-
- admin
- Site Admin
-
- Posts: 647
- Joined: Wed Aug 11, 2004 12:19 am
By Anonymous on Tuesday, May 4, 2004 - 09:08 am:
All-
Can anyone tell me or direct me to this information? I cannot find the pKa of heptanesulfonic acid in the CRC, Merck Index, or Lange's. Also, I've google searched this to no avail. Help? Thanks,
Jeff
-------------------------------------------------------------------------------------------------------
By Anonymous on Tuesday, May 4, 2004 - 10:25 am:
http://www.chem.uic.edu/ghosh/links.htm
-------------------------------------------------------------------------------------------------------
By Chris Pohl on Tuesday, May 4, 2004 - 11:14 am:
Jeff,
I couldn't find heptanesulfonic acid per se but I did manage to dig up some data on shorter analogs: methanesulfonic acid (pKa: 1.92), ethanesulfonic acid (pKa: 1.68) and propanesulfonic acid (pKa: 1.53) in Smith and Martell's "Critical Stability Constants". Based on extrapolation, one would expect heptanesulfonic acid to fall somewhere around 1.35
-------------------------------------------------------------------------------------------------------
By Anonymous on Wednesday, May 5, 2004 - 05:36 am:
Thanks, Chris. I knew someone would come through for me...
-Jeff
-------------------------------------------------------------------------------------------------------
By Anonymous on Sunday, May 9, 2004 - 12:27 am:
I have a similar question concerning trifluoroaetic acid, pentafluoropropionic acid, heptafluorobutyric acid and so on...
Which one is a stronger acid - I couldn't find pKa for the acids (except for TFA - 0.23)
-------------------------------------------------------------------------------------------------------
By Anonymous on Sunday, May 9, 2004 - 02:33 am:
Can we determine the pKa of a neutral molecule ?
-------------------------------------------------------------------------------------------------------
By Chris Pohl on Monday, May 10, 2004 - 02:58 pm:
I couldn't find much either but this link gives the pKa for TFA (0.3) and HFBA (0.4): http://scholar.lib.vt.edu/theses/availa ... ch_05b.pdf
-------------------------------------------------------------------------------------------------------
By Anonymous on Wednesday, May 12, 2004 - 02:29 pm:
yes, I know this document... Don't you think, that PFBA must be a stronger acid than TFA?
-------------------------------------------------------------------------------------------------------
By Chris Pohl on Wednesday, May 12, 2004 - 05:41 pm:
The pKa order for aliphatic acids is: formic; 3.75, acetic; 4.76, propanoic; 4.87, butyric; 4.82. From this trend you can see that a methyl group has a much bigger influence on the pKa than a methylene. The data listed in the reference above supports a similar trend for perfluoro carboxylic acids. TFA should be the strongest acid since it contains three fluorines on the carbon adjacent to the carboxylic acid. While one might expect additional fluorines to further increase acidity, moving the trifluoromethyl group further away from the carboxylate will reduce the inductive effect of the trifluoromethyl group which should have a larger influence on the acidity than a difluormethylene group adjacent to the carboxylate.
-------------------------------------------------------------------------------------------------------
By Anonymous on Friday, May 14, 2004 - 12:33 pm:
I would think that also. But are there any data about the inductive effect of fluorine? I mean this thing: isn't it possible, that three fluorines attached to the carbon have such inductive effect together, that CF3- group has possibly higher inductive effect that just F- !!!!
You cannot just compare CH3- group to CF3- group. However I need to find some data about such inductive effect.
-------------------------------------------------------------------------------------------------------
By Constantine Sychov on Wednesday, May 19, 2004 - 08:23 am:
Hi, Chris!
In advance sorry for bad Engish for I have limited time
In one of my researches I managed to establish the order in which the value of ion exclusion effect changes (using data obtained on different RP and sulphonated PSDVB phases and using principal components` method). It is (see the second column, factor 2):
Factor Factor
1 2
Tartaric -1,70592 -,62081
Formic -1,16409 ,83542
Malic -,63685 -,21728
Citric ,53431 -1,16964
Lactic -,13280 ,52930
Acetic ,16483 ,97271
Maleic ,76046 -1,75274
Succinic ,94138 ,40294
Fumaric 1,23868 1,02009
Ion exclusion is the maximum for citric and maleic and minimal for formic acetic and fumaric. Can you take it in accordance with pKa of these acids. Or the situation`s more complex? I wonder why maleic and fumaric acids act so differently.
Constantine.
