Water soluble quantum dots for bio-imaging An NMR study [PDF]

Department of inorganic and physical Chemistry Physics and chemistry of nanostructures

Water soluble quantum dots for bio-imaging An NMR study

Thesis submitted to obtain the degree of Master of Science in Chemistry by

Kim De Nolf

Academic year 2011 - 2012

Promoter: prof. dr. ir. Zeger Hens Copromoter: prof. dr. José C. Martins Supervisors: Sofie Abé, Antti Hassinen and Freya Van Den Broeck

Department of inorganic and physical Chemistry Physics and chemistry of nanostructures

Water soluble quantum dots for bio-imaging An NMR study

Thesis submitted to obtain the degree of Master of Science in Chemistry by

Kim De Nolf

Academic year 2011 - 2012

Promoter: prof. dr. ir. Zeger Hens Copromoter: prof. dr. José C. Martins Supervisors: Sofie Abé, Antti Hassinen and Freya Van Den Broeck

The author and promotors give the permission to use this thesis for consultation and to copy parts of it for personal use. Every other use is subject to the copyright laws, more specifically the source must be extensively specified when using from this thesis. De auteur en promotoren geven de toelating deze scriptie voor consultatie beschikbaar te stellen en delen ervan te kopi¨eren voor persoonlijk gebruik. Elk ander gebruik valt onder de beperkingen van het auteursrecht, in het bijzonder met betrekking tot de verplichting uitdrukkelijk de bron te vermelden bij het aanhalen van resultaten uit deze scriptie. Gent, Juni 2012

The promotors

The author

Prof. dr. ir. Z. Hens

Kim De Nolf

and Prof. dr. J. Martins

i

ii

Preface It has been a very exciting year in many ways. Full of courage, I started the last milestone before obtaining a master’s degree: a thesis. I would like to thank prof. Zeger Hens and prof. Jos´e Martins to make it possible for me to do this thesis. At first everything was very new and exciting. I had to learn to work with the spectrometer and how to synthesize quantum dots. I would like to give a special thanks to my supervisors: Antti and Sofie, who learned me the tricks of the trade of the quantum dot synthesis and Freya, who taught me how to obtain nice looking NMR spectra. Furthermore, I would like to thank all the members of the PCN and NMR STR groups because all of them helped me at some point this year and they made me feel very welcome in the groups. Later on I had to synthesize some polymers but it wasn’t possible for me to do the synthesis in our lab. Therefore I would like to express my gratitude to prof. Hoogenboom, Qilu Zhang and Tom Debruyne for providing the equipment and helping me with the synthesis of the polymer. Of course there were ups and downs during this year so I would like to thank my friends and family, and especially Zef for giving me the support I needed. Finally, I would like to go back to the period before university. I would like to thank Dirk Jacobs and my sister Sas for encouraging me to get a good education and Ines De Schepper for the general advice she gave me.

Kim De Nolf Gent, June 2012

iii

iv

Water Oplosbare Quantum Dots Voor Bio-beeldvorming – Een NMR Studie K. De Nolfa, Freya Van Den Broeckb, Antti Hassinena, Sofie Abéa, J. C. Martinsb en Z. Hensa a

Department Anorganische en fysische chemie, Universiteit Gent, 9000 Gent, België b Department Organische chemie, Universiteit Gent, 9000 Gent, België Quantum dots kennen al veel toepassingen. Zo, kunnen ze bijvoorbeeld gebruikt worden in het vakgebied van de bio-beeldvorming. Hiervoor moeten ze oplosbaar zijn in een waterig midden. Een methode, waarbij quantum dots ingesloten zitten in een polymeerschil, werd recent ontwikkeld door Parak et al. (1,2) Het doel van deze thesis is om op te helderen hoe de polymeerschil rond de Qdot gevouwd wordt en hoe het polymeer interageert met de organische liganden door middel van analyse met hoge resolutie NMR. Inleiding

Quantum dots (Qdots) zijn half geleider nanokristallen. Ze hebben een kristallijne strucutuur en een grootte van enkele nanometers. De nanokristallen kunnen gedispergeerd worden in een vloeistof als ze gestabiliseerd worden door liganden. Liganden zijn doorgaans organische moleculen die geabsorbeerd zijn aan het kristaloppervlak. Ze vormen een barrière tussen de Qdot en de omgeving. Qdots zijn zeer interessante materialen want hun eigenschappen varieren met hun grootte omwille van kwantum opsluiting. Zo zal de golflengte van het geabsorbeerde en geëmitteerde licht kleiner worden naarmate dat de Qdot kleiner wordt. Door deze eigenschap kunnen ze gebruikt worden in de biologie en de geneeskunde. Voor deze toepassingen moeten de deeltjes oplosbaar zijn in water.

Figuur 1. Als de Qdot groter wordt, zal het licht absorberen met een grotere golflengte. Deze figuur toont CdSe Qdots met verschillende groottes, gerangschikt van klein (links) naar groot (rechts).

Fluoresentie labeling is een van de toepassingen van Qdots in de biologie en geneeskunde.(1) Deze labels verstrekken de mogenlijkheid om bepaalde structurele eenheden van een cel te visualiseren. Hierbij worden Qdots gebonden aan een receptor molecule die op zijn beurt kan binden aan een target.

v

De Qdots hebben veel voordelen tegenover klassieke organische kleurstoffen omdat ze betere fotoluminescentie eigenschappen vertonen. Eerst en vooral vertonen ze minder photobleaching. Ten tweede is de vervaltijd van de Qdots veel langer dan de vervaltijd van de autofluorescentie. Autofluorescentie is de natuurlijke emissie van licht teweeggebracht door de biologische structuren zelf. Dit geeft als voordeel dat time-gate imaging gebruikt kan worden, waardoor de autofluorescentie achtergrond kan gereduceerd worden. Daarenboven is de breedte van het emissiespectrum heel nauw, waardoor Qdots met verschillende groottes simultaan kunnen gebruikt worden om verschillende structuren in de cel te visualiseren. Traditioneel worden Qdots gesynthetiseerd in organische solventen zoals tolueen omdat de deeltjesgrootte, de spreiding op de grootte en de vorm in dit soort solventen beter gecontroleerd kan worden. Om de Qdots in water over te brengen bestaan er verschillende methoden. Zo kan men de organische liganden uitwisselen voor andere liganden die een hydrofiele groep bezitten.(2) Anderzijds kunnen Qdots ook omsloten worden door een silica schil.(3) Met deze silica schil kan de Qdot gemakkelijk verbonden worden met een biomolecule. In de literatuur wordt echter ook een derde methode beschreven waarbij de Qdots met hun organische liganden door een polymeerschil omsloten worden.(4,5) Het gebruikte polymeer moet hydrofobe zijketens bevatten die kunnen interageren met de organische liganden, maar tegelijk moet de hoofdketen van het polymeer hydrofiel zijn om het oplossen in water mogelijk te maken. Deze methode is toepasbaar op verschillende soorten Qdots aangezien de organische liganden niet vervangen moeten worden. In dit onderzoek wordt dieper ingegaan op deze laatste methode en de karakterisatie van het complete water oplosbare systeem met HR NMR aangezien dit in de literatuur nog niet beschreven werd. Het doel van deze thesis is om op te helderen hoe de polymeerschil rond de Qdot gevouwd wordt en hoe het polymeer interageert met de organische liganden. Eerst worden CdSe en core/shell CdSe/CdS Qdots gesynthetiseerd. Deze laatste bestaan uit een CdSe kern omgeven door enkele CdS lagen. De liganden van deze Qdots zijn oleinezuur (OA). Daarna wordt de methode, beschreven in de literatuur, toegepast: een polymeer wordt gesynthetiseerd en de Qdots worden omsloten met het polymeer. Zowel de Qdots in tolueen, de polymeren als de Qdots omsloten met de polymeerschil worden onderzocht met verschillende technieken waaronder UV-VIS absorptie en luminescentie spectroscopie, TEM en NMR. Aangezien de CdSe/CdS core/shell deeltjes de beste resultaten geven, worden enkel deze vermeld in dit artikel. Ook het tweede polymeer wordt hier niet vermeld. Voor deze resultaten wordt verwezen naar de thesis. Experimenteel gedeelte Apparatuur en producten Alle NMR experimenten werden uitgevoerd op een Bruker 500 MHz AVANCE III spectrometer met een 1H frequentie van 500,13 MHz. Deze spectrometer is voorzien van een 5mm BBI-z of een 5mm TXI-z probe. De UV-VIS metingen werden uitgevoerd op Perkin Elmer spectrometer lambda 2. De JEM-2200FS Transmission Electron Microscope werd gebruikt om de microscoop afbeeldingen op te nemen. Poly(isobutyleen-alt-maleïnezuuranhydride), dodecylamine, CdO en sodium boraat werden aangekocht bij Sigma-Aldrich. Strem chemicals was de leverancier van S, Merck was de leverancier van octadecylamine. S en 1-octadeceen werden aangekocht bij Alfa Aesar. Tot slot werden de gedeutereerde solventen aangekocht bij Eurisotop.

vi

Synthese van CdSe/CdS Qdots Successive ion layer adhesion and reaction (SILAR) is een van de technieken die gebruikt kan worden om een core/shell Qdot te maken. Hierbij werd 0,1018 µmol CdSe Qdots opgelost in hexaan. Het mengsel werd opgewarmd in 1,5g octadecyl amine en 5g 1-octadeceen tot 100°C onder stikstof atmosfeer. Als het hexaan verdampt was, kon de temperatuur verhoogd worden tot 225°C. Eens dat deze temperatuur bereikt was, kon CdS monolaag per monolaag gevormd worden door stapsgewijze toevoeging van de precursors. De zwavel precursor werd eerst toegevoegd en na 10 minuten wachten werd de cadmium precursor toegevoegd. Na 10 minuten was de eerste monolaag gevormd. Het toevoegen van de precursoren werd vier keer herhaald, om vier lagen te synthetiseren. Cadmium oleaat precursor. Een 0,1M precursor werd bereid door CdO op te lossen in ODE waarbij oleinezuur (OA) werd toegevoegd in een 8:1 verhouding OA:Cd. Deze oplossing werd onder stikstof atmosfeer opgewarmd tot 160°C, tot alle CdO opgelost was. Zwavel precursor. Een 0.1M precursor werd gemaakt door zwavel op te lossen in ODE bij 160°C. Synthese van het polymeer Figuur 2 toont de structuur van het polymeer poly(isobutyleen-alt-n-laurylmaleamic zuur) (P(IB-alt-LMA)) dat gebruikt werd in dit onderzoek. Het werd gesynthetiseerd door poly(isobutyleen-alt-maleïnezuuranhydride) te modificeren met dodecylamine.

Figuur 2: Structuur van poly(isobutyleen-alt-n-laurylmaleamic zuur) en de bouwstenen: poly(isobutyleen-altmaleïnezuuranhydride) en dodecylamine.

15 mmol dodecylamine werd opgelost in 100mL THF. In een tweede kolf werd 20mmol poly(isobutyleen-alt-n-laurylmaleamic zuur) opgelost in THF. Als beide poeders opgelost waren, werden de twee oplossingen gemengd en opgewarmd tot 60°C. Na een uur werd het mengsel ingedampt tot een oplossing van 30 à 40 mL overbleef. Deze oplossing werd een nacht geroerd bij 60°C. Na indampen werd het mengsel volledig gedroogd en werd het terug opgelost in dichloormethaan.

vii

Procedure om Qdots in water op te lossen. Om de Qdots in water op te lossen worden de Qdots in tolueen gemengd met de polymeeroplossing in dichloormethaan, in een verhouding van 100 repeterende eenheden per nm² van effctieve oppervlakte van de Qdots. Het mengsel werd ingedampt bij 60°C en 200 mbar. Daarna werd de vaste film opgelost in een waterige oplossing. Dit proces wordt vergemakkelijkt door roeren en opwarmen tot 60°C. De waterige oplossing die gebruikt werd in dit onderzoekt is een oplossing van natriumboraat en kaliumhydroxide in millipore water. De pH van deze oplossing ligt dan rond de 9,24, wat de pKa van boorzuur is.

Resultaten en discussie Kwalitatieve analyse van de core/shell CdSe/CdS Qdots. De core/shell Qdots gesynthetiseerd voor dit onderzoek bestaan uit een CdSe kern van 3 nm en 4 CdS lagen. Dit resulteert in deeltjes van gemiddeld 5,69 nm. In de TEM afbeeldingen (figuur 3) is te zien dat de deeltjes niet clusteren. De fotoluminescentie kwantum opbrengst (PLQY) van de deeltjes in tolueen is 43,42%.

Figuur 3: TEM afbeeldingen van CdSe/CdS Qdots

NMR analyse van de core/shell CdSe/CdS Qdots. Het 1D proton NMR spectrum van de CdSe/CdS Qdots in tolueen-d8 is afgebeeld in figuur 4. Het deeltje zelf is niet te zien in het spectrum aangezien er geen protonen in het deeltje aanwezig zijn. Het ligand, OA, is wel te zien. De resonanties zijn toegekend zoals afgebeeld in de figuur, door middel van 2D spectra. De resonanties zijn verbreed aangezien OA gebonden is aan een Qdot, waardoor het minder bewegingsvrijheid. De diffusie coefficiënt van het OA gebonden aan het deeltje is 78,32 10-12 m²/s, wat overeenstemt met een hydrodynamische diameter van 9,6 nm. In het NOESY spectrum (figuur 5) zijn drie kruispieken te zien vanuit de resonantie bij 5,6 ppm naar de resonantie in de regio tussen 0,5 en 3,5 ppm.

viii

Figuur 4: 1D proton NMR spectrum van OA gebonden aan CdSe/CdS Qdots in tolueen-d8. De resonanties zijn toegekend zoals aangetoond in de figuur. De solvent resonanties zijn aangeduid met *.

Figuur 5: NOESY spectrum van OA gebonden aan CdSe/CdS Qdots in tolueen-d8

NMR analyse van (P(IB-alt-LMA). P(IB-alt-LMA) heeft een hydrofiele hoofdketen en hydrofobe zijketens. Daardoor is het polymeer oplosbaar is verschillende soorten solventen. In solventen zoals chloroform, zal het polymeer een conformatie aannemen waarbij de zijketens naar buiten zijn gericht en de hoofdketen niet in contact is met het solvent. In solventen zoals water daarentegen, zullen de zijketens aan de binnenzijde zitten en zal de hoofdketen in contact zijn met het solvent. Het solvent zal dus een invloed hebben op de NMR spectra. Figuur 6 toont de 1D proton NMR spectra van P(IB-alt-LMA) in D2O en in chloroform-d. Door middel van 2D spectra zijn de resonanties toegekend. In chloroform is de locatie van de protonen op plaatsen 2,3 en 4 niet exact bepaald worden. Dit is een gevolg van de conformatie van het polymeer. De protonen aan de binnenzijde hebben een lagere mobiliteit waardoor deze signalen breder zijn. ix

In D2O zijn er meer resonanties zichtbaar, maar nog steeds kunnen niet alle resonanties toegekend worden. De protonen aan de binnenzijde van de conformatie (5 tot 8) zijn nu breder dan in chloroform, maar ze kunnen toch toegewezen worden aangezien ze een grote intensiteit hebben. In beide solventen is een resonantie te zien bij 7,2 ppm dat afkomstig is van een onzuiverheid. Deze onziverheid was al aanwezig in poly(isobutyleen-alt-maleïnezuuranhydride), een van de bouwstenen van het polymeer. Uit DOSY blijkt dat deze onzuiverheid dezelfde diffusiecoefficient heeft als het polymeer. De diffusie coefficient van het polymeer in chloroform-d is 121,2 10-12 m²/s, wat overeenstemt met een hydrodynamische diameter van 6,21 nm. In D2O is de diffusiecoefficient 49 10-12 m²/s, overeenstemmend met een hydrodynamische diameter van 8,89 nm. Dit wilt zeggen dat het gemiddelde polmeer sneller diffundeert in chloroform dan in water.

