See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/305032999

Packaging Materials for Li-Ion Batteries

Conference Paper · September 2014

DOI: 10.13140/RG.2.1.1607.8320

CITATIONS READS

0 4,974

5 authors, including:

Qiaoyan Pan Reiner Weyhe

RWTH Aachen University Accurec GmbH, Germany
1 PUBLICATION 0 CITATIONS 20 PUBLICATIONS 721 CITATIONS

SEE PROFILE SEE PROFILE

Albrecht Melber Bernd Friedrich

ACCUREC RWTH Aachen University
4 PUBLICATIONS 26 CITATIONS 882 PUBLICATIONS 5,567 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Pure Metals (Zone melting, crystallization, distillation) View project

ecoLiga – Recycling and resynthesis of carbon materials from lithium batteries: Recovery, processing, reuse and adapted cell design View project

All content following this page was uploaded by Bernd Friedrich on 08 July 2016.

The user has requested enhancement of the downloaded file.

 PACKAGING MATERIALS FOR LI-ION BATTERIES

Investigation on packaging materials

for safe transport of spent Li-ion batteries

!
Qiaoyan Pan1, Reiner Weyhe1, Albrecht Melber1, Ieva Klavina1, Bernd Friedrich2

!
1: ACCUREC Recycling GmbH, D-45472 Mülheim an der Ruhr, Germany

2: IME Process Metallurgy and Metal Recycling, RWTH Aachen University, D-52056 Aachen, Germany

Abstract:
_____________________________________________________________________________________________________!
After one decade of fast growing sales of Li-ion batteries, these batteries reach End-of-Life (EOL) status in
significant volumes and return to sorting and recycling plants through various collection routes. As these Li-ion
batteries are still partly charged and contain high flammable components, the number of fire incidents has been
growing during transportation, storage and treatment. The causal analysis show that most cases could have been
avoided by using appropriate packaging.

!
This study compares functional properties of five market available packaging materials, respective insulation/
cushioning materials for spent Li-ion batteries by experimental work. After that, thermal behaviors of Li-ion
batteries under various practical orientated conditions of these packaging materials have been investigated. The
results show that sand, with lowest thermal diffusivity, is the far best among 5 materials to prevent uncontrolled
self-ignition (“Chain Reaction”) of bulk packaged EOL-Li-ion batteries.

_____________________________________________________________________________________________________
Keywords: Li-ion battery; Packaging; Transport; Thermal Runaway; Thermal property; Vermiculite; Sand

1. Introduction
not realizable. Especially after end of life (EOL), the
spent Li-ion batteries are in a mixture of different
In recent years, demand for Li-ion batteries has
conditions and their safety devices frequently are
increased enormously as a power source for portable
defective and cannot prevent fire incidents. Depending
devices, electric vehicles (EV) and other power driven
on precaution and packaging, these Li-ion batteries can
electronic applications, due to their excellent
go into thermal runaway during transportation and
characteristics of high voltage, high energy density, and
storage. The further progression of incident is heavily
light weight. However, with increasing use of lithium-
dependent on the surrounding conditions, especially the
ion batteries, reports of occasional fire incidents with
thermal insulation materials, which need to be
fatal ending appeared in the news [1].

investigated scientifically.

The safety issue of li-ion batteries has drawn great
In this study, the thermal runaway behavior of Li-
attention all over the world. In order to improve the
ion batteries was analyzed at different state of charge
safety performance of li-ion batteries, many safety
(SOC) and in various market available packaging
devices and actions have been developed, such as
materials in order to provide reference for safe
pressure release vent, thermal fuse, PTC, shutdown
packaging and transport of spent Li-ion batteries.
separator, non-flammable electrolyte, etc.[2] Despite of
the fact that these components have greatly improved
the safety performance of li-ion batteries, 100%
prevention of thermal runaway of li-ion batteries is still

« 1" »
 PACKAGING MATERIALS FOR LI-ION BATTERIES

2. General information about Li-ion • strong outer packaging, conforming to the

batteries’ packaging and transportation
packaging group II (medium danger)

! • empty space in packaging shall be cushioning
material that is non conductive and non flammable

2.1 Legal Background concerning !
transportation of EOL-Li-based batteries Moreover the transport documentation and marking
(ADR/IMDG Code)
shall include following information:

• the type of packing according group

According to the United Nations Economic and • tunnel category E (forbidden in tunnels category
Social Council (ECOSOC) Lithium and Li-Ion E, allowed in A, B, C or D)

batteries are classified as dangerous goods and • proper shipping name:

therefore specific rules for the transport are a.lithium metal batteries, UN 3090

implemented. The related regulations for road transport b.lithium-Ion batteries, UN 3480

are ADR (International Carriage of Dangerous Goods c.used lithium metal batteries, UN 3090
by Road) and IMDG Code (International Maritime WASTE LITHIUM METAL BATTERIES

Dangerous Goods Code) for maritime transportation. d.used lithium-Ion batteries, UN 3480
Both provide special provisions, with main WASTE LITHIUM-ION BATTERIES

requirements that are described below. Additionally, • transport category 2 (the maximum amount of the
packing instructions are defined, that correspond to the goods that can be carried before having to apply to all
battery type (< 500 g >; Li or Li-Ion; etc.) and its provisions ADR)

condition (UN-approved, new, used, defective, • special carrier documentation, requirements (e.g.
damaged, etc.).
ADR transport permit)

