26 replies to this topic

[ScottSakura]

Hello everyone, I've been testing the pH of my noodle samples and also submitting the same samples to the lab to test.  The samples are long-shelf-life noodles, pasteurized and lactic-acid dipped.  When the lab gets the samples, it consistently reads significantly higher. I tested 3 samples, which read 3.62 pH, 3.62 pH, and 3.63 pH. 

 

I use the same pH meter as the lab (ThermoScientific Orionstar A111) and electrode.  The electrodes are changed out within the 1-year mark, and buffers every year when the electrode is changed out.  I was recommended to change out the buffers every 3 months as this may affect pH, so the lab thought my pH meter was not properly calibrated.  So, on this day, the lab came down to my facility to calibrate my unit with their buffers -- they verified that nothing was wrong with my unit, calibration was perfectly fine, and we proceeded with pH testing.  

 

When the samples that were sent to the lab were tested, the results were 3.78 pH, 3.75 pH, and 3.85 pH, respectively.  The only thing different I can think of is that the ambient temperature in the warehouse when I tested the samples was about 54°F, and I assume the lab's room temperature is around 76°F. But would this cause 0.13 to 0.23 pH increase of variance?  I read that pH typically lowers as the temperature rises. Would this not apply to acidified foods?

 

Prior to this, I did a 12-business day pH equilibrium test on my product.  I tested the same lot over 12 business days, and the pH didn't fluctuate much.  We usually wait a couple hours after the noodles are pasteurized before checking the initial pH reading.  My study found that waiting 2 hours after cool-down period and at the 12-business day mark, the pH stayed about the same.  pH reading initially was 3.75 pH and on the last day, it was 3.76 pH. The days in between, were about the same as well; 3.74, 3.77, 3.73, etc., not much in change.

I don't understand how the lab's test results consistently produces a higher pH reading. The lab places the blame on me even though my pH meter was verified by the lab and was calibrated using their buffers.

Could it be that all pH meter varies unit by unit? If not, how does one explain this issue?

 

Thanks in advance.
 

[G M]

Have you tried packing up two sets of samples, one actually goes to the lab, and the other gets tested by you after a comparable 'shipping' delay.  I'm wondering if the condition of the sample (wet, chilled but not frozen?) allows its chemistry to alter slightly during the shipping & handling phase.  

 

Maybe even ship it to yourself to cover whatever might be happening in between.

Edited by G M, 01 April 2025 - 05:57 PM.

[marco89634]

Hi, one important thing to consider, in my opinion, is the difference in ambient temperature: 54°F in your facility vs. presumably 76°F in the lab. It is true. pH decreases with increasing temperature, but this principle applies mostly to simple aqueous solutions and calibration buffers, not necessarily to complex food matrices. In your noodles, a complex matrix, where starches, proteins, lactic acid and structured water interact, the pH temperature coefficient becomes matrix specific and may actually behave in the opposite way, making a change of 0.1-0.2 pH units with a 12°C temperature increase plausible. The ATC automatic temperature compensation of the pH meter only corrects for the electrochemical response of the electrode, not for the temperature-induced chemical changes in the sample itself. In practice, in a complex system such as food, pH changes due to temperature changes can arise either from electrochemical effects, corrected by ATC, or from changes in the chemical equilibrium of the sample, not corrected by ATC. To definitively resolve this issue, I would suggest preparing six identical samples from the same batch; measure two samples at your standard temperature. Send two coded samples to the lab without revealing the purpose of the test; store the last two in conditions that simulate the lab environment and measure their pH at the same time as the lab analyses. This approach will methodically isolate the temperature effect from other potential variability factors, allowing you to accurately determine the temperature-pH correlation specific to your noodle and providing a solid scientific basis for correctly interpreting both sets of data.

[GMO]

Sorry if I'm asking stupid questions but obviously noodles are solid.  How are you preparing the samples?  

 

A while back I was watching someone do a titration in a food lab and saw some awful pipetting technique.  Sometimes it's not about the instrument but how the sample is prepared.  So what I'd do is do some traditional fishbone problem solving.  Work through all the inputs which could be wrong.  Sample prep could be one.  Temperature another.  Aging sample chemistry a third.  There's probably more.  Knock them all off one by one.  