Figuur 6: 1D proton NMR spectra van P(IB-alt-LMA) in D2O (boven) en chloroform-d (beneden). De resonanties zijn toegewezen zoals aangeduid in de figuur. Solventresonanties en onzuiverheden zijn aangeduid met *.

Kwalitatieve analyse van de CdSe/CdS Qdots met P(IB-alt-LMA) in water. De UV-VIS absortie spectra (figuur 9, links) en de TEM afbeeldingen (figuur 8) tonen aan dat de grootte en vorm van de Qdots niet veranderd zijn tijdens de transfer naar water. De achtergrond in de TEM afbeeldingen heeft niet altijd dezelfde helderheid. De lichtere delen wijzen op de aanwezigheid van het polymeer. De DSL metingen (figuur 9, rechts) tonen aan dat de deeltjes stabiel zijn in water. De hydrodynamische diameter die uit de DSL metingen kan gehaald worden is 12,37 nm. De PLQY van de Qdots is slechts gedaald met ~6% tot 37,61%.

x

Figuur 7: TEM afbeeldingen van CdSe/CdS Qdots en P(IB-alt-LMA) in water

Figuur 9: links: UV-VIS absorptie spectrum van CdSe/CdS Qdots voor en na transfer naar water. Rechts: DLS meetingen van CdSe/CdS Qdots met P(IB-alt-LMA) in water.

NMR analyse van de CdSe/CdS Qdots met P(IB-alt-LMA) in water. Figuur 10 toont het 1D proton spectrum van CdSe/CdS Qdots met P(IB-alt-LMA) in water. Zowel de resonanties van het polymeer als de resonanties van het OA gebonden aan de Qdots zijn zichtbaar, maar de CH2 en CH3 resonanties tussen 0 en 4 ppm overlappen.

Figuur 10: 1D proton NMR spectrum van P(IB-alt-LMA) en OA gebonden aan de Qdots in D2O.

In DOSY zijn zowel de resonanties van het OA en het P(IB-alt-LMA) zichtbaar (figuur 11). Beiden hebben een diffusiecoeficient van 28,9 10-12 m²/s, wat overeen komt met een hydrodynamische diameter van 15,08 nm. Dit is een bewijs dat het de Qdots in een polymeerschil gesloten zitten. De resonantie bij 5,6 ppm is een ongekende onzuiverheid die enkel voorkomt in dit specifieke staal.

xi

Figuur 11: DOSY spectrum van P(IB-alt-LMA) en OA gebonden aan de Qdots in D2O.

Uit het NOESY spectrum, daarentegen, kan niet worden afgeleid dat P(IB-alt-LMA) met de liganden interageert (figuur 12). Er zijn vanuit de OA resonantie bij 5,2 ppm kruispieken te zien naar de regio tussen 0 en 4 ppm, maar daar overlappen de polymeerresonanties met de OA resonanties. Er kan dus geen besluit getrokken worden uit dit spectrum.

Figuur 11: NOESY spectrum van P(IB-alt-LMA) en OA gebonden aan de Qdots in D2O.

Om de eventuele interactie tussen P(IB-alt-LMA) en de liganden te bevestigen, werd besloten om een STD experiment uit te voeren. Aangezien het CdSe/CdS staal reeds geconsumeerd was voor andere experimenten, werd hiervoor het CdSe staal gebruikt, hetgeen zeer gelijkaardige resultaten vertoonde. Voor het STD experiment werd de OA resonantie bij 5,2 ppm geirradieerd. Figuur 12 toont drie refrentie spectra (Qdots in tolueen-d8, polymeer in D2O en Qdots met polymeer in D2O) en het STD experiment. Daaruit blijkt dat de saturatie werd overgedragen van het gebonden OA naar P(IB-alt-LMA) waarmee de interactie tussen de twee bevestigd is. Met enige voorzichtigheid kan gezegd worden dat dit resultaat ook voor de CdSe/CdS deeltjes zou gelden. xii

Figuur 13: (a) 1D proton NMR spectrum van OA gebonden aan de Qdots in tolueen-d8. (b) 1D proton NMR spectrum van P(IB-alt-LMA) in D2O. (c) 1D proton NMR spectrum van P(IB-alt-LMA) en OA gebonden aan de Qdots in D2O. (d) STD spectrum van P(IB-alt-LMA) en OA gebonden aan de Qdots in D2O.

Conclusie In de literatuur staat een methode beschreven om Qdots water oplosbaar te maken waarbij de Qdots ingesloten worden in een polymeerschil.(4,5) Er werd tot nu toe nog geen karakterisatie van dit water oplosbaar systeem beschreven. De karakterisatie van de Qdots in een polymeerschil door middel van hoge resolutie NMR is het onderwerp van deze thesis. Het doel van deze thesis is om op te helderen hoe de polymeerschil rond de Qdot gevouwd wordt en hoe het polymeer interageert met de organische liganden.

xiii

Eerst werden CdSe en core/shell CdSe/CdS Qdots gesynthetiseerd. De liganden van deze Qdots zijn oleinezuur (OA). De kwaliteit van deze deeltjes werd gecontrolleerd met behulp van TEM, UV-VIS absorptie en emissie spectroscopie. De spreiding op de grootte was goed en de PLQY was 43,42 %. Er kon besloten worden dat deze deeltjes geschikt waren voor dit onderzoek. De deeltjes werden ook geanalyseerd met NMR. Alle resonanties werden toegekend aan de protonen van OA. De diffusiecoefficiënt en de hydrodynamische diameter van de deeltjes werd uit de DOSY spectra gehaald. Een diffusiecoeficiënt van 78,32 10-12 m²/s stemt overeen met een hydrodynamische diameter van 9,6 nm. In het NOESY spectrum zijn 3 kruispieken te zien van de OA resonantie bij 5,6 ppm naar de regio tussen 0,5 en 3,5 ppm, waar de CH2 en CH3 resonanties zich bevinden. Daarna werd P(IB-alt-LMA) gesynthetiseerd door poly(isobutyleen-alt-maleïnezuuranhydride) te modificeren met dodecylamine. De resonanties werden toegekend en het gedrag van het polymeer in twee solventen werd bestudeerd. In solventen zoals chloroform, zal het polymeer een conformatie aannemen waarbij de zijketens naar buiten zijn gericht en de hoofdketen niet in contact komt met het solvent. De protonen aan de binnenzijde van deze conformatie hebben minder bewegingsvrijheid, waardoor ze breder zijn. Als gevolg hiervan zijn sommige resonanties niet van elkaar te onderscheiden. In solventen zoals water, daarentegen, zullen de zijketens aan de binnenzijde zitten en zal de hoofdketen in contact zijn met het solvent. De protonen aan de binnenzijde hebben minder bewegingsvrijheid maar deze resonanties zijn voldoende intens om nog zichtbaar te zijn. Vervolgens, werden de Qdots opgelost in water door middel van insluiting in P(IB-alt-LMA). De water oplosbare deeltjes werden aan een kwalitatieve analyse onderworpen. Hieruit bleek dat de deeltjes niet van vorm of grootte waren veranderd tijdens het overbrengen van de Qdots naar water. De PLQY daalde slechts met ~6% tot 37,61%. De DLS meetingen toonden aan dat de Qdots stabiel zijn in water. Ook werden deze Qdots geanalyseerd met NMR. In het 1D proton NMR spectrum waren de resonanties van P(IB-alt-LMA) en OA duidelijk aanwezig, maar de resonanties overlappen in de regio tussen 0 en 4 ppm. De diffusiecoefficiënt en hydrodynamische diameter van zowel het OA als P(IB-alt-LMA) kon uit de DOSY meetingen gehaald worden. OA heeft dezelfde diffusiecoefficiënt als P(IB-alt-LMA), waardoor kan besloten worden dat de Qdots in een polymeerschil gesloten zitten. Uit het NOESY spectrum, daarentegen, kan niet worden afgeleid dat P(IB-alt-LMA) met de liganden interageert (figuur 12). Er zijn vanuit de OA alkeenresonantie bij 5,2 ppm kruispieken te zien naar de regio tussen 0 en 4 ppm, maar daar overlappen de polymeerresonanties met de OA resonanties. Er kan dus geen besluit getrokken worden uit dit spectrum. Om de eventuele interactie tussen P(IB-alt-LMA) en de liganden te bevestigen, werd een STD experiment uitgevoerd op een staal dat nog voorhanden was, het CdSe staal. Daarbij werd de OA resonantie bij 5,2 ppm geirradieerd. Uit dit experiment blijkt dat de saturatie werd overgedragen van het gebonden OA naar P(IB-alt-LMA) waarmee de interactie tussen de twee bevestigd is. Met enige voorzichtigheid kan gezegd worden dat dit resultaat ook voor de CdSe/CdS deeltjes zou gelden.

xiv

Dankbetuiging Dit project werd financieel gesteund door het department Anorganische en fysische chemie van de universiteit Gent. De auteurs van dit artikel zouden graag Qilu Zhang en prof. Hogenboom bedanken voor hun hulp bij het synthetiseren van het polymeer. Referenties 1. Pellegrino, T., Manna, L., Kudera, S., Liedl, T., Koktysh, D., Rogach, A. L., Keller, S., et al. Nanotechnology., 4(4), p 703-707 (2004). 2. Lehmann, A. D., Parak, W. J., Zhang, F., Ali, Z., Röcker, C., Nienhaus, G. U., Gehr, P., et al., Small, 6(6), p753-762., (2010). 3. Parak, W. J., Gerion, D., & Pellegrino, T., Nanotechnology, 14, p R15 - R27 (2003). 4. Yu, W. W., Chang, E., Drezek, R., & Colvin, V. L., Biochemical and biophysical research communications, 348(3), p 781-786. (2006). 5. Aubert, T., Grasset, F., Mornet, S., Duguet, E., Cador, O., Cordier, S., Molard, Y., et al. Journal of colloid and interface science, 341(2), p 201-208. (2010).

xv

xvi

List of frequently used abbreviations Abbreviation

explenation

Cd

cadmium

CdOA

cadmium oleate

DMAP

4-Dimethylaminopyridine

DNA

deoxyribonucleic acid

DOSY

Diffusion Ordered Spectroscopy

FTIR

Fourier transform infra-red spectroscopy

NMR

High resolution solution state Nuclear Magnetic Resonance

NOESY

Nuclear Overhauser effect spectroscopy

OA

oleic acid

ODA

octadecylamine

ODE

1-octadecene

PL QY

photoluminescence quantum yield

Qdot

quantum dot

Se

selenium

TOP

trioctylphosphine

TOPO

trioctylphosphine oxide

xvii

Chapter 0. List of frequently used abbreviations

xviii

Contents preface

v

Dutch summary

v

List of frequently used abbreviations

xvii

1 Introduction 1.1 Quantum dots (Qdots) . . . . . . . . . . . . . . . . . . . . . 1.2 Applications of water soluble Qdots in biology and medicine 1.2.1 Study of cell mobility . . . . . . . . . . . . . . . . . . 1.2.2 Fluorescence labeling . . . . . . . . . . . . . . . . . . 1.2.3 Study of protein-DNA interactions . . . . . . . . . . 1.3 Water soluble quantum dots . . . . . . . . . . . . . . . . . . 1.3.1 Ligand exchange . . . . . . . . . . . . . . . . . . . . 1.3.2 Silica-coated Qdots . . . . . . . . . . . . . . . . . . . 1.3.3 Polymer coating . . . . . . . . . . . . . . . . . . . . . 1.4 Goal and content of this thesis . . . . . . . . . . . . . . . . . 2 Experimental techniques 2.1 The NMR Toolbox for nanoparticles . . . . . . . . . . . 2.1.1 NMR as spectroscopic technique for nanoparticles 2.1.2 1D proton NMR . . . . . . . . . . . . . . . . . . 2.1.3 Nuclear Overhauser effect spectroscopy (NOESY) 2.1.4 Diffusion Ordered Spectroscopy (DOSY) . . . . . 2.1.5 Saturation Transfer Difference NMR . . . . . . . 2.2 UV-Vis absorption spectrometry . . . . . . . . . . . . . . 2.3 Luminescence spectroscopy . . . . . . . . . . . . . . . . . 2.4 Other techniques . . . . . . . . . . . . . . . . . . . . . . 3 Synthesis and characterization of Qdots 3.1 CdSe Qdots . . . . . . . . . . . . . . . . . . . . . . . . 3.1.1 Synthesis I of zincblende CdSe Qdots . . . . . . 3.1.2 Qualitative analysis of CdSe Qdots (synthesis I) 3.1.3 NMR analysis of CdSe Qdots (synthesis I) . . .

xix

. . . .

. . . . . . . . .

. . . .

. . . . . . . . .

. . . .

. . . . . . . . . .

. . . . . . . . .

. . . .

. . . . . . . . . .

. . . . . . . . .

. . . .

. . . . . . . . . .

. . . . . . . . .

. . . .

. . . . . . . . . .

. . . . . . . . .

. . . .

. . . . . . . . . .

. . . . . . . . .

. . . .

. . . . . . . . . .

. . . . . . . . .

. . . .

. . . . . . . . . .

. . . . . . . . .

. . . .

. . . . . . . . . .

1 1 3 3 4 4 5 5 7 7 7

. . . . . . . . .

9 9 9 9 11 12 13 13 14 15

. . . .

17 17 17 20 20

Contents

3.2

3.1.4 Synthesis II of zincblende CdSe Qdots . . . . . . . 3.1.5 Qualitative analysis of CdSe Qdots (synthesis II) . 3.1.6 NMR analysis of CdSe Qdots (synthesis II) . . . . . Core/shell CdSe/CdS Qdots . . . . . . . . . . . . . . . . . 3.2.1 Synthesis of core/shell CdSe/CdS Qdots . . . . . . 3.2.2 Qualitative analysis of CdSe/CdS core/shell Qdots 3.2.3 NMR analysis of the CdSe/CdS core/shell Qdots .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

4 Synthesis and characterization of the polymers 4.1 Characterization of poly(isobutylene-alt-maleic anhydride) by means 4.2 Synthesis and characterization of P(IB-alt-LMA) . . . . . . . . . . 4.2.1 synthesis of P(IB-alt-LMA) . . . . . . . . . . . . . . . . . . 4.2.2 Characterization of P(IB-alt-LMA) by means of NMR . . . . 4.3 synthesis and characterization of P(IB-alt-UPA) . . . . . . . . . . . 4.3.1 synthesis of P(IB-alt-UPA) . . . . . . . . . . . . . . . . . . . 4.3.2 Characterization of P(IB-alt-UPA) by means of NMR . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

22 22 23 25 25 25 27

29 of NMR 29 . . . . 30 . . . . 30 . . . . 32 . . . . 34 . . . . 34 . . . . 35

5 Water solubilization of Qdots by means of a polymer shell 5.1 Water solubilization of Qdots by means of P(IB-alt-LMA) . . . . . . . . . 5.1.1 Protocol for dissolving nanoparticles in water . . . . . . . . . . . . 5.1.2 Calculating the volume of polymer needed for solubilization of Qdots 5.1.3 CdSe Qdots with P(IB-alt-LMA) in chloroform . . . . . . . . . . . 5.1.4 CdSe Qdots with P(IB-alt-LMA) in water . . . . . . . . . . . . . . 5.1.5 CdSe/CdS core/shell Qdots with P(IB-alt-LMA) in water . . . . . . 5.1.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2 Water solubilization of Qdots by means of P(IB-alt-UPA) . . . . . . . . . . 5.2.1 Protocol for dissolving nanoparticles in water . . . . . . . . . . . . 5.2.2 CdSe Qdots with P(IB-alt-UPA) in water . . . . . . . . . . . . . . . 5.2.3 CdSe/CdS core/shell Qdots with P(IB-alt-UPA) in water . . . . . . 5.2.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3 Saturation transfer difference NMR . . . . . . . . . . . . . . . . . . . . . .

