kilomentor

Pamoates are one of the pharmaceutically acceptable salts; however, they should only be considered for testing in an extended release formulation since the salts are almost always poorly soluble in water or stomach acid. The salts are prepared for retarding the dissolution of basic drugs. Stahl and Wermuth in Handbook of Pharmaceutical Salts Properties, Selection, and Use mention the pamoate salts of amitriptiline, benzphetamine, chlorpromazine, cyclguanyl, difenidol, dothiepin, imipramine, levomepromazine, metformine, noscapine, pamaquine, phendimetrazine, promazine, pyrantel, pyrvinium, and rhodoquine as being used in this way.

The acid was first described by Hosaeus in 1892 and a use was suggested for it in a patent to I.G. Farbenindustrie A.G. in 1992 where it was claimed as a method for manufacturing sparingly soluble, tasteless salts of nitrogenous basic compounds in particular salts of alkaloids such as strychnine and of bases of the ‘plasmochin’ type.

Kilomentor proposes that if one is faced with the problem of isolating an organic nitrogenous base by precipitation from any mixture, a reaction mixture say, the pamoate is probably the best first choice. A good second choice would be 2,2’-dihydroxy1,1’-dinaphthyl-3,3’-dicarboxylic acid, which is the compound similar to embonic acid but with the single difference that the methylene connecting the two naphthalene rings is gone. Both these compounds are either commercially available or easily synthesized. Syntheses are provided below from the paper by Barber and Gaimster, J. Appl. Chem., 2 October, 1952p. 565.

Embonic Acid

Method I- 2-hydroxy-3-naphthoic acid (750 g.) was suspended in glacial acetic acid (7.5 L.) and stirred at 5-100 C until dissolved. A mixture of glacial acetic acid (750 g.), 40% formaldehyde solution (450 g.) and concentrated sulfuric acid (71 g.) was added over20 minutes, the reaction being sufficiently exothermic to maintain the temperature between 95 and 100 C The suspension of embonic acid was stirred at 95-100 Cfor 30 minutes, allowed to cool to 70, filtered and washed first with ht glacial acetic acid (4.5 l.) and then distilled water until the washings were no longer ed to Congo red. The material was dried at 100 C to give embonic acid (700 g,)

Method II- 2-Hydroxy-3-naphthoic acid (500 g.) and 10% NaOH(1500 ml) were heated to 90 C with stirring; about 2/3of the solid dissolved. 40% formaldehyde solution (63 g.) was added, the temperature rising to 92 C, then a further 83 g. of 40% formaldehyde solution which caused a further rise in temperature to 95 C. No solid remained at this stage. After heating at 95 for a further 5 minutes, the solution crystallized spontaneously. The mixture was maintained at 95 C for 1 hour, cooled to 20 C and the sodium embonate filtered and washed with saturated brine (125 ml.) The damp sodium embonate (about 1.2 kg.) could be used as such or converted to the acid by dissolving in a mixture of water (3 l.) ad acetone (700 ml.), by heating to 50 C and adding glacial acetic acid (225 ml.)and then concentrated hydrochloric acid (bout 200 ml.) until the mixture was acid to Congo red. The precipitated embonic acid (480 g.) was filtered, washed with hot water until free of chloride, and dried at 100 C.

2,2’-Dihydroxy1,1’-dinaphthyl-3,3’-dicarboxylic acid

2,-Hydroxy-3-naphthoic acid (18.8 g.) was dissolved in a solution of sodium hydroxide (8.0 g.) in water (580 ml) and the solution was refluxed while a solution of ferric chloride (23 g.) of the hexahydrate) and conc. Hydrochloride acid (26 ml.) in water (29 ml.) was added drop-wise with stirring during a 20 minute period./the dark coloured reaction mixture was stirred at the boil for a further 30 minutes, then cooled, filtered and the filtrate rejected. After washing with a little water, the residue was dissolved in a slight excess of N-sodium hydroxide solution (200 ml.) The solution was treated with charcoal, filtered, acidified with concentrated hydrochloric acid and filtered. The yellow residue, after washing with water, was recrystallized from aqueous ethanol to give 2,2’-dihydroxy-1,1’-dinaphtrhyl-3,3’dicarboxylic acid (2.8 g.) as a pale-yellow hemi-hydrate m.p. 330-333 C.

US2397903 describes the poorly soluble salts with thiamine and dipyridoxine. US2641610 claims the use of the insoluble embonate salts of bis quartenary ammonium substances as a means of purifying and making the double salts with other anions by exchange.

Experimental details for making embonates either from relatively free bases or from mixtures of natural products are provided below for inspiration with your own problems.

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-ajpyrimidin-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 dimethyl formamide. 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 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, ea. ethanol with acetic acid, to a solution of disodium pamoate, pamoic acid 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 O.1M solution of haloperidol in an acidified ethanol (5% acetic acid) was added to 2.5 ml of a O.1M solution of disodium pamoate (2.5ml) 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 of 1:1 haloperidol pamoate salt.

Example 2:
2.5 ml of a 0.25M solution of haloperidol in an acidified ethanol (5% acetic acid) was added to 12.5 ml of a 0.05M solution of disodium pamoate 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.

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 acidifuied 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). Meldonium (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.

Molecules 2007, 12 1313

Extraction and precipitation of alkaloid-embonates

Homogenous dried leaves of a registered Finnish variety of C. roseus (1.0 g) were extracted for 30 minutes with 0.1 M hydrochloric acid solution (100 mL) in an ultrasonic bath (USF Finnsonic W 181,Ultra Sonic Finland). The mixture was then centrifuged at 2000 rpm for 10 min and the sediment was re-extracted with additional HCl (100 mL) for another 30 minutes. The combined supernatant from two repeated extractions was filtered and extracted with petroleum ether (200 mL) to eliminate chlorophyll and other lipophilic compounds. The acidic fraction was separated and an alkaline solution(pH 10.5) of 10 % embonic acid was slowly added for the precipitation of alkaloids as their embonate complexes. The pH of the resultant solution was increased to 5.0. The precipitate was separated simply by decantation and it was used as starting material for the semi-synthesis.

If you have a compound that is even quite a poor electron pair donor and that compound will not crystallize or will not crystallize to produce good quality crystals, what substance would be the best choice to test as a hydrogen bond donor? If you have a troublesome impurity in your product that can be predicted to be a better Lewis acid than your desired product, what would you propose as a good candidate to form a separable co-crystal with it?

