The Chemical Manufacturing Process Route Selection: The Inclusion of Phase Switching and its Relationship to Validation
kilomentor | 20 January, 2012 11:31
Neal G. Anderson in his monograph, Practical Process Research & Development, says nothing about validation but he does speak about the need “to freeze the final process early, that is, to identify early the most desirable final chemical transformation to prepare the drug substance.” He credits this concept to P. Shutts. [“Freeze the Commercial Process-Issues and Challenges” a talk contributed at the Third International Conference on Process Development Chemistry, Amelia Island, FA, March 26 1997]. Earlier in his book, Anderson says specifically: “In order to establish routine impurity profiles and levels, ideally the final isolation of drug substance should be optimized, and this process should be used for the preparation of material destined for toxicological studies and later Phase I studies. The types of impurities found in drug substance will be largely determined by the starting materials and reagents used in the final step to prepare the drug substance. Thus the ideal penultimate compound should be identified in investigations, and parallel development work should converge on this penultimate compound.” In this particular passage that the last step Anderson is talking about is not the formation of a pharmaceutical salt. Almost certainly he is talking about a covalent bond forming step that completes the final structure. Also Anderson is speaking about the ideal situation, but what this assumes is not just that the last step, from penultimate compound to actual API, does not change, but also that this final intermediate is so efficiently purified that the steps preceding do not contribute to its impurity profile other than as inconsequential trace substances such as would be below 0.1%. It would only be in such instance that different routes to this penultimate compound would not leave behind any of their own impurities in the drug substance. If this is translated into the vocabulary of process validation, this would declare that there should be no critical steps before the purification of the ultimate intermediate! This is indeed a truly ideal situation and would constitute an impractical goal for real-life penultimate compound purifications to routinely seek.
Even though Anderson does not actually use the word validation in his book, even if we accept that validation is the most significant economic outcome of a developed pharmaceutical process; Anderson does convey the idea that certain characteristics of the final process step that simplify validation, are very, very important.
To achieve this requires a very selective choice of the penultimate compound. It must be easily and efficiently isolated and purified. As Kilomentor has argued this suggests that the step in which it is prepared and worked up involve several phase switches, because it is phase switches that provide purification and itt is compounds containing acid and/or basic functionalities that provide the most common opportunities for simple phase switches.
Even though Anderson does not actually use the word validation in his book, even if we accept that validation is the most significant economic outcome of a developed pharmaceutical process; Anderson does convey the idea that certain characteristics of the final process step that simplify validation, are very, very important.
To achieve this requires a very selective choice of the penultimate compound. It must be easily and efficiently isolated and purified. As Kilomentor has argued this suggests that the step in which it is prepared and worked up involve several phase switches, because it is phase switches that provide purification and itt is compounds containing acid and/or basic functionalities that provide the most common opportunities for simple phase switches.
General |
Permalink |
Add comment |
Trackbacks (0)
A Question from the LinkedIn Organic Process Research & Development Group that Fosters Practicing Creativity
kilomentor | 14 January, 2012 15:24
A question was asked at the LinkedIn Organic Process Research & Development Networking Group http://www.linkedin.com/groupItem?view=&srchtype=discussedNews&gid=1902030&item=88459764&type=member&trk=eml-anet_dig-b_pd-ttl-cn&ut=07TQuSi7kXnl41 “How could one remove a small amount of ethoxytrialkylsilane from a solution of ethanol containing some water at a scale of a large commercial waste stream.
I do not know whether a good answer is possible for this question. What interests me is what instruction the question can provide to students of process chemistry about how one searches for possible constructive leads for research.What is the entire real question?
Larry Fertel who posed the problem says, “I've got a stream of ethanol contaminated with <0.09 wt. % of a low boiling, trialkylethoxysilane. Any ideas on how to remove this (adsorbtion, coordination with a flocculent, etc.)? Distillation is not feasible, as the b.p.s are similar. Note that this is on a commercial scale….There actually is 9% or so water in the stream, water content is not an issue….The problem/issue is that the silane has a b.p. within 2 deg C of the azeotrope, sorry for leaving that piece of info out….Let me say that the ethanol stream is many millions of gallons per month, so this is not a research project. The ethanol is a by-product of a commercial process, so replacing it is not an option. Any treatment must work within the economics of reselling the ethanol into the marketplace (i.e., can only cost so many cents/lb.) . One option we are looking at is to convert to ethyl acetate for reselling. We are considering carbon treatment, one can build large columns in parallel that can be used on a continuous basis and get switched around once the carbon gets saturated in the impurity (assuming it works)…. I work for a contract research company, we were approached by a customer with this issue (good for us ($$), bad for them!!) So we need to be the problem solvers. I worry about the kinetics of adding a small amount of a reagent to a small amount of impurity. Considering that it is a bimolecular reaction, wouldn't the rate be slow? Not sure of the spec needed for the final product, at this point we are at the "screening" level.”
The problem is that there is <0.09% of an unacceptable compound contaminating millions of gallons per month of ethanol/water which cannot be removed by distillation because the undesired trialkethoxysilane distils close to 78.2 C, the ethanol/water azeotrope.
The problem is not a solution of ethanol/water. It is the presence of the trialkylethoxysilane at the level of <0.09%. There must be some low level that is acceptable otherwise the problem is to reach an undetectable level at low cost and that is almost certainly impossible. We must assume that the actual level is around 0.09% and it needs to be substantially reduced to let us say 0.01%
The problem is that the process as designed gives this level. It is not that that the level is around 0.09% because of some deviation in the process. There was no time when the level was acceptable. The unacceptable level is always present in the exiting waste stream from the process.
To solve the problem we might argue that we cannot remove the ethanol/water from the stream. That is we might say we cannot act on the major content of the stream because that would be expensive but apparently that is not the case. Larry Fertel says that distillation is rejected because it does not work not because it is too expensive. This opens the possibility that if we could selectively convert the trialkylethoxysilane to something separable from ethanol/water it would constitute a possible solution. Similarly it would seem that somehow changing the distillation temperature of the silane would be a possible solution and it would be particularly advantageous if it were the impurity that vaporized at lower temperature. Adsorption or adsorbtion onto or into another phase would be advantageous and probably most preferred because the predominant ethanol-water would be unchanged.
The hydrolysis and adsorption ideas have been pretty well explored by Stone, Tyrell, and others in the LinkedIn discussion. Kilomentor would like to look at the distillation and reaction possibilities a bit more.
One way to increase the degree of separation between the points of distillation of the trialkylethoxysilane and the ethanol/water is to add something to the mixture which will enhance the rate of vaporization of the more lipophilic material. This is the idea of extractive distillation. At the same time one does not want to leave anything behind in the ethanol/water that you are trying to decontaminate. Suppose we were to add an essentially immiscible perfluorocompound (expensive) or a much more volatile pentane (cheap). Would the vaporizing added compound sweep with it the undesired silane? I am reminded that there is a patent that claims the removal of traces of DMSO by co-vapor entrainment with ??.
It has already been suggested that a hydrophobic styrene-divinylbenzene cross linked resin might work. What about just using beads of solid paraffin? These would be cheap, sturdy and insoluble. They could be created by dropping molten paraffin into the stirred bulf solvent. They would float on the surface of the ethanol/water and could be filtered off and melted to drive off the impurity.
I do not know whether a good answer is possible for this question. What interests me is what instruction the question can provide to students of process chemistry about how one searches for possible constructive leads for research.What is the entire real question?
Larry Fertel who posed the problem says, “I've got a stream of ethanol contaminated with <0.09 wt. % of a low boiling, trialkylethoxysilane. Any ideas on how to remove this (adsorbtion, coordination with a flocculent, etc.)? Distillation is not feasible, as the b.p.s are similar. Note that this is on a commercial scale….There actually is 9% or so water in the stream, water content is not an issue….The problem/issue is that the silane has a b.p. within 2 deg C of the azeotrope, sorry for leaving that piece of info out….Let me say that the ethanol stream is many millions of gallons per month, so this is not a research project. The ethanol is a by-product of a commercial process, so replacing it is not an option. Any treatment must work within the economics of reselling the ethanol into the marketplace (i.e., can only cost so many cents/lb.) . One option we are looking at is to convert to ethyl acetate for reselling. We are considering carbon treatment, one can build large columns in parallel that can be used on a continuous basis and get switched around once the carbon gets saturated in the impurity (assuming it works)…. I work for a contract research company, we were approached by a customer with this issue (good for us ($$), bad for them!!) So we need to be the problem solvers. I worry about the kinetics of adding a small amount of a reagent to a small amount of impurity. Considering that it is a bimolecular reaction, wouldn't the rate be slow? Not sure of the spec needed for the final product, at this point we are at the "screening" level.”
The problem is that there is <0.09% of an unacceptable compound contaminating millions of gallons per month of ethanol/water which cannot be removed by distillation because the undesired trialkethoxysilane distils close to 78.2 C, the ethanol/water azeotrope.
The problem is not a solution of ethanol/water. It is the presence of the trialkylethoxysilane at the level of <0.09%. There must be some low level that is acceptable otherwise the problem is to reach an undetectable level at low cost and that is almost certainly impossible. We must assume that the actual level is around 0.09% and it needs to be substantially reduced to let us say 0.01%
The problem is that the process as designed gives this level. It is not that that the level is around 0.09% because of some deviation in the process. There was no time when the level was acceptable. The unacceptable level is always present in the exiting waste stream from the process.
To solve the problem we might argue that we cannot remove the ethanol/water from the stream. That is we might say we cannot act on the major content of the stream because that would be expensive but apparently that is not the case. Larry Fertel says that distillation is rejected because it does not work not because it is too expensive. This opens the possibility that if we could selectively convert the trialkylethoxysilane to something separable from ethanol/water it would constitute a possible solution. Similarly it would seem that somehow changing the distillation temperature of the silane would be a possible solution and it would be particularly advantageous if it were the impurity that vaporized at lower temperature. Adsorption or adsorbtion onto or into another phase would be advantageous and probably most preferred because the predominant ethanol-water would be unchanged.
The hydrolysis and adsorption ideas have been pretty well explored by Stone, Tyrell, and others in the LinkedIn discussion. Kilomentor would like to look at the distillation and reaction possibilities a bit more.
One way to increase the degree of separation between the points of distillation of the trialkylethoxysilane and the ethanol/water is to add something to the mixture which will enhance the rate of vaporization of the more lipophilic material. This is the idea of extractive distillation. At the same time one does not want to leave anything behind in the ethanol/water that you are trying to decontaminate. Suppose we were to add an essentially immiscible perfluorocompound (expensive) or a much more volatile pentane (cheap). Would the vaporizing added compound sweep with it the undesired silane? I am reminded that there is a patent that claims the removal of traces of DMSO by co-vapor entrainment with ??.
It has already been suggested that a hydrophobic styrene-divinylbenzene cross linked resin might work. What about just using beads of solid paraffin? These would be cheap, sturdy and insoluble. They could be created by dropping molten paraffin into the stirred bulf solvent. They would float on the surface of the ethanol/water and could be filtered off and melted to drive off the impurity.
Remember guys, during the idea generation stage of problem solving criticism is supposed to be put on hold!
Now what can we get by playing with the idea of reaction of the impurity? What distinguishes the undesired impurity from the bulk? Molecular weight. The presence of silicon. The presence of a silicon oxygen bond. The presence of carbon-silicon bonds. The presence of silicon in the impurity seems to jump out at me. What is the strongest bond that silicon can form because that is what I want to do if I am going to drive this product into some other substance? The strongest silicon bond is one to fluorine. So perhaps the ion exchange resins that are being contemplated should have fluoride ion loaded on them. The conversion R3SiOEt + F- + H2O going to R3SiF + H-OEt + OH- would have an enthalpy advantage of about 27 kcal/mole. Hydrolysis in contrast has no enthalpy advantage, I think.
Anyway, problem solving is a combination of a good statement of the problem, a non-critical atmosphere for generating ideas, and most important well loaded retrievable associative memory banks. Note that you cannot come up with good ideas based on stuff you can’t recall but need to look up. Looking stuff up can be used to refine ideas but it cannot help to create them. You need to know things well enough that they can be recalled by associative memory when presented with the clues offered by the problem itself.
General |
Permalink |
Add comment |
Trackbacks (0)
The Strategy of using Split Runs in Chemical Process Development
kilomentor | 23 December, 2011 15:25
Split run does not appear to be a widely understood term; however, in my personal chemical process development experience it is a common term with a well defined meaning. Split runs are experiments done to test different work-up/isolation and/or purification procedures starting with equal portions from a single large reaction mixture. As a consequence, even if a reaction mixture is both complicated and uncharacterized, a chemist can compare different follow-on procedures for working it up without being concerned about reaction differences that may still exist from batch to batch because the reaction process itself has not yet been finalized and critic parameters completely controlled.
In the general case, the isolated yield fraction, (which when expressed as a percentage is often called simply the percentage yield), is the product of the reaction yield fraction (determined by an assay of the crude reaction mixture before work-up) multiplied by the isolation yield fraction (the fraction representing the effectiveness of the recovery of pure product from the crude reaction mixture). In the case of comparing split runs, the ratio of isolated split run yields is proportional to the ratio of the isolation yields because the pure reaction yield is the same for each, since they come from the same reaction batch. Thus, in comparing the recovered product from split runs, the difference represents only the effectiveness of the isolation protocols….it is independent of and not confounded by the exact details of the reaction conditions. In this way, even at an early stage in developing the particular reaction step, the development chemist can get some idea of the comparative effectiveness of different treatments of the crude reaction mixture.
Of course it is not just the best isolation yield that is of interest and of importance to the process development chemist but also the relative purities of the two products and the amounts of different impurities within each.
The benefit in using split runs results because the optimization of the reaction conditions and the screening of potential isolation methodologies can proceed in parallel. It is not necessary to wait until the optimized processing conditions have been selected to work on the best isolation/purification means. The work can proceed more quickly and an earlier completion date can be set for some project milestones.
This does not mean that reaction optimization and isolation optimization can be completely decoupled. This can be seen to be intuitively obvious if one realizes that in an extreme case, if the assay yield in the reaction becomes 100%, there will be no residual starting material and no by-products formed and so this could simplify the work-up beyond all normal proportion and perhaps allow the most mundane isolation to give a fine outcome.
In the general case, the isolated yield fraction, (which when expressed as a percentage is often called simply the percentage yield), is the product of the reaction yield fraction (determined by an assay of the crude reaction mixture before work-up) multiplied by the isolation yield fraction (the fraction representing the effectiveness of the recovery of pure product from the crude reaction mixture). In the case of comparing split runs, the ratio of isolated split run yields is proportional to the ratio of the isolation yields because the pure reaction yield is the same for each, since they come from the same reaction batch. Thus, in comparing the recovered product from split runs, the difference represents only the effectiveness of the isolation protocols….it is independent of and not confounded by the exact details of the reaction conditions. In this way, even at an early stage in developing the particular reaction step, the development chemist can get some idea of the comparative effectiveness of different treatments of the crude reaction mixture.
Of course it is not just the best isolation yield that is of interest and of importance to the process development chemist but also the relative purities of the two products and the amounts of different impurities within each.
The benefit in using split runs results because the optimization of the reaction conditions and the screening of potential isolation methodologies can proceed in parallel. It is not necessary to wait until the optimized processing conditions have been selected to work on the best isolation/purification means. The work can proceed more quickly and an earlier completion date can be set for some project milestones.
