Committee of Experts on the Transport of Dangerous Goods
and on the Globally Harmonized System of Classification
and Labelling of Chemicals

Sub-Committee of Experts on the Transport of Dangerous Goods

Forty-first session

Geneva, 25 June – 4 July 2012
Item 4 (a) of the provisional agenda

Electric storage systems: testing of lithium batteries

               T6 tests for lithium cells

                   Transmitted by the experts from China[1]

               Introduction

1.         At the thirty-eighth session of the Sub-Committee, the informal lithium battery working group submitted the working paper (ST/SG/AC.10/C.3/2010/81) that proposed the amendments section 38.3 of the Manual of tests and criteria concerning tests for lithium batteries and cells. In the paper, it proposed two tests for T6: impact and crush. Both tests simulate the mechanical abuse that may result in the internal short circuit to cells. The impact test applied to cylindrical cells not less than 20.0mmin diameter, and the crush test applied to the other cells.

2.         Chinese experts submitted informal document INF.35 at the thirty-eighth session, which proposed to apply impact tests for all the cylindrical cells. Experts in the meeting thought it was not practicable to apply the impact test to small cylindrical cells. Therefore, we suggested modifying the diameter limit from 20.0 mm to 18.0 mm at the meeting (ST/SG/AC.10/C.3/76, para. 40). At the thirty-ninth session, we submitted informal document INF.36, which proposed to modify the diameter to 18.0 mm. The Sub‑Committee suggested to submit an official proposal at the next session (ST/SG/AC.10/C.3/78, para.46).

3.         To illustrate the effects of impact and crush tests on cells, we compiled the test data for these two tests respectively, so that we can easily know the difference between crush test and impact test to the cells.

4.         First, we collected the impact test data on cylindrical cells, which included 261 groups of cells. Among these there were 221 groups of Li-ion cells with a diameter of 18.0 mm (18650), 14 groups of primary cells with a diameter of no less than 18.0 mm {[33.0 mm (33600) - 34.0 mm (341245、34615)], lithium-SOCl₂} and 26 groups of test data on primary cells with a diameter of less than 18.0 mm {[14.0 mm (14250、14500、14335) - 17.0 mm（17335、17500）], lithium-SOCl₂ and Lithium-MnO₂}. We found only one group of cells with the diameter of not less than 18.0 mm that could not pass the impact test, and the pass rate was more than 99%. For the primary cells with a diameter of less than 18mm, the rate was 65%. All the cells conducted to the test had an obvious temperature rise, for the smaller cells with a diameter of less then 18.0mm(14.0 mm-17.0 mm) which passed the test, the maximum temperature could reach 115 °C. Furthermore some cells caught fire and exploded. And the voltage of all the tested cells dropped to 0V. The data are shown in table 1 below:

Table 1：Impact test data

5.         Second, we chose 8 groups of cells with a diameter of less than 18.0 mm (14.0 mm and 17.0 mm) for crush test, 4 groups of cells for 14.0 mm and 4 for 17.0 mm. In the 4 groups of cells with a diameter of 14.0 mm, two groups of cells passed the above impact test and other two groups of cells failed. And it is the same for the cells with a diameter of 17.0 mm. All the 8 groups of cells passed the crush test. Also, we run the crush test on 32 groups of cells with a diameter of 18.0 mm including the group of cells which failed in impact test and 8 groups of cells with a diameter of 33.0 mm. Still, all the cells passed the test. Based on the test data, we could see that the crush test cannot make the cells temperature rise obviously, and there are no fire and no explosion. The data are shown in table 2 below:

Table 2：Crush test data

6.         Comparing table 1 and table 2, we note that the effects of impact test are more obvious than crush test for cylindrical cells. In other words, impact test is more effective to appraise the safety of cells during transportation than crush test. For safe transportation, we think the impact test is more reasonable for cylindrical cells, although we can understand that the experts at the thirty eighth session thought it is not practicable to apply the impact test to small cylindrical cells. We still think the cylindrical cells with a diameter of no less than 18.0 mm should conduct the impact test for the following reasons.

7.         The shell of cylindrical cells with a diameter of 18.0 mm is rigid. Crush test could not cause the internal short circuit as there is no temperature rise nor voltage drop, and there is only deformation according to the test results. On the contrary, there is 1 group of cells catching fire and exploding in the impact test we conducted. And the temperature of other cells rose to 85 °Cto 105 °C, the voltage dropped to 0V. It indicated the cells may have internal short circuit which correspond with the simulation purpose of this test. In addition, this type of cells are used broadly, the “18650” cells used in laptop computer also belong to this type. Keeping them safe is closely related to our life and very important. Therefore, we suggest that this type of cells shall be tested by more effective method: Impact test.

