AU9821398A - Paroxetine hydrochloride anhydrate - Google Patents

Description

ARA

USTRALIA-

PATENTS ACT 1M9 COMPLETE

SPECMFCAT!ON

FOR A STANDARD PATENT-

(ORIGINAL)

Name of Applirn~ .:Act~U nventors: Address for Service:.

invention Titlte: SmithKline Beecham PlC Viqtor Witold JACEWICZ and Neat WARD DAVIES COLLISON CAVE, Paten-Attoineys, I Little Collins:Street, Melbourne, 3000-.

Novel compounds The following stateme~nt is a full description,--of this. invenation,- including the best method of performing it known to us, This application is a divisional application derived from Australian PatentApplication No.

43328/96, the entire contents of which are incorporated herein by reference.

-The present invention relates to novel compounds, to processes for preparing them and to their use in treating medical disorders.

EP-B-223403 (Beecham Group plc) describes paroxetine hydrochloride hemihydrate and its use in treating certain medical disorders. Example 8 in this documint describes the preparation of paroxetine hydrochloride anhydrate as platelets melting at 118°C and with IR bands at 890, 1200, 1490, 3400 and 3640cm-1, by crystallisation from a water-containing solvent This material is hereinafter referred to as Form Z. Subsequent repetition of the preparation described in Example 8 has failed to yield any type of paroxetine hdrochloride anhydrate, and there is no clear teaching elsewhere in the document of any alternative route or modification to the process which would generate the anhydrate.

Paroxetine hydrochloride anhydrate is also purported to be disclosed in the International Journal of Pharmaceutics 42, (1988) 135 to 143, published by Elsevier. Theanhydrate is said to be produced by crystallising paroxetine hydrochloride from anhydrous propan-2-ol. Subsequent repetition of this process has resulted in a propan-2-ol solvate of paroxetine hydrochloride. That is to say that there is bound propan-2-ol in the product This bound propan-2-ol cannot be removed by conventional drying techniques such as vacuum oven drying.

S Paroxetine hydrochloride anhydrate substantially free of bound propan-2-ol, has not been lescfibed in the literature, nor has any method been disclosed which would yield such a product 2' as san inevitable result. A method for preparing paroxetne hydrochloride anhydrate substantially free of bound propan-2-ol has now been found. Furthermore, surprisingly, four new forms of paroxetine hydrochloride anhydrate have been found as have processes for their preparation.

These forms are hereinafter referred to as A, B, C and D respectively. The characterising data for the Forms A, B, C and D do not correspond to the characterising data provided in Example 8 3Q. .gofEP-A-223403.

S Accordingly, the present invention provides paroxetine hydrochloride anhydrate S'-ubstantially free of bound propan-2-ol.

The present invention also provides paroxetine hydrochloride anhydrate substantially free ~tf bound organic solvent.

hyd *h t ally The present invention also provides paroxetine hydrochloride anhydrate substantially free of propan-2-ol with the proviso that it is other than FormZ.

-la- *i P31122 Substantially free of bound organic solvent is to be interpreted to be less than the amount of propan-2-ol which would remain solvated, bound, within the crystal lattice of the product under conventional vacuum oven drying conditions.

The present invention also provides paroxetine hydrochloride solvates other than the__ propan-2-ol solvate as precursors in the preparation of paroxetine hydrochloride anhydrate substantially free of bound organic solvent. Examples of such solvates include solvates from alcohols (other than propan-2-ol) such as p: :pan-l-ol and ethanol; solvates from organic acids such as acetic acid; solvates from organic bases such as pyridine; solvates from nitriles such as acetonitrile; solvates from ketones such as acetone; solvates from ethers such as tetrahydrofuran and solvates from chlorinated hydrocarbons such as chloroform and solvates of hydrocarbons such as toluene.

Preferably, paroxetine hydrochloride anhydrate substantially free of bound propan-2-ol is provided in substantially pure form. Suitably, paroxetine hydrochloride anhydrate substantially free of bound propan-2-ol is provided with a purity of the paroxetine hydrochloride anhydrate of greater than 50%, preferably greater than 60%, more preferably greater than 70%, yet more .:'..preferably greater than 80% and even more preferably greater than 90%. Most preferably the .paroxetine hydrochloride anhydrate is provided in substantially pure form, paroxetine 2, .hydrochloride anhydrate substantially free of bound propan-2-ol is greater than 95% pure.

It should be understood that the present invention comprising paroxetine hydochloride anhydrate substantially free of bound propan-2-ol may contain unbound water that is to say 'water which is other than water of crystallisation.

Typically the amount of bound organic solvent on a weight for weight basis would be less than preferably less than more preferably less than even more preferably jess than yet more preferably less than 0.5% and most preferably less than 0.1%.

Generally, all percentages indicated herein are on a weight for weight basis unless otherwise stated.

Preferred forms of paroxetine hydrochloride anhydrate substantially free of bound propan-2-ol or substantially free of bound organic solvent include; i) paroxetine hydrochloride anhydrate in Form A; (as hereinafter defined) ii) paroxetine hydrochloride anhydrate in Form B; (as hereinafter defined) iii) paroxetine hydrochloride anhydrate in Form C; (as hereinafter defined) iv) paroxetine hydrochloride anhydrate in Form D; (as hereinafter defined) i P31122 iThe forms of paroxetine hydrochloride anhydrate may be distinguished from each other and the material formed as a result of carrying out the procedures mentioned in EP-B-0223403 and the International Journal of Pharmaceutics 42, (1988), 135 to 143, by crystalline shape, solvent analysis or techniques such as-IRmeldng point, X-ray-diffraction, NMRDSC, microscopy and any other analytical techniques which differentiate one form from another.

For example, Form A substantially free of solvent may be distinguished from other forms by the following analytical data. Form A has a melting point of about 123-125°C when obtained in similar purity to the material described in Example 1 which may be determined by conventional methods such as HPLC and significant IR bands (Figure 1) at about 513, 538, 571, 592, 613, 665,722, 761, 783, 806, 818, 839,888,906, 924, 947, 966,982, 1006, 1034, 1068, 1091, 1134, 1194, 1221, 1248, 1286, 1340, 1387, 1493, 1513, 1562, 1604, 3402, 3631 cm- 1 l The DSC exotherm, measured at 10°C per minute shows a maximum at about 126°C using an open pan and a maximum at about 121°C using a closed pan. Form A also has a substantially similar X-ray diffractogram to that shown in Figure 4, for example there are i characteristic peaks at 6.6. 8.0, 11.2, 13.1 degrees 2 theta and a substantially similar solid state NMR spectrum to that shown in Figure 7 for example with characteristic peaks at 154.3. 149.3, 141.6, 138.5 ppm.