 

Also consider the accuracy of your equipment.  Yes it's probably a bit more accurate than the differences you're seeing but not a hell of a lot.  So what is an acceptable tolerance?

 

Lastly, as ongoing verification, are you part of any proficiency schemes?  It's a great way of testing behaviours in your lab sampling indirectly down to the individual then coaching if there are gaps, or if there's a gap among everyone, recognising that there might be a kit issue.

[ScottSakura]

Have you tried packing up two sets of samples, one actually goes to the lab, and the other gets tested by you after a comparable 'shipping' delay.  I'm wondering if the condition of the sample (wet, chilled but not frozen?) allows its chemistry to alter slightly during the shipping & handling phase.  

 

Maybe even ship it to yourself to cover whatever might be happening in between.

 

 

I performed the 12-business day equilibrium test to mimic shipping and handling delays, but were relatively the same.  The noodle samples are wet, in an reduced-oxygen packaging.  

I'll have to try shipping myself a sample to see if storage/handling conditions during shipping changes anything.

Thanks for your input!

 

 

Hi, one important thing to consider, in my opinion, is the difference in ambient temperature: 54°F in your facility vs. presumably 76°F in the lab. It is true. pH decreases with increasing temperature, but this principle applies mostly to simple aqueous solutions and calibration buffers, not necessarily to complex food matrices. In your noodles, a complex matrix, where starches, proteins, lactic acid and structured water interact, the pH temperature coefficient becomes matrix specific and may actually behave in the opposite way, making a change of 0.1-0.2 pH units with a 12°C temperature increase plausible. The ATC automatic temperature compensation of the pH meter only corrects for the electrochemical response of the electrode, not for the temperature-induced chemical changes in the sample itself. In practice, in a complex system such as food, pH changes due to temperature changes can arise either from electrochemical effects, corrected by ATC, or from changes in the chemical equilibrium of the sample, not corrected by ATC. To definitively resolve this issue, I would suggest preparing six identical samples from the same batch; measure two samples at your standard temperature. Send two coded samples to the lab without revealing the purpose of the test; store the last two in conditions that simulate the lab environment and measure their pH at the same time as the lab analyses. This approach will methodically isolate the temperature effect from other potential variability factors, allowing you to accurately determine the temperature-pH correlation specific to your noodle and providing a solid scientific basis for correctly interpreting both sets of data.

 

Thanks for your input and the explanation of why the pH may increase. 

The lab tends to take various times from a 2-3weeks, to even 6-8 weeks.  The "lab" is CDPH-- California Dept. of Public Health.  So it's nearly impossible to test at the same time they do because they receive samples from all firms in the state of CA, and it would be hard for them to align timing with my timing.  They most likely have lab techs sitting there going through boxes and boxes of samples, testing them as they receive in order as fast an they can.

I did have a meeting with one of the higher-ups in Sacramento, along with the supervisor from the district office and the inspector via Microsoft Teams, but not once did they agree with the variables that can be on their end (or mine), like temperature.  They just said, pretty much, it's on me. So that's why I'm trying to figure out if I missed anything else on my end, or if it in fact, it is the temperature difference creating the variance.

I just wish the lab could accept this, instead of me running through hoops to figure out the variables on my end.

[ScottSakura]

Sorry if I'm asking stupid questions but obviously noodles are solid.  How are you preparing the samples?  

 

A while back I was watching someone do a titration in a food lab and saw some awful pipetting technique.  Sometimes it's not about the instrument but how the sample is prepared.  So what I'd do is do some traditional fishbone problem solving.  Work through all the inputs which could be wrong.  Sample prep could be one.  Temperature another.  Aging sample chemistry a third.  There's probably more.  Knock them all off one by one.  

 

Also consider the accuracy of your equipment.  Yes it's probably a bit more accurate than the differences you're seeing but not a hell of a lot.  So what is an acceptable tolerance?

 

Lastly, as ongoing verification, are you part of any proficiency schemes?  It's a great way of testing behaviours in your lab sampling indirectly down to the individual then coaching if there are gaps, or if there's a gap among everyone, recognising that there might be a kit issue.