This does not mean that reaction optimization and isolation optimization can be completely decoupled. This can be seen to be intuitively obvious if one realizes that in an extreme case, if the assay yield in the reaction becomes 100%, there will be no residual starting material and no by-products formed and so this could simplify the work-up beyond all normal proportion and perhaps allow the most mundane isolation to give a fine outcome.
General |
Permalink |
Add comment |
Trackbacks (0)
How thinking about Separating Acetic acid and Acetic Anhydride can sharpen our Chemical Process Development Skills
kilomentor | 16 December, 2011 10:50
Ganesh Wagh in the Chemical Process Development Group on LinkedIn asked the question, “What is the best way to separate a mixture of 60% acetic anhydride and 40% acetic acid to recover the acetic anhydride?” This solicited quite a few responses. The particular forum is now closed and although Kilomentor did contribute one suggestion to the forum I think the topic deserves some further discussion that doesn’t fit well into the question-comments format.
A plurality of people suggested either distillation of distillation under reduced pressure. These solutions may very well be the best if one has the proper fractional distillation equipment and it is not busy performing more valuable separations, but there is the implication in the question itself that, at least in the questioners hands, there was some difficulty with these solutions, since they are the first thing to come to a process chemist’s mind. Indeed Ganesh later noted that he did encounter problems in his hands.
Several people (Antonio Manka, Anton Frenkel, Marta Krawczyk, Pravinchandra Vankawala) suggested fractional distillation, either with some variation of the distillation condition for atmospheric pressure or under reduced pressure. I am not an expert on distillation. That is perhaps why it is not clear to me why distillation at reduced pressure would be expected to simplify separating these two liquids compared to an atmospheric pressure fractionation. Also, whatever the pressure under which the fractionation is done, these methods appear to require the complete removal of the acetic acid before the acetic anhydride distillation and that must leave some lost still bottoms and column residues that must subtract from the recovery. On top of this, I would imagine that fractional distillation must be slow and labour intensive. One suggestion, variation of the packing in the distillation column,which would change the height equivalent of a theoretical plate, is typically not readily accessible in a multipurpose plant particularly where the object is no more than reagent recycling.
Mark Bratt contributed the idea “Any thoughts as to a non-aqueous separation e.g. cyclohexane or methylcyclohexane with methanol or acetonitrile, with an organic base?” Although I don’t endorse trying a non-aqueous, two liquid extraction that uses methanol, which would be both miscible with and reactive towards acetic anhydride (this is almost certainly just a slip – it is true that methanol is very frequently a partner phase in such applications), the idea is suggestive of a type of strategy that is adapted and taken up by others later in the discussions. Perhaps we could simplify this idea. What would happen if we just added enough triethylamine to neutralize the acetic acid? Wouldn’t the triethylammonium acetate form a separate liquid phase aside from the acetic anhydride? Even if it didn’t, wouldn’t distillation of this mixture amount to an reactive distillation in which the acetic acid would be held back by its reversible interaction with the triethylamine allowing the acetic anhydride to distil over cleanly first? Acetic acid and triethylamine are reported to give a higher boiling azeotrope, with composition 69% acetic acid 31% triethylamine, that distils at 163 C. If triethylamine didn’t do the job what about neutralization with tributylamine?
Kees Hoek made a suggestionto add acetyl chloride that would, if it is worked, convert the acetic acid back to acetic anhydride thereby, in principle, multiplying the acetic anhydride recovery. My rough bond energy calculation for this reaction between acetyl chloride and acetic acid gives an enthalpy difference of close to zero so if the kinetics were practical it most likely would produce an equilibration of starting material and products. Driving out the hydrogen chloride would be expected to drive the equilibrium towards acetic anhydride according to LeChatelier. The usefulness might boil down ( pun intended) to how fast one can scrub hydrogen chloride. It might be attractive if one also had in the plant an alkaline waste that needed neutralizing.
Pedro Guivisdalsky, among other ideas, suggested adding toluene to form the binary lower boiling azeotrope with acetic acid that distils at 105.4 C. This is a variant on the extractive distillation idea whereby the acetic acid is made relatively more volatile. Waste toluene could be used and the toluene/acetic acid washed to recover the toluene.
John Knight, Scientific Director at Scientific Update, suggests neutralizing the acetic acid using the completely non-volatile basic ion-exchange resin. As he comments, this risks the exotherm that accompanies every mode of purification that features a neutralization. It would also involve bashing around a lot of insoluble resin that may be ground down and give some tedious still bottoms. Nevertheless, the core idea is that insoluble nonvolatile ionic exchange resins should come to mind when neutralization is considered in separations generally. Also, Dr. Knight again relates back to the idea of using some agent to convert the acetic acid to acetic anhydride. “What about phosphorus pentoxide”, I am musing to myself? The coproduct will be non-volatile and can be rinsed out of the still-pot after careful hydrolysis.
Many contributors to the discussion recommend looking at how acetic anhydride is made commercially as a pointer to an economic method. This is of course excellent advice since the cheapest starting material for acetic anhydride must be acetic acid. The methodology that is used for commercial product may of course be out of the question for recovering usable acetic anhydride in a multipurpose plant where the cost reduction for an accompanying process may be the actual motivation for the question.
Ilya Avrutov seems to have garnered the most endorsements for his suggestion that the acetic acid in the mixture could be made non-volatile by treatment with sodium carbonate, preferably anhydrous sodium carbonate. Others would include a drying agent. All understand the problem that the carbonate leads to water that must be taken into account. Ilya’s concept is another variant of a non-aqueous neutralization like that of Mark Bratt, above; it is simplified by not requiring further organic solvent additions but complicated because the coproduct of his neutralization can react with acetic anhydride to regenerate acetic acid.
This in turned stirs up another pertinent question. Are there cheap agents that can neutralize acetic acid without producing something that would react with acetic anhydride? Two chemicals that come to mind are calcium hydride and calcium carbide. These substances should come to mind whenever one wants to generate calcium salts of carboxylic acids for isolation or derivatization. In this instance the coproduct with calcium hydride would be hydrogen gas which would be innocuous but needs to be treated carefully. Using calcium carbide the byproduct would be acetylene gas that needs to be recognized and handled safely. In each case exactly how these solids are contacted with the acetic acid/acetic anhydride mixture would need to be worked out carefully. I suggest adding either as a slurry in more acetic anhydride.
I think I can still take credit for the most off-the-wall suggestion in the discussion. It was based on a simple question. What is the relation between the water hydrolysis of acetic anhydride and pH? Couldn't one stir the mixture of acetic anhydride and acetic acid with an aqueous salt solution buffered at the most unreactive pH and quickly extract the acetic acid into the aqueous phase? Stirring should be gentle to reduce interfacial contact and the temperature should be kept low to minimize reaction, and of course, separation of the layers should be done promptly. The idea here is that the difficulties that often stop us from reaching a simple solution simply may be mistaken. Here we assume that acetic anhydride will react with water and make a water wash unthinkable. But how well does this reaction actual take place and can it be slowed down enough to perform washing? Remember the Hinsberg reaction of a sulfonyl chloride is performed in contact with an aqueous alkaline solution!
Anyway, there were lots of ideas proffered. Thanks to everyone who contributed. My apologies if I created any distortion in presenting your contribution. The discussion was at http://www.linkedin.com/groupAnswers?viewQuestionAndAnswers=&discussionID=83697798&gid=1902030&commentID=60647184&goback=%2Egmr_1902030%2Eamf_1902030_76918230%2Eamf_1902030_49383814&trk=NUS_DISC_Q-subject#commentID_60647184
If you want to blast me at length please send e-mail to kilomentor@sympatico.ca. Short blasts can go in the comments here at kilomentor.
A plurality of people suggested either distillation of distillation under reduced pressure. These solutions may very well be the best if one has the proper fractional distillation equipment and it is not busy performing more valuable separations, but there is the implication in the question itself that, at least in the questioners hands, there was some difficulty with these solutions, since they are the first thing to come to a process chemist’s mind. Indeed Ganesh later noted that he did encounter problems in his hands.
Several people (Antonio Manka, Anton Frenkel, Marta Krawczyk, Pravinchandra Vankawala) suggested fractional distillation, either with some variation of the distillation condition for atmospheric pressure or under reduced pressure. I am not an expert on distillation. That is perhaps why it is not clear to me why distillation at reduced pressure would be expected to simplify separating these two liquids compared to an atmospheric pressure fractionation. Also, whatever the pressure under which the fractionation is done, these methods appear to require the complete removal of the acetic acid before the acetic anhydride distillation and that must leave some lost still bottoms and column residues that must subtract from the recovery. On top of this, I would imagine that fractional distillation must be slow and labour intensive. One suggestion, variation of the packing in the distillation column,which would change the height equivalent of a theoretical plate, is typically not readily accessible in a multipurpose plant particularly where the object is no more than reagent recycling.
Mark Bratt contributed the idea “Any thoughts as to a non-aqueous separation e.g. cyclohexane or methylcyclohexane with methanol or acetonitrile, with an organic base?” Although I don’t endorse trying a non-aqueous, two liquid extraction that uses methanol, which would be both miscible with and reactive towards acetic anhydride (this is almost certainly just a slip – it is true that methanol is very frequently a partner phase in such applications), the idea is suggestive of a type of strategy that is adapted and taken up by others later in the discussions. Perhaps we could simplify this idea. What would happen if we just added enough triethylamine to neutralize the acetic acid? Wouldn’t the triethylammonium acetate form a separate liquid phase aside from the acetic anhydride? Even if it didn’t, wouldn’t distillation of this mixture amount to an reactive distillation in which the acetic acid would be held back by its reversible interaction with the triethylamine allowing the acetic anhydride to distil over cleanly first? Acetic acid and triethylamine are reported to give a higher boiling azeotrope, with composition 69% acetic acid 31% triethylamine, that distils at 163 C. If triethylamine didn’t do the job what about neutralization with tributylamine?
Kees Hoek made a suggestionto add acetyl chloride that would, if it is worked, convert the acetic acid back to acetic anhydride thereby, in principle, multiplying the acetic anhydride recovery. My rough bond energy calculation for this reaction between acetyl chloride and acetic acid gives an enthalpy difference of close to zero so if the kinetics were practical it most likely would produce an equilibration of starting material and products. Driving out the hydrogen chloride would be expected to drive the equilibrium towards acetic anhydride according to LeChatelier. The usefulness might boil down ( pun intended) to how fast one can scrub hydrogen chloride. It might be attractive if one also had in the plant an alkaline waste that needed neutralizing.
Pedro Guivisdalsky, among other ideas, suggested adding toluene to form the binary lower boiling azeotrope with acetic acid that distils at 105.4 C. This is a variant on the extractive distillation idea whereby the acetic acid is made relatively more volatile. Waste toluene could be used and the toluene/acetic acid washed to recover the toluene.
John Knight, Scientific Director at Scientific Update, suggests neutralizing the acetic acid using the completely non-volatile basic ion-exchange resin. As he comments, this risks the exotherm that accompanies every mode of purification that features a neutralization. It would also involve bashing around a lot of insoluble resin that may be ground down and give some tedious still bottoms. Nevertheless, the core idea is that insoluble nonvolatile ionic exchange resins should come to mind when neutralization is considered in separations generally. Also, Dr. Knight again relates back to the idea of using some agent to convert the acetic acid to acetic anhydride. “What about phosphorus pentoxide”, I am musing to myself? The coproduct will be non-volatile and can be rinsed out of the still-pot after careful hydrolysis.
Many contributors to the discussion recommend looking at how acetic anhydride is made commercially as a pointer to an economic method. This is of course excellent advice since the cheapest starting material for acetic anhydride must be acetic acid. The methodology that is used for commercial product may of course be out of the question for recovering usable acetic anhydride in a multipurpose plant where the cost reduction for an accompanying process may be the actual motivation for the question.
Ilya Avrutov seems to have garnered the most endorsements for his suggestion that the acetic acid in the mixture could be made non-volatile by treatment with sodium carbonate, preferably anhydrous sodium carbonate. Others would include a drying agent. All understand the problem that the carbonate leads to water that must be taken into account. Ilya’s concept is another variant of a non-aqueous neutralization like that of Mark Bratt, above; it is simplified by not requiring further organic solvent additions but complicated because the coproduct of his neutralization can react with acetic anhydride to regenerate acetic acid.
This in turned stirs up another pertinent question. Are there cheap agents that can neutralize acetic acid without producing something that would react with acetic anhydride? Two chemicals that come to mind are calcium hydride and calcium carbide. These substances should come to mind whenever one wants to generate calcium salts of carboxylic acids for isolation or derivatization. In this instance the coproduct with calcium hydride would be hydrogen gas which would be innocuous but needs to be treated carefully. Using calcium carbide the byproduct would be acetylene gas that needs to be recognized and handled safely. In each case exactly how these solids are contacted with the acetic acid/acetic anhydride mixture would need to be worked out carefully. I suggest adding either as a slurry in more acetic anhydride.
I think I can still take credit for the most off-the-wall suggestion in the discussion. It was based on a simple question. What is the relation between the water hydrolysis of acetic anhydride and pH? Couldn't one stir the mixture of acetic anhydride and acetic acid with an aqueous salt solution buffered at the most unreactive pH and quickly extract the acetic acid into the aqueous phase? Stirring should be gentle to reduce interfacial contact and the temperature should be kept low to minimize reaction, and of course, separation of the layers should be done promptly. The idea here is that the difficulties that often stop us from reaching a simple solution simply may be mistaken. Here we assume that acetic anhydride will react with water and make a water wash unthinkable. But how well does this reaction actual take place and can it be slowed down enough to perform washing? Remember the Hinsberg reaction of a sulfonyl chloride is performed in contact with an aqueous alkaline solution!
Anyway, there were lots of ideas proffered. Thanks to everyone who contributed. My apologies if I created any distortion in presenting your contribution. The discussion was at http://www.linkedin.com/groupAnswers?viewQuestionAndAnswers=&discussionID=83697798&gid=1902030&commentID=60647184&goback=%2Egmr_1902030%2Eamf_1902030_76918230%2Eamf_1902030_49383814&trk=NUS_DISC_Q-subject#commentID_60647184
If you want to blast me at length please send e-mail to kilomentor@sympatico.ca. Short blasts can go in the comments here at kilomentor.
General |
Permalink |
Comments (3) |
Trackbacks (0)
An Example of a Possible Kilomentor Type Separation at Scale
kilomentor | 22 November, 2011 10:22
Hi Marto,
Thanks for the problem. Although you do not provide enough information to exactly identify the compounds that is OK. I understand there may be intellectual property or privacy issues. I think I can make some suggestions that are at least mentally stimulating and which certainly exemplify the kilomentor approach.
Marto has asked for suggestions of how to achieve a particular separation.
He wrote: “I am having a problem in the isolation of mixture containing biphenyl benzylacetate compound along with diarylated benzylacetate copound. I have been carrying out this reaction on 100 g- 200 g scale, the ratio of biphenyl and diaryl compounds is 10:1, isolation by recrystallization was unsuccessful. Both have similar retention factor in TLC. Suggest me any purification method for the isolation of these two compounds.”
First, I would hydrolyze the acetates. When compounds are strongly hydrophobic, separation should most often be tried on the most polar functional groups available.
