I have a big confusion over the mobile phase pH requirement by FDA/recommendation by other agencies. I don't even know if it is a requirement or a recommendation.

My colleague tells me if the pH of the mobile phase is to be controlled it has to be done for the buffer and not for the final mobile phase. That is, without adding the organic phase. Further, the lowest recommended pH one should work with is 2.00 for the aqueous buffer. Here it gets confusing since I used to think that the pH of aqueous buffer plus the organic phase is the one which actually matters for the degradation of the bonded stationary phase.

Though I agree that the final mobile phase pH can be accurately controlled by controlling the aqueous buffer pH but as far as going below pH 2.00 issue is concerned I think it should be applicable to final mobile phase pH.

For example if we have a buffer wiht pH 1.5 and after adding organic phase it goes above it should be fine. But my colleague tells me that the buffer should not be adjusted below 2.00

Please respond.

Thanks a lot in appreciation in advance.

Sean

We are dealing with two independent and unrelated things.
The first question is, how the pH should be measured such that one knows, where it is in reference to the pKa of the buffer. This should always be done in the purely aqueous part of the mobile phase. In water, acetic acid has a pKa of 4.75, and if I make a buffer of 4.75, I have an acetic acid buffer of maximum buffering capacity. If I do this in the presence of an organic solvent, I have no clue where the pKa of acetic acid is.
The second question is about the pH stability limits of a bonded phase. For some bonded phases, the manufacturer recommends not to use a pH less than 2. But this does not mean that the bonded phase is stable for 5 years at pH 2.1, and falls apart after 1 hour at pH 1.9. One will simply get reduced lifetime at pH values lower than 2, and acceptable lifetime at a pH higher than 2. The reference point is water - as above.
If the manufacturer of your column recommends to use a pH above 2, he means to do this measurement in water. This does not mean that your column will die immediately, if you go to pH 1.5. It just means that if you should get a column lifetime of a month at pH 2, you may get only a week at pH 1.5. If this is not a problem for you, go for it.

Sean,
The measurement of pH is a science in itself, like chromatography. Chemists have written whole books about pH measurement. Go to that literature to get answers to your questions.

In general praxis counts for more than theory. If a mobile phase eats your column up quickly, the pH matters little. First approximation gives pH value the most important cause of column deterioration and most would agree, but still suggest a counter ion change could alleviate the deterioration at a given (borderline) condition.

Organic solvents surpress the ionization of acids, that is, the hydrogen and counterion have a more difficult time staying separated in organic solvents. That fact should help the column. The organic solvent might aid the counterion and hydrogen get closer to the packing, however, and that would harm the column. The water in the mobile phase limits the pH change. No great change occurs until the amount of water becomes small, not a reverse phase HPLC situation. For example, no propyloxy ion forms in the presence of water. So pH stays about the same as if diluting with water.

Do worry about non-aqueous pH in your chromatography. The FDA suggestion or requirement comes about from the possible failure of the measuring instrument in mixed solvents, not from a consideration of accurately measuring the pH. If you mix a buffer, pH the solution, and then add the organic solvent, your solution will match solutions prepared in other labs even though they measured pH differently than you did.
Jim

Thanks Uwe and Jim for responding but does it mean there is no regulatory requirement for pH control of the aqueous buffer.

Another related thing is the buffering range of buffer salts. A buffer salt which controls a pH in the range of say 2.1 to 3.1 controls the pH of aqueous phase or that of final mobile phase (mixture of aqueous phase plus organic phase).

Thanks for responding.

Any other related discussion or article references are more than welcome.

Sean

Sean,
You're right. You have a regulatory requirement to control any variable that affects the quality of the product you release. So if your method controls a process variable or product assay that relates to efficacy, your method must control all variable that affect that assay, and if pH is one of those variables you must control it. For all that, however, the discussion drifted to the control of pH.

The first topic is measurement of pH. Every method of measuring pH does not give the same value as for other analytical methods. Organic compounds affect some pH electrodes more than others. That is, the pH could stay the same and its measured value would change.

The next topic is the affect of organic solvents on pH. Enough experiment evidence exists to make the claim that organic solvents do affect the pH. But that effect becomes convoluted with other factors during practical use. For example, with every buffer I know about, pH drifts with dilution as opposed to simple theory. That is, the pH drifts even when diluted with water, let alone an organic solvent.

The last topic is preserving the consistency of a mixed solvent buffer for the purpose of quality control. With some exceptions, the pH measured by various methods in water with glass/silver chloride or glass/calomel electrode pair will give the same values for the pH, so solutions prepared lab to lab will have the same pH. In whatever way the pH varies with the addition of an organic solvent, the pH will vary in the same manner lab to lab.