An answer may be extracted a paper by the late Margaret C. Etter, Acct. Chem. Res. 1990, 23, 120-126. This is not a paper that synthetic organic chemists or process development chemists are likely to read. The lead author was a crystallographer and solid-state chemist. What makes the question interesting for us is that solid stoichiometric compositions can be made from substances with as poor a Lewis basicity as aliphatic ethers. The compound that forms these complexes is symmetrical bis-N,N-(3-nitrophenyl)urea. The compound can be easily synthesized from 3-nitroaniline and any phosgene equivalent giving a solid with melting point 256-258 C. It can be crystallized from any of acetic acid, benzene, chloroform, dichloromethane, ethanol, 95% ethanol or ethylene glycol. Heating the complex will drive off the Lewis base if it is volatile under high vacuum. One of two polymorphs will form but the form is not important for making co-crystals.

The complexes with a donor can be formed in a suitable solvent that is evaporated or if the donor is a solid, such as triphenyl phosphine oxide, simply by grinding two solids together. Making complexes with a component that is only present as an impurity in a product mixture has not yet been tried. The trick could perhaps be used to remove a difficult impurity such as triphenylphosphine oxide, dicyclohexylurea, dimethyl sulfoxide, polyethylene glycol. Even if a slight excess of the dinitro-urea was needed to completely remove the impurity of concern from a crude product mixture. Subsequent removal of this urea in second treatment might be much simpler than getting rid of the original troublesome impurity.

From the list of solvents from which symmetrical bis N,N-(3-nitrophenyl)urea can be crystallized, it would appear that the compound is insoluble in hydrocarbons and these might be useful as anti-solvents to increase the yield of adducts.

In Stahl and Wermuths book, Pharmaceutical Salts: Properties, Selection and Use there is a further piece of advice beyond what Kilomentor has already written about concerning salt selection. Unlike the other advice it is provided by implication only and needs to be simply stated.

On pg. 181 of the book, the selection of an appropriate pharmaceutical salt for the candidate drug called RPR200765 is presented. The following details of that problem are provided. RPR200765 was a candidate drug substance to be used to treat rheumatoid arthritis. The drug would have had to be taken regularly for the rest of patients lives. It is a crystalline, weak base with a substituted pyridine ring system, a pKa of 5.3 and log P of 2.5. The anticipated pharmaceutically effective dose was expected to fall between 100-125 mg. One can calculate that the molecular weight of RPR200765 by itself was 488.48. The actual API material is identified in Bioorganic & Medicinal Chemistry Letters (200), 11(5) 693-696.

Four potential salts were identified in the example: mesylate, camphorsulfonate, hydrochloride and hydrobromide. What was particularly instructive is the comment concerning the camphorsulfonate. The authors wrote that the only disadvantage of the camphorsulfonate when compared to the mesylate (the first choice) was the increased molecular weight due to the larger counter ion. It was considered that this could create problems with experimental capsule or tablet later in development.

Camphorsulfonic acid has a molecular weight of 232. The molecular weight of the monocamphorsulfonate salt of RPR200765 would have been 720.48. Giving a dose of 100-125 mg on the free base basis (0.256 mmoles), as camphorsulfonate salt, would amount to giving a dose 184 mg of this pharmaceutical salt. Delivering a dose of 184 mg of API, it is said, was anticipated to be problematic. From this it is possible to generalize that the practical limit to the weight of API that can be confidently handled is about 184 milligrams in the highest strength. This seriously restricts the choices of pharmaceutical salts for medicines particularly where the neutral active has a low molecular weight because this means there will be more moles in the dose and so more moles of salt former.

To take a current example, the highest prescribed dose of the cancer drug imatinib is 400 mg as free base. The molecular weight of the free base is 493. If we imagine the salt with an acid of molecular weight 232(camphorsulfonate) the weight of active API would be 588 mg. This is already more than 3 times what this teaching advises one can be comfortable with for achieving a successful formulation. It is obvious that the acid used to form the pharmaceutical salt for imatinib is going to have to have a low molecular weight. Pharmaceutically acceptable acids with molecular weight below 100 are only: acetic, carbonic, formic, glycolic, hydrobromic, hydrochloric, isobutyric, lactic, methanesulfonic, nitric, oxalic, phosphoric, sulphuric and thiocyanic. Of these the only ones without other concerns are hydrochloric, methanesulfonic, phosphoric and sulphuric. Suddenly salt selection becomes a lot easier! In the case of imatinib, the methanesulfonate was chosen as the drug substance!

Put as the converse it means that besides camphorsulfonic acid also galactaric, glucoheptanoic, lactobionic, 2-naphthalene sulfonic, 1,5-naphthalenesulfonic, oleic, palmitic, pamoic, sebacic, stearic and tannic acids need not be initially considered for salt formation with candidate bases. Five of these are from the group of 30 called Class 1 acids, the most preferred acids based on safety considerations.

Similarly benethamine, benzathine and hydrabamine are distinctly less preferred for salt formation with acid candidates based on their molecular weights.

The objective of the Kilomentor Blog is education. One can learn the basis for chemical process development and organic synthesis in schools, either technical or university undergraduate. One can build upon these with chemistry or engineering postgraduate training, but there is no path onward from there, which is widely accessible. This may be because of a person’s location on the planet or financial means. The careers that spring up from these roots are fun and useful to society. They should be open to the best minds on the planet; to whoever is intrigued to practice these arts. I write what has been useful to me and what I have learned slowly and painstakingly over 40 years. Wherever you are, all you need is access to the internet and you can share for free my experience, my insights, and yes, my errors. My goal is to provide a level playing field world-wide for organic process scientists.

A month ago my fiftieth blog article was published. The number of viewings for Kilomentor has passed 27,000. Several months ago the Kilomentor Blog was for several weeks the first ranked article on Google when the search terms: chemical process development’ organic synthesis, were entered.

Writing for Kilomentor has encouraged me to read more widely and try to put in perspective what I read. It has driven me to ask myself what the core knowledge in this profession is. It has enabled me to work together with some of you, solving problems together.

Thanks for reading.

Clarke Slemon PhD

The Kilomentor

A powerful idea for pharmaceutical product development is contained in an article titled, Consider a new approach to pharmaceutical development, authored by Pradir Ki Basu, Ronald A Mack, ,and Jonathan M. Vinson available at

http://findarticles.com/p/articles/mi_qa5350/is_199908/ai_n21444525

Hereunder, Kilomentor discusses aspects of their core idea, but these comments can only be followed after reading the original article.

Much of the article presents arguments supporting the importance of cost efficiently discovering a synthetic method, scaling it up and putting into production a process for manufacturing a new pharmaceutical. This is the pharmaceutical business with the actual marketing and selling stripped away. Its importance to profitability does not need to be debated.

The present authors are concerned about the efficient execution of the plan that starts after the identification of a biologically active target that is a candidate to be a commercial drug and proceeds to the validation of manufacture that molecule at commercial scale.

