Crystal Engineering molekularer Festkörper

C ys al Enginee ing molekula e Fes kö pe Disse a ion zu E langung des akademischen G ades Dok o de Na u wissenscha en (D . e . na .) im P omo ionsp og amm Ma e ialchemie und Ka alyse an de Bay eu he G aduie enschule ü Ma hema ik und Na u wissenscha en (BayNAT) de Uni e si ä Bay eu h o geleg on Ch is ian Bu e ho gebo en in Bambe g Bay eu h 2013 ii i Die o liegende A bei wu de in de Zei on Augus 2009 bis Augus 2013 am Leh s uhl ü Ano ganische Chemie I (ACI) an de Uni e si ä Bay eu h un e Be- euung on He n P o . D . Jose B eu ange e ig . Volls ändige Abd uck de on de Bay eu he G aduie enschule ü Ma hema ik und Na u wissenscha en (BayNAT) de Uni e si ä Bay eu h genehmig en Disse - a ion zu E langung des akademischen G ades eines Dok o s de Na u wissenscha - en (D . e . na .). Disse a ion einge eich am: 24.09.2013 Zulassung du ch die P ü ungskommission: 26.09.2013 Wissenscha liches Kolloquium: 13.02.2014 Am ie ende Di ek o : P o . D . F anz X. Schmid P ü ungsausschuss: P o . D . Jose B eu (E s gu ach e ) P o . D . Jü gen Senke (Zwei gu ach e ) P o . D . Ma hias B euning (Vo si z) P o . D . Geo g Papas a ou ii iii Meine Familie Die g öß en E eignisse - das sind nich uns e lau es en, sonde n uns e s ills en S unden F ied ich Nie zsche (1844-1900) i Inhal s e zeichnis Inhal s e zeichnis Abkü zungs e zeichnis .............................................................................................. ii Zusammen assung ........................................................................................................ 1 Summa y ...................................................................................................................... 3 1. Einlei ung ............................................................................................................. 5 1.1. An o de ungen an A znei o men ...................................................................... 7 1.2. Wi ks o e in es e Fo m .................................................................................. 8 1.2.1. Beg i e und De ini ionen .......................................................................... 8 1.2.2. Wich ige Eigenscha en on Wi ks o en ................................................ 10 1.2.3. Pa en ech liche Aspek e .......................................................................... 13 1.3. Theo ie und P axis de K is allisa ion – The modynamik .............................. 15 1.3.1. U sache de Polymo phie ......................................................................... 15 1.3.2. The modynamik de Polymo phie ........................................................... 16 1.3.3. Klassische Keimbildungs heo ie .............................................................. 18 1.3.4. P ak ische K is allisa ion ......................................................................... 21 1.4. Un e such es Modellsys em ............................................................................ 24 1.4.1 Benzamid................................................................................................... 24 1.4.2 Benzoesäu e – Benzoa salze ..................................................................... 25 2. Synopsis ............................................................................................................. 27 2.1. Mo i a ion ....................................................................................................... 27 2.2. Ein luss de S öchiome ie bei de Co-K is allisa ion .................................... 28 2.3. Ein luss de Übe sä igung bei de Co-K is allisa ion .................................... 30 2.4. Ein luss de Ka ioneng öße bei de Co-K is allisa ion ................................... 31 2.5. Eink is alls uk u en on Na iumbenzoa und Kaliumbenzoa ...................... 33 2.6. Ein luss molekula e De ek e au die Phasenumwandlung ............................ 34 3. Publika ionen und Manusk ip e ......................................................................... 37 4. Wei e e Publika ionen ........................................................................................ 41 Inhal s e zeichnis i 5. Vo äge und Pos e bei äge ............................................................................... 43 6. Li e a u e zeichnis ............................................................................................ 45 Danksagung ................................................................................................................ 55 Anhang ....................................................................................................................... 57 Anhang A.1 ............................................................................................................ 57 Anhang A.2 ............................................................................................................ 65 Anhang A.3 ............................................................................................................ 73 Anhang A.4 ............................................................................................................ 83 Anhang A.5 ............................................................................................................ 97 E klä ung des Ve asse s .......................................................................................... 115 Abkü zungs e zeichnis ii Abkü zungs e zeichnis ADT Au oma ed elec on Di ac ion Tomog aphy ATR A enua ed To al Re lec ance c Konzen a ion DESY Deu sches Elek onen-Synch o on DFT-D Dich e unk ional heo ie mi semiempi ische Dispe sionswechselwi - kungsko ek u DSC Dynamische Di e enzkalo ime ie EMA Eu opean Medicines Agency FBRM Focused Beam Re lec ance Measu emen FDA Food and D ug Adminis a ion FTIR Fou ie T ans o m In a ed G x F eie En halpie on Fo m x HBz Benzoesäu e HIV Humanes Immunde izienz-Vi us IR In a o i in a enös KBz Kaliumbenzoa KZ Koo dina ionszahl LiBz Li hiumbenzoa M Mola MD Molekula dynamik MSZW Me as able Zone Wid h NaBz Na iumbenzoa NMR Nuclea Magne ic Resonance pK a Säu ekons an e PVM Pa icle Vision and Measu emen Einlei ung 6 kann es, wie e wähn hinsich lich de Bio e ügba kei zu Einsch änkungen kom- men. O enkundig gib es eine hohe Anzahl an Me hoden, die gewünsch en Eigen- scha en eines A zneimi els zu e eichen. Möglichkei en zu Op imie ung s ellen zum Beispiel de Einsa z und die Zusammense zung on Hil ss o en da , on denen es eine Vielzahl gib .11 Au diese Me hoden soll in diese A bei nich genaue einge- gangen we den. Es exis ie jedoch aus üh liche Li e a u zum Thema, au die e - wiesen sei.12-15 Diese A bei se z sich ausschließlich mi Me hoden, die den ak i en Wi ks o selbs be e en, auseinande . So gib es einige S a egien aus dem Be eich des „C ys al Enginee ing“, die in de Lage sind, e en uelle Nach eile hinsich lich Bio e ügba kei , e c. zu kompensie en und auch aus pa en ech liche Sich ü die pha ma- zeu ische Indus ie in e essan und wich ig sind: Polymo ph-Sc eening, Co-K is all-Sc eening, Salz-Sc eening, Amo phisie ung.16-18 Die En - wicklung eines A zneimi els is kos en- und zei in ensi . So be äg die En wicklungszei eines neuen A zneimi els (neue Wi ks o ) deu lich übe 10 Jah e und kos e e wa 500 Mil- lionen bis 2 Millia den USD.19 Die En wicklung gliede sich dabei in die eigen liche Fo schung nach dem Wi ks o und die En wicklung und Fo mulie ung des P oduk s ( gl. Abb. 1.1). Die P oduk en wicklung un e eil sich in die p äklinische und die klinische Phase. Ganz zu Beginn de Fo schung s eh dabei die Un e suchung on seh ielen (> 10000) Molekülen, hin au die gewünsch e Wi ksamkei .20 In de P o- duk en wicklungsphase, in de nu noch wenige geeigne e Moleküle wei e un e - such we den, olg bei es en Wi ks o en das „C ys al Enginee ing“: Salzauswahl, Polymo ph-Sc eening und die En wicklung de K is allisa ion.8 Die Au gabens ellung diese A bei um ass e die sys ema ische Un e suchung de K is allisa ion und Co-K is allisa ion an einem Modellsys em ein ache o ganische Subs anzen (HBz, de en Benzoa salze und Benzamid), die eilweise Anwendung als Konse ie ungsmi el inden. Abb. 1.1 Schema ische Übe sich übe die Daue de En wicklung eines A zneimi els sowie die einzelnen Phasen. In jede Phase is de Fokus dabei au ande e K i e ien ge ich e . Einlei ung 7 1.1. An o de ungen an A znei o men Folgende wich ige An o de ungen an ( es e) A znei o men (Table e, Kapsel, Salbe, In usion) we den ges ell :9 o S abili ä o Bio e ügba kei o Regula o ische Aspek e o Wi scha liche Aspek e o Technische Aspek e o Ma ke ingaspek e Abgesehen on de chemischen S abili ä des Wi ks o es is hie o allem auch die physikalische S abili ä , insbesonde e he modynamische S abili ä , de eingese z en Modi ika ion on Bedeu ung. S abili ä und Bio e ügba kei sind deswegen wech- selsei ig oneinande abhängig. Wie im nächs en Kapi el gezeig wi d, sind he mo- dynamisch s abile Modi ika ionen zwa o une wünsch en Phasenumwandlungen siche , jedoch können sie mögliche weise auch eine signi ikan ge inge e Bio e üg- ba kei au weisen. Ein p ominen es Beispiel hie ü s ell de HIV-P o easeinhibi o Ri ona i (Abbo ) da .21 So kam es bei de P oduk ion 1998 öllig une wa e zu Bildung eine neuen, s abile en Fo m, die eine zu ge inge Löslichkei au wies um die e o de liche Bio e ügba kei zu gewäh leis en.18 Regula o ische Aspek e be assen sich o allem mi de Zulassung des jeweiligen A zneimi els du ch die jeweils zu- s ändigen Zulassungsbehö den (z.B.: Eu opean Medicines Agency (EMA) ü Eu- opa ode die Food and D ug Adminis a ion (FDA) ü die USA). E wähnenswe sind hie bei insbesonde e auch un e schiedliche polymo phe Fo men on Wi ks o - en. Diese weisen zum Einen e schiedene physikalisch-chemische Eigenscha en au , zum Ande en können sie als eigens ändige Fo m pa en ie und somi auch e - ma k e we den.22 De Punk de echnischen Machba kei ziel auch au die mecha- nischen Eigenscha en des P oduk es und die Robus hei in de P oduk ionsphase ab. Hie bei spiel die eingese z e es e Fo m des Wi ks o es eine wich ige Rolle, wie im Folgenden noch au gezeig wi d. Wi scha liche und Ma ke ing Aspek e sollen hie nich e ie we den. Sie sind in de pha mazeu ischen Indus ie siche lich ebenso bedeu end wie die o he igen und sind deshalb de Volls ändigkei halbe hie e - wähn . Einlei ung 8 1.2. Wi ks o e in es e Fo m In de Fes kö pe chemie molekula e Ve bindungen kenn man un e schiedliche Va- ian en, in denen ein bei Raumbedingungen es e Wi ks o o kommen kann: o Polymo phe o Salze o Sol a e und Hyd a e o Co-K is alle o Amo phe Phasen Abb. 1.2 gib einen schema ischen Übe blick übe die e wähn en Va ian- en. Beach enswe is , dass es sich bei allen, auße na ü lich bei den amo phen Phasen, jeweils um k is alline Fes kö - pe handel . Jedoch is , wie be ei s e - wähn , de Übe gang on k is allin zu amo ph in ielen Fällen ließend und nich eindeu ig abzug enzen, wie sich am Beispiel des NaBz auch in diese A bei gezeig ha . Fü die Nomenkla u in de pha mazeu ischen Indus ie e - wähnenswe is zudem auch noch, dass die FDA alle oben au ge üh en Wi k- s o a ian en als Polymo phe bezeichne .23 1.2.1. Beg i e und De ini ionen Polymo phie. Die Polymo phied Abb. 1.2 ode de Polymo phismus is ein e b ei e es Phä- nomen in de Fes kö pe chemie. E besch eib das Au e en on mindes ens zwei es en, k is allinen Phasen eine Ve bindung, bei gleiche chemische Zusam- mense zung, du ch un e schiedliche Ano dnung de Moleküle ode allgemein Bau- s eine zueinande im Fes kö pe .24 zeig schema isch zwei Ano dnungsmög- lichkei en ü ein ache Baus eine au eine Ebene bezogen. Mi sche lich e wende e den Beg i in seinen A bei en übe isomo phe Me allsul a e e s mals im Jah e d polys (g iech.) = iel; mo phe (g iech.) = Ges al Abb. 1.2 Übe blick übe die e schiedenen Fes kö pe - a ian en eines Wi ks o moleküls. Es sind jeweils zwei mögliche polymo phe Fo men gezeig . Einlei ung 9 1822.25 Die e s e Beobach ung de Polymo phie an o ganischen Fes kö pe n wu de on F ied ich Wöhle und Jus us on Liebig im Jah e 1832 bei de Umk is allisa ion on Benzamid gemach .26 Wenn Polymo phie bei Elemen en au i , wi d dies als Allo opiee Salze. Falls in einem Molekül eine en sp echend geeigne e unk ionelle G uppe (z.B. Amino- ode Ca bonsäu eg uppe) o handen is , läss sich mi els P o onie ung ode Dep o onie ung du ch Umse zung mi eine en sp echenden Säu e bzw. Base ein Salz he s ellen. Gemäß De ini ion is ein Salz aus Ka ionen und Anionen au ge- bau .50 Salze sind bei pha mazeu ischen Wi ks o en wei e b ei e .51 Zu P o onie ung basische unk ionelle G uppen we den häu ig HCl, HB und HI ein- gese z . Häu ig e wende e sau e unk ionelle G uppen sind Sul onsäu en, Phospho , Fuma -, Zi onen-, Milch- und Essigsäu e.52 Als Ka ionen inden Alkali (Na+, K+), E dalkali (Ca2+, Mg2+), NH4+, Zn2+ Cholin, L-A ginin und L-Lysin iel ache Ve - wendung.8 bezeichne . Die Bedeu ung de Polymo phie ü die pha mazeu ische Indus ie zeig sich z.B. auch da in, dass 80% alle Wi ks o e Polymo phie zeigen.27 Jedoch spiel sie nich nu in de pha mazeu ischen Indus ie, sonde n auch in ande- en Zweigen eine bedeu ende Rolle. Polymo phie is so auch im Be eich on P lan- zenschu zmi eln,28-32 Pigmen en,33-37 nich linea -op ischen Ma e ialien,38-42 und auch bei Lipiden43-46 ( gl. Schokoladenhe s ellung47-49) wich ig. Sol a e und Hyd a e. Wenn in die K is alls uk u Lösungsmi elmoleküle au ( es- en) Plä zen eingebau we den, so wi d dies als Sol a bezeichne . Wenn es sich bei dem Lösungsmi el um Wasse handel , sp ich man on Hyd a en. Wei e hin kann man zwischen s öchiome ischen und nich -s öchiome ischen Sol a en un e - scheiden.53-55 Dieses Ve hal en i o bei Ve bindungen au , die Hohl äume in de K is alls uk u ausbilden. Ein wei e e Beg i , de ü Sol a e e wende wi d, is de Beg i de Pseudopolymo phie.56-58 Co-K is alle. Eng mi den Sol a en e wand sind Co-K is alle. Ans a einem Lö- sungsmi el (bei Raum empe a u lüssig) wi d eine neu ale, bei Raum empe a u es e Komponen e in das K is allgi e eingebau .59-61 Die Komponen en, im Falle on pha mazeu ischen Co-K is allen also Wi ks o molekül und zwei e (d i e,…) Komponen e (als co-c ys al o me bezeichne ) s ehen dabei in einem de inie en s öchiome ischen Ve häl nis zueinande . Die co-c ys al o me dü en, im Falle on e allos (g iech.) = ein ande es; ope (g iech.) = Umwandlung Einlei ung 10 A zneimi eln, nich - oxisch sein und soll en im Ideal all ü Lebensmi el zugelassen sein.62 Amo phe Phasen. Amo phe Phasen weisen keine ansla o ische Fe no dnung au . Meis ens liegen jedoch Be eiche mi Naho dnung o . De Übe gang zwischen amo ph und k is allin is meis ließend und man kann iele Zwischens u en un e - scheiden: nanok is alline Ma e ialien mi g oßem An eil an elaxie em, obe lächen- nahem Volumen und, wie be ei s e wähn , Subs anzen mi zunehmende Konzen a- ion on ausgedehn en De ek en ode nu niede dimensionale O dnung.1,6,63 Von besonde e Bedeu ung is , dass bei allen ie le z genann en Fo men selbs Polymo phie au e en kann.64 In diese A bei konn e dieses Ve hal en auch anhand polymo phe Co-K is alle au gezeig we den. Bei amo phen Phasen sp ich man hie dann on Polyamo phie.65-68 Auße dem in e essan is die Ta sache de Exis enz des sogenann en Salz-Co-K is all Kon inuums.69 Bei Salzen is de P o on T ans e on Säu e zu Base p ak isch oll- s ändig e olg , wäh end bei den Co-K is allen am ande en Ende des Kon inuums diese nich s a inde . De G ad des P o on ans e s wi d dabei on de Di e enz de pKa-We e und de Umgebung im K is all bes imm .70 1.2.2. Wich ige Eigenscha en on Wi ks o en Um e olg eich als A zneimi el am Ma k zugelassen zu sein, muss de be e ende Wi ks o zum Einen eindeu ig wi ksam sein und zum Ande en die oben genann en K i e ien wie S abili ä , Ve a bei ba kei und Bio e ügba kei besi zen.9 Diese K i- e ien we den o allem du ch die physikalisch-chemischen Eigenscha en des Wi k- s o es bes imm .7,71 Au die genann en K i e ien soll nun genaue eingegangen we - den. Wich ige physikalisch-chemische Eigenscha en, welche sich du ch die e wen- de e eingese z e Va ian e des API und die K is allisa ion e ände n lassen, sind: o in insische und ex insische Löslichkei und Lösungsgeschwindigkei o Pa ikel o m und Pa ikelg ößen e eilung o Dich e o Lu euch eemp indlichkei o elek os a isches Au ladungspo en ial o S abili ä gegen Wä me, D uck Einlei ung 11 S abili ä . Die Hal ba kei eines A zneimi els is on g undlegende Bedeu ung, denn bei nich aus eichende Hal ba kei is die Zulassung nich zu e eichen. Die S abili ä muss wäh end de gesam en angegeben Daue de Hal ba kei gewäh leis e sein. Die Auswahl de en sp echenden Fo m de API spiel dabei eine k i ische Rol- le. Sie muss dabei den bes imm en Umwel ein lüssen (Tempe a u , Lich , Bewegung, Lu euch e, Saue s o , Mik oo ganismen) s andhal en.7 Von den oben genann en Eigenscha en spielen hie o allem die Lu euch eemp indlichkei und die S abili- ä gegen Wä me und D uck eine wich ige Rolle. Ein Beispiel ü e besse e Be- s ändigkei gegen Lu euch igkei zeigen T ask e al. De Co-c ys al on Theophyllin mi Oxalsäu e is s abile gegen Lu euch e als das Anhyd a on Theophyllin.72 Gue ie i e al. konn en zeigen, dass die Auswahl des Gegenions ei- nen be äch lichen Ein luss au die chemische S abili ä des Lokalanäs he ikums P ocain ha .73 Ve a bei ba kei . Die Ve a bei ba kei des e wende en A zneis o es is bei de P oduk ion on g undlegende Bedeu ung. Hie bei spielen die Lu euch eemp ind- lichkei , die S abili ä gegen Wä me und D uck als auch die Möglichkei de elek o- s a ischen Au ladung eine wich ige Rolle. Beispielhalbe soll die Pa ikelg ößen e eilung eines es en Wi ks o es du ch Mahlen e inge we den um die Bio e ügba kei so zu s eige n.74-76 Falls diese jedoch d uckemp indlich eagie und sich somi ibochemisch in eine ande e Fo m umwandel ode sich s a k elek os a isch au läd , is dies mögliche weise nich ode nu un e e schwe en Be- dingungen möglich. Fü die Ve a bei ung wich ige Kenng ößen s ellen die Riesel ä- higkei bzw. das Fließ e hal en on Schü gü e n, die Schü dich e und die S amp - dich e da . Diese hängen im Wesen lichen on de Mo phologie, de Pa ikelg ößen e eilung, de Obe lächenbescha enhei und Feuch e ab.7 Diese Ei- genscha en sind wiede um on de eingese z en Fo m (Salz, Co-K is all), de K is- alls uk u und dem K is allwachs um abhängig. Bei de Table ie ung is die Komp- essibili ä des Wi ks o es on g undlegende Bedeu ung. Pa ace amol s ell hie bei ein in e essan es Beispiel da . De zei sind d ei polymo phe Fo men cha ak e i- sie .77-79 Indus ielle Anwendung inde dabei meis Fo m I ( he modynamisch s abil). Wie Nichols e al. gezeig ha , is Fo m II (me as abil) deu lich besse komp- imie ba und wü de somi eine Ve besse ung gegenübe dem bishe igen S anda d Fo m I als Wi ks o da s ellen.80 Ungüns ig is alle dings die Me as abili ä on Fo m II, welche die Hal ba kei he abse z .81 Einen in e essan en Ansa z ha dabei Einlei ung 12 die