 

Thanks for your input. I prepare the samples using a blender and distilled water to make a homogenous paste.  I measure the samples within 20-30 minutes of preparing the samples -- which the lab said it's fine.  If I prepped the samples hours prior, then it can possibly be an issue.

The acceptable tolerance is 0.2 pH.  As far as the person doing the in-house pH testing, it's only me. 

[ScottSakura]

Also, I did a small experiment to see if the amount of water used in the sample prep will affect the pH.  I found that using 100g of product with 20ml, 40ml, 60ml, 80ml, 100ml, 200ml, 500ml of DW did not change the pH.  This way, I can rule out the use of different amounts of water by me or the lab. 

Edited by ScottSakura, 01 April 2025 - 06:57 PM.

[dsglinski]

Could it be your distilled water vs the outside lab's?


Theoretically, freshly distilled water would be just about pH 7.0. This pH is likely to drop over time, especially in standing distilled water (such as if you use it from a 5 gallon container vs from a miliQ system).
 

I don't know if you've checked the pH of your distilled water source, but if its somewhat lower than 7, that could be your answer. Just a thought. 

[AltonBrownFanClub]

Also, I did a small experiment to see if the amount of water used in the sample prep will affect the pH.  I found that using 100g of product with 20ml, 40ml, 60ml, 80ml, 100ml, 200ml, 500ml of DW did not change the pH.  This way, I can rule out the use of different amounts of water by me or the lab. 

 

I went through a similar situation. Here are my suggestions to check:

• Have you confirmed the lab is also using distilled water? Seems obvious they would know to, but had to ask.
• How is your distilled water stored between tests? Are you using the 1-gallon plastic jugs? Are they sealed well?

I ask about your distilled water because the pH will change over time with exposure to CO2.

For example, since you are not doing as many samples as CDPH, it's likely your DW is exposed to air (CO2) for longer.

I assume since CDPH is doing so many tests every day, they use each container of DW quickly.

This would mean their water stays closer to 7.0, while yours could be as low as 5.0.

 

Here is a link that shows the math and effects of CO2 dissolved in distilled water. (pH change to 5-6.5)

Microsoft Word - pH_of_Distilled_Water 2.doc

 

I would try a fresh gallon of DW and compare the pH to the gallon you have already opened.

[ScottSakura]

Could it be your distilled water vs the outside lab's?


Theoretically, freshly distilled water would be just about pH 7.0. This pH is likely to drop over time, especially in standing distilled water (such as if you use it from a 5 gallon container vs from a miliQ system).
 

I don't know if you've checked the pH of your distilled water source, but if its somewhat lower than 7, that could be your answer. Just a thought. 

 

Thanks for your input.  Yes, our DW is about 7.0 pH.  I use gallon bottles of DW throughout the week, so it's always pretty fresh.

[ScottSakura]

I went through a similar situation. Here are my suggestions to check:

• Have you confirmed the lab is also using distilled water? Seems obvious they would know to, but had to ask.
• How is your distilled water stored between tests? Are you using the 1-gallon plastic jugs? Are they sealed well?

I ask about your distilled water because the pH will change over time with exposure to CO2.

For example, since you are not doing as many samples as CDPH, it's likely your DW is exposed to air (CO2) for longer.

I assume since CDPH is doing so many tests every day, they use each container of DW quickly.

This would mean their water stays closer to 7.0, while yours could be as low as 5.0.

 

Here is a link that shows the math and effects of CO2 dissolved in distilled water. (pH change to 5-6.5)

Microsoft Word - pH_of_Distilled_Water 2.doc

 

I would try a fresh gallon of DW and compare the pH to the gallon you have already opened.

 

Thanks for your input.

 

Yes, the lab is also using DW.  My DW is stored in gallon bottles, so fresh ones opened throughout the week.

Last time I checked the pH of DW, it was close to 7.0 pH, but just under.  I just checked again after reading few of the comments here, the one I have opened is closer to 6.2X, so I opened a new bottle and it's about 6.7X.  The pH meter was calibrated this morning as well.  How close to 7.0 pH should it be?  (My tap water is 7.4 pH)

Next time, I'll be sure to use a brand new bottle for testing to see if this changes anything.

