Ph Change (drift) during mashing.

HomeBrewTalk.com - Beer, Wine, Mead, & Cider Brewing Discussion Community.

Help Support Homebrew Talk:

Nordic Brewer

Member
Joined
Feb 21, 2018
Messages
8
Reaction score
0
Ph Change (drift) during mashing.

In different thread discussions, pH-drift has been explained as:
- pH meter is drifting or pH meter instability
- It takes time to reach chemical equilibrium in the mash, measurement taken to early.
- Or there may be some mash chemistry that is not taken into account in the calculators and spreadsheet used by many brewers.



I have not given much though to this issue until I did some water-adjustment and pH readings myself and experienced the same phenomenon.
Here is the result from my last brewday.
Later I found a scientific paper describing the same phenomenon. A discussion around that is found after my description of my pH readings.



Preconditions:
1. NO SPARGE, total water volume 30.2 litre
2. Temperature step infusion mashing with Speidel Braumeister equipment (electrical heating together with in-build water pump and temp regulator).
3. Source Water profile:
- Ca++ 6.25 mg/l
- Mg++ 1.50 mg/
- Na+ 0 mg/l
- Cl- 2.46 mg/l
- SO4-- 4.34 mg/l
- Alkalinity as CaCO3 12.5 ppm
- pH 7.18
- Ione balance 0.07 mEq/l

Water profile adjustment:
- + 3.4g CaCl2*2H20
- + 9.05g Lactic Acid 80%

Grist (Grist Crush 1.3mm):
- 4.6kg Pilsner malt
- 0.3kg Wheat malt
- 0.3kg Cara Pils

Using Brewer's friend water calculator gives an estimated mash pH = 5,20

4. Using Tesco pH meter with temp control with accuracy and precision +/- 0.01 pH.
- Automatic reading when measurement is stable.
- Measurement is done when sample is cooled down to 25C.
- pH meter calibrated in 2 step buffer before mashing
- pH meter turned on for each measurement to avoid drifting.
- pH meter is controlled against calibration buffer after mashing is completed.


Step-Mashing and Ph readings:

Dough-in at 37C.

- 20 min at 40C => pH= 5.22
- 10 min at 55C => No pH reading
- 0 min at 66C => pH= 5.46
- 60 min at 66C => pH= 5.51
- 20 min at 72C => pH= 5.51

What we see is that the pH continue to drift upward during the mashing and stabilize at 5.51.

The interesting part is the correlation between actual Mash pH = 5.22 after 20 min rest and estimated pH.
Estimated pH in Brewer's friend calculator is 5.20.
In other words, estimated and empirical pH reading is within accepted values at 40C.
It also seems that empirical mash buffer capacity is measured via tritration and done at a given temperature.
ref article by A.J. deLange 2013: http://wetnewf.org/pdfs/estimating-mash-ph.html
I believe that results from these and similar kind of empirical measurement again is used in different calculators and spreadsheets.


In different thread discussions, pH-drift has been explained as:
- pH meter is drifting or instability
- It takes time to reach chemical equilibrium in the mash.
- Or there may be some mash chemistry that is not taken into account in the calculators and spreadsheet used by many brewers.



I have been searching for scientific papers, where pH drift in the mash has been observed.
I found this one:
Recipitation of Protein During Mashing: Evaluation of the Role of Calcium, Phosphate, and Mash pH
by M. J. Lewis and N. Nelson Wahnon, Department of Food Science and Technology, University of California, Davis 95616

http://www.agraria.com.br/extranet/...precipitacao_de_proteina_na_mostura_-_ing.pdf


The interesting thing about this article is that it seems to correlate with my observed pH-shift measurement example above.

The experiment in the article was doing step mashing (also single step) and what they found out was this:
- pH was creeping upward with time and maxing out at 70C before creeping down again
- When they mixed CaCl at different concentrations into the mash the pH-creep function shifted down as expected (lower pH)
- They also observed a sharp increase in the concentration of dissolved proteins from 40C and into the beginning of the saccharification temp interval
and then decreasing sharply after.
- Ca++ concentration had a decrease over time.
- They also did a so called "Grain out mash" in parallel, where all the grains and particles are filtered out of the mash and then do the mashing.
- For the "Grain out mash" they found that both Ca++ and pH did not drift over time but dissolved protein pattern remained the same.

What does that mean? It could mean that dissolved protein concentration does not have impact on observed pH drift.

The article does not conclude over the cause of the pH drift over time but points to solid material left behind in the mash when doing "grain out" mash.