Form B substantially free of solvent may be distinguished from other forms by the S following analytical data, i.e.it has a melting point of about 138°C when obtained in similar purity to the material described in Example 7 which may be determined by conventional mrethods such as HPLC and significant IR bands (Figure 2) at about 538, 574, 614, 675,722, 762,782,815,833,884,925,938,970,986,1006,1039,1069, 1094,1114,1142,1182, 1230, :.1274,1304,1488,1510,1574,1604,1631 cm- 1 The DSC exotherm, measured at 10*C per minute, shows a maximum of about 137°C in both open and closed pans. Form B also has a substantially similar X-ray diffractogram to that :.:shown in Figure 5, for example, there are characteristic peaks at 5.7, 11.3, 12.4, 14.3 degrees 2 1 ':them and a substantially similar solid state NMR spectrum to that shown in Figure 8, for example with characteristics peaks at 154.6, 148.3, 150.1, 141.7, 142.7, 139.0 ppm.

Form C may be distinguished from other forms by the following analytical data, i.e. it has a melting point of about 164°C when obtain in similar purity to the material described in Example 8 which may be determined by conventional methods such as HPLC and has significant IR bands (Figure 3) at about 540,.574, 615, 674,720,760, 779, 802, 829, 840, 886, 935,965,984,1007, 1034, 1092, 1109, 1139, 1183, 1218,1240, 1263, 1280, 1507, 1540, 1558, 1598,1652 cm-

I

P31122 The DSC exotherm, measured at 10C per minute, shows a maximum of about 161°C in both open and closed pans.

Form C also has a substantially similar X-ray diffractogram to that shown in Figure 6, for example there axe characteristic peaks at 10.1, 12.1, 13.1, 14.3 degrees-2-theta-ard-a substantially similar solid state NMR spectrum to that in Figure 7, for example with characteristic peaks at 154.0, 148.5, 143.4, 140.4 ppm.

Form D may be distinguished from other forms by the following analytical data in that it exists as a semi-crystalline solid with a melting point of about 125°C when obtained in similar purity to the material described in Example 14 which may be determined by conventional methods such as HPLC Form D may also be characterised in that it has essentially similar physical characteristics when prepared from a toluene precursor solvate using methods generally described herein said toluene precursor solvate having significant IR bands at about 1631, 1603, 1555, 1513, 1503, U 1489, 1340, 1275, 1240, 1221, 1185, 1168, 1140,1113, 1101, 1076, 1037, 1007,986,968,935, -'924, 885, 841,818,783,760,742,720, 698, 672,612, 572,537 a-d 465 cm-1, and ".characteristic X-ray diffraction peaks at 7.2, 9.3, 12.7 and 14.3 degrees 2 theta.

The question of which particular form a particular sample of paroxetine hydrochloride anhydrate is would be readily determined by one skilled in the art using conventional techniques with reference to the data provided above that given in the examples and any other conventional "means.

Preferably forms A and B exist as needles ad form C edts as needles or prisms.

The present invention also provides a process for the preparation of paroxetine hydrochloride anhydrate substantially free of propan-2-ol which comprises crystallising paroxetine hydrochloride in either;, 4)b an organic solvent or mixture of organic solvents which form a solvate with the paroxetine hydrochloride and which are not removable by conventional drying techniques; or ii) an organic solvent or mixture or organic solvents which do or do not form a solvate with the paroxetine hydrochloride but which are removable by conventional vacuum oven drying; thereafter in the case of i) displacing the solvated solvent or solvents using a dis-placing agent and in the case of ii) by removing the solvent.

The present invention also provides a process for the preparation of the paroxetine hydrochloride solvates other than the propan-2-ol solvate which comprises crystallising P31122 paroxetine hydrochloride in an organic solvent or mixture of solvents which form a solvate with the paroxetine hydrochloride and which are not removable by conventional drying techniques.

The present invention also provides a process for the preparation of paroxetine hydrochloride anhydrate substantially free of bound organisolvent which-comprises displacingthe solvated solvent or solvents from a paroxetine hydrochloride solvate using a displacing agent In one preferred aspect of the invention crystallisation of paroxetine hydrochloride anhydrate is achieved by contacting a solution of paroxetine free base in an organic solvent or solvents with dry hydrogen chloride gas.

Alternatively, prior to the crystallisation of the paroxetine hydrochloride water may be removed by azeotropic distillation. Suitable solvents therefore include those which form an azeotrope with water such as pyridine and propan-2-ol. It should also be appreciated that nixtures of solvents can also be used to aid the azeotropic removal of water.

Thus, in another aspect of the invention paroxetine hydrochloride anhydrate is .crystallised by dissolving paroxetine hydrochloride hemi-hydrate in an appropriate solvent substantially free of water which forms an azeotrope with water. Suitably solvent is removed by distillation and fresh solvent substantially free of water is added until all of the water is removed.

Paroxetine hydrochloride hemi-hydrate or the free base thereof may be prepared :*::according to the procedures generally outlined in EP-B-0 223403.

The organic solvents should be substantially free of water to the extent that there is .insufficient water present at the time of crystallisation to effect conversion to the hydrochloride *:....emi-hydrate Organic solvents which are substantially free of water may be obtained in conventional manner. For example they can be dried using conventional techniques such as drying over molecular sieves or they can be purchased.

Factors which affect which form of the product will be obtained include the particular choice of organic solvent or solvents to be used will depend upon the particular form of the product which is desired.

It should also be appreciated that the method of solvent removal also depends upon the particular form of the product which is desired.

For process variant i) it should be appreciated that an organic solvent or solvents which form a solvate with the crystallised paroxetine hydrochloride and which are not removable by conventional drying techniques may be determined by a matter of routine experimentation.

Examples of such organic solvents include, but in no way are limited to, alcohols especially P31122 alkanols such.as propan- 2 -ol, ethanol and propan-1-oi; organic acids such as acetic acid; organic bases such as pyridine; nitriles such as acetonitrile; ketones such as acetone; ethers such as tetrahydrofuran and chlorinated hydrocarbons such as chloroform.

The paroxetine hydrochloride solvate produced by process variant i) is suitably isolated and dried by conventional methods such as drying in vacuo to remove some or all of the free or unbound solvent It should be appreciated that it is preferable and unexpected that the degree of drying is controlled such that only free solvent is removed. The bound solvent is then displaced with a displacing agent such as water or supercritical carbon dioxide. It is possible to use other displacing agents which may be selected by means of routine experimentation.

