CZECH TECHNICAL UNIVERSITY IN PRAGUE

Faculty of Biomedical Engineering

Department of Biomedical Technology

Kladno 2012

Interactions of nitrogen–vacancy centers with charged surfaces of functionalized nanodiamond particles for the detection of cellular

processes

Doctoral Thesis

Ph.D. Programme: Biomedical and clinical technology

Branch of study: Biomedical and clinical technology

Supervisor: prof. RNDr. Miloš Nesládek, CSc., HDR

Co-Supervisor: Ing. František Fendrych, Ph.D.

Vladimíra Petráková

- II -

ANNOTATION

This doctoral thesis presents the study of optical properties of nitrogen-vacancy (NV)

centers in variously functionalized fluorescent nanodiamond particles (FND). NV center in

diamond is a deep laying point defect which is the diamond bandgap, containing nitrogen

and a neighboring vacancy. In this thesis we used the very specific properties of these NV

centers engineered in diamond to be used as nanoscale sensors operating in cells. High

biocompatibility of nanodiamond, variable FND size ranging from ~ 5 nm, stable

luminescence from its NV centers and simple carbon chemistry for biomolecule grafting

make FND an attractive alternative to molecular dyes for drug-delivery. This thesis present

a principle of a novel method that can be used for remote monitoring of chemical processes

in biological environment based on color changes from photo-luminescent (PL) NV

centers in FND. It is proposed to drive the NV luminescence chemically, by alternating the

surface chemical potential by interacting atoms and molecules with the diamond surface.

This leads to changes in NV-/NV

0 PL ratio, or in other words the occupation of neutral and

negatively charged NV center, and allows construction of optical chemo-biosensors

operating in cells, with PL visible in classical confocal microscopes. This phenomenon is

demonstrated on single crystal diamond containing engineered NV centers and on oxidized

and hydrogenated ND in liquid physiological buffers for variously sized ND particles.

Hydrogenation of NDs leads to quenching of luminescence related to negatively charged

(NV-) centers and by this way produces color shifts from NV

- (638 nm) to neutral NV

0

(575nm) luminescence. The proposed method is further used for the optical detection of

variously charged polymers in buffer solution, showing the possibility of tracking charged

molecules in biological environment. Further research is focused on the study of the ND

cell uptake mechanism and demonstration of proposed biomolecular detection principles in

vitro.

ANOTACE

Tato doktorská práce prezentuje studii optických vlastností centra dusík-vakance (NV)

v závislosti na povrchové funkcionalizaci. Fluorescence nanodiamantových částic je

založena na centrech NV, což je hluboký bodový defekt v zakázaném pásu diamantu.

- III -

V této práci využíváme velice specifických vlastností těchto center NV vytvořených

v diamantu k vytvoření senzoru používaného v buňkách. Vysoká biokompatibilita,

rozličnost ve velikostech počínajících od 5 nm, stabilní luminiscence z bodových center

v mřížce a povrchová chemie založená na uhlíku dělají z nanodiamantu atraktivní

alternativu k běžně používaným fluorescenčním značkám. Tato disertační práce prezentuje

novou metodu založenou na posunech v luminiscenci center NV, kterou je možné využít

pro bezkontaktní monitorování chemických procesů, které se odehrávají v buněčném

prostředí. V práci je navržena možnost řídit luminiscenci chemicky, kdy díky interakcím

atomů a molekul s diamantovým povrchem dochází ke změnám elektrochemického

potenciálu na povrchu diamantu. Tyto změny vedou k posunům luminiscence center NV a

tím umožňuje sestavení optických biosenzorů, které operují v buněčném prostředů. Tyto

změny je možné pozorovat ve standardním konfokálním mikroskopu. Tento jev je

demonstrován na krystalickém diamantu, který obsahuje centra NV, na hydrogenovaných a

oxidovaných nanodiamantových částicích ve fyziologickém prostředí pro různé velikosti

krystalů. Hydrogenace nanodiamantů vede k zhášení luminiscence z negativně nabitého

centra NV a dochází k posunu luminiscence ke kratším vlnovým délkám, které jsou

charakteristické pro neutrální stav centra. Uvedená metoda je dále použita pro optickou

detekci přítomnosti nabitých molekul, což poukazuje na možnost využití těchto principů

pro optické sledování jednotlivých molekul v biologickém prostředí. Další výzkum je

zaměřen na studium pronikání nanodiamantů do buněk a demonstrace navrhovaných

mechanismů detekce biomolekul in vitro.

- IV -

ACKNOWLEDGEMENTS

It is a great pleasure to have the opportunity to thank all those people who made have

supported me and had their contributions in making this thesis possible.

First of all, I would like to express my warm and sincere thanks to my supervisor, Miloš

Nesládek, for the opportunity to work in his group. His guidance, hard work and never-

ending enthusiasm for new projects has encouraged me during my work. His ideas and

deep knowledge in the field of diamond research have been extremely beneficial for me.

He has given me the freedom and trust during my Ph.D. that motivated me to perform my

best. On the personal basis, I am extremely grateful for his respect to the family and

personal life.

I would like to thank to my co-supervisor, František Fendrych, for his support at the

Institute of Physics, AS CR, and for the great time at the conference in Budapest, 2010. I

am very grateful to Dr. Petr Šitner, the head of the Department of Functional Materials at

the Institute of Physics, AS CR, for the opportunity to work at the Institute.

I own my greatest thanks to my colleague at the Institute of Physics and friend, Andrew

Taylor, for introducing me to the laboratory techniques and equipment related to the

diamond research. His excellent practical skills were of a great value to me especially at

the beginning of my work. I thank to Andy also for his personal support and advices that

guided me through the difficult times.

I would like to thank to my colleagues from the institute of Organic Chemistry and

Biochemistry. My sincere thanks belong to my colleague and friend, Petr Cígler, for his

ideas and long and fruitful discussions. I always found his comments, questions and

suggestions in the manuscripts or reports very challenging and I always felt very relaxed

after answering his concerns. My great thanks also belong to Dr. Miroslav Ledvina for his

help concerning surface chemistry, support and collaboration within the projects. My deep

thanks belong also to other colleagues from the Institute of Organic Chemistry and

Biochemistry, mainly to Ivan Řehoř for the DLS and Zeta potential measurements and his

kind and patient effort during the measurements of the pH dependence of the zeta

- V -

potential, and also to Jan Havlík for his time consuming work on the preparation of

samples for annealing experiments.

My warm thanks also belong to colleagues from the Microbiological Institute, Anna

Fišerová, Veronika Benson and Jan Richter for their help with all biological experiments

and for their kind explanation of the methods and experiments.

My great thanks belong also to colleagues from the Institute of Nuclear Physics, AS CR,

Jan Štursa, Jan Kučka and Jan Ráliš for their work on the irradiation of the particles.

It is my pleasure to acknowledge all my current and previous colleagues in the group,

especially Laco Fekete, Jan Vlček, Ladislav Peksa, Martin Crhán, Jana Vejpravová, Miloš

Stefanović, Marie Vlčková and Michal Gulka for their support and providing a good

atmosphere in the lab. I would like to also thank to my colleagues at the Faculty of

Biomedical Engineering for providing a warm and friendly atmosphere, above all to

profesor Peter Kneppo, the head of the Department of Biomedical Technology, for his

support.

My studies would not go as smoothly without the help of the secretaries, Ivana

Beznosková, Helena Fujanová and Marcela Boháčková, who have always been ready to

help me with all administrative work. I own my great thanks to all of them!

Last, and most importantly, I wish to thank to my family. I cannot imagine being in the

position I am without the never-ending support and encouragement from my parents, Eva

Beránková and Pavel Beránek. I wish to thank them for the educational atmosphere that

they have made for me and my brother. Most of all, I wish to thank them for giving me and

my brother the freedom to choose our way of life and for their absolute support when

making important decisions. Thank you for being here for me and for being who you are.

My brother is a great friend for me and I would like to thank him for his encouragements,

for his believe in me and for many hours of discussions about science, life and fun.

My deep and heartfelt thanks belong to my husband and colleague Vašek Petrák, for being

the one who knows. Thank you for your love, support, and patience during the hardest

- VI -

times. I would also like to thank Vašek for his help with illustrations in this thesis and for

the help with Raman spectroscopy measurements made during my maternity leave.

Last, I would like to thank to my children, Alenka and Jonáš for their love, joy and

optimism they are giving me.

The financial support is gratefully acknowledges and was provided by the CTU grant No.

CTU 10/811700, financial support from the Academy of Sciences of the Czech Republic:

grant KAN200100801, and the European R&D projects (FP7 ITN Grant No. 245122 -

DINAMO and COST MP0901 – LD11078)

- VII -

TABLE OF CONTENT

1. Introduction and aims of the thesis ....................................................................... 1

1.1. Aims of the thesis ......................................................................................................................... 4

2. Diamond as a material for biology and medicine ............................................ 7

2.1. Classification of diamond.......................................................................................................... 7

2.2. Types of nanodiamonds: Origin of synthetic nanodiamonds ..................................... 9

2.2.1. HPHT Nanodiamonds ........................................................................................................ 9

2.2.2. Detonation Nanodiamonds ............................................................................................ 10

2.2.3. CVD Nanodiamonds .......................................................................................................... 11

2.3. Luminescent color centers in diamond ............................................................................. 12

2.3.1. Nitrogen-Vacancy center ................................................................................................ 13

2.3.2. Fabrication of NV centers ............................................................................................... 15

2.4. Biomedical applications .......................................................................................................... 16

2.4.1. Biocompatibility................................................................................................................. 16

2.4.2. Photoluminescence ........................................................................................................... 17

2.4.3. Recent biomedical applications ................................................................................... 18

3. Materials and methods .............................................................................................. 21

3.1. Characterization of morfology and size ............................................................................ 21

3.1.1. Atomic force microscopy ................................................................................................ 21

3.1.2. Dynamic light scattering and zeta potential measurement .............................. 22

3.2. Characterization of structure and optical properties .................................................. 25

3.2.1. Raman spectroscopy ........................................................................................................ 25

3.2.2. Fluorescence spectroscopy ........................................................................................... 26

3.2.3. Confocal microscopy ........................................................................................................ 28

3.3. Characterization of surface modifications ....................................................................... 29

3.3.1. Fourier Transform Infrared Spectroscopy (FTIR) ............................................... 29

3.3.2. Contact angle measurement .......................................................................................... 31

4. Results and discussions ............................................................................................ 33

4.1. Theoretical consideration of surface manipulation with optical defects by charge transfer .................................................................................................................................... 42

4.1.1. Hydrogenated diamond surfaces ................................................................................ 43

4.1.2. Band bending calculations ............................................................................................. 44

4.2. Demonstration of surface functional changes on the optical properties of the model system: implanted single crystal diamond ................................................................. 49

4.3. Formation of variously charged NV centres .................................................................... 52

4.3.1. Introduction of vacancies to the diamond lattice – different irradiation strategies .......................................................................................................................................... 56

4.3.2. Formation of NV centres: annealing study .............................................................. 59

4.4. Chemical control of the NV luminescence in nanodiamond ..................................... 63

4.4.1. Quenching of NV- luminescence on ND particles .................................................. 63

4.4.2. Fluorinated nanodiamonds ........................................................................................... 66

4.4.3. Size dependence of the luminescence of nanodiamonds ................................... 67

4.5. Sensing of charged molecules via NV luminescence .................................................... 71

4.5.1. pH dependence – charge switching ............................................................................ 74

4.6. Study of the cellular uptake ................................................................................................... 80

4.6.1. Oxidized nanodiamonds ................................................................................................. 80

- VIII -

4.6.2. Polymer-coated nanodiamonds ................................................................................... 83

4.7. Nanodiamond transfection system enabling detection of DNA delivery ............. 85

4.7.1. Monitoring of the formation of the ND-transfection system ............................ 86

4.7.2. Detection of the DNA release in cells ......................................................................... 90

4.7.3. Biocompatibility of the ND-transfection system ................................................... 93

4.7.4. Verification of successful transfection ...................................................................... 96

5. Conclusions and Perspectives ................................................................................ 98

6. References ................................................................................................................... 101

Appendix .................................................................................................................................. I

A. summary of author’s publications related to the topic ...................................................... I

- 1 -

1. Introduction and aims of the thesis

Bionanotechnology is a converging scientific field of physics, chemistry, and biology that

uses tools of nanotechnology to study living organisms. The first talk on nanotechnology

was given in the year 1959 by a physicist Richard Feynman, when he first introduced the

concept of manipulations with individual atoms and molecules. But the first major

development in nanotechnology goes to the end of 20th century.

How can we understand bionanotechnology? And what can this field bring us? The prefix

"Bio" stands for Biology which is study of all living organisms such as plants, insects, fish,

birds, elephants or cats; basically all of life on this planet. "Nano" is an abbreviation for

one nanometer (nm), which is one billionth of a meter. To imagine such small size, one can

help with analogies to macroscopic world. The diameter of human hair is approximately 50

000 nm, it means that if there is a nanoparticle with the size of 5 nm, one would need 10

000 of them to cover the diameter of one human hair. 1 nanometer is also how far the

fingernail grows in one second. In the respect to biology, the amazing thing about "nano"

dimension is fact that the impressive diversity of living species in the macro scale is built

from very similar building blocks in the nanoscale. When "zoomed" to the nanodimension,

we find that all living organisms are arranged from DNA, proteins, sugars, lipids and the

diversity is given by the various arrangements that is the result of interactions between

these components. "Technology" gives to this approach the knowledge, principles and

methods from science and engineering. Bionanotechnology can therefore bring us

knowledge about the elementary principles that lead to the specific arrangement and

function of these building blocks, single molecules, in living organisms as well as

discovering new tools for manipulations with them and use them for practical purposes,

such as diagnostic and therapeutic medical applications.

Nanoparticles and nano-devices can help us understand the processes that are happening on

the level of single molecules. But as the dimensions of nanoparticles and molecules are so

small, we have to use specific techniques to visualize them, and to manipulate with them.

Commonly used techniques are high resolution microscopes based on the scanning-probe-

microscopy (SPM) techniques, electron microscopy or others. However the most

commonly used techniques to visualize biomolecules or nanostructures are fluorescent

- 2 -

labels. Labeled molecule is then easily visualized via fluorescence microscope or other

fluorescence reading instrument.

The principles of fluorescence or fluorescence labeling are now widespread in everyday

life. One day my seven years old daughter came to me, showing a fluorescence marker that

can write invisible messages and colorfully shines under the light. I was impressed by her

natural use of the word “fluorescence”. That evening I realized that fluorescence can be

found in many unexpected aspects of life. I found that there is also fluorescent nail polish,

that the 2012 collection of Celvin Klein men’s underpants is called “Sexy Fluorescence”,

or that Fluorescein, the commonly used DNA or protein fluorescent marker, is used to dye

the Chicago River green on St. Patrik’s Day.

The first use of the term fluorescence goes back to 1852, when Sir George G. Stokes

described the fluorescent effects on the mineral fluorite. The first fluorescence microscopes

were developed in the beginning of 20th century, and fluorescence started to be used to

study binding of fluorescent dyes in living cells. [1] The specific fluorescence labeling of

antibodies with fluorescent dyes was developed by Albert Coons in 1940 which brought

new possibilities to visualize specific immunological processes. Nowadays, the highly

specific multiple labeling of individual organelles, macromolecules, or specific DNA

sequences with synthetic and genetically encoded fluorescent probes is used in real time

biological imaging. The amazing fact about fluorescent dyes is that it is not only the ability

of florescent dyes to help us visualize specific structure, but the ability to visualize

functional features of the studied system. Thanks to the discovery of the Green Fluorescent

Protein (GFP), we can incorporate the fluorescence in the genetic code which enables to

track gene expression [2]. The intense research led to discoveries of enormous number of

fluorescent dyes in the whole spectral range that help to overcome some drawbacks such as

high phototoxicity or optical instability.

The main characteristics of fluorescent dyes are maximum excitation and emission

wavelength that corresponds to the maximum in the absorption and emission spectra, the

ability of the dye to absorb photons (called Extinction coefficient), how many photons are

emitted per absorbed photons (Quantum yield), and Fluorescence Lifetime, which is

related to the decay between excitation and emission. Other important characteristics are

phototoxicity (how the dye is toxic to the cell after illumination), photobleaching (the

- 3 -

tendency for quenching of the fluorescence with illumination), or photoblinking. The

choice of the particular dye strongly depends on the characteristics that are required for

chosen application.

According to the company ThermoScientific, the most common fluorescent dyes are

fluorescein and rhodamine (and their derivatives) among the organic dyes, derivatives from

GFP from the biological fluorophores, and quantum dots. Among these, the organic dyes

are the most commonly used. The great advantage of synthetized organic dyes is high

availability at low cost and the great variability in the excitation and emission wavelength.

