Battery Outgassing Math

Quote of the Day

Madam, if you were my wife, I'd drink it!

- Winston Churchill's response to Lady Astor, who had said to him "If you were my husband, I'd poison your tea." Churchill and Lady Astor were famous for their feuding.


Introduction

Figure 1: Aftermath of a Hydrogen Gas Explosion in a Battery Vault.

Figure 1: Aftermath of a Hydrogen Gas Explosion in a Battery Vault.

I recently have received a number of questions about the outgassing of hydrogen gas that can occur from lead acid batteries when they are being overcharged. I thought it would be useful to review what is happening when a battery is outgassing. When being charged, batteries can release enough hydrogen gas to create an explosive hazard. Consider this report and Figure 1 as an example as to what can happen. With lead-acid batteries, hydrogen gas can be generated any time, but charging is when the greatest challenges are faced.

Background

Basic Chemistry

When you think about it, a lead-acid battery being charged looks a lot like a water electrolysis setup. You can split water into its hydrogen and oxygen components by applying an electrical potential greater than 1.48 V to water. Figure 2 shows a typical electrolysis setup (source).

Figure 2: Example of an Electrolysis Setup.

Figure 2: Example of an Electrolysis Setup.

Figure 2 also describes a battery being charged -- there are two terminals that are separated by water (plus some H2SO4) and the terminals have a voltage applied to them. Figure 3 shows a cross-section diagram of a lead-acid battery. Figures 2 and 3 are very similar.

Figure 3: Battery Cross Section Diagram.

Figure 3: Battery Cross Section Diagram.

Oxygen is generated at the positive terminal and hydrogen is generated at the negative terminal. Since we normally charge lead-acid batteries at a potential higher than 2.2 V, we always get some electrolysis along with the charging. That is why some batteries need to have their water replenished frequently. Those that do not need to have their water replenished incorporate some mechanism for gas recombination (see AGM and Gel Battery).

Equation 1 shows the basic electrolysis reactions.

Eq. 1 \displaystyle 2{{H}_{2}}O\to {{O}_{2}}\uparrow +4H+4{{e}^{-}} reaction at the positive electrode
\displaystyle 2{{H}_{2}}O+2e\bar{\ }\to ~{{H}_{2}}\uparrow \text{ }+\text{ }2\left( OH \right)\bar{\ } reaction at the negative electrode

How to Abuse a Lead Acid Battery

There actually are standards for how to abuse a battery (i.e. force it to runaway and outgas). The one I am most familiar with is the Induced Destructive Overcharge Test in IEC standard 952-1:1988. Here is a useful reference that discusses how to perform the test.

Example of Battery Abuse

It does not take much searching on the web to find examples of a lead acid battery that has undergone thermal runaway. Figure 4 shows an example of battery damage as the result of thermal runaway. I have seen some cases where the battery case got so hot that it melted.

Figure 4: Example of an Battery That Has Exploded (Wikipedia).

Figure 4: Example of an Battery That Has Exploded (Wikipedia).

Remember that overcharging, outgassing, and thermal runaway are all related. Thermal runaway is not a good thing to have to happen.

Analysis

Battery Example

The discussion that follows is about the Panasonic LC-127R2P 12V/7.2Ah, which is a very commonly used sealed lead-acid battery. I show this battery in Figure 5. This battery is composed of 6 cells connected in series. Each cell nominally generates 2.0 V, but the exact voltage varies with the batteries state of charge and can be anywhere from 1.75 V to 2.25 V when discharging. As this battery is filled with Sulfuric acid, this is a dangerous chemical for anyone to deal with, especially if it is released into the atmosphere. With this being said, it may be worth checking out a site like Storemasta, in the hopes of finding out more about chemical like Sulfuric acid.

Figure 5: Sealed Lead Acid Battery Example.