Preferably gaseous or liquid water may be used as a displacing agent It is important that the paroxetine hydrochloride solvate is contacted with enough water and for sufficient time to displace the solvent but insufficient to cause conversion to the hydrochloride hemi-hydrate- The amount of water, the form of the water, eg, liquid or gaseous and the length of time which the paroxetine hydrochloride solvate is contacted with the water differs from solvate to solvate. This depends largely upon the solubility of the solvate in question.

Particular ratios of paroxetine hydrochloride solvate to water are outlined in the examples hereinafter described (Examples 1, 4 to 6,9 to 11, 13 and 15). It should be appreciated that the pyridine solvate is believed to be more soluble in water than for example the propan-2-ol solvate. Thus the use of the common ion effect when using diluted hydrochloric acid may help ":'prevent dissolution of the solvate and subsequent conversion to the hydrochloride hemi-hydrate.

After contact with water to displace the bound solvent the product is suitably dried, for example, in vacuo at elevated temperature. Suitable drying may be over a desiccant such as 'phosphorus pentoxide.

When supercritical carbon dioxide is used it should be appreciated that the flow rate, temperature and pressure of the carbon dioxide may be controlled to give optimum solvent removal from the paroxetine hydrochloride solvate. Generally high pressure carbon dioxide may be used for example at about 2,500 psi. Elevated temperatures may also be preferably used such as between 50 to 80 0 C. More preferable between 55 to 75 0

C.

Process variant i) is preferably used to prepare Form A.

Preferably the crystallisation of the paroxetine hydrochloride anhydrate Form A precursor solvate may be facilitated by the addition of seeds of paroxetine hydrochloride anhydrate Form A precursor solvate.

P31122 Alternatively, sweds of paroxetine, hydrochloride anhydrate Form A may be used to ficilitate the crystallisation of paroxetine hydrochloride anhydrate Form A precursor solvates.

For process variant ii) it shouldbe-a-ppreciated-that-an iorgaric-solvent or mixture of organic solvents whidch does or does not form a soiv ate with the paroxetine hydrochloride but which is removable by conventional vacuum oven drying may be determined by a matter of routine experimentation.

An example of a solvent which forms a bound solvate with the paroxetine hydrochloride but which is removable by conventional vacuum oven drying is toluene- Toluene is preferably used to prepare Form D.

The crystallisation of paroxetine hydrochloride anhydrate Form D precursor solvates may be facilitated by the addition of seeds of paroxeri-ne hydrochloride anhydmate Form D precursor solvates Seeds of paroxetine hydrochloride anhydrate Form D may be used to facilitate the grsalsation of paroxewie hydrochloride anhydrate Form D precursorr solvates.

2d Examples ofl solvents which do not form a bound sol-eate with paroxetine hydrochloride Lbut which are removable by conventional vacuum oven drying are butan-l-ol and ethyl acetate.

Butan-l-ol is preferably used to prepare Form B and that butan-1-ol or ethyl acetate are preferably used to prepare Form C if Form B is required this may be prepared according to or analogously to the procedures outlined in Example 7.

Preferably the use of seeds of Form B may be used to facilitate the crystallisation of "FormB.

if Form C is required this may be prepared according to or analogously to the procedures outlined in Examples 8 and 12.

It should also be appreciated that the use of seeds of Form. C may be used to facilitate the crystallisation of Form C Seed crystals of Forms A, B, C and D may be prepared according to the procedures described herein or are freely available upon request to Corporate Intellectual Property.

SmithKline Beecham plc at New Frontiers Science Park, Third Avenue, Harlow, Essex, CM19 United Kingdom Form A is BRL 2906OF; Form B is BRL 29060G; Form C is BIRL P3112M 2906011;T- Form D is BRL 2906011 Samples of seeds of Forms A. B. C and D m-ay also be obtained f-rm t NCIMB, 23 St. Machor Drive, Aberdeen, ALB2 IRY, Scotland, United -Kingdom- Parxetine hydrochloride an-hydrate substantially fEm of propan- 2 -ol and Forms A, B, C and D (all of which are hereinafter referred to as the 'products of the invention") can be used to treat and prevent the following disorders: Alcoholism Anxiety Depression Obsessive Compulsive Disorder PRanic Disorder Chronic Pain dbesity S, enile Dementia Xuigrin Anorexia Social Phobia *.,Pm-Menstrual Syndrome (PMS) .:Molescent Depression 125' *'liichotillomania Dysthymia.

'Sistance Abuse These disorders are herein after referred to as "the Disorders"- 30.1bU~ fteDsresb diitrn r fetv n/rpohlci muto h The present invention further provides a method for treting and/or preventing any one The present, invention further provides a pharmaceutical composition for use in the treatment and/or prevention of the Disorders which comprises admixing the products of the invention with a phrumaceutically acceptable carrier.

The present invention also provides the use of the products of the invention for treating and/or preventing te Disorders -8- ~,~lrrr~w~-r~nrarruwarumur~-csmcys~ -n~ P31122 The present invention also provides the use of the pOducts of the invention in the manufacture of a medicament for treating and/or preventin the Disorders.

Preferred disorders include depression, OCD and panic.

The compositions of this invention are usually adapted for oral administration, but formulations for dissolution for parental administration are also within the scope of this invention.

The composition is usually presented as a unit dose composition containing from I to 200mg of active ingredient calculated on a free base basis, more usually from 5 to 100mg, for example 10 to 50mg such as 10, 12.5, 15, 20, 25, 30 or 40mg by a human patient Most preferably unit doses contain 20mg of active ingredient calculated on a free base basis. Such a composition is normally taken from I to 6 imes daily, for example 2, 3 or 4 times daily so that the total amount of active agent administered is within the range 5 to 400mg of active ingredient calculated on a free base basis. Most preferably the unit dose is taken once a day.

S Preferred unit dosage forms include tablets or capsules.

The compositions of this invention may be formulated by conventional methods of admixture such as blending, filling and compressing.

Suitable carriers for use in this invention include a dilutent, a binder, a disintegrant, a *'colouring agent, a flavouring agent and/or preservative. These agents may be utilised in conventional manner, for example in a manner similar to that already used for marketed ansidepressant agents.

Specific examples of pharmaceutical compositions include those described EP-B-0- 223403, and US 4,007,196 in which the products of the present invention are used as the active ingredients.

The following examples illustrate the present invention: i P31122 EXAMPLE 1 Crystalline Paroxetine Hydrochloride Anhydrate Substantially Free of Bound Propan-2-ol (Form A) i) Paroxetine hydrochloride propan-2-ol solvate Paroxetine hydrochloride hemihydrate [150 g] was stirred with propan-2-ol [1000 ml] and toluene [300 ml] in a round bottom flask and heated to boiling. Solvent was removed by distillation, the total volume being maintained by adding fresh propan-2-ol, until the boiling point had reached approximately 82 0 C, indicating that all the water had been removed.