39 39 39 40 40 42 44 45 47 47 48 51 53 55

6 Conclusion

57

Bibliography

61

A NMR data A.1 Poly(isobutylene-alt-maleic anhydride) A.2 P(IB-alt-LMA) in chloroform-d . . . . A.3 P(IB-alt-LMA) in D2O . . . . . . . . . A.4 P(IB-alt-UPA) in THF-d8 . . . . . . . A.5 P(IB-alt-UPA) in D2O . . . . . . . . .

65 65 68 71 73 76

xx

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

Chapter 1 Introduction In this chapter, the context of my research is given. First, the definition and properties of quantum dots are discussed. Next, some of the possible applications of quantum dots in medicine and biology are given. For these applications, it is required that the quantum dots are soluble in aqueous solution. Therefore, some methods to dissolve quantum dots in water are reviewed. Finally, the goal and content of my thesis is specified.

1.1

Quantum dots (Qdots)

A nanocrystal is a crystalline material with a size in the nanometer range. To disperse them as colloidal particles in a solvent, they are stabilized by ligands, which are often organic molecules. These ligands are adsorbed at the particle surface and act as an interface to the outside world. Especially semiconductor nanocrystals are interesting for a variety of applications since they behave as quantum dots (Qdots), which means that their properties vary continuously with their size due to quantum confinement. Qdots form a bridge between an atom and a bulk material in terms of the electronic energy level density. The electronic energy levels of an atom are known to be discrete whereas bulk semiconductor has a valence band and a conduction band, separated by a bandgap.[1] These bands are continuous since the level spacing is always smaller than thermal energy at operating temperatures. For Qdots however, the electronic energy levels will become discrete, the band gap will increase as the size gets smaller and the level spacing exceeds thermal energy at room temperature (figure 1.1). In a Qdot, the charge carriers (electrons and holes) are confined in the dimensions of the particle. According to the Heisenberg uncertainty relation (equation 1.1) the momentum of the particle is no longer well defined when the motion of the particle is restricted to a certain part of space; giving an increase in kinetic energy (equation 1.2).[2]

1

Chapter 1. Introduction

Figure 1.1: Schematic illustration of the density of states in semiconductor.

∆x∆p ≥

E=

p2 2m

~ 2

(1.1)

(1.2)

Hence, to create an electron-hole pair, more energy is needed when the particle is smaller and more energy is released by their recombination.[3] In other words, the absorption and emission spectra shift to shorter wavelengths as the size of the Qdot becomes smaller. The quantum size effects can thus be observed experimentally by means of the optical properties of the Qdot (figure 1.2). The photoluminescence quantum yield expresses the efficiency of the Qdot to absorb light followed by relaxation to the ground state by emitting light. Trap states at the surface of the inorganic semiconductor can cause unwanted alternative pathways for relaxation, reducing the photoluminescence quantum yield.[4] These trap states can be due to structural defects or chemical impurities.[5] Ligands can passivate these trap states since they bind to the surface atoms. The most widely used type of ligands are long chain organic ligands such as oleic acid (OA), dodecanethiol, oleylamine, trioctylphosphine oxide (TOPO), and phosphonic acids. However, passivation of the surface by embedding the core semiconductor in a shell of a different kind of semi-conductor provides even better stability.[6] The shell will serve as a physical barrier between the core and the surroundings. Hence, the Qdot is less sensitive to environmental changes or surface chemistry, thus leading to a better photoluminescence quantum yield. Since small particles with a large surface area tend to aggregate, colloidal Qdots need suitable ligands adsorbed at the surface to create a metastable state.[4] By means of steric hindrance, the aggregation of the Qdots is prevented. An important parameter defining a good ligand is the adsorption energy.[7] It is a measure for the affinity of the ligand towards 2

Chapter 1. Introduction

Figure 1.2: As the Qdot becomes larger, it will absorb light with a higher wavelength. This figure shows CdSe Qdots of different size ranked from small (left) to large (right).

the Qdot. If the energy is too low, the ligands will not stabilize the Qdots and the Qdots will tend aggregation. On the other hand, the energy should not be too large because this makes the surface inaccessible for growth. The interaction between ligands and the solvent molecules will determine the stability of the suspension.[4] Aliphatic ligands for example, will prefer interaction with a non-polar solvent thus Qdots with aliphatic ligands will be stable in non-polar solvents. On the other hand, the interaction of the aliphatic ligands with a polar solvent will be unfavorable and the particles will tend towards aggregation. The Qdots can thus be precipitated and resuspended by adding solvents. This is a suitable technique to purify the Qdot suspension.

1.2

Applications of water soluble Qdots in biology and medicine

Rapid development in the research of quantum dots has led to a wide range of applications, including these in biology and medicine. For these applications, the Qdots are used in an aqueous dispersion. Therefore, the Qdots must be soluble in aqueous solutions. A few examples of these applications are described below.

1.2.1

Study of cell mobility

It is shown that Qdots can be incorporated in living cells, while remaining fully luminescent. [8] The detailed process of the up-take mechanism is still unclear. Nevertheless, this uptake mechanism can be used to study cell mobility by a method based on phagokinetic tracks as suggested by Albrecht-Buehler.[9, 10] A cell migrates over a layer of markers and ingests them, leaving behind a blank track (figure 1.3). This method can be applied to distinguish slow moving healthy cells from faster moving cancer cells. 3

Chapter 1. Introduction

Figure 1.3: The images were collected with a confocal microscope using fluorescence detectors to examine the Qdots (B,E) and DIC (differential interference contrast) to visualize the cells (C,F). The merged images show the layer of Qdots and the cells (A,D). After 24h, large clearings in the Qdot layer are observed around the cells (D,E). The Qdot-filled tumor cells are also fluorescing brightly after 24h (E).[8]

1.2.2

Fluorescence labeling

Fluorescence labeling provides the opportunity to visualize structural units inside a cell which are not visible in a normal image. A fluorescent dye is attached to a receptor that will selectively bind to the target.[11] This receptor-target bond is based on a very fascinating phenomenon in biology that is called molecular recognition. Biological molecules can bind with each other with a high level of affinity and selectivity. It can be compared to a keyand-lock model. By attaching a Qdot to a certain receptor, these systems can be studied in detail. Fluorescent Qdots show a lot of advantages in comparison to traditional organic dyes due to their superior photoluminescence properties. First of all, they exhibit less photobleaching. Secondly, the decay time of the fluorescence of Qdots is longer than the decay time of autofluorescence. Autofluorescence is the natural emission of light by biological structures. Therefore time-gate imaging can be used to reduce the autofluorescence background in fluorescence imaging of cells. In addition, the bandwidth of the Qdots’ emission spectrum is very narrow. In this way, Qdots with different sizes can be used in parallel. Different types of antibodies will be labeled with Qdots of different size, resulting in a color difference. As a result, more than one structure in a cell can be visualized simultaneously.

1.2.3

Study of protein-DNA interactions

Interaction between DNA and proteins is the subject of much research in the medical research field.[12–15] Single-stranded DNA is a linear chain of nucleotides that consists of a sugar phosphate backbone and a purine of pyridine base. There are four different 4

Chapter 1. Introduction bases: adenine (A), guanine (G), cytosine (C) and thymine (T). The oligonucleotide is defined by the order of the bases. Oligonucleotides can form double-stranded DNA when two complementary oligonucleotides bind to each other.[11] Qdots can be covalently bound to DNA binding proteins. Jason Taylor and colleagues describe a way to study specific sequences on single DNA molecules.[16] Qdots conjugated to a histone protein will bind to the DNA in a random fashion (figure 1.4a). Since histone proteins bind by means of strong electrostatic interactions this occurs at arbitrary positions along the DNA sequence. However, if the Qdots are bound to the protein EcoRI, specific sequences in the DNA can be studied (figure 1.4b and c), since EcoRi recognizes specifically the GAATTC sequence, before cleaving the DNA strand. By attaching a Qdot to the protein at this interaction site, the interaction between the DNA and the protein can be studied by means of multicolor fluorescence imaging.

1.3

Water soluble quantum dots

Traditionally, the Qdots are synthesized in an organic solvent because there is a high degree of control over particle size, shape and size dispersion. After synthesis the Qdots should thus be transferred to water before they can be applied in the field of bio-imaging. A few ways to make Qdots water soluble are described below.

1.3.1

Ligand exchange

One approach to make quantum dots soluble in water is replacing the original ligands used during the synthesis with other ones (figure 1.5). Ligands are usually exchanged via a phase transfer procedure involving two immiscible solvents. A colloidal dispersion of Qdots is prepared in a non-polar solvent such as toluene. A polar solvent such as water serves as a solvent for the new ligands. These new ligands must be insoluble in the non-polar solvent. The two immiscible solutions are mixed and stirred. After a while, the Qdots transfer from one solvent to another, provided that the ligands are exchanged. In literature a lot of inorganic ligands are suggested for this purpose. In many cases, these ligands are bifunctional.[17] The first functional group will interact with the Qdot and the second one is hydrophilic or can be used to interact with biomolecules. Aminoethanethiol [18], oligomeric phosphines [19] and peptides [20] are some examples of such ligands. Unfortunately, an appropriate ligand has to be found for every individual material and the new ligand can alter the physical and chemical states of the surface, leading to a reduced photoluminescence quantum yield.[17, 21]

5

Chapter 1. Introduction

Figure 1.4: a. DNA (green) stretched on glass is incubated with histone-conjugated nanoparticles (yellow). b. Single nanoparticles bound to a specific site on a stretched DNA molecule. c. Specific binding of multiple nanoparticles on single DNA molecules. The experimental conditions were similar to those in (b.) except that higher nanoparticle concentrations and longer incubation times were used.[16]

Figure 1.5: Schematic representation of a Qdot with its original ligands (top), after ligand exchange (bottom left), with a silica shell (bottom middle) and with a polymer coating (bottom right).

6

Chapter 1. Introduction

1.3.2

Silica-coated Qdots

The Qdots can be embedded in a silica shell as well (figure 1.5). The first step of this alternative approach includes the exchange of the original ligand with silane molecules.[22] The functional group that binds to the Qdot surface has to be adjusted to the kind of Qdot that is used. The silica shell can then be grown by means of the silane ligands. A third step involves the cross-linking of the silane molecules to create a dense siloxane shell. With a silica shell the Qdots are soluble in a wide pH range and the optical properties are preserved.[23] The silica-coated Qdots can easily be conjugated to biomolecules as well.[24] The fact that the Qdot is not in physical contact with the surroundings is certainly an advantage of this approach since it provides photostability. However, since the synthesis of uniform thin silica shells is very complex, this method is still under development.

1.3.3

Polymer coating

Another method is suggested in the literature by Parak et al.[21, 25, 26] Here, the Qdot is coated with a polymer (figure 1.5). This polymer has hydrophobic side chains to interact with the organic ligands of the Qdot and a hydrophilic backbone to make the system soluble in water. It is applicable to any kind of Qdot since the original ligands are not replaced. Using this approach, the Qdot material does not come into contact with the body thus every kind of Qdot can be used. This advantage may widen the scope of the use of Qdots within bio-imaging.

1.4

Goal and content of this thesis

The method to dissolve particles in water, described in section 1.3.3, will be used in this thesis. In literature, no real characterization of the water soluble system has been reported. Therefore, the characterization of the water soluble system will be the subject of my research. The goal of this thesis is to elucidate how the polymer shell is folded around the Qdot and how the polymer interacts with the ligands. First, CdSe Qdots and CdSe/CdS core/shell Qdots are synthesized. Then, the method to dissolve quantum dots in water, as reported in literature, is performed: a polymer with hydrophilic backbone and hydrophobic side chains is synthesized and the Qdots are embedded in a polymer shell. The Qdots in toluene, the polymer itself and the water soluble system are analyzed by means of UV-Vis and luminescence spectroscopy, TEM and NMR. The experimental techniques and their application to Qdots are explained in chapter 2. In chapter 3 it is explained how the Qdots are synthesized and a quality control is performed 7

Chapter 1. Introduction before analysis of the Qdots by means of NMR. The synthesis and characterization of the polymers is explained in chapter 4. Subsequently, chapter 5 deals with the protocol to dissolve Qdots in water and the characterization of the water soluble system by means of NMR. Finally, The conclusion of the performed research can be found in chapter 6

8

Chapter 2 Experimental techniques The Qdots and the polymers that are synthesized need to be characterized by various techniques such as UV-Vis and luminescence spectroscopy, TEM and NMR. This chapter provides an overview of these experimental techniques and their applicability to Qdots.

2.1 2.1.1

The NMR Toolbox for nanoparticles NMR as spectroscopic technique for nanoparticles

High resolution solution state Nuclear Magnetic Resonance (NMR) is a very powerful tool to study the ligands of quantum dots.[27–30] The quantum dots themselves are not studied because this would only be possible with solid state NMR methods. In contrast to the Qdots, the organic ligand can be studied using solution state NMR. First of all, a full structural analysis can be performed by means of the different NMR techniques. Subsequently the behavior of the ligands can be studied. For example, a distinction can be made between free and bound ligands or one can also study whether the ligands are exchanging or not. All NMR experiments were performed on a Bruker 500 MHz AVANCE III spectrometer generating a 1H frequency of 500.13 MHz, equipped with a 5mm BBI-z or a 5mm TXI-z probe. The NMR samples in toluene-d8 were prepared by evaporating the original solvent followed by adding deuterated solvent. The samples in water were prepared by using D2 O from the start or adding 10% D2 O to the sample. The samples were always measured at room temperature (25°C).

2.1.2

1D proton NMR

The one dimensional proton NMR spectrum of ligands bound to a Qdot has a different appearance than a spectrum of the free species in solution. The first thing to notice is the 9

Chapter 2. Experimental techniques line width of the signals. This is determined by the relaxation rate of spin polarization in the xy plane, also known as T2 relaxation. Since the line width depends on the tumbling rate of a molecule in solution and larger molecules tumble more slowly than others, the tumbling rate of the bound ligands is smaller than the rate of the free ligands. Therefore, the line width of the signals corresponding to bound ligands will be broader. An example is shown in figure 2.1. (a) depicts the spectrum of free oleic acid (OA) in toluene. OA is a typical ligand used to stabilize Qdots in solution. Spectrum (b) is the proton spectrum of CdSe nanoparticles capped with OA in toluene. Some signals in spectrum (b) are broader than the corresponding signals in spectrum (a). These resonances correspond to OA bound to the nanoparticle surface, labeled as indicated in the figure. Note that the resonances of OA bound to the Qdot also have a slightly different chemical shift with respect to free OA. The cause of this shift is the change in chemical environment in which the ligand is located when going from solution to ligand shell. When the ligand is free in solution, it is completely surrounded by toluene molecules. This is an aromatic environment. When OA is bound to the Qdot, it is mostly surrounded by other ligands. This is an aliphatic environment. This difference in environment causes a change in chemical shift in the signals. In quantitative 1D proton NMR measurements, the integration value of an NMR resonance is proportional to the number of nuclei corresponding to that resonance. In this way, the concentration of a certain species in solution can be determined. The determination of the concentration is a relative measurement and instead of using an internal standard, it is easier to use the ERETIC (Electronic REference To access In Vivo Concentrations) method.[31] ERETIC generates a synthetic signal with a certain area in the spectra. Then the area of this synthetic signal is calibrated against a reference solution with a known concentration. In the second stage, a quantification of the unknown concentration can be performed indirectly against the reference. Subsequently, when the concentration of the ligands and the volume V of the sample is known, the ligand density can be calculated. The number of particles NQdots , the number of ligands Nligands and the surface area S can be calculated via equations 2.1, 2.2 and 2.3. These parameters make it possible to calculate the ligand density as shown in equation 2.4 NQdots = V CQdots Na

(2.1)

Nligands = V Cligands Na

(2.2)

2 S = 4πrQdot

(2.3)

10

Chapter 2. Experimental techniques

Figure 2.1: (A) One-dimensional 1H spectrum of OA in toluene-d8. (B) One-dimensional 1H spectrum of a CdSe Qdot suspension in toluene-d8. Next to the different protons of OA (labeled as indicated in the figure), we identify residual solvent resonances (†) and a proton pool related to H2O contamination (‡). The inset shows an overlay of the alkene proton resonance at 5.65 ppm at a concentration of 50 µM (blue line) and after a 30-fold dilution down to 1.67 µM (gray background).[29]

ligand density =

2.1.3

Nligands NQdots∗S

(2.4)

Nuclear Overhauser effect spectroscopy (NOESY)

Nuclear Overhauser effect spectroscopy can help with the identification of the ligands and it can visualize interactions as well. The nOe effect implies a polarization transfer from one spin to another via cross-relaxation. The nOe effect differs from spin-spin coupling because the nOe effect occurs through space rather than trough chemical bonds. This implies interactions between nearby nuclei can be measured. The magnitude and sign of the nOe depend on the molecules’ rotational dynamics.[31] Small molecules in low viscosity solvents will tumble rapidly and typically show a positive nOe (figure 2.2). This corresponds to a cross peak with a different phase than the signals on the diagonal in a NOESY spectrum. In contrast to small molecules, bigger species such as Qdots tumble slowly. Hence, the nOe of the ligands bound to the Qdot surface becomes negative and the cross peaks have the same phase as the signals on the diagonal. In between these two extremes there is a region where the molecules have an intermediate size and the nOe is very weak.