G uppe um William Jones e olg , die Pa ace amol mi u.a. Theophyllin co- k is allisie haben und so ähnlich gu e mechanische Eigenscha en e eichen konn- en, wie sie Fo m II besi z , abe zusä zlich auch die benö ig e S abili ä gewäh leis- e .82 Bio e ügba kei . De Beg i de Bio e ügba kei (auch biologische ode physio- logische Ve ügba kei ) is on zen ale Bedeu ung bei de He s ellung eines A z- neimi els. Die Bio e ügba kei wi d nich zule z o allem du ch die Löslichkei , Lösungsgeschwindigkei , Pa ikel o m und Pa ikelg ößen e eilung mi be- s imm . Neben de Löslichkei is o allem auch die Pe meabili ä (Du ch- gängigkei du ch die Zellmemb anen) on en scheidende Bedeu ung.9 Die Bio e ügba kei d ück in P ozen en aus, in welche Menge und wie schnell de un e ände e Wi ks o nach Ap- plika ion eine A znei o m im G oß- k eislau e schein .83 Die Bio e üg- ba kei wi d du ch die Bes immung de Konzen a ion des A zneis o es im Blu - plasma nach pe o ale Abb. 1.3 Applika ion e mi el .84 Hie bei wi d eine Plasmakonzen a i- ons-Zei -Ku e gemessen, die in (angelehn an 84) gezeig is . Die absolu e Bio e ügba kei ( ) abs F läss sich dann wie olg bes immen:84 ( ) % 100 i ex a asal abs ex a asal i D AUC FD AUC ⋅ = ⋅ ⋅ (1.1) Hie bezeichne i D und ex a asal D die e ab eich e Dosis in a enösg (i ) espek i e ex a asalh i AUC ; und ex a asal AUC die Fläche un e de Plasmakonzen a ionsku e in a enös espek i e ex a asal. Hin e g und dabei is , dass die in a enöse Ve ab- eichung des Wi ks o es eine Bio e ügba kei on 100% en sp ich . Da, wie e wähn , die Löslichkei und Löslichkei sgeschwindigkei einen eno men Ein luss au die Bio e ügba kei ha , kann diese du ch Wahl de en sp echenden pe (la .) = du ch, übe ; os, o is (la .) = Mund g in a (la .) = hinein, innen; ena (la .) = Vene, Blu ade h ex a (la .) = auße halb; as (la .) = Ge äß Abb. 1.3 Plasmakonze a ions-Zei -Ku e nach pe o ale Applika ion eines A zneimi els. Die maximale A znei- s o konzen a ion (cmax ) im Plasma wi d dabei nach de Zei max e eich . Einlei ung 13 Fo m de API bedeu end e besse we den. So sind Salze du ch den ionischen Cha- ak e meis besse löslich in Wasse und dami in Kö pe lüssigkei en.85-88 Hie is alle dings zu beach en, dass es meis ens biopha mazeu isch sinnlos is , Salze on basischen Wi ks o en zu e ab eichen, weil diese beding du ch das sau e Milieu im Magen selbs zu den Salzen eagie en.7 Einen ähnlichen Vo eil bie en Co-K is alle und me as abile polymo phe Fo men;89-91 diese sind du ch die besse e Löslichkei auch meis besse bio e ügba . So zeig en Weyna e al. o ku zem, dass Co- K is alle des nich s e oidalen An i heuma ikums Meloxicam signi ikan besse e Lö- sungsgeschwindigkei en au weisen als die eine Fo m.92 Fü die Ve besse ung de Bio e ügba kei inden sich iele wei e e Beispiele.93-98 Die Bedeu ung de Poly- mo phie wi d zum Beispiel bei den polymo phen Fo men de An ibio ika Te acyclin deu lich, bei de die me as abile Fo m II aus eichende Bio e ügba kei besi z , die he modynamisch s abile Fo m I hingegen nich .99 Amo phe Phasen sind he mody- namisch gesehen am ins abils en und bie en dahe eben alls eine höhe e Löslich- kei .100-103 Blagden e al. geben einen Übe blick übe die Möglichkei en des „C ys al Enginee ing“ zu Ve besse ung de Löslichkei und Löslichkei s a e on Pha mazeu- ika.104 1.2.3. Pa en ech liche Aspek e Die un e schiedliche Fo mulie ung on Wi ks o en, insbesonde e auch die e schie- denen K is all o men haben besonde e Bedeu ung bei de Pa en ie ung. Zen ale Beg i dabei is de de In ellec ual P ope y des geis igen Eigen ums. Das Pa en schü z das geis ige Eigen um und so g da ü , dass de /die E inde ü eine gewisse Zei das ausschließliche Nu zungs ech an diesem ha und dadu ch an diesem en - sp echend e dienen kann.9 P inzipiell sollen dadu ch die eilweise hohen zei lichen und inanziellen Au wendungen, die ü die E indung nö ig wa en, ausgeglichen we den. Um pa en ie we den zu können, muss die E indung im Wesen lichen d ei K i e ien e üllen:8,62 o Sie muss einen Nu zen haben o Sie muss neu sein o Sie muss einen e inde ischen We haben Die genaue Fo mulie ung diese K i e ien is on Land zu Land un e schiedlich, je- doch sind sie meis in i gendeine Fo m en hal en. Um nun au die oben besch iebe- nen Fes kö pe o men zu ückzukommen, läss sich sagen, dass p inzipiell jede( /s) Einlei ung 14 neu(e) ge undene Polymo ph, Sol a , Hyd a und Co-K is all, welche( /s) einen sig- ni ikan en Vo eil ha , pa en ie we den kann. Folglich können die e s en beiden Punk e e üll we den. De d i e Punk des e inde ischen We s s ell o den Knackpunk da , ob ein Pa en on de jeweiligen Behö de dann akzep ie wi d. G undsä zlich läss sich jedoch sagen, dass die geziel e K is allisa ion bes imm e Polymo phe (noch) nich gegeben is und dami imme eine gewisse e inde ische Tie e angenommen we den kann.105 In den e gangenen Jah zehn en gab es einige p ominen e Beispiele ü die ökonomi- sche Bedeu ung de Pa en ie ba kei einzelne polymo phe Fo men. So kam es eils zu e bi e en und länglichen Ge ich s e ah en. Einige seien hie genann :106 o Rani idin (Zan ac®, GlaxoSmi hKline) o Ce ad oxil (Du ice ®, B is ol-Mye s) o Pa oxe in (Paxil®, GlaxoSmi hKline) o Te azosin (Hy in®, Abbo ) Einlei ung 15 1.3. Theo ie und P axis de K is allisa ion – The modynamik In diesem Abschni soll genaue au die K is allisa ion an sich eingegangen we den. Zunächs soll jedoch die U sache ü das Au e en on Polymo phie genaue be- leuch e we den. Danach we den die he modynamischen Beziehungen on Poly- mo phen zueinande e läu e . De zwei e Teil dieses Abschni s beschä ig sich mi de K is allisa ion. Dabei wi d zunächs die Theo ie de K is allisa ion anhand de klassischen Keimbildungs heo ie besch ieben, danach geh es um die p ak ische K is allisa ion. 1.3.1. U sache de Polymo phie Die U sache ü das Au e en de Polymo phie is im komplexen Zusammenspiel on in a- (ko alen , ionisch) und/ode in e molekula en Wechselwi kungsk ä en zu inden. Bei molekula en Fes kö pe n e en olgende in e molekula e Wechselwi - kungen (WW) au : o Ionische WW o Wasse s o b ückenbindungen o Dipol-Dipol WW o Dispe sions WW Da auch das einzelne Molekül an sich zumeis nich s a is , kann es je nach Kom- plexi ä meh ode wenige ene ge isch un e schiedlich güns ige Kon o ma ionen annehmen. Die einzelnen Moleküle können sich nun im K is all selbs in e schiede- nen ene ge isch ähnlich güns igen Packungsmo i en ano dnen. Je nachdem, was meh zu i , kann man Polymo phie wei e in con o ma ional polymo phism107-109 ode packing polymo phism110-112 ein eilen. Die Komplexi ä des Zusammenspiels de oben genann en WW wi d deu lich, wenn man die heo e ische Vo he sage on K is alls uk u en (Polymo phe, Co-K is alle, Sol a e…) be ach e .113 Hie e such man be ei s exis ie ende K is alls uk u en und zusä zliche wei e e ene ge isch güns- ige K is alls uk u en om Zielmolekül mi compu e chemischen Me hoden (klas- sisch und quan enmechanisch) zu be echnen. T o z eils immense Fo sch i e in den le z en 10 Jah en is es bislang noch nich möglich K is alls uk u en siche o he - zusagen.114-118 Wenn dies zu e lässig möglich wä e, wü de dies einen eno men Fo - sch i ü die Fo schung und auch ü die Indus ie, besonde s auch die pha mazeu i- sche Indus ie, da s ellen.119-121 Doch auch wenn es die Möglichkei gäbe, neue K is- Einlei ung 22 und bei ge ingen Kos en du chge üh we den können.142-144 Die Auswe ung eine Pla e e olg mi els Raman-Spek oskopie ode Rön genpul e di ak ome ie, wo- du ch das en s andene P oduk cha ak e isie we den kann.145 Du ch den hohen P o- bendu chsa z bes imm , lassen sich so meis ens die ü die Indus ie ele an en Fo - men bes immen. Ein Co-K is all, Sol a - bzw. Salz-Sc eening läss sich auch mi de Hochdu chsa zme hode du ch üh en.146-148 Wenn man die gewünsch e Fo m, welche die gewünsch en physikalisch-chemischen Eigenscha en besi z , so e mi el ha , geh es an die P oduk ion und o allem die ep oduzie ba e He s ellung diese . Hie bie e es sich meis an in de sogenann en me as abilen Zone, du ch Imp en (Seeding) mi de gewünsch en Fo m, zu k is allisie en. Imp en in de me as abilen Zone. De A bei sbe eich ü das Imp en lieg in de sogenann en me as abilen Zone ( gl. Abb. 1.7).127 Diese wi d on de s abilen Zone und de ins abilen Zone eingeschlossen. U sache ü die me as abile Zone is die o - he disku ie e Keimbildungsba ie e. Die Abg enzung zwischen s abile und me a- s abile Zone wi d on de Sä igungs- ku e, an welche die Lösung ge ade gesä ig is , beg enz . Ab de me as abi- len Ku e is die Lösung übe sä ig . Die me as abile Zone wi d au de ande en Sei e on de ins abilen Zone du ch die Übe sä igungsku e beg enz . Diese besch eib die G enze, an de spon an Keimbildung e olg . Inne halb de me- as abilen Zone selbs komm es somi zu keine homogenen Keimbildung (keine spon ane Nuklea ion), sonde n ausschließlich zum K is allwachs um. Die B ei e de me as abilen Zone (MSZW) häng om S o sys em an sich, K is allisa ionsbedin- gungen und Ene gieein ag (Rüh geschwindigkei , Heiz- und Kühl a en, F emdpa i- kel) ab. Inne halb de me as abilen Zone is somi ein geziel es K is allwachs um eine gewünsch en Modi ika ion du ch sogenann es Imp en ode Seeding mi de gewünsch en Fo m möglich.149-152 Die P ozesskon olle e olg dabei zumeis übe in si u ATR-FTIR (a enua ed o al e lec ance- ou ie ans o m in a ed) Spek osko- pie, in-si u FBRM ( ocused beam e lec ance measu emen ) Messung und in-si u Lich mik oskopie PVM (pa icle ision and measu emen ).153-155 Als besonde s in e- essan ha sich in den le z en Jah en auch die he e ogene Keimbildung e wiesen. Abb. 1.7 Schema ische Übe blick übe die K is allisa- ion eine gewünsch en Fo m du ch Imp ung inne halb de me as abilen Zone. Einlei ung 23 He e ogene Keimbildung. Du ch das Zuse zen on he e ogenen Keimbildne n is es so gelungen neue unbekann e S uk u en zu k is allisie en.156 Hie wu den bislang sowohl o ganische und ano ganische Subs a e als auch Polyme e e olg eich e - wende . Die Keimbildungsba ie e bei de he e ogenen Nuklea ion is wesen lich ge inge und du ch die s uk u ellen Gegebenhei en de Subs a e is eilweise eine epi axiale Vo o ganisa ion und so epi axiales K is allwachs um des Zielmoleküls zu e wa en.157-159 Eine eng mi de he e ogenen Keimbildung e wand e Me hode is die Emulsionsk is allisa ion. Emulsionsk is allisa ion. Hie mach man sich den s uk u bildenden Cha ak e on G enz lächen zunu ze. Insbesonde e ü amphiphile Moleküle in e essan is die K is allisa ion in Öl-Wasse -Emulsionen, da das Molekül eine Tendenz zu An ei- che ung an de G enz läche pola -unpola besi z .160 Wei e b ei e is auch die K is allisa ion in Mik oemulsionen, mi de man K is alli g öße und Mo phologie, abe auch die Polymo phie beein lussen kann.161-163 Einlei ung 24 1.4. Un e such es Modellsys em In diese A bei wu den zwei Moleküle, Benzamid und HBz, de en molekula e S uk u eng mi einande e wand is , als Modellsys em un e such . Subs i uie e De i a e des Benzamid inden Ve wendung in de Pha maindus ie als Neu olep ika (z.B. Sulpi id®)164, An ieme ika (z.B. Me oclop amid®)165 ode auch Chemo he a- peu ika (z.B. P oca bazin®).166 De Lebensmi elzusa zs o HBz (E210) inde als Konse ie ungsmi el b ei e Anwendung in de Indus ie. Zusä zlich wu den e - schiedene Salze de HBz un e such : LiBz, NaBz (E211) und KBz (E212). Le z e e inden eben alls als Konse ie ungsmi el b ei e Ve wendung.167 1.4.1 Benzamid Benzamid wa das e s e molekula e, polymo phe Sys em, was F ied ich Wöhle und Jus us on Liebig e s mals im Jah e 1832 beme k en. Bei Abkühlung eine wäss ig kochenden Lösung on Benzamid bilde e sich zunächs eine nadel ö mige Fo m, die sich nach einigen S unden in eine plä chen ö mige Fo m umwandel e.26 Bislang konn en d ei Polymo phe s uk u ell cha ak e isie we den. Die bei Raumbedingun- gen he modynamisch s abile Fo m I (plä chen ö mige Mo phologie) wu de 1959 mi els Rön geneink is alls uk u analyse (RKSA) on Pen old e al. bes imm .168 Da id e al. gelang es im Jah e 2005 die seh ins abile Fo m II (nadel ö mige Mo - phologie) in Mischung mi Fo m I (17:83) aus Synch o onpul e da en s uk u ell zu cha ak e isie en.169 Schließlich gelang es Thun e al. 2007 die S uk u de eben alls me as abilen Fo m III (eben alls nadel ö mige Mo phologie), in Mischung mi Fo m I (50:50), aus Labo ön genpul e da en zu e mi eln.134 Au g und de angewand en K is allisa ionsbedingungen is die Fo m III de Fo m zuzuo dnen, welche Wöhle und Liebig be ei s 1832 beobach e en. Fo m I kann du ch langsames Abkühlen pha- sen ein aus Wasse k is allisie we den. Gemäß de Os waldschen S u en egel lassen sich du ch die Ve wendung höhe e Abkühl a en und somi höhe e Übe sä igung die me as abilen Fo men II und III k is allisie en. Bislang konn en diese nie phasen- ein k is allisie we den. In allen K is allisa ionsexpe imen en wu de imme nu eine Mischung aus Fo m I und III ode I, II und III e hal en.170 Geeigne e K is allisa ions- bedingungen ü die me as abilen Fo men, die zu keine Sekundä nuklea ion on de he modynamisch s abilen Fo m I üh en, sind ex em schwie ig zu inden. Einlei ung 25 1.4.2 Benzoesäu e – Benzoa salze T o z de g oßen s uk u ellen Ähnlichkei zu Benzamid, konn e bei HBz bislang keine Polymo phie ge unden we den. Die einzige bislang bekann e K is alls uk u wu de 1955 on Sim e al. mi els RKSA bes imm .171 Von den S uk u en de Alkalime allbenzoa e konn e bislang nu die on LiBz du ch Pla ne e al. 1994 be- s imm we den.172 Die S uk u en on NaBz und KBz konn en bislang nich be- s imm we den. 1974 konn e Flamme sheim die Exis enz eines Co-K is alls zwi- schen HBz und NaBz zeigen. E bes imm e die Zusammense zung zu 2 HBz ∙1 NaBz (2 Teile HBz und 1 Teil NaBz).173-175 Die S uk u diese Fo m konn e alle dings o z geglück e Indizie ung nich bes imm we den. Auch konn e e die Exis enz eine zwei en Fo m diese Phase au zeigen. Zusä zlich be ich e e übe Hinweise eines zwei en Co-K is alls mi de Zusammense zung 1 HBz ∙ 2 NaBz. Einlei ung 26 Synopsis 27 2. Synopsis De übe wiegende An eil de A zneis o moleküle besi z en wede sau en ode basi- schen Cha ak e und en häl olglich en sp echende unk ionelle G uppen (z.B. Ca boxyl- ode Amino unk ionen).9 Falls diese dann wiede um eine schlech e Lös- lichkei in Wasse besi zen, is die Bio e ügba kei o nich gewäh leis e . Ein mög- liche Ausweg is die Bildung eines Salzes. Diese sind, wie oben e wähn , meis deu lich besse bio e ügba . Fü Wi ks o moleküle mi basischem Cha ak e is die Fo mulie ung als Salz meis nich no wendig, da die Salzbildung im Magen du ch das salzsau e Milieu selbs e olg . Fü Wi ks o e mi Ca bonsäu e unk ionen wi d be o zug das en sp echende Alkalime allsalz (Na+, K+) o mulie . 2.1. Mo i a ion Die en sp echenden Alkalime allsalze sind o schlech k is allin und/ode weisen einen hohen G ad an Fehlo dnung au . De G und hie ü kann au das G ößen e - häl nis de beiden Ionen zueinande zu ückge üh we den. Die kleinen Alikalime- allionen s ehen o g oßen Wi ks o anionen gegenübe . Als Modellsys em is HBz gu geeigne , da es ähnliche Cha ak e is iken au weis . HBz selbs is nu seh mäßig in Wasse löslich (2,9 g/l bei 25 °C),176 die Salze hingegen ausgezeichne ( gl. NaBz: 660 g/l bei 20 °C,177 KBz: 556 g/l bei 20 °C)178. HBz is gu aus Lösung, Schmelze ode de Gasphase zu k is allisie en. Die Benzoa salze (besonde s NaBz) hingegen sind nu eilk is allin, was sich auch am Pul e di ak og amm zeig , das bis au den e s en Re lex nu b ei e übe lage e Hügel on Re lexen au weis und sich so nich indizie en läss (Abb. 2.3). NaBz lei- de dabei, wie iele pha mazeu ische Wi ks o e, an dem oben skizzie en P oblem: Selbs wenn die Ca boxyla g uppe als zweizähnige Ligand ungie , können mi eine 1:1-S öchiome ie die koo dina i en Bedü nisse des Ka ions nu zu ieden ges ell we den, wenn sich massi kan en e knüp e S uk u en ausbilden könn en. Dem s eh abe de s e ische Ansp uch des an die Ca boxylg uppe gebundenen o ga- nischen Res s en gegen. Du ch Co-K is allisa ion mi HBz lassen sich zusä zliche neu ale Liganden ein ü- gen, die dann im S ande sind, die Koo dina ionsbedü nisse abzusä igen. De Ve - lau de Co-K is allisa ion wi d dabei im Wesen lichen on d ei Fak o en bes imm : Synopsis 28 o Eingese z e S öchiome ie (HBz : Benzoa ) o K is allisa ions üh ung o Un e schiedliche Ka ioneng öße P ak isch we den die en sp echenden Co-K is allisa ionen en wede aus Lösung in s öchiome ischen Ve häl nissen de sogenann en Co-K is all Bildne (co-c ys al o me ) ode mi els Sol en D op G inding du chge üh .179 Bei diesem wi d eben alls eine s öchiome ische Mischung de (beiden) Bildne mi seh wenig Lösungsmi el (einige T op en) in ensi gemö se . Diese Technik is genauso e ek i wie das Sc eening aus Lösung, jedoch wi d kaum Lösungsmi el benö ig , was so zu eine besse en ökonomischen und ökologischen Bilanz üh .180 Bei de K is allisa ions üh ung wu de zudem zusä zlich noch gemäß de Os waldschen S u- en egel k is allisie . Bei höhe e Übe sä igung k is allisie en so be o zug me as abile Phasen aus, was auch ü die Bildung de Co-K is all-Phasen gil . Beim zwei en Modellsys em Benzamid konn en auße dem du ch die genaue Ein- s ellung de K is allwachs umsbedingungen g oße Eink is alle de me as abilen Fo m III k is allisie we den, die so ü eine ie gehende Cha ak e isie ung diese Phase he angezogen wu den. Hie konn e die Exis enz molekula e De ek e ku z o dem eigen lichen Schmelzpunk di ek gezeig und nachgewiesen we den. 2.2. Ein luss de S öchiome ie bei de Co-K is allisa ion Die Co-K is allisa ion on HBz mi NaBz wu de sys ema isch un e such . Hie ü wu den e schiedene s öchiome ische Ve häl nisse (3:1, 2:1, 1:1, 1:2, 1:3, 1:4, 1:5) on HBz mi NaBz mi els Sol en -D op G inding ibochemisch co-k is allisie . Abb. 2.1 Packungsmus e on Fo m A des 2 HBz ∙ 1 NaBz Co-K is alls. (a) Annähe nd hexagonale Packung on Bände n en lang de a-Achse. (b) Sei liche Ansich eines Bandes. Koo dina i e Bindungen sind schwa z und Wasse s o b ückenbindungen blau ges ichel . (c) Das Kalo enmodell e deu lich die dich e Packung de Na - ium Ka ionen. (C ys EngComm, 2012, 14, 3945-3950. - Rep oduced by pe mission o The Royal Socie y o Chemis y) Synopsis 29 Die K is alls uk u eines phasen einen Co-K is alls mi de Zusammense zung 2 HBz ∙ 1 NaBz konn e mi els RKSA bes imm we den ( gl. Abb. 2.1). Diese Phase wu de be ei s on Flamme sheim besch ieben, jedoch konn e diese die Phase lediglich indizie en, nich abe de en S uk u bes immen. Die S uk u is aus eindimensionalen Bände n au - gebau . Diese sind wiede um annähe nd hexagonal gepack . Die HBz-Liganden und die Benzoa -Liganden koo dinie en gleichzei ig an das Na ium und s ellen somi dessen Koo dina ionsbedü nis zu ie- den. Die Bände sind aus kan en e knüp en Ok aede pä chen au gebau . Diese Pä - chen sind du ch zweizähnige Liganden (HBz und Benzoa ) zu den Bände n e - knüp . Bei Abweichungen on de 2:1 S öchiome ie wu den imme Phasenge- mische e hal en, Hinweise au ande s zusammengese z e Co-K is alle konn en nich e hal en we den. An HBz eiche e Ve häl nisse (3:1) lie e en ibochemisch imme eine Mischung aus HBz und dem 2 HBz ∙ 1 NaBz Co - K is all ( gl. Abb. 2.2). Die Ve wendung ande e s öchiome ische Ve häl nisse mi höhe em NaBz An eil (1:1 bis 1:3) zeig eine Mischung aus dem 2 HBz ∙ 1 NaBz Co-K is all zusammen mi d ei Re lexen, die nich mi NaBz übe ein- s immen ( gl. Abb. 2.3). Eine Indizie- ung kann nich gelingen, de Beweis eine wei e en Phase kann also nich ange e en we den. Bei noch g öße em Abb. 2.2 Pul e di ak og amme de Co-K is allisa e und jeweiligen Re e enzen (blau: HBz; schwa z: HBz:NaBz = 3:1; o ange: HBz:NaBz = 2:1). (C ys EngComm, 2012, 14, 3945-3950. - Rep oduced by pe mission o The Royal Socie y o Chemis y) Abb. 2.3 Pul e di ak og amme de Co- K is allisa e und jeweiligen Re e enzen (o ange: HBz:NaBz = 2:1; o : HBz:NaBz = 1:1; lila: HBz:NaBz = 1:2; g ün: HBz:NaBz = 1:3; blau: HBz:NaBz = 1:4; hellg ün: HBz:NaBz = 1:5; schwa z: NaBz). (C ys EngComm , 2012, 14, 3945-3950. - Rep oduced by pe mission o The Royal Socie y o Chemis y) Synopsis 30 NaBz An eil (1:4 und 1:5) zeig das Di ak og amm einen g aduellen Übe gang hin nach NaBz. De e s e Re lex wande dabei hin zu g öße en d-We en mi abneh- mendem HBz : NaBz Ve häl nis. Dieses Ve hal en is bekann on Einlage ungs e - bindungen, die s a is ische Wechsellage ung zeigen. Die Exis enz eine möglichen 1 HBz ∙ 2 NaBz Phase kann zwa au g und de Da enlage nich de ini i ausgeschlos- sen we den, jedoch sp echen die E gebnisse ehe ü ein semik is allines Ma e ial. Eine eindeu ige Aussage wi d du ch den ehlgeo dne en Cha ak e des Di ak og ams on NaBz e schwe , de au niede dimensionale Baueinhei en hin- deu e , welche un e einande s a k ehlgeo dne gepack sind. Fü diesen Au bau sp ich auch die spä e besch iebene in diese A bei geklä e Eink is alls uk u on NaBz, die in de Ta eine Einlage ung e schiedene (nich s öchiome ische ) Men- gen on HBz zwischen die S abpackung möglich e scheinen lassen. 