The new approach that they propose positions ‘process vision’ as the core concept. It is the definition and exemplification of ‘process vision’ which is the article’s most significant accomplishment. The authors identify the defining characteristics of the process vision at different places in the article but for me, I cannot say I adequately understood it until I drew particular phrases together in my notes.

· “The process vision satisfies all essential requirements, including those for safety, quality, waste minimization, cost, time, and operability.”

· “The process vision is neither the process with maximum yield nor the one that gives maximum product purity…..it is neither a chemist’s vision, nor an engineer’s vision; it is not even the vision of the chemists and engineers together.”

· “It is a vision a vision that all stakeholders in development, manufacturing and marketing can share…..”

Reading between the lines and amplifying certain aspects, the process vision might be a policy statement that provides as a starting point, desirable standards by which team members of each stage of the plan (laboratory process, kilo lab, pilot plant, and manufacturing facility) strive to meet their downstream colleagues’ concerns from the outsell of their work.

The authors make this clearer with specific examples of the unique orientation and emphasis that players at the different stages have and which they want to bring into early assessment, early inevitable cross purposes, and early compromise or conflict resolution. They write, “Chemists think in terms of steps, reactions, yield, purity, and so on; engineers in terms of unit operations, physical properties, heat load, and the like; manufacturing personnel in terms of unit operations, in terms of throughput, waste, control issues and plant modifications that may be required to run a process; and marketing people in terms of net present value of the product, how much it can sell for etc.”

For me, what the authors are somewhat ambiguous about is the mechanism they recommend for achieving this ‘process vision’ even though over and over again in the article they return to this same theme:

“It is important ….to get stakeholders to develop….agreed-upon objectives of process development.”

“communication among….personnel is critical during process development.”

“We need to…. provid[e] development team members with systems or tools to facilitate communications among different disciplines.”

“Unless the manufacturing team is involved in the process development, they will not have confidence in the scale-up”.

“…manufacturing and commercial input at this stage [late stage discovery] is essential for choosing the optimum processing route”.

“Team members need to be involved setting targets for cost, manufacturability, waste and emission loads, development time….”

“These alternatives must be evaluated based on….criteria agreed upon by all stakeholders….”

“If stakeholders are involved in planning experiments, it’s likely that more useful data could be collected from fewer experiments.”

For me these snippets hint at or outright propose two different strategies.

One can try to bring a diverse project team, with participation beginning with late stage R&D and including representatives all the way up to marketing, together frequently enough to work out priorities and make decisions even at the experimental program level.

Alternatively, one can establish some sort of median or normal or starting-point performance criteria addressing the main, recurring concerns of process development, manufacturing, and marketing which will serve as a process vision statement that will act as a proxy for the multiform interests of the entire downstream project team and continuously represent their standard concerns to upstream collaborators.

According to this meaning, a process vision statement would be a tool commanding corporate authority that would continuously challenging upstream groups with the standard core concerns of the downstream members.

The authors marvelously illustrate this challenging throughout their article. What I interpret them to be saying is that the problem is not that different elements of the project team have concerns which inevitably seem to operate at cross purposes; but that the team members will reach solutions that satisfy all parties, so long as the area of tension is discovered early enough.

Kilomentor has a strong preference for the second alternative. Use of a process vision statement as a proxy for the perspectives and concerns of downstream project groups seems preferable to using meetings of a large group with the frequency needed to actually direct even the collection of particular data. For a company’s drug product projects to be successful and on-time, any process’s strategy must not conflict too greatly with the psychological needs and private professional goals of the individual team members. The people downstream in the project, whether they be in process development, manufacturing, or marketing, simply will not give a project the attention it needs until it arrives at the phase where they are being held singly and personally responsible. They are too busy concentrating their attention on what is on their plate already and extinguishing the fat that is already in the fire. This is human nature! Besides, pharmaceutical product projects can go on so long that some participants can realistically expect to no longer be involved when a late-stage discovery project limps into manufacturing or marketing. People may hope or plan to outrun the difficulties.

Equally problematically, the up stream professionals, working at a particular phase of the work on their own turf, would require an uncommon personal modestly to accept without rancor face-to-face demands that particular questions be answered on a priority basis.

A corporate ‘process vision’ statement takes the personalities and egos out. At the same time, the standards proposed by a process vision statement would command authority and yet not be carved in stone. They would exist to bring a persistent awareness of particular concerns. They would bring those different needs, which may be pulling at cross purposes to early attention, and they can be expected to bring the affected team members together to create or negotiate a solution.

This excellent thought provoking article by Basu, Mack and Vinson contains other important ideas which I hope to look at in later blogs.

Kilomentor continues with his review of the properties and procedures for the manufacture of pharmaceutical salts. The procedures are among the most import in chemical process development and organic synthesis.

It would seem from looking at the names of pharmaceutical products that phosphate anion is a fairly frequently used pharmaceutical salt former, but closer examination reveals that there are actually very few ionic pharmaceutical phosphate salts. Among drug substances called phosphates, the majority are covalent phosphate esters of an alcohol functional group.

Nevertheless, there is a place for the phosphate salts because the monophosphate is probably the most hydrophilic anion used to make pharmaceutical salts. Dihydrogen phosphate anion contains two very polarized hydrogen-oxygen bonds that energetically prefer to exist in a hydrogen bonding, high dielectric medium. When this hydrophilic anion is combined with a large hydrophobic cation, the result is almost always an insoluble salt. To balance this advantage the following disadvantages must be weighed:

· These salts have a high propensity to give several different hydrate pseudopolymorphs.

· phosphoric acid as a viscous oil or very low melting solid that is difficult to manipulate quantitatively.

· phosphoric acid is not miscible with the non-polar organic solvents which are the preferred media for the hydrophobic base partner

· the acid’s hygroscopicity makes weighing more difficult.

Evidence of the hydrophilicity of phosphate is provided by its selection for use in a standard procedure for making those acid addition salts of acids, which are themselves in the neutral form poorly stable, such as nitrates, thiocyanates, perchlorates and fluoroborates.

The procedure is taken from Brandstrom and Gustavii, Acta Chemica Scandinavica 23 (1969) 1215-1218.

To a two phase mixture of 1M aqueous phosphoric acid and methylene chloride or chloroform, add the free base which thereupon dissolves in the aqueous acid. Add sodium nitrate, sodium thiocyanate, sodium perchlorate or sodium terafluoroborate; mix the phases and extract the salt, anion and protonated base, into the organic layer quantitatively. Because the phosphoric acid monoanion is so hydrophilic it does not compete with these anions for extraction into the lipophilic organic layer even though it is present in enormous excess. That phosphoric acid is the bulk acidifying agent used testifies to its preference for the aqueous phase. The pKas of phosphoric acid are K₁ =7.107 x 10^{-3 ;} K₂ =7.99 x 10^-8; K₃= 4.8 x 10^-13.