11

Chapter 2. Experimental techniques

Figure 2.2: This graph depicts the variation of maximum theoretic homonuclear nOe as function of the rotational correlation time (τ c ). τ c is defined as the average time required for a molecule to rotate through an angle of one radian about any axis. A small τ c corresponds to a rapidly tumbling molecule and a large τ c corresponds to a slowly tumbling molecule.

2.1.4

Diffusion Ordered Spectroscopy (DOSY)

Diffusion Ordered Spectroscopy is a technique in which the 1D proton spectrum is plotted on the x-axis and the diffusion coefficient of the molecules that generate different signals on the y-axis (figure 2.3(a)). Diffusion coefficients are indicated on the spectra in µm2 /s. Small molecules, such as the solvent, will diffuse more rapidly, generating a large diffusion coefficient, while larger molecules or bound ligands, will diffuse more slowly, generating a small diffusion coefficient. By means of DOSY NMR the assignment of the resonances in the 1D proton NMR spectrum can be facilitated. With the diffusion coefficient, the hydrodynamic radius of the Qdots can be calculated (equation 2.5). kB embodies the constant of boltzmann, T is the temperature, η is the viscosity of the solvent and D is the diffusion coefficient. The hydrodynamic radius of a Qdot comprises the radius of the Qdot itself together with two times the length of the ligand (figure 2.3(b)). If the size of the Qdot itself is known by other techniques like UVVIS absorption spectrometry or TEM, the thickness of the ligand shell can be calculated from the difference in these radii. rh =

kb T 6πηD

12

(2.5)

Chapter 2. Experimental techniques

(a)

(b)

Figure 2.3: (a) Example of Diffusion Ordered Spectroscopy: DOSY spectrum of OA bound to CdSe Qdots. (b) The hydrodynamic radius of a Qdot comprises the radius of the Qdot together with two times the length of the ligand.

2.1.5

Saturation Transfer Difference NMR

In a Saturation Transfer Difference (STD) NMR experiment, two 1D proton NMR spectra are recorded. In the first one, a resonance will be selectively irradiated which will induce saturation in that molecule. If the irradiated species comes into contact with other species, the saturation can be transferred to that species. The intensity of the resonances experiencing the saturation will be attenuated. A second 1D proton spectrum is recorded where a frequency off-resonance is irradiated. The intensity of the resonances will thus be normal in this spectrum. If the difference is taken between the two spectra, only the resonances experiencing saturation will be visible. In this way, an interaction between two species can be confirmed.

2.2

UV-Vis absorption spectrometry

UV-Vis absorption spectrometry is used to determine the size of the particles by looking at the characteristic first excitation peak in the spectra. This peak represents the bandgap energy of the Qdot. Equation 2.6 shows the relation between the bandgap energy and the diameter d of zincblende CdSe Qdot. [32] Eg = 1.74 +

1 0.89 − 0.36d + 0.22d2

(2.6)

The concentration of nanoparticles in solution can be determined from the absorption spectra as well. For QDots with a given diameter d, the molar extinction coefficient of

13

Chapter 2. Experimental techniques zincblende CdSe QDs is given by equation 2.7.[32] ε = ad3

(2.7)

The constant a is given in table 2.1 for three different wavelengths where the absorption is not influenced by the crystal size. The molar extinction coefficient is expressed in L/cm mol if the diameter is expressed in nm. Using Lambert-Beer’s law, the concentration of the Qdot solution is easily determined (equation 2.8). A = εCL

(2.8)

In this equation, A is the absorbance, C the concentration and L the path length of light through the sample (typically 1cm). Thus the absorbance needs to be determined at a wavelength of 300, 320 and 340 nm and each molar extinction coefficient should be determined by using the correct diameter. All measurement were performed on a Perkin Elmer spectrometer lambda 2.

2.3

Luminescence spectroscopy

The photoluminescence quantum yield (PL QY) is a very important parameter for the application of Qdots in bio-imaging. It is used to express the efficiency of the photoluminescence process. In this work the PL QY is always studied relative to coumarin 2, which has a PLQY of 93%.[33] It is ideal to have similar or identical absorbances at the excitation wavelength for both the standard and the solution one wants to measure. The right absorbance is obtained by adjusting the concentration of the solutions. The PL QY is calculated as shown in equation 2.9, where Φ embodies the PL QY, I the integrated emission intensity, A the absorbance and n the refractive index of the solvent.[34] The measurements were performed on a time resolved and steady state photoluminescence setup of Edinburg instruments, where excitation is provided through a standard 450W Xe lamp. Φ = Φr

I Ar n2 Ir A n2r

(2.9)

Table 2.1: Constant a for three different wavelengths where the absorption is not influenced by the crystal size.

wavelength (nm) 340 320 300

14

a 19300 23800 31700

Chapter 2. Experimental techniques

2.4

Other techniques

Dynamic Light scattering (DLS) is used to analyze size distributions of colloidal particles in solution. Hence, particle aggregation can be detected and the hydrodynamic diameter can be approximated mathematically. With a laser as a light source, a timedependent fluctuation in the scattering intensity due to the Brownian motion is induced. The rate at which the intensity fluctuates, depends on the size of the dispersed particles. Images of the Qdots are generated by the transmission electron microscope (TEM). The TEM uses an electron gun as a probe source. The electron beam is scattered or transmitted trough the sample, creating an image with suitable resolution (up to 0.05nm) by a fluorescent screen or a Charged Coupled Device camera. All TEM images were recorded with the JEM-2200FS Transmission Electron Microscope.

15

Chapter 2. Experimental techniques

16

Chapter 3 Synthesis and characterization of Qdots Prior to testing the solubility of Qdots in water, the Qdots should be synthesized and characterized. Both CdSe and core/shell CdSe/CdS Qdots are synthesized and studied with various techniques, including NMR. In this chapter, the synthesis is covered and the Qdots undergo a first quality control using UV-VIS absorption spectroscopy and TEM before they are further analyzed using NMR spectroscopy.

3.1 3.1.1

CdSe Qdots Synthesis I of zincblende CdSe Qdots

The first step towards making Qdots water soluble is to synthesize the Qdots. Richard Capek, Iwan Moreels and coworkers described a hot-injection synthesis for zincblende CdSe Qdots in literature.[32] 0.72 mmole of a cadmium oleate precursor and 18.6 mL ODE were flushed with nitrogen at room temperature. After 10 minutes, the solution was heated and stirred for 30 minutes at 100°C. Then, the solution was heated to 260°C and 0.72 mmole of a Se precursor was injected. The nanocrystals were grown at 235°C and the reaction was quenched with a cold water bath when the particles had the desired size. A schematic representation of the experimental setup is shown in figure 3.1. After reaction, the particles were purified by precipitation and suspension. Equivalent amounts of toluene, isopropanol and methanol were added to the solution in a 1:1 ratio with respect to the amount of ODE. The mixture was centrifuged for about 20 minutes at 3500 rpm. The precipitate was dispersed in 1 or 2 mL toluene. In the second purification step the particles were precipitated with methanol (2:1 or 4:1 or 6:1 with respect to toluene). This step was repeated three times. Finally the particles were dissolved in 4mL toluene.

17

Chapter 3. Synthesis and characterization of Qdots

Figure 3.1: Schematic representation of the experimental setup.

Se-ODE precursor The Se precursor was prepared by dissolving 0.276 g Se in 35 ml ODE to obtain a 0.1 M solution. The solution was flushed under nitrogen atmosphere for 20 minutes at room temperature and for 40 minutes at 100°C. After flushing, the solution was stirred at 195°C under nitrogen atmosphere for 2h30. Cd(OA) precursor The Cd(OA) precursor was prepared by dissolving 0.449 g CdO and 7.909 g OA in 20.6498 g ODE (0.1M Cd in ODE with a OA:Cd ratio of 8:1). The solution was stirred under nitrogen atmosphere at 160°C until all CdO was dissolved. Sizing curve In this synthesis, the reaction time determines the size of the QDots. For every Se-ODE precursor, a first synthesis is performed to establish the sizing curve. Aliquots are taken at different reaction times, according to table 3.1. By means of the measured absorption spectra the size of the Qdots at different reaction times can be calculated. Figure 3.2(a) displays the absorption spectra of the different aliquots taken during the synthesis. The absorption wavelength or the size of the Qdot is then plotted in function of the reaction time (figure 3.2(b)), creating the sizing curve.

18

Chapter 3. Synthesis and characterization of Qdots

Table 3.1: A first synthesis is performed to compose the sizing curve. Aliquots are taken at different reaction times.

Aliquot 1 2 3 4 5 6 7

Time 20s 30s 50s 1min 20s 2min 10s 3min 30s 5min 30

Aliquot 8 9 10 11 12 13 14

Time 9min 14min 30s 23min 36min 1h 1h 35min 2h 30min

(a)

(b) Figure 3.2: (a) Representation of the absorption spectra of the different aliquots taken during the synthesis. The time in the legend indicates the time when the aliquot was taken after the start of the reaction. (b) Sizing curve. The wavelength of the first absorption peak is plotted as a function of the time when the aliquot is taken.

19

Chapter 3. Synthesis and characterization of Qdots

3.1.2

Qualitative analysis of CdSe Qdots (synthesis I)

Figure 3.3 shows the UV-VIS absorption spectrum of the CdSe particles synthesized with synthesis I. The first excitation peak is found at 521.5 nm, which corresponds to a diameter of 2.84 nm. The peak has a half width at half maximum of 16 nm, corresponding to a relative size dispersion of 6-7% which is appropriate for characterizing material properties at a later stage. On the other hand, the PLQY of 1-2% is rather low but this was expected since core Qdots are known to have a low PLQY after purification. Some TEM images of the synthesized zincblende Qdots are shown in figure 3.4. The Qdots are not clustered and the diameter measured in the TEM images is 3.12 nm. We can therefore conclude that these particles are a good starting point for the initial experiments in this study.

3.1.3

NMR analysis of CdSe Qdots (synthesis I)

In literature, the organic/inorganic interface of the zincblende CdSe Qdots has been studied thoroughly using solution state NMR.[29] The ligands have already been identified and quantified and the nature of the ligand-Qdot bond was studied. CdSe Qdots are stabilized by oleic acid ligands, bound to the Qdot surface as oleates. The overall stoichiometry of a colloidal Qdot and ligand self-exchange was studied as well. A 1D proton NMR spectrum of oleic acid bound to zb.CdSe Qdots is displayed in figure 3.5(b). The resonance at 5.7 ppm originates from the protons at the double bond of the oleic acid ligands. The other resonances are located in the region from 0 to 3 ppm which is the same region as where the polymer resonances will be located (see chapter 4). This can cause issues during the interpretation of the NMR spectra of the water soluble system. The ligand density can be calculated using the concentration of OA that was obtained with quantitative NMR measurements. An average of 3.78 ligands per nm2 are present at the CdSe Qdot surface. Below, the NOESY and DOSY spectrum of the zincblende Qdots are shown (figures 3.6(a) and 3.6(b)). These spectra are important references for the analysis of the Qdots in water. Note that there are clear and negative nOe cross peaks between the protons of the OA double bond and the CH2 ’s and CH3 , indicating that the OA ligand is part of a larger

Figure 3.3: UV-VIS absorption spectrum of the CdSe Qdots in toluene (synthesis I).

20

Chapter 3. Synthesis and characterization of Qdots

(a)

(b)

(c)

Figure 3.4: TEM images of zincblende CdSe Qdots (synthesis I).

(a)

(b) Figure 3.5: (a) 1D proton spectrum of oleic acid in toluene-d8. (b) 1D proton spectrum of CdSe nanoparticles capped with oleic acid in toluene (synthesis I).

21

Chapter 3. Synthesis and characterization of Qdots system. In addition, the diffusion coefficient of the OA resonances is 99.14 10−12 m2 /s, which corresponds to a hydrodynamic diameter of 7.729 nm. Also take note of the diffusion coefficient of the solvent resonances in DOSY (2088 10−12 m2 /s).

(a)

(b)

Figure 3.6: (a) NOESY and (b) DOSY spectrum of the ligands bound to CdSe Qdots in toluene-d8 (synthesis I).

3.1.4

Synthesis II of zincblende CdSe Qdots

Recently, a new hot-injection synthesis was developed in the PCN group at Ghent University. This synthesis can be performed under atmospheric conditions (not under nitrogen atmosphere) and completes within 5 minutes time. The size of the particle can be tuned by the chain length of the acid used as a ligand. The yield depends on the molar ratio Cd:acid:Se. For a yield of 100% the ratio should be 2:6:1. In a flask 0.025g CdO, 0.120g dodecanoic acid and 10mL of ODE were mixed. In another recipient 0.078g Se and 10 mL ODE were stirred. It is important not to dissolve the Se powder. Instead, the Se powder should be distributed homogeneously in the ODE. The CdO mixture was heated to 268°C. When that temperature was reached, 1mL of the Se dispersion was injected. The synthesis was performed at a temperature of 258°C. After the synthesis the dodecanoic acid ligands were exchanged with OA.

3.1.5

Qualitative analysis of CdSe Qdots (synthesis II)

The UV-VIS absorption spectrum of the CdSe particles synthesized with synthesis II is depicted in figure 3.7. The first excitation peak is located at 529.5 nm and the diameter 22

Chapter 3. Synthesis and characterization of Qdots of the particles is 2.84 nm. With a half width at half maximum of 13 nm and a relative size dispersion of 5-6% one can conclude that these particles are suitable for characterizing their material properties. However, the PLQY has a rather low value of 1-2% thus it can be said that these particles are suitable for the initial stage of this study.

3.1.6

NMR analysis of CdSe Qdots (synthesis II)

Figure 3.8 shows the 1D proton NMR spectrum of oleic acid bound to the CdSe Qdots of synthesis II. The resonances are assigned as indicated in the inset of the OA structure. Note that there appears to be a slight splitting of some the resonances in a sharper and a broader part, indicating that there could be a mixture of free and bound ligand present in the sample. However, only a single diffusion coefficient associated with these resonances is present in the DOSY spectrum. A diffusion coefficient of 129.88 10−12 m2 /s is equivalent to a hydrodynamic diameter of 5.90 nm. The slight splitting of the resonances could indicate that part of the ligands are in close contact with solvent molecules, giving rise to a sharper resonance. The ligand density is calculated as well. On average, 3.37 ligands per nm2 are present at the CdSe Qdot surface. In NOESY (figure 3.9(a)), there are clear nOe cross peaks between the protons of the OA double bond and the CH2 ’s and CH3 . It can be said that OA is part of a larger system since the nOe cross peaks of the OA resonances are negative.