The disadvantage is low optical stability (bleaching, blinking). The benefit of the

fluorescent proteins is that expression plasmids can be introduced into either bacteria, cells,

organs or whole organisms, to drive expression. However, the size of the fluorescent

protein can change the normal biological function of the cellular protein to which the

fluorophore is fused. Quantum dots are nanoscale-sized (2-50nm) semiconductors that,

when excited, emit fluorescence at a wavelength based on the size of the particle, and

therefore the emitted light shifts from blue to red as the size of the nanocrystal increases.

Since the quantum dot size can be tightly controlled, there is greater specificity for distinct

excitation and emission wavelengths than other fluorescent dyes. Quantum dots are

extremely photostable (they remained fluorescent for 4 months in a in vivo imaging study

[3]. However their use for the biological purposes is limited due to their cytotoxicity [4].

Their disadvantage is also relatively high cost in comparison to other fluorescent dyes.

For the purposes of the study of single molecules and interactions between single

molecules in biological environment, the fluorescent label must be biocompatible with

extreme optical stability. These features are not provided neither by synthetic organic dyes

or biological dyes due to the bleaching and blinking, or by quantum dots for their

cytotoxicity. Although photobleaching is required for some methods (fluorescence

recovery after photobleaching or others), it limits applications such as long term tracking

or tracking of single molecules. The nitrogen-vacancy (NV) center is a deep laying point

defect which is in the diamond bandgap, containing nitrogen and a neighboring vacancy. In

this theses we use very specific properties of these NV centers engineered in diamond to be

used as nanoscale sensors operating in cells. The NV center can provide alternative to

commonly used fluorescence dyes. The advantage of FND is the fact that the fND

fluorescence scales u linearly with number of NV centers. By that way, very highly

- 4 -

luminescent particles can be obtained. This NV luminescent point defect center in diamond

has a few exceptional optical properties that make them attractive in a wide field of

applications ranging from quantum photonics Error! Reference source not found. to

single particle tracking in cell [6][7][8][9]. Luminescence from NV centers is extremely

stable without any photobleaching or photoblinking [10][11][12]. The broad absorption

spectrum allows the use of various excitation sources reaching from blue to yellow. The

maximum of the emission intensity of the center is in the region from 650 to 750 nm with

very high quantum yield (> 0.99) so its fluorescence is well separated from cell

autofluorescence. NV centers can be fabricated in the small nanodiamond crystals of the

size ranging from 5 to 100 nm. No cytotoxicity was observed in nanodiamond particles and

they are believed to be fully biocompatible [13][14]. Nanodiamonds can be also easily

functionalized by the carbon/based surface chemistry [15][16]. Another advantage is that

the luminescence brightness of nanodiamonds, contrary to quantum dots, can be scaled up

linearly by increasing the NV concentration.

1.1. AIMS OF THE THESIS

The features of fluorescent nanodiamonds (NDs) make them a perfect tool for

bionanotechnology, allowing long term optical imaging on the molecular level. However,

is it possible to use nanodiamonds as an optical biosensor? To do so, one would have to

find a way to control the luminescence of NV centers by surface interactions in biological

conditions.

In this respect, the goal of this thesis is to develop a method for optical monitoring of the

biomolecules based on spectral changes in the nitrogen-vacancy (NV) luminescence with

the aim to design an in-cell detection probe. The final goal is to use the probe for targeting

cancer cells, where the drug-delivery event could be optically detected. To achieve it, this

thesis reports on the study of the optical properties of NV centers in the close surface

proximity as a function of surface functionalization and optimize the conditions for the

development of the nanodiamond-based optical probe for biodetection.

The thesis can be divided in to these parts:

- 5 -

1. Theoretical modeling of the effect of surface induced switching of the NV

luminescence based on the changes in the band structure near the surface.

2. Demonstration of the effect in the model system, high purity single crystal diamond

with shallow implanted NV centres.

3. Fabrication of highly luminescent nanodiamond particles: optimization of the

irradiation and annealing conditions

4. Study of the optical properties of fabricated NV centres in nanodiamond particles

as a function of surface termination.

5. Demonstration of the optical detection of the charged biomolecules via changes in

the NV luminescence.

6. Study of the nanodiamond cellular uptake.

7. Demonstration of the biomolecular detection principles in vitro.

This work describes how the surface chemistry effects can make the ND bulk

luminescence sensitive to chemical processes ongoing at the ND surface, aiming at using

ND for monitoring the chemical environment such as surface charges/pH, cellular

DNA/RNA hybridization, interaction with cell receptors, etc. The developed method is

based on the control of electronic chemical potential in a close proximity of ND surfaces,

influencing the occupation of the luminescent NV centers, that exist in neutral (NV0) or

negative charge states (NV-)[17][18]

with different PL properties. Therefor it is possible to

control emissions wavelength, i.e. 575 nm for NV0 and 636 nm for NV

-. This method

explained below yields advantages over recently demonstrated ND FRET [19] (Foerster

Resonance Energy Transfer) scaling as 1/r6, allowing to induce luminescence shifts

originating at energy transfer between ND donor and acceptor dye of size ~ 5 nm. In our

case, due to 1/r2 Coulombic interactions, we can observe chemically driven luminescence

shifts up to larger distances ~ 20 nm in depth, which is exactly calculated by modeling.

This allows construction of optical chemo-biosensors with sizes of about 40-50 nm easily

visible in classical confocal microscopes. This can be especially interesting for drug

delivery research and for monitoring chemical interactions occurring in cells based on

covalent or non-covalent interactions with charged molecules such as DNA and by various

surface terminations.

The thesis is divided in 4 main sections. The goal of the first part of the thesis: Chapter 2.

Diamond as a material for biology and medicine relates the thesis to the state-of-the-art

- 6 -

biomedical applications of nanodiamond. This part gives a brief review on the properties of

nanodiamond, gives the relation of these properties to the biomedical applications, and

presents recent biomedical applications of nanodiamond. Chapter 3. Materials and

methods describes the theoretical principle of methods that were used in this thesis and

relates their use to nanodiamond characteristics. The main part of the thesis, Chapter 4.

Results and discussions, presents the key results of the research related to the topic of the

thesis: study of the optical properties of NV centers in the close surface proximity as a

function of surface functionalization that shows novel principles of biomolecular detection

via changes in the luminescence spectra. This chapter is further divided to seven sub-

chapters each of them representing an independent problem that lead to the final goal of

the thesis. The brief summary of the results that are represented in each chapter is given at

the beginning of the section 4 and relates these results to the final aim. At the end,

Chapter 5 gives the summary of the work, states the limitations of discovered methods,

and discusses perspectives of this work. Author’s publications related to the topic are listed

in Appendix.

- 7 -

2. Diamond as a material for biology and medicine

Diamond is due to its remarkable physical properties attractive material with exceptionally

wide range of possible applications. The amazing variety of optical (diamond crystals exist

in variety of colors) and electronic properties (diamond acts as an electric insulator but can

exhibit semiconductor or metallic behavior when doped) is given mainly by point defects

and impurities in the diamond lattice. Since defects play such an important role in the

properties of diamond, the study of the origin of the defects, their properties, possible

artificial fabrication or their commercial use is the main part of the diamond-related

research.

Point defects in diamond play important role in many biomedical applications of diamond.

For example strong red luminescence of nanodiamonds origins from nitrogen-vacancy

defect center that can be easily fabricated in synthetic nanodiamonds. Therefore, this

chapter concentrates on the description of the physical and optical properties of diamond

that are related to the defect centers

The first part of this chapter describes basic physical and optical properties of diamond that

are related to defects incorporated in the diamond lattice. The physical classification of

diamond is stated and nanodiamonds from three types of diamond synthesis are briefly

introduced. The second part discusses properties of diamond from the perspective of

biomedical applications with the focus on the biocompatibility of nanodiamonds and their

photoluminescence.

2.1. CLASSIFICATION OF DIAMOND

The physical classification of diamond is based on the optical absorption of nitrogen, boron

and hydrogen-related defects and paramagnetic absorption of single substitutional nitrogen.

The classification system used now was defined by Dyer et al. (1965a) [20] and is shown

in Figure 1.

- 8 -

Figure 1: Physical classification of diamond

As Type I are classified diamonds where the nitrogen defects are dominant. About 98% of

natural diamonds contain nitrogen with concentration detectable in optical absorption. In

type I natural diamonds, the nitrogen is not distributed uniformly, forming areas of

increased concentration, or can gather in cluster of even about 10 atoms [21] (Berman et al.

1975).

Type I diamonds are divided in two groups, Ia and Ib, where in Ia diamond the nitrogen is

found in the form of aggregates (mainly A and B aggregate - see section 2.3 for details)

and in Ib diamond the nitrogen is predominantly in the form of single substitutional center.

It is believed, that for natural diamond, the nitrogen is originally incorporated in the form

of insolated substitutional in the lattice (Ib) but after long periods at geological

environment, the nitrogen forms aggregates characteristic of the type Ia diamond. Natural

diamonds type Ib are very rare (1 of 1000) and have attractive yellow color in a gem

diamond, known as "canary yellow". This color is given by an absorption in the visible

region, starting at approximately 550 nm with an increase towards shorter wavelengths

[20][22]. Nitrogen containing diamonds produced by high-pressure, high-temperature

synthesis (HPHT) are mostly type Ib with typical nitrogen concentration between 100 and

200 ppm.

- 9 -

Type Ia is further divided in type IaA diamonds where nitrogen is present as nearest-

neighbor substitutional pairs (A aggregates) and in type IaB diamonds where nitrogen is

present as B aggregate (4 substitutional nitrogen atoms symmetrically surrounding a

vacancy).

Type II comprises diamonds that do not contain nitrogen detectable by optical and

paramagnetic absorption. Type IIa diamonds are very pure with defects concentration

below 1018

cm-3

. Type IIb diamonds contain boron as a major impurity, giving diamond an

attractive blue color. These diamonds are p-type semiconductors.

2.2. TYPES OF NANODIAMONDS: ORIGIN OF SYNTHETIC NANODIAMONDS

Natural diamonds in a rough form, that is, uncut and unpolished are known since antiquity.

They were first reported in India 2700 years ago. The popularity of diamond as a gem

stone has begun to rice with development of faceting and polishing in the 14th and 15th

century. After Smithson Tennant in 1797 discovered that diamond is consists from carbon,6

there were many attempts to produce diamond from other (cheaper) allotropes of carbon.

Despite painstaking effort of many scientists, none reproducible results were reported

before 1950s. The first reproducible results were achieved in Sweden in 1953 and then by

General Electric in the USA in 1955 using the high-pressure high-temperature method.

Since then diamonds have found their place as a material for industrial applications, and

production of synthetic diamonds well exceeds the amount of mined diamonds [23].

Nowadays, there are three methods of industrial production of synthetic diamonds: high-

pressure high-temperature (HPHT) growth technique, detonation synthesis and chemical

vapor deposition (CVD) method.

2.2.1. HPHT Nanodiamonds

Synthesis of “man-made diamond” by method employing high-pressure high-temperature

(HPHT) was first described in 1955 by researchers from General Electric Company. Bundy

et al. employed the high-pressure high-temperature transformation of graphite [24]. This

method uses hydraulic press generating a pressure in range from 5 to 11 GPa while the

temperature is maintained in range from 1200 to 2200 °C. The addition of a metal catalyst,

- 10 -

typically nickel or chromium accelerates the process [25]. The crystals grown by the

HPHT technique are used for a wide range of industrial processes, including cutting and

machining mechanical components and for polishing and grinding of optics. A

shortcoming of the HPHT method is that it produces diamond only in the form of single

crystals ranging in size from tens of nanometers to about 1 centimeter. This limits the

range of possible applications. On the other hand, HPHT diamonds are of high quality

(larger HPHT diamonds can be used as gem stones) without any non-diamond carbon

structures (DLC, graphite, amorphous carbon) as seen in Figure 2. Most commonly, the

HPHT diamonds are type Ib with 100-200 ppm of single substitutional nitrogen atoms.

Figure 2: SEM image of HPHT ND particle (Source: Microdiamant)

2.2.2. Detonation Nanodiamonds

In 1963 a group of scientists in the Soviet Union discovered single crystals of diamond

particles in soot produced by detonating an oxygen-deficient TNT/hexogen composition in

an inert media. The average size of so called “detonation diamonds” has been

determined to be 4-5 nm [26]. For several reasons, including the security measures in the

USSR and lack of interest in nanotechnology at that time, the application of this type of

nanodiamond remained under-exploited until very recently. However, there has been a

rapid increase of interest in the use of the ND particles in a variety of applications in recent

years. The applications of the detonation ND are as diverse as quantum computing [27],

catalysis [28], or biomedical applications [29]. The other two examples of ND applications

are formation of hard coatings and composites, and polishing and seeding of substrates for

a CVD diamond growth.

Detonation nanodiamonds, produced by detonation of explosives under an oxygen-

deficient atmosphere have the smallest particle sizes among particulate synthetic

- 11 -

diamonds, mainly in the range of 4 to 5 nm [30]. However, the conditions in the reaction

chamber lead to a significant agglomeration of the material. The particles linked by the

usual electrostatic interactions and also via covalent bonds between surface functional

groups as well as by soot structures surrounding each primary particle (Figure 3).

Therefore, a thorough purification is needed to obtain graphite-free diamond. Krueger et al.

reported on a structure model for detonation diamond that explains the unusually strong

agglomeration and developed methods for an efficient deagglomeration and purification

[26].

Figure 3: A) HRTEM image of detonation diamond. The particles are surrounded by graphitic and soot-like material. B) Structure model of the diamond agglomerates. Picture

reprinted from [31].

2.2.3. CVD Nanodiamonds

Diamond growth by chemical vapor deposition (CVD), e.g. growth from low pressure

gases was first reported in 1950s. However, the rate of growth in these early experiments

was very low. This was caused by deposition of graphite. This leads to mixed sp3/sp

2

phases.[23] The breakthrough in CVD of diamond was discovery of Angus et al. who

found that the presence of atomic hydrogen in the gas mixtures leads to preferential etching

of graphite. This means that diamond growth is favored.[32] CVD growth of diamond has

become an active and extensive area of research beginning in the 1980s, leading to the

industrial manufacture and use of diamond materials in many applications

Typical source of carbon for diamond growth is methane; however other carbon containing

gases can be used. CVD growth of diamond requires means of activation of precursor

molecules. This is typically be done by thermal methods (e.g. hot filament) or microwaves

[33]. Growth of diamond normally requires that the precursor gas to be diluted in

- 12 -

hydrogen. Typical concentration of methane in the gas mixture is 0.5 to 2%. It is also

possible to induce dopant congaing gas (such as trimethylboron or diboran for doping with

boron) to the gas mixture.

The CVD method requires a substrate preparation prior to deposition to generate diamond

nuclei on the surface of substrates that can be enlarged by a subsequent growth. The

growth pretreatment processes, like mechanical abrading, ultrasonic seeding, or ion

bombardment, are used to prepare such nucleation centers on the surface of the substrate

[23]. The CVD technology allows deposition of thin polycrystalline diamond films on

areas up to 100 cm in diameter. Various deposition conditions yield films of various

parameters. Addition of other gasses (e.q. trimethyl boron) to the gas mixture leads to

incorporation of defects (e.g. boron) to the diamond (doping of diamond) that allows the

precise tuning of properties of such diamonds (e.g. electrical conductivity). CVD films can

be later milled to gain ND particles of various size and structure.

2.3. LUMINESCENT COLOR CENTERS IN DIAMOND

There are over one hundred optically active centers in diamond. These optical centers

make diamonds so attractive in their natural beauty by giving them various color shades.

The usefulness of diamond color centers is mostly given by their unique optical stability

that is combined to other outstanding physical properties of diamond. Diamond has the

widest optical transparency band of all known solids, being transparent in the ultraviolet,

visible and infrared spectral regions (except for the intrinsic vibrational absorption band

between 2.5 and 7 μm). This gives diamond the ability to provide optically active point

defect by their electronic and vibrational transitions. [34] The large bandgap energy is

favorable to luminescence of point defect that requires both the ground and excited

electronic states lie within the bandgap.

The majority of optically active point defects in diamond are related to nitrogen (as

nitrogen is the prominent impurity in diamond in general), but there are other centers, for

example silicon or nickel related centers that can be interesting for bio-labeling

applications and, in combination with nitrogen related centers, can serve as two color

labels. However, the most intense and the most important optical centers in diamond are

nitrogen-related centers. As mentioned above, nitrogen is a defect that exists either as a

- 13 -

single substitutional impurity or in aggregated form. The single substitutional nitrogen

(known as the C center) has the infrared vibration at 1344 cm-1

and it is a deep electron

donor, 1.7 eV below the conduction band minimum. Most common nitrogen aggregates are

pairs of neighboring substitutional atoms (the A aggregates) and groups of four nitrogen

atoms around a vacancy (the B aggregate) [35] that exist in natural diamonds. The

structure of these aggregates is shown in Figure 4. The most intensively studied and the

most interesting center from the aspect of applications is the nitrogen-vacancy center that

occurs in high energy irradiated Ib diamond.