The mixture was allowed to cool to approximately 50°C, when it spontaneously crystallised. The contents of the flask rapidly set to a thick paste which was diluted with 'propan-2-ol [approx. 500 ml] and stirred vigorously. The resulting suspension was g.llowed to cool to approximately 30°C and filtered under vacuum, taking care to avoid Sthe absorption of atmospheric moisture. The solvent-wet cake was dried in high vacuum i I over phosphorus pentoxide.

S.::'ield of solvated paroxetine hydrochloride 151 g, propan-2-ol content 13.0% (estimated by NMR).

SThe infra-red spectrum (Nujol mull) showed inter alia a characteristic band at 667 cm 1 1 Paroxetine hydrochloride anhydrate (Form A) aroxetine hydrochloride propan-2-ol solvate [110 g, propan-2-ol content 13.0%] was W rred in a beaker with water [275 ml] for 20 minutes. The mixture was filtered under vacuum and the damp solid dried in vacuum over phosphorus pentoxide to constant weight.

Yield of paroxetine hydrochloride anhydrate Form A 91.0 g Water content 0.13% propan-2-ol content 0.05% (estimated by NMR).

Melting point: 123-125 0

C

The DSC exotherm, measured at 10°C per minute, showed a maximum at about 126 0

C

using an open pan and a maximum at about 121*C using a closed pan.

The infra-red spectrum [Nujol mull] showed inter alia characteristic bands at 665, 3631 S and 3402 cmrl (seeFigure 1).

Elemental analysis: ,1 P31122 Requires for paroxetine hydrochloride anhydrate: C 62.38 H 5.79 N 3.83% Found: C62.10 H5.89 N 3.67% The sample was also examined by X-ray powder diffraction (see Figure 4) and solid state C13 NMR (see Figure 7).

EXAMPLE 2 Paroetine Hydrochloride Propan-2-ol Solvate Paroxetine free base (42.09 g) was dissolved in propan-2-ol (Fisons SLR grade, 210 ml).

Hydrogen chloride gas was bubbled into a cooled flask containing propan-2-ol (157 g) until 20.8 g hydrogen chloride had been absorbed. 39 g of this solution (containing approximately 4.6 g hydrogen chloride) was added rapidly to the paroxetine solution and the mixture stirred briskly. After about 1 minute crystallisation began and the mixture S' apidly set to an unstirrable paste, which was allowed to stand for one hour. The product *as collected by filtration, washed with propan-2-ol (50 ml) and dried under vacuum at ambient temperature to constant weight in a desiccator containing phosphoric oxide. The 2Q. ample was analysed by NMR spectrometry and found to contain approximately 6% ::'propan-2-ol by weight. Part of the sample was placed in a vacuum oven set at 50 0 C and fu" rther dried to constant weight, which took a further 4 days. NMR spectrometry showed that the sample contained approximately 2% propan-2-ol by weight XAMPLE 3 "'aroxetine Hydrochloride Propan-2-ol Solvate k. aroxetine free base (52.37 g) was dissolved in dry propan-2-ol (250 ml) and a solution .af hydrogen chloride gas in dry propan-2-ol (50 g of solution, containing approximately 5.8 g hydrogen chloride) was added rapidly with brisk stirring. After about 30 seconds Scrystallisation commenced, and the mixture was stirred for a further 30 minutes at :mbient temperature to permit complete crystallisation. The product was isolated by S'pcuum filtration, washed with 25 ml dry propan-2-ol, and dried in a desiccator containing phosphoric oxide at ambient temperature under vacuum.

After 3 days a sample was analysed by NMR and found to contain 10.5% propan-2-ol.

The rest of the material was dried for a further 3 days to constant weight under vacuum with fresh phosphoric oxide in the desiccator. NMR analysis showed that the product contained 5.7% w/w propan-2-ol.

-11- °S P31122 EXAMPLE 4 Crystalline Paroxetine Hydrochloride Anhydrate Substantially Free of Bound Pyridine (Formn A) i) Preparation of paroxetine hydrochloride pyridine solvate Paroxetine hydrochloride containing ca. 2% propan-2-ol 120.Og] was dissolved in hot pynidine [200 ml] and some of the solvent removed by distillation. The flask was scaled and allowed to cool, whereupon the pale red solution spontaneously crystallised. Ile thick suspension was stirred well, filtered, avoiding excessive exposure to atmospheric moisture, and the solid washed on the filter with pyridine [25 ml]. The product was dried under high vacuum over phosphorus pentoxide.

Yield 22.0 g.

Sb irscopic examination showed the product to be in the form of needle crystals, and ':i*aysis by NMR showed the presence of 15.2% by weight of pyridine (theory for a 1:1 -:'-Alvate 17.77%). The infra-red spectrum (Nujol mull) of the pyridine solvate differed 2Q lom, both that of hemihydrate and anhydrate Form A and in particular showed no :.*sdgnificant bands in the 3000 cm-i region. The pyridine solvate also gave a distinctive X -ray powder diffraction pattern.

ii) Preparation of paroxetine hydrochloride anhydrate (Form A) :.:.*aoxetine hydrochloride pyridine solvate [5.00 g] was added to 5 molar hydrochloric icid [25 ml] in a beaker and stirred for 5 minutes. The mixture was filtered, drained well, andvwashed with water [15 ml]. The crystals were dried under high vacuum over ~bshorus pentoxide.

iield 4.00 g ::TIhe infra-red spectrum (Nujol mull) was consistent with paroxetine hydrochloride .%hydrate Form A, and no pyridine could be detected by NMR analysis.

EXAMPLE S Paroxetine Hydrochloride Anhydrate Substantially Free of Bound Acetic Acid (Form A) i) Preparationo paroxetine hydrochloride acetic acid solvate Paroxetine hydrochloride containing approximately propan-2-ol [30.0g], was _dissole in hot glacial acetic acid [120 nil] and some of the solvent removed by distillation. The flask was sealed and allowed to cool overnight. The clear pale yellow solution was seeded with paroxetine hydrochloride anliydrate Form A, ultrasonicated and -12- .1 P31122 stired at room temperature for several hours. The mixture was allowed to stand for 24 hours, filtered and the product dried under high vacuum in a desiccator containing potassium hydroxide.

Yield 17.29 g.