Figure 3.7: UV-VIS absorption spectrum of the CdSe Qdots in toluene (synthesis II).

23

Chapter 3. Synthesis and characterization of Qdots

Figure 3.8: 1D proton NMR spectrum of oleic acid bound to the CdSe Qdots of synthesis II in toluene-d8. The resonances are assigned as indicated in the figure. * indicates solvent resonances, impurities and water.

(a)

(b)

Figure 3.9: (a) NOESY and (b) DOSY spectrum of the ligands bound to CdSe Qdots in toluene-d8 (synthesis II).

24

Chapter 3. Synthesis and characterization of Qdots

3.2 3.2.1

Core/shell CdSe/CdS Qdots Synthesis of core/shell CdSe/CdS Qdots

As mentioned in the introduction (chapter 1), a quantum dot can be passivated by embedding the semiconductor material of the core in a shell of a second semiconductor material. This will also improve the PLQY significantly. Hence, the nanocrystals have a higher potential for the applications in bio-imaging. To obtain core/shell nanocrystals with a high crystallinity, similar lattice parameters for the core and shell is a prerequisite. Successive ion layer adhesion and reaction (SILAR) is one of the techniques that can be used to make such a core/shell structure.[35] Here, 0.1018 µmoles zincblende CdSe Qdots dissolved in hexane were heated to 100°C in a mixture of 1.5g of ODA and 5g of ODE under nitrogen atmosphere. When the hexane had evaporated, the mixture could be heated to 225°C. Once that temperature was reached, monolayer by monolayer was formed by adding the precursors stepwise. The amount of precursor added in each step was calculated carefully. The S precursor was added first. After a period of 10 minutes the Cd precursor was added. Now one CdS layer was formed. These additions should be repeated a number of times, depending on the amount of layers one wants to synthesize. Figure 3.10 shows the absorption spectra of aliquots taken after formation of each CdS layer. The precursors For the SILAR technique, the same Cd precursor was used as described in section 3.1.1. A 0.1M S precursor was made by dissolving sulfur powder in ODE and heating the mixture to 160°C. It is crucial to add the precursors dropwise to prevent self nucleation. In this regard it is also important that the precursors are highly reactive.

3.2.2

Qualitative analysis of CdSe/CdS core/shell Qdots

The core/shell Qdots synthesized for this study consist of a CdSe core of 3 nm and 4 CdS layers. This results in a particle with total diameter of 5.7 nm. TEM images of the core/shell Qdots are shown in figure 3.11. The particles do not cluster. UV-VIS spectra of these particles have already been shown in section 3.2.1 (figure 3.10). PLQY has improved a lot in comparison with the core CdSe Qdots of section 3.1. The cores had a PLQY of 1-2% and core/shell Qdots now have a PLQY of 43%. Therefore, these core/shell Qdots are very suitable for this study.

25

Chapter 3. Synthesis and characterization of Qdots

Figure 3.10: Absorption spectra of aliquots taken during the SILAR synthesis at the start of the synthesis, after heating and after addition of each layer.

(a)

(b) Figure 3.11: TEM images of CdSe/CdS Qdots.

26

(c)

Chapter 3. Synthesis and characterization of Qdots

3.2.3

NMR analysis of the CdSe/CdS core/shell Qdots

A 1D proton spectrum of core/shell CdSe/CdS Qdots is depicted in figure 3.12. At first sight this spectrum looks quite similar to the spectrum of CdSe Qdots in toluene except for the two extra resonances that are present at approximately 1 and 1.5 ppm (indicated with ° in the figure). To know the origin of these resonances, the NOESY and DOSY spectra are studied (figure 3.13). NOESY (a) shows a very weak positive nOe between these two resonances but there is no interaction to the adjacent broader resonances. This suggest that the extra resonances originate from a small, non-interacting molecule. DOSY (b) shows that these resonances also have a different diffusion coefficient than the adjacent resonances of 318 10−12 m2 /s. A measurement of OA in toluene without nanoparticles yields a diffusion coefficient of 603 10−12 m2 /s. It can thus be concluded that the extra resonances are not bound to the surface but may be associated with the ligand shell and the adjacent broader ones are bound to the Qdot surface.

Figure 3.12: 1D proton NMR spectrum ligands bound to CdSe/CdS Qdots in toluene-d8. the resonances are indicated as depicted in the figure. * indicates solvent and impurities and ° indicates extra resonances of free ligands.

Figure 3.14 shows the 1D proton NMR spectrum of the ligands bound to CdSe/CdS Qdots without free ligand present. Considering the products being used during the synthesis, oleic acid is definitely bound to the Qdot surface since there is a broad double bond peak at 5.7 ppm in the spectrum. However, if the integration of the resonance at 5.7 ppm is set to 2, the integration of the CH3 resonance at 1.2 ppm exceeds 3. Probably another molecule is present in this sample.

27

Chapter 3. Synthesis and characterization of Qdots

(a)

(b)

Figure 3.13: (a) NOESY and (b) DOSY spectrum of the ligands bound to CdSe Qdots in toluene-d8.

Figure 3.14: 1D proton NMR spectrum of the ligands bound to CdSe/CdS Qdots in toluene-d8. * indicate solvent resonances and the OA resonances are assigned as indicated in the figure.

28

Chapter 4 Synthesis and characterization of the polymers In this thesis, Qdots are dissolved in water by means of a polymer coating (Section 1.3). Therefore, the synthesis of suitable polymers is necessary. In this chapter the synthesis of these polymers is covered and they are analyzed with NMR. One of the polymer building blocks, used in both syntheses, is also analyzed in order to simplify the analysis of the polymers.

4.1

Characterization of poly(isobutylene-alt-maleic anhydride) by means of NMR

Poly(isobutylene-alt-maleic anhydride) is used in the synthesis of both polymers. It is analyzed in order to simplify the analysis of the synthesized polymers. This polymer is an altering copolymer with varying stereochemistry of the two sterocenters in the maleic anhydride building block (stereocenters are indicated with a and b in the figure). As a result, multiple resonances may occur for each 1 H in the molecular structure (figure 4.1). The assignment of the resonances is treated in appendix A. The diffusion coefficient of this polymer can be extracted from the DOSY spectrum (appendix A). A diffusion coefficient of 247 10−12 m2 /s corresponds to a hydrodynamic diameter of 3.58 nm. Note that around 1.5 and 2.5 and between 7 and 7.5 ppm there are some impurities. The impurities have a different diffusion constant than the polymer itself.

29

Chapter 4. Synthesis and characterization of the polymers

Figure 4.1: 1D proton NMR spectrum of poly(isobutylene-alt-maleic anhydride). a and b indicate the stereocenters.

4.2

Synthesis and characterization of P(IB-alt-LMA)

4.2.1

Synthesis of poly(isobutylene-alt-n-laurylmaleamic acid)

Figure 4.2 shows the polymer that was synthesized during this research. Poly(isobutylenealt-n-laurylmaleamic acid) can be synthesized by modifying poly(isobutylene-alt-maleic anhydride) (figure 4.3(a)) in a reaction with dodecylamine (figure 4.3(b)). From now on, poly(isobutylene-alt-n-laurylmaleamic acid) will be referred to as P(IB-alt-LMA). 2,7g (15 mmol) dodecylamine powder was dissolved in 100mL anhydrous THF. 3.08g (20 mmol repeating unit) poly(isobutylene-alt-maleic anhydride) powder was dissolved in a second flask. Both solutions were mixed until the powders dissolved completely. The mixture was heated to 55-60°C under stirring. After one hour, the solution was concentrated to 30-40mL by evaporation of THF. The solution was left stirring overnight. The mixture was dried completely by evaporation on the next day and it was then redissolved in 40mL anhydrous dichloromethane, giving a 0.5M polymer solution. In practice, some fraction of the anhydride rings will still be intact after the reaction as shown in figure 4.2. This means the polymer is actually poly[(isobutylene-alt-n-Laurylmaleamic acid)-co-(isobutylene-alt-maleic anhydride)].

30

Chapter 4. Synthesis and characterization of the polymers

Figure 4.2: This polymer, with suitable properties for dissolving Qdots in water, is synthesized. P(IB-alt-LMA) has hydrophobic side chains to interact with the organic ligands of the Qdot and a hydrophilic backbone to make the system soluble in water.

(a)

(b)

Figure 4.3: By modifying poly(isobutylene-alt-maleic anhydride) (a) in a reaction with dodecylamine (b) a polymer with suitable properties for water solubilization is synthesized.

31

Chapter 4. Synthesis and characterization of the polymers

4.2.2

Characterization of P(IB-alt-LMA) by means of NMR

P(IB-alt-LMA) has hydrophobic side chains and a hydrophilic backbone. Therefore, it is soluble in different kinds of solvents. In a solvent like H2 O, the polymer will adopt a conformation in which the side chains will be on the inside and the backbone will be in contact with the aqueous environment (figure 4.4). In solvents like chloroform it will adopt a conformation where the side chains orient towards the outside. Hence, the solvent will have an impact on the solution conformation of the polymer, which is also apparent in the NMR spectra. A 1D proton NMR spectrum of P(IB-alt-LMA) featuring assignment in chloroform-d1 is shown in figure 4.5(a). A series of 2D experiments, used for the assignment of the resonances can be found in appendix A. The location of protons 2, 3 and 4 cannot be determined exactly. This is due to the conformation of the polymer. The protons on the inside will exhibit a lower degree of mobility. Therefore, these protons will be subject to faster relaxation than the other protons, resulting in broader peaks. The impurities found in poly(isobutylene-alt-maleic anhydride) are also present in this polymer. However these resonances are now broad, suggesting that they are enclosed by the polymer or are part of the molecular structure. In water, more resonances can be distinguished but not all of them can be assigned (see appendix A). (figure 4.5(b)). The protons on the inside of the structure (protons 5 to 8) will be subject to faster relaxation than the other protons, resulting in broader peaks. However, they can still be distinguished because these resonances are intense. The impurities are also present in this polymer with the same diffusion coefficient as the polymer.

Figure 4.4: In a solvent like H2 O, the polymer will a conformation like a micelle in which the side chains will be on the inside and the backbone will be in contact with water. In solvents like chloroform it will adopt a conformation with the side chains oriented towards the outside.

32

Chapter 4. Synthesis and characterization of the polymers

(a)

(b) Figure 4.5: 1D proton NMR spectrum of P(IB-alt-LMA) in chloroform-d1 (a) and in D2 O (b). * indicates an impurity or a solvent resonance

33

Chapter 4. Synthesis and characterization of the polymers Figure 4.6(a) shows the DOSY spectrum of P(IB-alt-LMA) in chloroform-d1. The diffusion coefficient of the polymer is 121 10−12 m2 /s, corresponding to a hydrodynamic diameter of 6.213 nm. This diameter is significantly bigger than the one of poly(isobutylene-alt-maleic anhydride) because the polymer now has hydrophobic side chains. The DOSY spectrum of P(IB-alt-LMA) in D2 O is shown in Figure 4.6(b). The diffusion coefficient of the polymer is 49 10−12 m2 /s, equivalent to a hydrodynamic diameter of 8.89 nm. This is about30% bigger than the hydrodynamic radius of the polymer in chloroform. As a result, the average polymer in water is moving slower than in chloroform. The second diffusion coefficient and associated resonances in the spectrum originates from unreacted amine (220 10−12 m2 /s, 1.89 nm).

(a)

(b)

Figure 4.6: DOSY spectrum of P(IB-alt-LMA) in chloroform-d1 (a) and in D2 O (b). Diffusion coefficients are indicated in µm2 /s.

4.3 4.3.1

Synthesis and characterization of P(IB-alt-UPA) Synthesis of poly(isobutylene-alt-3-(undecyloxycarbonyl) prop2-enoic acid)

Poly(isobutylene-alt-3-(undecyloxycarbonyl)prop-2-enoic acid) (Figure 4.7) is the second polymer that was used in this study and it can be synthesized by modifying poly(isobutylenealt-maleic anhydride) (figure 4.8(a)) with 10-undecen-1-ol (figure 4.8(b)). From now on, 34

Chapter 4. Synthesis and characterization of the polymers poly(isobutylene-alt-3-(undecyloxycarbonyl)prop-2-enoic acid) will be referred to as P(IBalt-UPA). 3.08g (20 mmol repeating unit) poly(isobutylene-alt-maleic anhydride) powder was dissolved in 100mL anhydrous THF. The solutions was mixed and heated to 50°C until the polymer dissolved completely. 15 mmol 10-undecen-1-ol, 60mmol pyridine and a spatula tip of 4-dimethylaminopyridine (DMAP) was added to the dissolved polymer. Pyridine and DMAP were added because the alcohol is a worse nucleophile than the amine used to synthesize the P(IB-alt-LMA). The solution was left stirring overnight and afterwards the mixture was dried completely by evaporation on the next day and was subsequently redissolved in 40mL anhydrous THF, rendering a polymer solution with a concentration of 0.5M. In practice, some of the anhydride rings can still be intact after the reaction. This means the polymer is actually poly[(isobutylene-alt-3-(undecyloxycarbonyl) prop-2-enoic acid)-co-(isobutylene-alt-maleic anhydride)].

4.3.2

Characterization of P(IB-alt-UPA) by means of NMR

P(IB-alt-UPA) has hydrophobic side chains and a hydrophilic backbone. Therefore, it is soluble in different kinds of solvents. However, it is not that versatile as P(IB-alt-LMA). It does not dissolve well in chloroform and it is not compatible with toluene. In a solvent like H2 O, the polymer will adopt a conformation in which the hydrophobic side chains will be on the inside and the hydrophilic backbone will be in contact with the aqueous environment (figure 4.4). In solvents like THF it adopts a conformation in which the side chains are oriented towards the outside. The solvent will thus have an impact on the NMR measurements of the polymer. Figure 4.9 shows the 1D proton spectrum of P(IB-alt-UPA) in THF-d8 (a) and in D2 O (b). The influence of the solvent is less pronounced than with P(IB-alt-LMA). The resonances of the hydrophilic backbone can be assigned in THF although they are small and broad. The resonances are assigned as shown in the figure. In water, the side chain resonances are somewhat broader, but again all resonances can be assigned to certain protons (figure 4.9(b)). Note that there are some pyridine resonances in the aromatic region between 7 and 9 ppm. In Figure 4.10, the DOSY spectra of P(IB-alt-UPA) in the two solvents are shown. In THF-d8 (a), the polymer resonances seem to have two diffusion coefficients. The first, 212 10−12 m2 /s matches with one polymer molecule with a hydrodynamic diameter of 4.17 nm. The second one, 1313 10−12 m2 /s, corresponds to 10-undecen-1-ol that has a hydrodynamic diameter of 0.67 nm. This is confirmed by the absence of the CH3 resonance at 1313 10−12 m2 /s since this resonance belongs to the polymer backbone and not to 10-undecen-1-ol.

35

Chapter 4. Synthesis and characterization of the polymers

Figure 4.7: This polymer, with suitable properties for dissolving Qdots in water, is synthesized. P(IB-alt-UPA) has hydrophobic side chains to interact with the organic ligands of the Qdot and a hydrophilic backbone to make the system soluble in water.

(a)

(b)

Figure 4.8: By modifying poly(isobutylene-alt-maleic anhydride) (a) with 10-undecen-1-ol (b) a polymer with suitable properties for water solubilization is synthesized.

36

Chapter 4. Synthesis and characterization of the polymers

(a)

(b) Figure 4.9: 1D proton NMR spectrum of P(IB-alt-UPA) in THF-d8 (a) and in D2 O (b).