Figure 4: Schematic structure of A aggregates, B aggregates and C aggregates

2.3.1. Nitrogen-Vacancy center

The nitrogen-vacancy (NV) center consists of a nearest-neighbor pair of a substitutional

nitrogen atom and a lattice vacancy (Figure 5). NV center exist in two charge state, neutral

(NV0) and negative (NV

-) with different optical and magnetic properties. The overall

structure of the center is an axial C3V symmetry. A nitrogen atom in the center has three

valence electrons bond to the carbon atoms and two unbounded valence electrons form a

lone pair. The two of three electrons from the vacancy form an electron pair with a

nitrogen lone pair resulting in one remaining unpaired electron. To form the negatively

charged NV- center, the additional electron is required.

- 14 -

Figure 5: The schematic structure and energetic levels of NV center. Taken from [35]

The NV center has a wide absorption spectra in the visible range that allows using of

various different lasers (from blue to yellow) to excite the emission of the center. NV0 and

NV- differ in the absorption and emission spectra. NV

- center emits bright red with zero

phonon line (ZPL) at 637 nm that is followed by broad side band luminescence with the

highest intensity around 700 nm, oppose to an orange luminescence from NV0 center with

ZPL around 575 nm also followed by broad side bands. Typical emission spectra of NV-

and NV0 centers is shown in Figure 6. The emission of NV centers is very stable without

any blinking or bleaching observed at room temperature.

Figure 6: Typical spectra of NV0 (blue) and NV- centre. Taken from [36]

The observed photochronism in NV centres [11][12] (one center can exist in negative and

neutral charge state) gave rise to new possible applications, making for example two color

- 15 -

marker from one single NV center. The control of the charge state of NV centers in the

biological environment could open new bio-detection possibilities and is a main topic of

this doctoral study.

2.3.2. Fabrication of NV centers

NV centers can be produced from the single substitutional nitrogen atoms (C centers) by

trapping the vacancy. This process can be accomplished either by nitrogen implantation to

IIa diamond (high purity) or by high energy beam (electrons, protons, alfa particles,

neutrons, etc.) irradiation of Ib type diamond (containing C centers) and subsequent

annealing in the temperature higher than 650 °C [35].

By nitrogen implantation, one can control the depth of the NV centres by tuning the

implantation energy. Vacancies are in this case created during the implantation process. At

the temperature around 650 °C, vacancies start to migrate in the lattice and are efficiently

trapped by nitrogen atoms resulting in the formation of stable NV centers (schematically

shown in Figure 7).

Thermal annealing at high temperatures (e.g., 800 °C) preferentially forms NV- centers in

the diamond lattice [37] [17]

Figure 7: Fabrication of NV centers in IIa diamond by nitrogen implantation

Production of NV centres in type Ib HPHT nanodiamond particles requires high energy

(e.g. 3MeV) irradiation with protons [13] or electrons [38][39] to create vacancies (as

ilustrated in Figure 8). An alternative method of creating vacancies in nanodiamond

involves the use of a much lower energy (e.g. 40 keV) beam made of He+ [8].

- 16 -

Figure 8: Fabrication of NV centers in Ib diamond by ion implantation

2.4. BIOMEDICAL APPLICATIONS

Biomedical applications of any material require specific features such as low toxicity,

biocompatibility, inertness or others. This chapter describes the properties of nanodiamond

that are essential for their use in biology and medicine and discuss other properties in the

respect to biomedical applications in general and to the applications related to this thesis.

The last part of this chapter lists the recent applications of nanodiamond in the field of

biology and medicine.

2.4.1. Biocompatibility

The possible use of nanoparticles for diagnostic and therapeutic applications has always

been very attractive. The increased interest about their usage with the increased presence of

nanomaterials in commercial product has raised concerns about its potential environmental

pollution and toxicity effects [40]. Oppose to metal or semiconductor based nanomaterials

(e.g. quantum dots), carbon based nanostructures were expected to be more biocompatible

and less toxic. However, carbon based nanoparticles themselves differ in biocompatibility.

For example, pronounced cytotoxicity has been found for single-walled nanotubes in

alveolar macrophages [41] but ND particles were found to be non-toxic to lung cells or

neural cells [42][37]. The biocompatibility of carbon based materials can be further

enhanced by suitable surface functionalization. ND is considered to have highest

biocompatibility of all carbon based nanomaterials. The long term examination of cells

with the content of ND showed low levels of reactive oxygen species and additionally,

cells grown on substrate coated with ND exhibited sustained viability and function [42].

- 17 -

The key property of ND is in this case its pure sp3 carbon composition, which retains

biologically inert.

Biocompatibility of nanodiamonds was studied under various conditions and so far

nanodiamonds are considered to be non-toxic and highly biocompatible, even though no

rigorous clinical trials have been done yet and the questions of expulsion of nanodiamonds

or their possible accumulation in organs remain to be answered.

There are also other factors than excellent biocompatibility that make diamond favorable

candidate for biomedical applications. First, as an allotrope of carbon, diamond can be

synthesized both physically and chemically [43][44]. Second, diamond has high rigidity

and low chemical reactivity. Third, surface of diamond can be easily terminated with

functional groups, making them suitable platforms for further attachments of various

biomolecules [31][45]. And fourth, as mentioned in previous sections, diamond

nanoparticles can contain highly fluorescent and optically stable color centers that allow

long-term observation of these particles individually in live cells [7][8].

2.4.2. Photoluminescence

Fluorescence markers are essential for biomedical imaging of living cells or cell

structures. Coventional fluorescent dyes and engineered fluorescent proteins have been,

due to their small size, used for these applications, but their main disadvantage is a poor

photostability that does not allow long term monitoring of cell structures with high

sensitivity. On the other hand, quantum dots have excellent optical properties such as

broad excitation spectra with very narrow emission, exceptionally high brightness and high

photobleaching threshold, but the presence of heavy metal ions, such as Cd+ (human

carcinogen), that results in high cytotoxicity, restricts their long term in vivo applications.

For these reasons, ND containing photoluminescence centers represent an attractive

alternative to these markers.

NV related luminescence is especially interesting for biological imaging. When imaging a

single molecule in cell, the process is usually accompanied by the high fluorescent

backgrounds that origin from photoexcitation of endogenous components, including

- 18 -

flavinc, nicotinamide adenine nucleotide and collagens [37]. These components absorb and

emit light in the wavelength range of 300 to 600 nm. The NV related luminescence

overcomes these limitations, the emission is well separated from cell autofluorescence (See

emission spectra in Figure 6), thereby is ideally suited for biological imaging aplications.

Another center that can be possibly suitable for bioimaging is the H3 (or N-V-N) center

that consists of two nitrogen atoms next to the vacancy. However, this center emits at

530 nm and is produced by irradiation of natural diamond powders with type Ia

characteristics [46].

2.4.3. Recent biomedical applications

ND as a fluorescent label

The development of nanodiamond optical labels has a large potential in many areas of

biomedicine and biotechnology. There are two approaches for using ND as a fluorescent

label: 1) to use luminescent color centers that are incorporated in the ND itself and 2)

additional fluorescent labeling with fluorophore tags. ND can be due to the stable

luminescence and nontoxicity used as a long term label that allows tracking of single

molecules or cell organels [37][9].

ND as a New Carbon-based Enterosorbent

Carbon and clay-containing adsorbents are commonly used in medical and

pharmacological industries to bind ingested toxins, which can be harmful for both animals

and humans [42]. Due to their properties, NDs were considered as potential enterosorbents

Enterosorbents are compounds that bind toxins in the gastrointestinal (GI) tract thereby

neutralizing the effect of these toxins. NDs were used for binding a group of mycotoxins

called aflotoxins [47].

Conjugated NDs in Ballistic Delivery

The usefulness of ND as a generalsolid phase support was demonstrated in experiments

using the biolistic particle delivery system for the delivery of bio-active molecules to yeast,

fall armyworms, cacti, and bananas [42]. Grichko et al. [48] reported the use of ND-

assisted ballistic delivery of, an ethylene antagonist, diphencyprone, to prevent the ripening

- 19 -

of bananas. ND showed many advantages, such as less toxicity, non-explosivity, water

solulibility and cost efficiency, compared to routinely used methods

ND for drug delivery systems

Clustered detonation NDs (2-8 nm), as active hydrogels, were used for chemotherapeutic

delivery. Huang, Ho et al. [49] confirmed that the ND hydrogel is nontoxic and

demonstrated that doxorubicin hydrochloride (DOX) , a drug used in cancer chemotherapy,

preserves its efficacy even after conjugation to ND. The drug could be efficiently carried

by the ND hydrogel into living cells such as human colorectal carcinoma cells and

additionally, a process of slow or sustained release of the drug was proven by DOX-

induced cell death measurement. The schematics of the principle of nanodiamond mediated

drug delivery is shown in Figure 9.

Figure 9: Schematics of the ability of ND for drug delivery. Reprinted from [50]

The fact, that ND were closely bonded to the ND hydrogel with the slow release after

carcinoma cells were targeted additionally highly increased the viability of rats used for in

vivo studies, showing the ability to target specific locations in the body to increase

treatment specificity, the reduction of the overall quantity of drug used at the active dosing

site which can significantly reduce patient toxicity and impact upon the immune system,

and the potential to reduce the concentration of the drug at healthy and unaffected sites,

resulting in fewer side effects that can complicate the efficacy of treatment. This result

indicates that ND is a therapeutically significant nanomaterial.

- 20 -

Bioanalytical applications

ND were also recently used for bioanalytical application, either nonspecific (absorption of

recombinant apoobelin and luciferase proteins that were later separated by

chromatographic methods [51]) or specific by further, more complex functionalization

with various proteins that resulted in novel probes for cellular targeting [52]. NDs can also

be linked to form ND-DNA probes for collection of the complementary DNA target, which

may be used in the future to construct ND-based DNA microarrays [42].

Other applications include the use of lysozyme conjugated NDs for antibacterial use [53],

the use of conjugated NDs for immunogenic effects [42] and for specific cell receptor

targeting [42].

This work concentrates on the sensing applications of nanodiamonds that can be targeted

for example in the cancer cells and on the development of drug delivery systems where the

drug release could be optically monitored.

- 21 -

3. Materials and methods

3.1. CHARACTERIZATION OF MORFOLOGY AND SIZE

3.1.1. Atomic force microscopy

The atomic force microscopy (AFM) is a high resolution scanning probe microscopy

(SPM) technique. The AFM measures very small forces (less than 1 nN) present between

AFM tip surface and a sample surface.Error! Reference source not found. These small

forces are measured by a measuring motion of a very flexible cantilever. Unlike a scanning

tunneling microscopy (STM) the AFM is capable of investigating surfaces of both

conductors and insulators on the atomic scale. The cantilever of the AFM has a sharp,

force-sensing tip at the end. The end of the tip interacts with the surface. As interaction

force between the cantilever tip and the surface varies, deflections are produced in the

cantilever. The deflection of the cantilever is precisely measured by the laser. (Figure 10)

The forces between atoms may be described by the Lennard-Jones potential

( ) [(

)

(

)

] (1)

The energy of interaction has a minimum value at equilibrium separation r0, and

separation is σ at ( ) . At separations greater than r0, the potential is dominated by

long-range attractive interactions that decay as a function of 1/r6, while at shorter distances

the interaction becomes increasingly dominated by short-range, repulsive interactions that

vary with 1/r12

. These intercalations are quantum-mechanical in nature and arise from the

interpenetration of the electron shells of the interacting atoms at small separations Error!

Reference source not found..

- 22 -

Figure 10 The schematic principle of AFM.

Three basic modes of AFM operation are recognized, depending on which forces are

acting on the tip during scanning. When the tip approaches the sample form further away,

at certain distance an attractive force is sensed. This utilizes a technique called non-contact

AFM. If the tip operates in a region where attractive forces predominate, the mode of

operation is called the contact mode. A technique, in which the tip is operated while

entering the repulsive, as well as attractive region, is called the tapping mode or semi-

contact mode. Each mode has advantages and disadvantages. The contact mode is the

simplest technique but is sensitive to noise, and the tip can easily be damaged on hard

surfaces, or the tip can damage soft samples. The non-contact mode uses high frequencies

and measures differences in frequency or amplitude which leads to the better signal to

noise ratio, but problems may arise when the tip sticks to the surface adsorbed liquid layer

[56]. The intermittent-contact mode AFM solves the sticking problem and can even be

applied in liquid environments.

In this study we use Integra (NTMDT, Russia) AFM. Our setup allowed measurements

on a 5 x 5 m or 125 x 125 m area. Silicon tips Etalon (NTMDT, Russia) with curvatures

of 10 nm were used in this study.

3.1.2. Dynamic light scattering and zeta potential measurement

Dynamic Light scattering (DLS)

Dynamic Light Scattering is a frequently used method to determine the size of particles

emulsions in a liquid. When a laser beam passes through a colloidal dispersion, the

- 23 -

particles scatter some of the light in all directions. When the particles are very small

compared with the wavelength of the light, the intensity of the scattered light is uniform in

all directions (Rayleigh scattering). For larger particles (above approximately 250 nm in

diameter) the intensity is angle dependent (Mie scattering) Error! Reference source not

found.. It is possible to observe time-dependent fluctuations in the scattered intensity using

a suitable detector such as a photomultiplier capable of operating in a photon counting

mode. These fluctuations are caused by the Brownian motion, and the distance between

them is therefore constantly varying. The intensity of the scattered light fluctuates at a rate

that is dependent upon the size of the particles as smaller particles move more rapidly.

Constructive and destructive interference of the light scattered by the neighboring particles

within the illuminated area gives rise to the intensity fluctuation. These intensity

fluctuations contain information about this Brownian motion. It is possible to calculate the

diameter of the particles via the Stokes Einstein equation, by analysis of the time

dependence, if the viscosity of the medium is known [57].

The diameter that is measured in DLS is called the hydrodynamic diameter (Figure 11) and

refers to the diffusion of the particle in a fluid. The hydrodynamic diameter is therefore the

diameter of a sphere that has the same translational diffusion coefficient as the particle

being measured. This means that the size measured by DLS can be larger that measured by

a SPM method where the particle is removed from its native environment.

Figure 11: Schematics of the hydrodynamic diameter

- 24 -

Zeta potential measurement

The development of a charge on the particle surface affects the distribution of ions in the

surrounding region. This results in increased concentration of ions of the opposite charge

to that of the particles close to the surface. As a consequence, an electrical double layer

exists around each particle. The liquid layer surrounding the particle has two parts: (1) the

inner region, called the Stern layer, where the ions are strongly bound and (2) the outer,

diffuse, region where they are less firmly attached. Within the diffuse layer there is a

distinct boundary inside which the ions and particles form a stable entity. When the

particle moves, ions within the boundary move with it. Any ions outside the boundary do

not travel with the particle. The potential that exists at this boundary is known as the Zeta-

potential (sometimes written as -potential). Position of the Stern and Zeta potential is

illustrated in Figure 12.

Figure 12: Schematic illustration of potential of a nanoparticle. Figure based on [58].

The zeta potential provides information about a potential stability of the colloidal system.

If all the particles in the suspension have a large absolute value of the zeta potential, they

tend to repel each other. However, if the particles have low zeta potential values then there

is no force to prevent the particles coming together and aggregating. It is common

knowledge that an unstable suspension has an absolute value of the zeta potential lower

- 25 -

than 30 mV. The particles with absolute value of the zeta potentials above 30 mV are

considered to be stable.

As a consequence of electrical charges on the surface, the particles will exhibit certain

effects under the influence of the electric field. The measurement of the zeta potential is

based on these effects. These effects, referred to as electrokinetic effects, are:

Electrophoresis: The movement of a charged particle relative to the liquid in which it is

suspended under the influence of an electric field.

Electroosmosis: The movement of a liquid relative to a stationary charged surface under

the influence of an electric field.

Streaming potential: The electric field generated when a liquid is forced to flow past a

stationary charged surface.

Sedimentation potential: The electric field generated when charged particles move relative

to a stationary liquid.

3.2. CHARACTERIZATION OF STRUCTURE AND OPTICAL PROPERTIES

3.2.1. Raman spectroscopy

The Raman scattering was first observed by Sir C.V. Raman in 1928. The Raman effect

relies upon the polarization of the electron cloud in chemical bonds by the incident

electromagnetic radiation, resulting in light-induced bond dipole moments. Thus it is the

polarizability, which is the important molecular parameter in determining Raman

intensities.