The literature provides another piece of evidence that ionic phosphates may be good choices for giving solid crystalline salts for a wide range of bases. Helene Perrier and Marc Labelle found the phosphates the second most preferred salts for isolating as salts a wide range of intermediates containing the 3-acylquinoline moiety [J. Org. Chem. (1999), 64, 2110-2113.

Some corroborative information about crystalline phosphate salts comes from an article analyzing salts found in the Cambridge structural Database. http://www.msm.cam.ac.uk/pfizer/pdf/Publications/P03%20(02)%20-%20Occurrence%20of%20Pharmaceutically%20Acceptable%20Anions%20and%20Cations%20in%20the%20Cambridge%20Structural%20Database.pdf

The phosphate dianion was found to have the highest percentage of its salts as hydrates of all the salts examined. According to the authors’ interpretation this suggests that increasing charge on a single ion leads to increasing hydrate formation. They reference another paper that suggests that hydrate formation is a result of an imbalance between the number of hydrogen bond donors and acceptors in a crystal. [Infantes l., Chisholm J. Motherwell S. Cryst. Eng. Comm. 2003, 5: 480-486.]

Some specific examples of procedures are given in the extended text below.

Introduction

Industrially important chemical transformation are usually conducted in solution; however, these processes often lead to partial loss of the solvent into the atmosphere. The equipment used conventionally in batch chemical processing is typically not adequate to prevent troublesome emissions of the most volatile solvent vapours. Additionally, the contaminated waste solvents are commonly sent for destruction rather than being recycled, increasing the likelihood of cumulatively damaging emissions. Chemicals in the atmosphere are sometimes serious pollutants and there is a need to reduce these leakages. Today many of the most common volatile solvents used in chemical process development are under a cloud. It has been proposed that solvents be more efficiently recovered and recycled but this is often discouragingly expensive. One suggestion has been to use dissolution media with much higher boiling points than the most popular solvents, because their vapour pressures are lower, but distillation of such solvents consumes more energy. Another suggestion has been to use special designed substances with solvent-like dissolution properties that can be reversibly chemically dissociated into more volatile fragmented products while they are removed from a reactor and which then can be recombined to regenerate the solvent-like substance. An example is the adduct between sulfur dioxide and perylene (1,3-penadiene) which is a liquid. Another example is the combination of sulfur dioxide, formaldehyde that together wit water produce hydroxymethylsulfonic acid. Often the recommendation is made that the industrial process be reengineered to use a more environmentally friendly solvent even if the relative volatility of the solvent is not reduced. What has not been recommended up until now is some means to use more volatile, lower-boiling solvents which can be easily disltilled with a low energy requirement so that the increased risk of pollution is avoided.

Kilomentor [Dr. Clarke Slemon] has filed a US patent application US61/069,688 to address this problem. The key to this new technology is a means of reducing the vapour pressure of the volatile solvent when it is not in use as a reaction solvent. The central claim of the application is the combination of a recycled, constrained solvent with a substantially less volatile complexing agent in a closed storage vessel. When the volatile solvent material is complexed, its vapour pressure is conveniently low for easy storage. When it is needed as a reaction solvent, heating the complex dissociates it and the solvent can be distilled into the reactor. When the solvent is no longer needed in the reactor, it can be distilled back into the reservoir where it recombines with the complexing agent.

Solvents that form such useful complexes are called constrained solvents by Kilomentor. The combination of a constrained solvent with a complexing agent is not new. Solvates of ammonia, for example, calcium chloride mono ammoniate or zinc chloride diammoniate have been long known. An intelligent and experienced reader, once presented with the inventive combination of a low boiling readily gaseous solvent, a complexing agent and a confining element in the context of the problem to be solved might easily assemble the useful combinations. The invention is the combination to address the opportunity (problem).

Particular pairs of a constrained solvent and complexing agent are: dinitrogen tetroxide and 1,4-dioxane; dinitrogen tetroxide and dimethylsulfoxide; sulfur dioxide and potassium bromide; sulfur dioxide and sodium iodide, or ammonia and calcium nitrate.

There are advantages besides ecological ones to working with a solvent under conditions where it is a constrained solvent. Some solvents are too volatile, explosive with air, flammable, poisonous, smelly or irritating to be used in regular processes where significant leakage into the atmosphere might more likely occur. Carbon disulfide for example is an excellent solvent with a well known unique combination of properties, but because of it flammability and low flash point it is unacceptable for process chemistry. Carbon disulfide (B.P. 46 C) if it could be put in combination with an appropriate complexing agent might be useful.

A possible advantage of this technology might be that purification by distillation would be inexpensive because the heat of vaporization for such liquids is low and because the boiling point is above 0 C the first stage of cooling can be a brine chiller. Residual uncondensed gas might be delivered below the surface of the non-volatile storage solvent.

This solvent could be stored in pressure resistant tanks mixed with a low vapour pressure environmentally more benign liquid that has a rapidly increasing solubility for the low boiling solvent with increasing pressure and which had a negatively deviating Raoult’s law vapour pressure.

By proper choice of reagents, co-reactants and catalysts, the very volatile solvent can be set up so that it can be distilled away from all the other components of the reaction mixture in such a state of purity that it would be usable, if not as a replacement for commercial grade solvent in every use, at least for a subsequent batch of the same product.

When the stored volatile solvent was needed in a repeat of the process step, it could be distilled out of its reservoir and condensed into the sealed preloaded reactor. The same cooling that is required to reach the reaction temperature is used to condense and retain the solvent. What is sacrificed is the ability to run reactions at 50 C and above. For these reactions the ecologically benign choice would have to be made from other alternatives.

Reactions that substantially proceed at ambient temperature but using the present technology are driven to completion by raising the temperature, could be driven to completion by concentrating the reaction mixture by starting the removal of the highly volatile solvent. This would dramatically increase the rate for reactions with a molecularity of two or higher. This includes most reactions that use a particular chemical reagent but does not include intramolecular rearrangements, hydrolyses or solvolyses.

The table shows solvents that might be used if a satisfactory complexing agent could be found.

Low boiling liquids that are already frequently used solvents are marked in bold in the table. The four highest boiling liquids are common organic solvents the first, second and third of which are typically avoided in process chemistry. Fluorine containing choices desirably handled in this confined manner because they may be ozone depleting substances: dibromodifluoromethane, dichlorofluoromethane, dibromodifluoromethane.