-------------------------------------------------------------------------------------------------------
By Chris Pohl on Sunday, May 23, 2004 - 06:59 pm:
Anon (5/14)
Regarding the pKa of of fluorocarboxylic acids, it's conceivable that the trifluoromethyl group could have a larger inductive effect than a single fluorine but my point was that this would be an exception to the general rule that substituents on the beta carbon have significantly less inductive effect on the pKa than is generally observed for substituents on the alpha carbon. The question was: why would one be expecting this and why when presented with data supporting the general rule are you not accepting this? Anyway, if you have a strong motivation to prove your hypothesis, why not determine pKa yourself. In general, it is difficult to determine pKa for such strong acids but for these compounds, you should be able to use either H or F NMR to determine pKa.
-------------------------------------------------------------------------------------------------------
By Chris Pohl on Sunday, May 23, 2004 - 07:23 pm:
Constantine,
While pKa is a significant factor in ion exclusion retention, hydrophobicity of the neutral form of the acid is also a major factor. From the data cited above (pKa order for aliphatic acids is: formic; 3.75, acetic; 4.76, propanoic; 4.87, butyric; 4.82) one would expect that the elution order would be formic, acetic, butyric and propionic with the latter two nearly coeluting. But in fact, butyric acid invariably elutes well after propionate under aqueous conditions, even when the pH of the mobile phase is low enough to drive both acids to the total permiation volume. In fact, unless you add solvent to the mobile phase, butyric acid always elutes after the total permiation volume, clearly showing that a "reversed phase" component to retention is also playing a role with butyric acid retention.
I suspect that the reason for the large difference between the retention of maleic and fumaric acid stems from differences in how the two molecules to interact with the pi electrons of the styrenic backbone of the ion exclusion phase. Similar differences are also seen in anion exchange separations of these two acids with styrenic ion exchange materials. In fact, this effect seems general with analogous effects being seen with cis and trans-aconitic acid.
-------------------------------------------------------------------------------------------------------
By Constantine Sychov on Monday, May 24, 2004 - 06:17 am:
Chris,
as for your first assumption - using PCA (principal components analysis) I extracted two factors (linear independent parametres)from logk` values on different phases. The first with average contribution about 70% is "reversed-phase" factor, it is factor 1 (look, values for maleic and fumaric acids are almost the same, as its going to be). The second factor 2 (contribution about 30%) reflects ANOTHER property, which may be called "ion exclusion". This property is INDEPENDENT.
As for your second assumtion. You see, maleic acid is not more retained, it is more EXCLUDED; so we observe the opposite effect.
And even if maleic acid was more retained (lets consider this theoretical possibility) - it can not be due to pi-interactions. I scaresely investigated the phenomena of pi-interactions in LC (see topic "pi-interactions" here) - they are very small in highly polar and aqueous environments even on neutral hypercrosslinked polystyrene, and they are almost negligable on sulphonated low crosslinked ones.
So the question remains. If I have time I`ll try to find pKa values for these compounds.
Anyways - thanks for discussion.
Constantine
-------------------------------------------------------------------------------------------------------
By Chris Pohl on Monday, May 24, 2004 - 08:02 pm:
Constantine,
Sorry you misunderstood my earlier message.
Regarding your first point about my "first assumption", you are reiterating precisely my point: "reversed phase" retention is a major component in the retention of many analytes in ion exclusion. I think we are agreement on this point (unless you mean that reversed phase retention is a major component for all analytes). I'm not sure why you thought I was saying these retention processes are interdependent. They are, of course, independent retention processes with each analyte exhibiting each retention mechanism to varying extents.
Second, I didn't say that maleic acid is more retained than fumaric acid due to pi electron interactions. My point was that fumaric acid (and other trans acids) exhibits an unusual amount of added retention due to pi electron interations between the analyte and the polymer backbone relative to the cis form of the same acid. In anion exchange, for example, pi electron interactions are definitely not negligable interactions in many materials based on aromatic polymers even though they may be generally "well hydrated". I think this stems from the fact that there are "knots" in the polymer matrix where hydration is extremely contrained. Simutaneous electrostatic interactions and pi electron interactions combine to enhance retention at a small number of poorly hydrated "high energy" sites. If you are interested in testing this, compare the retention via anion exchange of maleic and fumaric acid on a styrenic anion exchange material under condition where both analytes are fully ionized and the eluent species is extremely polar such as hydroxide or carbonate. In this case, you will frequently see an extreme retention difference for maleic and fumaric acids with fumaric eluting well after maleic. Under these conditions, adding even micromolar concentrations of an aromatic eluent component such as p-cyanophenol to the eluent will dramatically effect the retention of fumaric acid relative to maleic acid without significant changes in the retention of more polar analytes incapable of interacting with the pi electrons of the polymer backbone. Similar interactions are also involved in the unexpectly high retention of fumaric acid via ion exclusion.