2.3. Ein luss de Übe sä igung bei de Co-K is allisa ion Neben de S öchiome ie is auch die einges ell e Übe sä igung on g undlegende Bedeu ung ü das e hal ene K is allisa . Gemäß de Os waldschen S u en egel k is- allisie en bei höhe e Übe sä igung und nied ige Tempe a u zunächs die me a- s abilen Fo men be o zug aus.181 Du ch Ände ung de K is allisa ionsbedingungen is es in diese A bei gelungen, ein Polymo ph des 2 HBz ∙ 1 NaBz Co -K is alls (Fo m B) zu k is allisie en. Die höhe e Übe sä igung wu de du ch schlaga iges Ve damp en des Lösungsmi els Me hanol gene ie . Hie zu wu de eine heiße Lö- sung au eine o geheiz en Obe läche (75 °C) schlaga ig e damp . T o z de ha - schen Bedingungen (kleine e wachsene K is alli e) gelang es die S uk u on Fo m B mi els RKSA zu bes immen. Im Ve gleich zu he modynamisch s abilen Fo m A, die aus eindimensionalen Bände n au gebau is , is Fo m B aus eindimensionalen S äbchen au gebau ( gl. Abb. 2.4). Abb. 2.4 Packungsmus e on Fo m B des 2 HBz ∙ 1 NaBz Co -K is alls. (a) Annähe nd hexagonale Packung on S äbchen en lang de a-Achse. (b) Sei liche Ansich eines S äbchens. Koo dina i e Bindungen sind schwa z und Wasse s o b ückenbindungen blau ges ichel . (c) Kalo enmodell eines S äbchens e deu lich die dich e Pa- ckung de Na ium Ka ionen. (C ys EngComm, 2012, 14, 6744-6749. - Rep oduced by pe mission o The Royal Socie y o Chemis y) Synopsis 31 Diese sind eben alls annähe nd hexago- nal gepack . De Un e schied zwischen beiden Polymo phen inde sich im Au bau diese Un e einhei en. Im Ge- gensa z zu Fo m A, in de die Bände aus kan en e knüp en Ok aede pä chen au gebau sind, sind die S äbchen in Fo m B aus unendlichen Ok aede ke en au gebau . Jedes Ok aede is zwei ach kan en e knüp . Die he modynami- sche Beziehung zwischen beiden Poly- mo phen konn e du ch e schiedene Expe imen e und Rechnungen au geklä we - den. Dich e unk ional heo e ische Rechnungen am DFT-D (mi semiempi ische Dispe sionswechselwi kung)182 Le el zeigen, dass Fo m A e was ene ge isch güns i- ge als Fo m B (+0,08 kJ/mol) is . Expe imen ell läss sich dies übe eine du ch Lö- sungsmi el e mi el e Umwandlung zeigen. Eine 1:1 Mischung beide Fo men wandel sich in Anwesenhei on wenig E hanol binnen 40 Minu en bei Raum empe- a u olls ändig in Fo m A um ( gl. Abb. 2.5). Die eigen liche he modynamische Beziehung zwischen beiden Polymo phen konn e eben alls eindeu ig als enan io op expe imen ell bes imm we den: Au g und des hohen Sublima ionsdamp d ucks on HBz wa en aussagek ä ige DSC Messungen nich möglich, so dass die Bu ge schen Regeln so nich belas ba angewand we den konn en. Deshalb wu de ein Eink is all on Fo m A in eine gasdich e schlossene Kapilla e abge üll . Bei 110 °C wandel sich diese binnen eines Tages olls ändig in Fo m B um, was mi els Pul e di ak ome ie-Messungen beleg we den konn e. 2.4. Ein luss de Ka ioneng öße bei de Co-K is allisa ion Ein wei e e wich ige Fak o bei de K is allisa ion und Co-K is allisa ion on Sal- zen is die G öße des Ka ions im Ve gleich zum Anion. Fü das HBz/Benzoa Sys- em wu de deshalb de Ein luss de G öße de Alkalime allka ionen Li+, Na+, und K+ un e such . In diesem Zug wu de LiBz mi HBz co-k is allisie . Im Gegensa z zu NaBz wa in diesem Fall eine S öchiome ie on 1:1 ü die K is allisa ion eines Co- K is alls no wendig. Die S uk u dieses Co-K is alls 1 HBz ∙ 1 LiBz konn e ebe n- alls mi els RKSA bes imm we den. Die S uk u bes eh , ähnlich wie Fo m A des Co-K is alls 2 HBz ∙ 1 NaBz, eben alls aus Bände n. Li hium selbs is 4- ach koo - Abb. 2.5 Lösungsmi el e mi el e Umwandlung eine Mischung aus Fo m A und B. Das o e Di ak og amm en sp ich eine Fo m B und das g üne Di ak og amm eine Fo m A. Die schwa zen Di ak og amme sind nach en sp echende Zei gemessen. (C ys EngComm , 2012, 14, 6744-6749. - Rep oduced by pe mission o The Royal Socie y o Chemis y) Publika ionen und Manusk ip e 38 o Anhang A.3 In luence o Ca ion Size on he Co-c ys allisa ion o Benzoic Acid wi h Di e en Benzoa es C. Bu e ho , W. Milius und J. B eu, Z.Ano g.Allg.Chem., 2013, 2, 308-311. Das Konzep de Publika ion wu de on P o . Jose B eu, D . Wol gang Milius und mi e a bei e . Alle p ak ischen A bei en und Messungen wu den on mi du chge- üh . Die Publika ion wu de on P o . Jose B eu und mi e ass . De Eigenan eil beläu sich au e wa 90%. o Anhang A.4 Mic ophase Sepa a ion wi h Small Amphiphilic Molecules: C ys al S uc u e o P ese a i es Sodium Benzoa e (E 211) and Po assium Benzoa e (E 212) C. Bu e ho , T. Ma in, W. Milius und J. B eu, Z.Ano g.Allg.Chem., 2013, 15, 2816- 2821. Das Konzep de Publika ion wu de on P o . Jose B eu und mi e a bei e . Alle p ak ischen A bei en, Messungen und Auswe ungen wu den on mi du chge üh . Bei de Eink is alls uk u lösung on Na iumbenzoa un e s ü z e mich D . Wol - gang Milius. Bei de Rie eld Ve eine ung on Na iumbenzoa hal mi Thomas Ma in. Die Publika ion wu de on P o . Jose B eu und mi e ass . De Eigenan eil beläu sich au e wa 80%. Publika ionen und Manusk ip e 39 o Anhang A.5 The moanaly ical E idence o Me as able Molecula De ec s in Fo m I o Benzamide C. Bu e ho , T. Ma in, P. Ec o s, D. Zahn, P. Niemie z, J. Senke , C. Nä he und J. B eu, C ys . G ow h Des., 2012, 12, 5365-5372. Da s ellung des Eigenan eils: Das Konzep de Publika ion wu de on P o . Jose B eu, P o . Jü gen Senke , P o . Di k Zahn und mi e a bei e . Die p ak ischen A bei en, einige Messungen und die Eink is alls uk u e eine ung wu den on mi du chge üh . M. Sc. Thomas Ma in ha mi mi zusammen die empe a u abhängigen Pul e di ak og amme gemessen. Pha mazeu Philipp Ec o s ha die MD-Simula ionen und DFT-Rechnungen du chge- üh . Die SS-NMR Messungen und Auswe ungen ha M. Sc. Paul Niemie z ausge- üh . P o . Ch is ian Nä he ha die The moanaly ik und The momik oskopie gemes- sen. Die Publika ion wu de on P o . Jose B eu und mi e ass . P o . Ch is ian Nä he ug zum DSC-Teil, P o . Di k Zahn zum MD-Teil und P o . Jü gen Senke zum NMR-Teil bei. De Eigenan eil beläu sich au e wa 66%. Publika ionen und Manusk ip e 40 Wei e e Publika ionen 41 4. Wei e e Publika ionen Im Rahmen diese A bei en s anden olgende wei e e Ve ö en lichungen, welche nich Inhal diese kumula i en Disse a ion sind: o A. Baumga ne , C. Bu e ho , S. Koch und J. B eu, Clay Clay Mine ., 2009, 57, 271-277. o J. Thun, L. Sey a h, C. Bu e ho , J. Senke , R.E. Dinnebie und J. B eu, C ys . G ow h Des., 2009, 9, 2435-2441. Wei e e Publika ionen 42 Vo äge und Pos e bei äge 43 5. Vo äge und Pos e bei äge o Polymo phism o Benzamide Molecula C ys als: Syn hesis (Vo ag) C. Bu e ho , T. Ma in und J. B eu, 1. T e en SPP 1415 in Bad S a els ein (Klos e Banz), Juli 2010. o Kok is allisa ion on Benzoesäu e mi Na iumbenzoa (Pos e ) C. Bu e ho , W. Milius und J. B eu, 15. Vo ags agung de GDCh Fach- g uppe Fes kö pe chemie und Ma e ial o schung in Be lin, Sep embe 2010. o Polymo phie on Benzamid-Molekülk is allen – Syn hese (Vo ag und Pos e ) C. Bu e ho , T. Ma in und J. B eu, 1. Be ich skolloquium SPP 1415 in Boppa d Ok obe 2011. o Polymo phie on Benzamid-Molekülk is allen (Pos e ) C. Bu e ho , T. Ma in und J. B eu, in-si u-Wo kshop des SPP 1415 in Wis- ma , Mä z 2012. o Polymo phism in Co-c ys als: Co-c ys allisa ion o Benzoic Acid wi h Sodium Benzoa e (Pos e ) C. Bu e ho , W. Milius und J. B eu, 4 h Eu opean Con e ence on C ys al G ow h (ECCG4) in Glasgow, Juni 2012. o Neue polymo phe Fo m des Benzamids (Pos e ) C. Bu e ho , T. Ma in, I. And usenko, U. Kolb und J. B eu, 16. Vo ags a- gung de GDCh Fachg uppe Fes kö pe chemie und Ma e ial o schung in Da ms ad , Sep embe 2012. o Polymo phism in he Ionic Co-C ys al 2 HBz · 1 NaBz (Pos e ) C. Bu e ho und J. B eu, 16. Vo ags agung de GDCh Fachg uppe Fes kö - pe chemie und Ma e ial o schung in Da ms ad , Sep embe 2012. Vo äge und Pos e bei äge 44 Li e a u e zeichnis 45 6. Li e a u e zeichnis 1 K is allog aphie, W. Bo cha d -O , Sp inge Be lin Heidelbe g, 2009. 2 Chemie de We ks o e, H. B iehl, B.G. Teubne Ve lag / GWV Fach e lage GmbH Wiesbaden, 2008. 3 Mak omoleküle, H. G. Elias, Wiley-VCH, 2003. 4 Mak omolekula e Chemie: Ein Leh buch ü Chemike , Physike , Ma e ialwis- senscha le und Ve ah ens echnike , M. D. Lechne , K. Geh ke und E. H. No dmeie , Bi khäuse , 2003. 5 Polyme Chemis y: An In oduc ion, M. P. S e ens, Ox o d Uni e si y P ess, 1999. 6 Ein üh ung in die K is allog aphie, W. Klebe , H. J. Bau sch, J. Böhm und D. Klimm, Ve lag Technik Be lin, 1998. 7 Leh buch de Pha mazeu ischen Technologie, K. H. Baue , K. H. F ömming und C. Füh e , Wissenscha liche Ve lagsgesellscha S u ga , 2006. 8 Polymo phism in he Pha maceu ical Indus y, R. Hil ike , Wiley-VCH Ve lag Weinheim, Ge many, 2006. 9 Die Pha maindus ie, D. Fische und J. B ei enbach, Spek um Akademische Ve lag Heidelbe g, 2013. 10 Solid-S a e Chemis y o D ugs, S. R. By n, R. R. P ei e und J. G. S owell, SSCI Inc.: Wes La aye e, IN, 1999. 11 Fiedle - Encyclopedia o Excipien s: o Pha maceu icals, Cosme ics and Re- la ed A eas, H. P. Fiedle , Edi io Can o Ve lag, 2007. 12 B. Campisi, D. Chicco, D. Vojno ic und R. Phan-Tan-Luu, Jou nal o Pha ma- ceu ical and Biomedical Analysis, 1998, 18, 57-65. 13 H. Kalasz und I. An al, Cu . Med. Chem., 2006, 13, 2535-2563. 14 A. S. Na ang, D. Desai und S. Badawy, Pha m. Res., 2012, 29, 2660-2683. 15 R. G. S ickley, Pha m. Res., 2004, 21, 201-230. 16 C. B. Aake öy, Ac a C ys . B, 1997, 53, 569-586. 17 G. R. Desi aju, J. Chem. Sci., 2010, 122, 667-675. 18 S. Da a und D. J. W. G an , Na . Re . D ug Disco ., 2004, 3, 42-57. 19 C. P. Adams und V. V. B an ne , Heal h A ai ., 2006, 25, 420-428. 20 De elopmen and Con ol o Medicines and Medical De ices, R. J. Ha man, Pha maceu ical P ess, 2004. Li e a u e zeichnis 46 21 A. P. Lea und D. Faulds, D ugs, 1996, 52, 541-546. 22 J. O. Henck, U. J. G iesse und A. Bu ge , Pha m. Ind., 1997, 59, 165-169. 23 A. S. Raw, M. S. Fu ness, D. S. Gill, R. C. Adams, F. O. Holcombe und L. X. Yu, Ad . D ug Deli e Re ., 2004, 56, 397-414. 24 J. Haleblia und W. McC one, J. Pha m. Sci., 1969, 58, 911-929. 25 E. Mi sche lich, Ann. Chim. Phys., 1822, 19, 350-419. 26 F. Wöhle und J. on Liebig, Ann. Pha m., 1832, 3, 249-282. 27 A. G unenbe g, J. O. Henck und H. W. Siesle , In . J. Pha m., 1996, 129, 147- 158. 28 A. Bu ge und C. an den Boom, Mik ochim. Ac a, 2000, 135, 63-69. 29 U. J. G iesse , D. Weigand, J. M. Rollinge , M. Haddow und E. Gs ein, J. The m. Anal. Calo im., 2004, 77, 511-522. 30 E. Nauha, H. Saxell, M. Nissinen, E. Kolehmainen, A. Schä e und R. Schlecke , C ys EngComm, 2009, 11, 2536-2547. 31 E. Nauha, A. Ojala, M. Nissinen und H. Saxell, C ys EngComm, 2011, 13, 4956- 4964. 32 G. W. S ock on, R. God ey, P. Hi chcock, R. Mendelsohn, P. C. Mowe y, S. Rajan und A. F. Walke , J. Chem. Soc. Pe k. T. 2, 1998, 2061-2071. 33 M. U. Schmid , D. W. M. Ho mann, C. Buchsbaum und H. J. Me z, Angew. Chem. In . Ed., 2006, 45, 1313-1317. 34 C. H. Chang, R. M. Ch is ie und G. M. Rosai , Dyes Pigmen s, 2009, 82, 147- 155. 35 P. E k, Cu . Opin. Solid. S . M., 2001, 5, 155-160. 36 T. Kinu a, T. Sa o, N. Tajima, Y. Ma suba a, M. Miyazawa und Y. Imai, C ys EngComm, 2012, 14, 1016-1020. 37 C. P. P ice, A. L. G zesiak, J. W. Kamp und A. J. Ma zge , C ys . G ow h Des., 2003, 3, 1021-1025. 38 S. Blums engel, A. Bo ghesi, L. An olini, M. Jandke und P. S oh iegl, Syn he ic Me ., 2001, 124, 245-247. 39 P. Munshi, B. W. Skel on, J. J. McKinnon und M. A. Spackman, C ys EngComm, 2008, 10, 197-206. 40 M. Gi a d, N. Me cie , B. G a oueille, E. Ripaud, J. Luc und B. Sah aoui, C ys EngComm, 2008, 10, 968-971. Li e a u e zeichnis 47 41 S. P. Thomas, K. Naga ajan und T. N. G. Row, Chem. Commun., 2012, 48, 10559-10561. 42 T. V. Timo ee a, V. N. Nes e o , R. D. Cla k, B. Penn, D. F azie und M. Y. An ipin, J. Mol. S uc ., 2003, 647, 181-202. 43 A. Zaba a, I. Ama -Yuli und R. Mezzenga, Langmui , 2011, 27, 6418-6425. 44 P. K. Ta a da und M. J. Swamy, Chem. Phys. Lipids, 2009, 162, 25-33. 45 G. Mazzan i, S. E. Gu h ie, E. B. Si o a, A. G. Ma angoni und S. H. J. Idziak, C ys . G ow h Des., 2004, 4, 1303-1309. 46 N. Ga i, D. Libs e und A. Ase in, Food Func ., 2012, 3, 700-713. 47 C. Loisel, G. Kelle , G. Lecq, C. Bou gaux und M. Olli on, J. Am. Oil Chem. Soc., 1998, 75, 425-439. 48 O. Jo ano ic, D. J. Ka lo ic und J. Jako lje ic, Ac a Alimen . Hung., 1995, 24, 225-239. 49 P. F ye und K. Pinschowe , M s Bull., 2000, 25, 25-29. 50 IUPAC Compendium o Chemical Te minology, A. D. McNaugh und A. Wil- kinson, Royal Socie y o Chemis y Camb idge, 1997. 51 Handbook o Pha maceu ical Sal s: P ope ies, Selec ion, and Use, P. H. S ahl und C. G. We mu h, Wiley-VCH, 2008. 52 J. C. Espinosa-La a, D. Guzman-Villanue a, J. I. A enas-Ga cia, D. He e a- Ruiz, J. Ri e a-Islas, P. Roman-B a o, H. Mo ales-Rojas und H. Höp l, C ys . G ow h Des., 2012, 13, 169-185. 53 U. J. G iesse und A. Bu ge , In . J. Pha m., 1995, 120, 83-93. 54 G. A. S ephenson, E. G. G oleau, R. L. Kleemann, W. Xu und D. R. Rigsbee, J. Pha m. Sci., 1998, 87, 536-542. 55 R. L. Te, U. J. G iesse , K. R. Mo is, S. R. By n und J. G. S owell, C ys . G ow h Des., 2003, 3, 997-1004. 56 K. R. Seddon, C ys . G ow h Des., 2004, 4, 1087. 57 G. R. Desi aju, C ys . G ow h Des., 2004, 4, 1089-1090. 58 J. Be ns ein, C ys . G ow h Des., 2005, 5, 1661-1662. 59 A. D. Bond, C ys EngComm, 2007, 9, 833-834. 60 G. R. Desi aju, C ys EngComm, 2003, 5, 466-467. 61 J. D. Duni z, C ys EngComm, 2003, 5, 506. 62 Pha maceu ical Sal s and Co-c ys als, J. Wou e s und L. Qué é, Royal Socie y o Chemis y, 2011. Li e a u e zeichnis 54 170 J. Thun, L. Sey a h, C. Bu e ho , J. Senke , R. E. Dinnebie und J. B eu, C ys . G ow h Des., 2009, 9, 2435-2441. 171 G. A. Sim, J. M. Robe son und T. H. Goodwin, Ac a C ys ., 1955, 8, 157-164. 172 D. A. Pla ne , W. Pe e und D. Seebach, Chimia, 1994, 48, 138-141. 173 H. J. Flamme sheim, K is . Tech., 1974, 9, 299-311. 174 H. J. Flamme sheim, K is . Tech., 1974, 9, 313-326. 175 H. J. Flamme sheim, J. The m. Anal., 1975, 7, 571-585. 176 Siche hei sda enbla gemäß Ve o dnung (EG) N . 1907/2006: h p://www.ca l o h.com/media/_de-de/sdpd /P738.PDF 177 Siche hei sda enbla gemäß Ve o dnung (EG) N . 1907/2006: h p://www.ca l o h.com/media/_de-de/sdpd /8548.PDF 178 www.chem.unep.ch/i p c/sids/oecdsids/BENZOATES.pd 179 A. V. T ask, N. Shan, W. D. S. Mo he well, W. Jones, S. H. Feng, R. B. H. Tan und K. J. Ca pen e , Chem. Commun., 2005, 880-882. 180 A. V. T ask, W. D. S. Mo he well und W. Jones, Chem. Commun., 2004, 890- 891. 181 C ys alliza ion P ocesses: Molecula aspec s o he polymo phic c ys alliza ion o amino acids and lipids, M. Ki amu a, S. Ueno und K. Sa o, Wiley Se ies in Solu ion Chemis y, 1998. 