ion battery

Stack simulation results show stable output voltage and current.

The maximum current density of zinc–air battery in this study is about 40 mA/cm2 according to experiment result and that is the most important factor in the downsizing of cell electrode and electrolyte. However, spare output should be regarded because of other negative factors neglected for the performance like current density distribution, gravity, etc. [10]. Oxygen portion is not different during the operating time. Although temperature cell, high temperature helps the output performance of stack by preceded experiment until 80 °C. As you can see breakpoint in graphs of Fig. 7 (b), switching from secondary battery to primary battery is performed without a problem.

5 Conclusions

All specifications of existing UPS system are satisfied with suggested UPS system by zinc–air battery hybrid system. Hybrid system and isolating electrolyte give technical suitability and possibility for commercialization to zinc–air battery.

Many risk elements which spring from new technology can be handled as the development of technology in the metal–air battery. This development will increase technical feasibility for mass production in the materials. In addition, ease in maintenance, preoccupancy in eco-friendly technology, a possibility of application to other facilities is the advantage of metal–air UPS system.

References

  1. 1.
    D. Linden, T.B. Reddy, Handbook of Batteries, 3rd edn (McGraw-Hill Education, New York, 2002)Google Scholar
  2. 2.
    C.W. Lee, K. Sathiyanarayanan, S.W. Eom, H.S. Kim, M.S. Yun, Effect of additives on the electrochemical behaviour of zinc anodes for zinc/air fuel cells. J. Power Sources 160(1), 161–164 (2006)CrossRefGoogle Scholar
  3. 3.
    P. Pei, Z. Ma, K. Wang, X. Wang, M. Song, H. Xu, High performance zinc air fuel cell stack. J. Power Sources 249, 13–20 (2014)CrossRefGoogle Scholar
  4. 4.
    S. Maharudrayya, S. Jayanti, A.P. Deshpande, Flow distribution and pressure drop in parallel-channel configurations of planar fuel cells. J. Power Sources 144(1), 94–106 (2005)CrossRefGoogle Scholar
  5. 5.
    E. Deiss, F. Holzer, O. Haas, Modeling of an electrically rechargeable alkaline Zn–air battery. Electrochim. Acta 47(25), 3995–4010 (2002)CrossRefGoogle Scholar
  6. 6.
    Y. Li, H. Dai, Recent advances in zinc–air batteries. Chem. Soc. Rev. 43(15), 5257–5275 (2014)CrossRefGoogle Scholar
  7. 7.
    D.M. See, R.E. White, Temperature and concentration dependence of the specific conductivity of concentrated solutions of potassium hydroxide. J. Chem. Eng. Data 42(6), 1266–1268 (1997)CrossRefGoogle Scholar
  8. 8.
    H. Ma, B. Wang, Y. Fan, W. Hong, Development and characterization of an electrically rechargeable zinc–air battery stack. Energies 7(10), 6549–6557 (2014)CrossRefGoogle Scholar
  9. 9.
    F. Barreras, A.M. López, A. Lozano, J.E. Barranco, Experimental study of the pressure drop in the cathode side of air-forced open-cathode proton exchange membrane fuel cells. Int. J. Hydrogen Energy 36(13), 7612–7620 (2011)CrossRefGoogle Scholar
  10. 10.
    M. Geske, M. Heuer, G. Heideck, Z.A. Styczynski, Current density distribution mapping in PEM fuel cells as an instrument for operational measurements. Energies 3(4), 770–783 (2010)CrossRefGoogle Scholar

Copyright information

© The Korean Society of Mechanical Engineers 2019

Authors and Affiliations

  • Bonhyun Gu
    • 1
  • Sung Hyun Yoon
    • 2
  • Sung Kwan Park
    • 3
  • Suyoung Byun
    • 4
  • Suk Won Cha
    • 1
    Email author
  1. 1.Department of Mechanical EngineeringSeoul National UniversitySeoulRepublic of Korea
  2. 2.Department of Electrical and Computer EngineeringSeoul National UniversitySeoulRepublic of Korea
  3. 3.Department of Materials Science and EngineeringSeoul National UniversitySeoulRepublic of Korea
  4. 4.New Technology Research DepartmentKorea District Heating Corporation Research and Development InstituteYonginRepublic of Korea

Personalised recommendations