-------------------------------------------------------------------------------------------------------
By Constantine Sychov on Tuesday, May 25, 2004 - 08:14 am:
Chris,
I will think about you have said.
Now my first reaction is that I don`t think that the reason is pi-interactions. May be we came across with intra-molecular mesomeric cis-trans effect? I think we olny have to find exact pKas for these acids to continue useful discussion. So, I`ll try, I promise.
OR you see, the experiment also must be carried out on sulphopropilated silica-gel. Mmmm... here I have problems... I have very very old ones I never checked. As far as I`m specializing on chiral separations I only have good phases of this type.
As for addition of cyanophenol... it`s amazing fact.
And, Chris, do you have some linkes or do you have already obtained k` or better jusk tR values of standard org. acids on ion exchane phases (without cyanophenol and with it)? Such information could really help to clear situation.
Constatine
-------------------------------------------------------------------------------------------------------
By Constantine Sychov on Tuesday, May 25, 2004 - 08:28 am:
Chis,
maybe it is not ION exclusion but just exclusion? Maybe cis is hydrated much harder then trans?
-------------------------------------------------------------------------------------------------------
By Chris Pohl on Tuesday, May 25, 2004 - 07:18 pm:
Constantine,
When I get back in the office on Thursday, I'll post some pKa values for maleic and fumaric acids. Any other pKa values you would like?
I'll have to look and see if I can find a convenient example of the p-cyanophenol effect to post. If not I'll see if I can generate a fresh example.
I suppose that the unusual retention of fumaric acid could, in part, have something to do with hydration but wouldn't you think that hydration of the trans form would be easier given the geometry? If so, it should have the least retention? I'll look into that too when I'm back. The limiting equivalent conductance might help sort out hydration differences.
One last point on the pi electron interactions in ion exclusion: aromatic acids all exhibit exceptionally high retention on ion exclusion columns. So much so that such columns are almost never used for these analytes without the addition of solvent.
-------------------------------------------------------------------------------------------------------
By Constantine Sychov on Thursday, May 27, 2004 - 08:56 am:
Chris,
thanks you for help in solving of such "unprogmatic", but interesting task: to deduce the origin of so called "ion-exclusion effect". Of course, I`m interested in pKa of all components listed in previous table plus pKa of oxalic acid (it is not in list for I used log a values with relation to oxalic acid to avoid problems with void volumes - all the data was taken from literature). About hydration - yes it may seem that is not good assumpion but it worthy to be checked. About pi-interactions - the retention data on SAX silica-gel is needed to close the discussion on this topic.
-------------------------------------------------------------------------------------------------------
By Constantin Sychov on Tuesday, June 29, 2004 - 03:04 am:
Hi, Chris,
the situation is following,
"ion exclusion factor" for the test mix or organic acids rather poorly correlates with their pKa values, but bad correlation ( r = 0.76) is due to two acids - citric and fumaric. Without fumaric acid (too little excluded) correlation is satisfactory r - 0.91, without citric (too much excluded) it is good r = 0.95. In fist approximation, this fact can be explained, you see, by pi-interactions and size exclusion, accordingly, but it is suggestion to be proved.
Then, the contribution of "ion exclusion" appears different for different phases, I`m sure, it depends on pore diameter and degree of sulphonation.
-------------------------------------------------------------------------------------------------------
By Chris Pohl on Tuesday, June 29, 2004 - 03:55 pm:
Constantin,
It sounds like you must have found pKa data for your test mixture. Sorry I didn't post this information as I promised. Anyway, from my point of view there are two factors involved in ion exclusion retention: ion exclusion (this component should be correlated to pKa) and the solubility parameter of the analyte relative to that of the stationary phase. I think it's a mistake to focus on any sort of pore size effect since gels of the sort used in ion exclusion materials don't have any true pores (in spite of the fact that they are typically referred to as being microporous). They are in fact true hydro-gels, fully solvated by water. Because of the extremely low pH within the stationary phase (the pH of the stationary phase should be around zero), any analyte which enters the stationary phase will be essentially 100% nonionic while in the stationary phase. For this reason, any added (or diminished) retention of the neutral form of the analyte must reflect the relative solubility of various analytes in their neutral form in the stationary phase.
From this point of view, I would interpret the higher than expected retention of fumaric acid as evidence of relatively similar solubility parameters for solute and stationary phase, thus enhancing retention. Conversely, the unexpectedly low retention for citric acid is indicative of a bigger discrepancy between the solubility parameters of the solute and the stationary phase, thus diminishing retention.