182 B. Ci alle i, C. M. Zico ich-Wilson, L. Valenzano und P. Ugliengo, C ys EngComm, 2008, 10, 405-410. Danksagung 55 Danksagung Zue s möch e ich meinem Be eue P o . D . Jose B eu ü die Be ei s ellung des in e essan en Themas, de seh gu en Auss a ung am Leh s uhl, de seh gu en Be- euung sowie den o in ensi en und meis ziel üh enden Diskussionen danken. Danke ü das mi s e s en gegengeb ach e Ve auen. D . Wol gang Milius danke ich ü seine iel äl ige Un e s ü zung. E wa imme o en und gleich da, als ich Hil e benö ig habe, gleich ob es um F agen chemische ode k is allog aphische A ging. Ich danke He n D . Jü gen Thun ü seine s e ige O enhei und Diskussionsbe ei - scha meinen F agen hinsich lich Polymo phie und K is allisa ion gegenübe . Dank all meinen Labo kollegen Lena Geiling, Michael Sch emp , Thomas Ma in und Kilian Bä winkel he sch e im BZKG Labo meis eine locke e S immung und gu e Laune. De abwechslungs eiche BZKG Spo wi d mi s e s in E inne ung blei- ben. Fü die Un e s ü zung und g oßa ige Diskussionsbe ei scha in wissenscha lichen F agen sowie wei da übe hinaus möch e ich mich beim ganzen Leh s uhl de AC I, auch insbesonde e bei Sebas ian Koch, Jose Hausne , D . Wol gang Milius, Ma - hias S ö e , D . Hussein Kalo, Thomas Ma in, Be nd Pu z, Lena Geiling und Ma kus He ling. Wei e hin danke ich allen echnischen Anges ell en Lena Geiling, Be nd Pu z, Bea e Boje und Die e Will ü die du chge üh en Messungen, sowie auch den Sek e ä- innen Pe a Seidle und I is Rai hel ü die Hil e mi den zahllosen Fo mula en und ielem meh . D . Ma kus D echsle möch e ich ü die du chge üh en TEM Messungen danken. Wei e hin danke ich den Mechanik und Elek onik We ks ä en de Uni e si ä Bay- eu h ü ih en Einsa z und den Bau on Ge ä en und um das Thema C yo Spin Coa ing. Danksagung 56 Eine g oße Un e s ü zung in den einzelnen P ojek en wa en die s e s diskussions- eudigen und engagie en Koope a ionspa ne : Uni e si ä Bay eu h, Ano ganische Chemie III (P o . D . Jü gen Senke , Paul Niemie z und Kilian Bä winkel) ü NMR Messungen und ieles meh . Ch is ian-Alb ech s-Uni e si ä zu Kiel, Ano ganische Fes kö pe chemie I (P o . D . Wol gang Bensch und D . Nicole Pienack) ü die Vo be ei ung und Hil e bei den Messungen am DESY. Johannes Gu enbe g-Uni e si ä Mainz, Zen um ü hochau lösende Elek onenmik- oskopie (D . U e Kolb und I yna And usenko) ü die du chge üh e ADT Elek o- nenbeugung. Ch is ian-Alb ech s-Uni e si ä zu Kiel, Fes kö pe - und Koo dina ionschemie (P o . D . Ch is ian Nä he ) ü die Du ch üh ung und Diskussion de Messungen um Benzamid Fo m III. Uni e si ä Pade bo n, Physikalische Chemie (P o . D . Klaus Hube und D . Todo Hiko ) ü die Messung und Diskussion de Lich s euexpe imen e. Ka ls uhe Ins i u ü Technologie (P o . D . Michael Tü k, D . Eugenia B eininge , Sab ina Mülle und Nina Teubne ) ü die Du ch üh ung on RESS Expe imen en. F ied ich-Alexande -Uni e si ä E langen-Nü nbe g (P o . D . Di k Zahn und Phi- lipp Ec o s) ü die Compu e simula ionen und de en Diskussion. Eine gu e Un e s ü zung wa en auch alle Mi a bei e p ak ikan en (Inna Dewald, Pia Ruckdeschel, Janina Laue , Liyao Wang, Julia S öckl, Ch is ina Kuhn, Ka ha ina Jö g, Na ascha Weiß, Ma in Völkel und Anja Wunde ). Meinen El e n und B i a gil mein g öß e Dank. Meinen El e n danke ich zunächs ü die Un e s ü zung wäh end de gesam en Ausbildung und dank B i a konn en auch die schwie igs en Phasen in de Dok o a bei mühelos übe s anden we den. Anhang 57 Anhang Anhang A.1 Co-c ys allisa ion o benzoic acid wi h sodium benzoa e: he signi icance o s oi- chiome y C. Bu e ho , W. Milius und J. B eu, C ys EngComm, 2012, 14, 3945-3950. - Rep o- duced by pe mission o The Royal Socie y o Chemis y Anhang 58 Co-c ys allisa ion o benzoic acid wi h sodium benzoa e: he signi icance o s oichiome y{ Ch is ian Bu e ho , Wol gang Milius and Jose B eu* Recei ed 9 h Feb ua y 2012, Accep ed 16 h Ma ch 2012 DOI: 10.1039/c2ce25185j Cu en ly he syn hesis o co-c ys als is ecei ing conside able a en ion. We s udied co-c ys allisa ion o benzoic acid wi h sodium benzoa e in di e en a ios and wi h di e en c ys allisa ion echniques. A c ys alline co-c ys al, composed o wo equi alen s o benzoic acid and one equi alen o sodium benzoa e (2 HBz?1 NaBz) could be iden i ied and s uc u ally cha ac e ised. By concomi an coo dina ion o benzoa e and benzoic acid he coo dina ion needs o he ca ion could be sa is ied. Howe e , e en wi h a 2 : 1 a io, an oc ahed al en i onmen could only be ealised by sha ing an edge in a dime . The dime s a e connec ed o one-dimensional apes which in u n a e packed in a dis o ed hexagonal a angemen . Only wi h a a io o wo neu al benzoic acid ligands and one benzoa e can he coo dina ion equi emen s o he sodium ca ion and cha ge neu ali y be assu ed a he same ime. In oduc ion In he ield o c ys al enginee ing, co-c ys als ha e ecei ed inc easing a en ion o e he las decade. In pa icula , co- c ys als o ac i e pha maceu ical ing edien s wi h small o ganic molecules a e o in e es . 1–4 Physical p ope ies like solubili y, dissolu ion a e, mel ing poin , mois u e so p ion endency and comp essibili y a ec he bioa ailabili y, design, manu ac u ing, and he s abili y o he esul an dosage o m. Despi e high in e es in co-c ys als, he numbe o epo ed and cha ac e ised co-c ys als is a he low (0.45%) in compa ison wi h ‘‘no mal’’ one-componen molecula c ys als as poin ed ou ecen ly by Shan and Zawo o ko. 3 Along wi h he inc easing in e es in co- c ys allisa ion a deba e a ose in he li e a u e ega ding he e m co-c ys al i sel . 5–8 I is now commonly accep ed ha co-c ys als a e buil up by a leas wo componen s which a e solid unde ambien condi ions. Some au ho s es ic he second componen o neu al molecules bu o he s include sys ems whe e acid–base eac ions occu be ween he cons i uen s, he so called ‘‘sal -co- c ys al con inuum’’. 9 Fu he mo e some ionic co-c ys al sys ems ha e been cha ac e ised. 10,11 Di e en app oaches ha e been aken o syn hesise co- c ys als. A popula ou e o inco po a e bo h componen s in o a single phase is slow e apo a ion om solu ion which con ains s oichiome ic amoun s o he so-called co-c ys al o me s. Mechanochemical echniques also p o ed sui able; he ein p obably he easies and mos equen ly applied me hod is he sol en -d op-g inding echnique. 12,13 We in es iga ed he co-c ys allisa ion o benzoic acid (HBz) and i s co esponding sodium sal (NaBz) in di e en a ios and wi h di e en c ys allisa ion me hods. Bo h species a e widely used as p ese a i es in he ood indus y. 14 In e es ingly, his is a sys em whe e ac ually he co-c ys als a e easily accessible while he pu e sal , NaBz, does no c ys allise well. The e o e, un il now he c ys al s uc u e o NaBz could no be sol ed and e ined, despi e he a he simple molecula s uc u e. Sui able single c ys als o c ys al X- ay c ys allog aphy could no be ob ained. Mo eo e , he powde di ac ion pa e n (Fig. 7) is no well esol ed and a he ea u eless, indica ing massi e s uc u al diso de . Addi ionally, he NaBz s uc u e seems o lack h ee-dimensional ansla ional symme y as indica ed by a ea u eless PXRD ace o NaBz, which hampe s indexing. One eason o he e a ded c ys allisa ion beha iou may be ela ed o he simple 1 : 1 s oichiome y. In o de o ealise he common 6- old coo dina ion o sodium, he coo dina ion polyhed a would ha e o be connec ed in pa by sha ed edges, e en i he ca boxyla e g oup would ac as biden a e ligand. Un o una ely, he la ge a io o he olume o he benzoa e anion compa ed wi h he sodium ca ion 15 makes edge sha ing di icul . In any case, o NaBz indexing and hence s uc u e solu ion p o ed impossible up o now. A hyd a e o NaBz is also epo ed bu lacks s uc u al cha ac e isa ion p obably o he same ea- sons. 16 Ano he in e es ing ea u e o NaBz is he o ma ion o a smec ic liquid c ys al phase upon hea ing abo e 430 uC. 17 Con a y o NaBz, he c ys al s uc u e o HBz was de e mined o e 55 yea s ago. 18 In his c ys al s uc u e hyd ogen bonded HBz dime s can be ound. We we e in e es ed in he co-c ys allisa ion o bo h NaBz and HBz and pa icula ly in s udying he modes by which sodium sa is ies i s coo dina ion needs and how his changes he hyd ogen bonding mo i as compa ed o pu e HBz. Ino ganic Chemis y I, Uni e si y o Bay eu h, Uni e si a¨ ss abe 30, 95440 Bay eu h, Ge many. E-mail: [email p o ec ed]; Fax: +49 921 55-2788; Tel: +49 921 55-2530 {CCDC e e ence numbe 865834. Fo c ys allog aphic da a in CIF o o he elec onic o ma see DOI: 10.1039/c2ce25185j C ys EngComm Dynamic A icle Links Ci e his: C ys EngComm, 2012, 14, 3945–3950 www. sc.o g/c ys engcomm PAPER This jou nal is ßThe Royal Socie y o Chemis y 2012 C ys EngComm, 2012, 14, 3945–3950 |3945 Downloaded by UNIVERSITAT BAYREUTH on 05 Oc obe 2012 Published on 20 Ap il 2012 on h p://pubs. sc.o g | doi:10.1039/C2CE25185J View Online / Jou nal Homepage / Table o Con en s o his issue The benzoic acid–sodium benzoa e co-c ys al sys em In he yea 1974 Flamme sheim de e mined he physicochemical p ope ies o he sys em sodium benzoa e–benzoic acid. 19 Th ee ‘‘complexes’’ – as he named i a ha ime – be ween NaBz and HBz we e desc ibed and cha ac e ised by powde X- ay di ac ion (PXRD), X- ay single c ys al di ac ion, in a ed (IR) spec oscopy and di e en ial scanning calo ime y (DSC) measu emen s. The composi ion o he i s ‘‘complex’’ was de e mined o be 2 HBz?1 NaBz. I could be c ys allised om solu ion as well as by ibochemical c ys allisa ion me hods. Fo easons no known, un o una ely, Flamme sheim ailed o de e mine he c ys al s uc u e, 19 despi e success ul indexing (a= 22.29(7) A ˚;b= 14.55(5) A ˚;c= 5.89(2) A ˚;b=98u, space g oup o P2 1 /n). Fu he mo e, a ansi ion o his ‘‘2 : 1-complex’’ o a ‘‘high empe a u e modi ica ion’’ was desc ibed. This phase ansi ion was claimed o occu upon hea ing he c ys als o 120 uC. Finally, Flamme sheim also epo ed a second ‘‘com- plex’’ wi h he composi ion 1 HBz?2 NaBz. In a ecen a icle B i ain desc ibed he co-c ys allisa ion o HBz wi h NaBz in a 1 : 1 s oichiome ic a io. 20 The cha ac e - isa ion o he 1 : 1 s oichiome ic p oduc was pe o med by PXRD, DSC, Fou ie - ans o m in a ed (FTIR) and Raman spec oscopy. PXRD, howe e , was only in e p e ed in a ‘‘quali a i e way’’ by compa ing pa e ns, bu no indexing o e en s uc u e solu ion was o e ed. Resul s and discussion We in es iga ed he co-c ys allisa ion o HBz wi h NaBz by sys ema ically a ying he a ios, c ys allisa ion me hods and sol en s. Bo h sol en e apo a ion c ys allisa ion as well as he popula me hod o sol en -d op-g inding we e pe o med. We used wo di e en sol en s, d y me hanol and e hanol–wa e (4 : 1/ : ). In he case o me hanol he g inding p ocedu e was pe o med wi hin a glo e box unde d y ni ogen a mosphe e o a oid he o ma ion o possible hyd a es ( o de ails see expe imen al sec ion). Sol en e apo a ion c ys allisa ion yielded single c ys als o good quali y wi h s oichiome y 2 HBz?1 NaBz. While sol en -d op-g inding in pu e wa e yielded he same co-c ys al, g inding had o be inc eased signi ican ly o mo e han 60 min o comple e con e sion. Sho e g inding imes ga e mix u es o HBz and NaBz oge he wi h he 2 HBz?1 NaBz co-c ys al. Mos likely, he eason o he slowe con e sion is he poo solubili y o HBz in wa e (y2.85 g l 21 a 20 uC). 21 The co-c ys al can e en be ob ained ibochemically wi hou sol en as Flamme sheim showed. 19 C ys al s uc u e o he 2 HBz?1 NaBz co-c ys al A co-c ys al o he composi ion 2 HBz?1 NaBz ( wo neu al benzoic acid molecules pe Na-benzoa e) is o med in solu ion as well as by mechanochemical sol en d op g inding (see sol en - d op-g inding sec ion). The la ges and bes single c ys als could be g own by slowly e apo a ing a solu ion o an e hanol–wa e (4 : 1/ : ) mix u e, which con ained HBz and NaBz in a 2 : 1 s oichiome ic a io. C ys al s uc u e solu ion and e inemen u ned ou o be s aigh o wa d. Table 1 shows he c ys al- log aphic da a o he 2 HBz?1 NaBz co-c ys al. A compa ison o he cell pa ame e s epo ed by Flamme sheim only shows small de ia ions o he uni cell pa ame e s (see Table 2). The e o e, his s uc u e could be assigned as he same c ys al- lisa ion p oduc which he desc ibed as he ‘‘low- empe a u e complex’’ o NaBz wi h HBz. Desc ip ion o he c ys al s uc u e o 2 HBz?1 NaBz. The asymme ic uni o he 2 HBz?1 NaBz co-c ys al is shown in Fig. 1. The packing and i s cons uc ion om building uni s is shown in Fig. 2. The s uc u e is composed o one-dimensional apes (Fig. 2a), which a e packed in an app oxima ely hexagonal a ay. Be ween hese apes an de Waals o ces and one p– p-s acking in e ac ion a e he main in e ac ion o ces. The p–p in e ac ion dis ance be ween adjacen symme y equi alen phenyl ings (C(16)–C(21)) is 3.98(6) A ˚. A side iew on one o he apes shows ha sodium esides, as expec ed, in a dis o ed oc ahed al coo dina ion (see Fig. 2(b), Fig. 3 and Fig. 4). A space illing model (Fig. 2(c)) highligh s ha he packing in he apes is dense. No space is a ailable o a ange addi ional ligands a ound he Na-moie ies. Wi hin he apes, dime s o edge- sha ing oc ahed a become appa en . All ca boxylic (HBz, o ange) and ca boxyla e (NaBz, ed) g oups ac as monoden a e ligands in espec o a gi en dime . Ca boxyla e g oups o wo benzoa e anions a e in ol ed in he sha ed edge (Fig. 3). O he h ee HBz molecules in an oc ahed on, one is coo dina ed wi h i s hyd oxyl (O(2)), he o he wo ia hei ca bonyl oxygens (O(1), O(5)). Eigh ou o en ligands in ol ed in he coo dina ion sphe e o a dime connec he dime in bo h Table 1 C ys allog aphic da a o he co-c ys al 2 HBz?1 NaBz Fo mula C 21 H 17 NaO 6 Fo mula weigh 388.34 T/K 293(2) C ys al sys em Monoclinic Space g oup P2 1 /c a/A ˚5.8941(12) b/A ˚14.565(3) c/A ˚22.299(5) a(u)90 b(u) 97.30(3) c(u)90 V/A ˚˚1898.8(7) Z4 D c /g cm 23 1.358 m/mm 21 0.119 R in 0.0665 Re ln (all/ind) 3511/1867 R 1 /wR 2 (obsd da a: F 2 .2s(F 2 )) a 0.0374/0.0840 R 1 /wR 2 (all da a) a 0.0791/0.0947 La ges esidual/e A ˚ 23 0.172 a R1~P(F0{ jj Fckk )=PF0 jj ;wR2~½Pw(F0 jj {Fc jj )2=Pw(F0)21=2 Table 2 Compa ison o he cell pa ame e s wi h he published alues This wo k Flamme sheim 19 C ys al sys em Monoclinic Monoclinic Space g oup P2 1 /cP2 1 /n a/A ˚5.8941(12) 22.29(7) b/A ˚14.565(3) 14.55(5) c/A ˚22.299(5) 5.89(2) a(u)9090 b(u) 97.30(3) 98 c(u)9090 3946 |C ys EngComm, 2012, 14, 3945–3950 This jou nal is ßThe Royal Socie y o Chemis y 2012 Downloaded by UNIVERSITAT BAYREUTH on 05 Oc obe 2012 Published on 20 Ap il 2012 on h p://pubs. sc.o g | doi:10.1039/C2CE25185J View Online di ec ions o a one-dimensional ape ia b idging ca boxylic/ ca boxyla e g oups (Fig. 4). The wo emaining ca boxylic ligands ealise a H-bond be ween O(6) and O(4) (Fig. 4 and Table 3). This in e -dime bond ein o ces he dime . A second in a-dime H-bonding be ween O(2) and O(4) (Fig. 4 and Table 3) con ibu es o he o ma ion o he apes. Adjacen phenyl g oups wi hin he apes a e in ol ed in shi ed p-s acks (Fig. 2(c)), (in e ac ion be ween phenyl ings o (C(2)–C(7)) and (C(9)–C(14))). The p–ps acking dis ance is 3.87(3) A ˚. Sol en -d op-g inding Wi h a a io o 2 : 1, a phase pu e ma e ial was ob ained as indica ed by he PXRD ace (Fig. 5). In he p esence o bo h sol en s, d y me hanol and e hanol–wa e (4 : 1/ : ), co- c ys als o 2 HBz?1 NaBz we e ob ained as a pu e phase o bo h sol en e apo a ion and sol en d op g inding. In he li e a u e co-c ys als o a ios di e ging om 2 : 1 ha e been epo ed, e.g. 1 : 1. Such NaBz ich composi ions would only be able o sa is y he coo dina ion needs o sodium i he deg ee o condensa ion o oc ahed a could inc ease beyond he le el o edge sha ing dime s obse ed in 2 HBz?1 NaBz. Since he packing o ligands a ound he sodium ca ion is, howe e , al eady dense wi h 2 HBz?1 NaBz (Fig. 2(c)), i migh appea ha a pe iodic, c ys alline molecula packing o he ligands migh be di icul o ealise. Ne e heless we explo ed bo h NaBz ich and poo s oichiome ies by sol en -d op-g inding c ys allisa ions. The expe imen al de ails a e gi en in he expe imen al sec ion. NaBz poo s oichiome ies: HBz : NaBz = 3 : 1. Independen ly o he wo sol en s applied, d y me hanol and e hanol–wa e (4 : 1/ : ), a mix u e o he co-c ys al 2 HBz?1 NaBz and c ys alline HBz was ob ained. The PXRD ace obse ed is a simple o e lay (Fig. 6). Despi e he massi e o e lap, ee HBz can be clea ly iden i ied by a ew unique e lec ions, he mos p ominen appea ed a Fig. 1 ORTEP plo and he c ys allog aphic numbe ing scheme o he asymme ic uni o he co-c ys al 2 HBz?1 NaBz. Displacemen ellipsoids a e d awn a he 50% p obabili y le el. Fig. 2 Molecula packing in he 2 HBz?1 NaBz co-c ys al o HBz wi h NaBz. (a) App oxima ely hexagonal packing o one-dimensional apes along he a-axis. (b) Side iew on one o he apes. Coo dina ion bonds a e black and hyd ogen bonds blue dashed. (c) Space illing model o one ape highligh ing he dense packing a ound sodium. This jou nal is ßThe Royal Socie y o Chemis y 2012 C ys EngComm, 2012, 14, 3945–3950 |3947 Downloaded by UNIVERSITAT BAYREUTH on 05 Oc obe 2012 Published on 20 Ap il 2012 on h p://pubs. sc.o g | doi:10.1039/C2CE25185J View Online 27.79u2H, (3.21 A ˚). All e lec ions a e sha p and can be indexed by ei he 2 HBz?1 NaBz o c ys alline HBz. NaBz ich s oichiome ies: HBz : NaBz = 1 : 1 o 1 : 5. Up o a s oichiome ic a io o 1 : 3, e lec ions o he co-c ys al 2 HBz?1 NaBz a e clea ly isible. Addi ional week e lec ions (e.g. 3.75u 2H, (23.55 A ˚); 6.15u2H, (14.36 A ˚); 16.68u2H, (5.31 A ˚); 18.45u 2H, (4.80 A ˚); 22.80u2H, (3.90 A ˚), as e isks in Fig. 7) appea which a e mos in ense a a a io o 1 : 2. Despi e conside able e o , we ailed o index hese addi ional e lec ions. The pa e n o hese addi ional e lec ions di e s om he PXRD ace o NaBz p epa ed in a simila way, howe e , he mos in ense peak is iden ical. Since a a a io o 1 : 2, 2 HBz?1 NaBz is s ill clea ly p esen , he esidual ma e ial mus ha e a composi ion which is e en iche in NaBz han 1 : 2 and i would be expec ed ha he in ensi y o hese addi ional e lec ions should u he inc ease wi h a ios lowe han 1 : 2. Howe e , con a y o his expec a ion, he in ensi y o he addi ional e lec ions ins ead dec eases wi h u he inc easing NaBz con en . The PXRD ace o he 1 : 3 a io is al eady a he ea u eless, wi h a u he dec easing a io he pa e n g adually blends in o he pu e NaBz ace in a kind o me amo phic ansi ion. Ano he in e es ing ea u e is he g adual shi o he i s e lec ion a 3.75u2H o highe d- alues wi h dec easing HBz : NaBz a io. Such g adual shi s a e well known o andomly in e s a i ied in e cala ion compounds. Fu he mo e, he ea u eless appea ance o he PXRD ace o NaBz, which hampe s indexing, migh also be due o diso de . Al hough he exis ence o a c ys alline phase wi h a a io smalle han 2 : 1 can no de ini ely be uled ou a his s age, he obse a ions desc ibed mos likely poin o a semic ys alline Fig. 3 Coo dina ion sphe e o sodium in he co-c ys al 2 HBz?1 NaBz. The Na–O dis ances show he s ongly dis o ed oc ahed al coo dina- ion. Oxygen a oms belonging o ca boxylic g oups (HBz) a e in o ange, while oxygen a oms belonging o ca boxyla e g oups (NaBz) a e in ed. Fig. 4 O e iew o he connec ion o he edge sha ing oc ahed on pai s. The exac connec ion is desc ibed in he ex . Fo cla i y, phenyl ings o HBz and NaBz a e no shown. Hyd ogen bonds a e blue dashed. Table 3 Summa y o in e molecula in e ac ions (D–H…A; A ˚,u) ope a ing in he c ys al s uc u e o 2 HBz?1 NaBz DH AH …A/A ˚D…A/A ˚ (DHA)/uSymme y ope a ion O(2) H(2) O(4) 1.77 2.5840(19) 174.7 2x, 2y+1,2z+1 O(2) H(2) O(3) 2.59 3.147(2) 126.0 2x, 2y+1,2z+1 O(6) H(6A) O(4) 1.86 2.683(2) 176.6 2x, 2y+1,2z+1 Fig. 5 Powde di ac ion pa e n o he co-c ys al 2 HBz?1 NaBz. The g een ace displays he simula ed single c ys al di ac ion pa e n as a e e ence (calcula ed wi h FWHM = 0.2u2 The a). Fig. 6 PXRD aces o p oduc s c ys allised by sol en d op g inding om e hanol–wa e (4 : 1/ : ) (blue ace: HBz d op g inded wi h e hanol–wa e (4 : 1/ : ), black ace: HBz : NaBz = 3 : 1, o ange ace: HBz : NaBz = 2 : 1). 