When light scatters from an electron elastically, the energy of the scattered photon is equal

to the energy of the incident photon. The Raman is an inelastic effect that invokes an

absorption process in which the incident radiation is absorbed by an electronic ground state

of the molecule, leading to its excitation to virtual energy state and followed by emission

back to its first excited vibrational state. The energy difference between the incident and

emitted radiation is thus equal to one quantum of vibrational energy. Emitted photons are

called Stokes photons (Figure 13). An alternative situation is when the molecular vibration

is already in its vibrational state described by v = 1 and, after re-emission, photons of

- 26 -

higher energy than the exciting energy are radiated leading to the transition to the ground

state. This is known as the anti-Stokes process Error! Reference source not found..

Figure 13: Energy diagrams for the different possibilities for light scattering. (left) Elastic, Rayleigh scattering; (middle) Stokes scattering; (right) anti-Stokes scattering. Picture

modified form Error! Reference source not found..

The Raman spectra that are detected yield information on the optical transitions that lead to

bond polarization and are associated with movements such as its vibration, rotation,

stretching etc. By this way the Raman analysis can deliver valuable information on the

chemical nature and bonding of examined materials.

Raman spectroscopy is a very important tool in the investigation of carbon materials. Due

to the nature of the carbon bonding, the Raman spectrum is particularly sensitive to the

microstructure of the carbon. Therefore, the Raman spectroscopy is the leading method to

investigate the structure and quality of diamond, giving us quantitative information about

the content of sp3 carbon in the crystalline form (diamond Raman peak at 1332 cm-1

),

amorphous sp3 carbon (D-band), graphite (G-band), diamond like structures, as well as

content of dopants and many other characteristics.

If not stated otherwise, the Raman spectra shown in this thesis were taken on a Renishaw

InVia Raman Microscope, excitation wavelength 325 nm, at room temperature. Laser

power was 5 mW; spot focus with an X50 objective, 50 µm slit, and grating was 2400

lines/mm.

- 27 -

3.2.2. Fluorescence spectroscopy

Fluorescence spectroscopy is a spectroscopic method of characterization the fluorescence

of a sample. It is complementary with the absorption spectroscopy, but oppose to the

absorption spectroscopy, fluorescence spectroscopy cannot be used as a quantitative

technique.

The principle of fluorescence and absorption spectroscopy is explained by Jablonski

diagram (Figure 14). The species is first excited by the absorption of the photon from its

ground state to one of the vibrational excited electronic states. The process of light

absorption is extremely rapid, in the order of one femtosecond. The excited molecule then

loses vibrational energy until it reaches the lowest vibrational state of the excited electronic

state via vibrational relaxation and internal conversion. This radiationless process takes

place in about one picosecond (1 ps = 10-12

s).

Figure 14: Jablonski diagram illustrates the electronic states of a molecule and the transitions between them. The electronic states are arranged vertically by energy. They are grouped horizontally by spin multiplicity. In the left part of the diagram three singlet

states with anti-parallel spins are shown: the singlet ground state (S0) and two higher singlet excited states (S1 and S2). Singlet states are diamagnetic, as they do not interact with an external magnetic field. The triplet state (T1) is the electronic state with parallel

spins. A molecule in the triplet state interacts with an external magnetic field. Transitions between electronic states of the same spin multiplicity are allowed. Transitions between

- 28 -

states with different spin multiplicity are formally forbidden, but may occur owing to a process called spin-orbit coupling. This transition is called intersystem crossing.

Superimposed on these electronic states are the vibrational states, which are of much smaller energy Error! Reference source not found..

The lowest vibrational level of the excited state is therefore the starting point for

fluorescence emission to the ground state S0, non-radiative decay to S0 (internal

conversion), and transition to the lowest triplet state (intersystem crossing) as stated in the

Jablonski diagram. Fluorescence takes place on the nanosecond timescale (1 ns = 10-9

s)

and. It is clear from the Jablonski diagram that fluorescence always originates from the

same level, no matter of which electronic energy level is excited. It is for this reason that

the fluorescence spectrum is shifted to lower energy than the corresponding absorption

spectrum (Stokes shift). Basicaly speaking, the abosrption spectrum is the mirror to the

fluorescence spectrum with the difference, that the vibrational fine structure in a

fluorescence spectrum reports about vibrations in the ground state, and vibronic bands in

an absorption spectrum provides information on vibrations in higher electronic excited

states. The transition between the lowest ground vibrational energy level and the lowest

excited vibrational energy level is in the absorption/emission spectrum called a zero

phonon line.

The triplet state depicted in Figure 14 is related to phosphorescence. Once the molecule

has reached this state, it will reside for a very long time there (from microseconds to

seconds) before it will decay to the ground state. This is due to the spin-forbidden

transitions involved in the (excited) singlet-triplet and triplet-singlet (ground state)

transitions.

The measurement of an absorption spectrum is based on the Lambert-Beer law, and shows

the ability of the investigated sample to absorb light at different wavelengths. A

fluorescence spectrum represents the intensity of the fluorescence light emitted by the

sample as a function of emission wavelength. As fluorescence transitions start in most

cases from the lowest vibrational level of the first electronic excited state, characterises the

energetic structure of the electronic ground state.

- 29 -

3.2.3. Confocal microscopy

Unlike traditional optical microscopes, where a large area of sample is illuminated and

image is detected from a relative large volume, in confocal microscope the only light

detected is originated (reflected, transmitted or emitted) from a small sample illumination

volume at the focus of microscope objective. In a confocal microscope, the light coming

back from the illumination volume is focused down to another diffraction-limited spot,

which is surrounded by a narrow pinhole. The pinhole filters out the defocused light

originating from parts of the sample that are not coming from the illumination volume.

Because it is positioned at a point conjugate to the focal point in the sample, the pinhole is

said to be confocal to it, and the pinhole allows only the light from the illumination volume

to reach a detector (schematically in Figure 15).

Figure 15: A schematic of a typical confocal microscope. Light from a laser beam is reflected by a dichroic beam splitter and focused onto a spot on the sample. The optic axis

is along z direction. Light from the sample, at a lower wavelength, comes back up from the illumination volume via the objective, passes through the dichroic beam splitter, and is

focused onto a point, surrounded by a pinhole. The detected light then passes to the detector. The pinhole blocks all light originating from points not at the focus of the

microscope objective, so that only the light from the illumination volume is detected. Picture modified from Error! Reference source not found.

Laser is a common light source in the confocal microscopy that is through the objective

lens focused into a small (ideally diffraction limited) spot within the specimen (which

might be fluorescent – fluorescence confocal microscopy). The high resolution of the

confocal microscope is therefor given by two factors, the small focal volume (given by the

- 30 -

numerical aperture of the objective and the wavelength of the illumination light) and the

confocal pinhole aperture.

3.3. CHARACTERIZATION OF SURFACE MODIFICATIONS

3.3.1. Fourier Transform Infrared Spectroscopy (FTIR)

Infrared spectroscopy (IR) is in principle complementary technique to Raman

spectroscopy. Both of them are sensitive to the transitions between vibrational levels,

however each of the techniques can sense different types of molecular bonding, which is

given by the principal of the effect that specifies Raman and infrared spectroscopy. Raman

spectroscopy is a scattering spectroscopy which involves momentary distortion of the

electrons distributed around a molecular bond. That means that the molecule is temporarily

polarized, a momentarily induced dipole that disappears upon relaxation and reemission.

Raman spectroscopy is therefore sensitive to bonds that are polarizable such a carbon

allotropes. Infrared spectroscopy is absorption spectroscopy that requires the vibrational

mode of molecule to have a change in the dipole moment or change in the dipole

distribution associated with it. Then the radiation of same frequency interacts with the

molecule and promotes it to the excited state. Bonds visible in IR spectroscopy must have

vibrations that change the dipole moment of the bond such as functional groups od polar

bonds C=O, O-H, etc. For this reason, the FTIR is suitable for the study of chemical

surface functionalization.

Infrared (IR) spectroscopy is attractive as a for studying surface of samples or for

investigation of particles because of its versatility, broad range of application, relatively

low cost and low requirements on measurement conditions. Most modern surface IR

utilizes a Fourier transform IR (FTIR) spectrometer. The essential feature of an FTIR

spectrometer as compared to a dispersive instrument is that all of the light from the source

falls onto the detector at all the time during measurement. Wavelength identification is not

achieved using monochromators but through frequency analysis (Fourier analysis) of the

periodic signal at the detector produced by the Michelson interferometer or similar device.

An interference pattern is produced by splitting a beam of light into two paths, bouncing

the beams back and recombining them to create alternating interference fringes. The

intensity measured depends on the overall effects of the phase difference for each

- 31 -

component wavelength. The phase difference of course varies with each component

wavelength.

One of FTIR techniques is Attenuated total reflectance (ATR), in which the sample is

brought into close contact with the surface of a prism made of a material with a high

refractive index. A light beam approaching the interface from the optically denser medium

at large enough angle of incidence is totally reflected. However, the beam does penetrate a

small distance into the optically thinner medium (the sample). If the sample absorbs IR

radiation, an IR spectrum can be obtained. This method is depicted in Figure 16.

Figure 16: Principle of Fourier infer red spectroscopy in the ATR mode.

3.3.2. Contact angle measurement

A contact wetting angle developed at a liquid droplet on a solid surface is a result of the

interface/surface tensions (surface free energies) between liquid and solid surrounded by

vapor. The contact angle can be measured by producing a drop of pure liquid on a solid

substrate. The angle formed between the solid/liquid interface and the liquid/vapor

interface and which has a vertex where the three interfaces meet is referred to as a contact

- 32 -

angle. The Young's equation is used to describe the interactions between the forces of

cohesion and adhesion and measure what is referred to as the surface energy.

If the liquid is very strongly attracted to the solid the droplet will completely spread out

on the solid surface and the contact angle will be close to 0°. On highly hydrophilic

surfaces, water droplets will exhibit contact angles of 0° to 30°. On contrary, strongly

hydrophobic solids will have a contact angle close to up or even larger than 90°.

The wetting angle can be described as follows

(1.1)

where is the solid/liquid interfacial free energy, is the solid surface free energy

and is the liquid surface free energy (Figure 17).

Figure 17: Contact angle

The contact angle in this study was measured by adding a 1 l droplet from the manual

dispenser. The shape of the droplet on the substrate was captured by the camera and

analyzed by the software OCA 15 Plus (Dataphysics, Germany).

- 33 -

4. Results and discussions

With the increasing development of new nanomaterials, novel possibilities arise for their

use in biology and medicine. From a biological and medical applications viewpoint, the

interest in nanoparticles comes from the fact that they are small enough to interact with

cellular machinery and potentially to reach previously inaccessible targets. The use of

nanoparticles also opens a possibility to study biological processes on the single molecular

level, as the nanoparticles are small enough to "feel" single biomolecules. Their increased

sensing ability is mainly given by their high surface to volume ratio that results in specific

interactions with biomolecules in the biological environment that cannot be observed in the

bulk materials.

This section reports on the results achieved on the development of the new detection

principle using nanodiamond particles (NDs) [61]. It further demonstrates the use of this

method for the optical sensing of biomolecules adsorbed to the surface. As a final result,

the construction of the in-cell operating biosensor is presented with the ability for the

detection of DNA/RNA transfection. This biosensor can be used for the monitoring of the

drug delivery machinery with the final detection of the drug release/delivery. We further

discuss the use of proposed mechanism for the detection of the DNA hybridization that

would enable noninvasive, easy-to-use and cheap detection of the early stage colorectal

carcinoma.

Nanodiamond particles can serve as a new type of optical marker for cellular imaging

[6][7][8]. Strong ND photoluminescence (PL) originates from single photon producing

nitrogen-vacancy (NV) color centers consisting of a substitutional nitrogen atom next to a

vacancy, engineered artificially in the diamond lattice. The nanoscale effects related to

artificially engineered NV color centers, attracted important attention to diamond due to

applications ranging from quantum computing Error! Reference source not found. to

cell-imaging [7][8][9]. The luminescence from NV centers, incorporated in NDs larger

than 5 nm, is extremely stable without any photobleaching or photoblinking [10][11][12]

and compared to more known quantum dots, ND brings additional advantages such as high

biocompatibility [13][14] and simple C-surface chemistry [15][16]. This allows grafting of

- 34 -

biomolecules that are interesting for the cellular targeting[62][63] or biomolecular drug

delivery [64][65][66].

NV center exist in two charge state, neutral (NV0) and negative (NV

-) with different

optical and magnetic properties [17][18]. NV- center emits bright red with zero phonon line

(ZPL) at 637 nm that is followed by broad side band luminescence with the highest

intensity around 700 nm, oppose to an orange luminescence from NV0 center with ZPL

around 575 nm also followed by broad side bands. Single NV center can exist in both

charge states, i. e., one center can be switched from negative to neutral state and vice versa

[11][12]. If we were able to understand the mechanism of the charge switching, and we

found a way to control this mechanism, we would achieve a new method for optical

sensing. For biomedical application, we would gain a biocompatible and extremely stable

two-color label that has additional sensing ability.

The first section: 4.1. Theoretical consideration of surface manipulation with optical

defects by charge transfer, introduces a novel detection principle that is based on the

surface-charge induced modulation of photoluminescence of nitrogen vacancy centres in

diamond. This method is based on the modulation of the occupation of energy levels of

two charge states of NV centers. This can be done by shifting the position of Fermi level

above the defect's charge transition level. It means that when the Fermi level is shifted

below a charge transition level, the defect loses an electron and is converted to its neutral

or eventually other state. With this purpose, we utilized the unique properties of the

hydrogen-terminated surface of diamond which is called a surface transfer doping. H-

terminated diamond surfaces possess a high electric dipole moment which attracts polar

ions, leading to creation of a hole accumulation layer at the surface. This leads to

consequent upwards surface band bending, causing a generation of 2 dimensional hole gas

at the diamond surface and pinning of the Fermi level at the valence band maximum. By

this way, the Fermi level will be lowered below the energy of NV- ground state. By

controlling the band banding at the diamond surface one thus would be possible to control

the NV charge state to a certain depth and consequently the luminescence, as is in detail

described in Chapter 4.1.1 Hydrogenated diamond surfaces.

To determine exact band bending profiles, the Poisson equation with the Bolzmann

statistical distribution is solved in this chapter, calculating with N donors and charged NV

- 35 -

centers, present in the diamond, influencing the band-bending zone. Based on the solution

of Schrödinger equation using DFT calculations, NV0

and NV- centers ground state energy

levels were calculated. Chapter 4.4.2 Band bending calculations describes in detail

simulations of the band bending profile of hydrogenated diamond surface with the

transition levels of the NV defects performed for two types of model system: A) flat

surface with Gaussian distribution of nitrogen and NV centers in the surface proximity and

B) the spherical surface with homogenously distributed nitrogen and NV centers.

Parameters of the model are chosen to closely correspond with experiments, as for example

depth profiles of nitrogen distribution in the flat surface is given by the SRIM (Stopping

range of ions in matter) simulation for parameters of nitrogen implantation performed on

single crystal diamond plate. Simulations of the band bending profile show that electrons

are transferred from the valence band of diamond to compensate for a charge induced at

the diamond surface by adsorbates, which contribute to an upwards surface band bending.

The total electric field profile is then the balance between the surface adsorbates and the

deep donors (N, NV-). Calculations show a great enhancement of the surface effect for the

spherical surface that corresponds to the nanoparticle. In that case the effect of the NV0 and

NV- switching can be observed up to 18 nm in depth in comparison to a limited band

bending for the case of flat surface, where the reduction of the NV- luminescence can be

observed only up to several nanometers.

The theoretically predicted surface-induced charge switching of NV centers is

experimentally verified using a well-defined model system in Chapter 4.2. Implication of

the surface functional changes on the optical properties of the model system:

implanted single crystal diamond. The changes in the optical properties of NV centers

are shown on the defects implanted to specific depths to high purity single crystal

diamond. The spectra of the NV centers of hydrogenated and oxidized surface are

compared. Hydrogen termination followed by adsorbate attachment to the diamond surface

leads to upwards band bending, i.e. the hole accumulation layer, while oxygen termination

(having an electric dipole moment in the opposite direction) causes an opposite effect, i.e.

downwards band bending. Results show observable, but small changes in the NV-/NV

0

ratio with the surface termination with the clear dependence on the implantation energy.

This dependence is related to different levels of the nitrogen content (various efficiency of

creation of NV centers as a function of implantation energy) and is explained by performed

mathematical modeling.