In the case where there is a difference between the number of substituents ortho to the primary alcohol function, the compounds can probably be separated by the difference in the rates they form the O-sulfate derivatives. Sulfonate the mixture with only enough sulfonating agent ( say chlorosulfonic acid in one of the solvents pyridine or dimethylaniline; see the kilomentor blog on making O-sulfonates). The sulfonated material and the non-sulfonated are easily separated by acid-base extraction and the sulfonation is easily reversedto give two separated alcohols.
In the situations where there is no difference in the number of ortho substituents between the alcohols one must depend upon a difference in inductive and electronic effects and the best chance comes when the primary alcohol mixture is converted into two carboxylic acids. The required hydrolysis and oxidations of the esters can be done together without isolation and the mixture of two carboxylic acids can be obtained cleanly and easily by simple acid-base extraction.
I think there will be a good chance that these acids can be separated by what is called extractive crystallization. (kilomentor has also written a blog about extractive crystallization). Extractive crystallization works by taking advantage of any small difference in pKas of the two acids augmented by any small difference in the solubilities of the two acids in an immiscible organic solvent selected to augment this pKa difference. The goal is to get one compound overwhelmingly as a salt in water and the other compound overwhelmingly as free acid in the organic solvent.
In any case, even if this cannot be made to work, as free acids there are many more salts, both organic and inorganic, that are accessible to find substances that do fractionally crystallize.
Also when one has a mixture of acids one can screen a portfolio of enzymes to find one that esterifies only one of these acids. Thus one uses the most discriminating of reagents, enzymes, catalytically in their favorite reversible reaction, ester formation/ hydrolysis.
Once separated the alcohol-acetates can easily be reformed if that is what is required. The extra steps in both directions are simply probably high yield and the separations are trivial acid/base extractions.
If you are reforming the acetates remember that the ester (non alcohol) can probably be easily separated from any residual alcohol by treatment with lithium bromide or calcium bromide or calcium chloride in hexanes where the alcohol is likely to form an insoluble inorganic complex leaving the ester in hexane solution.
Thanks for the problem. Although you do not provide enough information to exactly identify the compounds that is OK. I understand there may be intellectual property or privacy issues. I think I can make some suggestions that are at least mentally stimulating and which certainly exemplify the kilomentor approach.
Marto has asked for suggestions of how to achieve a particular separation.
He wrote: “I am having a problem in the isolation of mixture containing biphenyl benzylacetate compound along with diarylated benzylacetate copound. I have been carrying out this reaction on 100 g- 200 g scale, the ratio of biphenyl and diaryl compounds is 10:1, isolation by recrystallization was unsuccessful. Both have similar retention factor in TLC. Suggest me any purification method for the isolation of these two compounds.”
First, I would hydrolyze the acetates. When compounds are strongly hydrophobic, separation should most often be tried on the most polar functional groups available.
In the case where there is a difference between the number of substituents ortho to the primary alcohol function, the compounds can probably be separated by the difference in the rates they form the O-sulfate derivatives. Sulfonate the mixture with only enough sulfonating agent ( say chlorosulfonic acid in one of the solvents pyridine or dimethylaniline; see the kilomentor blog on making O-sulfonates). The sulfonated material and the non-sulfonated are easily separated by acid-base extraction and the sulfonation is easily reversedto give two separated alcohols.
In the situations where there is no difference in the number of ortho substituents between the alcohols one must depend upon a difference in inductive and electronic effects and the best chance comes when the primary alcohol mixture is converted into two carboxylic acids. The required hydrolysis and oxidations of the esters can be done together without isolation and the mixture of two carboxylic acids can be obtained cleanly and easily by simple acid-base extraction.
I think there will be a good chance that these acids can be separated by what is called extractive crystallization. (kilomentor has also written a blog about extractive crystallization). Extractive crystallization works by taking advantage of any small difference in pKas of the two acids augmented by any small difference in the solubilities of the two acids in an immiscible organic solvent selected to augment this pKa difference. The goal is to get one compound overwhelmingly as a salt in water and the other compound overwhelmingly as free acid in the organic solvent.
In any case, even if this cannot be made to work, as free acids there are many more salts, both organic and inorganic, that are accessible to find substances that do fractionally crystallize.
Also when one has a mixture of acids one can screen a portfolio of enzymes to find one that esterifies only one of these acids. Thus one uses the most discriminating of reagents, enzymes, catalytically in their favorite reversible reaction, ester formation/ hydrolysis.
Once separated the alcohol-acetates can easily be reformed if that is what is required. The extra steps in both directions are simply probably high yield and the separations are trivial acid/base extractions.
If you are reforming the acetates remember that the ester (non alcohol) can probably be easily separated from any residual alcohol by treatment with lithium bromide or calcium bromide or calcium chloride in hexanes where the alcohol is likely to form an insoluble inorganic complex leaving the ester in hexane solution.
General |
Permalink |
Comments (1) |
Trackbacks (0)
What Might be the Best Solvent for Difficult Cleaning Jobs on the Reactor Walls of a Plant Reactor
kilomentor | 24 October, 2011 07:36
The walls of a large scale reactor can sometimes be difficult to clean. The problem is compounded because they are not easily accessible and cannot be inspected closely. Methods that can be applied in the laboratory for many reasons are off limits. Scrubbing is impractical, dangerous, and potentially damaging to the equipment. What is needed is a powerful but innocuous solvent that can work by vapor condensation not just below the surface of the refluxing liquid cleaner but above the surface and on the reactor walls where the reaction mixture may have splashed, caked and baked.
In the very old literature a common, inexpensive, and innocuous compound was claimed to be the best solvent known and one that would dissolve both organic and inorganic materials; salts as well as uncharged covalent molecules. It was molten boiling acetamide. Acetamide can be synthesized in situ in the reactor by heating ammonium carbonate and acetic acid and distilling out water. This in fact is the first preparation in the First Collective Volume of Organic Synthesis. Acetamide has bp760 222.0 C ; bp100 158 C; bp40 136 C; bp20 120 C; bp10 105 C; or bp5 92 C. As a white solid it has mp 82.3 C. The solubility is 2 grams per ml of water. Acetamide has been advocated as a “green” solvent [http://acs.confex.com/acs/green07/techprogram/S3384.HTM]The ninth edition of the Merck Index describes it as “Solvent; molten acetamide is an excellent solvent for many organic and inorganic compounds. Solubilizer; renders sparingly soluble substances more soluble in water by mere addition or by fusion.” Way back in 1933, Professor O.F. Stafford of the University of Oregon wrote that acetamide dissolved more different chemicals than any other known solvent. [J. Am. Chem. Soc., 1933, 55 (10), pp 3987–3988].
Process chemist sometimes forget that their responsibility is for the minimized overall cost of the process and this is much more than the chemicals only cost. The throughput per unit time is a major factor in the overall cost. That time includes the equipment cleaning time required between batch runs and between the end of one campaign and a new one for another product to be run in the same equipment. It makes little sense to invest extensive research efforts in reducing processing time when the same throughput efficiencies can be more easily achieved by reducing cleaning time between runs.
Many plants use a standard cleaning protocol implemented as an SOP. Only when it fails to remove all the contamination are special cleaning procedures resorted to. In some cases the standard cleaning or rinsing will even exaggerate the problem. For example, in the synthesis of adamantane described in Organic Synthesis Coll. Vol. III pg. 16-19, specific instructions are provided to avoid treating the vessel with water until acetone is used first to completely remove the tar.
The development chemists are the first to get an indication that unusual cleaning difficulties could arise after certain processing. Giving the plant scale-up people a heads-up and some suggestions will improve both teamwork and overall efficiency.
In the very old literature a common, inexpensive, and innocuous compound was claimed to be the best solvent known and one that would dissolve both organic and inorganic materials; salts as well as uncharged covalent molecules. It was molten boiling acetamide. Acetamide can be synthesized in situ in the reactor by heating ammonium carbonate and acetic acid and distilling out water. This in fact is the first preparation in the First Collective Volume of Organic Synthesis. Acetamide has bp760 222.0 C ; bp100 158 C; bp40 136 C; bp20 120 C; bp10 105 C; or bp5 92 C. As a white solid it has mp 82.3 C. The solubility is 2 grams per ml of water. Acetamide has been advocated as a “green” solvent [http://acs.confex.com/acs/green07/techprogram/S3384.HTM]The ninth edition of the Merck Index describes it as “Solvent; molten acetamide is an excellent solvent for many organic and inorganic compounds. Solubilizer; renders sparingly soluble substances more soluble in water by mere addition or by fusion.” Way back in 1933, Professor O.F. Stafford of the University of Oregon wrote that acetamide dissolved more different chemicals than any other known solvent. [J. Am. Chem. Soc., 1933, 55 (10), pp 3987–3988].
Process chemist sometimes forget that their responsibility is for the minimized overall cost of the process and this is much more than the chemicals only cost. The throughput per unit time is a major factor in the overall cost. That time includes the equipment cleaning time required between batch runs and between the end of one campaign and a new one for another product to be run in the same equipment. It makes little sense to invest extensive research efforts in reducing processing time when the same throughput efficiencies can be more easily achieved by reducing cleaning time between runs.
Many plants use a standard cleaning protocol implemented as an SOP. Only when it fails to remove all the contamination are special cleaning procedures resorted to. In some cases the standard cleaning or rinsing will even exaggerate the problem. For example, in the synthesis of adamantane described in Organic Synthesis Coll. Vol. III pg. 16-19, specific instructions are provided to avoid treating the vessel with water until acetone is used first to completely remove the tar.
The development chemists are the first to get an indication that unusual cleaning difficulties could arise after certain processing. Giving the plant scale-up people a heads-up and some suggestions will improve both teamwork and overall efficiency.
General |
Permalink |
Add comment |
Trackbacks (0)
The Importance of Understanding Process Validation for Pharmaceutical Process Chemists
kilomentor | 17 October, 2011 05:15
Process validation is a legal requirement applied to active pharmaceutical ingredient manufacturers by governments and it is administered by their regulators. Process development chemists contribute to the documentation of process validation.
Philosophical Differences between Process Chemists and Regulators
Process chemists when faced with a synthesis problem ask themselves the question, “How might this be made to work?” We then seek the optimal operating parameters for a scheme that achieves the desired physical reality. Regulators look at what we have achieved and ask themselves, “How could this fail?” Instead of looking at the optimal conditions that have been found, they are looking at the limits for failure of every critical parameter of the system we have optimized and are asking how this could upset the output quality of the system? We, as creators, struggle to succeed at jigging the system to give the result we need. We struggle to understand the coded messages of our experiments to deduce what a practical process would be. The regulators analyze what we have created to see at what points it could break down. Since any process can be made to fail by deviating enough from its teaching, this thinking is alien and frustrating to us.
Drug Substance vs Drug Product
As Dr. Oljan Repic has pointed out in his book,[ Principles of Process Research and Chemical Development in the Pharmaceutical Industry, John Wiley & Sons, Inc. p. 179-194], a chemical process for making an active pharmaceutical ingredient (API) from commercial bulk chemicals is very, very different from making a final dosage form from its API. This is because, in the former, chemical steps result in the creation of new substances by first combining a great many undesired materials (substrates, substrate impurities, reagents, catalysts, solvents etc) and then, after exposure to some reacting conditions, separating these and other undesirable materials (co-products, by-products, unanticipated and unidentified impurities) from the desired reaction product. Contrastingly, making a drug product from an API does not involve chemical reactions or purifications (other than drying). The drug product is some kind of mechanical mixture or micro-engineered assemblage of the inputs and so it is a simple combination of what is put into the process modulated by the manipulations. This is not to say that how an active ingredient is formulated into a drug product doesn’t affect the medicine’s effectiveness or even safety. It does. What I (and I think, Dr. Repic) are maintaining is that formulating is an inherently simpler change than synthesis and consequently failing to control the process of the former within the limits established is much more certain to cause the product to fail its specifications. Since there is no purification aspect in formulating, if it is garbage in, it will most certainly be garbage out. There is no possibility for purification in formulation so there is no capacity to rectify any serious deviations.
The synthesis of a pure chemical substance such as an API is nowhere near so dependent upon the starting material purity and the nitty-gritty of the process. In fact in the limit (100% purity), the properties of a pure chemical substance are completely independent both of its route of synthesis and the details of the most immediate transformation by which it is prepared. Even in the real world of actual APIs, the only evidence of the route of synthesis that remains in ‘pure’ product is the faint fingerprint of trace impurities that are still carried in it. Effective purification decreases this below any level of practical significance related to efficacy or safety and even, in many cases, below the level of analytical detection.
Even if it might come as a surprise to beginning formulators that how an API and the other excipients in a drug product recipe are combined can significantly alter the bioavailability of a tablet; it would not surprise even one tyro chemist that deviation from the details of a chemical process would cause the quality and yield of such product to suffer.
It is the Purification that Must Work
In a synthesis it is really only the final purification that needs to be effective to deliver a drug substance of the required purity, with the proviso of course that the analytical methods used to assess that purity need be validated. The process by which the final product is produced needs to be known essentially only so that the analytical methods can be tested for their capacity to detect and quantify the authentic likely/expected impurities.
This however is not the law. Regulators and politicians err on the side of caution and the validation rules they approve are more rigid than they need to be, because the rules derive historically from, and still closely resemble, those for formulated drug products. Process validation was originally conceived and implemented as a method to assure that each unit of a drug product could be guaranteed safe even when it could not be analyzed, either because the analysis would destroy the product or because the proper functioning of the product could not be analyzed. Drug security by validating the process was initially directed to drugs formulated for injection into the blood stream. If these were contaminated by bacteria the patient would likely quickly die. Even if one took a proper statistical representative sample of the product there still could be some fatally toxic units of product undiscovered. The solution was to control the process used to make the units so strictly that it would be extremely unlikely to make even one contaminated sample. This was the only course of action that could be effective because the fatal contamination came from the process (from the bacteria carried in the air) not from any starting materials.
Process validation also turned out to be valuable for assuring the efficacy of drug dosage forms where the delivery of the active ingredient from the dosage form ( ie tablet, capsule) into the blood stream of the patient was challenging. In these cases, analysis of the dosage form was not showing reliably whether a particular lot of drug product could deliver the clinically required exposure to the medicine. Drug product prepared by a validated process was found much more likely to produce the desired clinical outcome.
Process validation differs significantly from chemical analysis in its ability to provide protection from different types of contamination problems. A drug substance is almost always a pure chemical compound. It is a single molecular entity and can be analyzed; precisely, to confirm its constitution, and sensitively, to show the presence of homogeneously dispersed contaminants. Drug substances are not mechanical mixtures like flakes of this and that stirred together as is often the case for tablets and capsules. Such mixtures are statistically homogeneous only at the scale of a dose but heterogeneous at the particle scale. This is what one finds in a mixture that is pressed into a pill. Process validation is better than analysis to guarantee that heterogeneous, most often particulate, contamination is not present. The reason for this is that chemical analysis of an active pharmaceutical ingredient (API) requires taking a representative sample from the entire batch to perform the analysis upon. It is quite likely that even with an approved statistically sound sampling procedure a heterogeneously contaminated batch would not get any contaminant into the sample being analyzed. Take for example the situation where glass particles from a chipped piece of equipment contaminated a batch. The dense glass would migrate to the bottom of the container and might escape the sampling. Thus it would not be detected. A validated process, however, could make sure that there was no conceivable condition where any glass could remain in a final product. The final filtration before the final crystallization of an API is part of a process for just such a validation purpose.