So what could the cause of pH increase over time and temp increase be, if not caused by:
- pH meter is drifting or instability
- It takes time to reach chemical equilibrium in the mash.


We know there is a lot of particles and husk material in the mash together with dissolved and undissolved b-glucans and starch.
It is also known that beta-glucans and starch can be Acid Hydrolyzed but to which degree this take place in the mash is unknown.
In hydrolysis a larger molecule is broken down into simpler substances by the addition of water molecules.
When this process is carried out in the presence of acid, it is called Acid hydrolysis.
These reactions also speed up with increasing temp.

It is also known in organic chemistry that many acids can function as sources for the protons in this type of reactions such as phosphoric acid.

If this is the case here, free protons H+/H30+ could go into the acid hydrolysis reaction and we would observe that Ph decrease over time with increasing temp.
If acid hydrolysis is increasing with temp then the Ca++/phosphate acid equilibrium will also be influenced, which again will have impact on the Ca++ consentration:
A sharp decrease of Ca++ consentration was also observed with increasing mash temp in the study referred above.


Well what is the practical impact of this if the hypotesis is valid?
- Measure pH after ca 20min mashing before comparing with your estimated/calculated values and eventually do adjustments.
- Depending on mash schedule and method: The timing of acid addition is something to consider.

Looking forward to get some feedback on this.
 

bitteritdown

Well-Known Member
Joined
Nov 12, 2017
Messages
148
Reaction score
40
Since Phytic Acid is stored in the hulls try conducting a mash with hulled barley malt or malt some unhulled barley and add Na or K metabisulphate plus some hydrogen peroxide to the steep water.
 

ajdelange

Well-Known Member
Lifetime Supporter
Joined
Aug 5, 2010
Messages
11,959
Reaction score
2,725
Location
McLean/Ogden
Ph Change (drift) during mashing.

In different thread discussions, pH-drift has been explained as:
- pH meter is drifting or pH meter instability
This isn't pH drift. It is drift of the instrument used to measure pH drift. In considering actual pH drift in a mash it should not be included.
- It takes time to reach chemical equilibrium in the mash, measurement taken to early.
Mash, wort, fermenting beer and packaged beer are extremely complex chemical systems. Neither mash nor wort nor fermenting beer nor packaged beer ever reach chemical equilibrium though they appear to approach it asymptotically suggesting that there is an equilibrium state but we never, in any of the phases, quite get there. Nevertheless we rely on these 'phantom equilibrium' points to help us in determining how to brew a beer and whether we think we have succeeded or not.

- Or there may be some mash chemistry that is not taken into account in the calculators and spreadsheet used by many brewers.
As they know very little about the chemistry of the mash tun (or any other phase) the calculators cannot base predictions on any but the most general considerations of the chemistry actually taking place. It is obvious that mash pH prediction depends on having some understanding of the acid base reactions and this is the approach that they take assuming that a mix of acids and bases with given intrinsic pH and buffering capacity must conserve protons transferred and that the equilibrium pH is the pH at which the number of protons emitted by proton emitters are exactly equal to the number absorbed by proton absorbers. That pH can be calculated at equilibrium if the acid/base properties of each mash component is known. One of the main difficulties here is that while it is easy to do the calculations for the carbonic acid system species in liquor and mash it is not possible to do a similar calculation for a malt because the concentration and types of acids and bases in a malt is not known. We can measure a titration curve for a mass of each malt but one of the big problems with doing that is that the curves change with time. The pH of a mixture of a ground malt, DI water and a known amount of acid or base changes over time and this is usually explained by saying that it takes time for the water and acid or base mixture to penetrate the grist, solubilize the reacting grist components and finally react. Thus we have mechanical factors as well a kinematic factors at play and are really at a loss as to what to do except to wait until the pH stabilizes (and use a meter that we are sure is stable in making those measurements). But how long should we wait? Measuring mash pH would not be a useful thing to do if we had to wait 2 hours to do it and so we, from observation during titration experiments, note that things are generally stable after about half an hour though in many cases there is pH creep beyond that point. The process is further complicated by the fact that pH depends on temperature. It is not the same for a given malt if mashed in cold as it is if mashed in at a ß-glucan rest temperature, protein rest temperature or saccharification temperature (high or low). We can, in the course of a titration, determine the sensitivity of the malt being tested to temperature shift, and try to correct but it is a pretty iffy proposition. Now add the fact that calcium in the liquor reacts with some mash components (notably phytin but also with protein) precipitating insoluble salts and releasing some H+ in the process but the quantity of H+ released is unknown as a function of the amount of phosphate, protein and calcium present and can only be estimated usually from the work of Kolbach done prior to WWII which actually applies to knockout and not the mash.