- 36 -

To develop the nanodiamond luminescence probe, NV centres have to be primarily

fabricated in the nanodiamond lattice. For biological applications, the brightness of the

probe is one of the key factors enabling its standard use. As the brightness of

nanodiamonds depends on the number of NV centers in the particle, the optimization of the

process of creation of NV centres is necessary. There are three basic approaches to form

NV centers in diamond: i) implantation of nitrogen and annealing in ultra-high purity

diamond, ii) irradiation of the nitrogen-containing diamond and annealing and iii)

incorporation of NV centers in the process of CVD growth. The Chapter 4.3 Formation

of variously charged NV centers describes optimization of the formation of NV centers

in nitrogen containing nanodiamond particles. High energy irradiation leads to creation of

vacancy as the ion loses energy to the lattice. During thermal annealing, vacancy becomes

mobile while the single nitrogen remains immobile. When the vacancy is trapped to the

nitrogen, the thermally stable NV center is formed. In order to obtain ND with high

population of NV centers suitable for bio-labeling applications, the influence of particle

type, particle energy and annealing parameters was studied.

Chapter 4.3.1 Introduction of vacancies to the diamond lattice – different irradiation

strategies compares luminescence of nanodiamonds irradiated on cyclotron beams (p+, d

+,

3He

2+,

4He

2) with energies from 5 MeV to 25 MeV and microtron with electron beams of

energy range from 7 MeV to 23 MeV. The effect of the various irradiation conditions is

studied after annealing, by comparison of the NV luminescence emission normalized to the

Raman signal. The best luminescent characteristics were obtained by proton beam

irradiation with the energy deposit without Bragg peak. The creation efficiency was

approximately 10 times higher in the case of 130 nm particles in comparison to 35 nm

particles. The study of amorphisation of samples is also stated in this chapter. The presence

of high sp2 content in the samples, where the high stopping power created the maximum

concentration of vacancies, shows that the critical threshold for amorphisation of diamond

was reached in samples. In this respect it could be predicted that irradiation with lower

fluence, than in our case 9.1015

cm-2

, would lead to higher creation efficiency.

The influence of the annealing conditions on the creation of NV centers is discuss in

Chapter 4.3.2 Formation of NV centers: Annealing study. Commonly used annealing

parameters of nanodiamond particles are in the range of 700 – 800 °C for 1 – 2 hours. We

- 37 -

highly extend these conditions creating a matrix of annealing conditions, consisting of

various combinations of temperatures and respective times (700 - 950 °C, 0.5 – 8 h). The

most striking as well as practically applicable result of the study is finding of a discrete

maximum of NV photoluminescence in samples annealed at 900 °C for 1 hour. This result

shows unprecedented conditions enabling to generate up to 3-fold brighter particles

(compared to commonly used range of 700–800 °C for 1–2 hours) just by a subtle change

of annealing temperature and time.

Nanodiamond prepared by described procedures provide sufficient luminescence contrast

that is comparable to other biomarkers, and retains advantages related to their nano-

dimension for cellular tracking. Chapter 4.4 Chemical control of the NV luminescence

in nanodiamond describes how the surface chemistry effects can make the bulk

luminescence of nanodiamond sensitive to chemical processes ongoing at the surface. The

proposed method is based on the control of electronic chemical potential at the ND

surfaces, influencing the surface band bending earlier demonstrated on single crystal

diamond.

This mechanism is demonstrated on ND with hydrogenated and oxidized surface which

exhibits important differences in the surface chemical potential. The data presented in

Chapter 4.4.1 Quenching of NV- luminescence on ND particles, report on the three

situations: oxidized surface, hydrogenated surface and diamond after subsequent oxidation

performed by annealing in air. The result show the NV- luminescence fully quenched after

hydrogen termination, while NV0 luminescence is still visible. The NV

- luminescence

could be restored again by annealing in air (leading to oxidized surfaces). In addition we

observed a large influence of ND size on the photoluminescence intensity of both NV0 and

NV- defect centers. The process of surface charge transfer modulation of luminescence of

nanodiamonds was demonstrated for particles of size in the range of 8 – 100 nm. This is

significant advantage over Forester Resonance Energy Transfer (FRET), in the point of

using a standard confocal microscope used in biological research, in which the larger

particles are easy to observe, while particles ~ 5nm or smaller, which are necessary for

FRET interaction are difficult to distinguish on the resolution of standard optical

microscope. Successful demonstration of the mechanism of luminescence switching

presented in this chapter is one of the milestones of this thesis and leads towards novel

biomedical detection principles.

- 38 -

Luminescence of NV centres is known to be very stable without any photobleaching or

photoblinking. However, if the centers are incorporated in very small particles (below 5

nm), the blinking and quenching was observed. Chapter 4.4.2 Fluorinated

nanodiamonds show a new possibility to stabilize the luminescence of NV centres in the

close surface proximity. The difference in the PL properties of hydrogen and oxygen

terminated surface is the consequence of different electron affinity of hydrogenated and

oxidized surfaces. Since the variation in electron affinity is closely connected to the surface

electric dipoles that are caused by different electro negativities of elements terminating the

surface, introduction of highly electronegative element, such as fluorine, to the surface, the

quenching effect of the surface on the luminescence could be reduced.

This hypothesis is experimentally supported in Chapter 4.4.3. Size dependence of the

luminescence of nanodiamonds, in which the photoluminescence is studied for different

size fractions of variously terminated nanodiamonds (oxidized, hydrogenated and

fluorinated). A strong size dependence of PL properties can be found. Luminescence of H-

terminated ND quenched completely for 8-10 nm NDs, the 15-20 nm NDs showed clear

NV0 related luminescence with no NV

- luminescence, for larger ND NV

-/NV

0 ratio could

be precisely tuned. The luminescence of oxidized ND was found to be less sensitive to size

of ND, but changes in NV-/NV

0 ratio were also observed, with the favor to NV0

luminescence for smaller ND. The NV- luminescence was still dominant in all oxidized

samples. PL of fluorinated ND was found to be the most stable in all size ranges. The

intensity of NV- emission of fluorinated ND was in average 20% more intense than the

emission of oxidized ND.

To make use of the principle of charge-transfer induced modulation of luminescence for

detection of biological processes, the alternation of the possition of the Fermi level at the

surface should be driven by conditions/events that can occur in biological environment.

One of the possible alternatives is a non-covalent binding of charged biomolecules to the

surface of nanodiamond particle. This aspect is addressed in Chapter 4.5. Sensing of

charged molecules via NV luminescence. If the charged molecule is strongly attracted to

the surface, the charge transfer can occur, depending on the HOMO (highest occupied

molecular orbital) and LUMO (lowest unoccupied molecular orbital) energetic levels of

the charged molecule. The principles of charge-interaction of molecules in the close

- 39 -

proximity of the diamond are demonstrated on the interaction with four types of charged

polymers. The luminescence of NV- centers clearly decreased upon interaction with

positively charged molecules, while after addition of negatively charged polymers, the

luminescence was restored to the original level. The results presented in this chapter show

for the first time the ability of fluorescence nanodiamonds to interact with charged

polymers that leads to the changes in the NV luminescence. These results bring new

possibilities for the biomolecular detection and open a variety of new possible applications

of fluorescence nanodiamonds.

Additional verification of the impact of charged molecules on the luminescence of

nanodiamonds is shown in Chapter 4.5.1. pH dependence. pH indicates the concentration

of the solvated hydrogen ions (protons) and gives information about deprotonation of the

solvent. The design of the experiments with charged polymers allowed us to monitor the

charge switching of the polymer with changes in the pH. Results reported in this chapter

clearly show the correspondence between luminescence changes observed on

nanodiamond particles and the changes in the Zeta potential, both measured on the

particles coated with variously charged polymers as a function of pH. These results give us

an important understanding of the process of charge transfer from charged polymers to

nanodiamonds that is used further in the proposed biosensing application.

To be able to use nanodiamonds as an in-cell operating biosensor, we need to understand

the mechanism of cellular uptake. Results of the study of this mechanism are presented in

the Chapter 4.6. Study of the cellular uptake. As nanodiamond size is ranging from a

few nanometers to hundreds of nanometers, the most of particles are taken up by via

endocytosis. As shown in Chapter 4.6.1. Oxidized nanodiamonds, the incorporation of

oxidized FNDs was tested with IC21 macrophage and HT29 colorectal adenocarcinoma

cell lines or peripheral blood mononuclear cells (PBMC). While the IC21 cell line

intensely uptakes 130 nm particles even after short incubation, the HT29 adenocarcinoma

cell line takes at least 24 hours to show significant FND uptake. Higher incorporation of

the HT29 cells was observed for 35 nm particles as nanodiamonds were observed in 70 %

of cells.

Nanodiamonds are aimed for the use as a transfection system, in which the DNA

attachment and release could be monitored optically. For this purpose, nanodiamonds were

- 40 -

coated by polyethylenimine (PEI), a polumeric transfection enzyme. Chapter 4.6.2.

Polymer coated nanodiamonds show an increase cellular uptake of such polymer coated

nanodiamonds, when the most of the cells contained nanodiamonds. The same results were

observed on PEI and DNA coated nanodiamonds. This result reports on a simple method

of improving the ability of cellular uptake.

The final chapter in this section, Chapter 4.7 Nanodiamond transfection system

enabling detection of DNA delivery, reports on the possibility of using nanodiamond

particles as optical biological sensor based on the method of the surface charge induced

luminescence changes, developed within this thesis. In recent studies, polyethylenimine

(PEI) coated nanodiamonds (non-luminescent) showed increased efficiency of siRNA

transfection in comparison with plain PEI of the same molecular weight, while remaining

biocompatible. Based on these results, we show the construction of such transfection

system and the ability of this system to monitor the binding of the transfection polymer and

DNA to the ND particle that enables us to optically monitor the successful gene delivery.

This sensor is based on a simple fact that the DNA or RNA molecules are strongly

negatively charged and usage in combination with the PEI (positively charged) would lead

to reversible charge switching upon attachment/release of nucleic acids.

The formation of PEI-ND complexes is based on the electrostatic attraction of positively

charged PEI and oxidized ND surface. The optical monitoring of a formation of ND-PEI

complexes as well as ND-PEI-DNA complexes is reported in chapter 4.7.1. Monitoring of

formation of transfection system. Nanodiamonds of various surface treatment are used as

well as PEI of different molecular weigths, to understand and to optimize the charge

interactions that lead to the modulation of NV luminescence. The presence of positively

charged PEI led to decreasing or quenching of NV- photoluminescence. Subsequent

addition of negatively charged DNA compensated the positive charge, which resulted in

enhancement or restoration of NV- luminescence.

The final aim of the thesis is presented in chapter 4.7.2. Detection of the DNA release in

cells. In this section, we have demonstrated the construction of the fND biosensor

operating contactless in cells based on the discovered principle of the surface charge

interactions. The direct optical monitoring of DNA intracellular release from transfection

- 41 -

system at single particle level was demonstrated using the ND containing NV centers as a

delivery/sensing system.

Chapters 4.7.3. Biocompatibility of the ND-transfection system and 4.7.4. Verification

of successful transfection report on the results showing the proliferation, viability of cells

containing PEI and PEI-DNA coated nanodiamonds as well as the expresion of the DNA

delivered to cells, that was successful only for the case of the ND-PEI-DNA complex.

- 42 -

4.1. THEORETICAL CONSIDERATION OF SURFACE MANIPULATION WITH

OPTICAL DEFECTS BY CHARGE TRANSFER

The NV defect is a one of the sources of stable and bright luminescence in diamond,

responsible for pink coloration of diamond crystals when present in high. NV center exist

in two charge state, neutral (NV0) and negative (NV

-) with different optical properties. The

NV center has a wide absorption spectra in the visible range that allows using of various

different lasers (from blue to yellow) to excite the emission of the center. NV0 and NV

-

differ in the absorption and emission spectra. NV- center emits bright red with zero phonon

line (ZPL) at 637 nm that is followed by broad side band luminescence with the highest

intensity around 700 nm, oppose to an orange luminescence from NV0 center with ZPL

around 575 nm also followed by broad side bands.

An interesting fact is a possibility of a direct charge transfer of these two states,

accompanied with the changes of the zero phonon line from 575 nm to 636 nm i.e. from

NV0 to NV

-. Charge transfer was demonstrates in earlier works showing a manipulation

with the occupation of the NV center by a monochromatic light absorption with defined

energy, allowing to quench the radiative transition from 3E exited to A3 ground state

triplet states, as shown in Figure 18.

Figure 18: Room-temperature PL spectra with and without 2,8 eV illumination, and the difference spectrum, illustrating the light induced charge transfer of electrons from NV- to

NV0 centres (reprinted from [12] with permission of the author).

- 43 -

4.1.1. Hydrogenated diamond surfaces

H-terminated diamond surfaces possess a high electric dipole moment which attracts polar

ions such as water adsorbates [67][68], leading to creation of a hole accumulation layer at

the surface for undoped diamond of a high purity. This leads to consequent upwards

surface band bending, causing a generation of 2 dimensional hole gas (2DHG) at the

diamond surface and pinning of the Fermi level at the valence band maximum

(EVBM)[70][71]. This effect is called surface transfer doping [69] and is used in many

devices such as solution gate FETs [72][73]. One can employ a very similar effect, i.e.

changes of chemical potential of variously terminated diamond surfaces to drive close to

the surface laying NV centers that are present in the band bending zone, to induce changes

in their occupation. This effect has been clearly documented for single crystal,

polycrystalline and also nanocrystalline diamond[74][75].

To determine exact band bending profiles, one has to solve Schrodinger – Poisson

equation, calculating with N donors and charged NV centres, present in the diamond,

influencing the band-bending zone. Based on the solution of Schrödinger equation using

DFT calculations, NV0

and NV- centers, have different ground state energy levels 1.2 eV

and 2.0 eV (calculated by Goss et al.[76][77]). Fermi level (EF) energy will be lowered

below the energy of NV- ground state if this state loses electron and becomes unoccupied.

By such way this center will become inactive and cannot contribute to the luminescence

(electron cannot be excited from this state). By controlling the band banding at the

diamond surface one thus controls NV (or SiV and other) charge states to a certain depth

and consequently the PL spectra.

Color shifts in diamond affected by surface termination reported by our group [78] were

also observed in [79] but their origin, which we model, was not explained. In general, (A)

hydrogen termination followed by adsorbate attachment on undoped diamond surface leads

to upwards band bending, i.e. the hole accumulation layer, while (B) oxygen termination

(having an electric dipole moment in the opposite direction) causes an opposite effect, i.e.

downwards band bending. If the particular defects lay on ~ 10 nm scale from such surface,

the surface transfer doping [69][70] could influence the defect luminescence,. By such way

- 44 -

we could induce luminescence shifts via relative shift of NV0/NV

- levels with respect to

the EF (i.e. the state occupation). The situation is illustrated in Figure 19.

Figure 19: Schematic model of surface band bending for hydrogen terminated diamond containing NV centers with (a) low content of nitrogen, (b) high content of nitrogen.

Nitrogen in the diamond lattice acts as an electron donor, electrons will compensate the holes accumulated at the surface, effectively reducing the surface band bending and inhibiting the influence of hydrogen termination on photoluminescence. Additional nitrogen influences the position of the Fermi level EF in the band gap. (c) Oxidized

diamond surface with downward band bending, NV0 and NV- are occupied. x (nm) and y (nm) are regions of which the Fermi level crosses the NV- and NV0 energy, i.e. regions where the ground levels of NV- and NV0 are not occupied and therefore luminescence

cannot occur.

4.1.2. Band bending calculations

The electronic transition (i.e. optical excitation) from NV- or NV

0 ground states (

3A) to its

excited state (3E), occurs by absorbing a photon of energy AE

EEh 33 . This is only

possible if the NV0 or NV

- ground states are originally occupied by a single electron

(neutral NV0) or two electrons (negative NV

-). That means, EF must lay above the dark

ground level of particular NV0 or NV

- for the center [12]. To determine exactly the

position of Ef at the surface in N implanted diamond one has to take into account the

influence of NV- , NV

0 and N defects. And the electric filed at the surface induced by H-

termination and surface adsorbates. All these effects will contribute to the total charge

- 45 -

balance and influence the surface band bending (see Figure 19). If the concentration of the

donor-like centers (N or NV–) is increased, the holes from the accumulation layer at an H-

terminated surface can be transferred to deep donors by converting N to N+ and NV

- to

NV0.

The following model is calculated for N-shallow implanted diamond with energies 2-5

keV, followed by subsequent annealing to generate NV centers. In our model, electronic

transitions (i.e. optical excitation) from NV- or NV

0 ground states (

3A) to its excited state

(3E), occur by absorbing a photon of energy AE

EEh 33 . This is only possible if the

NV0 or NV

- ground states are originally occupied by a single electron (neutral NV

0) or two

electrons (negative NV-). That means, EF must lay above the dark ground level of

particular NV0 or NV

- for the center [12] compared to the 2DHG band bending model in

H-terminated undoped diamond [71], in N implanted diamond one has to take into account

the additional influence of NV- , NV

0 and N defects. These defects will contribute to the

total charge balance and influence the surface band bending (see Figure 19). If the

concentration of the donor like centers (N or NV–) is increased, the holes from the

accumulation layer at an H-terminated surface can be transferred to deep donors by

converting N to N+ and NV

- to NV

0. The upwards band bending is then consequently

reduced, depending on the concentration of adsorbates.