37

Chapter 4. Synthesis and characterization of the polymers In D2 O (b) the solvent has a diffusion coefficient of 849 10−12 m2 /s. The resonances of the polymer seem to have two diffusion coefficients: 105 and 10 10−12 m2 /s. These diffusion coefficients respectively correspond to one polymer molecule with a hydrodynamic diameter of 4.19 nm and possibly an aggregate of 10-undecen-1-ol with a hydrodynamic diameter of 42.61 nm. This is confirmed again by the absence of the CH3 resonance at 10 10−12 m2 /s since this resonance belongs to the polymer backbone and not to 10-undecen-1-ol. The resonance at 3.4 ppm, originating from 10-undecen-1-ol is also only present at the diffusion coefficient of 10.23 10−12 m2 /s.

(a)

(b)

Figure 4.10: DOSY spectrum of P(IB-alt-UPA) in THF-d8 (a) and in D2 O (b).

38

Chapter 5 Water solubilization of Qdots by means of a polymer shell 5.1

Water solubilization of Qdots by means of P(IBalt-LMA)

5.1.1

Protocol for dissolving nanoparticles in water

To dissolve the Qdots in water, P(IB-alt-LMA) in dichloromethane was mixed with a solution of quantum dots in toluene at a ratio of 100 repeating units of polymer per nm2 of effective Qdot surface. Detailed calculations are shown in section 5.1.2. The mixture was completely dried by slowly evaporating the solvent at 60°C and 200 mbar. The solid film was then redissolved in an aqueous solution. Stirring and heating to 60°C facilitated the process. To optimize the protocol, different aqueous solutions were tested. First millipore water was used without any result. The Qdots did not dissolve, not even at elevated temperature. Then a series of basic aqueous KOH solutions were made with pH ranging from 9 to 14. An intermediate pH of 12 led to the best result. However, none of these dispersions were luminescent. Finally a basic solution of millipore water, KOH and sodium borate (Na2 B4 O7 .10H2 O) was used, as suggested in literature.[25] With this solution, the Qdots were luminescent in water and they dissolved well. After dissolution, the pH of the solutions were ± 9.24, which corresponds to the pKa of boric acid.

39

Chapter 5. Water solubilization of Qdots by means of a polymer shell

5.1.2

Calculating the volume of polymer needed for solubilization of Qdots

To know how much of the polymer solution should be added to the Qdots, the total diameter of the QDots and the ligands should be known: dcore,ef f = dcore+ligandshell

(5.1)

The surface area can then be calculated as shown in equation 5.2. dcore,ef f 2 ) 2 = Cparticle Vparticle NA

Aef f = 4π( Ntotal,particle

(5.2) (5.3)

Equation 5.3 demonstrates how the total number of particles can be calculated. Here, C expresses the concentration and V the volume of nanoparticle solution used. If the number of particles is known (equation 5.3), the effective surface of all Qdots can be calculated: Aef f,total = Aef f Ntotal,particle

(5.4)

At a ratio of 100 repeating units per nm2 of effective surface, the number of repeating units needed is: Ntotal,polymer = 200 nm2 Aef f,total

(5.5)

Finally, the volume of polymer solution needed can be calculated as follows: Vpolymer =

5.1.3

Ntotal,polymer (NA Cpolymer )

(5.6)

CdSe Qdots with P(IB-alt-LMA) in chloroform

Prior to analyzing the CdSe QDots with P(IB-alt-LMA) in water, the system is first analyzed with NMR in a solvent where both the Qdots and the polymer are soluble. The ideal solvent for this experiment is chloroform. Figures 5.1 represents a 1D proton spectrum of this system in chloroform. Unfortunately, the resonances of both the Qdot and the polymer are present in the region between 0 and 2 ppm the resonances of both species overlap. The DOSY spectrum (figure 5.2) indicates that 3 species are present. The diffusion coefficients and the corresponding hydrodynamic diameters of these species are listed in table 5.1. One of the species corresponds to the polymer resonances, and two to the OA resonances. The hydrodynamic diameter of the polymer corresponds to the one of the polymer measured in chloroform in section 4.2.2. The hydrodynamic diameter of the two OA species 40

Chapter 5. Water solubilization of Qdots by means of a polymer shell confirms that both free and bound OA are present in this system. P(IB-alt-LMA) and the Qdots have a different diffusion coefficient thus these species do not migrate together. This is expected since they are both soluble in chloroform and it would not be energetically favorable to migrate together.

Figure 5.1: 1D proton NMR spectrum of the ligands bound to zincblende CdSe Qdots and P(IB-alt-LMA) in chloroform-d1.

Figure 5.2: DOSY spectrum of the ligands bound to zincblende CdSe Qdots and P(IB-alt-LMA) in chloroform-d1.

Table 5.1: This table shows the diffusion coefficients and hydrodynamic diameters of zincblende Qdots, free OA and P(IBalt-LMA) in chloroform.

Species CdSe Qdots Free OA P(IB-alt-LMA)

Diffusion coefficient (m2 /s) 89.63 10−12 415.30 10−12 133.31 10−12

41

hydrodynamic diameter (nm) 8.54 1.85 6.07

Chapter 5. Water solubilization of Qdots by means of a polymer shell

5.1.4

CdSe Qdots with P(IB-alt-LMA) in water

Qualitative analysis of CdSe Qdots in water The CdSe Qdots in toluene are transferred to water according to the protocol described in section 5.1.1. The Qdots do not change in size or shape since the UV-VIS absorption spectrum did not change after the transfer (figure 5.3). The particles are still luminescent, although the PLQY is significantly lower. Table 5.2 displays the PLQY before and after the transfer to water. A DLS measurement shows that the Qdots in water have excellent stability since the size distribution stayed the same over time. The hydrodynamic diameter can also be approximated by DLS (figure 5.4). It shows a hydrodynamic diameter of 10.06 nm. Table 5.2: This table shows The PLQY of CdSe Qdots, before and after transfer to water.

Qdots CdSe Qdots (synthesis I) CdSe Qdots (synthesis II)

PLQY in toluene (before transfer) 1.61 1.66

PLQY in water 0.14 0.14

Figure 5.3: UV-VIS absorption spectrum before and after transfer to water.

Figure 5.4: DLS measurement of zincblende CdSe Qdots embedded in a P(IB-alt-LMA) shell in water.

42

Chapter 5. Water solubilization of Qdots by means of a polymer shell NMR analysis of CdSe Qdots in water Figure 5.5 depicts the 1D proton NMR spectrum of CdSe Qdots (synthesis II) in water. The resonances of both the Qdot and the polymer are present but the region between 0 and 2 ppm contains overlapping resonances of both species, similar to section 5.1.3. In DOSY, the double bond resonance of OA is not visible (figure 5.6(a)) and the other OA resonances overlap with the polymer resonances. This means that the diffusion coefficient cannot be extracted directly. At first sight we cannot tell if the Qdots are embedded in a polymer shell. However, the hydrodynamic diameter extracted from DOSY is much larger than the diameter of the polymer alone in water (table 5.3). This indicates that the CdSe Qdots are embedded in the polymer shell. The hydrodynamic diameter is significantly larger than the one of DLS measurements. Note that the diffusion coefficient of the residual toluene resonances is much lower in a water environment than in toluene. This indicates that the toluene co-migrates with the Qdot system. This situation is favorable since toluene will like the hydrophobic environment of the OA and the side chains much more than the water environment. The fact that toluene does not have the same diffusion coefficient as the polymer-Qdot system confirms that the toluene molecules are exchanging between the water and the polymer-Qdot environment. The NOESY spectrum (figure 5.6(b)) is studied to confirm that the Qdots are embedded in the polymer shell. If the polymer is in proximity to the OA, we would expect a nOe cross peak to be visible between the polymer and OA resonances. There are some cross peaks visible between the OA double bond and the CH2 /CH3 region. However, it is not clear if these cross peaks of the OA resonance connect to the polymer or the OA resonances in the CH2 /CH3 region.

Figure 5.5: 1D proton NMR spectrum of P(IB-alt-LMA) and ligands bound to wurtzite CdSe Qdots in D2 O.

43

Chapter 5. Water solubilization of Qdots by means of a polymer shell Table 5.3: This table shows The diffusion coefficient and hydrodynamic diameter of P(IB-alt-LMA) in water, with and without CdSe Qdots.

Polymer Polymer alone in water Polymer with CdSe Qdots in water

Diffusion coefficient (10−12 m2 /s) 49 26.04

(a)

Hydrodynamic diameter (nm) 8.89 16.7

(b)

Figure 5.6: (a) DOSY spectrum of P(IB-alt-LMA) and ligands bound to zincblende CdSe Qdots in D2 O. (b) NOESY spectrum of P(IB-alt-LMA) and ligands bound to zincblende CdSe Qdots in D2 O.

5.1.5

CdSe/CdS core/shell Qdots with P(IB-alt-LMA) in water

Qualitative analysis of CdSe/CdS Qdots in water The CdSe/CdS Qdots are transferred to water according to the protocol described in section 5.1.1. TEM images (figure 5.7) and UV-VIS absorption spectra (figure 5.8) confirm that the shape and size of the Qdots stayed intact during the transfer. The background of the TEM images does not always have the same darkness. The lighter sections of the background could indicate the presence of the polymer. Image 5.7(c) clearly shows that some Qdots occur in pairs. This could mean that one polymer can embed two Qdots at once. Figure 5.9 shows the DLS measurement. It shows that the stability of the core/shell Qdots in water is good since the size distribution stayed similar over time and the hydrodynamic diameter is approximately 12.37 nm. This average diameter disproves the assumption made by analysis of the TEM images that some Qdots occur in pairs. The PLQY of the 44

Chapter 5. Water solubilization of Qdots by means of a polymer shell CdSe/CdS Qdots in toluene is 43%. After the transfer to water the PLQY decreases to 37%. In comparison to the core CdSe particles this is an excellent result. NMR analysis of CdSe/CdS Qdots in water The resonances of both the polymer and the ligands are again visible in the 1D proton NMR spectrum shown in figure 5.10. Unfortunately most resonances of the ligands and polymer overlap. Again, no conclusions can be drawn from the NOESY spectrum in figure 5.11(a) as the cross peaks of the protons of the OA double bond can originate from the interactions with the polymer or from the interactions between different OA molecules. On the other hand, the alkene resonance of the OA is visible in the DOSY spectrum (figure 5.11(b)). The diffusion coefficient (31 10−12 m2 /s) and the hydrodynamic diameter (14.01 nm) of the Qdots is thus known exactly. The OA bound to the Qdots has the same diffusion coefficient as the polymer resonances. Therefore, it can be concluded the Qdots are embedded in a polymer shell. The resonance at 5.59 ppm is an unknown impurity, that only occurred in this sample. This diameter is larger than the diameter extracted from the DLS measurement.

5.1.6

Conclusion

To dissolve Qdots in water, the Qdots are embedded in a P(IB-alt-LMA) shell, the mixture is dried and the basic sodium borate solution is added. Then the water-soluble system is studied with various analytical techniques. TEM and UV-VIS show that the particles remained intact during the transfer. The PLQY decreases a little for the core/shell Qdots and a lot for the core Qdots. DLS measurements confirm that the system in water is very stable. The hydrodynamic diameter of the system, extracted from the DOSY spectra, confirms that the Qdots are embedded in a polymer shell. On the other hand, there was no confirmation of this conclusion in the NOESY spectra due to overlap of the polymer

(a)

(b)

(c)

Figure 5.7: TEM images of CdSe/CdS Qdots after transfer to water by means of P(IB-alt-LMA).

45

Chapter 5. Water solubilization of Qdots by means of a polymer shell

Figure 5.8: UV-VIS absorption spectrum of CdSe/CdS Qdots before and after transfer to water.

Figure 5.9: DLS measurement of CdSe/CdS Qdots embedded in a P(IB-alt-LMA) shell in water.

Figure 5.10: 1D proton NMR spectrum of CdSe/CdS Qdots in D2 O.

46

Chapter 5. Water solubilization of Qdots by means of a polymer shell

(a)

(b)

Figure 5.11: (a) NOESY spectrum of P(IB-alt-LMA) and ligands bound to CdSe/CdS Qdots in D2 O. (b) DOSY spectrum of P(IB-alt-LMA) and ligands bound to CdSe/CdS Qdots in D2 O.

and OA resonances. In the next sections, an attempt is made to overcome the problem of the overlapping resonances.

5.2

Water solubilization of Qdots by means of P(IBalt-UPA)

5.2.1

Protocol for dissolving nanoparticles in water

This protocol is somewhat different than the one with P(IB-alt-LMA). It was mentioned before that the polymer is not compatible with toluene thus the Qdots had to be dried prior to adding the polymer solution. The polymer was mixed with the dried quantum dots at a ratio of 100 repeating units per nm2 of effective surface of the Qdots. Calculations are the same as for P(IB-alt-LMA), shown in section 5.1.2. The mixture was completely dried by slowly evaporating the solvent at 60°C and 200 mbar and the solid film was then redissolved in an aqueous solution. Stirring and heating to 60°C facilitated the process. The aqueous solution was the same one as for the first protocol: millipore water with sodium borate and potassium hydroxide. This rendered a dispersion with a pH of ± 9.24; this is the pKa of boric acid.

47

Chapter 5. Water solubilization of Qdots by means of a polymer shell

5.2.2

CdSe Qdots with P(IB-alt-UPA) in water

Qualitative analysis of CdSe Qdots with P(IB-alt-UPA) in water The CdSe Qdots are transferred to water with the protocol described in section 5.2.1. The UV-VIS absorption spectra (figure 5.13) and TEM images (figure 5.12) confirm that the size and shape of the particles did not change during the transfer to water. In the TEM images the brightness of the background is not always the same. As before, the lighter parts are probably due to the presence of the polymer. The core Qdots display a PLQY of 1-2% in toluene and this value decreses to 0.10 % in water. NMR of CdSe Qdots with P(IB-alt-UPA) in water The resonances of both the polymer and the ligands are again visible in the 1D proton NMR spectrum shown in figure 5.14. The resonances in the CH2 /CH3 region overlap but between 4.5 and 6 ppm the resonances of OA and the polymer are clearly separated. Unfortunately, there is also 10-undecen-1-ol present in the sample which sometimes overlaps with the resonances of the polymer. The DOSY spectrum is shown in figure 5.15. The OA signals are not visible in this spectrum, meaning that the diffusion coefficient cannot be extracted directly. There are four different diffusion coefficients present. 30.37 10−12 m2 /s most probably corresponds to a micelle of 10-undecen-1-ol since the resonances at 3.4 and 5.65 ppm, that are part of 10-undecen-1-ol, only occur at this diffusion coefficient. Its value is smaller than the one in section 4.3.2 because the concentration of 10-undecen-1-ol is probably lower in that sample. The solvent has a diffusion coefficient of 685 10−12 m2 /s. The polymer resonances are found at 83.03 10−12 and 49.27 10−12 m2 /s. It can be assumed that the diffusion coefficient 83.03 10−12 m2 /s, corresponding to a hydrodynamic diameter of 5.48 nm, originates from the polymer and a diffusion coefficient of 49.27 10−12 m2 /s, corresponding to a hydrodynamic

(a)

(b)

(c)

Figure 5.12: TEM images of CdSe/CdS Qdots after transfer to water.

48

Chapter 5. Water solubilization of Qdots by means of a polymer shell

Figure 5.13: UV-VIS absorption spectrum of CdSe Qdots before and after transfer to water by means of P(IB-alt-UPA).

Figure 5.14: 1D proton NMR spectrum of the ligands bound to zincblende CdSe Qdots and P(IB-alt-UPA) in D2 O.

49

Chapter 5. Water solubilization of Qdots by means of a polymer shell diameter of 8.48 nm, originates from the polymer chains containing a Qdot.

Figure 5.15: DOSY spectrum of the ligands bound to zincblende CdSe Qdots and P(IB-alt-UPA) in D2 O.