Edited by ScottSakura, 01 April 2025 - 07:54 PM.

[dsglinski]

You could confirm exactly what the outside lab is doing (as in, are they standardizing their diluent) and/or request that they pH the distilled water they're using as a diluent.

For now, you could adjust your in house testing to include a pH reading of your distilled water, then adjust up to a 6.9-7.1 (for example) pH range, and used that pH adjusted water as your diluent for your noodle samples. This would at least give you a standardized diluent pH for comparisons sake, and should be closer to the outside lab's distilled water, assuming they're closer to pH 7. 

Edited by dsglinski, 01 April 2025 - 08:09 PM.

[jfrey123]

The lab tends to take various times from a 2-3weeks, to even 6-8 weeks.  The "lab" is CDPH-- California Dept. of Public Health. 

[snip]
I just wish the lab could accept this, instead of me running through hoops to figure out the variables on my end.

 

I think I found the root cause, though it could just be my personal grievances with CA government in general misleading me to suspect they're not that great at running a lab.

 

In your shoes I would take a series of triplicate samples:  test one series yourself, send another series to CDPH, and send the third to a certified lab.

 

In your OP you mentioned CDPH calibrated your equipment, and again I would be suspect of this and want to get my equipment evaluated by the manufacturer if the independent certified lab results align with CDPH.  If the independent lab's result aligns with yours, I'd repeat the test a few more times and then compile that for a complaint/conversation with their superiors.

Edited by jfrey123, 01 April 2025 - 08:20 PM.

[ScottSakura]

You could confirm exactly what the outside lab is doing (as in, are they standardizing their diluent) and/or request that they pH the distilled water they're using as a diluent.

For now, you could adjust your in house testing to include a pH reading of your distilled water, then adjust up to a 6.9-7.1 (for example) pH range, and used that pH adjusted water as your diluent for your noodle samples. This would at least give you a standardized diluent pH for comparison's sake, and should be closer to the outside lab's distilled water, assuming they're closer to pH 7. 

 

Thanks for your input.

Curious, I read that lab-grade DW is more pure, etc., but do you know how much that would affect pH reading? For example, would the lab-grade DW be almost a perfect 7.0X pH versus my store bought DW at 6.7X?

(pH meter calibrated with a slope of 100.3%).

[GMO]

Thanks for your input. I prepare the samples using a blender and distilled water to make a homogenous paste.  I measure the samples within 20-30 minutes of preparing the samples -- which the lab said it's fine.  If I prepped the samples hours prior, then it can possibly be an issue.

The acceptable tolerance is 0.2 pH.  As far as the person doing the in-house pH testing, it's only me. 

 

At least the person variable is taken away.  Sorry but it's a bugbear.  Their lab is using a pipette properly aren't they in making the sample?  Volumetric "bulb" pipettes are the most accurate.  Microbiological ones are inaccurate as hell.  I assume you are pipetting?  And using volumetric flasks?  Or are you using weight measurements?  Basically what I'm hinting at is the error in all of the processes to make the sample.  The reason why I think that's worth looking into is that firstly it's common to have errors in that where people think they're not relevant (for this kind of small error they are) but secondly if you've checked both lab and your pH probes with the same buffer solutions, perhaps it's an error somewhere else?

[kingstudruler1]

Thanks for your input.

 

Yes, the lab is also using DW.  My DW is stored in gallon bottles, so fresh ones opened throughout the week.

Last time I checked the pH of DW, it was close to 7.0 pH, but just under.  I just checked again after reading few of the comments here, the one I have opened is closer to 6.2X, so I opened a new bottle and it's about 6.7X.  The pH meter was calibrated this morning as well.  How close to 7.0 pH should it be?  (My tap water is 7.4 pH)

Next time, I'll be sure to use a brand new bottle for testing to see if this changes anything.

 

 

Did your try and prepare the same sample with each of these and test?   My guess it the result would be different for each.  That might help confirm or eliminate the PH of water as the issue.   