In our model, the occupation of NV0 and NV

- states is calculated from the Boltzmann’s

statistical (concentration of NV centers is on the range of 1017

cm-3

) distribution, taking

into account the presence of N and NV centers at various concentrations. The electric field

profile on an H-terminated surface with surface adsorbates is calculated using the Poisson

equation for a dielectric medium and energy of NV0/-

centers determined from DFT

calculation by solving the Schrödinger equation [76][77]. These values for NV0

and NV-

energy levels (1.2 eV and 2.0 eV respectively), are relative to the valence band top. N acts

in our model as a deep electron donor with an energy level at 1.7 eV below the conduction

band [80] (Figure 19). It is evident that in case of doped diamond one can additionally

count with ionized impurities such as B- or P

+.

- 46 -

To study quantitatively the occupation of NV- and NV

0 ground states, we have carried out

precise mathematical modeling outlined below using Boltzmann statistics and solution of

Schrödinger equation for NV level energies. The boundary conditions are as follows:

The detail charge balance includes the surface acceptors at distance 0x with a surface

density of 1013

cm-2[80]

. For an H-terminated surface on a non-implanted SCD (single

crystal diamond) with the surface conductivity in the range of 10 k, as measured in our

case, we can assume that the Fermi level is pinned at the surface at EVBM [67].

We can write for the depth x dependent total space charge density x :

kT

EEeNx xVF

V exp

, (1)

where VN is the temperature dependent effective density of states at EVBM,

19107.2 VNcm

-3 at room temperature, e is elementary charge, k is Boltzmann

constant and T is thermodynamic temperature. FE and VE are energetic levels of Fermi

and valence band.

x

e

dx

EEd

r

xVF

02

2

, (2)

where 0 r is the permittivity of material and x is total space charge density given by

Boltzmann distribution (1).

The Poisson equation was solved numerically with boundary conditions

0

0ln

p

NkTEE V

VF

(3)

gxNxNNep DDA 10 (4)

where 0p is the total unscreened positive charge at 0x from (1) and AN is the effective

density of surface acceptors. The terms 1xNe D and gxNe D represent the net

original concentrations of N and NV- centers respectively in the sample after annealing.

is the conversion efficiency of N to NV centers for thermal annealing, which is different

for SCD and for ND[86][87] g is the relative occupation of NV- and NV

0 centers in bulk

diamond and the e the unitary electron charge.

- 47 -

The concentration of nitrogen xND is in the case of implanted SCD the density of donors

(i.e. N and NV- centers) is given by the SRIM (Stopping range of ions in matter)

simulation of the implantation nitrogen (Figure 21). For the case of HPHT NDs, we

assume that nitrogen and NV centers are homogenously distributed in the sample thickness

and the irradiation of 5.4 MeV protons is also homogenous in the volume. Only a part of

the implanted nitrogen is converted to NV centers during annealing [85][86], therefore the

parameter is introduced to the model. This parameter represents a bulk dependent

conversion efficiency of N in the lattice to NV centers. It has been shown recently that the

parameter depends strongly on the method of implantation, which can be also the reason

for observed difference in the NV PL intensity in both SCD and HPHT ND, as discussed

later. In our calculations was set to 0.1 for Ib HPHT NDs [86] and 0.01 for IIa implanted

and annealed SCD [87]. Our recent results show that efficacy as high as 2-5 % can be

achieved for generation of NV centers. Upon annealing and diffusion of vacancies, one can

estimate that N in the bulk of diamond will be proportionally converted to NV0 and NV

-

centers; therefore we introduce a parameter g, which is set in our standard stimulations to

0.5 (i.e. the concentration of NV- and NV

0 centers would be equally produced without any

surface bending). Using these input parameters, the band bending was calculated.

It is interesting to evaluate electric field distribution not only in flat samples but also in

round nanoparticles that have important applications in quantum informatics or

nanobiology. To calculate the surface band bending in case of spherical symmetric ND

particles, the above equations are transformed to spherical coordinates. The surface charge

is assumed to be homogenously distributed at the surface of the sphere.

kT

EEeN

e

dr

EEdrVF

VrVF

exp40

2

2

, (5)

where r is the radius of the sphere. The boundary condition are used as previously.

The results of the numerical simulation for the case of the H-termination of diamond

containing N and NV centres are shown in Figure 20. Electrons are transferred from the

valence band of diamond to compensate for a charge induced at the diamond surface by

adsorbates, which contribute to an upwards surface band bending. The total electric field

profile is then the balance between the surface adsorbates and the deep donors (N, NV-).

- 48 -

Our calculation yields the value of about 2 eV band bending at x=0 for undoped diamond

(i.e. diamond without any NV/N defects) which agrees well with the measured value [72].

However, if additional donors (N, NV-) are available near the surface, the band bending is

reduced proportionally to the defect or dopant concentrations.

Figure 20: Simulation of electron statistics in hydrogen terminated surfaces using density of states (DOS) model. Band diagrams show energetic profiles of the conduction band

minimum (EC), valence band maximum (EVBM), NV-, NV0 and N impurities relative to the Fermi level (at zero energy) for (a) Ib HPHT diamond contained 200 ppm of nitrogen with

surface carrier density 1013 cm-3, where 10 % of nitrogen is converted to NV centers [86][87]. In this case, we solve the Poisson equation for ND particles using spherical

coordinates. In this case we approximate diamond as a spherical particle. (b) IIa diamond

plate implanted with nitrogen (6 keV, 1013 ionscm-2), surface carrier density 1013 cm-3 with 1 % yield of conversion to NV center. (c) IIa diamond plate implanted with nitrogen (1 keV,

1013 ionscm-2), surface carrier density 1013 cm-3 with 0,1 % yield of conversion to NV centers. Cases b and c are solved using 1D model.

Figure 20 compares band-bending calculations, for two model situations, for HPHT NDs

and for implanted SCD. When comparing modeled data in Figure 20, it is clear that the

observed effect of PL shift is much weaker in SCD then in ND. The influence of H/O

termination on PL in implanted SC CVD diamond was discussed in our previous work

Error! Reference source not found. and also recently in [88]. The enhancement of the PL

- 49 -

shift in ND is in agreement with theory for size dependent effect in spherical particles

outlined below.

- 50 -

4.2. DEMONSTRATION OF SURFACE FUNCTIONAL CHANGES ON THE

OPTICAL PROPERTIES OF THE MODEL SYSTEM: IMPLANTED SINGLE

CRYSTAL DIAMOND

To demonstrate the possibility of surface charge driving, we have used ultra-high purity

single crystal CVD diamond (SCD), implanted by N. Four different sectors in a high purity

SCD were shallow implanted with nitrogen, using energies 1 keV and 6 keV. Created NV

centers were localized from around 3 nm to 10 nm of depth (Figure 21). The surface of the

SCD was later oxidized and hydrogenated.

Figure 21: a) Schematic map and TRIM simulation for SCD diamond with four sectors implanted by N+ ions. Sectors A and B were implanted with the same dose 1013 cm-2, but

different energies (1 and 6 keV, respectively). Sectors C and D were implanted twice at the same conditions (dose 1013 cm-2, energies 1 and 6 keV, respectively). The y-axis scale shows implanted atoms in atoms/cm3 per 1 carbon atom.[90] b) Contact angle of hydrogenated

and c) oxidized SCD.

Hydrogenation of samples was done in microwave-excited hydrogen plasma for 30

minutes at a temperature of 500°C and pressure of 1 mbar in a vibrating holder. Finally, the

H-terminated samples (droplets) were annealed for 120 minutes in air at 400°C to oxidize

the surface. At each step the luminescence spectra were measured at room temperature.

- 51 -

Figure 22 shows typical Raman and PL spectra of hydrogenated and oxidized SCD for

implantation energies 1 and 6 keV. Though the NV-/NV

0 ratio changes with the surface

termination, the observed changes are rather small. PL emission from NV- ZPL is also

clearly influenced by the implantation energy. For H-terminated SCD on sectors with 6

keV implantation energy (and higher surface conductivity), PL spectra showed decreased

emission from NV- ZPL compared to 1 keV implantation (Figure 22a). To explain the

reduction of the effect compared to ND we have performed mathematical modeling of the

effect.

c)

Figure 22: a, b) 514 nm-pumped Raman and PL spectra of treated SCD, showing the influence of different surface termination on NV0 and NV- related luminescence. NV0 ZPL

(575 nm) peaks and NV- ZPL (638 nm) peaks are visible for oxidized SCD. For SCD with hydrogenated surface, NV- related luminescence is decreased with an increase in NV0

related luminescence, as shown in the difference spectra. All spectra are normalized to the

- 52 -

diamond Raman peak (1332 cm-2) and background corrected. a) 1 keV and b) 6 keV N+ ion implantation energy. When normalized to second order diamond Raman peak, the

number of counts (i.e. intensity) of NV related luminescence is significantly lower for 1 keV implantation energy compared to 6 keV implantation energy. This fact suggests different efficiency of creation of NV centers depending on the implantation energy for the same total dose.This effect is discussed below and evaluated in the mathematical modeling of

observed effects (c). Fluorescence of NV defects implanted with 3 different energies. Dark stripes mark areas where the diamond surface is hydrogenated and bright areas indicated

places where the surface is oxygen terminated (c) Ref [88]

In parallel, the possibility to drive NV centers chemically was presented by Hauf et. al [88]

on the NV centres implanted to the single crystal using various implantation energy and

fluence (see Figure 22b). The same group presented recently the control of the charge state

of the single centre by an electrolytic gate electrode [91].

- 53 -

4.3. FORMATION OF VARIOUSLY CHARGED NV CENTRES

For the control of NV centers for applications to quantum computing but as well to

biological applications charged state of NV centers as well as number of active NV centers

is of primary importance. While, in principle, for quantum applications single NVs or their

arrangement in interacting networks is required, for biological applications very bright

luminescent nanodiamonds are needed. In this respect, there is a necessity to optimize the

process of formation of NV centers in order to get controlled number of NV centers as well

as their occupation. This section deals with fabrication method. There are three basic

approaches to form NV centers in diamond: i) implantation of nitrogen and annealing in

ultra-high purity diamond (Figure 23), ii) irradiation of type Ib diamond and annealing

(Figure 24) and iii) incorporation of NV centers in the process of CVD growth.

Figure 23: Schematic figure of the formation of NV centres in single crystal diamond by nitrogen implantation (a) and annealing (c). Vacancies are created in the path of the

implanted nitrogen atoms (b). This method is efficient for quantum applications due to the possibility of tuning of the implantation depth and precise control of the position of

nitrogen atoms.

- 54 -

Figure 24: Schematics of the formation of NV centres in Ib nanodiamond particles containing single substitutional nitrogen atoms by proton irradiation (a). Vacancies are produced by the proton penetration through nanodiamonds (b). NV centres are formed

during thermal annealing using temperatures above 600 °C due to the migration of vacancies (c). Some of the vacancies are trapped by the nitrogen atoms, others diffuse to

the surface.

The N implantation was discussed in the chapter 3.2. Here we concentrate on the method

ii). The CVD growth is mainly effective for Si incorporation as N incorporation usually

leads to deterioration of diamond quality for large N concentrations.

Preparation of luminescent particles

A commercial solution of ND (HPHT, Ib, Microdiamant, Switzerland) was lyophilized and

heated in the slow stream of air at 425°C for 5 h to remove any sp2 carbon-containing layer

[92]. The resulting pale grey powder was dispersed in water and deposited in the form of a

thin film on a target backing (10 mg cm–2

) for proton implantation. The ND was then

irradiated using an external proton beam of the isochronous cyclotron U-120M. The angle

of the target backing with respect to the beam direction was 10°. The fluency of the

delivered beam was 9.2x1015

cm–2

, beam energy 5.4 MeV and beam current 0.6 μA. The

irradiated ND was thermally annealed in vacuum at 710°C for 2 h and then oxidized in a

mixture of concentrated H2SO4-HNO3 (9:1, vol/vol) [8] at 90°C for 7 days. The reaction

mixture was diluted by deionized water, the NDs were separated by centrifugation and

- 55 -

washed subsequently by 0.1 M NaOH, 0.1 M HCl, and finally three times by water. The

solution was lyophilized providing highly luminescent NDs in the form of a stable

colloidal dispersion in water as confirmed by AFM and DLS (final size ~20-100 nm). The

colloidal dispersion was stable after 2 months with no sedimentation. A micro droplet of

the dispersion was dried on a quartz substrate, producing isolated ND particles, as

confirmed by AFM (see Figure 25). The rest of the material was lyophilized for later

hydrogenation. The AFM measurements were performed in tapping mode (111 kHz) with

an NTEGRA Prima NT MDT system equipped with a soft HA_NC etalon tip.

Figure 25: AFM image of single dispersed ND particles, dispersed on a quartz glass.

Raman spectra taken in each step of the process confirmed high quality of ND showing no

visible non-diamond structures in the particles (Figure 26). Therefor the PL changes are

connected to interactions with diamond surface without any additional influence of sp2 or

other non-diamond structures.

- 56 -

Figure 26: Raman spectra taken from untreated ND, ND after irradiation, ND after irradiation and annealing and ND after final oxidation.

Irradiation of type Ib diamond (most commonly HPHT) is the most favorable method for

the large-scale production of highly luminescent diamonds by generating vacancies and

recombining them with the N. HPHT nanodiamonds are commercially available at low

cost in the size ranging from single digit NDs to hundreds of nanometers. Another

advantage is the natural content of C centers (isolated substitutional nitrogen in the

diamond lattice) which allows to overcome problems with introduction of nitrogen such as

the low penetration depth in the case of nitrogen implantation (1 um/MeV [81][90]). This

limitation prevents large-scale production. High energy irradiation leads to creation of

vacancy as the ion loses energy to the lattice. In general, electrons and protons produce

predominantly isolated vacancies, neutrons and heavy ions produce regions of multiple

damage [22]. Additionally, electrons and neutrons exhibit high penetration depth (orders of

mm) in the contrast with rather low penetration depth in the case of protons or other

heavier charged particles (orders of micrometers for MeV energies). During annealing,

single nitrogen is immobile and vacancies are mobile. When the vacancy is trapped to the

nitrogen, the thermally stable NV center is formed. Formation of defect centers in diamond

by irradiation is well described on single crystal diamond; however no systematic study

was done on nanodiamond particles. In order to obtain ND with high population of NV

centers suitable for bio-labeling applications, the influence of particle type, particle energy

and annealing parameters was studied.

- 57 -

4.3.1. Introduction of vacancies to the diamond lattice – different

irradiation strategies

Accelerators, i.e. isochronous cyclotron U-120M and microtron MT 25, were used for the

production of the fluorescent nanodiamond (fNDs). The irradiations were performed both

on cyclotron beams (p+, d

+,

3He

2+,

4He

2) with energies from 5MeV to 25MeV and

microtron with electron beams of energy range from 7MeV to 23MeV. Particle density was

chosen to be approx. 1016

/cm2.

Target holders were designed for the purpose of irradiation in cyclotron, the electron

irradiation was performed using glass vial as the target. The layer of ND was prepared on

each target. Two different irradiation energies were chosen for each type of irradiation

particles in cyclotron (p+, d

+,

3He

2+,

4He

2); energy deposit with Bragg peak (ion beam is

stopped in the target) and without Bragg peak (ion beam flies through the target). Two size

ranges of nanodiamonds were used in this study: ND with the size distribution at 35 and

130 nm. All particles were subsequently annealed at 700 °C for 2 hours under the Ar

athmosphere and oxidized by wet oxidation [45]. To analyze the quality of samples, all

samples were characterized by luminescence and Raman spectroctroscopy.

The concentration of nitrogen in Ib HPHT ND was about 100-200 ppm. For the fluence of

1016

cm-2

, the maximal vacancy concentration of the atomic displacements was 1019

cm-3

(that corresponds to 100 ppm) [90].

Creation of vacancies: GR1 vs ND1

All irradiated samples used in the study showed clear PL originating from GR1 centre

(Figure 27). However, due to the fact that ND1 vacancies do not exhibit any PL, the effect

of the various irradiation conditions was studied after annealing; by comparison of the NV

luminescence emission normalized to the Raman signal. In addition, the amorphization of

samples is studied by Raman spectroscopy.

- 58 -

Figure 27: PL spectra of proton irradiated nanodiamonds before annealing (130 nm particles) showing detectable emission from GR1 centre – measured at room temperature,

excitation 636 nm.