The NOESY spectrum in figure 5.16(a) shows cross peaks between the resonance of the protons of the OA double bond (5.25 ppm) and the CH2 /CH3 region. It was expected to see a cross peak between this resonance and the resonances of the polymer at 4.9 and 5.8 ppm. However, this is not the case. In the spectra of CdSe Qdots in toluene, maximum 4 cross peaks could be identified between the protons of the OA double bond and that region. Here, the resonance at 5.25 ppm exhibits 6 cross peaks (figure 5.16(b)). Moreover, the resonance of 10-undecen-1-ol at 5.65 ppm shows a similar set of cross peaks to that region. This indicates that the OA bound to the Qdot is interacting with the polymer. Note that the resonance at 5.65 ppm also shows cross peaks to the pyridine resonances between 7 and 9 ppm. Pyridine has a high diffusion coefficient thus it is not part of the polymer-Qdot system, however, the pyridine will sometimes be located in the polymerQdot system and interact with the OA since cross peaks are visible between the resonances of these two species.

50

Chapter 5. Water solubilization of Qdots by means of a polymer shell

(a)

(b)

Figure 5.16: (a) NOESY spectrum of the ligands bound to zincblende CdSe Qdots and P(IB-alt-UPA) in D2 O. (b) Detail of the NOESY spectrum.

5.2.3

CdSe/CdS core/shell Qdots with P(IB-alt-UPA) in water

Qualitative analysis of CdSe/CdS Qdots with P(IB-alt-UPA) in water The CdSe/CdS Qdots are transferred to water via the protocol described in section 5.2.1. The UV-VIS absorption spectra (figure 5.18) and TEM images (figure 5.17) confirm that the size and shape of the particles did not change during the transfer to water. In the TEM images the brightness of the background is not always the same. The lighter parts are probably due to the presence of the polymer. In all three images, it seems that some Qdots occur in groups of two or three. This could mean that one polymer can embed more than one Qdot. However, the diffusion coefficient extracted from DOSY contradicts this (see further). The core/shell Qdots display a PLQY of 43% in toluene and this value decreses to 38 % in water. In comparison to the core Qdots, this is an excellent result. NMR of CdSe Qdots with P(IB-alt-UPA) in water The resonances of both the polymer and the ligands are again visible in the 1D proton NMR spectrum shown in figure 5.19. The resonances in the CH2 /CH3 region overlap but between 4.5 and 6 ppm the resonances of OA and the polymer are clearly separated. Unfortunately, there is also 10-undecen-1-ol present in the sample which sometimes overlaps with the resonances of the polymer. 51

Chapter 5. Water solubilization of Qdots by means of a polymer shell

(a)

(b)

(c)

Figure 5.17: TEM images of CdSe/CdS Qdots after transfer to water.

Figure 5.18: UV-VIS absorption spectrum of CdSe/CdS Qdots before and after transfer to water by means of P(IB-altUPA).

Figure 5.19: 1D proton NMR spectrum of the ligands bound to CdSe/CdS Qdots and P(IB-alt-UPA) in D2 O.

52

Chapter 5. Water solubilization of Qdots by means of a polymer shell In the DOSY spectrum, shown in figure 5.20, the OA signals are not visible. This implies that the diffusion coefficient of the Qdots cannot be derived directly from this spectrum. There are four different diffusion coefficients present. The solvent has a diffusion coefficient of 620 10−12 m2 /s. The 10-undecen-1-ol resonances are linked to two diffusion coefficients. One, at 33.50 10−12 m2 /s, most probably corresponds to a micelle of 10-undecen-1-ol molecules while the diffusion coefficient of 160.05 10−12 m2 /s, corresponding to a hydrodynamic diameter of 2.7 nm, most likely originates from single 10-undecen-1-ol molecules. The polymer resonances correspond to the last diffusion coefficient. It can be assumed that the diffusion coefficient of 44.42 10−12 m2 /s, corresponding to a hydrodynamic diameter of 9.81 nm, originates from the polymer chains containing a Qdot. The NOESY spectrum in figure 5.21(a) shows cross peaks between the resonance of the protons of the OA double bond (5.25 ppm) and the CH2 /CH3 region. It was expected to see a cross peak between this resonance and the resonances of the polymer at 4.9 and 5.8 ppm. This is not the case since the resonances of the polymer at 4.9 and 5.8 ppm do not exhibit any cross peaks. Similar to the NOESY spectrum of the CDSe Qdots, 5 cross peaks are visible (figure 5.21(b)). This is a larger number of cross peaks than expected for an interaction between two OA molecules. Moreover, the resonance of 10-undecen-1-ol at 5.65 ppm shows a similar set of cross peaks to that region. This indicates that the OA bound to the Qdot is interacting with the polymer.

5.2.4

Conclusion

To dissolve Qdots in water, the Qdots are embedded in a P(IB-alt-UPA) shell. This watersoluble system is studied with various analytical techniques. TEM and UV-VIS show that the particles remained intact during the transfer. The PLQY decreases a little for the core/shell Qdots and a lot for the core Qdots. The hydrodynamic diameter of the system, extracted from the DOSY spectra, demonstrates that the Qdots are embedded in a polymer shell. The NOESY spectra confirm this presumption since more that four cross peaks are visible between the proton resonance at 5.25 ppm of OA and the CH2 /CH3 region.

53

Chapter 5. Water solubilization of Qdots by means of a polymer shell

Figure 5.20: DOSY spectrum of the ligands bound to CdSe/CdS Qdots and P(IB-alt-UPA) in D2 O.

(a)

(b)

Figure 5.21: (a) NOESY spectrum of the ligands bound to CdSe/CdS Qdots and P(IB-alt-UPA) in D2 O. (b) Detail of the NOESY spectrum.

54

Chapter 5. Water solubilization of Qdots by means of a polymer shell

5.3

Saturation transfer difference NMR

A last experiment was conducted to confirm the interaction between the OA bound to the Qdots on the one hand and the polymer on the other. This experiment is explained in section 2.1.5 of chapter 2. The sample with the CdSe Qdots and a P(IB-alt-LMA) shell was chosen for this experiment. Figure 5.22 shows four spectra. The first one, (a), shows a reference spectrum of the Qdots without polymer in toluene-d8. (b) shows a reference 1D proton NMR spectrum of P(IB-alt-LMA) in D2 O. The third spectrum, (c), shows a 1D proton NMR spectrum of both the Qdots and the polymer in D2 O. The last spectrum depicts the STD experiment (d). The resonance of the protons at the double bond of OA at 5.22 ppm is irradiated. The saturation is transferred to certain protons on the condition that they are spatially close. The resonances of these protons are visible in the difference spectrum. The shape of the spectrum between 0 and 4 ppm resembles the spectrum of P(IB-alt-LMA) in D2 O. Moreover, there is a resonance visible at 2.63 ppm which is also present in the reference spectrum of P(IB-alt-LMA) in D2 O. This resonance is not present in the reference spectrum of the Qdots. We can conclude that the saturation was transferred from the OA to the polymer. This implies that these species are in proximity to each other and together with the diffusion coefficient extracted from DOSY, it confirms that the Qdots with OA ligands are embedded in a P(IB-alt-LMA) shell. In the aromatic region of the STD spectrum the pyridine resonances are visible as well. This implies that the pyridine is also in contact with the OA.

55

Chapter 5. Water solubilization of Qdots by means of a polymer shell

(a)

(b)

(c)

(d) Figure 5.22: (a) 1D proton NMR spectrum of OA ligands bound to CdSe Qdots in toluene-d8. (b) 1D proton NMR spectrum of P(IB-alt-LMA) in D2 O. (c) 1D proton NMR spectrum of OA bound to CdSe Qdots and P(IB-alt-LMA) in D2 O. (d) Saturation Transfer Difference spectrum of OA bound to CdSe Qdots and P(IB-alt-LMA) in D2 O.

56

Chapter 6 Conclusion A method to dissolve particles in water was described in literature by Parak et al [21, 25]. However, in literature, no real characterization of the water soluble system has been reported. Therefore, the characterization of the water soluble system was the main subject of this thesis. The goal of this thesis was to elucidate the composition of the Qdot embedded in a polymer shell. First, CdSe and CdSe/CdS Qdots were synthesized according to methods described in literature [32, 35]. The quality of these Qdots was analyzed by means of TEM, UV-VIS absorption and luminescence spectroscopy. The diameter and the size dispersion was good for the different kinds of Qdots. The PLQY on the other hand, was rather low for the CdSe Qdots but was good for the CdSe/CdS Qdots. It could therefore be concluded that these Qdots were suitable for this study. Then the Qdots in toluene were analyzed with NMR. The resonances were assigned to the protons of OA and the hydrodynamic diameter was extracted from the DOSY spectra. For the CdSe/CdS Qdots, it was not entirely sure that OA was the only ligand. Additional research is needed to identify the second ligand. Then, a polymer, poly(isobutylene-alt-n-laurylmaleamic acid) (P(IB-alt-LMA)), was synthesized according to methods described in literature [21, 25] and it was fully characterized by NMR. All proton resonances were assigned and the behavior of (P(IB-alt-LMA) in different solvents was studied. In solvents like water, the polymer adopts a conformation in which the side chains are on the inside and the backbone are in contact with the aqueous environment. In solvents like chloroform it will adopt a conformation with the side chains oriented towards the outside. The resonances of protons that are located on the inside of the conformation are broad due to lack of rotational freedom of movement. Next, the protocol to dissolve the Qdots with the polymer in water, described in literature, was tested and the water soluble system was analyzed with various techniques. TEM and UV-VIS show that the particles remained intact during the transfer to water. The 57

Chapter 6. Conclusion PLQY decreased to ± 37% for the core/shell Qdots and it decreased to ± 0.1% for the core Qdots. DSL measurements confirm that the Qdots in water are very stable. The hydrodynamic diameter of the systems, extracted from the DOSY spectra, confirm that the Qdots are embedded in a polymer shell. The hydrodynamic diameter of the residual toluene in the sample was significantly lower than expected. This is due to the fact that toluene was exchanging between the solvent and the Qdot environment. Unfortunately, there was no confirmation of this conclusion in the NOESY spectra due to overlap of the polymer and OA resonances. Therefore, an attempt was made to overcome the problem of the overlapping resonances by synthesizing a new polymer. A second polymer, poly(isobutylene-alt-3-(undecyloxycarbonyl) prop-2-enoic acid) (P(IBalt-UPA)), was synthesized. This polymer has a hydrophilic backbone and hydrophobic side chains with double bond at the end of each hydrophobic side chain. This polymer was fully characterized by NMR as well. P(IB-alt-UPA) displays the same behavior in different kinds of solvents as P(IB-alt-LMA) and again all resonances were assigned to the protons of the polymer. A slightly different protocol was developed to dissolve the Qdots in water by means of P(IB-alt-UPA). The water soluble system was analyzed with various techniques. TEM and UV-VIS show that the particles remained intact during the transfer to water. The PLQY decreased to ± 38% for the core/shell Qdots and it decreased to ± 0.1% for the core Qdots. The hydrodynamic diameter of the system, extracted from the DOSY spectra, demonstrated that the Qdots are embedded in a polymer shell. Unfortunately, there was no cross peak visible in the NOESY spectra between the double bond resonance of the polymer and the double bond resonance of the OA. However, this does not mean that there is no interaction between the polymer and the OA since the double bond resonance of the polymer did not display cross peaks. The OA double bond resonance did display six other cross peaks. Since maximum four cross peaks are expected for interaction between two OA molecules, this confirm that there is an interaction between the polymer and the OA. In the final stage of this research, an STD experiment was conducted to confirm the interaction between the OA bound to the Qdots on the one hand and the polymer on the other. The sample with the CdSe Qdots and a P(IB-alt-LMA) shell was chosen for this experiment. The resonance of the protons at the double bond of OA at 5.22 ppm was irradiated. The saturation was transferred to certain protons in the sample. The resonances of these protons were visible in the difference spectrum. The shape of the spectrum between 0 and 4 ppm resembles the spectrum of P(IB-alt-LMA) in D2 O. Moreover, there is a resonance visible at 2.63 ppm which is also present in the reference spectrum of P(IBalt-LMA) in D2 O. This resonance is not present in the reference spectrum of the Qdots. 58

Chapter 6. Conclusion We can conclude that the saturation was transferred from the OA to the polymer. This implies that these species are in close proximity to each other and together with the diffusion coefficient extracted from DOSY, it confirms that the Qdots with OA ligands are embedded in a P(IB-alt-LMA) shell. It can be concluded that the goal of this thesis was reached. The DOSY measurements confirmed that the Qdots in water are embedded in a polymer shell. Moreover, the NOESY measurements of the Qdots with P(IB-alt-UPA) showed interactions between the OA bound to the Qdots and the polymer. Finally, the STD measurement confirmed the interaction between the OA bound to the Qdot and P(IB-alt-LMA). DOSY and NOESY also confirmed that residual toluene and pyridine was exchanging between the solvent and the Qdotpolymer environment. Figure 6.1 shows a scheme of the composition of the Qdot-polymer system and the confirmed interactions.

nOe + DOSY

<

DOSY

nOe + DOSY nOe + STD

Figure 6.1: Scheme of the composition of the Qdot-polymer system and the confirmed interactions.

59

Chapter 6. Conclusion

60

Bibliography [1] Paul A Alivisatos. Nanocrystals: Building blocks for modern materials design. Endeavour, 21(2):56–60, January 1997. [2] Paul A Alivisatos. semiconductor clusters, nanocrystals and quantum dots. Science, 271:933–937, 1996. [3] AM Smith and Xiaohu Gao. Quantum Dot Nanocrystals for In Vivo Molecular and Cellular Imaging. Photochemistry and photobiology, 2004. [4] Zeger Hens, Jos´e C Martins, Iwan Moreels, and Bernd Fritzinger. Ligands for nanoparticles. In Comprehensive Nanoscience and Technology, chapter Ligands fo, pages 21 – 49. Oxford: Academic Press, 2011. [5] Wolfgang L. Kalb, Simon Haas, Cornelius Krellner, Thomas Mathis, and Bertram Batlogg. Trap density of states in small-molecule organic semiconductors: A quantitative comparison of thin-film transistors with single crystals. Physical Review B, 81(15):1–13, April 2010. [6] Margaret A. Hines and Philippe Guyot-Sionnest. Synthesis and Characterization of Strongly Luminescing ZnS-Capped CdSe Nanocrystals. The Journal of Physical Chemistry, 100(2):468–471, January 1996. [7] Yadong Yin and Paul A Alivisatos. Colloidal nanocrystal synthesis and the organicinorganic interface. Nature, 437(7059):664–70, September 2005. [8] Wolfgang J Parak, R Boudreau, M Le Gros, and D Gerion. Cell motility and metastatic potential studies based on quantum dot imaging of phagokinetic tracks. Advanced, 14(12):2000–2003, 2002. [9] G Albrecht-Buehler. The phagokinetic tracks of 3T3 cells. Cell, 11(2):395–404, June 1977. [10] G Albrecht-Buehler. Phagokinetic tracks of 3T3 cells: parallels between the orientation of track segments and of cellular structures which contain actin or tubulin. Cell, 12(2):333–9, October 1977. 61

Bibliography [11] WJ Parak, Daniele Gerion, and Teresa Pellegrino. Biological applications of colloidal nanocrystals. Nanotechnology, 14:R15 – R27, 2003. [12] J D McGhee and P H von Hippel. Theoretical aspects of DNA-protein interactions: co-operative and non-co-operative binding of large ligands to a one-dimensional homogeneous lattice. Journal of molecular biology, 86(2):469–89, June 1974. [13] H Ge. UPA, a universal protein array system for quantitative detection of proteinprotein, protein-DNA, protein-RNA and protein-ligand interactions. Nucleic acids research, 28(2):e3, January 2000. [14] T D Tullius and B a Dombroski. Hydroxyl radical ”footprinting”: high-resolution information about DNA-protein contacts and application to lambda repressor and Cro protein. Proceedings of the National Academy of Sciences of the United States of America, 83(15):5469–73, August 1986. [15] Donald R Helinskij. SUPERCOILED CIRCULAR DNA-PROTEIN COMPLEX IN ESCHERICHIA COLI : PURIFICATION AND INDUCED CONVERSION TO AN OPEN CIRCULAR DNA FORM *. 411:1159–1166, 1969. [16] J R Taylor, M M Fang, and S Nie. Probing specific sequences on single DNA molecules with bioconjugated fluorescent nanoparticles. Analytical chemistry, 72(9):1979–86, May 2000. [17] William W Yu, Emmanuel Chang, Rebekah Drezek, and Vicki L Colvin. Water-soluble quantum dots for biomedical applications. Biochemical and biophysical research communications, 348(3):781–6, September 2006. [18] Sander F Wuister, Ingmar Swart, Floris Van Driel, Stephen G Hickey, and Celso De Mello Donega. Highly Luminescent Water-Soluble CdTe Quantum Dots. Nano, 2003. [19] Sungjee Kim, YT Lim, EG Soltesz, and AM De Grand. Near-infrared fluorescent type II quantum dots for sentinel lymph node mapping. Nature, 22(1):93–97, 2003. [20] Fabien Pinaud, David King, Hsiao-Ping Moore, and Shimon Weiss. Bioactivation and cell targeting of semiconductor CdSe/ZnS nanocrystals with phytochelatin-related peptides. Journal of the American Chemical Society, 126(19):6115–23, May 2004. [21] Teresa Pellegrino, Liberato Manna, Stefan Kudera, and Tim Liedl. Hydrophobic nanocrystals coated with an amphiphilic polymer shell: a general route to water soluble nanocrystals. Nano Letters, 4(4):703 – 707, 2004.

