Small molecule thin films for photodynamic anti-bacterial therapy

(1)

Small molecule thin films for

photodynamic anti-bacterial therapy

Johan Kroeze, 10334378

VU University - Faculty of Sciences

Report Bachelor Project Physics and Astronomy - 15 EC Conducted between 01-04-16 and 01-07-16

Submitted 22-07-2016 Supervisor: Prof. Dr. E.L. von Hauff Second assessor: Prof. Dr. Ir. Erwin J.G. Peterman

Abstract

Photodynamic therapy (PDT) is the usage of photoactive materials for anti-microbial treatment. Organic semiconductors like metallo-phtalocyanines are found to be very efficient materials for PDT, so research into their properties is growing. The goal of this research is to establish a protocol for the thermal evaporation of thin copperphtalocyanine (CuPc) films, and to check what happens to the CuPc thin film properties when they are sterilized. To establish a production protocol, the evaporation chamber had to be calibrated by evaporating well-known materials. Zinc and C60 thin films were made, their thickness was measured and a tooling factor of 1 (zinc) respectively 2.13 (C60) was found. Then CuPc thin films were made by thermal evaporation and their optical properties characterised. The films were sterilized whereafter their optical properties were characterised again. Although a few films were mechanically destroyed during sterilization, the optical properties of the surviving films proved out to be not influenced by sterilization.

(2)

I. Samenvatting

Moleculen kunnen in aangeslagen toestand komen door licht van een bepaalde golflengte erop te schijnen: het licht wordt geabsorbeerd en de energie wordt door het molecuul opgenomen. Uiteindelijk zendt de stof het opgenomen licht weer uit, het verliest de extra energie en valt terug in zijn grondtoestand. Bij sommige stoffen gebeurt er op dat moment iets bijzonders: de energie die vrij komt wordt overgedragen aan het nabije zuurstof, wat daardoor radicaliseert. Dit betekent dat het zuurstof agressief wordt en het nabije weefsel aanvalt.

Een vrij nieuwe methode om ziektes te genezen is photodynamische therapie. Hierbij wordt het hierboven beschreven proces gebruikt om kwaadaardige (tumor)cellen, bacteriën en virussen te doden. Hiervoor zijn stoffen nodig met speciale eigenschappen. Dit onderzoek gaat over zo’n stof: koperphtalocyanine, ook wel CuPc genoemd, en hoe deze stof bruikbaar kan worden gemaakt voor photodynamische therapie.

Om CuPc te kunnen gebruiken voor photodynamische therapie moet er een heel dun laagje, van 20 á 200 nanometer dikte, op een plaatje worden aangebracht. Dit wordt gedaan door de stof te verdampen in een vacuüm verdampingskamer en op te vangen op glasplaatjes. Je krijgt dan zogenaamde dunne films. Deze dunne films absorberen bepaalde golflengtes van licht. In dit onderzoek is uitgezocht welke golflengtes dat zijn, hoe sterk ze die absorberen, en of dat allemaal niet veranderd als de plaatjes met CuPc worden gesteriliseerd.

Wanneer een stof wordt verdampt in de vacuüm kamer, wordt bijgehouden hoe dik de verdampte laag is. Dat wordt gedaan met behulp van speciale kristallen die worden gebruikt als sensoren in de vacuüm kamer. Deze kristallen trillen met een bepaalde frequentie die verandert als er massa op het oppervlak van het kristal komt. De kristallen zijn aangesloten op een machine die de frequentie van de kristallen meet en daarmee berekent hoe dik het laagje op het kristal - en dus op het glasplaatje - is. De kristallen bevinden zich echter niet op precies dezelfde plek als de glasplaatjes, waardoor er een factor verschil zit tussen de laagdikte op die twee. Deze factor moest worden gevonden zodat de verdampingskamer goed werkt: de verdampingskamer moest worden gekalibreerd. Dat is gedaan door een aantal dunne films te maken van verschillende stoffen en hun laagdikte na te meten. De verdampte stoffen zijn CuPc, zink en C60.

Het blijkt dat de gesteriliseerde dunne CuPc films goed bruikbaar zijn voor photodynamische therapie. De aborptie-eigenschappen veranderen niet.

Voor elke verdampte stof is een goede factor gevonden, maar wel voor elke stof een verschil-lende. Dat kan komen doordat bepaalde eigenschappen van CuPc en C60 nog niet zo goed bekend zijn, waardoor de laagdikte verkeerd berekend wordt, of doordat de verschillende stoffen een verschillende vorm hebben.