-------------------------------------------------------------------------------------------------------
By Constantin Sychov on Wednesday, June 30, 2004 - 08:28 am:
Chris,
I`ve got your point, and it is the same as mine. Firstly I didnt believe that "ion exclusion" mode for organic acids is true ion exclusion, for analytes are almost in non-ionic form. But the fact is - that lg alphfa correlates with pKa rather well (exept above mentioned couple of acids). Maybe, the property you are calling "solubility parameter" also depends on pKa?
I think, it can be easyly checked by eluting acids under neutral pH... but I didnt see any data of this kind.
-------------------------------------------------------------------------------------------------------
By Chris Pohl on Wednesday, June 30, 2004 - 09:09 am:
Constantin,
Before I posted my most recent comment above I did a bit of searching through tabulated data on solubility parameters. Solubility parameters are commonly used in polymer chemistry as a basis for predicting what solvents will dissolve polymers. Unfortunately, while I found solubility parameters for a number of common carboxylic acids I could not find these values for fumaric or citric acid. Furthermore, there is no tabulated data on the solubility parameters for sulfonated polystyrene so there's really no basis for testing this hypothesis unless you can find this sort of tabulated data elsewhere.
It seems, however, that you may have misunderstood one my points in my previous post. I would characterize ion exclusion as largely dominated by the pKa of the analyte. After all, if the analyte isn't at least partially in the neutral form in the mobile phase it will not have access to the ion exclusion gel particles. It's just that the solubility of the analyte in the stationary phase plays a secondary role in this separation process so one cannot simply predict elution order based on pKa alone.
-------------------------------------------------------------------------------------------------------
By Anonymous on Monday, July 5, 2004 - 03:03 am:
Chris,
Can you confirm your value of the pKa of methanesulfonic acid? I put this into a well-known pKa calculator and it gave me a value of MINUS 0.4. o.k. these things are sometimes wrong, but the calculator also has a database of experimentally determined values and quoted me a value of MINUS 1.86 as determined by I think Serjeant- a well known worker in the field. Well this could be wrong too; however, I did think methane sulfonic acid was stronger than you suggested. Also I don't see why propanesulfonic acid should be stronger than methanesulfonic acid, from your figures and a consideration of inductive effects.
Regards,
David
-------------------------------------------------------------------------------------------------------
By Chris Pohl on Saturday, July 10, 2004 - 02:07 pm:
David,
I agree with you that these values seem lower than expected. Perhaps there is some mistake although I have yet to find an error from this reference: Smith and Martell, Critical Stability Constants, 1977 (it's in the volume titled: Other Organic Ligands) I doublechecked these phases and again, pKa values were 1.92 for methanesulfonic acid, 1.68 for ethanesulfonic acid and 1.53 for propanesulfonic acid. By the way, the original reference that they cite for this data is: A. Covington and R. Thompson, J. Solution Chemistry, 1974, volume 3, p 603
-------------------------------------------------------------------------------------------------------
By David on Monday, July 12, 2004 - 08:50 am:
Chris,
Thanks for your answer. Here are 2 web references that quote the pKa of methanesulfonic acid as about MINUS 2.
http://www.cem.msu.edu/~reusch/VirtualText/acidity2.htm
http://www.mrgoodchem.com/Mr_%20Good%20 ... 0(MSA).htm
I thought that all these sulfonic acids were completely ionised over the range of RP-HPLC, and that was why they were chosen as ion pair reagents. Similarly, there are polymeric ion exchange HPLC columns with sulfonic acid groups(often propanesulfonic acid) and the whole idea of these groups is that they remain completely ionised over the range of operation of the column (which may be pH 1 -13 if the column is polymeric)
-------------------------------------------------------------------------------------------------------
By David on Monday, July 12, 2004 - 08:54 am:
Sorry, for some reason in the second link (MSA).htm bit missed out and the link above doesn't work directly. However, if you copy the whole lot and paste it into your browser it will work.
David
-------------------------------------------------------------------------------------------------------
By Chris Pohl on Monday, July 12, 2004 - 01:42 pm:
David,
As I said, the values from Smith and Martell do seem to indicate acidity values which are lower than expected. On the other hand, as I said, I haven't yet encountered a documented error in the data from Smith and Martell. For example, the first reference you mentioned as a number of values which do not agree with either Smith and Martell or Lange's Handbook of Chemistry (which I generally use as my primary reference for pKa values). These include discrepancies for oxalic acid listed as 1.2 although Lange's list it at 1.27 and Smith and Martell list it at 1.27 or 1.26; trifluoroacetic acid, listed in Lange's at 0.5 (several other references also list trifluoroacetic acid in the 0.3-0.4 range). I guess you could argue these aren't big discrepancies but the fact that they don't match those of the best sources makes me wonder where they got the data. All things being considered, it seems like the data in the two references you cited isn't very believable either. The value you cited at - 1.86 seems in the ballpark of what I might have expected although even this seems to be a bit extreme.