In particular, SP188 applies to the shipment of small • special requirements for driver (e.g. ADR driving
quantities of Lithium-metal batteries with less than 1.0 license)

grams lithium per cell and 2.0 grams lithium per • class 9 label (miscellaneous dangerous goods)

battery as well as small quantities of Lithium-Ion cells
with less than 20 Wh and batteries with less than 100
!
Under certain conditions, simplifications for
Wh.
transportation and packaging are applicable with
For larger batteries and larger quantities than those Special Provision:

defined in SP 188 the transport has to be made under a • SP636 (special provision for batteries from
fully regulated regime.
collection schemes; apply only to ADR)

The current ADR Regulations include Special • SP188 (special provision, when the batteries are
Provision - SP 661 for damaged and defected batteries. not subject to ADR provision)

Several authorities have further developed Multilateral resulting in Packing Instruction:

Agreement M259 governing the transport and • P903a (special packing instruction for used Li-
packaging conditions for these batteries. This batteries in max. 30kg boxes)

agreement is valid between signatories until 31st • P903b (special packing instruction for mixed used
December 2014, for non-signatories remain valid SP batteries including Li-batteries).

661.

From 2015 shall apply new provisions - SP 376 for
!
A summary of all current relevant Provisions and
damaged and defected batteries and SP 377 on the Packaging Instructions are illustrated with the decision
transport of waste lithium batteries, associated with tree in Annex I.

new packaging instructions P908/SP 376, LP904/
SP376 (special packing instruction for single damaged
!
2.2 Current situation and experiences

or defective battery and cells and these contained in
equipment) and P 909/SP377.

!
! End of life Li-Metal and Li-Ion accumulators are
generated predominantly at facilities listed and ranked
The general principles that have to apply when
in Table 1.

packing Lithium batteries for their transport are:

• cells or batteries shall be protected against short

circuit

• cells or batteries shall be fixed in packaging that
protects against damage, caused by the movement
within the packaging (e.g. vibrations)

« 2" »
 PACKAGING MATERIALS FOR LI-ION BATTERIES

Table 1 : Source of EOL Li-Primary and Li-Ion accumulators

Battery type Application Facility of generation Comment

Li-Primary consumer electronics collection at retailers and mainly with mixed used
municipalities portable batteries

industrial (meter etc.) professional installation/ good awareness of danger;

maintenance companies mainly adequate packaging

military spare part/logistic department high awareness, exemplary

packaging

Li-Ion portable consumer electronics collection at retailers and mainly with mixed used
batteries (mobile, laptop, etc.) municipalities portable batteries

power tools professional repair centre low awareness of danger;

mainly bulk packaging
without insulation
various WEEE! very low awareness;
inadequate packaging and
-recycling plants
regular illegal
transportation

Li-Ion industrial e-bikes dedicated brand repair centre low awareness of danger;
and e-mobile appl. and bike shops mainly bulk packaging
without insulation

hybrid and full electric auto-repair centres; car due to novel character of
vehicle recyclers (later stage) appl., today there is a lack of
market experience

Industrial and military users tend to package Li- occurred by a Li-thermal runaway e.g. when it is
Primary batteries after EOL, or after expire date with packaged in a drum or card box.

their original packing, mainly trays in card boxes. As Unsorted batteries generated at WEEE (Waste
far as these trays fix the batteries reliable even under Electrical and Electronic Equipment) recycling plants
more robust handling in the sector of waste do contain a fast increasing proportion of Li-primary
management, this provides a sufficient and safe and secondary batteries. They are either

packaging.
a) manually disassembled or

Consumer type Li-batteries, diluted in a bulk b) extracted from semi-mechanical sorting
standard mixture of EOL household batteries, do not machineries.

pose a significant risk during transportation and In both cases they have been abused, but even
storage. They are surrounded by Alkaline, NiCd and with manual dismantled WEEE scrap, the batteries
other non-reactive batteries. The high thermal heat show mostly a damage of outer casing, so that the
capacity of nonreactive batteries and subsequent function of safety electronic is not any more ensured.
potential endothermic reaction (e.g. water evaporation)
will limit the heat transfer and peak temperature

« 3" »
 INVESTIGATION ON PACKAGING MATERIALS

!! Depending on penetration of training activities through

!! responsible collection schemes, these dealers should be
equipped with instructions and adequate packaging
!! materials.

In some countries collection schemes start to
!! provide card boxes or drums with the intention to
!! transport these boxes in accordance with ADR. Without
any additional prevention measures, this precaution
!! seems not sufficient as it does increase awareness but
not decrease safety risk.

!! In recent years E-mobility in automotive sector has

! started with the use of Nickel-Metal-Hydride batteries.

Li-Ion batteries are now introduced mainly in latest
! Figure 1: Li-ion batteries from manual disassembly

generation of Hybrid cars and in a small number of
selected full EV cars. EOL Li-Ion batteries from E-
Due to high number of staff involved, training and
Mobility are not yet available, and European wide, less
packaging advices are well introduced. Mostly Li-
than 150 t/a of them have been registered in 2012,
batteries are insulated with vermiculite and stored in
mainly being off-spec or defective and send for
60/120 L plastic or 200 L steel drums.

recycling. Generator of waste is predominantly
WEEE treatment plants with semi-automated pre-
automotive industry and OEM´s, which tend to send
crushing of WEEE-waste subsequently sort out
those Li-Ion batteries/modules in original packaging or
batteries by hand. Although these batteries contain a
provisional wood boxes. In liner, insulation and
high concentration of Li-batteries, this battery stream is
cushioning materials are used rarely.

identified as EWC 200133 mixed batteries, and they
Either from legal or from practical point of view,
are sent without any insulating materials.

lithium batteries need to be well packaged with
insulation/cushioning materials for safe transportation
and storage. On the basis of ensuring safety, the
insulation/cushioning material should also reduce cost
continuously.