Process Validation: A Development Scientist’s Most Important Activity
Drug discovery chemists do not need to learn about process validation; process development chemists do. It is the latter’s work that is being examined. Nevertheless, scientists in general always like to know whatever they can about the synthetic routes they devise and discovery chemists may eventually transition into career development chemists. The additional experimentation that process validation requires does increase scientific knowledge about that process. It answers questions about the practical limits of those process conditions that most significantly impact final product quality. But aside from the scientific merits, it is important that pharmaceutical chemists, particularly process development chemists, understand process validation, since, because the pharmaceutical business is a highly government regulated activity, any API they might make cannot be included in a medicine for sale unless it is made by a validated process. This makes validation the most economically significant test to which a scaled up industrial process can be subjected and so all the chemist’s activities should be viewed from the perspective of its contribution towards success in validation.
Many and perhaps most companies that manufacture APIs exclude scientists below the managerial level from any contact with the government regulators who assess the quality and completeness of the validation process. Likewise, it is only senior technical people who sign off on or assemble the validation report, which is the core document of the validation exercise. Yet exposing employees, who have very specialized tasks, to the all encompassing validation program lets them see their precise role in it. Specifically, chemists, who are focused on process development, are likely to do much better in organizing their data, writing reports, etc. if they understand the larger purpose their data and conclusions will serve when process validation protocols are assembled and executed and the Development Report is folded into the Validation Report. Indeed it is highly recommended that general training in the concepts of process validation should be given to all individuals who are part of the development and manufacturing program not just to those who require such training as part of Good Manufacturing Practice (GMP).
The Corporate Validation Policy
Validation is so important to API producing companies that very often it is heralded in a written Corporate Validation Policy. Where this document exists it is the overarching document pertaining to the validation methodology. As such it sets down general principles that employees should adhere to, rther than the nitty-gritty details; nevertheless, it is a good place to start in examining that most important activity called process validation.
A lifecycle policy approach to the validation concept is regarded by many experts as preferable or even essential. Validation activities begin with the inception of the development program and logically should not end until the product is retired. A corporate validation policy should provide validation expectations over the lifetime of the process. It should state the company’s expectations for both the organization’s scientific and its documentation goals. Both of these classes of goals need to be kept in mind and adhered to from the first day of process development to the last. In other words the Product Development Report (PDR) should be a ‘living’ progress report that is continuously updated through the product lifecycle. Many firms make the mistake of creating a static, concrete, dated, PDR paper which is archived once that first regulatory approval is obtained. Because improvements are being made throughout the lifetime of the process, these may require additional validation work and regulatory filings. These changes should set in motion an update of the PDR.
Since every organization has somewhat different internal structure, any policy needs to allocate responsibilities for maintaining the product/process documentation including the mature product responsibilities relating to monitoring, trending, and change control.
The Validation Master Plan
The Validation Master Plan is the most senior planning document that pertains to a specific validation. It makes concrete and project specific what is set down in the Validation Policy only in a general way. It takes into account even the most recent changes in the organization’s structure, responsibilities and reporting chains. The Validation Master Plan assigns actual project milestones to named persons and groups. It sets dates for meeting those milestones. It sets out what activities are to be performed when and it decides the science that will take place at small and at large scales. If the validation is for a new drug substance the Validation Master Plan needs to be revisited and modified as the clinical phases advance.
All aspects of the validation documentation are important. The regulators look not for what the company considers most important; what is scientifically most significant; what has been presented most clearly and completely or where the maximum inherent risk is, but what has been overlooked because that is from their perspective where that process deviation leading to dangerous medicine could most likely arise.
Some Aspects of Process Validation that are not the Development Chemists’ Responsibility
Validation must demonstrate that the process being examined is performed under good manufacturing practice (GMP). Mercifully there are many aspects of GMP that are not the responsibility of development chemists. Every organization that manufactures product for sale, makes batches for clinical (human) tests, or to support regulatory filings must have a separate and independent quality control unit to test, approve, and reject products; set and approve specifications for raw materials and products; and write and approve these and other written procedures. Another requirement of GMP is that all personnel associated with production must be trained concerning GMP and the company must be able to prove that training with the signatures of those trained. You may have been asked to submit your resume to a company representative responsible for meeting GMP standards. The company is required to show that its scientists too are adequately credentialed for their work.
The facilities and equipment to be included in a process need to be qualified before they can be used in an acceptable validation exercise. These qualifications are usually unfamiliar to practicing chemists. They are matters that are implicit in our everyday activities and are completely taken for granted by us.
Facility qualification relates to the suitability of the physical site and its infrastructure and personnel to carry out the process that is to be validated. Most of these matters are strictly the domain of architects, contractors and engineers.
Drug substances in general have requirements for the facilities in which they are manufactured. The FDA will not allow someone to make a GMP drug substance in a bathtub in a garage! It insists upon written assurance that the manufacturing facility has adequate space, heating, ventilation, plumbing and maintenance.
The Development Report
The core document that development chemists do contribute for API validation is the Development Report. Process chemists recognize the Development Report as the history of the entire process development but it is at the same time a regulatory validation document of enormous importance. Validation would need to be done, if for nothing else, at least to convince regulators that the product is reproducibly safe, but API product safety is not the only goal of process development. Optimization aims to minimize cost, maximize throughput, achieve appropriate safety, reduce environmental burden and deliver consistent, convenient operation of the steps. Consequently some IPCs and some specifications are not critical in the regulatory sense of controlling final product quality. Deviations from these do not put the API quality at risk. Some of these aforementioned standard ranges alert batch sheet reviewers, when they are changing non-randomly, to potential creeping deviations (trends) that can have economic, safety, environmental or simply consistency consequences, without there being quality ones. This means that an API could fail a noncritical IPC and still be safe and effective to use. Failing a specification or IPC that correlates with a critical process parameter (CPP) and in turn a critical quality attribute (CQA), on the other hand, would mark a failed batch.
A synthesis can in strict theory be traced back to the most elementary of chemical building blocks- even atoms themselves. What needs to be decided for validation is how many of the steps in the retrosynthetic direction need to be validated to insure the required safety. This ‘depth’ of the validation study establishes what chemical transformations or treatments are included within the validation and what steps are accepted simply as the making of starting materials to be controlled by their material specifications independent of their particular synthesis.
Today, it is a good bet that most synthetic API processes are multi-step with at least one complex chemical transformation. Similarly it is unlikely to be disputed that the final process step that creates an API, and the subsequently entailed purification, needs to be validated. Beyond this there is no consensus answer as to when to apply process validation for intermediate steps in a multi-step process. Each process it seems must be evaluated as a separate case. So many factors are in play. This does not mean that a coherent policy or procedure cannot be established for a company describing how it should do that evaluation. In fact, putting such a decision tree in place can lead to better, faster, even less expensive, choices.
The process validation for an API is at least the sum of separate validations; one for each intermediate step that contains a critical parameter. Thus determination of what steps are to be included in the full validation amounts to identifying what steps contain at least one critical parameter. This is much the same as asking what the starting materials are for the steps that will be validated because steps for making an API starting material do not need validation. Terminology can be the source of as much disagreement as the chemistry in these discussions so definitions are important. A ‘step’ leads either to an isolated or non-isolated intermediate. Other important determinations to get straight are: what is(are) the ultimate intermediate(s) or final intermediate(s), and what overall scheme of reactions and purifications will be required to produce the final API.
Determining what constitutes a starting material is debatable. According to Roger W. Koops, [Process Validation of Synthetic Chemical Processes for the Production of Active Pharmaceutical Ingredients (APIs), Journal of Validation Technology, Volume 8 Number 2 January 2002], “a starting material should be a readily available item that has a standard grade (or grades) associated with it, and has been well characterized. If produced under contract by a vendor, it should be produced by a known and established process, and the end product should be, again, capable of achieving a standard grade. Second, the vendor should be qualified (under a vendor qualification program)to to ensure that a material meets consistent quality standards. A rudimentary quality agreement should be established to outline change notification and quality requirements. What must be avoided are unshared, unannounced process changes that can change the impurity profile of the downstream drug. If a material is manufactured by a contract party and the process was supplied by the innovator, the material is simply a third-party manufactured intermediate. The contractor now becomes an extension of the innovator, and the transferred process must be included in the full validation evaluation”.
Starting materials, generally speaking, are inputs to a synthesis whose atoms are partially incorporated into the drug substance. They do not need to have their own syntheses validated. Their quality is accepted based on their certificates of analysis. In contrast, intermediates are created directly or indirectly from starting materials. They arise downstream in the process with respect to some starting materials. Most intermediates arise in steps that need validation, but at least in principle there can be exceptions. A step starting with an intermediate and resulting in another intermediate does not need to be validated if the interconversion process step has no critical parameters. That is, the step is so rugged that no matter how it is executed it does not affect the purity of the final drug product. This would be a rare case indeed. The reaction would need to be ultra tolerant of reaction parameter deviations and/or the isolation would need to be exceptionally powerful at excluding impurities.
As the overall chemical development proceeds an impurity profile for the final product should emerge The validation policy should emphasize the importance of characterizing process impurities as early in the development process as possible. From this, each intermediate process step can be evaluated as to its influence on the purity of the end product. If an intermediate needs to be controlled because it could contribute to an adverse quality profile of the API, process validation should be applied to its related process step.
Each intermediate needs to be evaluated for how it contributes to the final API, in regard to the impurity profile and the specific process for that intermediate. Besides isolated intermediates, non-isolated intermediates should also be evaluated, since steps comprising these may be ones where even stricter controls may be required. For example, an intermediate may not be isolated due to structural instability when not in solution. In this case, concentration and/or potency of the intermediate may be a critical attribute that needs to be controlled.
Either the corporate validation policy or the particular validation plan needs to provide guidance whether the validation exercise at plant scale should be designed so as to “stress” the accepted parameters by operating at the maximum or minimum accepted level (the edge of failure), or to aim for the set point target of each parameter. Some companies have attempted to matrix the parameter limits in an effort to test the extreme limits of process parameters.
This author holds the view that the laboratory development phase is the place to establish the boundaries of the operating ranges, and how combinations between extremes of parameters affect the final outcome of the process. Process validation, by definition, is used to demonstrate the limits of consistency of the process as designed and thus should aim for the set points of operating parameters. From a manufacturing standpoint, it is advantageous to operate based upon the targeted parameter values for batch consistency.
Because the Process Development Report (PDR) is such a key element of validation documentation, it is important for the corporate validation policy to set out at what points interim reports need to be submitted during process development because interim reports make the compilation of the PDR easier to manage.
The Corporate Validation Policy may need to provide some guidance concerning choosing critical parameters since their number strongly affects the breadth and hence the expense of the experimentation.
Some organizations choose to call critical only parameters, that affect the impurity profile and the ability of a material to pass specifications. However, the ability of an output product to meet specification is not the sole indicator that its process is running smoothly. There are sometimes parameters that, for example, effect yield, without affecting the final product’s properties. Nonetheless, variation of yield beyond normal experimental error points towards a process out of control. Whatever the inclusion/exclusion rules, the basis of decision can be either a general policy prescription or something considered only on a case-by-case basis at the level of the Validation Master Plan (VMP).
Validation Protocols
Validation Master Plan amalgamates validation protocols. The hallmark of a process step that is ‘under control’ is that is predictable. The sequence of operations that constitute the step are unequivocally established in advance, the acceptable operating ranges of the critical parameters are set down in advance, and the critical specifications of the output material are set down in advance. So what sets a validation batch apart from any other pilot experiment is that there can be no recourse to any procedural or corrective activity that is not set out in advance in the protocol. The organization cannot redefine success in a process step being validated after the requisite batch is complete and so eliminate deviations between what was predicted and what transpired. The regulators are the referees who decide whether, when the organization executes the protocol, it follows its own procedures and meets its own standards.
Although some experts urge that the protocol should be drafted to be ‘as concise as possible’, protocols are normally known for their verbosity. A process validation protocol will describe in excruciating detail the ‘how to’ for the validation exercise. The protocol resembles a master record in many ways but provides a greater level of descriptive detail regarding the execution of the process, the monitoring of the steps containing critical parameters and sampling instructions for the in-process checks, particularly critical in-process checks. In every case, either a procedure should be described in detail in the protocol or a reference should be provided in the protocol to an already established procedure for handling the task (ie a standard operating procedure). The instructions should not just be sufficiently unambiguous that it would direct a trained operator to choose the correct action. It should be such that a regulatory reviewer can only conceive a single proper performance of the instruction. Certain styles of drafting can condense the presentation without losing explicitness. Where applicable, data should be entered directly into the protocol. If an exercise is particularly complex, a series of protocols may be preferable. Protocols should have places in them for reviewers’ initials or signatures.
One way of looking at the components of a validation protocol is to see that they reproduce many of the parts of a summary report about a set of batches, but unlike the report the protocol is written in advance (remember after developing a process you are supposed to be able to predict everything that is crucial to product safety in advance).
The protocol should include many sections that process chemists would likely consider report padding. There should be an introductory section defining the scope of the protocol relating back to the overall Validation Master Plan. The question where does this particular protocol fit into the overall validations should also be answered. The responsibilities for execution, testing, review and approval for the overall validation exercise need to be reproduced here even if they also appear elsewhere. Also, for each particular protocol the department and job responsibilities need to be restated and the entire protocol needs to be approved by the top technical officer and the top quality officer. The purpose of the sign-off is to make unequivocally clear that top management has seen and bought into the exact protocol exercise that is to be executed.
Because a description in words alone may be more difficult to follow than an illustration, a process flow diagram (PFD) for the particular step or portion of a process that is being validated should be provided annotated with starting materials, reagents, solvents, process conditions, transfers, by-product concentrates, waste streams, and purifications etc. If there are any recycle loops or patches prepared in advance these need to be presented here in advance. The operations should be assigned an identification code so that further elaborations can be related unequivocally to the correct portion of the PFD. If seed crystals are used where they come from, what quantity they are what specifications they meet need to be set down. It is generally important and advantageous to identify critical process steps containing one or more critical process parameters on the PFD document. The idea is to give the operators no ‘wiggle room’ as to how the execution of the exercise is to proceed. This in turn gives regulators confidence that you have confidence in your process understanding and control.
A complete protocol should provide a bill of materials for every consumable item used during the process. Particular attention should be given to be sure that even materials that are not part of the chemical transformations also be there, such as filters, filter aids, packaging for chemicals before and after the main chemical transformation, seed crystals etc..Some materials have both a name and an identification code. A reference to the specification or COA should be in the listing. If the PFD has the operations in the sequence labelled, this label should be here as a cross reference.
All the equipment used in the process needs to be referenced. The actual documentation does not need to be reproduced in the report but the IQ, OQ and PQ report numbers need to be given as well as the cleaning procedures and the cleaning validation reports and instructions. The ranges of parameters, particularly any critical parameters that will be used should be noted and highlighted. Of course, the required operating range needs to be within the validated range for each equipment. If there is any special procedures that involve the equipment during the process these procedures must be detailed here or referenced accurately.
Facility documentation only needs to be referenced along with the DQ, IQ, OQ and PQ materials. The only recapitulation of the facility validation would refer to the particular importance of any unusual facility capability for the protocol being validated. One example of such a special competence would be the inclusion in the facility of Class C clean rooms.