All-
Can anyone tell me or direct me to this information? I cannot find the pKa of heptanesulfonic acid in the CRC, Merck Index, or Lange's. Also, I've google searched this to no avail. Help? Thanks,
Jeff
-------------------------------------------------------------------------------------------------------
By Anonymous on Tuesday, May 4, 2004 - 10:25 am:
http://www.chem.uic.edu/ghosh/links.htm
-------------------------------------------------------------------------------------------------------
By Chris Pohl on Tuesday, May 4, 2004 - 11:14 am:
Jeff,
I couldn't find heptanesulfonic acid per se but I did manage to dig up some data on shorter analogs: methanesulfonic acid (pKa: 1.92), ethanesulfonic acid (pKa: 1.68) and propanesulfonic acid (pKa: 1.53) in Smith and Martell's "Critical Stability Constants". Based on extrapolation, one would expect heptanesulfonic acid to fall somewhere around 1.35
-------------------------------------------------------------------------------------------------------
By Anonymous on Wednesday, May 5, 2004 - 05:36 am:
Thanks, Chris. I knew someone would come through for me...
-Jeff
-------------------------------------------------------------------------------------------------------
By Anonymous on Sunday, May 9, 2004 - 12:27 am:
I have a similar question concerning trifluoroaetic acid, pentafluoropropionic acid, heptafluorobutyric acid and so on...
Which one is a stronger acid - I couldn't find pKa for the acids (except for TFA - 0.23)
-------------------------------------------------------------------------------------------------------
By Anonymous on Sunday, May 9, 2004 - 02:33 am:
Can we determine the pKa of a neutral molecule ?
-------------------------------------------------------------------------------------------------------
By Chris Pohl on Monday, May 10, 2004 - 02:58 pm:
I couldn't find much either but this link gives the pKa for TFA (0.3) and HFBA (0.4): http://scholar.lib.vt.edu/theses/availa ... ch_05b.pdf
-------------------------------------------------------------------------------------------------------
By Anonymous on Wednesday, May 12, 2004 - 02:29 pm:
yes, I know this document... Don't you think, that PFBA must be a stronger acid than TFA?
-------------------------------------------------------------------------------------------------------
By Chris Pohl on Wednesday, May 12, 2004 - 05:41 pm:
The pKa order for aliphatic acids is: formic; 3.75, acetic; 4.76, propanoic; 4.87, butyric; 4.82. From this trend you can see that a methyl group has a much bigger influence on the pKa than a methylene. The data listed in the reference above supports a similar trend for perfluoro carboxylic acids. TFA should be the strongest acid since it contains three fluorines on the carbon adjacent to the carboxylic acid. While one might expect additional fluorines to further increase acidity, moving the trifluoromethyl group further away from the carboxylate will reduce the inductive effect of the trifluoromethyl group which should have a larger influence on the acidity than a difluormethylene group adjacent to the carboxylate.
-------------------------------------------------------------------------------------------------------
By Anonymous on Friday, May 14, 2004 - 12:33 pm:
I would think that also. But are there any data about the inductive effect of fluorine? I mean this thing: isn't it possible, that three fluorines attached to the carbon have such inductive effect together, that CF3- group has possibly higher inductive effect that just F- !!!!
You cannot just compare CH3- group to CF3- group. However I need to find some data about such inductive effect.
-------------------------------------------------------------------------------------------------------
By Constantine Sychov on Wednesday, May 19, 2004 - 08:23 am:
Hi, Chris!
In advance sorry for bad Engish for I have limited time
In one of my researches I managed to establish the order in which the value of ion exclusion effect changes (using data obtained on different RP and sulphonated PSDVB phases and using principal components` method). It is (see the second column, factor 2):
Factor Factor
1 2
Tartaric -1,70592 -,62081
Formic -1,16409 ,83542
Malic -,63685 -,21728
Citric ,53431 -1,16964
Lactic -,13280 ,52930
Acetic ,16483 ,97271
Maleic ,76046 -1,75274
Succinic ,94138 ,40294
Fumaric 1,23868 1,02009
Ion exclusion is the maximum for citric and maleic and minimal for formic acetic and fumaric. Can you take it in accordance with pKa of these acids. Or the situation`s more complex? I wonder why maleic and fumaric acids act so differently.
Constantine.