               Proposal

8.         Based on the discussion above, we suggest to amend the diameter limit in impact test scope, and read as below: (new text underlined, deleted text struck out)

“38.3.4.6.2Test procedure - Impact (applicable to cylindrical cells greater not less than 20 18.0mmin diameter)

38.3.4.6.3Test procedure - Crush (applicable to prismatic, pouch, coin/ button cells and the cylindrical cells not more less than 20 18.0mmin diameter)

Note: Diameter here refers to the design parameter (for example the diameter of          18650 cells is 18.0 mm). ”

联合国危险货物运输专家和全球化学品统一分类和标签制度专家委员会

危险货物运输专家分委员会

第四十一次会议

日内瓦, 2012年6月25日-7月4日
议程第4（a）项

电存储系统：锂电池测试

锂电池T6测试               

                   由中国专家提交 [2]

               介绍

1.               .在38次会议上锂电池非正式工作组提出了对UN38.3的修订草案（ST/SG/AC.10/C.3/2010/81），其中T6中有二种测试：碰撞和挤压， 这二种测试的目的都是模拟机械滥用而导致电池芯内短路，其中碰撞适用于直径不小于20mm的圆柱型电池芯，挤压适用于其它种类电池芯。

2.               在38次会议上,我们提交了UN/SCETDG/38/INF.35，提议将碰撞测试用于所有的圆柱形电池芯。但会上的一些专家认为碰撞测试用于小的圆柱形电池芯不现实，因此，我们当时建议将直径限制由20.0mm 改为18.0mm，见ST/SG/AC.10/C.3/76-40。在第39次会议上我们提交了UN/SCETDG/39/INF.36，提出建议将直径限制改为18.0mm。大会希望我们可以提交正式文件，见ST/SG/AC.10/C.3/78-46。

3.               为了更好地说明碰撞和挤压测试对电池芯的作用，我们分别整理了两种测试的测试数据，从而可以更直观的对两种测试的测试效果进行比较。

4.               首先，我们收集了一批对圆柱形电池芯的碰撞试验数据，共261组。其中221组是直径18.0mm的圆柱形锂离子电池芯（18650）的数据，14组是直径大于18.0mm[33.0mm(33600)-34.0mm(341245、34615)]的圆柱形锂金属电池芯（锂亚硫酰氯）的数据，26组是直径小于18.0mm[14.0mm(14250、14500、14335)-17.0mm（17335、17500）] 的圆柱形锂金属电池芯(锂亚硫酰氯和锂二氧化锰)的数据。在这批试验数据中，我们发现被测的直径大于或等于18.0mm的电池芯中只有一批没有通过测试，通过率是99%以上。小于18.0mm的电池芯的合格率则为65%。所有被测电池芯均发生了明显温升，特别是小于18.0mm的电池芯最高温度可达到115℃。此外，对直径小于等于18.0mm的被测电池芯都有着火及爆炸现象发生，所有被测电池芯电压均降至0V。具体见下表1：

表1：碰撞测试数据

5.               然后，我们选择了八组直径小于18mm的圆柱形电池芯做挤压测试，直径14.0mm和17.0mm各四组，直径14.0mm的四组电池芯中，二组通过了上面的碰撞试验，二组没有通过，直径17.0mm的四组电池芯也相同。我们发现这八组电池芯都可以通过挤压测试。我们又分别对32组直径为18.0mm 和8组 33.0mm 的电池芯做挤压测试，这些电池都做过碰撞测试，其中也包括碰撞中不合格的一批电池，结果都通过了测试。由数据可以看出挤压测试后的电池芯温升很不明显，也基本没有着火、爆炸等不良现象发生，电压降也基本没有。具体数据见下表2：

表2：挤压测试数据

6.               由表1和表2的数据比较可知：对于圆柱型电池芯，碰撞试验的测试效果明显要强于挤压测试。也就是说，碰撞测试对圆柱型电池芯可以更为有效的评价其安全性。因此，从安全性角度来说，我们认为对所有圆柱形电池芯适用碰撞测试是更为合理的。但在第38次会议上一些专家提出对于小的圆柱型电池芯用碰撞测试不现实，我们也表示理解。尽管如此，我们仍认为至少对于直径大于等于18.0mm的圆柱形电池芯仍应该用碰撞来进行安全测试，原因如下：

7.               直径18.0mm的圆柱形电池芯外壳很硬，挤压测试的结果是没有温升、没有电压降和只是有一点点变形，完全不能造成电池芯内短路，无法达到测试的目的。而对直径18.0mm电池芯的碰撞测试，有一批电池在试验中起火、爆炸，其他电池芯表面温度上升到85℃到 105℃之间，电压都下降到0V，说明圆柱形电池芯可能发生了内短路，与测试的目的相符。这一类的电池芯目前应用范围非常广，包括大批量笔记本电脑电池所用的18650电池芯，它们的安全性和我们的日常生活息息相关，保证其安全性十分重要。因此，我们希望可以用更有效的碰撞测试方法来做为该类电池安全和品质的保证。

提议

8.               基于以上说明，我们建议改变UN38.3中T6测试对圆柱型电池芯的直径限制条件，具体如下：

38.3.4.6.2    试验程序 – 碰撞 (适用于直径大于或等于18.0mm20mm的圆柱形电池芯)

        ......

38.3.4.6.3      试验程序 – 挤压（适用于方形，软包装，纽扣电池芯和直径小于等于18.0mm20mm的圆柱形电池芯）

注：这里的直径以电池的设计尺寸而定（如18650 电池设定直径为18.0mm）。