Perhaps your company will use only your method. In that case your validation covers the preparation of the buffer, method of measurement for the pH, and the acceptable limits of pH in the buffer. I haven't read the regs in a while but in general the FDA wants good drugs not 'measuring the pH in a certain manner', that is, your first duty is to the consumers and the FDA does its best to product them. If you have a good reason for measuring the pH in a different manner and have proven it to work in your validation the inspectors will accept it.
Jim

As far as I know, the FDA does not prescribe how the pH should be measured. However, if you need to measure the pH in an unusual way, you should specify this in your procedure.
What I had described before is the common way of dealing with the issue without getting into unnecessary difficulties.

Sean,

I covered several related issues in articles in the American Laboratory.
If you give me your address, I'll send you the collection of articles.

Uwe

A good book on pH measurement....Helmuth Galster, pH Measurement. It may still be in print.

another good book is Determination of pH by Bates but it is not in print. Bates worked at NIST and he developed the currently accepted definition of pH as well as buffers to support the definition. Note, that pH is a defined quantity, not a fundamental property of matter. If you can't find the books, a literature search on Bates will get you the same information. In Chem Reviews there is a good ,but not easily read, summary of Bates' work.

There is a near endless, but useful, discussion of this topic in the archives. I don't know if you can search on a contirbutor's name but if you can I was a contributor. I think it occured about 13 months ago.

The most reproducible and least subject to misinterpretation way to prepare buffers is by weight and forget measuring pH. This statement will once again bring folks out of the woodwork to rail against the proposal, but, the buffers used to standardize pH electrodes are prepared by weight at NIST. If they think this is the most accurate way to prepare a buffer, chromatographers should consider the merits.

Many will find this fact unsettling, but pH is defined by NIST as what the meter says when you stick the electrode in something. In most cases the number has no fundamental meaning, especially in a nonaqueous solution. So, one should not be surprised when a 0.01M solution of HCl reads one value in water and quite another in methanol. The pKa of HCl, the activity coefficients, junctions potentials (which could be 1 or 2 pH unit equivalents) and standard state of hydrogen electode couple are all different between these two solutions and all these factors affect the pH reading on the meter.

Bill, here is one out of the woodwork. But surprise, recently I prepared a buffer, used in an iodination, by weighing ONLY, for technical reasons...(needed buffering in an existing solution which should not be diluted). Still, normally, its not unwise to check the pH with a pH meter, it doesnīt hurt (unless you stick the electrode in the buffer instead of taking part of the buffer out, or change ionic strength in correcting, etc.) and lets you know whether the activity coefficients are not playing havok. It shouldnīt be forgotten that activities and concentrations might be quite different. ....Just canīt sleep well, normally, without measuring the pH with a pH meter.
Incidentally, a pH stick has been useful to check on when a non-neutral solution has emerged from the HPLC system or such . . . .

Hans

Bill
Good rules are good rules, but all good rules have exceptions...
It might be difficult to make an ammonium chloride buffer at pH 9 by accurately weighing the ammonia...
Uwe

Uwe

Referring back to your first response, does it mean that when organic phase is added to an aqueous buffer of pH 2.0 the real pH is still 2.0 and adding the organic phase has no effect on pH.

In other words, adding 500 mL or 1000 mL methanol to 500 mL of pH 2.0 buffer will actually not change the pH, and the working pH would still be considered pH 2.0 though on a pH meter the final pH of these two different show a different pH.

If this is not true and the final pH is the real working pH, then how to choose between a different salt for the final pH range (say pH 5.0) than for aqueous buffer (say pH 2.0) since they have different buffering range.

I hope I am making it clear enough to be understood.

Thanks and appreciations for your responses everyone.

Sean

Now HW, I have no objection to measuring pH. It has its place. And a stock of pH strips is truely essential. But, without a fresh supply of students to harass the forum was getting dull and a tree needed shaking.

If you provide me a procedure that involves buffer preparation I would prefer that you provide weights and volumns. Then my technician can reproduce this buffer to easily one part in a thousand, there is no question of whether the pH was measured before or after solvent, and the buffer can be prepared in the least amount of time, a factor more important in some environments.

At my company, ideally if we develop from scratch, we'll adjust pH of aqueous portion first, then add organic. If we adapt from literature, we do as original authors, we just write instructions in an un-ambiguous way. And that's what the FDA wants: validate with your procedure, and make it bulletproof, don't let a tech get confused by lazy writing ("Hey, I know what I mean").