The presence and concentration of GR1 and ND1 centers in natural and synthetic single

crystal diamond, produced by radiation damage, has been studied using optical absorption

spectroscopy and electron paramagnetic resonance. Although these techniques, unlike

photoluminescence spectroscopy, can be used to determine quantitatively the amount of

defect centres, their use is limited in the case of nanodiamond particles (due to high

reflection, scattering, low signal). The possible option for confirmation of the presence of

defect centre is photoluminescence. This method cannot be used directly to evaluate the

number of defects, however it is possible to quantitatively compare the number of defect

centres, when we normalize the photoluminescence to the Raman signal [22].

By irradiation of nitrogen rich diamond, the number of created negatively charged

vacancies (ND1 centre) highly exceeds the number of neutral vacancies (GR1 centre)

[22][93], because substitutional nitrogen acts as an electron donor to the vacancy. If the

concentration of nitrogen is higher than few ppm, the number of ND1 centres dominates

over the GR1 centre [94]. When the concentration of nitrogen is 75ppm, the GR1

absorption is undetectable [22]. GR1 centres have characteristic absorption and emission

zero phonon line around 741 nm. ND1 has a characteristic absorption line at 3,15 eV and

exhibit no photoluminescence, instead emits an electron.

- 59 -

Resulting NV PL of variously irradiated ND

In the case of 130 nm particles, the only positive influence of Bragg peak deposit was in

the case of deuteron irradiation. In the case of 30 nm particles, the influence of Bragg peak

deposit was negative in all cases. The best luminescent characteristics were obtained by

proton beam irradiation with the energy deposit without Bragg peak. The creation

efficiency was approximately 10 times higher in the case of larger particles. In the Raman

study, there was systematic increase in the sp2 content (G band) in the samples irradiated

with Bragg peak in comparison to the samples without Bragg peak (Figure 28).

Figure 28: (a) Normalized luminescence intensity of NV- center for various irradiation parameters measured on 35 nm (red columns) and 130 nm (black columns) ND, (b) Raman

spectra of proton irradiated ND show increased sp2 content after irradiation with the Bragg peak (red line) compared to the irradiation without the Bragg peak. The damage is

much lower in the case of 130 nm NDs.

- 60 -

The evolution of the amorphised structure upon thermal annealing depends critically on the

damage density created during irradiation. In regions, where the vacancy density is above a

critical threshold, the diamond crystalline structure is permanently converted to a sp2 –

bonded phase. On the other hand, when the damage density is below the critical threshold,

thermal annealing has the effect of converting the amorphised structure back to the

crystalline diamond phase with the residual point defect formed in the crystal [95]. The

critical threshold in previous studies performed on single crystal diamond was set to the

fluence of 5.1015

cm-2

for the case of Helium irradiation [95], 5.1016

cm-2

for the case of

proton irradiation [96] and 2.1012

cm-2

for the case of nitrogen implantation to IIa

diamond[85].

The presence of high sp2 content in the samples, where the high stopping power at the

Bragg peak created the maximum concentration of vacancies, shows that the critical

threshold for amorphisation was reached in these samples. In this respect it could be

predicted that irradiation with lower fluence, than in our case 9.1015

cm-2

, would lead to

higher creation efficiency, especially in the case of smaller nanodiamond particles.

4.3.2. Formation of NV centres: annealing study

For biological studies, the use of very bright luminescent nanodiamond (ND) particles is

needed. A critical step in preparation of luminescent ND is the annealing procedure. Yet

again, no systematic study was performed on nanodiamond particles. Commonly used

annealing parameters of nanodiamond particles are in the range of 700 – 800 °C for 1 – 2

hours [7][8][9]

2D matrix of annealing conditions, consisting of various combinations of temperatures and

respective times (700 - 950 °C, 0.5 – 8 h). Results are shown in Figure 29 for two sizes of

type-Ib ND particles of 35 and 130 nm in diameter. NDs were irradiated by protons

(approx. 5 MeV). The samples were annealed in argon atmosphere and subsequently

oxidized in mixture of HNO3 and H2SO4 (85 °C, 3 days).

- 61 -

Figure 29: Normalized relative luminescence intensities of NV– and NV0, ratios of NV–/NV0 luminescence, and “%” of sp3 carbons for 35 nm (upper images) and 130 nm (lower images) ND particles as a function of annealing time and temperature. Black dots

- 62 -

represent the matrix of annealing conditions, darker color represents brighter samples. Interestingly, the NV– in 45 nm fNDs are preferentially formed at lower temperatures and shorter times (compare the ratios of NV–/NV0 luminescence). This trend is not pronounced

in 140 nm fNDs.

When comparing data presented at Figure 29a and Figure 29b, we can observe clear

difference in the annealing influence on 35 and 130 nm particles. Higher annealing time

leads to higher creation efficiency of NV centres in 130 nm ND for temperatures below

850 °C. In contrast, 35 nm particles do not show such trend and annealing longer than 1-2

hours does not lead to further increase in the intensity of NV luminescence. Also the trend

of moderate increasing of NV PL with increasing temperature (up to 900 °C) observed in

130 nm particles is not generally observed in the case of 35 nm particles. For 35 nm, there

is a trend of the increase in the NV intensity with increasing temperature ends at

temperatures above 800 °C, except for the sharp discrete maximum observed at 900 °C. If

we consider annealing times, the 130 nm particles behave similarly to single crystal

diamond in which the loss in the GR1 and ND1 absorption and increase in the NV

absorption is monitored even after 20 hours of annealing of irradiated Ib sample [97].

Decrease in the intensity of NV PL after annealing above 900 °C can be explained by the

formation of the NVN center. This explanation is supported by presence of NVN (H3)

center that was monitored in the small fraction (below 15%) of nanodiamonds annealed in

temperatures 900 °C and higher, which is in agreement with studies performed on single

crystal diamond [22]. When we compare the ratio of NV-/NV

0 luminescence in Figure 29,

the NV- are preferentially formed at lower temperatures and lower times. This effect is

more pronounced in 35 nm ND, but the general trend is also observed in 130 nm ND.

We suppose that the increase in the NV PL intensity for longer annealing times for 130 nm

NDs is related to the diffusion and charge state of vacancies. Our assumption is based on

the fact, that in Ib HPHT diamond with high concentration of nitrogen, mainly ND1

(negatively charged vacancy) centers are formed, but the capture of vacancies only occurs

through the motion of GR1 (neutral vacancy). During annealing, the ND1 needs to be first

converted to GR1 which is then mobile [98]. For these reasons, higher annealing times are

more suitable for the NV centres formation in Ib diamond. In the case of 35 nm NDs, the

reason for the decrease in the NV PL intensity for longer annealing time and higher

- 63 -

temperature, could be explained by the diffusion of the vacancies to the surface that would

be reasonably higher for smaller NDs.

NV induced PL of 35 nm particles is on average approximately 10x lower than NV PL of

130 nm particles, when normalized to the Raman signal, which corresponds the volume

fraction, differences between 35 and 130 nm particles. It should be noted that other effects

can be involved in the total luminescent yield, besides the number of created NV centers,

such as the quenching effect of the surface, leading to the “dark state” of NV centres, as

discussed further, which is of importance especially for small particles. However, studies

[85][86] have also used PL as an evaluating method and therefore results presented here,

discussing only optically active NV centers, are fully comparable. The efficacy of

generating of NV centers is in our case about 2% for the conditions used. Highest reached

efficiency of NV creation in our studies was 7%.

The questions still arise at the presence of the discrete maximum of NV PL observed in

samples annealed at 900 °C for 1 hour (900/1). The NV PL intensity of these samples is

two and three times higher in the 130 nm and 35 nm ND respectively in comparison to an

average sample. To confirm the repeatability of this result and to rule out an experimental

error, the NV PL was measured on 20 additionally prepared samples (5 times 900/1, 5

times 700/1 for both, 130 and 35 nm NDs), obtaining similar results.

For very small ND the surface can influence the formation efficiency of NV centers if it is

charged [85]. However this effect is difficult to study as during the irradiation the surface

charges can be fully compensated by incoming ion fluxes.

- 64 -

4.4. CHEMICAL CONTROL OF THE NV LUMINESCENCE IN NANODIAMOND

Here we describe how the surface chemistry effects can make the ND bulk luminescence

sensitive to chemical processes ongoing at the ND surface. The proposed method is based

on the control of electronic chemical potential at the ND surfaces, influencing the surface

band bending, as calculated in the sections 3.1 and 3.2. By that way one can change the

occupation of the luminescent NV centers, that exist in neutral (NV0) or negative charge

states (NV-)

with different PL – zero phonon lie (ZPL) emission wavelength, i.e. 575 nm

for NV0 (ZPL) and 637 nm for NV

- (ZPL).

This mechanism is demonstrated on ND with hydrogenated and oxidized surface which

exhibits important differences in the surface chemical potential. We validate the model and

ND size effect by comparing the results with a defect-free, chemical vapor deposition

(CVD) grown IIa single crystal diamond (SCD) with electronic-grade surface polish (< 0.1

nm) which was implanted to the nm depth with nitrogen and subsequently annealed to

convert N to NV centers by trapping of vacancies (In details described in Chapter 4.3).

Electric filed penetration related and corresponding size effects of ND particles on

luminescence is modeled mathematically in chapter 4.1. We show clearly why the size

effect of nanoparticles is important with comparison to bulk SC CVD diamond [88]Error!

Reference source not found. as this allows a significant increase in the color shift

magnitude.

4.4.1. Quenching of NV- luminescence on ND particles

The above described effect of NV luminescence color shift is especially interesting for

applications to biology. Particles of 20-50 nm size, made of High Pressure High

Temperature (HPHT) synthetic Ib diamond are very suitable for biologic applications due

to the possibility of production of a high amount of stable NV centers. These provide

sufficient luminescence contrast that is comparable to other biomarkers, and retains

advantages related to their nano-dimension for cellular tracking. To investigate the effect

of PL shifts, HPHT ND particles containing approximately 100 ppm of N and with a peak

size distribution at ~ 40 nm and 100 nm were irradiated by protons using energy of 5.4

MeV to produce vacancies and subsequently annealed at to produce NV centers. Raman

spectra taken in each step of the process confirmed high quality of ND, similarly to single

crystal diamond, showing no visible non-diamond sp2 bonds that could negatively

- 65 -

influence the surface charge interactions (i.e. reduce the sensitivity of the surface

termination to the surface chemical potential as is the case for detonation ND) . The

measurements below were executed in a biological buffer solution (pH 7). Figure 30a

shows the experimental data obtained for 40 nm ND for 3 different situations: oxidized

surface, hydrogenated surface and diamond after subsequent oxidation performed by

annealing at 400°C in air. According to our model, the NV- luminescence with ZPL at 638

nm is fully quenched after hydrogen termination, while the NV0 luminescence is still

visible. The NV- luminescence can be restored again by annealing in air (leading to

oxidized surfaces). Further on, Figure 30b shows the PL spectra taken at room temperature

of H-terminated ND upon heating in air at gradually elevated temperature. A reverse

process, i.e. backward transition from NV0 luminescence to NV

- dominated luminescence

can be generated upon heating (Figure 30b). The first change occurs at about 200°C. This

temperature agrees well with the experimentally observable desorption of adsorbates [83]

confirmed by the loss of the 2DHG surface conductivity (i.e. band bending). At 400°C

permanent changes in the surface termination occur leading to the loss of surface hydrogen

due to oxidation, leading NV- to dominate the PL spectra.

Figure 30: (a) Changes in NV- and NV0 luminescence induced by various terminations of 40 nm sized ND particles. Spectra of oxidized, hydrogenated, and annealed surfaces at

400°C (leading to restoration of original surface termination). The hydrogenation of larger particles (~40 nm) resulted in luminescence shift towards NV0 luminescence

(hydrogenated 1), for the smaller particles (< 20 nm), the luminescence quenched completely (hydrogenated 2). (b) Luminescence changes of hydrogenated ND (40 nm) upon annealing in air at different temperatures. Samples were heated to the target temperature (using a ramp of 25°C per minute), kept at the set temperature for 30

- 66 -

minutes and then cooled down to the room temperature. All spectra were taken at room temperature and baseline corrected.

In addition we observe a large influence of ND size on the PL intensity of both NV0 and

NV- defect centers. Figure 31 shows the size effect for hydrogenated ND (Figure 31). The

size of ND was tuned by extraction using centrifugation. Both NV- and NV

0 centers came

from the same irradiation badge and the PL intensity is normalized to Raman line signal in

each ND. For very small particles, even the NV0 luminescence can be strongly reduced.

For larger particles, the NV-/NV

0 ratio is altered by the ND size. This is in accordance to

the model presented in the section 4.1. Biomedical applications of NDs require specific

particle size, which is for example different for endosomatic penetration of fNDs through

cellular membrane as for a penetration of fND to cell nucleus for which ultra-small fND is

required. Size effects are therefore of importance especially for very small particles for

biological applications.

Figure 31: (c) PL spectra of hydrogenated ND of various size (20-100 nm) measured in physiological solution (pH = 7) showing the ND size effect on the PL changes in

hydrogenated/oxidized diamond and the possibility to monitor PL changes in liquid which is essential for biosensor work. The NV- to NV0 ratio could be tuned with the size of ND. All

spectra were normalized to diamond Raman peak and water Raman background was subtracted. d.-g. Confocal image of single ND particles upon various treatments. For

oxidized ND, both NV0 (d) and NV- (e) luminescence is clearly visible in contrast to hydrogenated ND, where NV0 luminescence (f) is dominant and NV- luminescence (g) is barely visible. Confocal PL images correspond to luminescence collected from 570 nm to

610 nm spectral range for NV0 and from 630 nm to 750 nm for NV-.

- 67 -

The luminescence of individual particles as well as measurements of ND PL in cells were

performed by a confocal microscope Olympus FV-1000, excitation wavelength 553 nm,

laser power 15 mW. Raman and luminescence spectra were measured at Renishaw InVia

Raman Microscope; excitation wavelength was 514 nm with 25 mW laser power. Spectra

were taken at room temperature and normalized to the diamond Raman peak.

The measurements in liquid were performed in Hellma fluorescence cuvette (type no.

105.252-QS) in aqueous solution (0.2 mg in 1 ml). Size selection was performed by

centrifugation: Aqueous colloids were made from particles by immersion 1 mg of ND in

200 ml of deionized water and dispersing with a high-power ultrasonic horn (Hielscher

UP400S, Sonotrode H3) using 400 W at a 1:1 (on/off) cycle for 2 hours under liquid

cooling. The temperature of the solution was below 25 °C. Solution was centrifuged for 60

minutes (rcf: 14,000 x g). The solution was divided into three fractions with peak

distribution around 20 nm, 50 nm and 100 nm, as confirmed by AFM and DLS.

For measurements of luminescence changes upon consecutive annealing of hydrogenated

ND in air, samples were heated to the target temperature (using a ramp of 25°C per

minute), kept at the set temperature for 30 minutes and then cooled down to the room

temperature. At each step the luminescence spectra were measured at room temperature.

4.4.2. Fluorinated nanodiamonds

The difference in the PL properties of hydrogen and oxygen terminated surface is the

consequence of different electron affinity of hydrogenated and oxidized surfaces.

Hydrogen terminated surface has highly negative electron affinity and hence the upwards

band bending, the positive electron affinity of oxygen terminated surface leads to

downward band bending that leads to stabilizing of the luminescence. Since the variation

in electron affinity is closely connected to the surface electric dipoles that are caused by

different electro negativities of elements terminating the surface, one could expect that by

introducing highly electronegative element, such as fluorine, to the surface, the quenching

effect of the surface on the luminescence could be reduced.

Besides the possible impact of fluorination on the luminescence properties, fluorinated

nanodiamonds could be also used as a probe for NMR imaging, and have high impact on

- 68 -

novel nucleic acid grafting strategies. Another application is a use of fluorinated ND in the

development of highly hydrophobic ND that are required in specific drug delivery systems.

Although there are specific plasma techniques of fluorination, these techniques were

mainly applied to flat CVD diamond plates. To achieve high surface coverage also on

HPHT NDs, original method was developed, based on the reaction with gaseous fluorine in

the presence of hydrogen was modified to the reaction process with the gaseous fluorine in

liquid hydrogen fluoride at elevated temperatures. The reaction scheme of fluorination is

shown in Figure 32.

Figure 32: Reaction scheme of fluorination

This procedure resulted in the high surface coverage density fluorinated HPHT

nanodiamonds. According to elementary analysis, the content of fluorine was 1,7%, which

corresponds to the atomic coverage above 37%. The presence of fluorine was further

proved by FTIR spectroscopy and 19

F NMR spectroscopy. In order to address the basic

mechanism of the process of fluorination and to examine the optical properties of mixed

fluorine-hydrogen and fluorine-oxygen terminated surface, the oxidized and hydrogen

plasma-reduced nanodiamonds were used as a primarily source for fluorination. The final

density of the fluorine substitution of oxidised ND was double (37 %) in the comparison to

hydrogenated ND.