3948 |C ys EngComm, 2012, 14, 3945–3950 This jou nal is ßThe Royal Socie y o Chemis y 2012 Downloaded by UNIVERSITAT BAYREUTH on 05 Oc obe 2012 Published on 20 Ap il 2012 on h p://pubs. sc.o g | doi:10.1039/C2CE25185J View Online ma e ial. We p opose ha NaBz as ob ained by d op g inding is composed o low dimensional building uni s which a e packed in a diso de ed way and which migh be capable o in e cala ing a ious (non-s oichiome ic) amoun s o neu al HBz molecules. Expe imen al NaBz (pu i y ¢99.0%) and HBz (pu i y .99.5%) we e pu chased om AppliChem. E hanol was sou ced om VWR (BDH P olabo) and me hanol (H 2 O,0.01%) om Sigma- Ald ich. Me hanol was d ied o e molecula sie es, while all o he chemicals and sol en s we e used wi hou u he pu i ica ion. Dis illed wa e was used. PXRD aces we e eco ded using a STOE STADI P (Cu Ka1 adia ion, ansmission geome y) di ac ome e . The samples we e illed in capilla ies (diame e 0.5 mm). Single-c ys al X- ay di ac ion da a we e collec ed using a STOE IPDS I ins umen (293 K, Mo Ka adia ion). Selec ed c ys allog aphic da a a e lis ed in Table 1. The c ys al s uc u e was sol ed and e ined using SHELXTL 5.1 (B uke AXS). All igu es we e d awn wi h he DIAMOND p og amme. C ys allisa ion om solu ion Single c ys als we e ob ained om a supe sa u a ed solu ion o 4.885 g (40.00 mmol) HBz and 2.882 g (20.00 mmol) NaBz in 35 ml e hanol–wa e mix u e (4 : 1/ : ). This mix u e was hea ed up o 50 uC, un il all s a ing ma e ial was dissol ed. A e wa ds he solu ion was cooled e y slowly (o e 3 h) o oom empe a u e. Then he solu ion was le a oom empe a u e. A e h ee days spicula , colou less c ys als could be collec ed. These we e il e ed o he mo he liquid and d ied a oom empe a u e. All o he expe imen s we e pe o med in he same way by using di e en a ios o HBz and NaBz (1 : 1; 2 : 1; 1 : 2) and by using he e hanol–wa e mix u e (4 : 1/ : ) o me hanol as sol en . T ibochemical c ys allisa ion The popula me hod o sol en -d op-g inding wi h a ypical hand-mo a was applied. Di e en a ios o HBz and NaBz (3 : 1; 2 : 1; 1 : 1; 1 : 2; 1 : 3; 1 : 4; 1 : 5) ( o al quan i ies: 1500 mg) we e we ed wi h he co esponding sol en (e hanol– wa e mix u e (4 : 1/ : ) o me hanol) and hen g ound o d yness. In he case o me hanol he g inding p ocedu e was pe o med wi hin a d y ni ogen a mosphe e. The capilla ies o PXRD we e illed wi hin a glo e box o a oid con ac wi h ai humidi y. Conclusions Co-c ys allisa ion expe imen s o HBz wi h i s co esponding sodium sal we e pe o med wi h he help o di e en c ys al- lisa ion echniques. The ionic co-c ys al 2 HBz?1 NaBz was c ys allised by di e en me hods ( ibochemical and solu ion) and i s c ys al s uc u e was de e mined by single-c ys al s uc u e e inemen . By concomi an coo dina ion o benzoa e and HBz, he coo dina ion needs o he ca ion could be sa is ied. Howe e , e en wi h a 2 : 1 a io, an oc ahed al en i onmen could only be ealised by sha ing an edge in a dime . The dime s a e connec ed o one-dimensional apes which in u n a e packed in a dis o ed hexagonal a angemen . Re e ences 1 O. Alma sson and M. J. Zawo o ko, Chem. Commun.,2004, 1889–1896. 2Polymo phism in he Pha maceu ical Indus y, R. Hil ike , Wiley- VCH Ve lag, Weinheim, Ge many, 2006. 3 N. Shan and M. J. Zawo o ko, D ug Disco e y Today, 2008, 13, 440–446. 4 P. Vishweshwa , J. A. McMahon, J. A. Bis and M. J. Zawo o ko, J. Pha m. Sci., 2006, 95, 499–516. Fig. 7 PXRD aces o p oduc s c ys allised by sol en d op g inding om e hanol–wa e (4 : 1/ : ). (b own ace: pu e HBz, o ange ace: HBz : NaBz = 2 : 1, ed ace: HBz : NaBz = 1 : 1, pu ple ace: HBz : NaBz = 1 : 2, da k g een ace: HBz : NaBz = 1 : 3, blue ace: HBz : NaBz = 1 : 4, ligh g een ace: HBz : NaBz = 1 : 5, black ace: pu e NaBz). The inse shows a magni ica ion o he ange om 2u2H o 9u2H. The disappea ance o peaks ela ed o he 2 HBz?1 NaBz co- c ys al wi h dec easing HBz : NaBz a io can be moni o ed. This jou nal is ßThe Royal Socie y o Chemis y 2012 C ys EngComm, 2012, 14, 3945–3950 |3949 Downloaded by UNIVERSITAT BAYREUTH on 05 Oc obe 2012 Published on 20 Ap il 2012 on h p://pubs. sc.o g | doi:10.1039/C2CE25185J View Online Solu ion media ed ans o ma ion. To expe imen ally de e mine he he modynamic anking a oom empe a u e, we illed a 1 : 1 a io o bo h o ms in o a capilla y, which was hen soaked wi h e hanol, sealed, and moun ed on a powde di ac ome e equipped wi h a as and high esolu ion de ec o . Wi hin 40 min o m B is comple ely ans o med in o o m A (Fig. 7) a oom empe a u e which unequi ocally p o es o m B o be me as able a his empe a u e. This anking is in ag eemen wi h he densi y ule, since bo h expe imen al and heo e ical densi ies (DFT-D) o o m A a e highe han he densi ies o o m B (A: 1.387 g cm 23 (173 K) and 1.450 g cm 23 (0 K, DFT-D); B: 1.326 g cm 23 (173 K) and 1.386 g cm 23 (0 K, DFT-D). While he solu ion-media ed ans o ma ion is also in line wi h he o al ene gies ob ained in he DFT-D calcula ions, i deli e s, howe e , no in o ma ion on he na u e o he ans o ma ion (enan io opic o mono opic). Bu ge in oduced se e al ules o de e mine he he mody- namic ela ion in polymo phs. 36 One o he mos impo an ules, he hea -o - usion ule could no be applied because eliable DSC measu emen s a e no easible. Because bo h Fig. 4 Coo dina ion sphe es o sodium in o m A and B o he co-c ys al 2 HBz?1 NaBz. Oxygen a oms belonging o ca boxylic g oups (HBz) a e in o ange, while oxygen a oms belonging o ca boxyla e g oups (NaBz) a e in ed. The Na–O dis ances show he s ongly dis o ed oc ahed al coo dina ion in bo h o ms (bo h measu ed a 173 K). The dis ances in o m B a e longe compa ed o he he modynamically s able o m A a oom empe a u e. Fig. 5 Compa ison o he connec ion o oc ahed a in o m A and B o he co-c ys al 2 HBz?1 NaBz. Fo cla i y easons, phenyl ings o HBz and NaBz a e no shown. Hyd ogen bonds a e dashed in blue. (a) One- dimensional ape o edge-sha ing oc ahed a in o m A. (b) In ini e ods o edge-sha ing oc ahed a in o m B. Table 2 Summa y o in e molecula in e ac ions (D–H…A; A ˚,u) ope a ing in he c ys al s uc u e o o m B o he co-c ys al 2 HBz?1 NaBz DHAH …A/A ˚D…A/A ˚ (DHA)/uSymme y ope a ion O(2) H(2) O(3) 1.73 2.551(4) 174.7 x,y,z O(5) H(5) O(4) 1.74 2.558(5) 178.7 x+1,y,z Fig. 6 IR spec oscopy o he wo polymo phs o he co-c ys al 2 HBz?1 NaBz. The black ace co esponds o o m A and he ed ace o he me as able o m B. Because o he di e en packing he di e ences a e ob ious. Table 3 Compa ison o he o sional angles (u) in o m A and B Fo m A Fo m B HBz 1 6.18 (O1–C1–C2–C7) 0.45 (O2–C1–C2–C7) 8.30 (O2–C1–C2–C3) 2.19 (O1–C1–C2–C3) HBz 2 2.54 (O6–C15–C16–C17) 14.67 (O5–C15–C16–C17) 6.53 (O5–C15–C16–C21) 18.13 (O6–C15–C16–C21) Bz 2 11.83 (O3–C8–C9–C14) 27.52 (O4–C8–C9–C14) 12.14 (O4–C8–C9–C10) 29.40 (O3–C8–C9–C10) This jou nal is ßThe Royal Socie y o Chemis y 2012 C ys EngComm, 2012, 14, 6744–6749 |6747 Downloaded by UNIVERSITAT BAYREUTH on 05 Oc obe 2012 Published on 09 Augus 2012 on h p://pubs. sc.o g | doi:10.1039/C2CE25562F View Online polymo phs eadily loose benzoic acid upon hea ing we e ained om de e mining mel ing en halpies. Flamme sheim e e ed o a ‘‘high- empe a u e’’ modi ica ion in his wo k. 37 He claimed ha 2 HBz?1 NaBz o o m A unde goes a phase ansi ion in o a ‘‘high- empe a u e’’ o m upon hea ing he c ys als o 120 uC. Bu he also men ioned se e e p oblems caused by sublima ion. Fo his eason we a emp ed a solid–solid phase ansi ion o a la ge c ys al o o m A in a gas- igh sealed qua z-capilla y. The capilla y was annealed o one day a 110 uC. Fig. 8 shows ligh mic oscopy images o a single c ys al be o e and a e he annealing. PXRD o he annealed c ys al indica es a comple e ans o ma ion o o m A in o o m B. As sugges ed by he opaqueness o he annealed c ys al, he ans o ma ion did no occu single c ys al o single c ys al bu a mic oc ys alline powde was ob ained. Phase pu i y was e i ied by pe o ming a Pawley e inemen using TOPAS 38 o he uni cell pa ame e s (Fig. 9). Thus, he he modynamic ela ion o o m A and B can be assigned as enan io opic. Compa ing a lis o d- alues published by Flamme sheim o he ‘‘high- empe a u e’’ mod- i ica ion (Table 5), his modi ica ion appea s o be iden ical o o m B. Expe imen al NaBz (pu i y ¢99.0%) and HBz (pu i y .99.5%) we e pu chased om AppliChem. Me hanol was sou ced om Sigma-Ald ich. All chemicals and sol en s we e used wi hou u he pu i ica ion. PXRD aces we e eco ded using a STOE STADI P (CuK a1 adia ion, ansmission geome y) di ac ome e equipped wi h a DECTRIS My hen 1 K silicon s ip de ec o . The samples we e illed in capilla ies (diame e 0.5 mm and 2.0 mm). Single-c ys al X- ay di ac ion da a we e collec ed using a STOE IPDS I ins umen (293 K, MoK a adia ion). Selec ed c ys allog aphic da a a e lis ed in Table 1. The c ys al s uc u e was sol ed and e ined using SHELXTL 5.1 (B uke AXS). All igu es we e d awn wi h he DIAMOND p og amme. IR spec oscopy was pe o med Table 4 Summa y o he CASTEP geome y op imisa ion. The de ia- ion o he op imised s uc u e pa ame e s as compa ed o he expe imen ally de e mined c ys al s uc u es a 173 K a e gi en in b acke s Fo m La ice pa ame e s DE(kJ mol 21 ) Aa/A ˚5.770 (21.21%) 0 b/A ˚14.331 (21.16%) c/A ˚21.621 (22.33%) a(u) 90.00 (0.0%) b(u) 95.85 (21.37%) c(u) 90.00 (0.0%) V/A ˚ 3 1778.5 Ba/A ˚6.912 (0.30%) +0.08 b/A ˚11.624 (23.79%) c/A ˚13.169 (21.52%) a(u) 107.80 (21.98%) b(u) 99.98 (20.53%) c(u) 105.82 (1.87%) V/A ˚ 3 930.5 Fig. 7 Solu ion media ed ans o ma ion o a mix u e o o m A and B a oom empe a u e (slu y wi h e hanol). The ed ace co esponds o pu e o m B and he g een ace o pu e o m A. The black aces co espond o he slu y o he mix u e o o m A and B measu ed a e di e en ime in e als. Fig. 8 Ligh mic oscopy images o he enan io opic phase ansi ion o he co-c ys al 2 HBz?1 NaBz. On he le side a la ge single-c ys al o o m A, which is he he modynamically s able o m a oom empe a u e, was pu in o a capilla y. The capilla y was sealed and hea ed o 110 uC o one day. A e hea ing he la ge single c ys al o o m A ans o med in o many small mic o-c ys alli es. These can be assigned as c ys alli es o o m B by powde di ac ion. Fig. 9 Pawley e inemen (applying TOPAS 38 ) o uni cell pa ame e s o o m B o he co-c ys al 2 HBz?1 NaBz ob ained by annealing a single c ys al o o m A a 110 uC. All e lec ions can be indexed by he uni cell pa ame e s showing phase pu i y and comple e con e sion o o m A in o B (R wp = 6.248; ze o poin = 20.026(4)u2 The a; a= 6.9065(5) A ˚;b = 12.2522(7) A ˚;c= 13.4431(7) A ˚;a= 111.346(5)u;b= 101.121(5)u;c= 101.953(6)u) 6748 |C ys EngComm, 2012, 14, 6744–6749 This jou nal is ßThe Royal Socie y o Chemis y 2012 Downloaded by UNIVERSITAT BAYREUTH on 05 Oc obe 2012 Published on 09 Augus 2012 on h p://pubs. sc.o g | doi:10.1039/C2CE25562F View Online on a Pe kinElme Spec um 100 FT-IR spec ome e wo king wi h he UATR (uni e sal a enua ed o al e lec ance) echnique. The DFT-D calcula ions we e ca ied ou using he CASTEP code. 39 The gene alised-g adien app oxima ion (GGA) wi h he Pe dew–Bu ke–E nze ho (PBE) unc ional was used. 40 A plane- wa e basis se wi h an ene gy cu o o 900 eV was applied and he co e elec ons we e ep esen ed by pseudopo en ials. Fo he geome y op imisa ion he con e gence ole ances o ene gy, maximum o ce, maximum displacemen we e 2.0 610 25 eV pe a om, 5.0 610 22 eV A ˚ 21 and 1.0 610 23 A ˚, espec i ely. The allowed s ess ole ance was 0.1 GPa. Flash e apo a ion c ys allisa ion Single c ys als we e ob ained om a solu ion o 2.442 g (20.00 mmol) HBz and 1.441 g (10.00 mmol) NaBz in 15 ml me hanol. This mix u e was hea ed o 65 uC, un il all s a ing ma e ial was dissol ed. A e wa ds y3 ml o he ho solu ion we e sp ead on a hea ed (y75 uC) c ys allising dish (Øy15 cm). Wi hin seconds me hanol e apo a es and colou less c ys als can be obse ed (see Fig. 1). Conclusions The he modynamic ela ionship be ween he wo polymo phs o he ionic co-c ys al 2 HBz?1 NaBz was de e mined o be o enan io opic na u e wi h a ans o ma ion empe a u e below 110 uC. A oom empe a u e, o m B is me as able. While o m A is composed o one-dimensional apes, o m B is buil om in ini e ods. The coo dina ion sphe e o sodium, howe e , in bo h co-c ys als is oc ahed al and he packing a ound i is dense. Acknowledgemen s We hank he Deu sche Fo schungsgemeinscha (SPP 1415), o inancial suppo . The au ho s hank Ka l Kemp and P o . D Raine Schobe o measu emen o he IR spec a. We would like o hank an anonymous e e ee o aluable commen s. Re e ences 1 J. Haleblia and W. McC one, J. Pha m. Sci., 1969, 58, 911–929. 2 D. B aga, F. G epioni and L. Maini, Chem. Commun., 2010, 46, 6232–6242. 3 H. G. B i ain, J. Pha m. Sci., 2008, 97, 3611–3636. 4Polymo phism in he Pha maceu ical Indus y, ed. R. Hil ike , Wiley- VCH Ve lag, Weinheim, Ge many, 2006. 5 K. Sa o, Chem. Eng. Sci., 2001, 56, 2255–2265. 6Pha maceu ical Sal s and Co-c ys als, ed. J. Wou e s and L. Que´ e´, Royal Socie y o Chemis y, 2011. 7 S. R. Chembu ka , J. Baue , K. Deming, H. Spiwek, K. Pa el, J. Mo is, R. Hen y, S. Span on, W. Dziki, W. Po e , J. Quick, P. Baue , J. Donaubaue , B. A. Na ayanan, M. Soldani, D. Riley and K. McFa land, O g. P ocess Res. De ., 2000, 4, 413–417. 8 Y. Kobayashi, S. I o, S. I ai and K. Yamamo o, In . J. Pha m., 2000, 193, 137–146. 9 J. O. Henck, U. J. G iesse and A. Bu ge , Pha m. Ind., 1997, 59, 165–169. 10 S. R. Vippagun a, H. G. B i ain and D. J. W. G an , Ad . D ug Deli e y Re ., 2001, 48, 3–26. 11 C. B. Aake o¨y and D. J. Salmon, C ys EngComm, 2005, 7, 439–448. 12 O. Alma sson and M. J. Zawo o ko, Chem. Commun.,2004, 1889–1896. 13 P. Vishweshwa , J. A. McMahon, J. A. Bis and M. J. Zawo o ko, J. Pha m. Sci., 2006, 95, 499–516. 14 N. J. Babu and A. Nangia, C ys . G ow h Des., 2011, 11, 2662–2679. 15 N. Schul heiss and A. Newman, C ys . G ow h Des., 2009, 9, 2950–2967. 16 D. B aga, F. G epioni, L. Maini, S. P ospe i, R. Gobe o and M. R. Chie o i, Chem. Commun., 2010, 46, 7715–7717. 17 D. B aga, F. G epioni, G. I. Lamp on i, L. Maini and A. Tu ina, C ys . G ow h Des., 2011, 11, 5621–5627. 18 L. Lecle cq, I. Suisse, G. Nowog ocki and F. Agbossou-Niede co n, J. Mol. S uc ., 2008, 892, 433–437. 19 S. Ai ipamula, P. S. Chow and R. B. H. Tan, C ys . G ow h Des., 2010, 10, 2229–2238. 20 J. Be ns ein, R. J. Da ey and J. O. Henck, Angew. Chem., In . Ed., 1999, 38, 3440–3461. 21 F. Wo¨hle and J. on Liebig, Ann. Pha m., 1832, 3, 249–282. 22 W. I. F. Da id, K. Shankland, C. R. Pulham, N. Blagden, R. J. Da ey and M. Song, Angew. Chem., In . Ed., 2005, 44, 7032–7035. 23 J. Thun, L. Sey a h, C. Bu e ho , J. Senke , R. E. Dinnebie and J. B eu, C ys . G ow h Des., 2009, 9, 2435–2441. 24 J. Thun, L. Sey a h, J. Senke , R. E. Dinnebie and J. B eu, Angew. Chem., In . Ed., 2007, 46, 6729–6731. 25 P. P. Bag, M. Pa ni and C. M. Reddy, C ys EngComm, 2011, 13, 5650–5652. 26 P. P. Bag and C. M. Reddy, C ys . G ow h Des., 2012, 12, 2740–2743. 27 J. B eu, W. Seidl, D. Hu ne and F. K aus, Chem.–Eu . J., 2002, 8, 4454–4460. 28 P. A. Williams, G. E. Hughes, G. K. Lim, B. M. Ka iuki and K. D. M. Ha is, C ys . G ow h Des., 2012, 12, 3104–3113. 29 C. Bu e ho , W. Milius and J. B eu, C ys EngComm, 2012, 14, 3945–3950. 30 Nume ical Da a and Func ional Rela ionships in Science and Technology - Vapo P essue o Chemicals, ed. J. Dykyj, J. S oboda, R. C. Wilhoi , M. F enkel and K. R. Hall, Sp inge , Be lin Heidelbe g, Ge many, 2000. 31 I. Dance and M. Scudde , C ys EngComm, 2009, 11, 2233–2247. 32 E. R. McNellis, J. Meye and K. Reu e , Phys. Re . B, 2009, 80. 33 A. Tka chenko and M. Sche le , Phys. Re . Le ., 2009, 102. 34 B. Ci alle i, C. M. Zico ich-Wilson, L. Valenzano and P. Ugliengo, C ys EngComm, 2008, 10, 405–410. 35 B. G. P omme , M. Co e, S. G. Louie and M. L. Cohen, J. Compu . Phys., 1997, 131, 233–240. 36 A. Bu ge and R. Rambe ge , Mic ochim. Ac a, 1979, 72, 259–271. 37 H. J. Flamme sheim, K is . Tech., 1974, 9, 299–311. 38 TOPAS Academic Technical Re e ence, Ve sion 4.1, A. A. Coelho, B uke AXS GmbH, Ka ls uhe, Ge many, 2007. 39 S. J. Cla k, M. D. Segall, C. J. Picka d, P. J. Hasnip, M. J. P obe , K. Re son and M. C. Payne, Z. K is allog ., 2005, 220, 567–570. 40 J. P. Pe dew, K. Bu ke and M. E nze ho , Phys. Re . Le ., 1996, 77, 3865–3868. Table 5 Compa ison o selec ed d- alues epo ed by Flamme sheim o he ‘‘high- empe a u e’’ phase o he co-c ys al 2 HBz?1 NaBz and he me as able o m B desc ibed in his wo k. The ag eemen shows ha bo h a e likely o be iden ical Flamme sheim 37 This wo k 12.06 11.966 10.92 10.874 10.59 10.557 5.62 5.596 5.46 5.437 5.24 5.226 4.94 4.948 4.51 4.498 This jou nal is ßThe Royal Socie y o Chemis y 2012 C ys EngComm, 2012, 14, 6744–6749 |6749 Downloaded by UNIVERSITAT BAYREUTH on 05 Oc obe 2012 Published on 09 Augus 2012 on h p://pubs. sc.o g | doi:10.1039/C2CE25562F View Online Anhang 73 Anhang A.3 In luence o Ca ion Size on he Co-c ys allisa ion o Benzoic Acid wi h Di e en Benzoa es Copy igh 2013 Wiley. Used wi h pe mission om C. Bu e ho , W. Milius und J. B eu, Z.Ano g.Allg.Chem., 2013, 2, 308-311. Anhang 74 ARTICLE DOI: 10.1002/zaac.201200464 In luence o Ca ion Size on he Co-c ys allisa ion o Benzoic Acid wi h Di e en Benzoa es Ch is ian Bu e ho , [a] Wol gang Milius, [a] and Jose B eu* [a] Dedica ed o P o esso Ha mu Bä nighausen on he Occasion o His 80 h Bi hday Keywo ds: Co-c ys als; C ys al s uc u e; Benzoic acid; Li hium benzoa e; Li hium Abs ac . In he pha maceu ical indus y many new ac i e pha maceu- ical ing edien s (APIs) a e ma ke ed as ca boxylic sal s because o enhanced solubili y and dissolu ion a es. These sal s a e, howe e , o en ha d o c ys allise and/o exhibi a low deg ee o c ys allini y. The eason may be he la gely di e ing sizes o ca ion and anion. One way ou o he esul ing “coo dina ion dilemma“ is he o ma ion o co-c ys als o he sal wi h i s neu al acid which deli e s addi ional coo dina ion si es o he ca ion. In his line, he c ys al s uc u e o a new co-c ys al o benzoic acid (HBz) and li hium benzoa e (LiBz) In oduc ion Mos APIs equi e su icien solubili y and dissolu ion a es o assu e bioa ailabili y and e ec i e d ug concen a ions in body luids. [1–3] The e o e, many APIs a e ma ke ed as ca - boxylic sal s, usually in 1:1 s oichiome y o ca ion:anion. Wi h 1:1 s oichiome y and since he molecula olumes o small ino ganic ca ions a e much lowe han he olumes o he la ge o ganic anions, i is di icul o a ange enough anions a ound he ca ion o sa is y he coo dina ion needs o he ca - ions. [4] Consequen ly, such sal s a e o en di icul o c ys al- lise. He e we s udy benzoa es as simple model compounds o explo e his p oblem. Fo ins ance, he s uc u es o such sim- ple compounds like NaBz and KBz could no be de e mined ye . In he case o NaBz, he powde di ac ion pa e n is no well esol ed, showing only ew and b oad e lec ions which indica e s uc u al diso de and only pa ial o de ing. One eason o he p oblema ic c ys allisa ion beha iou may be ela ed o he 1:1 s oichiome y. In o de o ealise he common six old coo dina ion o sodium, he coo dina ion polyhed a would ha e o be connec ed by sha ed edges and co ne s, e en i he ca boxyla e g oup would ac as biden a e ligand. Un o - una ely, he la ge a io o he olume o he benzoa e anion * P o . D . J. B eu Fax: +49-921-55-2788 E-Mail: [email p o ec ed] [a] Depa men o Ino ganic Chemis y I Uni e si y o Bay eu h Uni e si ä ss . 