Voor verder onderzoek moeten de eigenschappen van CuPc en C60 beter in kaart worden gebracht. Ook moet er gekeken worden naar de invloed van de vorm van de verdampte stof op het verdampingsproces.

Figure 1: Schematische weergave van de werking van photodynamische therapie om een tumor te genezen. Stap 1:

Injecteer de werkende stof. Stap 2: De werkende stof concentreert zich in de tumor. Stap 3: Activeer de stof met licht. Stap 4: De tumor wordt vernietigd.

(3)

II. Introduction

Organic semiconductors have good potential for electronic and photonic applications. One of these applications is photodynamic therapy (PDT), which is an anti-microbial therapy that uses photoactive materials to treat bacterial and fungial infections, to fight plaque and sometimes even to treat forms of cancer [Konopa, K. and Goslinsky, T., (2007)]. Copperphtalocyanine (CuPc) is an organic semiconductor that is used for PDT due to it’s interesting absorption- and photoproperties. In this thesis CuPc was investigated for its potential in PDT. To do this, thin CuPc films were made thermally evaporated and their optical properties were characterised. Then they were sterilized, after which their optical properties were characterised again. Some of the films were mechanically destroyed during sterilization, but most survived.

The thickness of the films in process is monitored by quartz cyrstals inside the evaporation chamber. The evaporation chamber used was not yet calibrated at the start of this research, so a side goal became the calibration of the evaporation chamber. This was done by evaporating well-known materials and measuring their layer thickness using interferometry. Halfway in the process the research stumbled upon some fundamental hardware problems concerning the CuPc evaporations, which made it impossible to apply the general calibration to the CuPc evaporations. Therefore a specific calibration for CuPc evaporations had to be done.

III. Theory

i.

Electronic structure of organic semiconductors

Organic semiconductors are carbon-based materials that are sp2hybridized: two p orbitals and one s orbital form three sp2orbitals. With the hybridization of the material π bonds form between the atoms, which causes an delocalised electron-density in the molecule. Organic semiconductors with π-bonds have low-energy exited states called exitonic states because π-bonds are weak, making the material semi-conductive and crystal-like. An exitonic state is a neutral electronic excitation that has a lower excitation energy than the energy that is required to excite an electron from the valence band into the conduction band [Schwoerer. M. and Wolf, H.C., (2007)].

A higher energy state can be a triplet or a singlet state. As a rule, the singlet states have a bit higher energy then the triplet states and have counterpaired electronic spins, meaning the total spin is equal to zero. Triplet states have a somewhat lower energy then triplet states and have two parallel aligned electron spins, making the total spin equal to one, see figure 2 [Schwoerer. M. and Wolf, H.C., (2007)].

The ground state of organic semiconductors is always a singlet state. When jumping from singlet to triplet states or vice versa, a spin-flip occurs. This is called intersystem crossing and is generally forbidden, but organic semiconductors can break this prohibition because of the weak spin-orbit coupling that is caused by the weak π-bonds. If heavy atoms are present in the molecule the chance of intersystem crossings becomes even greater [Schwoerer. M. and Wolf, H.C., (2007)]. When this happens in the presence of oxygen, which has a triplet ground state, it can react and transform to highly radical singlet oxygen, which causes localized photodamage and cell death [Konopa, K. and Goslinsky, T., (2007)].

Absorption=I0−T−R (1)

When light of specific wavelengths is directed at an organic semiconductor it can be absorbed, thereby exciting the material. By measuring the reflectance and transmittance of the material in a range of wavelengths, an absorption spectrum can be made with equation 1, which gives

(5)

information about the wavelengths that excite the material into a singlet or triplet state. Here I0is

the intensity of the reference lightbeam, T is the percentage transmission and R is the percentage reflection.

Figure 2: Higher energy singlet (S1) and triplet (T1, T2) excited states, the groundstate (S0) and vibronic states shown

schematically. Singlet states are diamagnetic and have total spin quantum number S=0, while triplet states have S=1. That is why transitions between singlet and triplet states are generally forbidden. Here the groundstate is a singlet state, which means the lowest energy triplet state has a long lifetime. Oxygen, for example, has a triplet groundstate, and a long lifetime lowest energy singlet state.

IV. Methods

i.