Two of these possibilities are not new to consideration as solvents, dinitrogen tetroxide and sulfur dioxide are two volatile dipolar aprotic solvents. In fact, I have on my book shelf a thin volume called Chemistry in Non-Aqueous Solvents, by Harry H. Sisler, Reinhold Publishing Company, 1961. In the pertinent chapters there is no discussion of possibility that these solvents could be easily recycled. This is the aspect, which provides the new perspective.

Dintrogen Tetroxide

Dinitrogen tetroxide melts at –12.C and its normal boiling point is 21.3 C thus its liquid range is convenient for its use as a solvent. The liquid may be readily supercooled and has been cooled as low as –110 C without it crystallizing. Its critical temperature is 158.2 C and its critical pressure is 100.0 atm. The density of dinitrogen tetroxide is 1.49 g/cc. at 0 C. The electrical conductance of liquid ditrogen tetroxide is very low. The specific conductance at 17 C is 2.36 X10-13. There is an equilibrium between ditrogen tetroxide and two molecules of the paramagnetic nitrogen dioxide. Although there is very little of the monomeric triatomic compound in the liquid at the boiling point there is about 16% in the gas phase.

A possible key to how dinitrogen tetroxide could be recycled is that it forms several fairly stable solvates with higher boiling liquids which comprise a large percentage of dinitrogen tetroxide. With p-dioxane dinitrogen tetroxide forms a 1:1 salt of melting point +45.2 C and a less stable one with 1,3-dioxane of mp 2 C. These complexes contain 60.7% dinitrogen tetroxide by weight. Thus to generate 100 ml of solvent it would only be necessary to decompose 250 gm of complex. The solid complex is heated to melting and the solvent distilled away from the dioxane residue. When the reaction in which it was mediating is complete the dinitrogen tetroxide can be distilled and condensed back into the dioxane solution where it reacts and returns to the solid state. The flask of solid is secure in the refrigerator in a stoppered flask. To se it as a solvent it needs to be appreciated that primary alcohols, amines, alkenes, and amides all react with dinitrogen tetroxide.

Sulfur Dioxide

Sulfor dioxide is widely used in the petrochemical industry as a solvent because of its ability to discriminate between function group classes, dissolving alkenes and aromatic hydrocarbons while having little solubility for saturated hydrocarbons. Means for recovering sulfur dioxide on scale are therefore most likely well developed by our engineering colleagues.

The boiling point of sulfur dioxide is –10.02 C and its freezing point is –75.46 C. Its density at –10 C is 1.46 g/cc. Sulfur dioxide displays some useful solvent properties for metathesis reactions and is a good solvent for Friedel Craft reactions in part because AlCl₃ dissolves readily in it.

The characteristic with which we are particularly focussed here however is the possibility that the sulfur dioxide could be trapped as a reversible adduct for storage. Looking at the data available in the Sisler book one can see that the potassium bromide solvate might be a good choice. The solvate combines 4 equivalents of sulfur dioxide with one formula weight of potassium bromide working out to a sulfur dioxide content of about 68%. Thus to prepare 100 ml of liquid sulfur dioxide would require 215 gm of the solvate. At –1 C the vapour pressure over this solid is already 1 atmosphere so it would need to be refrigerated strongly to keep it confined as the complex. More stable compounds however contain a smaller weight percent of sulfur dioxide. Aluminum chloride forms a disolvate with sulfur dioxide. As with the dinitrogen tetroxide case, the gas can be distilled away from the reservoir solid and condensed as liquid in the reactor and then distilled back into mixture r in the reservoir when the solvent was no longer needed.

The sulfate salt is the second most common pharmaceutical salt behind the hydrochloride. Bisulfate salts are quite acidic so the base from which one is made needs to be acid stable.

Sulfuric acid is a diprotic acid. It can form two different stoichiometric salt types: the 1:1 bisulfate salt and the 2:1 sulfate salts in which two moles of amine are protonated by each of the two protons of H₂SO₄. The pKas of sulphuric acid are –3 and 1.92 with almost five orders of magnitude difference between the acidity of the first and second hydrogen. Most pharmaceutical salts are of the 1:1 bisulfate type. Sulfates are most often made by the addition of an, at least partially aqueous, solution of acid because neat acid is not soluble in apolar solvents and it has some dehydrating capability which can lead to by-products when sufuric acid is in excess. Typical organic solvents used in making sulfates are methanol, ethanol, 1-propanol, 2-propanol, acetone and mixtures thereof. Acetone however is not recommended because an excess of acid causes the oligomerization of acetone creating color in the solution.

Kilomentor anticipates that by providing some examples of pharmaceutical sulfate salt preparations with some commentary to draw attention to important aspects of the methods a skilled experimentalist should have no difficulty making others.

US7230016 PREPARATION OF PIOGLITAZONE SULFATE

24.g of sulfuric acid was added slowly, at room temperature, to 250 ml of methanol followed by addition of 80 g of pioglitazone base with stirring. The mixture turned into a clear solution. 250 ml of ether was slowly added followed by 500 ml of heptane. A solid precipitated, and the suspension was stirred for 3 hours. The solid (98.4 g, yield was 96.5%) was collected by filtering and washed once with ether. The solid had a mp: 1113.5-116.5° C. (recrystallized from methanol).

This example illustrates the addition of the base to the organic solution of sulphuric acid in methanol. Although a small amount of methyl hydrogen sulphate might form this not a problem because MeOSO2OH is a pharmaceutically acceptable counterion. Note also that in the procedure the chemistry provided three opportunities to obtain crystals. Pioglitazone hydrogen sulfate might have precipitated from the methanol solution itself after partial dissolution. The salt might have crystallized when the methanol was diluted 50:50 with diethyl ether. The final opportunity occurred when the solution was diluted 1:1 with heptane and this was successful. Notice that the methanol could not be diluted with heptane directly. Two phases would have resulted. This is an example of a well designed approach to getting crystalline solid. If crystals still had not formed the solution would have been concentrated.

WO06040728A1: Preparation of 1-(2-(4-benzyl-4-hydroxy-piperidin-1-yl)-ethyl)-3-(2-methyl-quinolin-4-yl)-urea

Example 1
1-(2-(4-benzyl-4-hydroxy-piperidin-1-yl)-ethyl)-3-(2-methyl-quinolin-4-yl)-urea

(1 equivalent) is dissolved in ethanol at a concentration of 25% w/w and the mixture is heated at 50°C. Aqueous sulfuric acid (1M, 1.1 equivalents) is added. Optionally, the crystallization is initiated by a wet seed of Example 1 (0.5%). The suspension is cooled to 0°C with a cooling rate of 15 C°/h and maintained at this temperature at least 1 hour before filtration and washing with aqueous ethanol (50 % W/V). The solid is dried at 30°C under a wet stream of nitrogen (50% RH) to provide the title compound with a purity of 97.7% with a yield of approximately 90%.