30 95440 Bay eu h, Ge many Suppo ing in o ma ion o his a icle is a ailable on he WWW unde h p://dx.doi.o g/10.1002/zaac.201200464 o om he au- ho . © 2013 WILEY-VCH Ve lag GmbH & Co. KGaA, Weinheim Z. Ano g. Allg. Chem. 2013,639, (2), 308–311308 wi h composi ion 1HBz·1LiBz was de e mined and was compa ed o published ionic co-c ys als o o he alkali sal s wi h espec o he s uc u e de e mining c i e ia: coo dina ion numbe o he ca ion, a io o HBz:Bz – , mode o coo dina ion o he ca boxylic g oups (mono- o biden a e), and connec i i y (edge- o co ne -sha ing) be ween neigh- bou ing polyhed a. The new c ys al s uc u e is closely ela ed o o m A o 2HBz·1NaBz. Bo h c ys al s uc u es con ain dime s wi h sha ed edges which a e u he connec ed by b idging biden a e ligands o apes. compa ed wi h he sodium ca ion hampe s connec ion o poly- hed a. In any case, o NaBz indexing and hence s uc u e solu ion p o ed impossible up o now. [5] One way ou o he “coo dina ion dilemma“ is he o ma ion o co-c ys als o he sal wi h i s neu al acid which deli e s addi ional coo dina ion si es o he ca ion. While in his pape we ocus on he s uc- u al implica ion o co-c ys al o ma ion, in he pha maceu ical con ex co-c ys als ha e ecen ly ecei ed conside able a en- ion because i was ound ha in many cases physical p ope - ies like solubili y, dissolu ion a e, e c. we e supe io o he pu e compounds. [6–10] Fo such co-c ys als o ca boxylic acid sal s in gene al and benzoa es in pa icula , he c ys al s uc- u es will be de e mined and a e cha ac e ised by he sub le balance o he ollowing ac o s: a) size o he ca ion which in u n de e mines he coo dina ion numbe , b) a io o HBz:Bz – , c) mode o coo dina ion o he ca boxylic g oups (mono- o biden a e), d) connec i i y (edge- o co ne -sha ing) be ween neighbou ing polyhed a. Consequen ly wi h any gi en ca ion size, only speci ic a ios o HBz:Bz – will be capable o ealise h ee-dimensional pe iodic s uc u es. He e we epo he so a unknown s uc u e o a co-c ys al o LiBz and com- pa e i wi h published s uc u es o he o he alkali me al sal s. Resul s and Discussion The smalles alkali me al ca ion Li + ( olume: 2 Å 3 ) [11] was ound o equi e a 1:1 (anion : neu al acid) s oichiome y o c ys allise and esides in a ou old coo dina ion. Addi ionally, a se ies o ibochemical co-c ys allisa ions (applying sol en - d op-g inding) [12] we e pe o med using s oichiome ic a ios In luence o Ca ion Size on he Co-c ys allisa ion o Benzoic Acid wi h Di e en Benzoa es (2:1; 3:1; 1:2; 1:3) be ween HBz and LiBz (see Expe imen al sec ion). He e always a mix u e o he 1:1 co-c ys al p oduc oge he wi h excess HBz o LiBz, espec i ely, was ob ained as indica ed by X- ay powde di ac ion (see Figu es S1, S2, and S3 in he Suppo ing In o ma ion). Table 1 shows he c ys- allog aphic da a and Figu e 1 he asymme ic uni o he co- c ys al 1HBz·1LiBz (one o mula uni HBz and one o mula uni LiBz) as ORTEP plo . The molecula packing is shown in Figu e 2. As i is shown in Figu e 2(a), he s uc u e consis s o apes, which a e s acked along he a-axis. Benzoa e and benzoic acid ligands a e a anged al e na ing on he sides o he apes. Be ween hese apes an de Waals o ces and π–π-s acking in e ac ions a e he main in e ac ion o ces. The π–πin e ac ion dis ances be ween adjacen phenyl ings (C(2)–C(7)) and (C(9)–C(14)) a e 4.28(6) Å and 4.85(2) Å. Table 1. C ys allog aphic da a and e inemen de ails o he co-c ys al 1HBz·1LiBz. Empi ical o mula C 14 H 11 LiO 4 Fo mula weigh 250.17 Tempe a u e /K 173(2) C ys al sys em iclinic Space g oup P1 ¯ a/Å 5.3629(11) b/Å 8.6472(17) c/Å 13.516(3) α/° 99.36(3) β/° 93.73(3) γ/° 95.70(3) V/Å 3 613.3(2) Z2 ρ calc /Mg·m –3 1.355 μ(Mo-K α )/mm –1 0.098 R in 0.0619 R σ 0.0416 No. e lns. 7752 Unique e lns. 3246 GOOF (F 2 ) 1.095 R 1a) (Iⱖ2σ) 0.0512 wR 2b) (Iⱖ2σ) 0.1206 R 1a) (all da a) 0.0685 wR 2b) (all da a) 0.1343 a) R 1 =Σ||F o |–|F c ||/Σ|F o |. b) wR 2 =[Σ[w(F o2 –F c2 ) 2 ]/Σ[w(F o2 ) 2 ]] 1/2 . Figu e 1. ORTEP plo and he c ys allog aphic numbe ing scheme o he asymme ic uni employed o he co-c ys als 1HBz·1LiBz. Dis- placemen ellipsoids a e d awn a he 50% p obabili y le el. A side iew on one o he apes shows ha li hium esides in a dis o ed e ahed al coo dina ion (see Figu e 2(b) and Fig- u e 3(a)). A space illing model (Figu e 2(c)) highligh s he dense packing in he ape. No space is a ailable o a ange addi ional ligands a ound he li hium moie ies. Wi hin he Z. Ano g. Allg. Chem. 2013, 308–311 © 2013 WILEY-VCH Ve lag GmbH & Co. KGaA, Weinheim www.zaac.wiley- ch.de 309 Figu e 2. Molecula packing in he 1HBz·1LiBz co-c ys al. (a) Pack- ing o one-dimensional apes along he a-axis. (b) Side iew on one o he apes. Coo dina ion bonds and hyd ogen bonds a e shown as dashed lines. (c) Space illing model o one ape highligh ing he dense packing a ound li hium. apes, dime s o edge-sha ing e ahed a become appa en . The dime s con ain in e sion cen es in he middle o he sha ed edges. All ca boxylic (HBz) and ca boxyla e (LiBz) g oups ac as monoden a e ligands in espec o a gi en dime . Symme y C. Bu e ho , W. Milius, J. B eu ARTICLE equi alen ca boxyla e g oups a e in ol ed in he sha ed edge (Figu e 3). The HBz molecule in a e ahed on is coo dina ed wi h i s ca bonyl oxygen a om (O(1)). Six ligands a e in ol ed in he coo dina ion sphe e o a dime and connec he dime in bo h di ec ions o he one-dimensional ape ia wo b idging ca boxyla e g oups (Figu e 3 (a)). An in a-dime H-bonding be ween O(2) and O(4) (Figu e 2(b), Figu e 3(a) and Table 2) con ibu es o he o ma ion o he apes. This in e -dime bond ein o ces he dime . Adjacen phenyl g oups wi hin he apes a e in ol ed in shi ed π-s acks (Figu e 2(c)), (in e ac ion be- ween symme y ela ed phenyl ings o (C(2)–C(7)) and (C(9)–C(14))). The π–πin e ac ion dis ance he e is 5.36(3) Å. In he ollowing we compa e his s uc u e wi h published co-c ys al s uc u es o o he alkali me al benzoa es. [5,13–15] Table 3 summa ises his compa ison wi h espec o he s uc- u e de e mining c i e ia men ioned in he in oduc ion: coo di- na ion numbe , a io o HBz : Bz – , mode o coo dina ion o he ca boxylic g oups (mono- o biden a e), and connec i i y (edge- o co ne -sha ing) be ween neighbou ing polyhed a. Figu e 3 shows he connec ion o he polyhed a (p ima y building uni ) o di e ing seconda y building uni s in he a i- ous co-c ys als. In espec o bo h he connec ion o p ima y and seconda y building uni s, 1HBz·1LiBz and o m A o he 2HBz·1NaBz, a e mos simila . Bo h con ain dime s o edge sha ed polyhed a ( e ahed a in he case o Li + s. oc ahed a in he case o Na + ) and ealise he same in e connec ion scheme Figu e 3a and Figu e 3b): Biden a e ligands connec he dime s in o in ini e apes. The apes a e hen s acked by an de Waals o ces. Al hough he ca ions a e qui e simila in size (2 Å 3 s. 3Å 3 ), as expec ed, Li + p e e s e ahed al coo dina ion which allows he same in e connec i i y bu wi h di e ing s oichiom- e y. In e es ingly he sodium compound was ound o be dimo - phic. [5,13] The second polymo ph, while s icking o he same coo dina ion and a io o HBz / Bz – , is buil om in ini e col- umns. The highe deg ee o condensa ion o coo dina ion poly- hed a in he seconda y building uni s is hen necessa ily com- pensa ed by ewe biden a e ligands (Table 3). The c ys al s uc u es o he po assium and ammonium com- pounds a e isos uc u al despi e he a he la ge di e ence in ca ion olume (10 Å 3 s. 24 Å 3 ). As in o m B o he sodium compound, he c ys al s uc u e consis s o in ini e columns o edge-sha ing oc ahed a. The s oichiome y is, howe e , 1:1 (anion:neu al acid) indica ing ha he deg ee o connec i i y mus be highe o he po assium compound as compa ed o o m B o he sodium co-c ys al. While in he s uc u e o he la e , neighbou ing columns a e s acked by an de Waals in e ac ions, in 1HBz·1KBz neighbou ing columns a e con- nec ed by biden a e ca boxylic g oups in o laye s (Figu e 3d). Thus, he highe connec i i y in he po assium compound is u ilised o ealise a highe dimensionali y. Conclusions Compa ison o he a ious c ys al s uc u es o ionic co- c ys als o alkali sal s sugges ha he pa icula coo dina ion www.zaac.wiley- ch.de © 2013 WILEY-VCH Ve lag GmbH & Co. KGaA, Weinheim Z. Ano g. Allg. Chem. 2013, 308–311310 Figu e 3. O e iew o he connec ion o coo dina ion polyhed a in he co-c ys als: (a) 1HBz·1LiBz (b) 2HBz·1NaBz ( o m A) (c) 2HBz·1NaBz ( o m B) (d) 1HBz·1KBz. The exac connec ion is de- sc ibed in he ex and Table 3. Fo cla i y, phenyl ings o HBz and Bz – a e no shown. Hyd ogen bonds a e shown as dashed lines. Table 2. Summa y o in e molecula in e ac ions (D–H···A/Å,°) ope a ing in he c ys al s uc u e o he ionic co-c ys al 1HBz·1LiBz. DHAH···A/Å D···A/Å ⬍(DHA) /° Symme y ope a ion O(2) H(2) O(4) 1.79 2.6158(16) 166.7 x,y,z equi emen s o he ca ions and dense packing cons ain s can bo h be se ed by egula ing he a io o HBz:Bz – and/o mode o coo dina ion o he ca boxylic g oups (mono- o biden a e). In luence o Ca ion Size on he Co-c ys allisa ion o Benzoic Acid wi h Di e en Benzoa es Table 3. Compa ison o he packing ea u es in he ou ionic co-c ys als. 1HBz·1LiBz 2HBz·1NaBz ( o m A) 2HBz·1NaBz ( o m B) 1HBz·1KBz Ra io HBz / Bz – 1/1 2/1 2/1 1/1 Ca ion Li + Na + Na + K + Volume /Å 3 23310 Coo dina ion numbe 4 6 6 6 Coo dina ion dis o ed e ahed al dis o ed oc ahed al dis o ed oc ahed al dis o ed oc ahed al Connec ion o polyhed a dime s wi h 1 sha ed edge dime s wi h 1 sha ed edge ods wi h 2 sha ed edges ods wi h 2 sha ed edges Seconda y building uni s apes apes columns laye s o in e connec ed columns N #monoden a e HBz ligands 1140 N #monoden a e Bz–ligands 0000 N #biden a e HBz ligands 0203* N #biden a e Bz–ligands 3323* # numbe o ligand o pa icula ype pe polyhed on, * HBz/Bz – no dis inguished Expe imen al Sec ion HBz (pu i y ⬎99.5%) was pu chased om G üssing, LiBz (pu i y ⱖ 99.0%) om Al a Aesa . E hanol was sou ced om VWR (BDH P o- labo). All chemicals and sol en s we e used wi hou u he pu i ica- ion. Dis illed wa e was used. X- ay powde di ac ion aces we e eco ded using a PANaly ical X- Pe P o (Cu-K α adia ion, e lec ion geome y) di ac ome e . Single- c ys al X- ay di ac ion analyses we e pe o med using a STOE IPDS II ins umen (Mo-K α adia ion) equipped wi h an Ox o d c yos eam sys em. Selec ed c ys allog aphic da a a e lis ed in Table 1. The c ys al s uc u es we e sol ed and e ined using SHELXTL 5.1 (B uke AXS). All non-hyd ogen a oms we e e ined aniso opically wi h he hyd o- gen a oms added o hei geome ically ideal posi ions and e ined iso- opically excep ha he hyd ogen a oms o he hyd oxyl g oups we e loca ed om he elec on densi y and hen e ined using a iding model. All igu es we e d awn wi h he DIAMOND p og am. C ys al- log aphic da a o he s uc u e in his pape ha e been deposi ed wi h he Camb idge C ys allog aphic Da a Cen e, CCDC, 12 Union Road, Camb idge CB2 1EZ, UK. Copies o he da a can be ob ained ee o cha ge on quo ing he deposi o y numbe CCDC-911587 (h p://www.ccdc.cam.ac.uk) 1HBz·1LiBz Single c ys als we e ob ained om a supe sa u a ed solu ion o HBz wi h LiBz in s oichiome ic 1:1 a io ( o al quan i y: app ox. 4500 mg) in 35 mL e hanol-wa e -mix u e (4:1/ : ). The mix u e was hea ed un- il all s a ing ma e ial was dissol ed. Then he solu ion was cooled slowly (o e 3 hou s) o oom empe a u e. A e wa ds he solu ion was le a oom empe a u e. A e 3–5 days spicula , colou less c ys- als we e ound. These we e il e ed o he mo he liquid and d ied a oom empe a u e. T ibochemical C ys allisa ion The popula me hod o sol en -d op-g inding wi h a ypical hand-mo - a was applied. Di e en a ios o HBz and NaBz (3:1; 2:1; 1:1; 1:2; Z. Ano g. Allg. Chem. 2013, 308–311 © 2013 WILEY-VCH Ve lag GmbH & Co. KGaA, Weinheim www.zaac.wiley- ch.de 311 1:3) ( o al quan i ies: app ox. 1500 mg) we e we ed wi h he sol en (e hanol–wa e mix u e (4:1/ : )) and hen g ound o d yness. Suppo ing In o ma ion (see oo no e on he i s page o his a icle): Powde di ac ion pa e ns o 1HBz·1LiBz and he sol en -d op- g inding p oduc s. Acknowledgmen s We hank he Deu sche Fo schungsgemeinscha (SPP 1415), o inan- cial suppo . Re e ences [1] H. G. B i ain, Polymo phism in Pha maceu ical Solids, In o ma Heal hca e, Inc., 2009. [2] R. Hil ike , Polymo phism in he Pha maceu ical Indus y, Wiley- VCH Ve lag Weinheim, Ge many, 2006. [3] J. Wou e s, L. Qué é, Pha maceu ical Sal s and Co-c ys als, Royal Socie y o Chemis y, 2011. [4] A. L. Rohl, D. M. Mingos, J. Chem. Soc., Dal on T ans. 1992, 3541. [5] C. Bu e ho , W. Milius, J. B eu, C ys EngComm 2012,14, 3945. [6] N. Blagden, M. de Ma as, P. T. Ga an, P. Yo k, Ad . D ug Deli e . Re . 2007,59, 617. [7] S. L. Childs, M. J. Zawo o ko, C ys . G ow h Des. 2009,9, 4208. [8] N. Schul heiss, A. Newman, C ys . G ow h Des. 2009,9, 2950. [9] N. Shan, M. J. Zawo o ko, D ug Disco . Today 2008,13, 440. [10] P. Vishweshwa , J. A. McMahon, J. A. Bis, M. J. Zawo o ko, J. Pha m. Sci. 2006,95, 499. [11] D. M. P. Mingos, A. L. Rohl, Ino g. Chem. 1991,30, 3769. [12] A. V. T ask, W. D. S. Mo he well, W. Jones, Chem. Commun. 2004, 890. [13] C. Bu e ho , K. Bä winkel, J. Senke , J. B eu, C ys EngComm 2012,14, 6744. [14] I. A. Ox on, T. S. Came on, O. Knop, A. W. McCulloch, Can. J. Chem. 1977,55, 3831. [15] J. M. Skinne , G. M. D. S ewa , J. C. Speakman, J. Chem. Soc. 1954, 180. Recei ed: Oc obe 19, 2012 Published Online: Janua y 4, 2013 Z. Ano g. Allg. Chem. 2012 · © WILEY-VCH Ve lag GmbH & Co. KGaA, 69451 Weinheim, 2012 · ISSN 0044–2313 SUPPORTING INFORMATION Ti le: In luence o Ca ion Size on he Co-c ys allisa ion o Benzoic Acid wi h Di e en Benzoa es Au ho (s): C. Bu e ho , W. Milius, J. B eu* Re . No.: Z201200464 Mic ophase Sepa a ion wi h Small Amphiphilic Molecules Figu e 1. PXRD aces o NaBz (a) and KBz (b). The uppe aces co espond o compounds as pu chased, he lowe aces o e-c ys allized ma e ials. Fo NaBz he numbe o e lec ions inc eased signi ican ly and he e lec ions a e much sha pe a e long- e m annealing. In he case o KBz di e ences be o e and a e e-c ys alliza ion a e much mo e sub le. The inse shows, ha ne e heless he asymme y o lambda-shaped e lec ions caused by s acking de aul s in he p is ine ma e ial (uppe ace) is cu ed and new e lec ions appea upon e-c ys alliza ion by slow e apo a ion om me hanol (lowe ace). ion o sodium o po assium can only be ealized, i he coo - dina ion polyhed a sha e edges and co ne s. Un o una ely, he space equi ed o he la ge o ganic benzoa e anion hampe s joining o he polyhed a. A way ou o he dilemma is he o ma ion o co-c ys als o HBz wi h NaBz, which inc eases he numbe o ligands pe ino ganic ca ion. [11] A co-c ys al wi h he composi ion 2HBz·1NaBz ( wo o mula uni s HBz and one o mula uni NaBz) could be s uc u ally cha ac e ized. Fu he mo e a sec- ond polymo ph o his co-c ys al ( o m B) could be c ys- allized by lash e apo a ion and also s uc u ally cha ac- e ized. [12] In a e iew we compa ed hese wo polymo phs wi h analogue co-c ys als o HBz and benzoa es wi h di e en ca ions. [13] Fo he c ys al s uc u e o NaBz we p oposed ha i is com- posed o low dimensional building uni s, which a e packed in a diso de ed way. [11] By applying long e m annealing a 420 °C o 5 mon hs we now ob ained iny needle-like c ys als o NaBz, which we e ne e heless sui able o single c ys al s uc u e de e mina ion. Resul s and Discussion Op imiza ion he Condi ions o C ys al G ow h o he Alkali Me al Benzoa es By employing c ys alliza ion me hods om solu ions, u iliz- ing di e en sol en s, o NaBz only poo ly c ys alline ma e- ial could be ob ained as indica ed by he X- ay powde di - ac ion pa e n (XRPD). Employing lowe deg ees o supe - sa u a ion (applying slow e apo a ion c ys alliza ion) o NaBz s ill a poo ly c ys alline powde [Figu e 1(a)] was ob ained. Howe e we we e able o g ow small needle-like single c ys- Z. Ano g. Allg. Chem. 2013, 2816–2821 © 2013 WILEY-VCH Ve lag GmbH & Co. KGaA, Weinheim www.zaac.wiley- ch.de 2817 als o su icien quali y o s uc u e de e mina ion by apply- ing long e m annealing a 420 °C in an o en wi h a e y accu- a e empe a u e con ol (accu acy ⫾0.5 °C) o e 165 d. Long ime annealing o p omo e c ys alliza ion is well es ab- lished. [14,15] Appa en ly he high empe a u es allow o su - icien mobili y o he molecules o heal he de ec s and a - ange in o a 3-dimensionally o de ed s uc u e. To p e en de- g ada ion by oxida ion o NaBz, he powde was sealed in a gas igh qua z-ampule unde acuum. The PXRD ace ob- ained o NaBz a e annealing shows signi ican ly mo e and much na owe e lec ions indica ing a mo e c ys alline ma e- ial [Figu e 1(a)]. While addi ional e lec ions appea , a ew e lec ions a e al eady p esen in he powde pa e n be o e he annealing p ocedu e indica ing ha some s uc u al ea u es a e e ained du ing annealing. Addi ionally he mo phology o he annealed ma e ial di e ed conside ably (Figu e S1, Sup- po ing In o ma ion). While he ma e ial be o e annealing con- sis s o i egula shaped and a he small pa icles, a e wa ds he mo phology is needle-like wi h well-de ined c ys al aces and edges. Ne e heless, he long needles a e e y hin, ben , and agile. A la ge numbe o c ys als had o be sc eened be o e a c ys al sui able o da a collec ion could be iden i ied. Con a y o NaBz, KBz could be c ys allized by a s anda d solu ion c ys alliza ion me hod. The bes c ys als could be ob- ained om me hanol (0.05 mol·L –1 ) as la ge pla y c ys als, applying slow e apo a ion c ys alliza ion. Like wi h NaBz, he c ys als di e conside ably in mo phology and size be o e and a e he e-c ys alliza ion (Figu e S1). Howe e , as i can be seen om SEM images he c ys als a e buil up om s acked laye s, which a e sligh ly shi ed and o a ed agains each o he . These s acking aul s un along he caxis o he c ys al s uc u e. The PXRD aces o KBz be o e and a e e- c ys alliza ion di e li le on a i s sigh . A close look, how- C. Bu e ho , T. Ma in, W. Milius, J. B eu ARTICLE e e , e eals sub le bu impo an di e ences [inse o Fig- u e 1(b)]. The p is ine ma e ial su e s o asymme ic, lambda- shaped hk-bands, o ins ance he 11-band caused by s acking aul s ha occu equen ly wi hin he cohe ence leng h o he X- ay beam. Addi ionally, a i a ional e lec ion a 24.24°2θis obse ed o his semi-o de ed ma e ial ha makes indexing o he ace impossible due o he non-B agg na u e o his peak. Upon e-c ys alliza ion by slow e apo a ion om me hanol he asymme y is cu ed and new e lec ions (111,112) appea in he egion o he 11-band, while he non-B agg in ensi y an- ishes comple ely. While he s acking o de imp o es a g ea deal as obse ed in he PXRD ace, some s acking aul s emain especially in la ge single c ys als. The e o e, a la ge numbe o c ys als had o be sc eened be o e a c ys al sui able o da a collec ion and s uc u e solu ion could be iden i ied. Mos “single c ys als” o KBz could no be indexed due o a signi ican numbe o s acking aul s emaining. The di use in ensi y ela ed o he s acking aul s ne e heless induces sligh ly highe esidual alues in he e inemen . C ys al S uc u e o Sodium Benzoa e Despi e he small c ys al size and consequen ly a he low in ensi ies, c ys al s uc u e solu ion and e inemen o NaBz u ned ou o be s aigh o wa d. Table 1 shows he c ys allo- g aphic da a and Figu e 2(a) he asymme ic uni o he c ys al s uc u e o NaBz as ORTEP plo . The molecula packing is shown in Figu e 3 (le column). The s uc u e consis s o apes, which un along he baxis. The hyd ophilic ca boxyla e g oups o he benzoa e ligands a e o ien ed owa ds he co e o he ape, whe e addi ionally all Na + eside. The hyd ophobic Table 1. C ys allog aphic da a o he single c ys al e inemen o NaBz and KBz. NaBz KBz C ys al shape needle pla ele Molecula o mula C 35 H 25 Na 5 O 10 C 14 H 10 K 2 O 4 Fo mula weigh 576.40 320.42 Tempe a u e /K 293(2) 173(2) C ys al sys em monoclinic o ho hombic Space g oup P2 1 /n(No. 