!
3. Thermal runaway behavior of Li-ion
battery cells

!
3.1 Thermal runaway mechanism

!
Generally, thermal runaway occurs when an exo-
thermic reaction gets out of control in a battery. As a
result, an increase in temperature will accelerate the
! reaction rate, causing a further increase in temperature

! Additionally to the missing insulation, most

Figure 2: Mixed batteries from WEEE-plant
and reaction rate, which may result in an explosive
reaction [1].

The intrinsic heat source for Li-ion battery’s thermal
batteries are heavily damaged and partly deformed due runaway is associated with electrical energy and
to mechanical forces of impact or crushing mills. From chemical energy. The electrical energy can be valued
legal point of view, the containing damaged Li- by the capacity and the state of charge (SOC) of the
batteries need to be packed according to Multilateral battery, while the chemical energy is a material issue.
Agreement M259 of Section 1.5.1 ADR. However in In other words, the chemical energy mainly depends on
practice, they are rarely packed according to the the properties of the substances used in the battery.
Agreement.
Researchers [3] [4] had summarized some main heat
Dedicated collection points for Li-batteries, for sources from the reactions in Li-ion batteries (see Table
example e-bike dealer, generate unique waste streams 2). These reactions do not take place in an exact given
of batteries in significant and increasing quantities. order, some of them may occur simultaneously. The
These batteries return because of defective function or onset temperature and total energy of these reactions
damage, and therefore implicate danger of fire. varies with the components and the SOC [5].
« 4" »
 PACKAGING MATERIALS FOR LI-ION BATTERIES

The resulting temperature of a lithium ion cell is in thermal runaway. While the amount of heat
determined by the heat balance between the amount of generated depends on the battery itself (SOC and
heat generated and that dissipated by the cell [1]. If the battery substances), the heat dissipated depends
cell can dissipate the heat generated, the reactions mainly on the environment [5] [6]. Therefore, to
described in Table 1 will not occur or stabilize at a prevent the battery from thermal runaway, an
certain temperature level and the cell will environment with good heat dissipation has a

progressively cool down. However, if the heat beneficial effect, while a heating environment should
generated is more than what can be dissipated (such as be avoided.
in adiabatic environment, or even worst, in a heating
environment), the exothermic processes will proceed
under adiabatic-like conditions and the cell’s
temperature will increase rapidly, eventually resulting
Table 2: Summary of reactions in Li-ion battery [3] [4] [7]

Exothermic reactions Temperature range Energy released

!(CH2OCO2Li)2 → Li2CO3 + C2H4 + CO2 + 1/2O2

1 SEI layer decomposition 80-130 °C 186-257 J/g

2Li + (CH2OCO2Li)2 → 2Li2CO3 + C2H4

!
2 Intercalated carbon+ electrolyte
2Li + C3H4O3 (EC) → Li2CO3 + C2H4
110-290 °C 350-1714 J/g

2Li + C4H6O3 (PC) → Li2CO3 + C3H6

2Li + C3H6O3 (DMC) → Li2CO3 + C2H6
3 Lithiated carbon+ fluorinated binder 220-350 °C 1025-1647 J/g
!
Li + C2H2F2 → LiF + C2HF + 1/2H2
!
4 Electrolyte/Solvent decomposition
LiPF6 → LiF + PF5
130-350 °C 155-530 J/g

PF5 + H2O → PF3O + 2HF

C2H5OCOOC2H5 + PF5 → C2H5OCOOPF4HF + C2H4
C2H5OCOOC2H5 + PF5 → C2H5OCOOPF4 + C2H5F
C2H5OCOOPF4 → HF + C2H4 + CO2 + PF3O
C2H5OCOOPF4 → C2H5F + CO2 + PF3O
C2H5OCOOPF4 + HF → PF4OH + CO2 + C2H5F
C2H5OH + C2H4 → C2H5OC2H5
!
5 Positive active material decomposition
Li0.5CoO2 → 1/2LiCoO2 + 1/6Co3O4 + 1/6O2
181-270 °C 290-1600 J/g

Li1-xNiO2 → (1-2x)LiNiO2 + xLiNi2O4 (x≤0.5)

Li1-xNiO2 → [Li1-xNi(2x-1)/3][Ni(4-2x)/3]O(8-4x)/3+(2x-1)/
3O2 (x>0.5)
Mn2O4 → Mn2O3 + 1/2O2
The oxygen released may react with solvent:
5/2O2 + C3H4O3 (EC) → 3CO2 + 2H2O
4O2 + C4H6O3 (PC) → 4CO2 + 3H2O
3O2 + C3H6O3 (DMC) → 3CO2 + 3H2O
!
6 Lithium metal reactions* !! !
(1)29055 J/g
2Li + 2H2O → 2LiOH + H2 ――(1) ~180 °C (2)12054 J/g
2Li + 1/2O2 → Li2O ――(2) (melting point of Li) (3)7903 J/g
3Li + 1/2N2 → Li3N ――(3)
*The mechanism about battery lithium metal reactions is not sufficient in Literature. It is indicated in paper [8] that the SEI layer on lithium
protects the metal from reaction until the melting point. Here the enthalpies [J/g Li] of reactions of Li+ H2O / O2 / N2 are given as reference.
It has to be noted that the reaction of Li with nitrogen has a similar intensive reaction enthalpy like Li with oxygen. Nitrogen can therefore
not be used as inert protecting gas in thermal treatment of Li-based batteries.

