WEF Discussions — Now on LinkedIn!

WEF is pleased to announce the creation of free LinkedIn groups associated with its technical discussion forum topics. Through these LinkedIn groups, members will be able to view and participate in discussions, as well as communicate with one another via direct messages. Share your experiences and knowledge, ask questions and respond to other discussions as frequently as you like!

You must have a LinkedIn profile to join a LinkedIn group; you can create your profile here. If you already have a LinkedIn profile, all you need to do is submit a request to join one or all of the following groups:

Water Environment Federation (Main) | Biosolids | Collection Systems | Nutrients | Stormwater 
Utility Management | Water Reuse | Water for Jobs | Watershed Management | Laboratory Practices 

RSS Feed Print
BOD Bottle Method? Grad Cylinder Method? Hybrid?
Perry Brake
Posted: Wednesday, December 16, 2009 4:55 PM
Joined: 12/16/2009
Posts: 69

When assessing a WWTP lab last week, I found a situation where the BOD technician was adding a specific volume of concentrated buffer/nutrient solution (made by diluting the 6-liter pillows with only 200 mL of reagent grade water) rather than following the procedure for either the bottle method, or the graduated cylinder method.  After adding this known volume of buffer/nutrient solution, seed, and sample (and inhibitor if doing CBOD), the bottle was then filled with reagent grade water.  The objective was to make sure every bottle had the same amount of buffer and nutrients in it.


From a legal standpoint, doing this would certainly weaken if not destroy the defensibility of BOD data in a court of law because the lab isn't following the method.  But can anybody see anything wrong with it from a technical standpoint?  I think I can, but I won't taint your thought process with my own thoughts.



Gregg Mitchell
Posted: Thursday, December 17, 2009 8:43 AM
Joined: 9/21/2009
Posts: 6

The only way this would work is if the resultant solution strength in each bottle was the same as if making the dilution water properly.  It seems like a complicated way of doing a rather easy solution mixing and dispensing method.  SM prescribes the concentration of the nutrient buffers and how to prepare the dilution water.  Just make the dilution water as stated and dispense into the BOD bottles.  If there is a sample clean enough to use a large amount for the test, just add nutrient buffers to it at the same concentration as dilution water.  For instance, I always add nutrient buffer to my final effluent samples because I use 200, 250, and 300 mls of sample for the test.  Because these bottles get little to no dilution water added to them, this is the only way they will have the same concentration of nutrients as the dilution water itself.


Gregg M

Posted: Thursday, December 17, 2009 5:31 PM

I cannot see a technical argument against this: it seems to be a reasonable and valid approach to solving a problem.  Given the vagaries of the BOD test, the fact that this is not exactly per the method would seem to be a good idea.


However, we all know that in a court the evidence is based on what the jury or judge will understand. It might be difficult to make an average jury or judge understand the ins and out of buffer preparation and dispensing; but if a determined attorney who want to discredit the data sets off on a "Did you or did you not follow the written method exactly?" line of questioning he will eventually get the answer "No.", and the judge and jury can understand that.  I suppose that the overwhelming technical argument is: BOD is defined as what you get when you do A, B, C, D etc as written.  And if you missed step C, or messed with it, then it's not a BOD.

Keith Chapman
Posted: Tuesday, December 22, 2009 11:09 AM
Joined: 10/2/2009
Posts: 35

I thought you were retired, Perry!  No matter.  I like the way in which this person has IMPROVED the method by finding a way to better ensure that each sample has nearly the same amount of nutrients.  It would be way too much work for more than a few samples at a time, though.  We measure some BOD's on River samples by the "direct" method wherein we fill the BOD bottles with sample entirely (300 ml sample).  It has always bothered me that there is no added seed nor nutrients.  On the other hand, I think it is a valid way to measure oxygen demand of a river or stream.  Hope everyone has a good christmas.  Here is a little song to add to your festivity:


Christmas 2010


Oh B-O-D, Oh B-O-D

You are so biochemical

A little sample and then some seed

So plain yet diabolical

To set you up is such a breeze

Just five short days at 20 degrees

Oh B-O-D, Oh B-O-D

Pray the blanks don’t deplete on me.


First get the bottles all lined up

Then calibrate the meter

Aerate the water; prep the seed

And don’t forget the Winkler

Pipet your samples in “just so”

Make sure they have enough D.O.

Now load the samples on a cart

Incubate them five days in the dark.