Analysis by NMR showed the presence of 13.5% by weight of acetic acid (theory for a 1:1 solvate 14.10%). The infra-red spectrum (Nujol mull) of the acetic acid solvate differed from both that of paroxetine hydrochloride hemihydrate and anhydrate Form A and in particular showed a strong band at 1705 cm-1 indicative of bound acetic acid and no significant bands in the 3000 cm-1 region. The acetic acid solvate also gave a distinctive X-ray powder diffraction pattern.

ii) Preparation of paroxetine hydrochloride anhydrate (Form A) Paroxetine hydrochloride acetic acid solvate [1.00 g] was treated with 5 molar hydrochloric acid [5 ml] and stirred for 5 minutes. The mixture was filtered, drained well, and the crystals dried under high vacuum in a desiccator containing phosphorus :."pentoxide. 1 yield 0.S0 g .the infra-red spectrum (Nujol mull) confirmed that the product was paroxetine :hydrochloride anhydrate Form A. Analysis by NMR showed the presence of approximately 0.4% acetic acid. Microscopic examination showed the material to be in the form of fragmented needles.

.XAMPLE 6 Paroxetine Hydrochloride Anhydrate Substantially Free of Bound Acetonitrile SForm A) Preparation of paroxetine hydrochloride acetonitrile solvate Paroxetine hydrochloride anhydrate Form A prepared using the method of Example 1 0.

8 g) was dissolved in warm anhydrous acetonitrile (40 ml) in a conical flask, sealed, and cooled in the refrigerator for I hour, during which time some crystals separated. The mixture was ultrasonicated, returned to the refrigerator, and left overnight. The contents set to a thick paste. The following morning the paste was broken up using vigorous shaking and ultrasonication, and the mixture filtered. The product was dried under high vacuum in a desiccator containing phosphors pentoxide.

Yield 9.30 g, acetonitrile content 2.5% (by NMR).

ii) reparation of paroxetine hydrochloride anhydrate (Form A) -13- P31122 -Paroxetine hydrochloride acetonitrile solvate (4.23 g) was stirred in water (20.6 g) for minutes. The solid was collected by vacuum filtration, washed on the filter with water ml) and dried in a vacuum oven containing phosphorus pentoxide at 50 0

C.

Yield 3.75 g The IR spectrum showed that the product was paroxetine hydrochloride anhydrate. Form

A.

Acetonitrile content approximately 0.5% (by NMR).

EXAMPLE 7 Paroxetine Hydrochloride Anhydrate (Form B) Paroxetine free base (10-0 g] was dissolved in butan-l-ol [25 ml]1 at room temperature and a solution of hydrogen chloride gas [1.25 g] in butan-l-ol [15 ml] was added. The 'clear pale red/brown solution was sealed and stored in the refrigerator overnight. A 2Q s=Al amount of crystalline material formed on the base of the flask and ulurasonication .was used to bring about crystallisation of the bulk. The mixture was again stored in the refrigerator overnight, then allowed to warm to room temperature and filtered. The lmxuct was dried under high vacuum in a desiccator containing phosphorus pentoxide.

Microscopic examination with a polarising microscope showed the sample to be in the form of feather shaped crystals.

Melting point: 137-138-C The NIMR (CDCl 3 spectrum conformed to that of a standard sample of paroxetine hydrochloride.

t elemen tal analysis was consistent with anhydrous paroxetine hydrochloride: for C 19

H

21 NC1FO 3 C 62.38 H 5.79 N 3.83 Ci 9.69% found: C 62.08 H 5.75 N 3.81 Ci 9.62% The:X-ray powder diffactograxn confirmed that the sample was crystalline (see Figure The diffractogram differed from both that of hemnihydrate and anhydrate Form A.

The IR spectrum (Nujol mull) also differed from both that of hemnihydrate and anhydrate Form A, (see Figure;2).

The DSC cxotheMn measured at 10*C per minute, showed a maximum at about 137TC in both open and closed pans.

-14- 7-7- P31122 The sample was also examined by solid state C13 NMR (see Figure 8).

EXAMPLE 8 Paroxetine Hydrochloride Anhydrate (Form C) Paroxetine hydrochloride hemihydrate [300 g] and toluene [1200 ml] were heated under reflux and the water removed using a Dean and Stark apparatus. When no further water could be collected, the bulk of the toluene was removed by distillation and replaced with anhydrous butan-1-ol. Distillation was continued until the still temperature reached about 1170C, indicating that all the toluene had been removed. The mixture was diluted to about 1200 ml with butan-l-ol and allowed to cool. At about 42 0 C, seeds of paroxetine hydrochloride anhydrate Form B (needles) were added. Although crystallisation then began, it was observed that the product was in the form of well formed prisms, indicating that the product was crystallising in a different form to the seeds added.

The mixire was allowed to stand overnight, then filtered. The crystals were washed on the filter with butan-1-ol, then dried in vacuum at 50 0 C over phosphorus pentoxide.

Yield 250 g.

Melting point: 162-164C Analysis by NMR (CDC13) confirmed that the product was paroxetine hydrochloride and showed the presence of a trace of butan-1-ol (ca 0.1% by weight). The infra red spectrum (Nujol mull) differed from either Form A or B, (see Figure 3).

'Water content 0.06% (KF) The elemental analysis was consistent with anhydrous paroxetine hydrochloride: Required for C 19

H

2 1 NC1FO: C 62.38 H 5.79 N3.83 Cl 9.69% Found: C 62.23 H 5.67 N 3.83 Cl 9.74% DSC exotherm, measured at 10C per minute, showed a maximum at about 161°C in both open and closed pans.

The X-ray powder diffractogram confirmed that the sample was crystalline (see Figure The diffracogram differed from both that of anhydrate Form A and anhydrate Form

B.

The sample was also examined by solid state C13 NMR (see Figure 9).

-15- Y -15- P31122 EXAMPLE 9 Paroxetine Hydrochloride Anhydrate Substantially Free of Bound Acetone (Form

A)

i) Paroxetine Hydrochloride Acetone solvate Paroxetine free base (10.51 g) was dissolved in acetone (40 ml, dried with 4A molecular sieves), and a solution of hydrogen chloride gas (1.31 g) in dry acetone (10 ml) added with stirring. Crystallisation occurred spontaneously within one minute, and the mixture quickly became unstirrable. After approximately half an hour the product was filtered, placed in a desiccator over phosphorus pentoxide, and dried at ambient temperature overnight Weight of product: 11.24 g. Acetone content (estimated by NMR) 4% wt/wt. The infrai "ed spectrum showed a characteristic band at 667 cm1.

pproximately half the product was placed in a vacuum oven set at 50 0 C and dried ther to constant weight. NMR analysis of the resulting product indicated the presence .:of 1.2% acetone wt/wt.

ii) Paroxetine Hydrochloride Anhydrate (Form A) A sample of the acetone solvate (5.18 g) was stirred for 10 minutes in water (20 ml), S filtered, and dried at 50°C in a vacuum oven containing phosphorus pentoxide.