« 5" »
 PACKAGING MATERIALS FOR LI-ION BATTERIES

3.2 Experiment work

!
Table 3: Voltage/weight of 18650-type cell

In order to understand and define standard
before and after furnace test!
conditions of thermal runaway behavior of Li-ion
batteries, 18650-type Li-ion cells, which are widely SOC Weight Voltage
used in laptops and electric tools, have been tested by Before After Weight
ca. Before test After test
heating in vacuum furnace.

! 50% test test loss

! 45.5 g 36.6 g 20% 3.63 V 0.00 V

!
!
!
!
!
!
!
!
!
!
!
! The experiment was carried out in a vacuum furnace
Figure 3: Set-up in vacuum furnace

equipped with an external condenser, a vacuum pump, !

Figure 5: Temperature and pressure at vacuum furnace test
a pressure measurement device and a temperature
control & measurement system. The set-up in the
vacuum furnace is shown in Figure 3.

!
(18650-type Li-ion battery with 50%SOC)

! As the temperature rises, the internal partial

pressure of volatile organics increases. After about 30
3.3 Experiment result and discussion

! min a small pressure peak at cell surface temperature of
159°C is recorded. At this point, the cell released the
3.3.1 Thermal runaway behavior single volatile organics through the safety vent. Therefore, the
18650-type Li-ion cells
furnace pressure and the temperature around the cell
Figure 4 shows one typical sample of new showed a small rise while the cell itself slightly cooled
purchased batteries, before and after the test. In this down due to the extraction of evaporation enthalpy.

test, a cylindrical 18650 cell (rated voltage: 3.6 V, rated After that, the cell was continuously heated up by a)
capacity: 3400 mAh) was contacted with 2 the external heater and b) inner chemical reactions. The
thermocouples: TC1 to measure the surface increasing temperature gradient at 35-37min indicates
temperature of the cell (T Cell surface); TC2 to that the cell is heated by itself. At T(Cell surface)
measure the furnace temperature (10 mm above the 242°C, the cell surface temperature started to rise
cell, T furnace).
sharply and finally reached a peak of 687°C. The so-
Information about the cell’s voltage and weight call “thermal runaway” process has happened. It took
before and after test is given in Table 3. Curves of cell 20 seconds for the temperature of the cell to rise from
& furnace temperatures and furnace pressure over time 242°C to 687°C. A large volume of gases (O2,
are shown in Figure 5.

! hydrocarbons, CO2, H2O …) were produced by the

! reactions described in Table 2, leading to a sharp

! pressure rise.

!
!
!
!
!
! (a) Before test
! !

(b) After test

Figure 4: 18650-type battery (50% SOC)




before and after furnace test

« 6" »
 PACKAGING MATERIALS FOR LI-ION BATTERIES

The heater of the furnace was switched off after

thermal runaway and cooled down gradually. The cell
kept its shape after the thermal runaway. The cell
temperature was so high that Al foils melted down and
were carried out of the cell by gas eruption (the silver
metal drops on the head of the cell, see Figure 4).
However, some other test with the same cell type, but
with 100% SOC, the thermal runaway process was
much more violent so that the top lid of the positive
pole burst open and the escaping gas carried the Cu
foils out of the cell (see Figure 6).

! !! Figure 8: Temperature and pressure of Li-ion batteries

!!!
!!
at different SOC

It can be seen that the thermal behavior is clearly

! (a) Before test (b) After test

Figure 6: 18650-type battery (100% SOC) before

dependent on battery SOC. a) Concerning peak
temperature, battery of 100% SOC showed highest
and after vacuum furnace test
temperature (579°C), followed by 50% SOC battery
with 539°C. The battery with 10% SOC showed no
3.3.2 Effect of Li-ion battery’s State of
explicit cell surface temperature peak and followed the
Charge (SOC) on thermal runaway behavior temperature gradient of the furnace. b)An additional
To further verify the influence of SOC on the SOC-specific feature is the on-set temperature: the
thermal runaway, three cells with different SOC (100%, battery of 100% SOC showed lower on-set temperature
50%, 10%) were tested. Exemplary results are shown (ca. 275°C) than battery of 50% SOC (ca.320°C). c)
with table 4 and complementary Figure 7 and Figure 8.
Additionally it can be seen from table 4 that the highest
! weight loss occurred at 100% SOC battery (27%),
! followed by battery at 50% SOC (16%) and battery at
! 10% SOC (12%).

! It is easy to understand that, with higher SOC
! more electrical energy is available in a cell, and thus
! higher peak temperature can be generated in thermal
! (b) After test

! runaway. Besides, as the SOC increases, more lithiated
carbon exists in the cell. It is also indicated in paper [2]
(a) Before test

Figure 7: Li-ion batteries at different SOC (100%, 50%, 10%)

! that the positive materials have lower thermal stability
in delithiated state (high SOC). These active materials

!
before and after vacuum furnace test
will react in thermal runaway process and thus lead to
higher weight loss than that of a cell at low SOC.

! !
! 3.3.3 Conclusions

Table 4:Voltage/weight of cells at different SOC before

and after vacuum furnace test
In vacuum furnace tests, 18650-type Li-ion
! batteries of different SOC release the first pressure in a
temperature range of 120°C-190°C. It is found that the
No. SOC Weight, g Voltage, V highest cell surface temperature at thermal runaway is
in range of 340°C-690°C.