62

Bibliography [22] Feng Zhang, Emma Lees, Faheem Amin, Pilar Rivera Gil, Fang Yang, Paul Mulvaney, and Wolfgang J Parak. Polymer-Coated Nanoparticles: A Universal Tool for Biolabelling Experiments. Small (Weinheim an der Bergstrasse, Germany), 7(22):3113– 3127, September 2011. [23] Daniele Gerion, Fabien Pinaud, Shara C Williams, Wolfgang J Parak, Daniela Zanchet, Shimon Weiss, and Paul A Alivisatos. Synthesis and Properties of Biocompatible Water-Soluble Silica-Coated CdSe / ZnS Semiconductor Quantum Dots. Society, pages 8861–8871, 2001. [24] Tangi Aubert, Fabien Grasset, St´ephane Mornet, Etienne Duguet, Olivier Cador, St´ephane Cordier, Yann Molard, Val´erie Demange, Michel Mortier, and Hajime Haneda. Functional silica nanoparticles synthesized by water-in-oil microemulsion processes. Journal of colloid and interface science, 341(2):201–8, January 2010. [25] Andrea D Lehmann, Wolfgang J Parak, Feng Zhang, Zulqurnain Ali, Carlheinz R¨ocker, G Ulrich Nienhaus, Peter Gehr, and Barbara Rothen-Rutishauser. Fluorescentmagnetic hybrid nanoparticles induce a dose-dependent increase in proinflammatory response in lung cells in vitro correlated with intracellular localization. Small, 6(6):753–62, March 2010. [26] Cheng-An J Lin, Ralph a Sperling, Jimmy K Li, Ting-Ya Yang, Pei-Yun Li, Marco Zanella, Walter H Chang, and Wolfgang J Parak. Design of an amphiphilic polymer for nanoparticle coating and functionalization. Small (Weinheim an der Bergstrasse, Germany), 4(3):334–41, March 2008. [27] Iwan Moreels, Jose C Martins, and Zeger Hens. Ligand adsorption/desorption on sterically stabilized InP colloidal nanocrystals: observation and thermodynamic analysis. Chemphyschem : a European journal of chemical physics and physical chemistry, 7(5):1028–31, May 2006. [28] Iwan Moreels, Bernd Fritzinger, and JC Martins. Surface chemistry of colloidal PbSe nanocrystals. Journal of the American Chemical Society, 130(45):15081–15086, 2008. [29] Bernd Fritzinger, Richard K Capek, Karel Lambert, Jos´e C Martins, and Zeger Hens. Utilizing self-exchange to address the binding of carboxylic acid ligands to CdSe quantum dots. Journal of the American Chemical Society, 132(29):10195–201, July 2010. [30] I Moreels, Y Justo, B De Geyter, and K Haustraete. Size-Tunable, Bright, and Stable PbS Quantum Dots: A Surface Chemistry Study. ACS nano, 5(3):2004–2012, 2011. [31] Thimothy Claridge. High resolution NMR techniques in organic chemistry. Elsevier, 2nd edition, 2009. 63

Bibliography [32] R Karel Capek, Iwan Moreels, and Karel Lambert. Optical properties of zincblende cadmium selenide quantum dots. The journal of physical chemistry. B, 114(C):6371– 6376, 2010. [33] T Lopez Arbeloa and F Lopez Arbeloa. Hydrogen-bonding effect on the photophysical properties of 7-aminocoumarin derivatives. The Journal of Physical Chemistry, 97(18):4704–4707, 1993. [34] M. Teresa Gandolfi Marco Montalti, Alberto Credi, Luca Prodi. Luminescence Spectroscopy Measurements. In handbook of photochemistry, volume 1, pages 561–581. CRC Press Taylor & Francis group, 2006. [35] Renguo Xie, Ute Kolb, and Jixue Li. Synthesis and characterization of highly luminescent CdSe-Core CdS/Zn0. 5Cd0. 5S/ZnS multishell nanocrystals. Journal of the American, 127(24):7480–7488, 2005.

64

Appendix A NMR data In this chapter, The assignment of NMR spectra of the different polymers that were synthesized is shown.

A.1

Poly(isobutylene-alt-maleic anhydride)

The structure of Poly(isobutylene-alt-maleic anhydride) is depicted in figure A.1. In one repeating unit, there are two CH’s, one CH2 and two methyl groups. Note that there are 2 stereocenter in the structure. Probably the polymer will be a racemic mixture of RR, SS, RS and SR configuration. The RR and the SS configuration will form a diastereomeric pair, meaning that these will give rise to the same resonances in NMR spectroscopy. RS and SR configuration also form a pair. At first sight, the 1D proton NMR spectrum shows more resonances than expected (figure A.1). This is a consequence of the racemic mixture. The assignment of the stereochemistry is not a part of this study. Note that there are some resonances in the aromatic region as well. There are no aromatic protons in the polymer thus these protons are originating from an impurity. This is confirmed by the DOSY measurement (figure A.3(b)). By looking at the phase of the cross peaks in the HSQC spectrum (figure A.3(a)), it is clear that the CH3 protons are located in the region between 1 and 1.5 ppm. Four individual CH3 resonances can be distinguished, in agreement with the fact that for each diastereomeric pair there are two CH3 resonances. Four CH2 resonances are located between and 1.8 and 2.5 ppm. Two of them connect to a carbon at 42.32 ppm and two connect to a carbon at 43.33 ppm. In COSY, the two proton resonances with the same carbon chemical shift value couple with each other. Each of these resonances represents one proton of the CH2 in a diasteromeric pair.

65

Appendix A. NMR data

Figure A.1: 1D proton NMR spectrum of poly(isobutylene-alt-maleic anhydride) in THF-d8. The sterocenters are indicated in the figure with a and b.

Figure A.2: HSQC spectrum of poly(isobutylene-alt-maleic anhydride) in THF-d8.

66

Appendix A. NMR data

(a)

(b)

Figure A.3: (a) COSY spectrum of poly(isobutylene-alt-maleic anhydride) in THF-d8. poly(isobutylene-alt-maleic anhydride) in THF-d8.

(b) DOSY spectrum of

Four CH resonances are found in the region between 3 and 3.5 ppm. Using HSQC and HMBC (figure A.4), it is clear that only the protons that connect with the carbons with a higher chemical shift value couple with the CH3 region. These can thus be assigned to CH number 2 for two diastereomeric pairs.

Figure A.4: HMBC spectrum of poly(isobutylene-alt-maleic anhydride) in THF-d8.

67

Appendix A. NMR data

A.2

P(IB-alt-LMA) in chloroform-d

The structure of Poly(isobutylene-alt-n-laurylmaleamic acid) is depicted in figure A.5. In one repeating unit, there are two CH’s, twelve CH2 ’s and three methyl groups. By looking at the phase of the cross peaks in the HSQC (figure A.7) it is clear that the proton resonances at 0.85 and 1.05 ppm are methyl groups and the other resonances are CH2 groups. The CH protons are not visible in the proton spectrum. The COSY spectrum (figure A.6(b)) shows that the resonance at 0.85 ppm is in contact with the large resonance at 1.23 ppm. Therefore, we can identify the resonance at 0.85 as methyl group number 8 and the big resonance at 1.23 ppm represents a collection of protons of the side chain of the polymer (number 7 in the figure). There are also cross peak between the resonance at 1.23 ppm and the one at 1.52 ppm. This last resonance represents the protons in position 6. This resonance couples with the resonance at 3.22 ppm which represents the protons at position 5. The protons at position 2,3 and 4 can not be distinguished. In chloroform, these protons are located on the inside of the composition where they can not move easily. Therefore these signals are very broad. It can be said that they are located between 1.30 and 3.20 ppm. The chemical shift and the assignment of the resonances is shown in table A.1.

Table A.1: This table shows the chemical shift and the assignment of the resonances in the 1D proton NMR spectrum of P(IB-alt-LMA).

# 1 2 3 4 5 6 7 8

chemical shift 1.05

3.22 1.52 1.23 0.85

68

Appendix A. NMR data

Figure A.5: 1D proton spectrum of P(IB-alt-LMA) in chloroform-d.

(a)

(b)

Figure A.6: (a) NOESY spectrum of P(IB-alt-LMA) in chloroform-d. (b) COSY spectrum of P(IB-alt-LMA) in chloroformd.

69

Appendix A. NMR data

Figure A.7: HSQC spectrum of P(IB-alt-LMA) in chloroform-d.

Figure A.8: HMBC spectrum of P(IB-alt-LMA) in chloroform-d.

70

Appendix A. NMR data

A.3

P(IB-alt-LMA) in D2O

With the assignment of P(IB-alt-LMA) in chloroform, explained in section A.4, in mind, we can now assign the resonances of the 1D proton NMR spectrum of P(IB-alt-LMA) in D2 O. Because most resonances have the same chemical shift in water as in chloroform, the explanation of the assignment of these resonances is not needed. However, there are some differences visible. At 1.88, 2.20 and 2.65 ppm there are resonances that were not present in chloroform. These resonances belong to the protons at positions 2, 3 and 4. In NOESY it is clear that these resonances connect with each other. The resonance at 2.65 ppm has the strongest coupling with the resonance at 1.05 ppm which means that the resonance at 2.65 ppm represents the proton an position 2. The two other resonances can not be distinguished with absolute certainty. The chemical shift and the assignment of the resonances is shown in table A.2.

Table A.2: This table shows the chemical shift and the assignment of the resonances in the 1D proton NMR spectrum of P(IB-alt-LMA) in D2O.

# 1 2 3 4 5 6 7 8

chemical shift 0.98 2.65

3.09 1.49 1.25 0.84

71

Appendix A. NMR data

(a)

(b) Figure A.9: (a) 1D proton spectrum of P(IB-alt-LMA) in D2 O. (b) NOESY spectrum of P(IB-alt-LMA) in D2 O.

72

Appendix A. NMR data

Figure A.10: HSQC spectrum of P(IB-alt-LMA) in D2 O.

A.4

P(IB-alt-UPA) in THF-d8

The structure of P(IB-alt-UPA) is shown in figure A.11. One structural unit contains 3 CH’s, 11 CH2 ’s and 2 CH3 ’s. The 1D proton spectrum of this polymer in THF-d8 is shown in figure A.11. It is immediately clear that the resonances between 4.5 and 6 ppm, in the double bond region, can be assigned to the protons at positions 1 and 2. Looking at the phase of the cross peaks in the HSQC (figure A.13(b)), it is clear that the proton resonance at 4.95 ppm is originating from a CH2 (protons at position 1) and the resonance at 4.95 ppm is originating from a CH (protons at position 2). COSY (figure A.13(a)) shows a clear cross peak from these resonances to the resonance at 2.07 ppm. It can thus be concluded that this resonance belongs to the protons at position 3. The resonance at 2.07 ppm also shows a coupling to the resonance at 1.35 ppm which is originating from the protons at position 4. The phase of the resonances in HSQC confirm these conclusions. COSY shows a coupling between the resonance at 1.35 and the one at 1.67. The latter also couples with the resonance at 4.04 ppm. Therefore, the resonance at 1.67 ppm corresponds to the protons located at position 5 and the one at 4.04 ppm corresponds to position 6. With the spectrum of poly(isobutylene-alt-maleic anhydride) in mind, it is clear that the methyl groups (9) are located between 1 and 1.30 ppm. These show a coupling in COSY with the resonances around 1.93 ppm, corresponding to protons 8. Finally, resonance 8 couples with the resonances around 3.39 ppm originating from the protons at position 7. DOSY (figure A.12(a)) shows that some residual 10-undecen-1-ol is left. This manifests itself in additional resonances, for example at 3.49 ppm. The chemical shift and the assignment of the resonances is shown in table A.3.

73

Appendix A. NMR data

Table A.3: This table shows the chemical shift and the assignment of the resonances in the 1D proton NMR spectrum of P(IB-alt-UPA) in THF-d8.

# 1 2 3 4 5 6 7 8 9

chemical shift 4.96 5.84 1.95 1.35 1.67 4.04 3.26 1.94 and 2.43 1.17

Figure A.11: 1D proton spectrum of P(IB-alt-UPA) in THF-d8.

74

Appendix A. NMR data

(a)

(b)

Figure A.12: (a) DOSY spectrum of P(IB-alt-UPA) in THF-d8. (b) NOESY spectrum of P(IB-alt-UPA) in THF-d8.

(a)

(b)

Figure A.13: (a) COSY spectrum of P(IB-alt-UPA) in THF-d8. (b) HSQC spectrum of P(IB-alt-UPA) in THF-d8.

75

Appendix A. NMR data

A.5

P(IB-alt-UPA) in D2O

With the assignment of P(IB-alt-UPA) in THF, explained in section A.4, in mind, we can now assign the resonances of the 1D proton NMR spectrum of P(IB-alt-UPA) in D2 O. Because most resonances have the same chemical shift in water as in THF, the explanation of the assignment of these resonances is not needed. However, there are some differences visible. First of all, the resonances that were on the inside of the structural conformation will now be on the outside. This implies that the broad resonances will become more sharp and the resonances that are on the inside, will be less sharp. As residual 10-undecen-1ol is also present in this sample, this will manifest itself in extra resonances. This is for example the case between 5.5 and 6 ppm. The broad resonance originates from the polymer (protons at position 2) and the sharp resonance originates from the reagent. Protons 1 are again located at 4.85 ppm, although the resonance is suppressed by the water suppression. Finally, the resonances of protons 7 have shifted to a lower chemical shift value. The chemical shift and the assignment of the resonances is shown in table A.4.

Table A.4: This table shows the chemical shift and the assignment of the resonances in the 1D proton NMR spectrum of P(IB-alt-UPA) in D2O.

# 1 2 3 4 5 6 7 8 9

chemical shift 4.85 5.82 1.95 1.24 1.55 3.99 2.59 and 3.05 2.14 0.95

76

Appendix A. NMR data

(a)

(b)

Figure A.14: (a) NOESY spectrum of P(IB-alt-UPA) in D2 O. (b) HSQC spectrum of P(IB-alt-UPA) in D2 O.

77