 

Purchasing expensive lab water will not help, even HPLC grade water will decrease in PH over time.   However, in not sure purchasing store water is ideal either. 

 

If you do find that the ph of the distilled water is the issue, you could adjust as others suggested or purchase distiller or other lab water sysem for on demand use.  not usually inexpensive.   

 

I also second the idea of triplicate samples - with one going to a new third party lab.  

[ScottSakura]

At least the person variable is taken away.  Sorry but it's a bugbear.  Their lab is using a pipette properly aren't they in making the sample?  Volumetric "bulb" pipettes are the most accurate.  Microbiological ones are inaccurate as hell.  I assume you are pipetting?  And using volumetric flasks?  Or are you using weight measurements?  Basically what I'm hinting at is the error in all of the processes to make the sample.  The reason why I think that's worth looking into is that firstly it's common to have errors in that where people think they're not relevant (for this kind of small error they are) but secondly if you've checked both lab and your pH probes with the same buffer solutions, perhaps it's an error somewhere else?

 

Thanks for your input.  The lab said they put 100g of product in a plastic bag of some sort, and use a garlic press to smash the noodles.  No pipettes are used, I don't use them as well.  After I blend the noodles and DW in a blender, I pour them in a 100ml cup.  

I'm not sure of the brand of the buffer solution that the lab uses, but the lab came down to use their buffers on my pH meter to calibrate and the results were still off, as usual. 

[ScottSakura]

Did your try and prepare the same sample with each of these and test?   My guess it the result would be different for each.  That might help confirm or eliminate the PH of water as the issue.   

 

Purchasing expensive lab water will not help, even HPLC grade water will decrease in PH over time.   However, in not sure purchasing store water is ideal either. 

 

If you do find that the ph of the distilled water is the issue, you could adjust as others suggested or purchase distiller or other lab water sysem for on demand use.  not usually inexpensive.   

 

I also second the idea of triplicate samples - with one going to a new third party lab.  

 

Thanks for your input.  I also bought lab-grade deionized water to test, and it didn't make a difference in the results.

I'll have to test again, using the different DW bottles I have right now (different pH levels) and see the results.

[GMO]

Thanks for your input.  The lab said they put 100g of product in a plastic bag of some sort, and use a garlic press to smash the noodles.  No pipettes are used, I don't use them as well.  After I blend the noodles and DW in a blender, I pour them in a 100ml cup.  

I'm not sure of the brand of the buffer solution that the lab uses, but the lab came down to use their buffers on my pH meter to calibrate and the results were still off, as usual. 

 

Ooh now this is getting more interesting.   Blending vs. stomaching or pressing the sample may not get the same result.  In micro labs they normally use things called "stomacher bags" which mush up the sample.  This kind of thing:

 

Stomacher® 400 Circulator Lab Blender | Seward

 

I certainly think you should remove that variable and prepare the sample the same way.

[ScottSakura]

I'm losing my mind. I just ran a few tests with the ThermoScientific meter I mainly use, and I have a Mettler-Toledo and Apera Instruments units as back ups, which I also used for testing.  The readings for DW is all over the place.  The variance between all 3 units is huge.
 

​Note:The electrode for the TS is just over 2 months old, the MT electrode is about the same age, the AI unit is the oldest, about 1.5 year old.

 

I just ran a few tests, first testing the DW.  The pH dropped in pH significantly from yesterday's readings of 6.2X pH (old DW), 6.7X pH (new DW)--both tested on the TS unit.

 

                  (TS)         (MT)        (AI)

Old DW:  5.43 pH,  7.58 pH,  8.96 pH

New DW: 5.42 pH  7.50 pH,  8.73 pH

 

                                                                                        (TS)     (MT)   (AI)

They were all calibrated at the same time, slope read 100.3%, 103%, 98%. 

 

I tested the same sample of noodles and here are the numbers I got:

                (TS)       (MT)        (AI)

Old DW:  3.52 pH,  3.67 pH,  3.72 pH
New DW:  3.54 pH,  3.71 pH,  3.69 pH

 

Although the readings for the DW water were all over the place, the noodle sample variance was much less. 