Before After Loss Before After Regarding to the effect of SOC, it is found that the
test test test test higher SOC is, the earlier the battery starts its thermal
Cell 1 100 % 42.3 31.0 27 % 4.00 0.00 runaway process, and ends up at higher peak
Cell 2 50% 42.0 35.2 16 % 3.70 0.00 temperature with consequence of higher weight loss.

Cell 3 10% 42.3 37.2 12 % 3.18 0.00
!
!
« 7" »
 PACKAGING MATERIALS FOR LI-ION BATTERIES

4. Basic properties of packaging materials

!
As written in Chapter 2, for legal and safety reasons !
cushioning and insulation materials (taken as
“packaging materials” in this paper) are necessary for
!
transportation and storage of Li-ion batteries. This ! !
(1) Sand
chapter mainly investigates the thermo-physical
properties of market available packaging materials to !
provide a basic understanding of their thermal !
performances.

! !
4.1 Generic literature information
! (2) Vermiculite
Vermiculite is the main predominant packaging
material for EOL Li-ion batteries in Europe. In this
! !

study, five market available packaging materials !

(shown in Figure 9) are tested and evaluated.

!
1.Sand: fire-dried quartz sand, main chemical
composition: SiO .

2
!
2.Vermiculite (standard application): ! !
(3) Sorbix
exfoliated vermiculite particles, main chemical
composition: SiO , Al O , MgO.

2 2 3
!
3.Sorbix: an oil binding agent based on !
calcium silicate hydrate for fire protection,
main chemical composition: SiO , CaO, Al O .

2 ! 2 3

4.Absorbent: a kind of clumping cat litter, ! (4)Absorbent

main chemical composition: bentonite
(hydrous alumina silicates), clinoptilolite, ! !

sepiolite.

5.PyroBubbles: an extinguishing agent, main
!
chemical composition: SiO incorporating with
2 !
N .

2

The main basic physical properties of the materials

!
are listed in Table 5.
! !
(5)PyroBubbles

! !
!
Table 5: Basic properties of packaging materials based on literature [ 9 -19 ]

No. Material Thermal Density Tmelt Heat capacity Particle

Moisture*
conductivity size

W/(m·K) kg/m3 °C J/(kg·K) mm %wt

1 0.15-0.25(dry),
Sand 1600 1600-1700 830 0-1 0.07
0.25-2(moist)
0.06(25°C)

2 Vermiculite 85-90 1315 840-1080 0-3 1.30
-0.18 (500°C)
3 Sorbix not available 420 not available ca.700 0-4 1.55
4 Absorbent 0.34 600 >1200 1150 0-8 1.16
5 PyroBubbles 0.075 (20°C) 190-230 1100 700 1-5 0.05
*The moisture was measured by weight difference before and after drying (2h, 200°C)

« 8" »
 PACKAGING MATERIALS FOR LI-ION BATTERIES

4.2 Measurement of thermal conductivity !

To set reproducible, comparable experiments, all
All these materials have sufficient electric resistance materials were dried in an oven at 200°C for 10 hours.

to avoid short circuit. Regarding to the purpose of fire In this test the heat is vertically from the bottom
protection, this study partly focuses on the thermal heater to the top Cu cooling plate. The horizontal heat
properties of these materials.
transfer can be neglected because of the insulation
The main parameters describing the thermal layer outside the vessel. After the system has reached
properties of a material are the thermal conductivity λ the equilibrium temperature, the vertical heat flow can
(W/m·K), the heat capacity Cp (J/kg·K) and the be considered equal to the heat flow absorbed by
thermal diffusivity α (m2/s).
water. Derivation of thermal conductivity is shown
However, the public available data of thermal below.

conductivity are incomplete and even incomparable !
because of their different measurement standards. Thus, • • !
the thermal conductivities of these materials have been q specimen = q water ! (4)
measured in the lab.

! !
• (Tbottom − Ttop ) π 2 ! (5)
q specimen =λ⋅
H
⋅ D
4 !
4.2.1 Experiment setup and preparation

• • !
! q water = m water ⋅ Cpwater ⋅ (T − T ) !
2 1
(6)

The heat conductivity λ of these 5 materials was !

measured with steady-state method. The schematic (T −T ) π • !
diagram and the picture of setup are shown in Figure λ ⋅ bottom top ⋅ D 2 = m water ⋅ Cpwater ⋅ (T2 − T1 )
H 4 ! (7)
10.
!
! •
4 ⋅ m water ⋅ Cpwater ⋅ H ⋅ (T2 − T1 ) !
! λ=
!
(8)
! !!
π ⋅ D 2 ⋅ (Tbottom − Ttop )
!
! !!
!
! !
! ! ― Heat flow transfers through test specimen, W;

!
! ―Heat flow absorbed by water, W;

!
! !
! ― Thermal conductivity of specimen, W/(m·K);

! ― Bottom/top temperature of specimen, °C;

!
! ― Height of specimen, m;
!
! ! ― Diameter of specimen, m;
!
! ― Mass flow of water, kg/s;
!
! ―Heat capacity of water, 4177 J/(kg·K) (25°C);

!
! ― Temperature of inlet water;

! !
! ― Temperature of outlet water.