Critical Parameters
Only at this point do we get to elements that process chemist would consider the essence of the scaled up inventiveness. Specific page references should be made to the Development Report where the critical parameters are established and the data used to establish them is organized. Various ranges of the critical parameter need to be discussed. The master procedure range is the range that will be set out in the master batch sheet. Within that range will be the set point which is the numerical setting used on the regulating equipment. Under regular operating conditions the set point can only be transiently met because of the equipment’s inherent limitations to overshoot and undershoot. The master procedure range of course cannot be less than the range set by the equipment’s inherent limitations. The next wider range is the verified range which is the range actually experienced in the plant. The Development Report may make predictions of a wider verifiable range from laboratory experiments. Using development data, a prediction will also be made of the process limits for the parameter. This is the edge of failure exceeding which the API is expect to suffer quality repercussions. Another limit that may be mentioned is the practical range of the equipment with respect to the critic parameter. For example in a steam-heated reactor the practical limit for external heating would be 150 C. For each critical parameter the controls that will be applied to maintain it need to be listed.
In-Process Tests
Every batch sheet comprises in-process tests that are used to assess the progress of the processing. A list is required of all the in process tests within the protocol showing at what point the sample for testing is taken, what the test method will be, who will perform the test and how the results will be reported. If the test is a go/no go test which is repeated until the go signal is obtained this should be explained. Achieving a particular result may be a critical parameter. If this is the case that needs to be made clear. For example neither the reaction temperature nor the time may be critical but the outcome of an in-process test for reaction completion may be critical. Thus a lower temperature and longer time or higher temperature and shorter time may each achieve the completeness required to pass a critical IPC.
Sampling Plan
Process development chemists know that sampling heterogeneous reaction mixtures does not give results representative of the mixture in its entirety. Even testing of the separated phases individually is fraught with difficulties. Solids that are to be sampled are rarely homogeneous. Even homogeneous solid mixtures can demix with mild shaking. As a consequence if one is going to have a significant analytic result, the sampling must achieve a representative sample. Within the protocol, the plan should indicate the sampling location, the amount of sample to be taken, how it is to be collected, how it is to be stabilized if necessary, how stored, and how transported to the testing facility. Both operators and regulators need to know before the exercise exactly how the sampling is planned. Of course at what point the test will be done must be indicated unequivocally. If there is a sampling deviation, it is helpful if it is reported in the batch sheet before any result for that sample is available.
Acceptance Criteria
All criteria that must be within predefined ranges must be listed in advance both critical parameters and IPC results and testing on the final product.
Deviations and Investigations
The most desirable outcome from executing a validation protocol is that all critical parameters are held within the operating range and the resultant intermediate has test measurements within the range of intermediates that have provided safe and effective IPA. That is what we aim for. What if some critical parameters are measured outside the master procedure range but not outside the verified range? What if some critical parameters have strayed outside the verified range but the intermediate seems unchanged in properties from intermediate prepared without deviations? What if the migration out of bounds was brief or very small in magnitude? What is very important is that as much as possible the assessment of the impact of a possible deviation should be made before the validation is actually executed. Identifying a parameter as critical when it is not critical is unlikely after a thorough development but getting the limits of failure somewhat incorrect is rather likely. One does not want to abandon or fail a validation batch when in fact it will result in pure safe final IPA. Falling outside of most critical parameter ranges is easily avoided but sometimes the desired operating range falls close to a limit of failure because the manufacturing needs to achieve some objective not related to final product safety or effectiveness but to minimizing cost, improving safety or reducing waste. It is in these situations where preliminary considerations of less desirable outcomes can be invaluable.
Equipment and Facility Qualification
Equipment Design qualification (DQ) relates to equipment/machines. Design qualification is normally the responsibility of the equipment vendors. Successful Installation qualification (IQ) demonstrates that the equipment has been properly installed. Operational qualification (OQ) demonstrates that the equipment can achieve the operating parameters intended when exposed to typical substrates. An OQ does not not need to specifically test the conditions of a particular process. That is the function of performance qualification (PQ). In general, it is best to have all facility and equipment qualifications completed prior to process validation (certainly all IQ/OQ should be completed). Performance Qualification (PQ) can be performed during process validation, but it is generally advisable to have completed this testing prior to process validation since it would add additional risk to the exercise. However, there may be some instances where it is useful to concurrently perform an equipment or facility PQ during process validation. In those cases, a separate PQ protocol should be drafted and referenced in the process validation protocol.
Analytical Validation
The validation of the required analytical methods that are used to follow the trends of the critical parameters during the process experimentation may be regarded as a sub-program to the overall validation project. Secure conclusions can only be derived from reliable data collected using validated methods, so methods validation needs to proceed hand-in-hand with the examination of the chemical transformation processes themselves.
Cleaning Validation
Equipment cleaning methods need to be included as part of the Validation Report. Cleaning validation is also sufficiently complex that it is better treated as a sub-discipline with its own report sub-section. Some API producers prefer to have standard cleaning procedures for their equipment which are always executed irrespective of the process that is being cleaned up. These methods are part of the SOPs for operations. Where these methods alone cannot be expected to satisfy the cleaning requirements additional cleaning may be set in place as part of the Master Batch Record following a particular exposure. Another organization might specify the cleaning treatment based on what the chemical step has exposed equipment to. That decision can be part of the Corporate Validation strategy. Cleaning methods can be developed in parallel with the chemical process development and cleaning validation can conveniently occur either prior to or concurrent with process validation. These decisions can also be policy decisions or part of a Validation Master Plan for a particular validation.
Cleaning of reactors in which any broadly insoluble, polymeric, or tarry by-product is formed can expend valuable reactor time and penalize throughput. When process chemists include such steps in a process going to scale up they should take a special responsibility to work with production personnel to develop efficient cleaning or at least pass along a warning of the difficulty and whatever they already know about overcoming it.
Conclusion
In conclusion, although process chemists need not become validation experts and although they may not be personally enmeshed in the validation exercise, some overall familiarity with the subject, from their own perspective and emphasizing their unique role, can enhance their work and their enjoyment of it.
General |
Permalink |
Add comment |
Trackbacks (0)
Is Glycerin a General Green Solution to Solvent Recovery and Replacement for Process Chemistry?
kilomentor | 13 June, 2011 07:15
At laboratory scale, when one solvent needs to be replaced with another, the solution contents are placed in a r.b. flask, set spinning on the vacuum rotary evaporator with appropriate heating and condensing facilities and when the first solvent has been completely evaporated then the required new replacement solvent is added and the solutes brought back into solution by swirling, scrapping and heating. There is no equivalent procedure for solvent exchanges at scale. In the plant situation, because stirring and heat transfer become ineffective below the minimum stirrable volume, the solvent and substrates combined volumes can never be taken below the minimum stirrable volume. Thus, solvent is never completely removed so no essentially solvent free state arises. When the solvent replacement is of a more volatile one with a less volatile one, the latter solvent can be used to chase the former in some form of distillation; but when the replacement solvent is of more volatile one, tricks, such as using azeotropes, come into play. Even where an azeotrope can work there are still fractions of mixed solvents that become environmentally troublesome waste. It is proposed that a general solution to this problem is to use the minimum stirrable volume of glycerin as a chaser for the first solvent. Glycerin is cheap, biodegradable and has a bp of 182C @ 10 mm Hg. It would be expected to remain behind in a standard distillation when combined with any of the common organic reaction solvents. Even DMSO (bp 189), DMF (bp153), NMP (bp 81-82 10mm hg) would be expected to be chased by glycerin. All the first solvent could be distilled because the volume in the reactor would not go below the minimum stirrable volume. The less volatile solutes from the starting solvent would be retained in the glycerol although not necessarily kept in solution. Advantageously glycerol is immiscible with many common organic solvents such as hexane, methylene chloride, acetone, chloroform, benzene, probably toluene. Thus the solutes can be extracted from the glycerin phase into the new lower boiling solvent. If the transfer is inhibited by partition coefficients, the glycerin can be diluted with some water to perhaps improve the transfer. Whatever glycerin is taken into the new lower boiling solvent can be removed using a drying agent that strongly binds it such as calcium chloride. The distilled first solvent contaminated with traces of glycerin upon simple treatment might be ready for reuse. Thus the first solvent is no longer a waste and there are no mixed fractions of solvents to dispose of. The waste glycerin is a biodegradable material and the quantity used is no more than the minimum stirrable volume of the reactor.
General |
Permalink |
Comments (8) |
Trackbacks (0)
More Information about using Inorganic Salt Complexes to simply Purify Alcohols at Scale
kilomentor | 16 May, 2011 08:19
Barry Sharpless et al. were the first to teach the separation of alcohol mixtures and alcohols from non-alcohols using calcium and manganese salt complexes. The original J. Org. Chem. paper [J. Org. Chem. 40(9) 1252=-1257 (1975)] unluckily recommended as a most preferred modality, using one mole of anhydrous calcium chloride per two moles of alcohol. This turns out to have been a good pressumption based on theory, but not such a good practical recipe. Soon after Sharpless’s publication, Cook et al. [Biochemical and Biophysical Research Commun. 68(1) 143-148 (1976)] showed in a crystal structure that 2:1 alcohol:calcium would be theoretically expected. Mentally extrapolating from what the Cook paper says about the probable structures, strings of calcium chloride ions are interspersed by layers of hydrophobic organic substructures where the thickness of these hydrophobic layers is somewhat related to the length and shape of the hydrocarbon substructures. How fast and how purely these clay-like layers form probably depends upon how rapidly the alcohols get into the layers and how quickly they sort themselves out between layers. This is however conjecture. In any case, in the light of more experience, this 2:1 stoichiometry recommendation turned out to be poor advice, Professor Sharpless has told me. Large excesses of calcium chloride, generally, work best.
It is not just some alcohols that complex inorganic salts such as calcium chloride, calcium bromide, manganese chloride, lithium bromide etc. ; all alcohols do but there are differences in affinities and separation is only possible where precipitate/crystallize occurs. This formation of solid complexes by stirring the substrate mixture, and the inorganic salts together with an appropriate organic liquid (most often hexane) depending upon the specic situation may be driven by relative rates or thermodynamics.
Alcohols are not the only functional groups that form complexes with such inorganic salts, many other functional groups do, but not as strongly and they don’t often precipitate. The interaction between these other functional groups do cause the inorganic salts to dissolve as some sort of soluble complexes or miscelles but precipitation/crystallization is not typically observed.
Whether an insoluble complex forms depends also on the character of the salt chosen to combine with it. L-menthol for example formed complexes with calcium bromide and manganese chloride but apparently not with calcium chloride.
Lower alcohols seem to enhance the rates of complexation perhaps by getting the inorganic salts dissolved as complexes in the organic liquid although there must be something more than this because ethanol catalyst was used in the exchanges of Table 1 in the J. Org. Chem. paper even though the starting complex would be somewhat soluble in the hexane liquid. Yet still catalysis is found to be useful.
Based on this preliminary understanding, certain questions that affect the ruggedness and the practical breadth of the method come to mind.
Are 1,2- or 1,3-diols very strongly and dependably complexed and precipitated? It would be useful to have an almost generally useful method to drop out such compounds from a reaction mixture.
Will the presence of other complexed functional groups in the compounds to be separated tend to screw up such a complexing/ precipitating scheme?
For example, the geraniol/citronellol mixture works since the other functional groupt here (alkene) does not form complexes; but what if both molecules contained alcohols and say ketones? Ketones do form complexes but the complexes don’t seem to precipitate (although cholestenone does form an isolable complex with calcium bromide). Actually, not all ketones appear to form complexes. The solvents methyl isobutyl ketone (MIBK) and methyl amyl ketone can be used as the liquid phases for forming complexes and so must participate very poorly (or it would swamp the effect).
Other examples of complexes between inorganic salts and organic compounds containing an alcohol group were known. In BE1555968 reference is made to isolating sterols by heating in hydrocarbon solvents with 14-16 molar equivalents of anhydrous zinc chloride and isolating the solid. This is referenced to in DE827199. Also in BE1164769, there is described a process for isolating sterols from mixtures by dissolving the mixture in a hydrocarbon, adding to the solution an aqueous solution of a metal salt suitable for complex formation, removing the water to a great extent by azeotropic distillation, and after cooling the mixture, isolating and splitting up the precipitated adduct in the usual way.
In the patent BR1555968, 3-hydroxy steroids and 3 oxo-steroids in various natural oil mixtures were separated by treatment with calcium bromide dehydrate in MIBK or n-amyl methyl ketone. It is unclear why the 3-oxo-steroids gave complexes under these conditions since all other literature suggests that ketones form weak complexes and/or these complexes are soluble and do not precipitate.
Kilomentor thanks Professor Sharpless for answering questions about this technique and for directing me to the Cook et al. paper.
General |
Permalink |
Comments (1) |
Trackbacks (0)
Saturated Calcium Chloride Brine for Separations?
kilomentor | 11 May, 2011 15:54
Calcium chloride is very soluble in water. It is a slow, inefficient but ‘thirsty’ drying agent. In moist air the anhydrous solid is deliquescent. Quite a few solvents can be expected to be immiscible with its aqueous brine solution and these could serve as an immiscible partner phase for liquid-liquid extractions. Saturated aq. calcium chloride can be expected to behave similarly to saturated sodium chloride solution; it forms a separate phase with isopropanol or pyridine, liquids with which water alone is miscible. But what molecular structural types would interact and pass into a calcium chloride brine that would not as easily or more easily dissolve in plain water?
This question seems appropriate because it would be logical to think that the complexing power of calcium ion would be largely erased when comingled with so much of the Lewis base, water. Yet this is apparently not so, if expired patent US 3225113 is to be believed. According to this patent, calcium chloride brine forms complexes with hydrocarbon ethers, when the ratio of the number of carbons oxygens is <8:1. The best complexes are when the ratio is 2:1 as in polyethylene glycol ethers.
This complexing could be useful in work-ups that require a solvent switch where an ether or glyme solvent needs to be removed in favour of another solvent. Thus, when using soluble polyethylene glycol polymer as solvent/diluent in synthesis there may be an option besides the prior art procedure of precipitating these polymer with hexanes or diethyl ether. Aqueous saturated calcium chloride is very likely to precipitate the glyme as a filterable solid.
Patent US3225113 shows that a solution of calcium chloride in water can precipitate quantitatively glyme solvents in the presence of another hydrocarbon-like substance. Filtration and washing the solid with the non-polar hydrocarbon cleans up the glyme calcium chloride complex. The aqueous and hydrocarbon phases can be themselves separated since they are immiscible. This complexing also occurs between aqueous calcium chloride and 1,4-dioxane suggesting a way to solvent switch from dioxane.
It is thus also not a foregone conclusion that no organic species would occupy the calcium chloride/ water layer in a two phase combination with a hydrocarbon or halogenated solvent. A substance that complexes with calcium ion might enter such an aqueous salt layer if its complex is stronger than the water one. Glymes apparently meet the criterion. 1,2- or 1,3-diols or cyclic polyethers might also prefer aqueous calcium chloride more than water alone or might precipitate out when so treated. This would be a particularly propitious means to recover expensive cryptate cyclic ether ligands.