Weight of product: 4.63 g. NMR analysis indicated the presence of 0.6% acetone wt/wt.

The infra-red spectrum corresponded to the spectrum of paroxetine hydrochloride anhydrate Form A and showed a characteristic band at 665 cm 1 EXAMPLE Paroxetine Hydrochloride Anhydrate Substantially Free of Bound Ethanol (Form

A)

i) Paroxetine Hydrochloride Ethanol Solvate Paroxetine free base (11.25 g) was dissolved in absolute ethanol (40 ml), and a solution of hydrogen chloride gas (1.9 g) dissolved in absolute ethanol (20 ml) added with stirring. There was no sign of crystallisation after 10 minutes, so the clear solution was seeded with paroxetine hydrochloride anhydrate Form A. After 30 minutes there was still no sign of crystallisation, so the solution was evaporated at reduced pressure to approximately half volume and re-seeded. This time slow crystallisation was observed, 16- 1 P31122 and the mixture was left for a further hour. The esulting crystalline mass was dried at ambient temperature in a vacuum desiccator containing phosphorus pentoxide.

Weight of product: 11.87 g. Ethanol content (estimated by NMR) 4% wt/wt. The infrared spectrum showed a characteristic band at 667 cm- 1 A small sample was placed in a vacuum oven set at 50 0 C and dried further. NMR analysis of the resulting product indicated the presence of 0.7% ethanol wt/wt. The infra-red spectrum corresponded to the spectrum of paroxetine hydrochloride anhydrate Form A and showed a characteristic band at 665 cm- 1 ii) Paroxetine Hydrochloride Anhydrate (Form A) A sample of the ethanol solvate (5.3 g) was stirred for 10 minutes in water (20 ml), filtered, and dried overnight at ambient temperature in a desiccator containing phosphorus pentoxide.

Weight of product 4.56 g. NMR analysis indicated the presence of less than 0.4% t"hanol wtIwt The infra-red spectrum corresponded to the spectrum of paroxetne 2Q...hydrochloride anhydrate Form A and showed a characteristic band at 665 cm. .EXAMPLE 11 Paroxetine Hydrochloride Anhydrate Substantially Free of Bound Chloroform (Form A) i) Paroxetine hydrochloride chloroform solvate Paroxetine free base (8.54 g) was dissolved in chloroform (30 ml), and a solution of hydrogen chloride gas (1.05 g) dissolved in chloroform (10 ml) added with stirring.

i There was no sign of crystallisation after 5 minutes, so the clear solution was seeded with '..*paroxetine hydrochloride anhydrate Form A. After 15 minutes there was still no sign of rystallisation, so hydrogen chloride gas was bubbled through the solution until the orange colour disappeared. After one hour signs of very slow crystallisation could be seen, with large needle crystals visible to the eye. The mixture was left to crystallise in a S. Stoppered flask for a further hour, then filtered and dried at ambient temperature in a S acuum desiccator containing phosphorus pentoxide.

Weight of product: 5.65 g. Chloroform content (estimated by NMR) 12.5% wt/wt. The infa-red spectrum showed a characteristic band at 667 cm- A small sample was placed in a vacuum oven set at 50°C and dried further. NMR analysis of the resulting product indicated the presence of 3.4% chloroform wt/wt.

ii) Paroxetine hydrochloride anhydrate (Form A) -17- P31122 A sample of the chloroform solvate containing 12.5% chloroform (2.0 g) was stirred for minutes in water (8 ml), filtered, and dried overnight in a vacuum oven at Weight of product: 1.09 g. NMR analysis indicated the presence of approximately 0.8% chloroform wt/wt The infra-red spectrum corresponded to the spectrum of paroxedne hydrochloride anhydrate Form A and showed a characteristic band at 665 cm1.

EXAMPLE 12 Paroxetine Hydrochloride Anhydrate (Form C) Paroxetine free base (8.5 g) was dissolved in ethyl acetate (40 ml) and hydrogen chloride .gas was bubbled in until the weight of the flask and contents had increased by 1.1 g.

There was no sign of crystallisation after 15 minutes, so the clear solution was seeded with paroxetine hydrochloride anhydrate Form A. After stirring for a further one hour, signs of very slow crystallisation could be seen. The mixture was left stirring overnight to crystallise in a stoppered flask, then filtered and dried at ambient temperature in a vacuum desiccator containing phosphorus pentoxide.

Weight of product: 7.56 g. Ethyl acetate content (estimated by NMR) 0.4% wt/wt. The infra-red spectrum was different from both paroxetine hydrochloride hemihydrate and anhydrate Form A and consistent with the infra-red spectrum obtained in Example 8.

.EXAMPLE 13 2:7 Paroxetine Hydrochloride Anhydrate Substantially Free of Bound Propan-1-ol S (Form A) i) Preparation of paroxetine hydrochloride propan-1-ol solvate 36 Paroxetine free base [10.6 g] was dissolved in propan-1-ol [30 ml] and hydrogen chloride gas (1.25 g) passed into the solution. The warm solution was seeded with paroxetine S hydrochloride anhydrate Form B and ultrasonicated, whereupon the pale red solution rapidly crystallised. The thick suspension was diluted with propan-1-ol (25 ml), filtered avoiding excessive exposure to atmospheric moisture, and the product dried in vacuum over phosphorus pentoxide.

:'"Yield 10.3 g.

Analysis by NMR showed the presence of approximately 7% by weight of propan--ol.

The infra-red spectrum (Nujol mull) showed that the product was not Form B, but a solvated species with a significant band at about 667 cm-l. Thepropan-l-ol solvate also gave a distinctive X-ray powder diffraction pattern.

ii) Preparation of paroxetine hydrochloride anhydrate (Form A) -18- i P31 117 Paroxetine hydrochloride propan-l-ol solvate [5.24 g] was stirred in water [25 ml] for minutes. The mixture was filtered and the product washed with water [10 The crystals were dried in high vacuum over phosphorus pentoxide at 50*C2 Yield 4.35 g The infra-red spectrum (Nujol mull) confwred that the product was the anhydrate Form Analysis by INM showed the presence of ca 0.25% by weight of propan-1-oI.