Figure 5: Sealed Lead Acid Battery Example. This is an AGM (Absorbed or Advanced Glass Mat) battery. The mat is sandwiched between the plates and is saturated with sulfuric acid. The mat retains the acid in the event of a breakage, making the battery spill-proof. You can also get AGM Solar Batteries.

It really is a workhorse product -- I have never had any issue with it. Like all lead-acid batteries, you just need to treat it nicely.

Rate of Outgassing

I thought it would be a good exercise to show how much hydrogen and oxygen a battery can generate. Note that most references focus on the generation of hydrogen because that is the gas that is flammable. IEEE 484 is the standard governing the installation practices for lead-acid batteries and it states that

5.4 Ventilation

... Maximum hydrogen evolution rate is 0.127 mL/s per charging ampere per cell at 25 °C and standard pressure (760 mmHg). The worst-case condition exists when forcing maximum current into a fully charged battery. ...

The hydrogen evolution rate is important to know because the Lower Explosive Limit (LEL) concentration for H2 gas is 4%. Knowing the rate of hydrogen gas generation and the volume of the battery enclosure allows us to determine the amount of ventilation required for an explosion-proof installation. We can show where this result comes from applying a bit of basic chemistry. Figure 6 shows my derivation. In this derivation, I show how to compute the H2, O2, and total gas (H2 and O2) generation rates per cell.

Figure 6: Derivation of IEEE 950 Value for H2 Gas Generation Per Amp of Current.

Figure 6: Derivation of IEEE 484 Value for H2 Gas Generation Per Amp of Current. A Total Gas Volume Value is Also Generated For Comparison with Panasonic Data in Figure 7.

Using the derivation of Figure 6, Equation 2 states the complete equation for the total gas generation rate (O2 and H2) from a battery composed of multiple cells.

Eq. 2 {{R}_{Gas}}={{R}_{O2}}+{{R}_{H2}}=11.4\frac{\text{mL}}{\text{minute}\cdot \text{cell}\cdot \text{Ampere}}\cdot {{N}_{Cells}}\cdot {{I}_{Charge}}

where

  • RGas is the total gas generation rate.
  • NCell is the number of cells in the battery.
  • ICharge is the charging current.

While Equation 2 is stated for the total gas generation rate, the same basic equation form holds for O2 and H2 individually, just change the constant term from 11.4 mL/(min·A·cell) to:

  • 7.6 mL/(min·A·cell) for H2 generation only
  • 3.8 mL/(min·A·cell) for O2 generation only

Empirical Data

Figure 7 shows the outgassing graph for the Panasonic LC-127R2P 12V/7.2Ah Sealed Lead Acid Battery. This graph shows the total amount of gas generated, which means both O2 and H2. You can see from the handwriting on the graph that this is not an official graph -- I got this from an electrochemist with Panasonic who had measured the outgassing characteristic. The little black arrows on the graph indicate which axis the individual curve corresponds to. The x-axis is stated in units of "CA". CA describes the charging current as a fraction of the A-hr capacity of the battery (the A-hr capacity is treated as a current value). Using the 7.2 A-hr battery for an example, 0.1 CA = 0.1·7.2 A = 0.72 A charging current.

Figure 7: Hydrogen Gas Emissions from a 7.2 A-hr Sealed Lead Acid Battery.

Figure 7: Hydrogen Gas Emissions from a 7.2 A-hr Sealed Lead Acid Battery.

Figure 7 deserves one other comment -- I have no idea why the battery vendor's electrochemist used a logarithmic x-axis. Gas generation is linear with charging current. The use of a logarithmic axis makes it look like something nonlinear is going on. In the following discussion, I will take the data and re-plot it on a linear graph (Figure 8).

Theoretical Versus Empirical

Figure 8 shows a comparison of the measured gas generation rate versus the predicted gas generation rate. Notice that the theoretical rate is higher than the measured rate. This is because 100% of the charging current into the battery does not go into electrolysis -- some must go into charging. To recover the water lost because of electrolysis, sealed lead acid batteries contain one of two types of gas recombining technology which will ensure that low levels of generated gas will be recombined into water. However, the theoretical rate is useful because it established an upper bound for the amount of H2 that can be generated.