If Wednesday morn you put them in

Next Monday all your woes begin

All the dilutions that you tried

Were either too low or else too high

All your controls are out of range

The calculations just too strange

Oh B-O-D you are so great             

The test that we all love to hate


Perry Brake
Posted: Tuesday, December 22, 2009 12:04 PM
Joined: 12/16/2009
Posts: 69

I DID retire, Keith, back in 2006, but I gave Dept of Ecology four years of my life by continuing to do lab assessments as a volunteer.  My four years were up last week and I have officially replaced my white gloves and note pad with a fishing pole!


Back to the nutrients/buffers issue.  I concluded that the lab tech's reasoning was flawed, and explained my position in the assessment report as follows:


[Explained difference between bottle and cylinder methods, followed by…]

The XXXXX lab uses a different technique that perhaps could be called a hybrid of the two previously mentioned procedures.  Concentrated solutions of buffer/nutrient (commonly referred to as “pillows”) are dissolved in 200 mL of water (normal dilution of those pillows is with six liters of water).  20 mL of the resulting solution are added to every bottle.  The rationale for doing this was stated to be a perceived need to get the same amount of buffer and nutrient reagents into each 300-mL BOD bottle.  However, the total volume of the bottle is not the proper criterion for determining how much nutrient/buffer solution is required.  Rather, it is the volume of the sample that is the key factor because it is the bacteria in the sample (and seed) that need the nutrients, and it is sample composition that determines the need for buffering.  To illustrate, and to simplify calculations, consider that the nutrient/buffer pillow contains reagents at a combined concentration of 10 milligrams per liter of sample, such that 20 mL would contain 0.2 mg of buffer/nutrient reagents.  Consider also that to be an effective nutrient/buffer solution, it must be present in the filled bottle at a concentration of greater than 1.0 mg/L of sample (a hypothetical concentration used only for illustration).  The table below shows that when the sample size exceeds 200 mg/L of sample, the amount of nutrient/buffer does not meet or exceed the threshold 1.0 mg/L of sample.


 Sample Size           Concentration of Buffer/Nutrients per liter of Sample


    6 mL                         0.2 mg/0.006 L = 3.33 mg/L of sample

    100 mL                      0.2 mg/0.100 L = 2.0 mg/L of sample    

    200 mL                      0.2 mg/0.200 L = 1.0 mg/L of sample

    250 mL                      0.2 mg/0.250 L = 0.8 mg/L of sample

    295 mL                      0.2 mg/0.295 L = 0.68 mg/L of sample



The table above clearly shows a declining amount of nutrient/buffer material per liter of sample for the larger sample sizes, and even though these numbers are hypothetical, they also show how at some sample volumes…>200 mL in the method…a special nutrient/buffer pillow is needed to assure sufficient nutrients and buffers are present in the bottle.


But the point I emphasized during the on-site assessment is that the lab must follow the method to assure legal defensibility of the data.

James Royer
Posted: Wednesday, December 23, 2009 7:20 AM
Joined: 9/21/2009
Posts: 98

I would not worry very much about any decline in nutrients in the BOD bottle for the BOD test. The only sample where there could be a problem is a low range standard. I have not seen a real sample where there was not more than enough nutrients already in the sample. The more sample the nutrients. We have to work hard to remove these nutrients.


Perry is right. The only thing is making defensable data. You do not want to get on a witness stand and try to explain what you did and what affect it had on the result. If you said it had no effect then they would wonder why you changed it then. They would then probably throw out the data. It is just easier to follow the standard procedure.


If there is a good reason to try something you should approach the Standard Methods Committee to try and have the change written into the method.



Perry Brake
Posted: Wednesday, December 23, 2009 10:24 AM
Joined: 12/16/2009
Posts: 69

Making sure there is enough buffer/nutrient available is important for all samples, but you need not do anything special for a low-range GGA.  The GGA bottle would have aproximately 290 mL of dilution water which has sufficient buffer/nutrient to do the job.  But is IS important to do something special for  low-range samples, such as a very clean effluent.  For example, if your effluent is so good that you need to (or would like to) use 290 mL of sample, after adding, say, 5 mL of seed, you are left with room for only 5 mL of dilution water.  That small volume of dilution water does not contain enough buffer/nutrient to keep the bugs happy for 5 days. That's why Standard Methods says whenever you exceed 200 mL of sample in the bottle, you must introduce the nutrient/buffers directly into the bottle and fill with reagent grade water.  Hach makes a special nutrient/buffer pillow for such use.