What do you guys make of this?

 

[ScottSakura]

Ooh now this is getting more interesting.   Blending vs. stomaching or pressing the sample may not get the same result.  In micro labs they normally use things called "stomacher bags" which mush up the sample.  This kind of thing:

 

Stomacher® 400 Circulator Lab Blender | Seward

 

I certainly think you should remove that variable and prepare the sample the same way.

 

Thanks for your input.  After having a virtual meeting with the lab, they did mention that it shouldn't matter how the sample is prepared whether using a garlic press (what the lab uses) or a blender, as long as the noodles itself ends up in a homogenous paste or evenly and thoroughly blended through.  They believe it's something else.

Also, I'm not sure if I want to spend thousands of dollars, and end up with the same results as the lab didn't mention they use that. 

Edited by ScottSakura, 03 April 2025 - 04:11 PM.

[GMO]

Thanks for your input.  After having a virtual meeting with the lab, they did mention that it shouldn't matter how the sample is prepared whether using a garlic press (what the lab uses) or a blender, as long as the noodles itself ends up in a homogenous paste or evenly and thoroughly blended through.  They believe it's something else.

Also, I'm not sure if I want to spend thousands of dollars, and end up with the same results as the lab didn't mention they use that. 

 

No, I get it, I wasn't suggesting you spend a tonne.  But what about taking your pH kit to the lab and testing the sample in their lab with their method?  Or finding out what their exact method is and replicating it (probably something far cheaper).  I'm not so sure that blending vs. crushing wouldn't give you a difference.  The pH measurement will only measure what is in solution, not what is still solid.  So to my mind, blending is far more likely to get more into solution than crushing.  I would absolutely 100% rule it out.  You're not looking for a big difference here.  

[ScottSakura]

No, I get it, I wasn't suggesting you spend a tonne.  But what about taking your pH kit to the lab and testing the sample in their lab with their method?  Or finding out what their exact method is and replicating it (probably something far cheaper).  I'm not so sure that blending vs. crushing wouldn't give you a difference.  The pH measurement will only measure what is in solution, not what is still solid.  So to my mind, blending is far more likely to get more into solution than crushing.  I would absolutely 100% rule it out.  You're not looking for a big difference here.  

 

Thanks for your input.  

 

I did think about doing that; taking my unit to the lab to have them test with it and using their method.  It's about a 6-7 hour drive, so one day when time permits, I may have to do that.

 

You're right on the cheaper part, if I try using what they use. A garlic press and placing it in a bag and adding DW (I forgot what the bag was called, but it was similar to a zip lock design).  I guess that's worth a try even though the lab said it shouldn't make a difference.  But, you're saying that blending in DW will cause the pH to be lower than using a garlic press and adding DW?  I'll get a garlic press and try that out, definitely worth a shot.

[GMO]

I always think anything is worth a try.  And in an old factory I worked in, when we were doing RCA the first question was "are we to standard".  I LOVED that question because it just forces you to think of all the simple stuff first.  

 

I'm also reminded of an issue where someone in ops had an idea about where we had a micro problem in our plant.  I looked at it went "pah!  No way!"  And I was very wrong, once we'd spent more time and more money and effort ruling out everything else.  It was a good reminder to just try things sometimes.  If nothing else you can at least prove to any superiors, auditors etc that you've been super thorough.

[ScottSakura]

I always think anything is worth a try.  And in an old factory I worked in, when we were doing RCA the first question was "are we to standard".  I LOVED that question because it just forces you to think of all the simple stuff first.  

 

I'm also reminded of an issue where someone in ops had an idea about where we had a micro problem in our plant.  I looked at it went "pah!  No way!"  And I was very wrong, once we'd spent more time and more money and effort ruling out everything else.  It was a good reminder to just try things sometimes.  If nothing else you can at least prove to any superiors, auditors etc that you've been super thorough.

 

Yes, you're right, it's worth a shot, especially with a garlic press being such a cheap item.  Thanks!  Also, would you happen to know why my 3 different pH meters gave out reading that are so far apart even they were all calibrated at the same time?