Materials and preparation

Before evaporating, glass samples from 25x25mm were cleaned in the following order. • 5 minutes ultrasound bath in pure H20

• Five minutes ultrasoundbath in acetone • Five minutes ultrasoundbath in isopropanol • Fifteen minutes ozone oven

The samples were put in a MBraun nitrogen filled chamber –the glovebox –and from there into the Siemens Simatic HM1 evaporation chamber. Each evaporation contained two or three substrates. The material to be evaporated was also put in the evaporation chamber directly from the glovebox. The evaporation chamber evacuated till at least 1∗10−6 millibar before heating up the source for evaporation. Quartz crystals inside the evaporation chamber are connected to an Inficon SQC-310C deposition controller to monitor the layer thickness of the film in process. For this to work well, two paramaters of the evaporated material have to be set in the Inficon deposition controller: the density and Z-factor of the evaporated material. After evaporation, the film thickness of each film was measured using a Veeco Wyko NT9100 interferometer, and the transmittance and reflectance of each film was measured using a PerkinElmer Instruments Lambda 900 spectrometer (UV/Vis/NIR).

(6)

Material Density Z-factor purity Evaporation

(g/cm3₎ _(percent) _{temperature (celsius)}

CuPc 1.60 15 99 320

Zn 7.04 0.514 99 300

C60 1.65 3.45 99.9 450

Table 1: Evaporated materials and their properties. The properties of zinc are well known, as are the density of CuPc

and C60. The Z-factor of CuPc and C60 were found in the literature to be 15 respectively 3.45 [Botha, A. F., (2010)]. All materials were over 99 percent purity and ordered at Sigma Aldrich.

CuPc has a density of 1.60 g/cm3 and a Z-factor of 15 was found [Botha, A. F., (2010)]. Unfortunately the highest Z-factor the deposition controller could set was 9.999, so we set it to 9.999 for the CuPc evaporations. This means the evaporation chamber could not calibrated with the CuPc, nor could a general calibration be applied to the CuPc evaporations. For the calibration of the evaporation chamber zinc was evaporated, because its density and Z-factor are well known to be 7.04 g/cm3_{and 0.514. Because zinc is not an organic but a metal, a C60 control calibration}

was done. C60 is a organic material a density of 1.65 g/cm3and a Z-factor of 3.45 [Botha, A. F., (2010)]. The CuPc, zinc and C60 were all 99 percent or more pure. All the specifics are to be found in table 1.

Figure 3 shows the structure of CuPc. The special thing about phtalocyanines is that the middle spot is open for substition of metals. In copperphtalocyanine that place is filled by a copper atom.

Figure 3: The structure of copperphtalocyanine (CuPc). Phtalocyanine is a material which has an ’empty’ spot in the

middle, which can be filled by any kind of metal. CuPc is phtalocyanine with a substituted copper atom.

ii.

Thermal evaporations of thin films

General principles

Organic semiconductors which consist of only one type of molecules exhibit pure van der Waals bonding, which is mainly responsible for the cohesion within molecular solids. Therefore, the properties of individual molecules in such materials remain intact during evaporation. This means that when evaporating a molecular solid like CuPc, the properties of individual molecules stay intact and when the molecules hit the substrate, they form a crystal. This makes thermal evaporation a very viable technique for the creation of organic thin films. In this research the materials are evaporated at vacuum unto glass substrates. Evaporation at vacuum is preferable above evaporation at normal pressure because of two reasons. Firstly, any gas present in the evaporation chamber can contaminate the thin film that is created. Secondly, the minimal temperature needed to evaporate any material is lowered at lower pressures.

(7)

Quartz crystals inside the vacuum chamber and Inficon software are used to monitor the thickness of the film in process. These crystals have a resonating frequency, changing with the mass on the surface of the crystal. By measuring the change in frequency, the film layer thickness on the crystal is calculated with equation 2, the Sauerbrey equation. To make this calculation the density and Z-factor of the evaporated material must be known. The inficon software knows the rest of the parameters, it measures the frequency of the crystal and it does the calculation.