14) Pca2 1 (No. 29) a/Å 15.113(3) 11.481(2) b/Å 6.4048(13) 3.9265(8) c/Å 34.476(7) 30.055(6) α/° 90 90 β/° 100.18(3) 90 γ/° 90 90 Volume /Å 3 3284.5(11) 1354.9(5) Z20 8 Calcula ed densi y /mg·m –3 1.457 1.571 μ/mm –1 0.161 0.708 R in 0.2250 0.0935 GOF 0.647 1.099 Re lec ions (unique/ 6280/1187 2434/1851 unique⬎2σ) R 1 /wR 2 [obsd. da a: I⬎0.0699/0.1383 0.0695/0.2154 2σ(I)] wR 2 (all da a) 0.1857 0.2260 La ges esidual/e·Å –3 0.275 0.983 www.zaac.wiley- ch.de © 2013 WILEY-VCH Ve lag GmbH & Co. KGaA, Weinheim Z. Ano g. Allg. Chem. 2013, 2816–28212818 phenyl moie ies o he ligands coa his co e. The apes e- semble a wo m-like micelle [Figu e 3(c)]. These micelles a e packed in hexagonally close esembling he hexagonal mic o- phase o su ac an s o block-copolyme s. Be ween he apes an de Waals o ces and π–π-s acking in e ac ions a e he main in e ac ion o ces. Fo ins ance, along he aaxis π–π in e ac ions [dis ance be ween adjacen phenyl ings (C(2)– C(7)) and (C(9)–C(14)) is 5.77(9) Å] in e connec he adjacen apes. The weak in e ac ion o ces in e connec ing he apes migh explain he agile and ib ous na u e o he small c ys- als. The numbe o symme y independen en i ies in he uni cell (Z⬘= 5) is excep ionally high e lec ing he di icul ies in sa is ying he coo dina ion. The e a e e y ew c ys als s uc- u es epo ed in he CSD wi h suchlike high Z⬘numbe . [16,17] S eed in es iga ed and ound c ys al s uc u es wi h Z⬘⬎4 o be ex emely a e. [18] Z⬘= 6 and Z⬘= 8 a e sligh ly mo e ypical and co espond o he exis ence o pseudosymme y. One example o pseudosymme y is he c ys al s uc u e o choles e ol monohyd a e. [19] Along he apes he sodium ca ions eside a wo di e en heigh s [Figu e 4(a)]. The coo dina ion sphe es o inne so- dium ca ions is oc ahed al (Na[6]), he ca ions a he im o he apes, howe e eside in a sligh ly dis o ed quad a ic p ism (Na[5]). Rows o Na[5], Na[6], Na[6], Na[6], Na[5] a e con- nec ed by sha ed edges and a e il ed by ca. 38° in espec o he baxis. Adjacen ows a di e en heigh a e connec ed by sha ed edges as well. Na–O dis ances o all polyhed a a e summa ized in Table S1 (Suppo ing In o ma ion). Adjacen phenyl g oups wi hin he apes a e in ol ed in shi ed π-s acks [Figu e 3(b)], [in e ac ion dis ance be ween adjacen phenyl ings o (C(2)–C(7)) and (C(16)–C(21)) is 4.01(4) Å; (C(16)– C(21)) and (C(30)–C(35)) is 4.01(5) Å; (C(2)–C(7)) and (C(9)–C(14)) is 4.02(5) Å; (C(9)–C(14)) and (C(23)–C(28)) is 4.05(3) Å, espec i ely]. To p o e phase pu i y o he annealed NaBz, addi ionally a Rie eld e inemen applying he benzoa e molecules as igid bodies was pe o med. The Rie eld plo is shown in Figu e 5 and he esul s o he e inemen a e summed up in Table 2. The quali y o he Rie eld i is high indica ing a phase pu e ma e ial. C ys al S uc u e o Po assium Benzoa e Mos single c ys als showed p onounced di use s eaks and indexing p o ed impossible. F om abou 100 sc eened c ys als only wo could be indexed. Wi h he da a se collec ed o he chosen c ys al, howe e , c ys al s uc u e solu ion and e ine- men u ned ou o be s aigh o wa d. Table 1 shows he c ys- allog aphic da a and Figu e 2(b) he asymme ic uni o he c ys al s uc u e o KBz as ORTEP plo . The molecula pack- ing is shown in Figu e 3 ( igh column). As wi h NaBz, he compound unde goes “mic ophase sepa a ion” upon c ys alli- za ion, howe e , ins ead o he hexagonal, he lamella phase is ealized wi h KBz. The laye s a e s acked along he caxis. The hyd ophilic pa s o he compound again sepa a e in o he co e o he laye s and he phenyl moie ies o he benzoa e li- Mic ophase Sepa a ion wi h Small Amphiphilic Molecules Figu e 2. ORTEP plo s and he c ys allog aphic numbe ing schemes o he asymme ic uni s o he single c ys al s uc u es o NaBz (a) and KBz (b). Displacemen ellipsoids a e d awn a he 50% p obabili y le el. Please no e he excep ionally high numbe o symme y independen molecules (Z⬘= 5) in he asymme ic uni o NaBz. Figu e 3. Molecula packing o he NaBz (le column) and KBz ( igh column) c ys al s uc u es. (a) App oxima ely hexagonal packing o apes unning along he baxis. π–πin e ac ions be ween he apes a e dashed in g ey. (b) Side iew on his single ape. (c) Single ape ep esen ing a wo m-like micelle. (d) Laye s s acked along he caxis. (e) Side iew on his single laye . Coo dina ion bonds a e dashed in black. ( ) Single laye . gands a e a he su ace o he laye s. The laye s a e s acked by weak an de Waals o ces, which is he eason o he s acking aul s obse ed. Z. Ano g. Allg. Chem. 2013, 2816–2821 © 2013 WILEY-VCH Ve lag GmbH & Co. KGaA, Weinheim www.zaac.wiley- ch.de 2819 Wi hin he laye [Figu e 3(e)] π–πin e ac ions a e ealized along he baxis [dis ance be ween adjacen phenyl ings (C(2)–C(7)) is 3.92(7) Å and (C(9)–C(14)) is 3.92(7) Å, C. Bu e ho , T. Ma in, W. Milius, J. B eu ARTICLE Figu e 4. Coo dina ion o he ca ions in NaBz (a) and KBz (b). The coo dina ion polyhed a a e in e connec ed by sha ed edges. The coo dina ion numbe is no ed in squa e b acke s, while he labels o he a oms a e gi en in pa en heses. Figu e 5. Rie eld e inemen (applying TOPAS Academic) o he single c ys al s uc u e o NaBz (a) and KBz (b). Wi h NaBz he e inemen is showing he good i be ween he obse ed and calcula ed da a. Shown a e he obse ed pa e n (O), he bes Rie eld- i p o ile (–), he ick ma ks (|) indica e he posi ions o he B agg e lec ions o he uni cell o NaBz. Δmeans he di e ence. The inse shows he egion om 44 o 52°2θ. Wi h KBz he Rie eld e inemen shows ob iously he s acking aul s in he egion 23 o 26°2θ. Table 2. Rie eld e inemen de ails o he labo a o y PXRD da a ob- ained. Pa ame e NaBz KBz Tempe a u e /K 293(2) 173(2) a/Å 15.045(2) 11.4643(5) b/Å 6.3970(7) 3.9426(1) c/Å 34.287(4) 29.871(1) β/° 100.13(1) 90 Volume /Å 3 3248.8(7) 1350.2(9) Z20 8 Z⬘52 Da a ange 1.5–70 1.5–70 R B agg /% 0.90 1.30 R wp /% 2.13 5.0 espec i ely] and ein o ce he bonding wi hin he laye s. All ca ions a e coo dina ed in a dis o ed e agonal p isma ic way. www.zaac.wiley- ch.de © 2013 WILEY-VCH Ve lag GmbH & Co. KGaA, Weinheim Z. Ano g. Allg. Chem. 2013, 2816–28212820 The wo c ys allog aphically independen po assium ca ions eside a di e en heigh s in he laye s. By sha ing edges, wo chains unning along ba e o med wi h he e agonal basal su aces acing each o he [Figu e 4(c)]. The wo chains a e in e connec ed by addi ional sha ed edges. All oge he one single p ism sha es six edges wi h i s nea es su ounding neighbo s. The K–O dis ances o he wo polyhed a a e sum- ma ized in Table S2. To check o phase pu i y o he e-c ys allized KBz, ad- di ionally a Rie eld e inemen (TOPAS Academic [20] ) apply- ing he benzoa e molecules as igid bodies was pe o med. The Rie eld plo is shown in Figu e 5 and he esul s o he e ine- men a e summed up in Table 2. The quali y o he Rie eld i is high indica ing a phase pu e ma e ial. E en he 2θ ange o he 11-band (inse ) is well ep oduced. Ne e heless, he backg ound be ween he 111 and he 112 e lec ions is oo high Mic ophase Sepa a ion wi h Small Amphiphilic Molecules indica ing some emaining s acking aul s e en a e e- c ys alliza ion. Conclusions Labo ious op imiza ion o he c ys alliza ion condi ions al- lowed o signi ican ly educe s uc u al diso de and pa ed he way o sol ing he s uc u e o he comme cially mos ele an ood p ese a i es E 211 (NaBz) and E 212 (KBz). Benzoa e ep esen s a small amphiphilic molecule wi h a pola ca box- ylic head g oup and a non-pola phenyl moie y. And indeed o bo h compounds a kind o mic ophase sepa a ion was ob- se ed upon c ys alliza ion. Fo NaBz a kind o hexagonal close packing o od micelles is ealized, while in he case o KBz a lamella a angemen was ound. Wi h su ac an s and block-copolyme s [21] changes o mic ophases a e igge ed by concen a ion, empe a u e, o he ela i e leng hs o he blocks. Compa ing he c ys al s uc u e o he wo compounds NaBz and KBz, i is wo h no ing ha he minu e change o he ela i e olumes o ino ganic ca ions and o ganic anion al eady igge a change in he mic ophase sepa a ion ealized upon c ys alliza ion. Expe imen al Sec ion C ys alliza ion: NaBz (pu i y ⱖ99.0%) and KBz (pu i y ⱖ99.0%) we e bo h pu chased om Fluka. Me hanol AnalaR NORMAPUR was sou ced om VWR BDH P olabo and was used wi hou u he pu i i- ca ion. Small needle-like single c ys als o NaBz we e ob ained by annealing. NaBz (ca. 0.5 g) was sealed in a gas igh qua z-ampule and hea ed in a u nace wi h a e y accu a e empe a u e con ol (accu acy ⫾0.5 °C) [Ca boli e LHT 6/30 wi h wo empe a u e con ol uni s (uni 1: Eu o- he m E3216P1 (PID (P opo ional In eg al De i a i e) egula o ; uni 2: Eu o he m E2132] o 165 d a 420 °C. Tempe a u e con ol uni 2 was se o 422 °C o p e en o e hea ing. Please no e ha abo e his empe a u e slow decomposi ion o NaBz was obse ed. Single c ys als o KBz we e c ys allized om a solu ion o KBz in me hanol (0.05 mol·L –1 ) by e y slow e apo a ion c ys alliza ion in a pe o a ed snap cap essel a oom empe a u e. A e 7 d, small pla y c ys alli es appea ed. They we e il e ed o he mo he liquid und d ied a oom empe a u e. PXRD: Measu emen s we e pe o med using a STOE STADI P (Cu- K α1 adia ion, ansmission geome y) di ac ome e equipped wi h a as , high esolu ion silicon s ip de ec o DECTRIS My hen1K. PXRD aces a low empe a u e we e measu ed on he same ins umen equipped wi h an Ox o d C yos eam 700 Se ies empe a u e con ol uni . SEM: Scanning elec on mic oscopy was pe o med wi h a Zeiss LEO1530 FESEM. Single C ys al X- ay Di ac ion: Da a we e collec ed wi h a STOE IPDS I ins umen (173 K and 293 K, Mo-K α adia ion) equipped wi h an Ox o d C yosys ems C yos eam sys em. Selec ed c ys allog aphic Z. Ano g. Allg. Chem. 2013, 2816–2821 © 2013 WILEY-VCH Ve lag GmbH & Co. KGaA, Weinheim www.zaac.wiley- ch.de 2821 da a a e lis ed in Table 1. The c ys al s uc u e was sol ed and e ined using SHELXTL 5.1 (B uke AXS). The c ys al o KBz was ound o be acemically winned wi h a olume ac ion o 24% o he mino i y domain. All igu es we e d awn wi h he DIAMOND p og am. C ys- allog aphic da a o he s uc u e in his pape ha e been deposi ed wi h he Camb idge C ys allog aphic Da a Cen e CCDC, 12 Union Road, Camb idge CB2 1EZ, UK. Copies o he da a can be ob ained ee o cha ge on quo ing he deposi o y numbe CCDC 957991- 957992 (h p://www.ccdc.cam.ac.uk). Suppo ing In o ma ion (see oo no e on he i s page o his a icle): SEM images o he c ys als and ca ion – anion dis ances o he di - e en coo dina ion polyhed al. Acknowledgemen s We hank he Deu sche Fo schungsgemeinscha (SPP 1415) o inan- cial suppo . Re e ences [1] H. G. B i ain, Polymo phism in Pha maceu ical Solids, 2nd ed. In o ma Heal hca e, Inc., 2009. [2] R. Hil ike , Polymo phism in he Pha maceu ical Indus y, 1s ed., Wiley-VCH Ve lag, Weinheim, Ge many, 2006. [3] J. Wou e s, L. Qué é, Pha maceu ical Sal s and Co-c ys als, 1s ed., Royal Socie y o Chemis y, 2011. [4] D. F. Schoo s, M. Desme , T. Reiss, D. Ma golskee, H. Cheng, P. La son, R. Amin, G. Some s, B i . J. Clin. Pha maco. 1995,40, 277–280. [5] J. Thun, W. Milius, B. Wedel, A. Ridde , P. Moe sdo , J. B eu, C ys EngComm 2009,11, 1306–1308. [6] A. L. Rohl, D. M. Mingos, J. Chem. Soc. Dal on T ans. 1992, 3541–3552. [7] E. Luck, Zbl. Bak . Mik. Hyg. B 1985,180, 311–318. [8] Uni ed Na ions En i onmen al P og amme (UNEP), 2001, Ben- zoa es, O ganiza ion o Economic Coope a ion and De elopmen Sc eening In o ma ion Da ase . UNEP Publica ions. A ailable: h p://www.inchem.o g/documen s/sids/sids/BENZOATES.pd . [9] R. Van Deun, J. Ramaeke s, P. Nockemann, K. Van Hecke, L. Van Mee el , K. Binnemans, Eu . J. Ino g. Chem. 2005, 563– 571. [10] G. A. Sim, J. M. Robe son, T. H. Goodwin, Ac a C ys allog . 1955,8, 157–164. [11] C. Bu e ho , W. Milius, J. B eu, C ys EngComm 2012,14, 3945– 3950. [12] C. Bu e ho , K. Bä winkel, J. Senke , J. B eu, C ys EngComm 2012,14, 6744–6749. [13] C. Bu e ho , W. Milius, J. B eu, Z. Ano g. Allg. Chem. 2013, 639, 308–311. [14] Y. C. Kim, I. J. Chung, Polym. J. 1993,25, 1257–1265. [15] L. L. Kukkonen, I. M. Reaney, D. Fu niss, A. B. Seddon, Phys. Chem. Glasses 2001,42, 265–273. [16] J. Be ns ein, J. D. Duni z, A. Ga ezzo i, C ys . G ow h Des. 2008,8, 2011–2018. [17] S. Mahapa a, T. S. Thaku , S. Joseph, S. Va ughese, G. R. Desi- aju, C ys . G ow h Des. 2010,10, 3191–3202. [18] J. W. S eed, C ys EngComm 2003, 169–179. [19] B. M. C a en, Ac a C ys allog ., Sec . B 1979,35, 1123–1128. [20] A. A. Coelho, TOPAS Academic Technical Re e ence, Ve sion 4.1, B uke AXS GmbH, Ka ls uhe, Ge many, 2007. [21] L. Leible , Mac omolecules 1980,13, 1602–1617. Recei ed: Augus 28, 2013 Published Online: Oc obe 7, 2013 Z. Ano g. Allg. Chem. 2013 · © WILEY-VCH Ve lag GmbH & Co. KGaA, 69451 Weinheim, 2013 · ISSN 0044–2313 SUPPORTING INFORMATION Ti le: Mic ophase Sepa a ion wi h Small Amphiphilic Molecules: C ys al S uc u e o P ese a i es Sodium Benzoa e (E 211) and Po assium Benzoa e (E 212) Au ho (s): C. Bu e ho , T. Ma in, W. Milius, J. B eu Re . No.: Z201300436 Mic ophase Sepa a ion wi h Small Amphiphilic Molecules: C ys al S uc u e o P ese a i es Sodium Benzoa e (E 211) and Po assium Benzoa e (E 212) Ch is ian Bu e ho §, Thomas Ma in§,Wol gang Milius and Jose B eu.§* § Ino ganic Chemis y I, Uni e si y o Bay eu h, Uni e si ä ss aße 30, 95440 Bay eu h, Ge many. Suppo ing In o ma ion Fig. 1. SEM Images o NaBz (a) and KBz (b). On he le side he compounds as pu chased and on he igh side he e-c ys allized ma e ials a e shown. In he case o NaBz (a, igh ) b i le, bend, needle-like c ys alli es we e ob ained a e long e m annealing. In he case o KBz (b, igh ) by slow e apo a ion c ys alliza ion om me hanol la ge, pla y c ys als showing s acking aul s along he c-axis a e ob ained. Table S1. Na-O dis ances in he polyhed a. In b acke s he co esponding symme y ope a ions a e gi en, while in he cases o (x, y, z) no in o ma ion is gi en. Na(1) Na(2) Na(3) Na(4) Na(5) coo dina ion numbe 6 6 6 5 5 O(1)-Na(x) (Å) 2.4868(53) and 2.3640(25) (1-x, 2-y, -z) 2.3906(69) O(2)-Na(x) (Å) 2.5521(57) and 2.3188(66) (1-x, 1-y, -z) 2.3866(66) (1-x, 1-y, -z) O(3)-Na(x) (Å) 2.5085(58) 2.3111(66) (x, -1+y, z) 2.3246(63) O(4)-Na(x) (Å) 2.4547(72) 2.6199(54) 2.3423(65) O(5)-Na(x) (Å) 2.3239(71) 2.4492(55) 2.3879(70) O(6)-Na(x) (Å) 2.4508(63) (1-x, 1-y, -z) 2.3208(66) (x, 1+y, z) 2.6417(57) O(7)-Na(x) (Å) 2.4651(60) 2.2594(71) O(8)-Na(x) (Å) 2.6453(72) 2.5904(56) 2.3113(65) (x, 1+y, z) O(9)-Na(x) (Å) 2.3031(69) 2.3848(64) (x, 1+y, z) O(10)-Na(x) (Å) 2.5933(65) (x, 1+y, z) 2.3175(68) (x, 1+y, z) 2.6625(56) (x, 1+y, z) Table S2. K-O dis ances in he polyhed a. In b acke s he co esponding symme y ope a ions a e gi en, while in he cases o (x, y, z) no in o ma ion is gi en. K(1) K(2) coo dina ion numbe 6 6 O(1)-K(x) (Å) 2.7682(89) 2.6079(75) and 2.7523(75) (x, 1+y, z) O(2)-K(x) (Å) 3.0504(89) 2.6684(70) (0.5+x, 1-y, z) and 2.7046(70) (0.5+x, 2-y, z) O(3)-K(x) (Å) 2.6936(78) (0.5+x, 1-y, z) and 2.7886(78) (0.5+x, 2-y, z) 3.1386(90) (0.5+x, 2-y, z) O(4)-K(x) (Å) 2.7423(65) and 2.7462(65) (x, -1+y, z) 2.6844(64) (0.5+x, 2-y, z) mo phology o he la ge c ys als ob ained, o he fi s ime allowed o eco e ing o a phase pu e sample o o m III by mechanical sepa a ion. Phase pu i y was c oss-checked by PXRD (Figu e S2, Suppo ing In o ma ion) and he uni cell was efined. The phase pu e sample o o m III was used o collec IR- and Raman spec a and o eco d he moanaly ical da a and compa e i wi h da a o o m I. P e iously, we ied o iden i y cha ac e is ic ib a ions o o m III by compa ing spec a o pu e o m I and mix u es o o m I and o m III. 13 Un o una ely, he IR spec a showed no significan diffe ences. Compa ing he spec a o he pu e phases (Figu e 4) howe e e eals dis inc shi s o equencies. This will allow o decon olu ion o bands o e lapping in mix u es ( o ins ance a app oxima ely 1180 cm−1and 920 cm−1) whe e he equencies o pu e phases diffe by as much as 8 cm−1, and consecu i e quan i a i e analysis by simple IR. Raman spec a (Figu e 5) e en showed well sepa a ed cha ac e is ic bands o bo h o m I and III which will offe e en lowe de ec ion limi s in mix u es. DSC (Figu e 6 and Table S2, Suppo ing In o ma ion) o sepa a ed pu e o m III in closed Al pans (hea ing a e:1 °C/ min) ga e an onse empe a u e o mel ing o To= 126.1 (±0.2) °C and he ela ed en halpy was de e mined o be ΔH= 187.8 J·g−1. No indica ions o o he phase ansi ions we e ound; in pa icula , no ans o ma ion o o m I p io o mel ing could be obse ed. Please no e ha mechanical sepa a ion limi ed he a ailable sample sizes o o m III. Figu e 2. Ligh mic oscopy o he c ys alliza ion p oduc o benzamide in benzene (cooling induced nuclea ion ollowed by c ys al g ow h by slow e apo a ion). Pla ele s (ma ked by he ed box) co espond o o m I, while needles (ma ked by he g een box) co espond o o m III. Figu e 3. Compa ison o he c ys al s uc u e o o m III o benzamide as de e mined by labo a o y powde da a (50:50 blend wi h o m I) (a) and single-c ys al da a (b). Selec ed in e molecula angles and dis ances be ween phenyl planes a e shown. Figu e 4. IR spec a o o m I ( ed) and o m III (g een) o benzamide. Figu e 5. Raman spec a o o m I ( ed) and o m III (g een) o benzamide. Figu e 6. Typical DSC cu es o o m I ( ed) and o m III (g een) measu ed wi h 1 °C/min. The endo he mal signal a o m I has go one b oad shoulde be o e he mel ing, he e a e wo successi e endo he mal e en s. In he inse his can be poin ed ou by using small hea ing a es (0.1 °C), and he shoulde can be esol ed as a second endo he mic peak. In compa ison he mel ing o o m III showed a single peak. C ys al G ow h & Design A icle dx.doi.o g/10.1021/cg3009706 |C ys . G ow h Des. 2012, 12, 5365−53725368 Consequen ly, he e o s o he moanaly ical alues o o m III a e much highe as compa ed o alues o o m I. The diffe ences o empe a u es o he peak maximum he e o e ce ainly a e insignifican . The onse empe a u e o he fi s endo he mic e en obse ed o o m I was ound o be 123.5 (±0.2) °C and he ela ed en halpy was de e mined o be ΔH= 187.7 (±0.9) J·g−1(Figu e 6 and Table S2, Suppo ing In o ma ion). This endo he mic signal o om I was, howe e , unusually b oad and showed a shoulde . Lowe ing he hea ing a e o 0.1 °C/ min (inse o Figu e 6) imp o ed he esolu ion and p o ed ha clea ly wo successi e endo he mic e en s a e in ol ed wi h an onse empe a u e o he second peak o abou 125.6 (±0.2) °C. The na u e o he endo he mic e en was, howe e , no appa en a his poin . En apped benzene sol en molecules can be clea ly uled ou as a possible cause. Amoun s o sol en sufficien o p oduce he a he s ong he mal signal would clea ly be de ec able by a weigh loss in he TG, o by IR- and NMR-spec oscopy. No e idence was, howe e , ound wi h any o hese analy ical echniques. An ob ious explana ion would be an enan io opic ans- o ma ion o o m I in o o m III wi h he la e being he high empe a u e phase. Consul ing Bu ge s law 23 was impossible because bo h hea s o usion as well as mel ing poin s we e iden ical wi hin expe imen al e o s. Compa