-------------------------------------------------------------------------------------------------------
By Chris Pohl on Sunday, May 23, 2004 - 06:59 pm:
Anon (5/14)
Regarding the pKa of of fluorocarboxylic acids, it's conceivable that the trifluoromethyl group could have a larger inductive effect than a single fluorine but my point was that this would be an exception to the general rule that substituents on the beta carbon have significantly less inductive effect on the pKa than is generally observed for substituents on the alpha carbon. The question was: why would one be expecting this and why when presented with data supporting the general rule are you not accepting this? Anyway, if you have a strong motivation to prove your hypothesis, why not determine pKa yourself. In general, it is difficult to determine pKa for such strong acids but for these compounds, you should be able to use either H or F NMR to determine pKa.
-------------------------------------------------------------------------------------------------------
By Chris Pohl on Sunday, May 23, 2004 - 07:23 pm:
Constantine,
While pKa is a significant factor in ion exclusion retention, hydrophobicity of the neutral form of the acid is also a major factor. From the data cited above (pKa order for aliphatic acids is: formic; 3.75, acetic; 4.76, propanoic; 4.87, butyric; 4.82) one would expect that the elution order would be formic, acetic, butyric and propionic with the latter two nearly coeluting. But in fact, butyric acid invariably elutes well after propionate under aqueous conditions, even when the pH of the mobile phase is low enough to drive both acids to the total permiation volume. In fact, unless you add solvent to the mobile phase, butyric acid always elutes after the total permiation volume, clearly showing that a "reversed phase" component to retention is also playing a role with butyric acid retention.
I suspect that the reason for the large difference between the retention of maleic and fumaric acid stems from differences in how the two molecules to interact with the pi electrons of the styrenic backbone of the ion exclusion phase. Similar differences are also seen in anion exchange separations of these two acids with styrenic ion exchange materials. In fact, this effect seems general with analogous effects being seen with cis and trans-aconitic acid.
-------------------------------------------------------------------------------------------------------
By Constantine Sychov on Monday, May 24, 2004 - 06:17 am:
Chris,
as for your first assumption - using PCA (principal components analysis) I extracted two factors (linear independent parametres)from logk` values on different phases. The first with average contribution about 70% is "reversed-phase" factor, it is factor 1 (look, values for maleic and fumaric acids are almost the same, as its going to be). The second factor 2 (contribution about 30%) reflects ANOTHER property, which may be called "ion exclusion". This property is INDEPENDENT.
As for your second assumtion. You see, maleic acid is not more retained, it is more EXCLUDED; so we observe the opposite effect.
And even if maleic acid was more retained (lets consider this theoretical possibility) - it can not be due to pi-interactions. I scaresely investigated the phenomena of pi-interactions in LC (see topic "pi-interactions" here) - they are very small in highly polar and aqueous environments even on neutral hypercrosslinked polystyrene, and they are almost negligable on sulphonated low crosslinked ones.
So the question remains. If I have time I`ll try to find pKa values for these compounds.
Anyways - thanks for discussion.
Constantine
-------------------------------------------------------------------------------------------------------
By Chris Pohl on Monday, May 24, 2004 - 08:02 pm:
Constantine,
Sorry you misunderstood my earlier message.
Regarding your first point about my "first assumption", you are reiterating precisely my point: "reversed phase" retention is a major component in the retention of many analytes in ion exclusion. I think we are agreement on this point (unless you mean that reversed phase retention is a major component for all analytes). I'm not sure why you thought I was saying these retention processes are interdependent. They are, of course, independent retention processes with each analyte exhibiting each retention mechanism to varying extents.
Second, I didn't say that maleic acid is more retained than fumaric acid due to pi electron interactions. My point was that fumaric acid (and other trans acids) exhibits an unusual amount of added retention due to pi electron interations between the analyte and the polymer backbone relative to the cis form of the same acid. In anion exchange, for example, pi electron interactions are definitely not negligable interactions in many materials based on aromatic polymers even though they may be generally "well hydrated". I think this stems from the fact that there are "knots" in the polymer matrix where hydration is extremely contrained. Simutaneous electrostatic interactions and pi electron interactions combine to enhance retention at a small number of poorly hydrated "high energy" sites. If you are interested in testing this, compare the retention via anion exchange of maleic and fumaric acid on a styrenic anion exchange material under condition where both analytes are fully ionized and the eluent species is extremely polar such as hydroxide or carbonate. In this case, you will frequently see an extreme retention difference for maleic and fumaric acids with fumaric eluting well after maleic. Under these conditions, adding even micromolar concentrations of an aromatic eluent component such as p-cyanophenol to the eluent will dramatically effect the retention of fumaric acid relative to maleic acid without significant changes in the retention of more polar analytes incapable of interacting with the pi electrons of the polymer backbone. Similar interactions are also involved in the unexpectly high retention of fumaric acid via ion exclusion.
-------------------------------------------------------------------------------------------------------
By Constantine Sychov on Tuesday, May 25, 2004 - 08:14 am:
Chris,
I will think about you have said.