!
!
! Figure 10: Set-up for thermal conductivity measurement
!
!
« 9" »
 PACKAGING MATERIALS FOR LI-ION BATTERIES

4.2.2 Experiment result

calculated and listed in Table 6, where specific heat
capacity Cp has been determined by literature research,
The measured thermal conductivities (λ) are shown and bulk density ρ has been measured in the laboratory.

in Table 6. If a packaging material has low thermal As shown in Table 6, sand has the smallest thermal
conductivity, a thermal running battery surrounded by diffusivity among other materials and hence can be
it can dissipate heat at a low rate. Therefore, as considered as a preferable material to prevent “chain
described in 3.1, batteries in a material with lower reaction” of batteries. Absorbent ranks the 2nd place
thermal conductivity will easier runaway thermally. with a same order of magnitude as sand. PyroBubbles,
Based on this, sand, with highest thermal conductivity Sorbix and Vermiculite have relatively large thermal
of the 5 materials, is expected to be the best material to diffusivity, which is not benefit to prevent “chain
prevent thermal runaway.

! reaction” of batteries.

!
! !
Table 6: Thermo-physical properties of dry packaging materials

4.3 Thermal diffusivity

!
Usually, in heat transfer analysis, thermal diffusivity
No. Packaging λ (dry, ρ Cp α
is the thermal conductivity versus density and specific
heat capacity at constant pressure [20]. It measures the material 25°C-280°
ability of a material to conduct thermal energy relative C)
to its ability to store thermal energy. The formula is W/(m·K) kg/m3 J/(kg·K) m2/s
given by:

! 1 Sand 0.31 1530 830 2.44E-7

! 2
3
Absorbent
PyroBubbles
0.24
0.21
820
270
1150
700
2.55E-7
1.11E-6
λ ! 4 Sorbix 0.18 170 ca.700 1.51E-6
α= ! (9) 5 Vermiculite 0.15 90 ca.960 1.74E-6
ρ ⋅ Cp
! !
!
― Thermal diffusivity, m2/s;

! !
However, the result above is based on dry materials.
― Thermal conductivity, W/(m·K);
!
― Density, kg/m3;

In reality, the diffusivity of these materials will vary
with their moisture. Because the specific heat capacity
of water is relatively high [ Cp(H2O) = 4181 J/(kg·K),
― Specific heat capacity, J/(kg·K).
25 °C [21] ], material that has higher moisture will
! have larger integrated volumetric heat capacity. On the
! other hand, the thermal conductivity will also vary with
increasing water content [ λ(H2O) = 0.609 W/(m·K),
Substances with high thermal diffusivity rapidly
adjust their temperature to that of their surrounding 27°C [22] ]. As a result, the comprehensive effect on
temperature because they conduct heat quickly in thermal diffusivity is unknown. Wet sand is found to
comparison to their volumetric heat capacity.
have higher diffusivity than that of dry sand in
A Li-ion battery at thermal runaway can be treated literature [23]. But the effect of moisture on other
as an unsteady heat source. Thus, thermal diffusivity is packaging materials’ thermal diffusivity remains
more reasonable than thermal conductivity to describe unknown. Therefore, the thermal properties of
the influence of a battery at thermal runaway on the investigated packaging materials with practical
surroundings. In packaging material with thermal moisture have been compared in chapter 5.
diffusivity, the high temperature frontline caused by
thermal runaway can move fast through the material,
reaching other potential surrounding batteries. As a
consequence, thermal runaway of surrounding batteries
and then a “chain reaction” are more likely. For a safe
storage and transportation of Li-ion batteries, the
thermal diffusivity of packaging material should be
therefore as small as possible.

!
The thermal diffusivities of the materials tested are
« 10
" »
 PACKAGING MATERIALS FOR LI-ION BATTERIES

5. Thermal runaway behavior of single

5.2 Experiment result

Li-ion battery in selected packaging materials
The dynamic temperature distribution in different
packaging materials is shown in Figure 12.

In this chapter the previous investigated packing !
materials have been tested under practical moisture !
shown in Table 5.
!
! !
!
5.1 Experiment setup
!
!
Single 18650-type Li-ion cells have been forced
!
into thermal runaway in different packaging materials
! !
in order to compare their performances.

!
As shown in Figure 11, a 150 W cylindrical heater, a
!
100% SOC 18650 type Li-ion battery and 6
!
thermocouples were placed within packaging material
!
in a 60 L ADR approved plastic drum (Φ 36 cm x 63
!
cm). The heater is connected to an external temperature
!
control system with accuracy of +/-1 °C. Thermocouple
!
T1 measured the battery surface temperature. T6
!
measured the heater surface temperature. T2, T3, T4
!
and T5 were placed above the battery within the
! !
packaging material. The distances between battery and
!
heater and between each thermal couple were 15mm,
!
respectively.

!
At the beginning of the experiment, the heater was
!
heated to 560 °C, which simulates the temperature of a
!
Li-Ion battery at thermal runaway due to e.g. short
!
circuit, and then held on until the thermal runaway of
!
battery started.

! !
! ! !

! !
! !
! !
! !
! !
! !
! !
! !
! ! !
!
!
!
!
!
!
!
! !
! ! !
Figure 11: Setup for testing of thermal runaway behavior of single !
Figure 12: Temperature distribution in different

Li-ion battery in selected packaging materials

! packaging materials

« 11
" »
 PACKAGING MATERIALS FOR LI-ION BATTERIES

It can be seen that,

(a) at same experimental setup, it took much longer
time to ignite the battery in sand (Vermiculite: 66 min,
PyroBubbles: 71 min, Sorbix: 103min, Absorbent:
140min, Sand: 379 min. ) For the test in sand, due to
the time limit, the distance between battery and heater
was even shortened to 10 mm to ignite the battery;

(b) after the occurrence of thermal runaway, the
temperature of packaging material at 15mm distance
from battery, T(15 mm), raised significantly in
Vermiculite, PyroBubbles and Sorbix; while in
Absorbent and Sand it raised slightly ( ∆T(15mm)=174
°C (Vermiculite), 160 °C (PyroBubbles), 158 °C Figure 13: Dynamic cell surface temperature in

(Sorbix), 67 °C (Absorbent) and 21 °C (Sand) ). This different packaging materials

can be explained by their thermal diffusivities. Sand,
with the smallest thermal diffusivity, will get the
!
smallest heat influence from a heat source and will 6. Chain reaction of bulk Li-ion batteries

have slightest heat influence to the surrounding.