∆m A = Nqρq πZ fL arctan[Z tan(πfU−fL fU )] (2)

fL =Frequency of loaded crystal (Hz)

fU=Frequency of unloaded crystal (Hz)

Nq=Frequency constant for AT-cut quartz crystal (1.668∗1013 Hz·Å)

∆m=Mass change (g)

A=Area of quartz crystal (cm2₎

ρq=Density of quartz crystal (ρq=2.648 g/cm3)

Z=Z-factor=qρ_ρqµq

fµf

ρf =Density of the film (g/cm3)

µq=Shear modulus of quartz crystal ( uq=2.947∗1011 g∗cm−1∗s−2)

µf =Shear modulus of film (g∗cm−1∗s−2)

Since the quartz crystals and the substrates cannot be at the exact same position, the film thickness on the crystals will not be exactly the same as the film thickness on the substrates. There is a factor difference called the tooling factor. This is a purely geometrical factor, which can be found by remeasuring the deposited film thickness and comparing this to the measured layer thickness on the quartz crystals, using equation 3. Here, TF is the tooling factor, Thinter f erometryis

the layer thickness that was measured using interferometry and Thin f iconis the layer thickness on

the quartz crystals.

TF= Thinter f erometry Thin f icon

(3)

Evaporation chamber setup

The evaporation chamber consists of a few essential elements: The material to be evaporated is put into the source, where it gets heated up. While the evaporation is in process, quartz crystals inside the evaporation chamber in combination with the Inficon deposition controller are used to monitor the rate of the evaporation and the thickness of the layer in process. When the desired evaporation rate is reached, the substrate shutter is opened so that the material reaches and attaches to the substrates. Figure 4 shows the setup schematically. It must be noted that the evaporation chamber that was used in this research was a bit asymmetric.

(8)

Figure 4: Schematic: Inside an evaporation chamber. When vacuum is reached the source is heated up, evaporating the

material inside onto the substrates. Quartz crystals monitor the film thickness in process. The source shutter can open and close the source. The evaporation chamber that was used did not have a sensor shutter, but it did have a substrate shutter to open or close the path from source to substrate.

iii.

Measurements of thin film thickness

The deposited film thickness is measured using interferometry, a technique which uses the interference of reflected light to determine the thickness of a sample. When a beam of light is directed at the sample, part of the light will reflect directly at the outer surface, another part will reflect at the inner surface and the rest will pass through. The two reflected light beams have travelled different lengths and thus have a different phase, as illustrated in figure 5. This causes the light to infer, enhancing or diminishing each others intensity depending on the thickness of the film [Hariharan, P., (2007)].

Figure 5: Light waves travel differenth lengths, causing interference.

To verify the validity of interferometry measurements on glass samples, one sample of every C60 evaporation was measured, then coated with an thin layer of gold and then remeasured.

iv.

Absorption spectroscopy

After the sterilization of the CuPc samples their reflectance and transmittance were measured in order to make their absorption spectra. The samples were one by one mounted in front of a reflecting sphere with a small hole. Then light of wavelengths between 280nm and 860nm was

(9)

directed at them, from which a part passed through the sample and through the hole. This part got trapped inside the reflecting sphere until it reached a sensor, measuring the intensity of the transmission and comparing it to a reference measurement. The same was done again, but now the sample was mounted at the back of the reflecting sphere, while measuring the intensity of the reflected light at each wavelength and comparing this with the reference measurement. The reference measurement was made beforehand by measuring the intensity of light of wavelengths between 280nm and 860nm that was directed at a clean glass sample that was mounted in front of the reflecting sphere. The transmission and reflection spectra are added up and then subtracted from the reference measurement. Using equation 1, this gave the absorption spectra of the cleaned CuPc samples. Normalized absorption spectra were made in the same way, but then normalizing the added up transmission and reflection spectra and then substracting them from the reference measurement.

The absorption A of light in a material is lineairly related to the molar absorptivity e of the material, the concentration c of the material and the path length or thickness d of the sample. Equation 4, the Beer-Lambert Law describes this relationship. Here I0 is the intensity of the

incident beam of light and I is the intensity of the transmission. I

I0

(10)

V. Results

i.

Calibration of the thermal evaporation equipment

Table 2: General calibration: Thicknesses of multiple Zn and c60 films. Thickness inficon is the layer thickness that

was read from the deposition controller and thickness interferometry is the measured layer thickness. The tooling factor is calculated by dividing the measured thickness by the inficon thickness. Zinc evaporation 1 to 4 were flawed and must be discarded.