The example illustrates the addition of the acid to an excess of base. The addition is performed warm. An aqueous sulphuric acid reagent is used and it is added to a water miscible solvent in this case ethanol. Using seeds of the salt product is optional here. The example prescribes a cooling rate that will lower the temperature to the final filtration temperature over somewhat more than 3 hours. This is followed by a hold time to ensure that all the material that can crystallize has come out before the filtration. The wash solution is a mixture of solvents similar to that from which the solid is crystallized. Often a slightly less polar wash solution is used than the mixture from which the crystals are produced. This gives some assurance that the wash will not redissolve the solid. Although it is not reported the wash solvent is usually cooled to the temperature of the slurry that was filtered originally. On scale, this is done simply by loading the wash solvent mixture into the crystallizer. Because the solvent is a mixture with water, there is no danger of condensing damaging moisture into the wash solution. The example illustrates using a moist gas stream to dry the solid without dehydrating it.

Example 2
1-(2-(4-benzyl-4-hydroxy-piperidin-1-yl)-ethyl)-3-(2-methyl-quinolin-4-yl)-urea sulfate trihvdrate.

To a suspension of 1-(2-(4-benzyl-4-hydroxy-piperidin-1-yl)-ethyl)-3-(2-methyl-quinolin-4-yl)-urea (21.36 kg) in CH₃OH (178 L) is added aqueous H2SO4 (6 L, 9.91%) during 10 min. The clear solution is filtered and further aqueous H2SO4 (33.8 L, 1.07 M) is added during 45 min. The solution is cooled to -2°C during 1.5 h and stirred at -5 to -9°C for 1 h. The formed precipitate is filtered, washed with cooled CH30H (- 5°C, 54 L) and dried under a stream of nitrogen provide 1-(2-(4-benzyl-4-hydroxy-piperidin-1-yl)-ethyl)-3-(2-methyl-quinolin-4-yl)-urea sulfate of formula l as a non-defined hydrate. A slurry of the so obtained salt in H2O (16.2% w/w) is stirred for 3 days at 25°C. Filtration and drying at 30°C under a wet stream of nitrogen (50% RH) provides the title compound.

This example replaces the ethanol with methanol and is in most particulars very much the same. Here unhydrated gas was used in the drying ad there apparently was some dehydration. Stirring a slurry in water for an extended period recreates the hydrate illustrating a method of preparing a pseudopolymorph hydrate. Drying to the trihydrate was successful when the relative humidity was controlled at 50%.

Example 4
1-(2-(4-benzyl-4-hydroxy-piperidin-1-yl)-ethyl)-3-(2-methyl-quinolin-4-yl)-urea sulfate dihvdrate

1-(2-(4-benzyl-4-hydroxy-piperidin-1-yl)-ethyl)-3-(2-methyl-quinolin-4-yl)-urea (15.4 kg, 1 equivalent) is dissolved in ethanol (78 L) and the mixture is heated at 50°C. Aqueous sulfuric acid (1M, 1.1 equivalents) is added during minutes. The crystallization is initiated by a wet seed of Example 1 (1%) as described below. The suspension is cooled to 1°C with a cooling rate of 14C°/h and maintained at this temperature at least 11 hours before filtration and washing with aqueous ethanol (50 % W/W, 50 L). The solid is dried at 30°C under a wet stream of nitrogen (33-40% RH) to provide the title compound with a purity of 99.4% with a yield of approximately 79%.
The wet seed used in the above procedure is prepared by mixing - 1-(2-(4-benzyl-4-hydroxy-piperidin-1-yl)-ethyl)-3-(2-methyl-quinolin-4-yl)-urea sulfate dihvdrate (Example 1,) with a saturated solution (421 9) of 1-(2-(4-benzyl-4-hydroxy-piperidin-1-yl)-ethyl)-3-(2-methyl-quinolin-4-yl)-urea sulfate dihvdrate (Example 1, 73.9 9) in aqueous ethanol (50 % W/W, 810 9).

Example 5

1-(2-(4-benzyl-4-hydroxy-piperidin-1-yl)-ethyl)-3-(2-methyl-quinolin-4-yl)-urea sulphate dihvdrate

1-(2-(4-benzyl-4-hydroxy-piperidin-1-yl)-ethyl)-3-(2-methyl-quinolin-4-yl)-urea (1.01 kg, 1 equivalent) is dissolved in ethanol (3.05 kg) under stirring (200±20 rpm) and the mixture is heated at 50°C. Aqueous sulfuric acid (1 M, 1.1 equivalents) is added during 20 minutes. The crystallization is initiated by a wet seed of Example 1 (1 %) as described below. The obtained mixture is maintained at 50°C for about 15 minutes, then it is cooled to 0°C with a cooling rate of 15°C/h and maintained at this temperature for least 1 hour before filtration and washing with aqueous ethanol (50 % W/W, 3 kg). The solid is dried in a conductive agitated dryer at a temperature of 35± 3°C under a wet stream of nitrogen (45±5% RH), optionally under stirring (max. rpm) in case the cake humidity is below 25%, to provide the title compound with a purity of 99.8% with a yield of approximately 94%.

The wet seed used in the above procedure is added in two shots and is prepared by mixing 1-(2-(4-benzyl-4-hydroxy-piperidin-1-yl)-ethyl)-3-(2-methyl-quinolin-4-yl)-urea sulfate dihvdrate (Example 1, 6.5 9) with a saturated solution (13.9 9 for the first shot, plus 15.6 9 for subsequent rinsing and second shot) of 1-(2-(4-benzyl-4-hydroxy-piperidin-1-yl)-ethyl)-3-(2-methyl-quinolin-4-yl)-urea sulfate dihvdrate (Example 1, 7.0 9) in aqueous ethanol (50 % W/W, 50.0 9) for about 2 minutes. The first shot of wet seed is prepared at least 5 minutes before use to ensure that the seed is correctly wetted.

Example 5 illustrates that in a process description of crystallization the mixing times, cooling rates and stirring need to be precisely controlled. The need to properly moisten the seed crystals with the solvent is also illustrated. If the seeds’ surface does not wet properly they cannot catalyze crystal growth properly.