! in selected packaging materials

Another important observation needed to be
considered is the maximum battery surface peak
!
To be even closer to industrial practical conditions,
temperature (see Figure 13). One can see that this the thermal behavior of bulk of Li-ion batteries in
Tmax is significantly lower in sand than in other selected packaging materials has been investigated.

materials (sand: 372 °C, Absorbent: 512 °C,
PyroBubbles, Sorbix, vermiculite: above 550 °C). It
!
can be explained by the thermal conductivity of the 6.1 Experiment setup

materials. In sand, which has large thermal
conductivity, the heat from the battery is delivered fast !
to the surrounding materials and thus the cell surface In this test, 2 layers of 18650-type Li-ion batteries
temperature is limited.
(rated capacity: 2200 mAh) were placed within
! insulation/cushioning materials (Figure 14, a). In order
to simulate practical condition, batteries at different
The performances of the packaging materials are not
fully consistent with the sequence of their diffusivities SOC were used.

in Table 6. Sand and Absorbent showed consistent To create an original thermal-runaway battery, two
thermal performance with that in the table, while 100% SOC cells in the bottom layer have been ignited
Sorbix demonstrated better performance than by the heater described previously in 5.1. The heater
PyroBubbles and Vermiculite. PyroBubbles and has been switched off right after the two cells beside
Vermiculite showed very similar performance. started thermal runaway. Cells of different SOC were
Therefore, according to the comprehensive placed just next to these two thermal runaway batteries
performances, the practical thermal diffusivities can be to investigate the “chain reaction”. For further
in such sequence: Sand < Absorbent < Sorbix < verifying the effect of SOC, these cells were arranged
Vermiculite / PyroBubbles.
in different SOC order: right side―100% to 0%, left
! side―0% to 100% (see Figure14, b).

! Based on the status of the 2nd layer after the test, it
is possible to define a Minimum Safe Distance (MSD)
In conclusion, with the smallest thermal diffusivity
and largest thermal conductivity, thermal running for the packaging material. The tests have been
battery in sand exhibits lowest surface temperature and performed by increasing the distance between the 1st
smallest heat influence to the potential surrounding and the 2nd layer in a series of tests. In the first test the
batteries.
distance between the layers was positioned at 5mm.
! After the first experiment, the first layer has been
! checked that the thermal runaway took place and the
! second layer was examined regarding to its weight and
! voltage.
!
« 12
" »
 PACKAGING MATERIALS FOR LI-ION BATTERIES

If the batteries in the second layer lost their weight of the cells had been blown off. However, on the left
or voltage, they were noted as failed. In that case the side, the “chain reaction” has stopped at the 0% SOC
distance between the layers has been increased from battery.

5mm to 10mm and the test has been repeated. The !
enlargement of the distance in 5mm steps has been !
repeated until the second layer kept undamaged. In that
!
way a MSD could be determined for each insulation
!
material.

! !
! !
! !
! !
! !
!

!
! Figure 15: Effect of SOC on chain reaction

!
!
! In conclusion, while batteries with high SOC can

! easily lead to “chain reaction” of thermal runaway,

! fully discharged batteries can reduce the probability of

! a “chain reaction”.

!
! !
!
! 6.2.2 Minimum safe distance for Li-ion
! packaging materials

! !
Table 7 shows the result of MSD for all 5 packaging
!! (a)Total setup

materials. With only 10mm of MSD, sand shows

!! obviously lower safe distance and lower cell peak

temperature T1 than all other materials. Absorbent took
!! the 2nd position, with notable increasing T1. Sorbix,
Vermiculite and Pyrobubbles have significant higher
!! but similar minimum safe distance of 25-35 mm and
accordingly higher T1. This result is consistent with
!! previous discussion of their thermal diffusivities.

!
! Table 7: Minimum safe distances for Li-ion packaging materials

! ! No.
Packaging Safe distance T1* T2*

!
(b) Arrangement of 1st layer before test

Figure 14: Set-up for bulk Li-ion batteries’ thermal runaway test

1
material
Sand 10mm (±2mm) 526.8°C 111.8°C

! 2 Absorbent 15mm (±2mm) 662.2°C 115.1°C

! 3
4
Sorbix
Vermiculite
25mm (±2mm)
30mm (±2mm)
763.9°C
771.6°C
120.1°C
112.5°C
6.2 Experiment result
5 Pyrobubbles 35mm (±2mm) 724.0°C 87.7°C

! !
!
*T1: The highest temperature of batteries in 1st layer at thermal
6.2.1 Effect of SOC on chain reaction
runaway.