Evaporation Sample Thickness Thickness Tooling Comments

Inficon (nm) interferometry (nm) Factor

Zn 1 1a 50 577 11.54 Oxidized

Zn 1 1b 50 631 12.62 Oxidized

Zn 2 2a 85 2203 25.90 Oxidized

Zn 2 2b 85 1962 23.08 Oxidized

Zn 3 3a 26 X X Did not attach

Zn 3 3b 26 X X Did not attach

Zn 4 4a 100 X X Did not attach

Zn 4 4b 100 X X Did not attach

Zn 5 5a 250 204 0.82 Zn 5 5b 250 275 1.10 Zn 6 6a 350 382 1.09 Zn 6 6b 350 449 1.28 Zn 7 7a 200 126 0.63 Zn 7 7b 200 230 1.15 Zn 8 8a 150 127 0.85 Zn 8 8b 150 163 1.09 C60 1 1a 100 215 2.15 C60 1 1b 100 255 2.55 C60 1 1c 100 246 2.46

C60 1 1c Gold 100 248 2.48 Gold coating

C60 2 2a 50 82 1.64

C60 2 2b 50 86 1.72

C60 2 2c 50 79 1.58

C60 3 3a 25 55 2.20

C60 3 3b 25 58 2.32

C60 3 3c 25 56 2.24

For the calibration of the evaporation chamber, eight zinc evaporations and three C60 evaporations were done. Each sample was remeasured to calculate the tooling factor with equation 3. The results are summarized in table 2. The first four zinc evaporations can be discarded: the first two were done with oxidized zinc and in the third and fourth evaporation the zinc didn’t attach to the glass substrates well. The fifth to eighth zinc evaporations went well and an average tooling factor of 1.00 was found. The C60 control evaporations all went well, and an average tooling factor of 2.13 was found. As seen in table 2, one sample of each C60 evaporation was measured, then

(11)

coated with a thin layer of gold and then remeasured. This was done to validate the interferometry measurements on glass. No big deviations in thickness were found between the gold plated C60 samples and the normal C60 samples.

Table 3: CuPc calibration: Thicknesses of multiple CuPc films. Thickness inficon is the layer thickness that was read

from the deposition controller and thickness interferometry is the measured layer thickness. The tooling factor is calculated by dividing the measured thickness by the inficon thickness.

Evaporation Sample Thickness Thickness Tooling factor

Inficon (nm) interferometry (nm) CuPc 1 1 37 220 5.95 CuPc 1 2 37 228 6.16 CuPc 1 3 37 228 6.16 CuPc 2 4 12 53 4.42 CuPc 2 5 12 55 4.58 CuPc 3 6 20 110 5.5

Three CuPc evaporations were done, containing six samples in total. The thickness of each of these samples was measured using interferometry, from which a specific CuPc tooling factor was calculated to be around 5.37. The results are summarized in table 3. The first evaporation turned out to be around 220nm thick instead of 37nm. A proper tooling factor for this thickness would be around 6.0. The second evaporation turned out to be around 55nm thick instead of 12nm. A proper tooling factor would be around 4.5 for this thickness. The third evaporation gave a 110nm film, instead of 20nm. A proper tooling factor for this thickness would be around 5.5. Figure 6 shows that the connection between the initial and the remeasured layer thickness seems to be lineair, as it should be.

Figure 6: The CuPc layer thicknesses as measured on the quartz crystals vs those measured using interferometry. The

(12)

thickness are shown. The dotted line is the absorption spectrum of a non-sterilized reference film, the straight lines are the absorption spectra of sterilized films.

Figure 7: Absorption spectrum of CuPc thin films of different thicknesses. The dotted line is the absorption spectrum

of the non-sterilized reference film, the straight lines are the absorption spectra of the sterilized films. The thicker the layer, the higher the peaks.

Each film has three peaks, the highest at 330nm, the second at 615nm and the third at 690nm. The peaks get higher as the layer thickness gets thicker and the sterilized-film peaks are a bit broader than the reference peak. At a wavelength of 330nm the reference film of 105nm has a absorption peak of 45 percent, the 53nm sterilized film has a absorption peak of 17 percent, the 110nm sterilized film has a absorption peak of 75 percent and the 220nm sterilized film has a absorption peak of 85 percent. At a wavelength of 615nm the reference film has a absorption peak of 45 percent, the 53nm sterilized film has a absorption peak of 10 percent, the 110nm sterilized film has a absorption peak of 70 percent and the 220nm sterilized film has a absorption peak of 83 percent. At a wavelength of 690nm the reference film has a absorption peak of 27 percent, the 53nm sterilized film has a absorption peak of 3 percent, the 110nm sterilized film has a absorption peak of 45 percent and the 220nm sterilized film has a absorption peak of 75 percent.