US20060194833A1: Crystalline 1H-imidazo[4,5-b]pyridin-5-amine, 7-[5-[(cyclohexylmethylamino)-methyl]-1H-indol-2-yl]-2-methyl, sulfate (1:1), trihydrate and its pharmaceutical uses

According to the method, ER807447 is first suspended in water to form an aqueous suspension. Sulfuric acid is added to the aqueous suspension to form a solution while keeping the internal temperature of the solution below 25° C. The solution typically has a yellow color. The solution may optionally be filtered to remove particulates from the solution. Other techniques for removing particulates known in the art, centrifuging, etc. may be used as the filtering step. The solution is then slowly warmed until E6070 crystallizes from solution. The solution may be warmed to about 100° C. Typically crystal formation occurs at temperatures of about 70° C. Preferred rates of warming typically range from about 30 minutes to 5 hours. Longer or shorter times may be used, particularly depending upon the batch size. E6070 may not crystallize as readily from highly dilute solutions.

To enhance crystallization, an anti-solvent may be used in the method of making the crystalline E6070 or to recrystallize crystalline E6070. The recrystallization procedure is described in Example 5. In the above method, the anti-solvent may be added to the aqueous suspension before sulfuric acid addition or to the solution after sulfuric acid addition and the optional filtration step. Useable anti-solvents and their use are known in the art. Typical anti-solvents include water-miscible anti-solvents such as, for example, methanol, ethanol, 1-propanol, 2-propanol, acetone and mixtures thereof. When an anti-solvent is used, the solution may become cloudy. It is generally not necessary to warm the solution to as high of temperatures as when just using an aqueous solution.

The procedure above illustrates forming a bisulfate from water. Filtration or other clarification of the formed solution is illustrated. Removing insolubles removes nuclei that can catalyze improper nucleation. The example illustrates that a bisulfate salt in water may actually be supersaturated but the rate of nucleation may be impracically slow. Heating the solution increases the rate of nucleation and causes the insoluble salt to come out. If the sulphate is a high molecular weight molecule that should give an insoluble sulphate in water, perhaps heating will enhance the rate of seed formation as here.

The use of an antisolvent is also illustrated. Note that water miscible organics are often antisolvents for sulphate salts because sulphate salts are so hydrophilic.

Tizanidine Monosulfate (5-chloro-4-f2-imidazolin-2-ylamino')-2,l,3-benzothiadiazole monosulfate )

In a second preparation, tizanidine monosulfate was prepared by the following method: to solid tizanidine (9.957 g; 39.24 mol) was added a solution of sulfuric acid (5.438 g; 55.45 mol) in acetonitrile (175 mL). The yellow solid rapidly converted to a white crystalline solid. The mixture was heated to 60°C and stirred for 90 minutes. The mixture was cooled to room temperature and the solid was subsequently filtered and washed with additional acetonitrile (50 mL). The solid was collected via filtration and air-dried.

Tizanidine monosulfate comprises a 1:1 ratio of ionized tizanidine to sulfate counterion. In this example sulfuric acid in acetonitrile is used

Rosiglitazone Sulfate

Example 1:

5-[4-[2-(N-methyl-N-(2-pyridyl)amino) ethoxy]benzyl] thiazolidine-2,4-dione sulfate

5-[4-[2- (N-Methyl-N-(2-pyridyl)amino) ethoxy]benzyl]thiazolidine-2,4- dione (20.0 g) in glacial acetic acid (50 ml) was stirred and heated to 75°C until a clear solution 5 was observed. Concentrated sulfuric acid (1. 5 ml) was added and the stirred solution cooled to 21 °C. After evaporation of solvent under reduced pressure, methanol (100 ml) was added and the mixture stirred at 21°C for 48 hours. The solid was collected by filtration, washed with methanol (50 ml) and dried under vacuum to give 5-[4-[2-(N-
methyl-N-(2-pyridyl)amino)ethoxy]benzyl]thiazolidine-2,4- dione sulfate (10.7 g) as a crystalline solid.

Melting point: 184 - 189°C.

DSC: Tosser = 184.4°C, Tpeak = 189.1 °C Elemental Analysis: 15 Found: C; 52.96 H; 4.94 N; 10.23 S; 11.78 Theory: (C36H40N6O,oS3) C; 53. 19 H; 4.96 N; 10.34 S; 11.83

In this example glacial acetic acid is used as solvent for the free base and heating is require to get a clear solution. The experimentalist apparently expected the sulfate to precipitate at about ambient but it did not. It is important that in this example the basic group was tertiary because the combination of acetic acid and sulfuric acid could cause acetylation with primary or secondary amines. Undeterred the experimentqalist removed some acetic acid under vacuum and replace it with the antisolvent methanol. This time cooling gave the desired solid derivative.

Example 2:

5-[4-l2-(N-methyl-N-(2- pyridyl)amino) ethexylbenzyl] thiazolidine-2,4 20 dione sulfate

5- [4-[2-(N-Methyl-N-(2- pyridyl)amino)ethoxy]benzyl] thiazolidine-2,4-dione (40.0 g) in glacial acetic acid (100 ml) was stirred and heated to 70°C until a clear solution was observed. Concentrated sulfuric acid (3.1 ml) was added and the mixture stirred for 10 minutes at 70°C, then cooled to 21°C with stirring. The solvent was evaporated under reduced pressure, followed by the addition of methanol (100 ml) and the mixture was stirred at 21 °C until crystallization was complete. The product was collected by filtration, washed with methanol (200 ml) and dried under vacuum over phosphorus pentoxide for 4 hours at 50°C to give 5-[4-[2-(N-methyl-N-(2 pyridyl)amino)ethoxy] benzyl]thiazolidine-2,4-dione sulfate (36.9 g) as an off white crystalline solid.

Example 3:

5-[4-[2-(N-methyl-N-(2-pyridyl)amino) ethexy]benzyl] thiazolidine-2,4 dione sulfate

Concentrated sulfuric acid (1.94 ml) was added to a stirred suspension of 5-[4- [2 35 (N-methyl-N-(2-pyridyl)amino) ethoxy]benzyl] thiazolidine-2,4-dione (25.0 g) in methanol (1000 ml) at 56°C. The reaction mixture was stirred at 60°C until a clear solution was observed, then cooled to 21 °C and stirred at this temperature for 16 hours.
The product was collected by filtration, washed with methanol (100 ml) and dried under vacuum at 21°C for 3 hours to afford 5-[4-[ 2-(N-methyl-N- (2 pyridyl)amino)ethoxy]benzyl]thiazolidine-2,4- dione sulfate (19.5 g) as a white crystalline solid.

By far the most frequently successful pharmaceutical salt is the hydrochloride. In fact the hydrochloride salt is selected 50% of the time when chemists look for an acceptable salt. Typically, there must be a very good reason for not adopting this salt. If a hydrochloride crystallizes, one typically needs a n excellent reason not to use the hydrochloride.