Whether a particular polyether is likely to be strongly complexed by an aqueous calcium chloride solution may be judged by a simple test described and applied in the aforementioned patent. In it, 5 cc. of a saturated calcium chloride solution was added to a test-tube containing the ether. No heat was added to the reaction zones. The temperature rise at the end of two minutes is measured. Symmetrical dioxane gave a rise from 26 to 42 C. The increase for bis[2(2methoxyethoxy)ethyl] ether was from 28 to 71 C with solid complex separation.
Aqueous calcium chloride stirred with a potentially complexing compound in an immiscible organic solvent that does not complex calcium chloride, such as a hydrocarbon or MIBK, could it seems partially precipitate some material, dissolve some in the aqueous –salt phase and dissolve some in the hydrocarbon or MIBK phase. Filtration would separate the solid and the two liquid layers could be themselves separated giving three distinct phases. Uncomplexed organic will most likely be concentrated in the hydrocarbon or MIBK solution.
This method might be applicable to the frequent problem changing from a high boiling polar aprotic solvent such as DMF, DMA, N-methylpyrrolidone or DMSO, to a low boiling hydrocarbon solvent by distilling away the dipolar aprotic solvent using a small volume of a higher boiling glyme solvent to chase it, adding the low boiling hydrocarbon phase and then precipitating the glyme complex with an aqueous calcium chloride solution. Filtration would give two phases: the low oiling hydrocarbon liquid containing the reaction mixture and the residual saturated calcium chloride brine.
General |
Permalink |
Add comment |
Trackbacks (0)
Use of free radical inhibitors or antioxidants to increase the overall yield of organic synthesis steps.
kilomentor | 25 March, 2011 13:30
The use of radical inhibitors or antioxidants to improve yields does not appear to have many precedents in organic synthesis. A key word search provided only two references both related to the stabilization of m-chloroperbenzoic acid to thermal degradation during the epoxidation of resistant olefins. Y. Kishi, M. Aratani, H. Tanino, T. Fukuyama and T. Goto, J.C.S. Chem. Comm. 1972 64 and D.M. Tal, Steroids (1989), 54(1), 113-22. Synthetic chemists apparently assume that free radical reactions do not occur unless free radical initiators are present in the reaction mixture or unless the reaction mixture is irradiated. It might seem they think it can’t happen unless they are intending it to happen. Obviously this is not true! Free radical reactions can take place not just during the contemplated reaction phase but during the work-up of the reaction when we might think that all the reacting is stopped. Actual the opportunity is greater in the work-up phase this phase usually takes more time, particularly when the process is being scaled up. Are free-radical reactions inhibited by particular pH ranges of the solvent medium? No they are not. The most frequent type of free radical reaction is oxidation and only the relative amounts of different species that can be oxidized are affected by pH not particularly the oxidation rates. Oxidation often produces coloured products when it can introduce new unsaturation into molecules. The presence of unexpected colour in a reaction is suggestive of an unanticipated oxidation. I recall that in the preparation of some aniline compounds the procedure teaches the addition of hydrogen sulfide to the aqueous phase during isolation to prevent colour development from exposure to air during work up and crystallization. The usual response to coloured product is to use charcoal in the recrystallization rather than trying to prevent coloured by-products in the first place. The knock against the prophylactic (preventative) use of free radical inhibitors is that they are just one more substance to have to clean up during the work up, isolation and purification. Perhaps these radical inhibitors could be immobilized on an insoluble polymer so they could be filtered off after use. Has this been done to anyone’s knowledge?
General |
Permalink |
Comments (2) |
Trackbacks (0)
The Pharmaceutical Salt called Pamoate or Embonate
kilomentor | 26 February, 2011 14:47
One of the pharmaceutically acceptable salts is the pamoate also called embonate. Such salts are usually very poorly soluble in water and for this reason are used as biologically non-toxic derivatives in extended release solid dosage forms where the active is only supposed to appear gradually in the blood stream. If a good solubility is required these salts are not candidates. kilomentor hse provided some experimental procedures that have been used in the patent literature to make a variety of these salts and has added in some cases the mmoles and the weights in milligrams to assist in understanding what is going on. WO9425460A1 Risperidone Example I
A solution of 3- [2- [4-(6-fluoro- 1,2-benzisoxazol-3-yl)- I-piperidinyl) ethyl] -6,7,8,9-tetrahydro-2-methyl-4H-Pyrido[1,2-a]pyrimidin-4-one,19.70 g (0. 048mol) in ethanol (600ml) was added to a solution 18.64 g of pamoic acid (0. 048mol) in N,N-dimethylformamide (400ml). (1g/22 ml )
The mixture was stirred for 3 hours. The resulting precipitate was filtered off by suction, washed with ethanol and dried, yielding 3 1 g (8.1 %) of 3-[2-[4-(6-fluoro- 1,2benzisoxazol-3-yl)- I -piperidinyl)ethyl) -6,7,8,9-tetrahydro-2-methyl-4H-pyfido[ 1, 2ajpyrimidin-4-one 4,4'-methylenebis[3-hydroxy-2-naphthalenecarboxylate) (1: 1); mp. 269.2'C. This is a very poor yield of salt; just 8.1%. Pamoic acid apparently is soluble in dimethylformamide. This is useful information. The risperidone was dissolved in the usual ethanol. Perhaps the experimentalist did not wait long enough for the solid to all precipitate. They filtered after 3 hours. WO05016261A2 Example 1:
The pamoate salt of haloperidol (mw 375.86) can be prepared by treatment of haloperidol with pamoic acid or pamoate salt in solvent. Haloperidol pamoate can be prepared by adding a solution of haloperidol in an appropriate solvent, ie. ethanol with acetic acid, to a solution of disodium pamoate (mw 432.36) , pamoic acid (mw 388.36) or other pamoate salt and leaving undisturbed for 1-3 or more days until precipitation. Alternatively, other methods such as evaporation, slow or fast cooling or stirring solutions can also be used to precipitate salt.
Specifically, 2.5 ml of a 0.1M solution of haloperidol (94 mg) in an acidified ethanol (5% acetic acid) was added to 2.5 ml of a 0.1M solution of disodium pamoate)(108 mg) in ethanol/water (50/50). The mixture was allowed to sit at room temperature for 1-3 days. The resulting precipitate was filtered off by suction, washed with ethanol and dried in a vacuum oven at 60°C, yielding 240mg (1`26% of theoretical) of 1:1 haloperidol pamoate salt.
Example 2:
2.5 ml of a 0.25M solution of haloperidol (0.625 mmoles, 235 mg) in an acidified ethanol (5% acetic acid) was added to 12.5 ml of a 0.05M solution of disodium pamoate (0.625 mmoles, 270 mg) in ethanol/water (75/25). The mixture was allowed to sit at room temperature for 1-3 days. The resulting precipitate was filtered off by suction, washed with ethanol and dried in a vacuum oven at 60°C, yielding 206mg of 2:1 haloperidol pamoate salt. The stoichiometry is for a 1:1 adduct and haloperidol only has one basic nitrogen that can be neutralized. If the salt is 1:1 the yield would be 47%.If the dibasic pamoic acid neutralizes two haloperidols thern the yield is 59%since half the pamoic acid must remain without a reactive partner.
Example 3:
2.5 ml of a 0.25M solution of haloperidol in an acidified ethanol (5% acetic acid) was added to 6.25 ml of a O.1M solution of disodium pamoate in ethanol/water (50/50). The mixture was allowed to sit at room temperature for 1-3 days. The resulting precipitate was filtered off by suction, washed with ethanol and dried in a vacuum oven at 60°C, yielding 264mg of 2:1 haloperidol pamoate salt. - 1 1
Example 4:
5.ml of a 0.05M solution of haloperidol in an acidified ethanol (5% acetic acid) was added to 2.5 ml of a O.1M solution of disodium pamoate in ethanol/water (50/50). The mixture was allowed to sit at room temperature for 1-3 days. The resulting precipitate was filtered off by suction, washed with ethanol and dried in a vacuum oven at 60°C, yielding 119 mg of 1:1 haloperidol pamoate salt.
Example 6:
A (0.05 - 0.5M) solution of aripiprazole in an acidified ethanol is added to a (0.05 - 0.5M) disodium pamoate solution in a mixture of water/ethanol (100/0 0/100). The mixture is allowed to sit at room temperature for 1-3 days. The resulting precipitate is filtered off by suction, washed with solvent and dried in a vacuum oven at 60°C. These methods teach the method of adding the base acidified with 5% acetic acid in ethanol to the disodium pamoate in ethanol/water. The disodium salt is more soluble and so this method depends upon the acidification of sodium pamoate with acetic acid to create the pamoic acid in situ where it can interact with the amine in the presence of acetic acid. The more insoluble amine pamoate crystallizes. These examples illustrate the fact that pamoates often must be allowed to change form from a gel like form to crystalline over some time. Heating sometimes accelerates this change. WO04017970A1 (C) Preparation of 3-(3-methoxyphenyl)-3-(3- dimethylaminopropyl]-4,4-dimethyl-piperidine-2,6-dione pamoate salt (anhydrous) A solution of AGN-2979 bisulphate salt obtained in Step B (1 mmole, 430 mg) in 10 ml of water was mixed with methylene chloride (20 ml) and basified with aqueous ammonium hydroxide (29% w/w). After separation of the layers, the aqueous phase was extracted twice with methylene chloride. The combined organic phases were dried over anhydrous magnesium sulphate and the solvent was evaporated under reduced pressure. The residue was dissolved in ethanol (10 ml) and mixed with a hot solution of pamoic acid (embonic acid, 390 mg,1 mmole) in hot ethanol (30 ml) and the mixture was heated to reflux. After cooling, the pamoate salt crystallised and the salt was recrystallised in hot ethanol to give a pale yellow powder (melting point = 146°-150°C. The procedure separates free base, evaporates to an oil and dissolves it in ethanol. It is mixed with a hot solution of pamoic acid dissolved in hot ethanol. The embonate came out in crystalline form on cooling. WO05075454A2 FORMS OF 4-(4-METHYLPIPERAZIN-1-YLMETHYL)-n-[4-METHYL-3-(4-PYRIDIN-3-YL)PYRIMIDIN-2-YLAMINO)PHENYL]-BENZAMIDE - IMATINIB Example 10
4.l(4-Methyl-1 -piperazinyl)methyl]-N-[4-methyl-3-[ [4-(3-pyridinyl)-2- pyrimidinyl]amino]phenyl]- benzamide, pamoate A mixture of 4-[(4-methyl-1- piperazinyl) methyl]-N-[4-methyl-3-[[4-(3-pyridinyl)-2- pyrimidinyl]amino] phenyl]-benzamide (4.94 g, 10 mmol) and 4,4'-methylenebis[3-hydroxy-2- naphthoic acid (Fluke, Buchs, Switzerland; 3.88 g, 10 mmol) in ethanol (50 mL) is heated.
Water (25 mL) is then added. Upon cooling, the product crystallizes and is filtered-off and dried to afford 4-[(4-methyl-1- piperazinyl)methyl]-N- [4-methyl-3-[[4-(3-pyridinyl)- 2- pyrimidinyl]amino]phenyl]-benzemide, pamoate as a pale- yellow solid, having the following analytical properties: Analysis found: C, 69.12; H. 5.62; N. 10.88%; H2O, 2.50%. Calculated for C52H47N7O7- 1.26 H2O: C, 69.04; H. 5.52; N. 10.84%; H2O, 2. 51%. Heating pamoic acid in ethanol will create some solubility. The solids must have dissolved since the addition of water is usually done to the point of turbidity and then the crystals allowed to come out as the solution cools. WO05012233A1 MELDONIUM SALTS, METHOD OF THEIR PREPARATION AND PHARMACEUTICAL COMPOSITION ON THEIR BASIS EXAMPLE 10
Meldonium pamoate (1:1; x H20). Meldoniurn (5.46 g, 30 mmol) and pamoic acid (5.82 g, 15 mmol) are mixed with water and acetone (15 ml), the formed suspension is evaporated, 30-40 ml toluene is added to the residual viscous mass, it is grated, and evaporation is repeated. If the residue is insufficiently dry, treatment with toluene is repeated. Mp. 128-133°C (decomp.). H NMR spectrum (DMSO-d6), 6, ppm: 2.41 (2H, t, CH2COO-); 3.14 (2H, t, CH2N); 3.25 (9H, s, Me3N+); 4.75 (2H, s, -CH=(pam)) , 7.12 (2H, t, Harom); 7.26 (2H, td, Harom); 7.77 (2H, d, Harom); 8.18 (2H, d, Harom); 8.35 (2H, s, Harom). Found, %: C 62,90; H 5,83; N 4,98. Calculated, %: C 63,07; H S,84; N 5,07. Initially H:O content in the sample was 1.71%; after 24 hours maintenance at 100% humidity sample mass increased by 9% due to absorbed water. Pamoic acid is not particularly soluble in either water or acetone. Evaporation would readily remove the acetone. The water would only be grudgingly removed as an azeotrope with toluene. WO0008016A1 PAROXETINE SALTS Example 32 : Preparation of paroxetine pamoate 1: 1 salt.
A solution of paroxetine base in toluene (5 ml, 2. 10 g) was added to a solution of pamoic acid (2.48 g) in pyridine (40 ml), and the mixture was stirred at ambient temperature for 30 minutes. The solvent was then removed by distillation at reduced pressure, the residual oil diluted with toluene (30 ml) and the solvent again removed by distillation at reduced pressure. This procedure was repeated two more times. The solid product was washed with hot diethyl ether (c. 100 ml x 3) , and filtered under nitrogen to give a pale yellow solid. The product was washed twice more with diethyl ether (2 x 100 n- A), and then with methanol (30 ml), and finally dried under vacuum. Yield = 3.27 g,
IR nujol mull:
Bands at 1636, 1558, 1508, 1459, 1377, 1183, 1036, 830, 722 CM-1.
Example 33 : Preparation of paroxetine pamoate 2:1 salt.
A solution of paroxetine base in toluene (10 ml, 4.2 g) was added to a solution of pamoic acid (2.48 g) in pyridine (40 ml). The mixture was stirred at ambient temperature for 30 minutes. The solvent was then removed by distillation at reduced pressure, the residual oil diluted with toluene (30 ml) and the solvent again removed by distillation at reduced pressure. This procedure was repeated two more times. The solid product was washed with diethyl ether (c. 50 ml), and filtered under nitrogen to give a white solid. This solid was washed twice more with diethyl ether (2 x 10 ml), and then dried under vacuum. Yield 6.7 g.
IR nujol mull:
Bands at 1641, 1461, 13 77, 1181, 1035, 829, 757 cm- 1.