VI. Discussion

If the density and Z-factor of an evaporated material are correct, each evaporation should give the same tooling factor. This however, is not the case. The zinc evaporations gave an average tooling factor of 1.00, the C60 evaporations gave an average tooling factor of 2.13 and the CuPc evaporations gave an average tooling factor of 4.48. As the densities of the different materials and the Z-factor of zinc are well-known the most probable explanations for these deviations are twofold.

In the first place a wrong Z-factor for CuPc and C60, as these Z-factors are fairly unknown: only one source for these Z-factors was found, so it could be that this source was wrong.

Secondly, as for the difference between the zinc and C60 tooling factors, this could also be explained by the asymmetry of their different form. The C60 was a powder and the zinc was in the form of a cyllindrical stick. When the powder is put in the source it distributes and heats up evenly, hence it does not have a preferred direction of evaporation. The stick however, stands

(13)

tilted to the side in the source and only has contact with the source on one side. The zinc does not heat up evenly and the stick blocks itself when evaporating. For clarification see figure 8.

In other words: none of the calibration results is reliable for the establishment of a trusthworthy general tooling factor for the evaporation chamber, but the inner compliance within the series of materials is clear: in the results it is observed that the different tooling factors that were found per material are very close and very reproducible and have little significant deviation, even for the CuPc evaporations of which it is known that the tooling factor set in the deposition controller was wrong.

Figure 8: Schematic of the way the zinc cyllindrical stick is placed in the source. This could lead to assymetric

evaporation and thus to a deviation in tooling factor.

VII. Conclusion

In the interest for research in photodynamic therapy, a protocol for the thermal evaporation of copperphtalocyanine thin films was investigated. When a density of 1.60 g/cm3and a Z-factor of 9.999 are set in the Inficon depostion controller, the tooling factor is somewhat dependent of the desired layer thickness and averages to 5.37. The average tooling factor for cyllindrical zinc stick evaporations is around 1.00 and the average tooling factor for C60 evaporations is around 2.13. There will be deviations up to 25 percent in the layer thickness of any evaporated material.

Although no reliable general tooling factor for any of the materials was realised, the results from this paper can be used as a guideline for future production of thin CuPc, zinc and C60 films.

CuPc thin films were made and sterilized after which their optical properties were characterised and compared with those of a non-sterilized CuPc thin film. Though a few films were mechanically destroyed during the sterilization, the absorption spectra of the films that survived do not seem to be influenced by the sterilization. The absorption peaks are at 330nm, 615nm and 690nm, where the highest peak is about at 330nm and the lowest is at 690nm. The highest peak was about 85 percent and was found in CuPc sample 1, the thickest sample (220nm).

VIII. Outlook

The Z-factors that were used for C60 and CuPc in this paper were based on one other paper that was found in the literature, and they may be wrong, so more research into the Z-factors of C60 and CuPc is recommended. Besides that, it would be interesting to find out more about the effect that a wrong Z-factor in the deposition controller has on the tooling factor for that evaporation.

(14)

References

[Konopa, K. and Goslinsky, T., (2007)] Photodynamic Therapy in Dentistry, Critical reviews in oral biology and medicine, 868(8):694-707.

[Botha, A. F., (2010)] An Investigation into the Research and Development of Nanostructured Photovoltaic Cells

[Mali, S.S., Dalavi, D.S., Bhosale, B.N., Betty, C.A., Chauhan, A.K. and Patil, P.S., (2012)] Electro-optical properties of copper phthalocyanines (CuPc) vacuum deposited thin films, RSC Advances, 2, 2100-2104

[Schwoerer. M. and Wolf, H.C., (2007)] Organic Molecular Solids [Hariharan, P., (2007)] Basics of Interferometry. Elsevier Inc.

(15)

i.

Acknowledgements

I would like to express my gratitude to Dr. Elizabeth von Hauff, my supervisor during this project, for sharing her expertise, giving me advice, providing feedback on a few earlier versions of this thesis and embracing me in the Physics of Energy Group.

I would also like to thank Martin Slaman for teaching me how the lab works and how to use all the equipment, and for keeping the lab running.

Furthermore, a thanks goes out to everyone of the Physics of Energy Group for their support, fruitful discussions and for a great atmosphere.

Last but not least a big thanks goes out to my family, who supported and helped me get through this. A special thanks to Henk Kroeze, who proofread my thesis multiple times and helped me structure the text.

Small molecule thin films for photodynamic anti-bacterial therapy