Now my first reaction is that I don`t think that the reason is pi-interactions. May be we came across with intra-molecular mesomeric cis-trans effect? I think we olny have to find exact pKas for these acids to continue useful discussion. So, I`ll try, I promise.
OR you see, the experiment also must be carried out on sulphopropilated silica-gel. Mmmm... here I have problems... I have very very old ones I never checked. As far as I`m specializing on chiral separations I only have good phases of this type.
As for addition of cyanophenol... it`s amazing fact.
And, Chris, do you have some linkes or do you have already obtained k` or better jusk tR values of standard org. acids on ion exchane phases (without cyanophenol and with it)? Such information could really help to clear situation.
Constatine
-------------------------------------------------------------------------------------------------------
By Constantine Sychov on Tuesday, May 25, 2004 - 08:28 am:
Chis,
maybe it is not ION exclusion but just exclusion? Maybe cis is hydrated much harder then trans?
-------------------------------------------------------------------------------------------------------
By Chris Pohl on Tuesday, May 25, 2004 - 07:18 pm:
Constantine,
When I get back in the office on Thursday, I'll post some pKa values for maleic and fumaric acids. Any other pKa values you would like?
I'll have to look and see if I can find a convenient example of the p-cyanophenol effect to post. If not I'll see if I can generate a fresh example.
I suppose that the unusual retention of fumaric acid could, in part, have something to do with hydration but wouldn't you think that hydration of the trans form would be easier given the geometry? If so, it should have the least retention? I'll look into that too when I'm back. The limiting equivalent conductance might help sort out hydration differences.
One last point on the pi electron interactions in ion exclusion: aromatic acids all exhibit exceptionally high retention on ion exclusion columns. So much so that such columns are almost never used for these analytes without the addition of solvent.
-------------------------------------------------------------------------------------------------------
By Constantine Sychov on Thursday, May 27, 2004 - 08:56 am:
Chris,
thanks you for help in solving of such "unprogmatic", but interesting task: to deduce the origin of so called "ion-exclusion effect". Of course, I`m interested in pKa of all components listed in previous table plus pKa of oxalic acid (it is not in list for I used log a values with relation to oxalic acid to avoid problems with void volumes - all the data was taken from literature). About hydration - yes it may seem that is not good assumpion but it worthy to be checked. About pi-interactions - the retention data on SAX silica-gel is needed to close the discussion on this topic.
-------------------------------------------------------------------------------------------------------
By Constantin Sychov on Tuesday, June 29, 2004 - 03:04 am:
Hi, Chris,
the situation is following,
"ion exclusion factor" for the test mix or organic acids rather poorly correlates with their pKa values, but bad correlation ( r = 0.76) is due to two acids - citric and fumaric. Without fumaric acid (too little excluded) correlation is satisfactory r - 0.91, without citric (too much excluded) it is good r = 0.95. In fist approximation, this fact can be explained, you see, by pi-interactions and size exclusion, accordingly, but it is suggestion to be proved.
Then, the contribution of "ion exclusion" appears different for different phases, I`m sure, it depends on pore diameter and degree of sulphonation.
-------------------------------------------------------------------------------------------------------
By Chris Pohl on Tuesday, June 29, 2004 - 03:55 pm:
Constantin,
It sounds like you must have found pKa data for your test mixture. Sorry I didn't post this information as I promised. Anyway, from my point of view there are two factors involved in ion exclusion retention: ion exclusion (this component should be correlated to pKa) and the solubility parameter of the analyte relative to that of the stationary phase. I think it's a mistake to focus on any sort of pore size effect since gels of the sort used in ion exclusion materials don't have any true pores (in spite of the fact that they are typically referred to as being microporous). They are in fact true hydro-gels, fully solvated by water. Because of the extremely low pH within the stationary phase (the pH of the stationary phase should be around zero), any analyte which enters the stationary phase will be essentially 100% nonionic while in the stationary phase. For this reason, any added (or diminished) retention of the neutral form of the analyte must reflect the relative solubility of various analytes in their neutral form in the stationary phase.
From this point of view, I would interpret the higher than expected retention of fumaric acid as evidence of relatively similar solubility parameters for solute and stationary phase, thus enhancing retention. Conversely, the unexpectedly low retention for citric acid is indicative of a bigger discrepancy between the solubility parameters of the solute and the stationary phase, thus diminishing retention.
-------------------------------------------------------------------------------------------------------
By Constantin Sychov on Wednesday, June 30, 2004 - 08:28 am:
Chris,
I`ve got your point, and it is the same as mine. Firstly I didnt believe that "ion exclusion" mode for organic acids is true ion exclusion, for analytes are almost in non-ionic form. But the fact is - that lg alphfa correlates with pKa rather well (exept above mentioned couple of acids). Maybe, the property you are calling "solubility parameter" also depends on pKa?