EXAMPLE 1L4 Paroxetine Hydrochloride Anhydrate (Form D) i) Preparation of paroxetine hydrochloride toluene solvate Paroxetine hydrochloride hemihydrate 100 g) was stirred under ieflur, in toluene 1 1000 ml] and the water removed using a Dean and Stark apparatus- The solution was allowed to cool, seeded 'with paruxetine hydrochloride Form A, and ultrasonicated. Crystallisation was not induced, but after stirig for 40 minutes at room temperature the contents of the flask suddenly set to a thick .paste. The product was collected by filtration and dried in vacuum over phosphorus pentoxide.

knalysis of the product by NMR showed the presence of abo ut 10% wtlWt Oftoluene. The a toluene solvate gave a distinctive IR spectrum, showing a characteristic band at 672 cra 1 Thce abovre procedure was repeated, seeding with the toluene solvate, and the product dricd in vacuum-over phosphorus pentoxide- .Yield of toluene solvate 106.7 g Analysis of the product by NMR showed dhe presence of about 10% wxiwt of toluene- The product gave a distinctive X-ray powder diffractogran.

T-ii) Desolvation of the toluene solvate .ThCIb toluene solvate [20.0 g] was heated for 18 hours at 80 0 C in vacuum, over phosphorus entoide. Analysis by- NNR showed the presnce of about 03% wt/wt of toluene.

Watercontent: 0.08% (KF) Meltingpoint: ca125 0

C

P31122 EXAMPLE Paroxetine Hydrochloride Anthydrate Substantially Free of Bound Tetrahydrofuran (Form A) i) Paroxetine hydrochloride tetrahydrofuran solvate.

Paroxetine free base (10.26 g) was dissolved in dry tetrahydrofuran (35 nml), and a solution off hydrogen chloride gas (1.3 g) dissolved in dry tenahydrofuran (15 mil) added with brisk stirring. After a short period when the solution remained clear, rapid crystallisation commerced so that within a few minutes the mixture became unstirrable.

After a further half hour, the product was collected by filtration and dried at ambient temperature in a vacuum desiccator containing phosphorus pentoxide.

Weight of product: 12.31 g. Tetrahydrofuran content (estimated by NMR) 11.4% wt/wt.

The infra-red spectrum showed a characteristic solvate band at 667 cmi.k A small sample was placed in a vacuum oven set at 50CC and dried over the weekend.

NMR analysis of the resulting product indicated the presence of 1.3% tetrahydrofuran Parcoxetne, hydrochloride anhydrate (Form A) A sample of the tetrahydrofuran solvate containing 11.4% tetrahydrofuran (5.0Og) was stirred for 10 minutes in water (20 ml), filtered, and dried in a vacuum oven at 501C *::Weight of product: 3.79 g. MR analysis indicated the presence 6f approximately 0.02% :'letrahydrofuran wt/wL The infra-red spectrum corresponded to the spectrum of paroxetine hydrochloride anhydrate Form A and showed a characteristic band at 665 cm- 1 3 iXAMPLE 16 Parozetine Hydrochloride Anhydrate Substantially Free oflBound Propan-2-ol (Form A) Paroetine. hydrochloride propan-2-ol solvate (70 mg, containing 11.6% propan-2-ol) (Examples 1 or 3) was treated with a stream of carbon dioxide (3 mniminute, 55 0 C and 2,500 psi). After minutes the propan-2-ol content was reduced to and after a total of 120 minutes it was further reduced to The temperature: was then raised to 75* 3 ad after 30 minutes the prv~pan-2-o1 content was found to be 0.13%. After a further 60 minutes at 75 0 C the propan-2-ol content was 0.07%.

hr a separate experiment 70 mg of propan-2-ol solvate was extracted with carbon dioxide (3 mlfiutc, 75 0 C and 2,500 psi). After 150 minutes the propan-2-ol content was found to be 0.19%.

P311"2 This experiment was repeated on a larger sample of the solvate (350 mg) under the same conitions, and the propan-2-ol content was found to be 0.16% after 150 minutes.

EXAMPLE 17 Crystallisation or paroxetine hydrochloride anhydrate Form C from 2-butanone by seeding Paroxetmne hydrochloride anhydrate Form C [7.0 g] was heated to boiling in anhydrous 2butanone [40 mlu and the solution allowed to cool to ca 40"C. Seeds of Form C were added and the stirred mixture allowed to cool to room temperature. The product was collected by filtration, washed with anhydrous 2-bucanone [20 rol) and dried in an oven at 100'C Weight of dried product 5.95 a Melting point 162 -163 0

C

*iThe infra-red spectrum (Nujol mull) was consistent with paroxedne hydrochloride anhydrate 4orC.

29%.

tXAM'PLE 18 rystallisation of paroxetine hydrochloride from toluene by seeding Paroxetine hydrochloride anhydrate Form C [20.0 g] was dissolved in boiling toluene (200 m!] :nd approximately 50 ml of the solution added to each of 4 conical flasks. Each flask was .::heated again to boiling, allowing some toluene vapour to reflux out, in order to remove seeds.

Flask 1 was immediately sealed with a ground glas stopper and set aside to cool. The remaining flasks were sealed with foil, and allowed to cool somewhat, before adding seed crystals as Flask 2 was seeded with paroxetirie hydrochloride toluene solvate Flask 3 was seeded with paroxetine hydrochloride anhydrate Form B Flask 4 was seeded with paroxetine hydrochloride anhydrate Form C The added seeds remained undissolved. The flasks were sealed with ground glass; stoppers, stirred gently for a few seconds then set aside to cool-- Flask 2 was observed to crystallise very readiy, while in Flasks 3 an d 4 crystallisation took place more slowly. At this point Flask 1 remained completely clear, and all 4 flasks were left at room temperatur overnight. The following morning Flask 1 contained only a few crystals, while Flasks 2, 3 and 4 had extensively -21- P31 127 Flask I 'was stirred gently for several hours, during which dim the bdlk of the paroxedine hydrochloride crystallised.

The product from each flask was collected by filtration and dried at 50"C under vacuun.