 Figure 8: Comparison of Theoretical Versus Empirical Gas Generation Rates.


Figure 8: Comparison of Theoretical Versus Empirical Gas Generation Rates.

Conclusion

Lead-acid battery outgassing is one of the least understood characteristics of this chemistry. Hopefully this note helps explain how outgassing works and how to estimate the amount of hydrogen generated.

References

Good Paper on Battery Outgassing
Ventilation Example

Save

Save

Save

Save

Save

Save

Save

Save

Save

 
This entry was posted in Batteries, Electronics. Bookmark the permalink.

18 Responses to Battery Outgassing Math

  1. Fritz says:

    Nice; thanks for providing this!

     
  2. PJ says:

    Thank You

     
  3. Dan M. says:

    Figure 6 makes reference to "IEEE 950". I searched for this, but can't find it. What is IEEE 950?

     
    • mathscinotes says:

      That is a typo and it should read IEEE 484. The derivation refers to the quote from IEEE 484 just above the derivation. I know where my inadvertent "950" came from. I work all day with GR-950 from Telcordia and I sometimes use it when I mean something else. I have to watch that.

      mathscinotes

       
      • Dan M. says:

        Thanks so much for clearing that up for me. I wanted to plug this into a spreadsheet, but wanted to double check some of the IEEE standards I have.

         
  4. RJ says:

    Can you please explain how to calculate the charging current to be used in equation 2.
    Can this formula be used I = Power (in Watts)/Voltage

    Voltage in this case would be the sum of the voltages of all individual cells?

     
    • mathscinotes says:

      Hi RJ

      The voltage across the battery is the sum of the voltages of the individual cells. It is also true that power = current * voltage (P = I*V).

      Battery manufacturers often specify the charging time for their batteries in terms of current. This is why we often see batteries charged from current-limited power sources. Electrically, the chargers act like current sources – at least until battery voltage rises to desired "full" cell voltage (2.2V to 2.4 V per cell for a lead-acid battery).

      mathscinotes

       
  5. Tim Hughes says:

    Excellent article, although your description of gas "absorbtion" is a little misleading, as to what really is happening.
    The Panasonic batteries you are using here, are what are termed "starved electrolyte recombinant" lead acid batteries (similar technology larger format batteries are often termed "AGM" = Advanced Glass Matt).
    It is a bit misleading to say gases are "absorbed" at low overcharge rates.
    In starved electrolyte cells, there is a very short *gas path* between the positive and negative plates, that allows the gasses to rapidly diffuse through the glass matt separator, which is barely damp with electrolyte ("starved"). This allows almost 100% gas recombination at low overcharge rates and very low water loss. There is usually still a very,very tiny H2 loss. The fine glass fiber matt, is important, because the glass helps the transport and recombination process. The panasonic batteries are "low pressure" recombinant batteries. That is the bunsen valve they use, opens at a pressure of 1.5-3psi, at higher levels of overcharge , at which stage the cells recombination efficiency is lower, and they lose electrolyte and after relatively small water loss, they lose capacity.
    There are also high pressure (15-25psi) recombinant Pb batteries, which can take much higher levels of overcharge,before venting takes place (usually with Jelly roll plates and cylindrical cell cases) . Enersys Cyclon batteries (Originally "Gates Energy Products"), are examples of this type. The other common sealed Pb battery types are "Excess electrolyte" . These are gelled cells (eg Sonnenschein) which contain ~30% excess electrolyte and routinely lose electrolyte under float charge. Initially the loss rate is high,but as they "dry out" they develop micro channels in the gell, that allows gas diffusion between plates. Recombination can then take place, although recombination efficiency is lower than starved cells Again the gell contains fumed silica, which helps transport and recombination and the pressure valves are low pressure.