Most of the spreadsheets aren't nearly sophisticated enough to incorporate some of the things just mentioned. Have you ever seen one that asks what temperature you are mashing at? Or what the temperature coefficient for the malt you are mashing is? The author of one of the most commonly used spreadsheets wasn't even aware that Kolbach's work applied to the kettle and not the mash tun. Nor do the users of the popular (or less known) spreadsheets have malt titration data available to put into a robust algorithm even if they were using one. If you want malt titration data, you will have to get it yourself. The spreadsheets are thus forced to use things like the name of the malt, its color or titration data done by someone else on a similar malt and the miracle here is that these calculators perform as well as they do. But don't expect their predictions to be better than ±0.2 if that good.




+ 9.05g Lactic Acid 80%
I'm not sure what this means but assuming that it means you weighed out 9.05 grams of the solution it would contain 0.8*9.0 = 7.2 grams of HLac. Adding this amount of lactic acid to a mash composed as you described suggests that the pH of the mash might be 5.31 - 5.47 depending on whether I consider an acidic (relatively) or basic pilsner malt and that there is some resemblance between the base and other malts I used in my calculation and those you are actually using. This would be at about a half hour after strike.


Using Brewer's friend water calculator gives an estimated mash pH = 5,20
That seems a bit low but I cannot claim that the numbers I gave above are any more accurate.




Step-Mashing and Ph readings:
- pH meter turned on for each measurement to avoid drifting.
I wouldn't do this. Keep the meter on. That way it doesn't have to 'warm up' each time a measurement is made. It should be at and stay at equilibrium temperature to insure that the conditions at the interface with the instrumentation amplifier and A/D are constant. This is not a major factor in pH measurement accuracy, however.

Dough-in at 37C.

- 20 min at 40C => pH= 5.22
- 10 min at 55C => No pH reading
- 0 min at 66C => pH= 5.46
- 60 min at 66C => pH= 5.51
- 20 min at 72C => pH= 5.51

What we see is that the pH continue to drift upward during the mashing and stabilize at 5.51.[/QUOTE]

I, of course, cannot say for sure but I suspect that what happened here is that some of the lactic acid was still in the liquid i.e. it had not time to completely penetrate the grist and react. At 20 minutes I would expect that the pH would be closer to its asymptotic value than this but it is not surprising, in cases where acid is added, to see a fairly substantial drop in pH relative to an early measurement over time. Perhaps the fact that this is no-sparge has something to do with the fact that the time seems longer than normal.





The interesting part is the correlation between actual Mash pH = 5.22 after 20 min rest and estimated pH.
Estimated pH in Brewer's friend calculator is 5.20.
In other words, estimated and empirical pH reading is within accepted values at 40C.
As discussed earlier, I wouldn't put too much reliance on that estimate. You have determined, assuming that all pH measurements were conducted properly and it appears they were, that the asymptotic pH for this mash is close to 5.51. That does not seem unreasonable based on the estimated mash pH as high as 5.47. This high value was obtained using Rahr Pils malt which has a pretty high pHdi (5.8) which, in combination with a high buffering capacity (46 mEq/kg•pH) pulls up mash pH.

To be continued.
 
Last edited:

ajdelange

Well-Known Member
Lifetime Supporter
Joined
Aug 5, 2010
Messages
11,959
Reaction score
2,725
Location
McLean/Ogden
Continued...

It also seems that empirical mash buffer capacity is measured via tritration and done at a given temperature.
ref article by A.J. deLange 2013: http://wetnewf.org/pdfs/estimating-mash-ph.html
I believe that results from these and similar kind of empirical measurement again is used in different calculators and spreadsheets.
Brewers friend is based on titration measurements but rather primitive ones (it was pioneering work) relative to what we now know. The model is, for example, linear and malt titration curves are not linear. The others calculators try to use color, malt type etc. as proxies for actual malt measurements.






- For the "Grain out mash" they found that both Ca++ and pH did not drift over time but dissolved protein pattern remained the same.

What does that mean? It could mean that dissolved protein concentration does not have impact on observed pH drift.
I don't think that you will be able to explain what you see based on protein (or phytin) reaction with calcium. The probable effect of calcium on your mash pH is about 0.03 pH. What you are seeing appears to be perfectly normal. You just have to realize that your estimator gave you a pH estimate appreciably lower than your actual malts gave you. Again, quite normal.