James Royer
Posted: Wednesday, December 23, 2009 11:52 AM
Joined: 9/21/2009
Posts: 98

If you have 290 ml sample in a BOD bottle, seeded, then the BOD of the sample could not be over 8 mg/L. At the ideal ratio of 200 BOD: 5 N: 1 P which makes for good treatment, then we only need 0.2 N and .04 P in the BOD bottle. I have not seen very many real samples that do not contain .02 N and .04 P except for some industrial waste.


The only organisms that are not happy are the nitrifiers as there is no ammonia. We do not want to see nitrification anyway as we are only nitrifing the added ammonia and that does not relate to the sample at all.


As for buffering, most all samples contain enough alkalinity to keep the samples buffered. Effluent that has completely nitrified might deplete all the alkalinity. The pH of this sample would be low so if you adjust the pH up then you might need the added buffering.


In reality, we only add the additional nutrients because the method calls for it.

Luis Manriquez
Posted: Thursday, January 7, 2010 4:33 PM
Joined: 1/6/2010
Posts: 20

Perry Brake is a great BOD info source and I am hesitant to go against his judgment. But I could not see the logic in his table listing the mg/l of nutrients per sample volume used.


All volumes set up for BOD are predicated on obtaining a depletion within the ranges of a minimum of 2 and a final DO of 1. So except for clean effluents all samples properly set up have a certain amount of DO demand which is in almost the same range, no matter how much or how litle sample is used. So the same one volume of nutrients/buffer can be used to cover this demand.


To illustrate this further, think about the different dilutions set up of a single sample such as 1, 3, 5 mls. The ratio of nutrients per sample volume is vastly different for each, right?


On the other hand, one could argue that the nutrients are meant only to create optimum conditions for the seed used. Since the same amount of seed is used in all samples, it is logical that the same equal amount of nutrients be used.

Another point could be made that the test conditions used for the GGA and samples have to be the same. This means that all bottles should get the same amount of nutrients/buffer regardless of the sample size introduced.


For clean effluents that are set up with more than 200 mls we simply add a  buffer pillow for 300 mls Hach Cat# 14160-66.


We also check and adjust the pH for non-routine samples prior to set up for BOD. I know that that is what the buffer is supposed to be for, but our auditors require that. We also bring all the samples to within 3 degrees of 20C prior to set up.


Having said all that, I agree that the method should be followed to the letter and if a variance is still desired it should be obtained in writing from your regulator.


Luis Manriquez


Perry Brake
Posted: Thursday, January 7, 2010 9:39 PM
Joined: 12/16/2009
Posts: 69

Luis, I have the utmost respect for your opinion, too...you have been in front of a bench covered with BOD bottles much longer than I.  But in your second paragraph, you say "... So except for clean effluent...one volume of nutrients/buffer can be used to cover this demand."  That was exactly my point...whenever you use more than 200 mL of sample, as you would for a clean effluent, you must, according to the method, use the special 300-mL pillow to make sure you have enough nutrient/buffer in the bottle.  And that is why you CAN'T just add a constant volume of concentrated nutrient/buffer for ALL bottles.  I emphasize the "ALL" because it is those clean samples that cause the problem.

James Royer
Posted: Friday, January 8, 2010 9:38 AM
Joined: 9/21/2009
Posts: 98

I think you are missing the point of the analysis. The result is proportional the the amount of biodegradable organics in the sample. The organisms will consume the oxygen in the bottle if they are nurtured with an excess of nutrients and the pH stays in a normal range.


Balancing nutrients to the organics is not needed as we only need an excess of the nutrients. Any effluent from a well operated WWTP will have suffecient nutrients in the sample to make the orgamisms happy. Most of the time the nutrients are the problem in the discharge.


I think that adding nutrients is included in Standard Methods just in case you are doing an analysis on a special chemical or industrial discharge that would be defecient in nutrients. Anyway, extra nutrients will not hurt the analysis.


Maybe the next addition of Standard Methods can clarify this? The 18th ed. required ammonia addition to stored dilution water and the 20th ed. does not. Ammonia nutrient should not be a factor as nitrification should be inhibited.

Perry Brake
Posted: Friday, January 8, 2010 10:37 AM
Joined: 12/16/2009
Posts: 69

I don't think I'm missing the point of the analysis.  You say "... if they are nurtured with an excess of nutrients and the pH stays in a normal range."