The hydrochloride is a preferred choice because chloride does not have any activity of its own, unlike bromide, nitrate and others. Hydrochloric acid is a very significant acid in the stomach. By salt exchange hydrochlorides are formed to some extent no matter what the counter ion of an API is in the pharmaceutical product.

Hydrochloric acid is a strong mineral acid strong enough to quantitatively protonate even weak bases.

Hydrochlorides characteristically are substantially more soluble than the free bases used to make them, so the hydrochloride typically improves the bioavailability.

Hydrochlorides can be prepared in aqueous solution, in protic organic solvents, in aprotic organic solvents, and in non-polar solvents because hydrogen chloride can exist in both a covalent form in apolar solvents or as ionized protons and chloride ions in more polar solvents. The actual acidity varies being equal to the acidity of the conjugate acid of the solvent molecule. That is to say hydronium ions exist in water, protonated alcohol ions in alcohol, protonated acetic acid in glacial acetic acid or protonated ethyl acetate molecules in ethyl acetate. The multiple forms of HCl result in multiple techniques for the addition of the hydrogen and chloride ions to the pharmaceutical base we need to make into a salt.

Hydrogen chloride gas can be passed into neat organic solvents to prepare titratable molar solutions that are quite stable. Hydrogen chloride in lower alcohols is not stable for a long time and must e used soon after it is formed. More often the gas is added to the base dissolved in the lower alcohol. HCl forms quite stable solutions in IPA which can be sotred at ambient temperatures for several days. Hydrochloric acid solutions in non aqueous solution can be made by adding acetyl chloride into ethanol wher a quantitative reaction occurs to give hydrogen chloride and an equamolar amount of ethyl acetate.

A recent PCTpatent teaches the creation of hydrogen chloride in situ from trimethylsilyl chloride and any solvent with a silylated functionality or any inert solvent containing a slight excess over the silyl chloride of a silylable group.

Hydrochlorides can be made by reaction of the organic base with an equivalent of ammonium chloride. The stronger organic base preferentially takes the hydrogen chloride and the ammonia gas is liberated and may exit the reaction mixture.

If the free base of concern has some solubility in water, the hydrochloride can be made from aqueous hydrochloric acid and the base in water. Often heating is required to get the free base dissolved and the hydrochloride salt separates on cooling. In aqueous solution the solubility of the amine hydrochloride can be decreased by the addition of additional inorganic water soluble chloride to increase the common chloride ion. The addition of inorganic salts also increases the precipitation by the salting out effect.Excess hydrochloric acid can be used to decrease the solubility of the desired salt so long as the pharmaceutical chloride is stable in strong aqueous acid.

Standard aqueous solutions of hydrochloric acid can be added directly into the base dissolved in a water miscible polar solvent such as methanol, ethanol, propanols, butanols, acetone, 2-butanone, acetonitrile,

The most powerful and widely practiced method of making a hydrochloride salt in the laboratory is to add gaseous hydrogen chloride into a diethyl ether solution of the free base. If you think a hydrochloride salt might not be crystalline, this technique is likely to provide evidence one way or the other. It is not a practical process method to make the slat but it will give evidence that the solid salt is possible and will provide some seed crystals for other preparative methods.

If a solid pharmaceutical hydrochloride is formed, the next goal is to obtain it in a satisfactory recovery. Pharmaceutical bases are typically expensive moieties and losing material in a low recovery hydrochloride formation is undesirable. When a solution of a hydrochloride and the pharmaceutical base has been formed but only a small amount or no crystals at all precipitate three strategies are possible

1. cool the solvent to decrease the overall solubility in the solvent volume
2. all an antisolvent which changes the quality of the solvent system and lowers the solubility
3. add an antisolvent and then cool

As a rule of thumb if the recovery is 80% or more at ambient temperature, simply cooling the solvent can be expected to give an excellent recovery; if the recovery is 40-80% at ambient cooling should be applied and then an altisolvent judiciously added; but when the starting recovery is less than 40% an antisolvent should be added to just the cloud point and then cooling should be applied.

Treatment is completely reversible. What you cool down you can rewarm so this is usually tried first.

If treatment 2 is used it is useful for the analysis of the results to take a sample of the solid obtained just by cooling in order to measure the purity at that point, then add the antisolvent to give a practical recovery and compare the purity of the product when using an anti-solvent with the purity before that addition.

When the solid formation in the single solvent condition is low or none, an anti-solvent is added to the cloud point either to a hot or ambient solution and then controlled cooling is applied to try for crystallization.

Mixtures of solvents are not preferred in scale up processes because it introduces the nedd for an in process test to guarantee the proper solvent ratio. Second crops are more difficult to obtain from a mixed solvent in a simple procedure. Nevertheless situations where a mixed solvent gives the best purification and recovery do occur. It is advantage if two solvents differ substantially in boiling point. This allows recovery of pure solvents from the filtrate.

Solvents that form an azeotrope with water have the advantage that it is easier to be sure that the crystallization is under anhydrous conditions.

An article entitled “Pharmaceutical Manufacturing Goes Green” appeared in the recent March 1 2008 issue of Genetic Engineering & biotechnology News (GEN). It was stated that “advances in enzymatic catalysis of synthetic reactions, solvent substitutions, and recycling of by-products and waste may not only reduce the environmental impact of pharmaceutical processes but, with the potential to have a positive effect upon synthetic efficiency and overall productivity, can also decrease waste streams, lessen energy input, and minimize the need for hazardous reagents”.

It is only human nature that academic scientists who have hitched their career aspirations to green chemistry will advocate the greenness of a process as the essential basis for process evaluation; but realistically all that can be hoped for is that industrial process chemists will increasingly include green technology and its potential advantages to the corporation among all the basis upon which a preferred process is selected. There is, however, good reason to expect that this will happen.

As the GEN article notes realistically, ‘This may be a particular challenge in organic chemistry, in which many of the classical chemical reactions have been in use for decades to produce small molecule drugs”.

The industry does have an advantage in this re-education program since at this time there is a particularly rapid turnover of chemists because so many of them are reaching retirement age. The ranks must necessarily be quickly filled up with younger more environmental conscious and hopefully trained in universities very aware of the need for green chemistry.

Because of the patent laws and health regulations in the main, world pharmaceutical markets, the emphasis in the pharmaceutical industry must be on the synthetic route which is overall the most rapid, inexpensive, scaleable and rugged. The high risk of failure and the cost of even a successful development necessitate that pharmaceutical companies maximize the number of years that they have a monopoly in the marketplace before patent expiry. This forces these organizations towards using the first dependable, economic and scaleable process that will deliver a product that can