Flask 1 (not seeded) Weight of product: 4_25-g Appearance: short needles/rods Infra red spectrum: consistent with paroxetine hydrochloride anhydrate Form C Melting point 161 1620C Flask 2 (seeded with toluene solvate) Weight of product: 3.80 g 2a*. Appearance: Ion- fine needles Infrared spectrum: consistent with paroxetine hydrochloride toluene soilvate, Solvent content: 11I% AVt/wt oluene by NNMR Melting point: initial melt at about 70 0 C, followed by resolidification and further mel at 161 -162 0

C

kask 3 (seeded with anhydrate Form B) Weight of product. 4.20 g Appearance: needles hit e pcrm consistent with paroxetine hydrochloride anhydratc Form B Solvent conent 0.8% wt/wt toluene by INMR Melting point 138 1400C2 P31122 Flask 4 (seeded with anhydrate Form Weight of product 4-93 g Appearance: needles Infra red spectrum: con sistent with paroxedxne hydrochleride anhydrate Form C Solvent content: 0-8% wtlwt toluene by NrM.R Melting point: 161 162 0

C

Example 19 -Crystaliine Paroxetine Hydrochloride Anhydrate Substantially free of Bound Propan-2-ol .:Vacuum oven dried paroxetine hydrochloride propan-2-0l solvate containing 2_6% propan-2-oI g) was placed in a glass tube. The tube was immersed in a water bath set at 501C and tiitrogen gas, saturated with water vapour at a temperature of 40'C, was passed through the sample. After 10 hours a small sample was removed an analysed by NMR. which showed that the level of propan-2--ol had fallen to The temperature of the bath surrounding the tube :wa increased to 80'C and the tem raturc at which the gas being passed through the sample -':was saturated was increased to 70 0 C. After 10 hours the contents of the tube were sampled again and analysed by NMwhich showed that the level of propan-2-.ol had fallen further to -Paroxetine Hydrochlaride Anhydrate Substantially Free of Bound Acetone (Form A) i)preparation of p toeine hydrochloride acetone solvate A suspension of paroxetint hydrochloride anhydrate Form C (prisms) (5.0 g) in -acetone (75 ml) vim heated tD boiling to give a mass of fine needles. Ile flask was sealed and allowed to stand overnighit at room temperature. The solvent wz~ removed at low temperature using a rotary evaporator and replaced by hexane (100 ml). The solvent was again removed at low temperan=r to give the acetone solvate as a crystalline residue- Analysis by NNM showed the presence of acetone (12.2% by, weight). and the IR spectrum (Nujol mull) showed characteristic bands at 667 and 1714cm-l- P31122 ii) Preparation of paroxetine hydrochloride anhydrate (Form A) fromn the acetone solvate Paroxetine hydrochloride Form C (5.3 g) was converted to the .etone solvate by a similar procedure to that described above. Water (50 ml) was added, and the resulting suspension shaken gently for 10 minutes. The white solid was collected by filtraton, drained thoroughly and dried in a vacuum oven at 50 Yield 4.60 g. Acetone ccmtent (NMR) 0. 1% by weight.

The JR spectrum (Nujol mull) conformed to that of a standard sample of paroxetine hydrochloride anhydrate Form A.

Example 21 Paroxetine Hydrochloride Anhydrate Form D i) preparation of paroxc.tine hydrochloride tolucne solvatc.

An anhydrous SWlutidn of paroxetine hydrochloride in toluene was prepared by refluxing a :mixture of paroxetine hydrochloride hemihydrate in toluene in a Dean and Stark apparatus until :omore watier could be collected.: The solution was allowed to cool and seeded with paroxetine 26 *:hydrochloride toluene solvate. The product was collected by filtration, washed with toluene and dried vacuum oven at 50 C. Analysis by NMR showed the presence of 18% by weight of ,:toluene. The infra-red spectrum, reccrded at 22 IC using a Perkin-Elmer 172OX FT-JR spectrometer coupled to a Spectra-Tech IR-Plan microscope, is shown in Figures 1OA and lOB.

ii) preparation of paroxetine hydrochloride anhydrate Form D.

sthall sample of paroxetine hydrochloride toluene solvate (toluene content 18% wt/wt) was -heated at 80 *C and the paroxetine hydrochloride anhydrate Form D produced examined by -infra-red microspectrometry using a Perkidn-Elmer 172OX FT-JR spectrometer coupled to~SetaTc RPa irsoe h eutn nr-e pcnmi hw nFgrs1l an I IBI 4 -24-

Claims (10)

1. Paroxetmne hydrochloride anhydrate substantially free of bound propan-2-ol.

2. Paroxetine hydrochloride anhydrate substantially free of bound organic solvent.

3. Paroxetine hydrochloride anhydrate substantially free of bound propan-2-ol with the proviso that it is other than Form Z.

4. Paroxetine hydrochloride solvates, other than the propan-2-ol solvate. Paroxetine hydrochloride solvates as defined in claim 4 wherein the solvate is selected from the group consisting of alcohols (other than propan-2-ol), o rganic acids, organic bases, nitriles, ketone, ethers, chlorinated hydrocarbons and hydrocarbons.

C C c '4

6. Paroxetine hydrochloride solvate as defined in claim 4 wherein the solvate is selected from the group consisting of propan-l-ol, etnol, acetic acid, pyridine, acetonitrile, acetone, tetrahydrofliran, chloroform and toluene.~

7. Paroxetine hydrochloride anhydrate substantially free of bound propan-2-ol in substantially pure -form. %3P

8. Acompound asdefined in anyoof claimfs -1to 3.which is selected from the groulp consisting. of; crystalline paroxetin hydrochloride anhydrate substantially free of bound pyridine (Form paroxetine hydrochloride anhydrate substantially free of bound acetic acid (Form paroxetine hydrochloride anhydrate substantially free of bound acetonitrile (Form piroxetine hydrochloride anhydrate (Form paroxetine hydrochloride anhydrate (Form C).paroxetine hydrochloride anhydrate substantially free of bound acetone (Form A), pafoxetilD hydrochlrieanhydrate substantially free of bound ethanol (Form paroxetine hydrochloride anhy'rt substantially. free of bound chloroform (Form paroxetine hydrochloride anhydrate (Form paroxetine hydrochloride anhydrate substantially free of p\O MO«33 2-91V.I5I-'12Jl 26 S bound propan--ol (Form paroxetine hydrochloride anhydrate (Form and paroxetine hydrochloride anhydrate substantially free of bound tetrahydrofuran(Form A).

9. A process for the preparation of paroxetine hydrochloride anhydrate substantially free of propan-2-ol which comprises crystallising paroxetine hydrochloride in either; i) an organic solvent or mixture of organic solvents which form a solvate with the paroxetine hydrochloride and which are not removable by conventional drying techniques; or ii) an organic solvent or mixture or organic solvents which do or do not form a solvate with the paroxetine hydrochloride and which are removable by conventional vacuum oven drying; I thereafter in the case of i) displacing the solvated solvent or solvents using a displacing agent and in the case of ii) by removing the solvent.

10. A process for the preparation of the paroxetine hydrochloride solvates other than the propan-2-ol solvate which comprises crystallising paroxetine hydrochloride in an organic solvent or mixture of solvents which form a solvate with the paroxetine hydrochloride and which are not removable by conventional drying techniques. 20 DATED this fifteenth day of December 1998 SmithKline Beechln~ m pc. By its Patent Attorneys DAVIES COLISON CAVE i