A solution of 3- [2- [4-(6-fluoro- 1,2-benzisoxazol-3-yl)- I-piperidinyl) ethyl] -6,7,8,9-tetrahydro-2-methyl-4H-Pyrido[1,2-a]pyrimidin-4-one,19.70 g (0. 048mol) in ethanol (600ml) was added to a solution 18.64 g of pamoic acid (0. 048mol) in N,N-dimethylformamide (400ml). (1g/22 ml )
The mixture was stirred for 3 hours. The resulting precipitate was filtered off by suction, washed with ethanol and dried, yielding 3 1 g (8.1 %) of 3-[2-[4-(6-fluoro- 1,2benzisoxazol-3-yl)- I -piperidinyl)ethyl) -6,7,8,9-tetrahydro-2-methyl-4H-pyfido[ 1, 2ajpyrimidin-4-one 4,4'-methylenebis[3-hydroxy-2-naphthalenecarboxylate) (1: 1); mp. 269.2'C. This is a very poor yield of salt; just 8.1%. Pamoic acid apparently is soluble in dimethylformamide. This is useful information. The risperidone was dissolved in the usual ethanol. Perhaps the experimentalist did not wait long enough for the solid to all precipitate. They filtered after 3 hours. WO05016261A2 Example 1:
The pamoate salt of haloperidol (mw 375.86) can be prepared by treatment of haloperidol with pamoic acid or pamoate salt in solvent. Haloperidol pamoate can be prepared by adding a solution of haloperidol in an appropriate solvent, ie. ethanol with acetic acid, to a solution of disodium pamoate (mw 432.36) , pamoic acid (mw 388.36) or other pamoate salt and leaving undisturbed for 1-3 or more days until precipitation. Alternatively, other methods such as evaporation, slow or fast cooling or stirring solutions can also be used to precipitate salt.
Specifically, 2.5 ml of a 0.1M solution of haloperidol (94 mg) in an acidified ethanol (5% acetic acid) was added to 2.5 ml of a 0.1M solution of disodium pamoate)(108 mg) in ethanol/water (50/50). The mixture was allowed to sit at room temperature for 1-3 days. The resulting precipitate was filtered off by suction, washed with ethanol and dried in a vacuum oven at 60°C, yielding 240mg (1`26% of theoretical) of 1:1 haloperidol pamoate salt.
Example 2:
2.5 ml of a 0.25M solution of haloperidol (0.625 mmoles, 235 mg) in an acidified ethanol (5% acetic acid) was added to 12.5 ml of a 0.05M solution of disodium pamoate (0.625 mmoles, 270 mg) in ethanol/water (75/25). The mixture was allowed to sit at room temperature for 1-3 days. The resulting precipitate was filtered off by suction, washed with ethanol and dried in a vacuum oven at 60°C, yielding 206mg of 2:1 haloperidol pamoate salt. The stoichiometry is for a 1:1 adduct and haloperidol only has one basic nitrogen that can be neutralized. If the salt is 1:1 the yield would be 47%.If the dibasic pamoic acid neutralizes two haloperidols thern the yield is 59%since half the pamoic acid must remain without a reactive partner.
Example 3:
2.5 ml of a 0.25M solution of haloperidol in an acidified ethanol (5% acetic acid) was added to 6.25 ml of a O.1M solution of disodium pamoate in ethanol/water (50/50). The mixture was allowed to sit at room temperature for 1-3 days. The resulting precipitate was filtered off by suction, washed with ethanol and dried in a vacuum oven at 60°C, yielding 264mg of 2:1 haloperidol pamoate salt. - 1 1
Example 4:
ml of a 0.05M solution of haloperidol in an acidified ethanol (5% acetic acid) was added to 1 ml of a 0.25M solution of disodium pamoate in ethanol/water (50/50). The mixture was allowed to sit at room temperature for 1-3 days. The resulting precipitate was filtered off by suction, washed with ethanol and dried in a vacuum oven at 60°C, yielding 107 mg of 1:1 haloperidol pamoate salt.
Example 5:
5.ml of a 0.05M solution of haloperidol in an acidified ethanol (5% acetic acid) was added to 2.5 ml of a O.1M solution of disodium pamoate in ethanol/water (50/50). The mixture was allowed to sit at room temperature for 1-3 days. The resulting precipitate was filtered off by suction, washed with ethanol and dried in a vacuum oven at 60°C, yielding 119 mg of 1:1 haloperidol pamoate salt.
Example 6:
A (0.05 - 0.5M) solution of aripiprazole in an acidified ethanol is added to a (0.05 - 0.5M) disodium pamoate solution in a mixture of water/ethanol (100/0 0/100). The mixture is allowed to sit at room temperature for 1-3 days. The resulting precipitate is filtered off by suction, washed with solvent and dried in a vacuum oven at 60°C. These methods teach the method of adding the base acidified with 5% acetic acid in ethanol to the disodium pamoate in ethanol/water. The disodium salt is more soluble and so this method depends upon the acidification of sodium pamoate with acetic acid to create the pamoic acid in situ where it can interact with the amine in the presence of acetic acid. The more insoluble amine pamoate crystallizes. These examples illustrate the fact that pamoates often must be allowed to change form from a gel like form to crystalline over some time. Heating sometimes accelerates this change. WO04017970A1 (C) Preparation of 3-(3-methoxyphenyl)-3-(3- dimethylaminopropyl]-4,4-dimethyl-piperidine-2,6-dione pamoate salt (anhydrous) A solution of AGN-2979 bisulphate salt obtained in Step B (1 mmole, 430 mg) in 10 ml of water was mixed with methylene chloride (20 ml) and basified with aqueous ammonium hydroxide (29% w/w). After separation of the layers, the aqueous phase was extracted twice with methylene chloride. The combined organic phases were dried over anhydrous magnesium sulphate and the solvent was evaporated under reduced pressure. The residue was dissolved in ethanol (10 ml) and mixed with a hot solution of pamoic acid (embonic acid, 390 mg,1 mmole) in hot ethanol (30 ml) and the mixture was heated to reflux. After cooling, the pamoate salt crystallised and the salt was recrystallised in hot ethanol to give a pale yellow powder (melting point = 146°-150°C. The procedure separates free base, evaporates to an oil and dissolves it in ethanol. It is mixed with a hot solution of pamoic acid dissolved in hot ethanol. The embonate came out in crystalline form on cooling. WO05075454A2 FORMS OF 4-(4-METHYLPIPERAZIN-1-YLMETHYL)-n-[4-METHYL-3-(4-PYRIDIN-3-YL)PYRIMIDIN-2-YLAMINO)PHENYL]-BENZAMIDE - IMATINIB Example 10
4.l(4-Methyl-1 -piperazinyl)methyl]-N-[4-methyl-3-[ [4-(3-pyridinyl)-2- pyrimidinyl]amino]phenyl]- benzamide, pamoate A mixture of 4-[(4-methyl-1- piperazinyl) methyl]-N-[4-methyl-3-[[4-(3-pyridinyl)-2- pyrimidinyl]amino] phenyl]-benzamide (4.94 g, 10 mmol) and 4,4'-methylenebis[3-hydroxy-2- naphthoic acid (Fluke, Buchs, Switzerland; 3.88 g, 10 mmol) in ethanol (50 mL) is heated.
Water (25 mL) is then added. Upon cooling, the product crystallizes and is filtered-off and dried to afford 4-[(4-methyl-1- piperazinyl)methyl]-N- [4-methyl-3-[[4-(3-pyridinyl)- 2- pyrimidinyl]amino]phenyl]-benzemide, pamoate as a pale- yellow solid, having the following analytical properties: Analysis found: C, 69.12; H. 5.62; N. 10.88%; H2O, 2.50%. Calculated for C52H47N7O7- 1.26 H2O: C, 69.04; H. 5.52; N. 10.84%; H2O, 2. 51%. Heating pamoic acid in ethanol will create some solubility. The solids must have dissolved since the addition of water is usually done to the point of turbidity and then the crystals allowed to come out as the solution cools. WO05012233A1 MELDONIUM SALTS, METHOD OF THEIR PREPARATION AND PHARMACEUTICAL COMPOSITION ON THEIR BASIS EXAMPLE 10
Meldonium pamoate (1:1; x H20). Meldoniurn (5.46 g, 30 mmol) and pamoic acid (5.82 g, 15 mmol) are mixed with water and acetone (15 ml), the formed suspension is evaporated, 30-40 ml toluene is added to the residual viscous mass, it is grated, and evaporation is repeated. If the residue is insufficiently dry, treatment with toluene is repeated. Mp. 128-133°C (decomp.). H NMR spectrum (DMSO-d6), 6, ppm: 2.41 (2H, t, CH2COO-); 3.14 (2H, t, CH2N); 3.25 (9H, s, Me3N+); 4.75 (2H, s, -CH=(pam)) , 7.12 (2H, t, Harom); 7.26 (2H, td, Harom); 7.77 (2H, d, Harom); 8.18 (2H, d, Harom); 8.35 (2H, s, Harom). Found, %: C 62,90; H 5,83; N 4,98. Calculated, %: C 63,07; H S,84; N 5,07. Initially H:O content in the sample was 1.71%; after 24 hours maintenance at 100% humidity sample mass increased by 9% due to absorbed water. Pamoic acid is not particularly soluble in either water or acetone. Evaporation would readily remove the acetone. The water would only be grudgingly removed as an azeotrope with toluene. WO0008016A1 PAROXETINE SALTS Example 32 : Preparation of paroxetine pamoate 1: 1 salt.
A solution of paroxetine base in toluene (5 ml, 2. 10 g) was added to a solution of pamoic acid (2.48 g) in pyridine (40 ml), and the mixture was stirred at ambient temperature for 30 minutes. The solvent was then removed by distillation at reduced pressure, the residual oil diluted with toluene (30 ml) and the solvent again removed by distillation at reduced pressure. This procedure was repeated two more times. The solid product was washed with hot diethyl ether (c. 100 ml x 3) , and filtered under nitrogen to give a pale yellow solid. The product was washed twice more with diethyl ether (2 x 100 n- A), and then with methanol (30 ml), and finally dried under vacuum. Yield = 3.27 g,
IR nujol mull:
Bands at 1636, 1558, 1508, 1459, 1377, 1183, 1036, 830, 722 CM-1.
Example 33 : Preparation of paroxetine pamoate 2:1 salt.
A solution of paroxetine base in toluene (10 ml, 4.2 g) was added to a solution of pamoic acid (2.48 g) in pyridine (40 ml). The mixture was stirred at ambient temperature for 30 minutes. The solvent was then removed by distillation at reduced pressure, the residual oil diluted with toluene (30 ml) and the solvent again removed by distillation at reduced pressure. This procedure was repeated two more times. The solid product was washed with diethyl ether (c. 50 ml), and filtered under nitrogen to give a white solid. This solid was washed twice more with diethyl ether (2 x 10 ml), and then dried under vacuum. Yield 6.7 g.
IR nujol mull:
Bands at 1641, 1461, 13 77, 1181, 1035, 829, 757 cm- 1.
Pamoic acid is soluble in pyridine presumably as a pyridinium salt. It can be recrystallized from dilute aqueous pyridine. It is also soluble in nitrobenzene.
General |
Permalink |
Trackbacks (0)
Simple, Rapid Optimization of a Chemical Process Step
kilomentor | 10 February, 2011 14:53
This is a repost trying to overcome presentation difficulties. This discussion pertains to the most efficient method to discover higher yield/lower cost procedures for a chemical transformation; this is the individual reaction step optimization aspect of process development. For those to whom optimization strategies are a new subject, this article is not a good place to start. The discussion is geared towards chemists who have applied, or tried to apply such optimization methods as factorial designs, fractional factorial designs, D-optimal designs, the technique of steepest ascent, or simplex optimization. Although using optimization methods I have had some spectacular successes by comparison with one-variable-at-a-time or the intuitive approaches, I have been partially dissatisfied for two reasons: the reagent choice, solvent choice and catalyst or other excipient (acid/base etc) variables are discrete and not continuous.the randomness required to make good statistics seems to throw away most of the chemist’s book learning With respect to the first point, once I had by some means settled upon the reagent and general protocol for the reaction, optimizing the yield by adjusting continuous variables proceeded well but I felt that getting to the particular reagent and its methodology of use was by happenstance and I had not given other choices a fair shake. With respect to the second point, because statistics asks one to consider at the outset a large reaction space it takes quite a few experiments to get back, quite often to where one intuitively would have begun. When a chemical synthesis route is selected, the responsible chemists have a certain confidence that after some experimentation each chemical transformation can be made to proceed in an acceptable yield and the intermediates all recovered in an acceptable purity. Those chemists that chose a particular route from among all other possibilities, have this confidence in each of the required chemical steps because, based on their experience, inference, intuition and analogies, they expect that the combination of the particular starting material, particular reagent, particular solvent(s) and particular reaction conditions will deliver a satisfactory result. What needs to be worked out is the research program; the menu of experiments that will lead expeditiously to validating this desired result. I propose a pragmatic and rational development program should start with the information, which gave the creators of the route confidence that the chemical step being studied was do-able. As process chemists, the first thing we would like to know is, if we test the most preferred combinations of reagent/solvent/ and auxiliary chemicals using typical reaction parameters, whether we are likely to achieve a satisfactory result or whether we need to find, devise and test other combinations in order to achieve the minimum acceptable yield and purity. I think there is a statistically sound method to answer this question and that is what is discussed below. Dr. Charles Hendrix, in Chemtech, August 488-496, 1980, published an article, Through the response surface with test tube and pipe wrench teaching that in the beginning of a study completely random experimentation within a defined reaction space can actually quantify how likely it will be to achieve a yield matching or exceeding a target yield, which the chemist can specify. This article is the most useful I have found in my career for chemical process optimization. Hendrix makes the point that the problem with statistical optimization is that too many of the experiments are performed far away from the higher yield region and too few close to it. The beauty of the Hendrix approach is that the initially defined reaction space can, it seems, be the sum of several smaller reaction spaces, which allows these smaller reaction spaces to be dedicated to non-continuous variables. Taking a simple example, if I am studying an acetylation, one part of the acetylation reaction space could be using acetic anhydride and one part of the reaction space should be done with acetyl chloride. After 13 randomly selected conditions selected from the whole reaction space, I would be able to express as a percent the likelihood of reaching a specific yield target by looking within the reaction space- acetyl chloride or acetic anhydride used within a particular range of stoichiometry, temperature, and other continuous variable conditions. If I am happy with the odds of achieving my target yield, I can continue experimentation within the reaction space defined, using any of the standard optimization methods and usually starting at my best random result. If I am unhappy with the odds of achieving my target yield, I can change the reaction space by adding new discrete variables, such as for example 4-dimethylaminopyridine catalysis, and repeat 13 random experiments in the new reaction space and get a new estimate of success. Hendrix’s approach, as particularly applied to include discontinuous variables, addresses my two dissatisfactions. The chemist’s knowledge of reagent choices that will work, used in the initial route election, is now used to choose the sub-domains for the random search. Second, one gets an early estimate of the likelihood of success so that one can quickly start optimizing in the correct reaction sub-domain starting with the best result so far achieved. I found this easier to understand by working through a simple example showing how a process research chemist would experiment to test whether one of a number of reagent/solvent choices would be likely to achieve a yield of 80% or higher. The specific question that I will use in the test is: Can we expect to be able to selectively oxidize 7-methyl-2,6-octanediol, to 6-hydroxy-7-methyloctan-2-one in greater than 80% yield using one of the following literature methods (A B C D or E): A. Douglas F. Taber, John