I think, it can be easyly checked by eluting acids under neutral pH... but I didnt see any data of this kind.
-------------------------------------------------------------------------------------------------------
By Chris Pohl on Wednesday, June 30, 2004 - 09:09 am:
Constantin,
Before I posted my most recent comment above I did a bit of searching through tabulated data on solubility parameters. Solubility parameters are commonly used in polymer chemistry as a basis for predicting what solvents will dissolve polymers. Unfortunately, while I found solubility parameters for a number of common carboxylic acids I could not find these values for fumaric or citric acid. Furthermore, there is no tabulated data on the solubility parameters for sulfonated polystyrene so there's really no basis for testing this hypothesis unless you can find this sort of tabulated data elsewhere.
It seems, however, that you may have misunderstood one my points in my previous post. I would characterize ion exclusion as largely dominated by the pKa of the analyte. After all, if the analyte isn't at least partially in the neutral form in the mobile phase it will not have access to the ion exclusion gel particles. It's just that the solubility of the analyte in the stationary phase plays a secondary role in this separation process so one cannot simply predict elution order based on pKa alone.
-------------------------------------------------------------------------------------------------------
By Anonymous on Monday, July 5, 2004 - 03:03 am:
Chris,
Can you confirm your value of the pKa of methanesulfonic acid? I put this into a well-known pKa calculator and it gave me a value of MINUS 0.4. o.k. these things are sometimes wrong, but the calculator also has a database of experimentally determined values and quoted me a value of MINUS 1.86 as determined by I think Serjeant- a well known worker in the field. Well this could be wrong too; however, I did think methane sulfonic acid was stronger than you suggested. Also I don't see why propanesulfonic acid should be stronger than methanesulfonic acid, from your figures and a consideration of inductive effects.
Regards,
David
-------------------------------------------------------------------------------------------------------
By Chris Pohl on Saturday, July 10, 2004 - 02:07 pm:
David,
I agree with you that these values seem lower than expected. Perhaps there is some mistake although I have yet to find an error from this reference: Smith and Martell, Critical Stability Constants, 1977 (it's in the volume titled: Other Organic Ligands) I doublechecked these phases and again, pKa values were 1.92 for methanesulfonic acid, 1.68 for ethanesulfonic acid and 1.53 for propanesulfonic acid. By the way, the original reference that they cite for this data is: A. Covington and R. Thompson, J. Solution Chemistry, 1974, volume 3, p 603
-------------------------------------------------------------------------------------------------------
By David on Monday, July 12, 2004 - 08:50 am:
Chris,
Thanks for your answer. Here are 2 web references that quote the pKa of methanesulfonic acid as about MINUS 2.
http://www.cem.msu.edu/~reusch/VirtualText/acidity2.htm
http://www.mrgoodchem.com/Mr_%20Good%20 ... 0(MSA).htm
I thought that all these sulfonic acids were completely ionised over the range of RP-HPLC, and that was why they were chosen as ion pair reagents. Similarly, there are polymeric ion exchange HPLC columns with sulfonic acid groups(often propanesulfonic acid) and the whole idea of these groups is that they remain completely ionised over the range of operation of the column (which may be pH 1 -13 if the column is polymeric)
-------------------------------------------------------------------------------------------------------
By David on Monday, July 12, 2004 - 08:54 am:
Sorry, for some reason in the second link (MSA).htm bit missed out and the link above doesn't work directly. However, if you copy the whole lot and paste it into your browser it will work.
David
-------------------------------------------------------------------------------------------------------
By Chris Pohl on Monday, July 12, 2004 - 01:42 pm:
David,
As I said, the values from Smith and Martell do seem to indicate acidity values which are lower than expected. On the other hand, as I said, I haven't yet encountered a documented error in the data from Smith and Martell. For example, the first reference you mentioned as a number of values which do not agree with either Smith and Martell or Lange's Handbook of Chemistry (which I generally use as my primary reference for pKa values). These include discrepancies for oxalic acid listed as 1.2 although Lange's list it at 1.27 and Smith and Martell list it at 1.27 or 1.26; trifluoroacetic acid, listed in Lange's at 0.5 (several other references also list trifluoroacetic acid in the 0.3-0.4 range). I guess you could argue these aren't big discrepancies but the fact that they don't match those of the best sources makes me wonder where they got the data. All things being considered, it seems like the data in the two references you cited isn't very believable either. The value you cited at - 1.86 seems in the ballpark of what I might have expected although even this seems to be a bit extreme.