ing he ene gy o bo h s uc u es a e ene gy minimiza ion applying high-le el DFT calcula ions using dispe sion co ec ion (see Modeling De ails sec ion), o m III was ound o be dis a o ed o e o m I by as li le as 0.5 kJ/mol (pe molecule). While his indica es ha a leas a 0 K o m I should be he mo e s able s uc u e, howe e , such a small ene gy diffe ence canno ully de e mine he ela i e he modynamic s abili y a highe empe a u es. T ying o es ablish he ela i e he modynamic s abili y by he momic oscopy, we un o una ely s uggled wi h he high apo p essu e o benzamide close o he mel ing poin . Bo h o ms s a ed subliming be o e mel ing a a empe a u e ha depended s ongly on he size o he c ys als. To limi sublima ion e en ually he momic oscopic expe imen s we e pe o med wi h he samples fixed be ween wo glass pla es ha we e closed wi h glue a he edges. In his case no app eciable sublima ion was obse ed be o e mel ing. E en hen i was impossible o decide which o he wo o ms mel s fi s . This is no su p ising in he ligh o he insignifican diffe ences o empe a u es o he peak maximum in DSC. In wo expe imen s whe e bo h c ys als we e in con ac i looked like o m I g ew a he cos o o m III and ha o m III mel s be o e I (Figu e 7). Bu because o he con inuous and nea ly pa allel mel ing o bo h o ms no defini i e conclusions can be d awn om he momic oscopy. Nex , a sample o pu e o m I was hea ed in a DSC expe imen wi h a a e o 0.1 °C/min up o 124.4 °C whe e he maximum o he fi s endo he mic peak occu s and hen annealed o 30 min be o e i was analyzed by PXRD. No eflec ions o o m III could be obse ed, sugges ing ha no enan io opic con e sion o o m I in o o m III happened a he empe a u e o he fi s endo he mic peak. Finally, a long- e m solid equilib a ion o a mix u e o o m I in o o m III a a empe a u e o 100 °C moni o ed by in si u PXRD in a closed capilla y was pe o med. Because o he high apo p essu e o benzamide a his empe a u e, i was c ucial o ensu e ha he collec ed diff ac ion da a we e ep esen a i e o he whole olume o he sample. The e o e, special ca e was paid o accu a e empe a u e con ol (accu acy ∼0.1 °C), o minimiz- ing he empe a u e g adien , and o limi ing he sample olume as closely as possible o he olume ac ually i adia ed by he beam. O e a pe iod o almos one mon h a comple e con e sion o o m I was obse ed (Figu e 8). In line wi h DSC annealing expe imen s, his esul sugges ed ha o m I emains he he modynamically s able o m o e he whole empe a u e ange. Figu e 7. The momic oscopy o single c ys als o o m I (pla ele ) and III (needle) in con ac upon hea ing (hea ing a e 0.5 °C/min). I indica es ha o m III mel s a lowe empe a u es and o m I g ows a he cos o o m III bu no defini i e answe can be gi en. Figu e 8. PXRD aces o a mix u e o m I and III annealed a 100 °C o e a pe iod o 25 days. The mix u e o o m I and III con e s comple ely o o m I as indica ed by he disappea ance o he mos p ominen eflec ions being singula o o m III (ma ked by a ows). C ys al G ow h & Design A icle dx.doi.o g/10.1021/cg3009706 |C ys . G ow h Des. 2012, 12, 5365−53725369 Since all expe imen s consis en ly sugges ed ha o m I was no enan io opically con e ed in o o m III a ele a ed empe a u es, we ied o un a el he na u e o he fi s endo he mic peak obse ed o o m I on an a omis ic le el applying MD simula ions and solid-s a e NMR spec oscopy. MD Simula ions. MD simula ions on he basis o empi ical o ce fields we e execu ed o bo h polymo phs as a unc ion o empe a u e. Along his line, pe iodic simula ion models, each comp ising 576 benzamide molecules, we e in es iga ed a ambien p essu e and o e a ange o empe a u es. Because o he limi ed ime-scales accessible o MD simula ions, ela i ely high hea ing a es we e needed o obse e c ys al mel ing wi hin ou small simula ion models. Howe e , by scanning hea ing a es o 0.2, 0.1, 0.05, and 0.0025 K/ps i was possible o disc imina e a ificially la ge hys e esis effec s om sys em- a ic ends inhe en o he diffe en benzamide polymo phs (Figu e 9). Indeed, upon hea ing o ou de ec - ee, single-c ys alline, pe iodic models (exhibi ing no su aces o g ain bounda ies) bo h s uc u es displayed mel ing wi hin empe a u e diffe - ences o less han 10 °C, i.e., less han he e o ma gins we an icipa e o he modeling p ocedu e. Despi e he la ge de ia ion wi h espec o he mel ing empe a u es as ob ained om expe imen s (based on powde samples), ou simula ions a e ne e heless help ul o a ionalize he DSC expe imen s om a quali a i e poin o iew. Indeed, bo h models we e ea ed in ull analogy and hus allow a ela i e compa ison ha is much mo e eliable han he absolu e alues o he mel ing empe a u es. While he mel ing mechanism o o m III was obse ed as a con en ional diso de nuclea ion and g ow h pic u e, we iden ified p enuclea ion e en s in o m I ha may accoun o he ‘shoulde ’obse ed om he calo ime ic measu emen s a 123.5 °C be o e mel ing. By local b eaking o he hyd ogen bonded ne wo k, single benzamide molecules we e obse ed o unde go ( empo a y) local la ge angle eo ien a ion leading o he popula ion o de ec s in he else wise single-c ys alline s uc u e. The o ma ion o such de ec s equi es he b eaking o hyd ogen dime bonds and is hence subjec ed o a conside able ene gy ba ie , which is only c ossed upon hea ing ∼70 °C below he ac ual mel ing poin . Thus, we sugges he ene gy up ake obse ed om DSC expe imen s is ela ed o he popula ion o me as able de ec si es which a e kine ically inaccessible a lowe empe a u e (Figu es 10, S3, S4, and S5, Suppo ing In o ma ion). Indeed, as much as 60−70 kJ/mol de ec o ma ion ene gy was obse ed (compa e also o NMR expe imen s). ■NMR To collabo a e he simula ion esul s and o shed u he ligh on he o igin o he endo he mic e en close o he mel ing peak o o m I empe a u e dependen 1H and 13C line-shape analyses and 1H spin−la ice elaxa ion s udies we e ca ied ou o his phase. While he o me echnique p obes slow (10−7s ≤τ≤10−3s), la ge angle jumps o he benzamide molecules, spin−la ice elaxa ion de ec s elaxa ion p ocesses on he o de o he in e se lamo equency. Since he MD simula ions sugges ed he o ma ion o mobile poin de ec s caused by flip p ocesses o he molecule which will effec i ely dampen he la ice ib a ions especially elaxa ion measu emen s a e p omising. In his espec he la ge gy omagne ic a io o p o ons inc eases he sensi i i y allowing p obing e en small de ec concen a ions. The e olu ion o he p o on spec a du ing hea ing is shown in Figu e 11. A lowe empe a u es b oad solid-s a e spec a ypical o igid benzamide molecules a e obse ed up o a empe a u e o 113 °C. Upon u he hea ing a na ow line a ises a 118 °C and apidly gains in ensi y un il he sample mel s a 126 °C. Be ween 113 and 123 °C he ull wid h a hal maximum (FWHM) is a ound 1 kHz and hen na ows o 400 Figu e 9. Mel ing empe a u e as ob ained om MD simula ions, including ex apola ion o anishing hea ing a es (Tm o m I: 306 °C, Tmel o m III: 316 °C). Figu e 10. De ec o ma ion a 510 K o o m I ( ed); snapsho a 510 K. Figu e 11. 1H wide-line spec a as unc ion o empe a u e. Abo e 113 °C a na ow line appea s indica ing ha a pa o he molecules is unde going a nea ly iso opic eo ien a ion. The solid s a e pa keeps i s line shape un il mel ing, while he na ow line gains in in ensi y. C ys al G ow h & Design A icle dx.doi.o g/10.1021/cg3009706 |C ys . G ow h Des. 2012, 12, 5365−53725370 Hz in he mel a 126 °C and abo e. The line-shape o he b oad esonance does no change significan ly o e he en i e empe a u e ange (RT −126 °C). Since he na ow signal appea s al eady well below Tm≈126 °C(Δ=13°C) pa ial mel ing seems o be an unlikely explana ion a leas below 120 °C. This hypo hesis is suppo ed by he significan change in he line wid h o he na ow esonance abo e 123 °C when he mel ing p ocess se s in. The same effec is obse ed o he 13C wide-line spec a as a unc ion o empe a u e (Figu e S6, Suppo ing In o ma ion). Thus a pa o he benzamide molecules wi hin o m I become mobile al eady in he solid s a e. In acco dance wi h his obse a ion he 1H spin−la ice elaxa ion imes exhibi ed a biexponen ial beha io . The ac ion o he as elaxa ion p ocess (Figu e 12) co esponds o he in ensi y a io de i ed om he wide-line spec a. Abo e 119 °C he amoun o he mobile species inc eased apidly. In he same empe a u e egion he p epeak appea ed in he DSC expe imen s o o m I. The long and sho mean <T1> elaxa ion imes a e shown in Figu e 13a. I can be clea ly seen ha he mean elaxa ion ime o he solid phase <T1l> dec eases wi h inc easing empe a u e (slow mo ion limi ), whe eas he elaxa ion ime o he mobile pa T1k shows he opposi e beha io ( as mo ion limi ), becoming la ge nea he mel ing poin . Consequen ly he molecules in ol ed in he slow elaxa ion p ocess mo e wi h a a e cons an smalle han he la mo equency while he ones pa icipa ing in he as elaxa ion eo ien wi h a a e which is la ge han ω0. The la e obse a ion is in line wi h he o ma ion o poin de ec s in o m I whe e he molecules a e capable o slow, nea ly iso opic eo ien a ions (Figu e 10). Such a beha io ine i ably leads o a b eak-up o he molecula pai s ypical o he s uc u e o o m I. This ag ees wi h he finding o he MD simula ions ha he o ma ion o he de ec s equi es b eaking o he H-bonds since hese bonds would p ima y supp ess a ee molecula mo ion. In con as in he mel he molecula eo ien a ions a e expec ed o be much as e han ω0. A mo e de ailed examina ion o he <T1l> plo e eals wo addi ional in e es ing ea u es. The fi s aspec is he cu e p og ession i sel . Up o a empe a u e o 121 °C he dec ease o he mean elaxa ion ime ollows a linea beha io un il i d ops o sho e elaxa ion imes (Figu e 13b inse ). Ou explana ion o his unexpec ed d op o he elaxa ion ime akes in o accoun he apidly inc easing amoun o de ec s al eady shown in he inse in Figu e 12. Below 121 °C he pe cen age o he de ec s is low and hei influence on he long <T1>canbeneglec ed.F om122°C on he de ec concen a ion becomes high enough ha he as elaxa ion p ocess seemingly educes <T1l>. This is p obably caused by spin-diffusion be ween p o ons in he de ec egion wi h a sho T1and p o ons belonging o immobile molecules packed in he egula la ice (long T1). The spin diffusion media ed p ocess leads o an efficien equilib a ion o he elaxa ion imes on in e media e leng h scales up o a ew nanome e s. Such an in ima e mixing on he o he hand is no possible assuming ha pa ial mel ing only su ace de ec s would be esponsible o he na ow line below he mel ing poin . Bo h p ocesses would only affec he su ace bu no he bulk o he solid phase and would lead o a significan smalle T1 educ ion. This is also in line wi h in e p e ing he c componen displayed in Figu e 12 as a measu e o he de ec concen a ion well below he mel ing poin . Fu he mo e, in he low empe a u e egion he ime cons an T1l ollows an A henius beha io wi h an ac i a ion ba ie o 55(3) kJ/mol (Figu e 13b). No e ha his alue is in easonable ag eemen wi h he de ec o ma ion ene gy as Figu e 12. F ac ion c o he as elaxing componen o he biexponen ial spin−la ice elaxa ion da a as a unc ion o he empe a u e. The inse demons a es ha he amoun o he mobile phase inc eases apidly wi h an exponen ial end (compa e inse ). Figu e 13. (a) Mean spin−la ice elaxa ion imes <T1> o he as (T1k) and slow (T1l) elaxa ion componen s as a unc ion o 1/T. (b) Plo o he mean elaxa ion ime <T1l> wi h a linea fi in he “low” empe a u e egion. The slope co esponds o an ac i a ion ene gy o 55(3) kJ/mol. The inse closes up on he de ia ion o he <T1l> om he linea end (blue line as guide o he eye). C ys al G ow h & Design A icle dx.doi.o g/10.1021/cg3009706 |C ys . G ow h Des. 2012, 12, 5365−53725371 obse ed om MD simula ions. The la ge EAindica es ha he benzamide molecules unde go a slow la ge angle jump a low empe a u es. Ei he his p ocess is oo slow o affec he empe a u e dependen 1H and 13C wide-line spec a (Figu es 11 and S6, Suppo ing In o ma ion) o i does no influence he line wid h. In he la e case only 2D jump models emain like a 180°jump o he phenyl ing. In any case his ac i a ion ba ie migh be assumed as he lowe bounda y o he ac i a ion ene gy o he poin de ec c ea ions and is qui e close o he alue ex ac ed om he MD simula ions. ■CONCLUSION We ou lined op imized syn heses p o ocols leading o single c ys alline benzamide o o m I and o m III and hus allowing p o ound he moanalysis and i s in e p e a ion by a combina- ion o heo y and expe imen . Despi e he obse a ion o an endo he mic signal well below he mel ing poin obse ed by DSC, o m I o benzamide was confi med o be he modynami- cally s able whe eas o m III is me as able. As sugges ed by MD simula ions and suppo ed by he 1H solid-s a e NMR expe imen s, he he mosignal is a he connec ed o a significan popula ion o locally mo e mobile molecules in a me as able de ec s a e which could only be obse ed o o m I, while o m III mel s di ec ly ia a con en ional diso de nuclea ion and g ow h mechanism. Al hough he o ma ion o de ec s o he pa icula case o benzamide p o ed no o be ela ed o a phase ansi ion, he gene al impo ance o de ec s o solid−solid- ans o ma ions has been p oposed in he li e a u e. To he bes o ou knowledge, his is he fi s ime ha such de ec s in molecula c ys als, which ha e been pos ula ed o play a significan ole in selec i e polymo ph ans o ma ions ia sol en -d op g inding, 24 could ac ually be p o en bo h expe imen ally ia DSC and solid s a e NMR and om MD simula ions. ■ASSOCIATED CONTENT * SSuppo ing In o ma ion Addi ional DSC, NMR, and PXRD da a and c ys allog aphic files in CIF o ma o o m III. This ma e ial is a ailable ee o cha ge ia he In e ne a h p://pubs.acs.o g. ■AUTHOR INFORMATION Co esponding Au ho *E-mail: [email p o ec ed], [email p o ec ed], [email p o ec ed], [email p o ec ed] e langen.de. Funding We hank he Deu sche Fo schungsgemeinscha (SPP 1415) o financial suppo . No es The au ho s decla e no compe ing financial in e es . ■ACKNOWLEDGMENTS We would like o hank Ka l Kemp and P o . Raine Schobe o measu emen o he IR spec a, Clemens P esche and Leonid Dub o insky o measu emen o he Raman spec a. ■REFERENCES (1) Wohle , F.; on Liebig, J. Ann. Pha m. 1832, 249−282. (2) Aake oy, C. B. Ac a C ys allog . B 1997, 569−586. (3) Bi adha, K.; Su, C. Y.; Vi al, J. J. C ys . G ow h Des. 2011,4, 875−886. (4) Blagden, N.; Da ey, R. J. C ys . G ow h Des. 2003,6, 873−885. (5) Desi aju, G. R. J. Chem. Sci. 2010,5, 667−675. (6) Blagden, N.; de Ma as, M.; Ga an, P. T.; Yo k, P. Ad . D ug Deli e y Re . 2007,7, 617−630. (7) Haleblia, J.; McC one, W. J. Pha m. Sci. 1969,8, 911−929. (8) Hilfike , R. Polymo phism in he Pha maceu ical Indus y, 1s ed; Wiley-VCH Ve lag: Weinheim, Ge many: 2006. (9) Henck, J. O.; G iesse , U. J.; Bu ge , A. Pha m. Ind. 1997,2, 165− 169. (10) Pen old, B. R.; Whi e, J. C. B. Ac a C ys allog . 1959,2, 130− 135. (11) Da id, W. I. F.; Shankland, K.; Pulham, C. R.; Blagden, N.; Da ey, R. J.; Song, M. Angew. Chem., In . Ed. 2005,43, 7032−7035. (12) Thun, J.; Sey a h, L.; Senke , J.; Dinnebie , R. E.; B eu, J. Angew. Chem., In . Ed. 2007,35, 6729−6731. (13) Thun, J.; Sey a h, L.; Bu e ho , C.; Senke , J.; Dinnebie , R. E.; B eu, J. C ys . G ow h Des. 2009,5, 2435−2441. (14) Be ns ein, J.; Da ey, R. J.; Henck, J. O. Angew. Chem., In . Ed. 1999,23, 3441−3461. (15) Kulka ni, S. A.; McGa i y, E. S.; Meekes, H.; e Ho s , J. H. Chem. Commun. 2012,41, 4983−4985. (16) Chaudha i, S. R.; Su yap akash, N. J. Mol. S uc . 2012, 163− 168. (17) Thun, J.; Schoe el, M.; B eu, J. Mol. Simul 2008,10−15, 1359− 1370. (18) We used he pseudopo en ials C,N,O,H .pbe. jkus.UPF om he h p://www.quan um-esp esso.o g dis ibu ion. (19) G imme, S. J. Compu . Chem. 2006,15, 1787−1799. (20) Giannozzi, P. e al., J. Phys. Condens. Ma e 2009,21, 395502. (21) Benoi , D. M.; Ec o s, P.; Duchs ein, P.; B eu, J.; Zahn, D. Chem. Phys. Le . 2011,4−6, 274−277. (22) D inskikh, S.; Benini, G.; Senke , J.; Vogel, M.; Wiede sich, J.; Kudlik, A.; Rossle , E. J. Phys. Chem. B 1999,10, 1727−1737. (23) Bu ge , A.; Rambe ge , R. Mik ochim. Ac a 1979,3−4, 259−271. (24) T ask, A. V.; Shan, N.; Mo he well, W. D. S.; Jones, W.; Feng, S. H.; Tan, R. B. H.; Ca pen e , K. J. Chem. Commun. 2005,7, 880−882. C ys al G ow h & Design A icle dx.doi.o g/10.1021/cg3009706 |C ys . G ow h Des. 2012, 12, 5365−53725372 The moanaly ical E idence o Me as able Molecula De ec s in Fo m I o Benzamide Ch is ian Bu e ho , § Thomas Ma in, § Philipp Ec o s, # Di k Zahn, # Paul Niemie z, ‡ Jü gen Senke , ‡ Ch is ian Nä he , † and Jose B eu. § * § Ino ganic Chemis y I, Uni e si y o Bay eu h, Uni e si ä ss aße 30, 95440 Bay eu h, Ge many. # Theo e ical Chemis y/Compu a ional Chemis y Cen e , Uni e si y o E langen-Nu embe g, Nägelsbachs aße 25, 91052 E langen, Ge many. † Ino ganic Chemis y, Ch is ian-Alb ech s-Uni e si y Kiel, Max-Ey h-S aße 2, 24098 Kiel, Ge many. ‡ Ino ganic Chemis y III, Uni e si y o Bay eu h, Uni e si ä ss aße 30, 95440 Bay eu h, Ge many. Suppo ing In o ma ion Table S1. Selec ed c ys allog aphic da a o he single c ys al e inemen o o m III o benzamide. o m III c ys al shape needle molecula o mula C 7 H 7 NO o mula weigh 121.14 empe a u e (K) 173(2) c ys al sys em monoclinic space g oup P2 1 /c (No. 14) a (Å) 5.0585(10) b (Å) 5.4614(11) c (Å) 22.833(5) α (deg) 90 ß (deg) 103.85(3) γ (deg) 90 olume (ų) 612.5(2) Z 4 calcula ed densi y (mg·m - 3 ) 1.314 µ (mm - 1 ) 0.089 R in 0.0721 GOF 0.830 e lec ions collec ed/unique 4399/1184 inal R indices [I > 2σ(I)] 0.0398/0.0790 R indices (all da a) 0.0898/0.0917 la ges esidual/e (Å - 3 ) 0.165 Table S2. Resul s o he DSC measu emen s T o /°C ∆ ∆∆ ∆H / Jg - 1 Fo m I (1. hea ing) a 123.5 (0.2) / 125.6 (0.2) 187.7 ( 0.9 ) Fo m I (2 and 3. hea ing) a 123.4 (0.2) 18 6 . 3 ( 0.7 ) Fo m III (1. hea ing) c 126.1 (0.2) 187.8 T o = onse empe a u e; ∆H = mel ing/ ansi ion en halpy; Fo he empe a u es he maximum de ia ion and o he en halpies he a e age e o o he a e age alue is gi en in pa en heses. Two onse empe a u es a e gi en o hose measu emen s whe e he wo successi e e en s can easonable be sepa a ed a ∆H was de e mined om 3 di e en samples measu ed a 1, 3, 5 and 10 °C/min. b ∆H was de e mined om 3 di e en samples measu ed 1 °C/min. Fo hese ew measu emen s no a e age e o o he a e age alue can be gi en. Figu e S1. O e lay o DFT elaxed o m III s uc u es: (le along b-axis, igh along a-axis) ed: single c ys al e inemen , g een: powde e inemen . Figu e S2. PXRD pa e n o he sample o o m III ob ained by sepa a ing he needle shaped c ys alli es om he pla ele shaped c ys alli es o o m I. The g een icks indica e he expec ed posi ions and ela i e in ensi ies. Figu e S3. T ajec o y snapsho s: mel ing o o m I s a ing a 580 K (phenyl no shown o cla i y). Figu e S4. T ajec o y snapsho s: mel ing o o m III s a ing a 592.5K (phenyl no shown o cla i y). Figu e S5. 0 Kel in op imiza ion o a de ec ( ed) in o m I. le : snapsho a 510 K , igh : op imized 0 K s uc u e wi h de ec .