C. Amedio, Jr. and Kang-Yeoun Jung. J. Org. Chem. 52, 5621 (1987). P2O5 / DMSO / Triethylamine (PDT): A Convenient Procedure for Oxidation of Alcohols to Ketones and Aldehydes. The method is applicable on a large scale, is selective and uses neither cryogenic methods nor heavy metals. B. E. J. Corey and C.U. Kim. Tet. Lett. 12, 919 (1973). Oxidation of Primary and Secondary Alcohols to Carbonyl Compounds using Dimethyl Sulfoxide-Chlorine Complex as Reagent. The complex reacts with primary and secondary alcohols followed by a tertiary amine to give a ketone. Double bonds are chlorinated. The reaction occurs at –45 C. C. E.J. Corey, Ernie-Paul Barrette and Plato A. Magriotis, Tet. Lett. 26(48), 5855 (1985). A New Cr(VI) Reagent for the Catalytic Oxidation of Secondary Alcohols to Ketones. A new process is described for the oxidation of secondary alcohols to ketones using peroxyacetic acid in the presence of a catalytic amount of 2,4-dimethylpentane-2,4-diol cyclic chromate. As little as 2 mole percent of catalyst is often needed. the reaction proceeds in methylene chloride/carbon tetrachloride mixtures. the catalyst is produced in carbon tetrachloride solution. The peracid used was in ethylacetate solution. For isolation the mixture was diluted with 9:1 hexane / ether and filtered through silica gel to remove the chromium species. The reaction occurs at near zero Centigrade. The simplicity and economy of the method recommend it for large scale work. The method is likely to be sensitive to steric conditions around the alcohol group. D and E. Michael P. Doyle, Robert L. Dow, Vahid Bagheri, and William J. Patrie, Tet. Lett. 21 2795 (1980). Selectivity in Oxidation of Diols. Oxidation of 2,2-disubstituted –1,4-butanediols by the combination of nickel(II) bromide and benzoyl peroxide or by trityl tetrafluoroborate produce ββ-disubstituted-γ-butyrolactones with exceptional selectivity. The less sterically hindered alcohol is preferentially oxidized by these reagents.All of the reagents are known from literature precedent to be able to oxidize secondary alcohols to ketone. In the example the particular problem is whether we can expect to selectively oxidize a less hindered secondary alcohol is the presence of another more hindered secondary alcohol. We want to know how likely it is that if we explore the reaction space which includes all five of the above literature methods (D and E are in the same paper), we will find a set of conditions which will give a yield of the product (or the equivalent cyclized hemiketal) greater or equal to 80%. In order to answer this question using Monte-Carlo methods, Hendrix teaches that we must very first very precisely specify the reaction space to be explored. For simplicity in these cases we could, for example, say that wherever a tertiary amine is used in any of the methods, it will be triethylamine. Specifying the limits for each particular sub-domain reaction space we could specify: Condition A reaction space comprises (0.9-2.7 equivalents of phosphorus pentoxide); (1.—3.0 equivalents of DMSO); methylene chloride solvent and 1.75-5.25 equivalents of triethylamine at 0-30ÚC.Condition B would encompass 1-4 mol. equivalents of chlorine combined with 5 mol equivalents of DMSO compared to the chloride and 1-4 equivalents of triethylamine compared to the chloride equivalents. The solvent is to be methylene chloride and the reaction temperature –45ÚC with quenching at –45 to 25 CÚ with immediate neutralization of the excess oxidant.Using method C The chromium catalyst complex should be between 1-4 mol %; the peracetic acid oxidant between 1.5- 3.0 mol. equivalents; the temperature between –20 to +10 CÚ; and the solvent mixtures of methylene chloride and carbon tetrachloride. The time should be up to 12 hours.Using Method D, the reagent combination Ni(II)bromide (1.25- 3.5 mol. equivalents) / benzoyl peroxide (1-4 mol.equivalents), the solvent will be acetonitrile and the temperature will be 40-80ÚC and the time up to 48 hours.Using the reagent trityl tetrafluoroborate, condition E, (1.5-3.5 mol. equivalents), the solvent will be acetonitrile and the temperature range from 40-80ÚC. with a time up to 48 hours. It is important to realize that the reaction condition choices made for the statistical test are not the only ones possible. They are the conditions most optimistic from the perspective of the process chemist. By now performing 13 reactions under conditions randomized within that space, we can predict from the results how likely it will be to obtain the minimum required yield for the step. If the likelihood is excellent, the optimization can be continued within this constrained reaction space using some such process as a directed simplex starting from the best results and moving in the sub-reaction space of the particular reagent selected. If the probability of reaching the target yield is estimated to be too low then the process chemist needs to explore a larger reaction space or a different reaction space to achieve the desired result with a high likelihood. To continue working this example I will choose, at random, 13 experiments within the five-reagent reaction space I have defined. First I randomly selected how many experiments will be performed in each sub domain. My random selection was that I should perform 3 of the Condition A phosphorus pentoxide/DMSO; two of Condition B which is chlorine/DMSO; five of Condition C which is the chromium catalyst complex; one of Condition D which is the benzoyl peroxide/nickel(II)bromide oxidation and two of Condition E which uses the trityl tetrafluoroborate reagent. Now I choose using random numbers the actual values of the continuous variables for each sub-domain. In the example: For Condition A: Experiment #12.7 equivalents of phosphorus pentoxide; 2.0 equivalents of DMSO; 3.5 equivalents of triethylamine at a temperature of 0 C Experiment #20.9 equivalents of phosphorus pentoxide; 2.0 equivalents of DMSO; 1.75 equivalents of triethylamine at a temperature of 15 C Experiment #31.9 equivalents of phosphorus pentoxide; 3.0 equivalents of DMSO; 3.5 equivalents of triethylamine ar a temperature of 15 C For Condition B: Experiment #41 molar equivalent of chloride and 1 molar equivalent of DMSO and 2 molar equivalents of triethylamine in methylene chloride forming the reagent and doing the initial reaction at –45 C with warming to 0 C and quenching the residual oxidant. Experiment #52 molar equivalent of chloride and 2 molar equivalent of DMSO and 4 molar equivalents of triethylamine in methylene chloride forming the reagent and doing the initial reaction at –45 C with warming to 20 C and quenching the residual oxidant. For Condition C Experiment #61-4 mol % complex; 1.5-3.0 equiv. peracetic acid; -20 to +10 C; ratio methylene chloride/CCl4 1:1 to 5:1 Experiment #74 mol % complex; 2.0 equiv. peracetic acid; -10 C; ratio methylene chloride/CCl4 4:1 Experiment #83 mol % complex; 1.5 equiv. peracetic acid; -10 C; ratio methylene chloride/CCl4 3:1 Experiment #91 mol % complex; 2.5 equiv. peracetic acid; 0 C; ratio methylene chloride/CCl4 5:1 Experiment #102 mol % complex; 3.0 equiv. peracetic acid; +10 C; ratio methylene chloride/CCl4 3:1 For Condition D Experiment #11Ni(II) bromide(2.0 mol. equivalents)/benzoyl peroxide (3 mol.equivalents) the solvent will be acetonitrile and the temperature will be 80 C and the time up to 48 hours. Condition E Experiment #12(2.0 mol. equivalents), the solvent will be acetonitrile and the temperature range 60 C. with a time up to 48 hours. Experiment #13(1.5 mol. equivalents), the solvent will be acetonitrile and the temperature 80 C. with a time up to 48 hours. I will now consider three different scenarios for the results of the thirteen experiments.Suppose the results were as shown below designated Results A
| Experiment # | Yield | Rank Yield | Rank | (100-Rank)/(N+1) |
| 1 | 41 | 11 | 1 | 7.1 |
| 2 | 50 | 14 | 2 | 14.3 |
| 3 | 11 | 25 | 3 | 21.4 |
| 4 | 14 | 34 | 4 | 28.6 |
| 5 | 80 | 41 | 5 | 35.7 |
| 6 | 34 | 59 | 6 | 42.9 |
| 7 | 84 | 61 | 7 | 50.0 |
| 8 | 25 | 70 | 8 | 57.1 |
| 9 | 85 | 73 | 9 | 64.3 |
| 10 | 73 | 80 | 10 | 71.5 |
| 11 | 70 | 84 | 11 | 78.5 |
| 12 | 61 | 85 | 12 | 86.7 |
| 13 | 88 | 88 | 13 | 92.8 |
General |
Permalink |
Comments (1) |
Trackbacks (0)
Optimization continued
kilomentor | 10 February, 2011 14:50
Results B
Results C
Following the teaching of Hendrix in the article noted above if one makes 13 random observations and then ranks them there is a 7.2% chance (100-92.8) of finding a yield greater than the 13th measurement (which is the highest yield obtained) Here that is 88% within the reaction space defined for these five reagent/condition combinations taken together. So long as the data is reproducible (and the 88% example should be repeated immediately) we know immediately that at least an 88% yield can be obtained using the conditions of experiment #13. Experiment #13 in our example was (1.5 mol. equivalents) of trityl fluoroborate, the solvent acetonitrile and the temperature 80 C with a time up to 48 hours.Using different numbers, suppose the results were as shown below designated There is an 7.2% chance (100-92.8)of finding a yield greater than 94% within the reaction space defined for these five reagent/condition combinations taken together. So long as the data is reproducible (and the 94% example should be repeated immediately) we know immediately that at least a 94% yield can be obtained using the conditions of experiment #10. Experiment #10 was 2 mol % of the chromium diol complex; 3.0 equiv. peracetic acid; +10 C; methylene chloride/CCl4 ratio 3:1 These results are actually all more optimistic than are very likely to be obtained. Using a different set of random numbers suppose the results were as shown below designated Results C.From these results it can be concluded that there is only a 7% probability to find a yield of greater than 60% in this complex reaction space. The only promising lead is that only a single experiment was conducted with discrete variables ofr Condition D that is Ni(II) bromide(2.0 mol. equivalents)/benzoyl peroxide (3 mol.equivalents),solvent acetonitrile and the temperature 80 C and the time up to 48 hours. If the reaction space was now restricted to these discrete conditions the probability of a high yield might be improved. This is an update of an original publication as of February 24, 2007. The author has not actually used the proposed method in process optimization. It is a theoretical proposal unlike other advice of Kilomentor, which is based on practical application and experience. I expect it will work and prove to be the most efficient method to optimize a process step. Comments and suggestions from experienced process chemists are welcome.
| Experiment # | Yield | Rank Yield | Rank | (100-Rank)/(N+1) |
| 1 | 30 | 15 | 1 | 7.1 |
| 2 | 63 | 23 | 2 | 14.3 |
| 3 | 65 | 26 | 3 | 21.4 |
| 4 | 26 | 30 | 4 | 28.6 |
| 5 | 37 | 37 | 5 | 35.7 |
| 6 | 15 | 42 | 6 | 42.9 |
| 7 | 50 | 50 | 7 | 50.0 |
| 8 | 23 | 63 | 8 | 57.1 |
| 9 | 75 | 65 | 9 | 64.3 |
| 10 | 94 | 75 | 10 | 71.5 |
| 11 | 42 | 76 | 11 | 78.5 |
| 12 | 91 | 91 | 12 | 86.7 |
| 13 | 78 | 94 | 13 | 92.8 |
| Experiment # | Yield | Rank Yield | Rank | (100-Rank)/(N+1) |
| 1 | 0 | 0 | 1 | 7.1 |
| 2 | 45 | 0 | 2 | 14.3 |
| 3 | 0 | 11 | 3 | 21.4 |
| 4 | 11 | 15 | 4 | 28.6 |
| 5 | 15 | 23 | 5 | 35.7 |
| 6 | 32 | 24 | 6 | 42.9 |
| 7 | 53 | 32 | 7 | 50.0 |
| 8 | 33 | 33 | 8 | 57.1 |
| 9 | 44 | 42 | 9 | 64.3 |
| 10 | 24 | 44 | 10 | 71.5 |
| 11 | 60 | 45 | 11 | 78.5 |
| 12 | 42 | 53 | 12 | 86.7 |
| 13 | 23 | 60 | 13 | 92.8 |
General |
Permalink |
Comments (2) |
Trackbacks (0)
The Kilomentor Approach-More Easily Isolated Intermediates-Fewer Technically Demanding Physical Separations.
kilomentor | 01 February, 2011 09:19
The Kilomentor approach to process development is geared towards the simple, rugged and dependable process step and avoids the technically demanding step, which uses sophisticated and expensive equipment. This keeps with the tenor of the times wherein so much process work is being moved to the developing economies where at least for now labour is plentiful, so long as the work process is rugged.
The Kilomentor approach to every process step is to divide it into the reaction phase, the success of which is measured by:
(i) the assay yield (the quantity of desired product in the reaction mixture as a percentage of theory)
(ii) the isolation yield which assesses the success of the effort to separate a product of practical purity out of the reaction soup (again expressed as % of the theoretical possible
The standard overall yield That is all that is normally reported is the product of these (as fractions) expressed as a percentage.
Because of our historic addiction to the assessment of synthetic elegance by counting reaction steps-with the fewer the better, there has been a bias against isolating intermediates as functional group derivatives and then decomposing them back to the original product intermediate. Because we so often only looked at the overall yield, we could not specifically see that we were leaving good product behind just because we were using less effective isolation means in order to minimize the reaction steps.
At the same time, we often convinced ourselves that even this advantage forming a functional group derivative to improve the isolated yield, would be given back in the step of decomposing the derivative to recover the original functionality. For example, the isolation of a pure ketone intermediate might be done by a painstaking vacuum fractional distillation without adding any additional reaction steps to the’ academic’ step count. The same ketone might readily and quantitatively form a solid oxime or hydrazone. Yet we would make arguments to ourselves that the backwards hydrolysis of clean oxime to ketone could be difficult and inefficient and messy when really actual hydrolysis technology made the cleavage trivial and efficient.
Another bad reason that we prefer challenging physical separations (fractional distillation or chromatography) to chemical derivatization is that the reagents for chemical derivatization would add to the chemical cost, while the challenging separation uses labour, time, and equipment which are not considered in the early, small-scale, chemicals-only costing.
General |
Permalink |
Comments (2) |
Trackbacks (0)
Navigation
Menu
Search
Calendar
| « | January 2012 | » | ||||
|---|---|---|---|---|---|---|
| Mo | Tu | We | Th | Fr | Sa | Su |
| 1 | ||||||
| 2 | 3 | 4 | 5 | 6 | 7 | 8 |
| 9 | 10 | 11 | 12 | 13 | 14 | 15 |
| 16 | 17 | 18 | 19 | 20 | 21 | 22 |
| 23 | 24 | 25 | 26 | 27 | 28 | 29 |
| 30 | 31 | |||||
Recently...
- The Chemical Manufacturing Process Route Selection: The Inclusion of Phase Switching and its Relationship to Validation
- A Question from the LinkedIn Organic Process Research & Development Group that Fosters Practicing Creativity
- The Strategy of using Split Runs in Chemical Process Development
- How thinking about Separating Acetic acid and Acetic Anhydride can sharpen our Chemical Process Development Skills
- An Example of a Possible Kilomentor Type Separation at Scale
- What Might be the Best Solvent for Difficult Cleaning Jobs on the Reactor Walls of a Plant Reactor
- The Importance of Understanding Process Validation for Pharmaceutical Process Chemists
- Is Glycerin a General Green Solution to Solvent Recovery and Replacement for Process Chemistry?
- More Information about using Inorganic Salt Complexes to simply Purify Alcohols at Scale
- Saturated Calcium Chloride Brine for Separations?
- Use of free radical inhibitors or antioxidants to increase the overall yield of organic synthesis steps.
- The Pharmaceutical Salt called Pamoate or Embonate
- Simple, Rapid Optimization of a Chemical Process Step
- Optimization continued
- The Kilomentor Approach-More Easily Isolated Intermediates-Fewer Technically Demanding Physical Separations.
Categories
Syndicate
Powered by LT - Design by BalearWeb