     
    • mathscinotes says:

      Thank you for such wonderful information! The electrochemist at Panasonic mentioned similar things, but I struggled as a non-expert in this area to best describe what was happening. I am traveling now, but I shortly I will update the post to (1) remove the term "absorbed" and (2) refer my readers to your comment.

      Again, thanks for a very useful comment.

      mathscinotes

       
  6. Gavin says:

    I have a LC-R12R2 battery and are charging it using a Meanwell PB-120P-13C charger. Output of charger is 13.8v - 7.2a. Could this charger ruin or reduce the battery life of my battery.

     
  7. Pingback: Temperature-Compensated SLA Battery Charging Voltage | Math Encounters Blog

  8. Pingback: Battery Room Ventilation Math | Math Encounters Blog

  9. Kiran says:

    @mathscinotes.. Thanks !!! Very informative article to give idea about gas evolved during charging and how important it is to have proper room ventilation.Just wondering is the calulation same for old Plante type cells and new VRLA battery ? from what Tim Hughes ( Thanks to him too for his explanation!!)has explained it seems there shall be less hydrogen liberated for VRLA where AGM separators are used.Is there any thumb rule for rough calculation of room size and required air changes ?

     
    • mathscinotes says:

      Hi Kirian,

      Tim is correct that less hydrogen is liberated from a VRLA than the Plante cells. You can see that in Figure 8, where both the theoretical curve and measured gas emission levels are plotted. At low current levels, the AGM emits nothing because the battery glass mat can recombine at the rate of generation.

      The Plante cells pretty much follow the theoretical curves, minus whatever gas emission that did not occur because some current goes into residual charging. From an engineering standpoint, everyone that I know sizes their HVAC to ensure that they can evacuate the theoretical maximum amount of gas emission, i.e. they ignore recombination and residual charging.

      I have not seen a rule of thumb for air changes requirements for battery gas evacuation. I will think about – it seems like there should be.

      mark

       
  10. Tim Hughes says:

    "Tim is correct that less hydrogen is liberated from a VRLA than the Plante cells."
    I believe at very low overcharge rates part of the reason there is still a tiny hydrogen loss maybe because it is a smaller molecule and so harder to contain and diffuses faster than oxygen, so some escapes the case before recombination. The remaining oxygen probably either vents if pressure builds or oxidizes plate.

     
    • mathscinotes says:

      Hi Tim,
      I certainly can believe that some hydrogen can escape prior to recombination. I have some evidence for ability of gasses like hydrogen and helium to escape. I did some testing a few years ago on battery enclosures where we used helium to simulate hydrogen -- we did not want to risk explosion. As part of the testing, we tried to seal up the enclosures so that no gas could escape except through the ventilation system. I quickly found out there was no way to stop the helium from escaping at low levels from a sealed enclosure.

      mark

       
  11. Jerome says:

    Good morning, thank you for this publication, very insightful
    I am based in France and striving to understand how electrolysis works for vented lead acid batteries - following an explosion we had at one location. And also what the ventilation needs are.
    I am confronted with 2 different standards :
    - European Standard EN 50272-2 uses a formula to estimate ventilation needs based on a current (boost or floating) which could be less than the actual charging current and I do not understand why. They use Igas = 5 mA/Ah/cell in floating and 20 mA/Ah in boost mode to estimate the level of H2 released by one cell of a vented lead acid battery. Then multiply by the battery capacity C to derive the charging current generating gases. Then multiply by the number of cells and batteries and a factor 0.05 to get the minimum room ventilation required in m3/h. Q(m3/h)=0.05xNcellsxIgasxC
    - the other standard (French one) is more stringent since they estimate the ventilation flow needed by using the full charging current. For instance, in the case I am looking at, the charging current is 15A for 220 Ah, which yields I gas = 68 mA/Ah.

    For safety reasons I am enclined to use the most stringent values, but in reality, which standard is right according to you ?

     

Leave a Reply

Your email address will not be published. Required fields are marked *