*T2: The highest temperature of batteries in 2nd layer in case of
Figure 15 shows an exemplary result of the effect of
SOC on chain reaction. The insulation/cushioning
material in this experiment was vermiculite. After the
! These experiments show only qualitatively the MSD
“safe distance’’.

experiment, the right side batteries were all failed (lost of packaging materials, which may change with the
weight), which means thermal-runaway “chain battery type. In other words, with batteries of higher
reaction” had taken place in this part and the electrolyte capacity or higher caloric battery components, the

« 13
" »
 PACKAGING MATERIALS FOR LI-ION BATTERIES

7. Summary and outlook

(1) When thermal runaway of a Li-ion battery
occurs, the battery surface temperature can raise
explosively to 340-690°C.

(2) The SOC of a Li-ion battery has great influence
on its thermal runaway behavior. Batteries of higher
SOC show more risk with lower thermal runaway onset
temperature and higher thermal runaway peak
temperature. While low SOC Li-ion batteries release
their highly flammable electrolyte, which also brings
hazard, but have no explicit explosive cell temperature
growth. Additionally, after thermal runaway, batteries
at high SOC have higher weight loss than that at low
SOC.

(3) With highest thermal conductivity and lowest
thermal diffusivity, sand is the best among the tested
insulation materials to prevent “chain reaction” of
batteries’ thermal runaway and hence can be considered
as the safest packaging material. Absorbent ranks after
sand. Sorbix, Vermiculite and PyroBubbles have
similar performance and rank last.

!
Summarizing, sand is an appropriate electric and
thermal isolating material for packaging of Li-Ion
batteries. It is easy to access, with economically most
attractive prime cost, easy to separate at sorting/
recycling facilities and due to low tendency of

!
absorption of impurities, it is also frequently reusable.

!
!
!

« 14
" »
 PACKAGING MATERIALS FOR LI-ION BATTERIES

References
!
! !
! !
[1] Q. Wang, P. Ping, X. Zhao, G. Chu, J. Sun, C. [12] http://www.klein-daemmstoffe.de/vermiculit-
Chen, Thermal runaway caused fire and explosion of vermiculite.html

lithium ion battery, J. Power Sources 208 (2012)

210-224
[13] http://www.vermiculite.org/

! !
[2] P.G. Balakrishnan, R. Ramesh, T. Prem Kumar, [14]http://www.brandschutz-passin.de/Oelbindemittel-
Safety mechanisms in lithium-ion batteries, J. Power Sorbix-Basic.html

Sources 155 (2006) 401-414
!
! [15]https://shop.maagtechnic.de/maag/Datasheets/
[3] Y. Chen, Z. Tang, X. Lu, C. Tan, Research of Sorbix-Basic_CH_ger.pdf

Explosion Mechanism of Lithium-ion Battery, J.
Progress in Chemistry, 18 (6) (2006) 823-831

!
! [16] http://www.sfm.state.or.us/CR2K_SubDB/MSDS/
! CAT_LITTER.PDF

!
[4] R. Spotnitz, J. Franklin, Abuse behavior of high-
[17]http://www.inchem.org/documents/icsc/icsc/
power, lithium-ion cells, J. Power Sources 133 (2003)
eics0384.htm

81-100

! !
[5] The European Association for Advanced [18] http://www.genius-patent.de/Genius/
index.php/product.html

Rechargeable Batteries, Safety of Lithium ion
batteries, http://www.rechargebatteries.org/lithium- !
ion-battery-safety/
[19] http://www.fire-shield.de/Fire/en/downloads/

!

[6] S. Wang, Y. Fu, L. Lu, X. Liu, Thermal simulation [20] http://en.wikipedia.org/wiki/
on temperature changes for lithium-ion cells, Chinese Thermal_diffusivity#cite_note-2

Journal of Power Sources, 2010, 01: 41-44+91

!! !
[21] Mark] http://en.wikipedia.org/wiki/
Heat_capacity#Table_of_specific_heat_capacities

[7] B. J. McBride, M. J. Zehe, S. Gordon, NASA
Glenn Coefficients for Calculating Thermodynamic
!
[22] http://www.engineeringtoolbox.com/thermal-
Properties of Individual Species, NASA Glenn conductivity-liquids-d_1260.html

Research Center, Cleveland, OH United States, Report
No.: NASA/TP-2002-211556, Sep 01, 2002

!
! [23] http://de.wikipedia.org/wiki/
! Temperaturleitfähigkeit
[8] T. Kawamura, A. Kimura, M. Egashira, S. Okada,
J.-I. Yamaki, Thermal stability of alkyl carbonate
mixed-solvent electrolytes for lithium ion cells, J.
Power Sources 104 (2002) 260–264

!
[9] http://www.engineeringtoolbox.com/thermal-
conductivity-d_429.html

!
!
[10] http://www.engineeringtoolbox.com/specific-
heat-capacity-d_391.html

!!
[11] http://www.engineeringtoolbox.com/density-
materials-d_1652.html

!
« 15
" »
 PACKAGING MATERIALS FOR LI-ION BATTERIES

Li-batteries: UN 3090 / 3091 / 3480 / 3481

is the battery tested according 38.3 UN-Manual ?

Yes No

damaged or defective ? transportation

only accord.
SV 310
No Yes
full ADR
SV 188 (original packaging) ? transportation
only with competent
authority confirmation
Yes No M 259
P 903 P 908/ LP904
full ADR
No used Yes
batteries?

<> 500g < 500g

P 903a P 903b

> 333kg < 333kg

SV 188 SV 230 SV 230 SV 636

no P 903 P 903a no
ADR full ADR full ADR ADR

Dienstag, 26. November 13

Annex I: Summary of relevant Provisions and Packaging Instructions  

« 16
" »

View publication stats