 It's that "excess of nutrients" that is exactly what you are NOT assured of when there is a large volume (Standard Methods says more than 200 mL) of sample that moght not have sufficient nutrients (the pH is probably OK, but possibly not) for the 5-day incubation.  If I am missing the point of the analysis, I guess Standard Methods is, also.   Oops...that's probably not a good analogy!

Luis Manriquez
Posted: Friday, January 8, 2010 11:16 AM
Joined: 1/6/2010
Posts: 20



Reviewing my response and yours, I feel I missed the point of the thread.

When we follow the SM BOD method as written we get a different amount of nutrients/buffer deppending upon how much sample is used.



For example, a sample set up with 3 mls sample has twice the amount of buffer/nutrients as a sample  set up with 150 mls. Then when we set up a 300 mls sample and add the 300ml buffer pillow we are back to almost the same concentration as the 3 ml sample.


SM does not specify that we use the 300 ml buffer pillow for undiluted samples. However it is an accepted part of the BOD anaysis.

On the other hand, if ALL the samples were prepared in a graduated cylinder first using dilution water and then a 300ml buffer pillow were added to each BOD bottle, I am sure this procedure would be found acceptable.



What the analyst in question was doing was the equivalent of adding a 300 ml buffer pillow to ALL the samples, except that he was using a 6 liter pillow dissolved in 200 mls. You did not mention what volume he used, but I calculate he should have been using 3 mls per bottle.


So that technique is better than SM's. But my comment is still valid: Get your regulator's approval first.



James Royer
Posted: Friday, January 8, 2010 11:23 AM
Joined: 9/21/2009
Posts: 98

We do add extra nutrients because Standard Methods says to add them. We know from other analyses that there is an abundance of nutrients in our plant effluent. I have not seen any real sample of effluent where there was a defeciency of nutrient to do BOD testing.


What are your thoughts on Luis's sig figures on 9 initial DO and .2 depletion?

Posted: Friday, January 8, 2010 2:35 PM
Joined: 12/31/2009
Posts: 40

Luis Manriquez wrote:


SM does not specify that we use the 300 ml buffer pillow for undiluted samples. However it is an accepted part of the BOD anaysis.




Actually it does specify.  5210B 5 c 2 states "When a bottle contains more than 67% of the sample after dilution, nutrients may be limited in the diluted sample and subsequently reduce bilogical activity.  In such samples, add the nutrient, mineral, and buffer solutions (¶s 3a-e) directly to diluted sample at a rate of 1 mL/L (0.30 mL/300 mL bottle) or use commercially prepared solutions designed to dose the appropriate bottle size."

Luis Manriquez
Posted: Friday, January 8, 2010 3:36 PM
Joined: 1/6/2010
Posts: 20

Dsmith: I stand corrected. Thank you.


While thinking about this buffer inequality problem, I realized that the instruction to seed the dilution water found in SM would also cause a serious discrepancy.

For example, when setting up samples for 3, 6, 9 mls and then add the pre-seeded dilution water we would be adding about the same amount of seed as the GGA bottle. But if we set up an effluent sample at say, 75, 150, and 300 mls, the 75 ml bottle will have 3/4 seed, the 150 will have 1/2 seed, and the 300 will have the full seed volume as it will have to be seeded separately.


This is kind of similar to the buffer problem and I never gave it a thought until Perry Brake stirred my brain.

Perry Brake
Posted: Friday, January 8, 2010 3:39 PM
Joined: 12/16/2009
Posts: 69

"...there is an abundance of nutrients in our plant effluent."


James, I'm sorry to hear that.  Your discharge permit manager must not take kindly to that!


James Royer
Posted: Friday, January 8, 2010 4:24 PM
Joined: 9/21/2009
Posts: 98

In the 21st ed, Std. Meth. seeding the BOD bottle or seeding the dilution vessel should have a constant amount of seed material so that the seed correction is constant. It does not mention seeding the dilution water.


As for nutrients, this is Illinois and the EPA is just starting to look at setting nutrient limits.

Luis Manriquez
Posted: Saturday, January 9, 2010 6:38 PM
Joined: 1/6/2010
Posts: 20

Dsmith and J. Royer:

I guess I need to get a copy of the 21st Edition SM.

Our permit states that we have to follow the older SM edition, so it does not change my limits.