- It takes time to reach chemical equilibrium in the mash.
This is, IMO, what you are seeing but let's say 'approach' rather than 'reach'.


In hydrolysis a larger molecule is broken down into simpler substances by the addition of water molecules.
When this process is carried out in the presence of acid, it is called Acid hydrolysis.
These reactions also speed up with increasing temp.

It is also known in organic chemistry that many acids can function as sources for the protons in this type of reactions such as phosphoric acid.
The acid is only a catalyst here. It is not consumed. Also phosphoric acid is not am organic acid but, of course, organic acids can be a proton source for catalyzing hydrolysis. An example would be the citric acid used to invert sucrose. Note that in most recipes it is neutralized after the inversion is complete because it is not neutralized by the reaction. No protons are absorbed.

If this is the case here, free protons H+/H30+ could go into the acid hydrolysis reaction and we would observe that Ph decrease over time
Were protons consumed pH would go up.


Well what is the practical impact of this if the hypotesis is valid?
As it is worded it is not a valid hypothesis and so there is no practical conclusion to be drawn.

- Depending on mash schedule and method: The timing of acid addition is something to consider.
You want to add the acid as soon as possible (i.e. in the liquor) in order to get as close to the asymptotic )notice I did not say 'equilibrium') pH as soon as possible
 
Last edited:
OP
N

Nordic Brewer

Member
Joined
Feb 21, 2018
Messages
8
Reaction score
0
Ph.JPG

Thanks for a detailed and very interesting answer.
A lot of the background information regarding water calculators out there was new to me.
Regarding the advice keeping the PH meter on, I will remember.

Anyway the calculators are good as a starting point when I know what I'm dealing with
and a good basis for later fine tuning/adjustments.

Regarding the grist in my example: All the grist are from Bestmalz (Germany).
Yes 0.8*9.0g = 7.2 grams of HLac was used.

I have tried to sum up the discussion on a conceptual level without involving all the underlying math, by describing it with high level functions (se figure above).


To simplify: We assume pH(time) will correspond to pH as a function of time for a given grist composition, given water composition at a given mash temp.


- The function pH(time) tell us that the pH will stabilize if we wait long enough at a given mash temp.
- If time=0 B+C+D should give Ph of water before mixing it with the grist.
- As time increase the function D will move towards 0. In other words if you wait long enough pH(time) = B+C+0 and the water calculator (B) adjusted for
for error (C) should give you correct pH.

If we take my example regarding actual Ph-readings it is possible to "adjust" C and D so the result correspond to actual reading.
- In other word adjusting C because we assume the water-calculator is not precise enough regarding grist buffer values.
- Adjust the gradient of D (derivate of D) due to Poor circulation of the worth through the grist (take longer than normal to let the chemistry settle).

All your arguments above may actual be true regarding my brew set up and the water-calculator I have used, in other words C and D is not 0 at time=20 min,
and by adjusting C and D you can get a result near my final and last pH readings.
All good so far but by adjusting C and D to reach my final Ph reading, we get a problem explaining why my Ph reading at Ph(20min) = 5.2 = B = (B+0+0),
because by adjusting C and D to reach my final Ph, C and D will no longer be 0 at Ph(20min).

The main reason I started this thread is a Scientific paper describing Ph change with temp regarding "In-grain" mashes and no Ph change with temp regarding
"grain-out" mashes. The term "in-grain" & "grain-out" is described in the paper.(Se figure 8 & Discussions).


If this X-factor is real, we could actual have a function described as pH(time,temp) in my figure where the E-component is the "x-factor".
pH(time,temp) where:
- The function pH(time,temp) take into account that if we wait long enough and let the chemistry settle out at each mash-temp interval, we will see that pH also will change with temperature.


If one take the function pH(time,temp) it is should be possible to set up experiments where we minimize C and D and hence verify if E exist or not.
And if E exists it should be interesting to find out what the chemical explanation is.
 

ajdelange

Well-Known Member
Lifetime Supporter
Joined
Aug 5, 2010
Messages
11,959
Reaction score
2,725
Location
McLean/Ogden
It's widely accepted, and easily demonstrated that E exists. But data on what the slope is (change in pH per degree temperature change) is is sparse. In order to try to get a bit more I made test mashes by putting small amounts of finely milled malt in distilled water in stainless steel beakers in a water bath and recorded pH and temperature over time. I would start at a reasonable strike temperature wait until stable then reduce temperature over time and then increase again. The pH vs temperature plots were not nearly as nice as one might hope for but nice enough that slope could be deduced. It was from these experiments that I determined that the slope is different for different malts.
 
Top