Sean,
If I make an acetate buffer at pH 4.75, it is at the point of maximum buffer capacity for the concentration. This is where I would like to be. If I now add organic solvent and measure the pH again in the presence of the organic solvent, the pH will have shifted to a higher pH. But the buffer capacity remains at exactly the same strength, since the pK is shifting to the same spot due to the addition of the organic solvent.
If I make a phosphate buffer at pH 2, the exact same thing is happening.
Think about it in the following way: if you add an organic solvent, your pH scale is shifting relative to water. Now you don't know any more, where the pK of your buffer is. Why do you need to know the pK? Because you use a buffer to control retention, and buffers work best at their pK.
This is roughly the same statement as your statement that "the working pH is still considered to be pH 2 though on a pH meter the final pH ... shows a different pH".
The bottom line is - don't worry about the measurement in an organic solvent. It is complicated and irrelevant for all practical purposes.

Bill

I think you needed to read my questions more carefully and with more concentration before giving your judgement on who's who.

My question was related to buffering capacity and pK and it has been nicely answered by Uwe. (Thanks Uwe!!)

And by the way everybody has room for learning which makes everyone a student.

I guess a wrong part of tree got shaken.

Sean

Sean, my last note was harassing HW Mueller as part of our on going buffer discussions. I am trying to get him to retire his pH meter and he gleefully points out when I answer posts without my glasses. He would have gotten the inside student joke. It had nothing to do with you. (Periodically students post with the hope that forum participants will do homework for them and it often sets off a spirited exchange that adds interest to the forum.)

Bill, I am sorry for misunderstanding the things. No hard feelings. Peace out!!

Sean

Accepted and OK. I did in fact post a lengthy answer to your question, separate from my confusing note, but it vanished, which happens from time to time. I wish this site had some better software. "www.wwforum.com" for woodworkers would be nice format to copy.

Another question about the pH:-

For calibrating the pH meter in the lab, is the buffer solution or buffer powder (powder can be dissolved in the lab to make solution) for pH 1.00 or lower available commercially???

I am using a solvent with phosphate salt to control the pH and the pH of the solvent is 1.20
My pH meter is calibrated in the range of 2.00 to 4.00 If the solution for pH 1.00 is not available, how to convince the FDA??

Thanks

you are saved! What you need is the classic "Clark and Lubs" buffer system which covers pH 1 to 10.2 in 0.2 pH increments. It is described in a NIST publication (actually a NBS publication as it is old).

Note to all, it is prepare with weights and volumns!

for pH 1.00 you add 25 mL of 0.2M KCl (for ionic strength control) and 67.0 mL of 0.2M HCl and dilute to 100mL total with water. If you wonder that HCl is a buffer consult the archives of a few months ago for buffer capacity.

If you don't have an easy means of standardizing the HCl simply buy 1M standardized HCl and dilute.

For pH 1.2 buffer for LC you most likely could simply use phosphoric acid and forget the salt.

Thanks Bill, If I use phosphoric acid in my solvent and specify x mL of phosphoric acid and not specify the pH, could I just avoid specifying pH

I don't know anything about GLP/GMP, so from a regulatory perspective I don't know the answer to your question.

In a nonregulatory setting, a procedure that called for the preparation of a mobile phase by adding xx mL of yy% phosphoric acid would be scientifically sound. NIST and NIST traceable buffers are prepared by weights and volumn, so it would be resonable for the regulatory agency to allow this approach.

In an industrial lab setting we used to do many LC analyses for aliphatic acids. The aqueous mobil phase recipe we wrote and used was simply xx mL of concentrated phosphoric acid per liter of water. It is certainly by far the most reproducible approach, and way faster than adjusting to a pH.

Back from taking an Easter rest, canīt pass this up. Bill, I thought I was harassing you, shucks.
Also, I am using some pills from Merck which are dissolved in H2O to make a standard buffer (for checking the pH meter). As statet in the earlier buffer discussion? one needs very good, primary standard type materials to make a buffer with weighing only. Some substances are extremely pH dependent, less so ionic strength dependent. So if you start out with super H3PO4 you may pick up H2O in time, etc. You can still use it if you check your pH with a meter, your ionic strength may be off a bit.... Just one example. A small amount of such contamination is no issue if you are in a good buffering range.
Luckily regulatory issues are not necessary in this lab either, but I have followed some of these discussions with bewilderment. One gets the impression that one has to follow rules, not necessarily have a working analysis. This has been mentioned before also: A single chromatographic technique is insufficient to ascertain a concentration, no matter how many rules are followed. Furthermore, if the rules cause one to worry as to whether it is sufficient to report the weight make-up of a buffer, than the rules need extensive revision.

Hans