4.4.3. Size dependence of the luminescence of nanodiamonds

Size dependence of NV luminescence on variously terminated ND is shown in the

following section. Several fractions of variously terminated fND were separated by

centrifugation from the original polydisperzed ND solution, gaining fNDs of the size: 8-10

nm, 10-20 nm, 20-50 nm, and >100 nm. We were using functionalisation of the surface to

prevent aggregation of ND in aqueous solution, leading to stable dispersions for ND within

- 69 -

all size ranges used in the experiment. A strong size dependence of PL properties (Figure

13) can be found. In the case of hydrogenated NDs, the luminescence of H-terminated ND

quenched completely for 8-10 nm NDs, the 15-20 nm NDs showed clear NV0 related

luminescence with no NV- luminescence, for larger ND NV

-/NV

0 ratio could be precisely

tuned. The PL of oxidised ND was found to be less sensitive to size of ND, but changes in

NV-/NV

0 ratio were also observed, with the favour to NV

0 luminescence for smaller ND.

The NV- luminescence was still dominant in all oxidised samples. PL of fluorinated ND

was found to be the most stable in all size ranges.

These results showed in Figure 33, point towards the possibility of precise tuning of the

NV-/NV

0 ratio as well as the possibility of the conversion of NV centres to the dark state.

As discussed above, the charge state of NV centres depends on the position of the Fermi

level with respect to the energy level of the defect center that defines the energy at which

the defect takes up or loses an electron. If the Fermi level is shifted above the energy level

of the defect, an electron is taken up by the defect that results in the charge switching-off

the defect center. Figure 34 shows results of the modeling for the case of spherical

nanodiamond particle with marked values of the sizes of NDs used in the experiment for

the case of hydrogenated surface that provides an acceptor electronic state for the bulk

electrons, resulting in the surface band bending. The creation of dark state for particle sizes

of about 8 nm (i.e. double size of the band bending decay) is clearly evident as modeled in

Figure 34.

- 70 -

Figure 33: Luminescence intensity of variously terminated ND as a function of size. The fluor/Ox represent nanodiamonds that were fluorinated after oxidization leading to

partially oxidized/partially fluorinated surface, the Fluor/Hydro represents NDs where hydrogenated surface (a) The peak intensity of the NV- and NV0 zero phonon line (b)

Corresponding photoluminescence spectra.

.

Figure 34: Energy levels of NV0 and NV- centers in the hydrogenated spherical ND particles showing the relative position to the Fermi level as a function of radius, modeled

- 71 -

as described above. The red lines represent the mean size distributions of ND used in the experiment.

One can also clearly see that fluorination of the oxidized ND does not have are quenching

effect on the NV0 or NV

- luminescence for small particles. This can be clearly explained

by the downward band bending, i.e., fixing the EF below the NV0 states. Also, when

compared fluorinated ND with oxidized ND, ND partially covered with fluorine exhibit

higher NV- luminescence. The intensity of NV

- emission of fluorinated ND was in average

20% more intense than the emission of oxidized ND. Fluorination of hydrogenated NDs

lead to the surface partially covered by fluorine and partially by hydrogen. Surface

coverage by fluorine was in this case 23%, which indicates a great variety in the surface

electron affinity [99]. This fact we attribute to the surface stabilization of the fluorinated

nanodiamonds, allowing diminution of local variation of surface electric fields. This brings

a possibility of development of ND sensors, sensitive to both positive and negative charge

fields. This aspect of the detection principle is addressed in the next chapter.

- 72 -

4.5. SENSING OF CHARGED MOLECULES VIA NV LUMINESCENCE

In previous chapters, it was shown that luminescence from NV centres can be influenced

by surface termination. This effect is related to changes in the electro-chemical potential of

the surface that enables manipulation of the Fermi level in the crystals. The NV charge

switching was demonstrated by alternation the surface termination (hydrogen, oxygen,

fluorine). It means that the event that shifted the Fermi level was covalent bonding of

hydrogen or oxygen groups to the surface.

To make use of this principle for detection of biological processes, the alternation of the

surface chemical potential should be driven by conditions/events that can occur in

biological environment. One of the possible alternatives is a non-covalent binding of

charged biomolecules to the surface of nanodiamond particle.

It is now well-recognized that the surfaces of biomaterials (e.g., implants and medical

devices) are immediately covered by biomolecules (e.g., proteins, natural organic

materials, detergents, and enzymes) when they come in contact with a biological medium.

The absorption of biomolecules to such surfaces determines the subsequent cellular/tissue

responses. Due to their extremely high surface to volume ratio, nanoparticles (NPs) in

general have a very active surface chemistry in comparison to bulk biomaterials; hence, in

biological applications they tend to reduce their large surface energy by interaction with

the medium components in which they are dispersed. Thus, the dispersion of NPs in a

polymer-containing medium results in their surfaces being covered by a complex layer of

polymers.Error! Reference source not found.

The switching between charge states of NV centres is related to the position of the Fermi

level with respect to the energy level of NV centre. This effect can be enhanced by charged

molecules or particles that are closely bound to the surface of nanodiamond. Examples are

charged polymers, attached to the diamond surface. If the charged molecule is strongly

attracted to the surface, the charge transfer can occur, depending on the HOMO (highest

occupied molecular orbital) and LUMO (lowest unoccupied molecular orbital) energetic

levels of the charged molecule.

- 73 -

The principles of charge-interaction of molecules in the close proximity of the diamond

can be demonstrated on the model system with well-defined conditions. As a model system

we chose solution of nanodiamonds with four types of charged polymers. As positively

charged molecules, poly diallyldimethyl ammonium chloride (PDADMAC) and

polyallylamine (PAA.HCl) were chosen. As negatively charged molecules, the polyacrylic

acid sodium salt (PANa) and the polystyrene sulfonic acid sodium salt (PSSNa) were used.

Polymers were purchased from Sigma-Aldrich (Prague, Czech Republic). Polymers were

diluted in DI water to 31.2 mM concentration and added to the colloidal dispersion of ND

in DI water (1 mg/1 ml) in the ratio 1:9 (vol:vol). Solution was sonicated in the ultrasonic

bath for 20 minutes and left over night. For the purpose of the zeta-potential

measurements, the ND-polymer solution was further diluted to achieve final concentration

of nanodiamonds 0.1 mg/ml. The interaction between polymers and ND is schematically

shown in Figure 35.

Figure 35: The electric field formed in the close surface proximity of ND after the interaction with charged polymers.

It was found that most sensitive to the charged interactions are fluorinated ND, showing

significant decrease in NV- PL after the addition of positively charged polymers (Figure

33, Figure 35).

- 74 -

Figure 36: PL spectra showing interaction of oxidized and fluorinated ND with negatively (left) and positively (right) charged polymers. Clear decrease of NV- luminescence was

observed for F-fND after attachment of positively charged polymers.

Figure 36and Figure 37 show the changes in the PL spectra of fluorinated and oxidised ND

upon interactions with two different kinds of charged polymer. The luminescence of NV-

centres clearly decreased upon the interaction with positively charged molecules, while

after addition of negatively charged polymers, the luminescence was restored to the

original level.

Figure 37: PL spectra showing interaction of oxidized (left) and fluorinated (right) ND with negatively and positively charged polymers. Clear decrease of NV- luminescence was

observed for oxidized and fluorinated ND after attachment of positively charged polymers. All spectra are normalized and the background water and polymer spectra were

subtracted.

- 75 -

The presented results show for the first time the ability of fluorescence nanodiamonds to

interact with charged polymers that leads to the changes in the NV luminescence. These

results bring new possibilities for the biomolecular detection and opens a variety of new

possible applications of fluorescence nanodiamonds. However, further verification is

necessery to exclude the possible quenching effects of polymers itself and to clearly proof

that the changes in the NV PL are related to the charge of interacting polymers. These

proofs are given in the next chapter.

4.5.1. pH dependence – charge switching

When we discuss surface charge interaction, pH is one of the necessary aspects that need to

be discussed. pH indicates the concentration of the solvated hydrogen ions (protons) and

gives information about deprotonation of the solvent. If the pH is neutral, the concentration

of cationic hydrogen ions and anionic hydroxide ions are in balance. In simple terms, the

lower pH is indicated (pH < 7), the more positively charged ions can interact with the

surface. And opposite, the higher pH, the higher fraction of negatively charged OH- ions is

present.

The design of the experiments with charged polymers allowed us to monitor the charge

switching of the polymer with changes in the pH. The charge of the polyallylamine and

polyacrylic acid sodium salt is tunable by pH, while poly diallyldimethyl ammonium

chloride and polystyrene sulfonic acid sodium salt are strongly cathionic/anionic in the

whole scale of pH.

Figure 38 show the pH dependence of the NV PL. We can observe reduction in the NV-

PL (638 nm ZPL) for lower pH as the concentration of the hydrogen atoms increase. The

NV- PL fully disappeared for the pH of 2,56 showing the sensitivity to the intensive

deprotonation of the ND surrounding.

- 76 -

Figure 38: Pl spectra of oxidized nanodiamonds as a function of pH. The NV- luminescence is reduced in acidic pH.

The same experiments were performed on the polymer-coated nanodiamonds. The samples

of various pH were prepared separately. The final value of pH was measured prior the PL

measurement to prevent errors in the pH value.

Figure 39 shows the pH dependence of the ND coated with positively charged polymers,

polyallylamine (PAAHCl) and poly diallyldimethyl ammonium chloride (PDADMAC).

We can clearly see a difference in the luminescence behavior of nanodiamonds coated with

these two polymers. While the luminescence of the PDADMAC coated NDs does not

change significantly with pH and the NV- luminescence is quenched in the whole range of

pH, the luminescence of PAAHCl coated NDs varies with increasing pH. This corresponds

to the predicted behavior as the PDADMAC polymer is fully deprotonized in the whole

spectral range, however the positive charge of the PAAHCl is reduced in higher pH.

- 77 -

Figure 39: PL spectra of oxidized nanodiamonds coated with positively charged polymers. While NDs coated with strongly cationic polymer (PDADMAC) remain with quenched NV- luminescence in the whole pH scale, PL of NDs coated with PAAHCl that loses the positive charge in the pH above 8 is changing. The NV- luminescence is quenched for the pH below

8,8 and starts to appear again for higher pH.

As shown in Figure 40 the luminescence behavior of the NDs coated with negatively

charged polymers is similar to spectra taken on oxidized nanodiamonds. These results also

clearly support our theory as the deprotonation of the solution is given by pH, the negative

charge, carried by polymers, is fully compensated by hydrogen atoms. However, the

PSSNa coated nanodiamonds exhibited lower sensitivity to the pH change. Both, NV0 and

NV- luminescence remain visible up to pH of 3.5. This is further explained by the zeta

potential measurements.

- 78 -

Figure 40: PL spectra of oxidized nanodiamonds coated with negatively charged polymers.

NV- to NV

0 ratio was calculated from all spectra using a deconvolution of the signal to the

typical spectra of the NV0 and the NV

- luminescence. The summarized graph is shown in

Figure 41. Binding of charged polymers to nanodiamonds as well as charge switching of

polymers by pH was monitored by the Zeta potential measurement. The data are shown in

Figure 42.

When we compare the values plotted in Figure 41 with the zeta potential of examined

samples, there is a clear correspondence in the trend of the data.

- 79 -

Figure 41: Summary of the NV- to NV0 ratio of PL of polymer-coated nanodiamonds as a function of pH.

Figure 42: The zeta potential of polymer coated NDs as a function of pH. Zeta potential show clearly adsorption of the positively/negatively charged polymers on the surface of

nanodiamonds.

When we discuss the correspondence between the zeta potential and the observed

luminescence changes, we need to understand the principles of the zeta potential

measurements as well as the mechanism of the charge transfer from polymers to diamond

that is connected to the spectral PL changes.

- 80 -

Figure 43 schematically show the potential distribution of nanoparticle. Further description

of this method is presented in the Chapter 3.1.2.

Figure 43: Schematic illustration of potential of a nanoparticle. Figure based on [58].

While the Zeta potential measurement gives information about the potential at the edge of

the electrical double layer, the surface charge transfer can occur only at the very surface of

the particle. It means that the measured zeta potential give us only limited view on the

surface potential itself and give us only an approximate feeling of the charge processes

ongoing on the surface.

- 81 -

4.6. STUDY OF THE CELLULAR UPTAKE

The aim of our study is to use fluorescent nanodiamonds (FNDs) as an in-cell operating

biosensor. One of the key tasks is therefore the study of mechanism of the nanodiamonds

cellular uptake. In general, there are several possible ways of penetration the cell

membrane: i) passive diffusion through membrane, ii) targeted penetration through protein

channels, and iii) penetration via endocytosis. As FNDs size is ranging from a few

nanometers to hundreds of nanometers, the most of particles are taken up by clathrin-

mediated endocytosis [13] or phagocytosis, in the case of macrophage cells. If the surface

of ND is specifically functionalized to target specific protein channels, NDs are taken up

directly by protein channels. [14]. Due to,FNDs size, the passive diffusion is the most

unlikely mechanism of ND cell uptake and no study has reported this transport of NDs to

cells.

4.6.1. Oxidized nanodiamonds

The incorporation of oxidized FNDs was tested with IC21 macrophage and HT29

colorectal adenocarcinoma cell lines or peripheral blood mononuclear cells (PBMC). As in

previous experiments, two different size fractions of NDs were used: 130 nm FNDs and 35

nm FNDs. The cells were cultivated for 24 hours under standard conditions. To prevent

aggregation, FNDs were added in serum-free medium. Figure 44 shows an intake of 130

nm particles in the IC21 macrophage cells. FNDs of this large size are strongly

incorporated in cells. This data clearly show that most of the FNDs are incorporated and

only a small fraction of FNDs was observed in cell medium. Similar results were obtained

using 35 nm FNDs. Incorporation of smaller NDs is observed even 30 minutes after

incubation, which indicates a great ability of the macrophage cell to uptake nanodiamonds.

- 82 -

Figure 44: IC21 macrophage cells with uptaken FND particles (40x). (a) transmission figure, (b) fluorescence imaging showing PL of NDs, (c) combination of (a) and (b). IC21 macrophage cell line strongly phagocytes FND particles. The cells were incubated for 24 hours in H-MEMd medium containing 25ug/ml of 100 nm FNDs. Particle incorporation is

visible as soon as after 5 hours of incubation.

As the aim of this work is the use of nanodiamonds for the design of probes for monitoring

of cancer-related processes, the influence on leukocytes (PBMC) was also tested.

Figure 45: PBMC with FNDs. The image shows PBMC incubated with FND sample for 24 hours. It is clearly visible that nanoparticles were not taken up by lymphocytes. The only

uptake is visible in granular cells (probably macrophages and APCs).

From the results shown in Figure 45 is clear that lymphocytes were not affected and do not

uptake nanodiamonds. Granular cells (macrophages and APCs) did however incorporate a

detectable amount of FNDs. This is an expectable result, as particularly macrophages are

prone to ingestion of foreign particles, such as nanodiamonds. These results were obtained

on both, 130 and 35 nm particles.

Unlike the previously studied cells, the HT29 cell line showed only limited uptake ability

for 130 nm particles. Figure 46 shows cells after 5 and 24 hours of incubation. This

- 83 -

relatively low tendency to incorporate FNDs possess a problem that led us to the necessity

of finding a suitable chemical structure to bind onto the ND surface to target the particles

to the cancer cells and increase their uptake. Our preliminary results on chemically

functionalized FNDs indicate that bombesin is particularly suitable for improving cellular

uptake of FNDS into cancerous cells.

Figure 46: FND intake by HT29 adenocarcinoma cell line. While the IC21 cell line intensely uptakes nanoparticles even after short incubation, the HT29 adenocarcinoma cell line

takes at least 24 hours to show significant FND uptake. The right figure (b) shows HT29 cell colony after 5h incubation with 130 nm FNDs. It is clear that there are no

nanodiamonds within the cell. After 24 hours (a), the FND particles were detected inside the HT29 colony (localization proved by confocal 3D scan).

Comparatively higher incorporation of the HT29 cells was observed for 35 nm particles

(Figure 47), where after 24 hours of incubation, nanodiamonds were observed in 70 % of

cells.

Figure 47: FND, with the size of 35 nm, intake by HT29 adenocarcinoma cell line. The right figure (b) shows HT29 cell colony after 5h incubation with 50 nm FNDs. After 24 hours (a),

the FND particles were detected inside the HT29 colony.

- 84 -

The apoptosis and proliferation was examined on all samples. Nanodiamonds did not

influence the cell viability and therefore can be considered as suitable material for a cell

delivery and detection probe applications.

Cell culture

Human colorectal carcinoma cell line HT-29 was obtained from American Type Culture

Collection (HTB-38; Manassas, VA). Cell culture were maintained in RPMI 1640 media

supplemented with 10% fetal bovine serum and antibiotics (Biosource, Camarillo, CA) in

5% CO2 and 37°C. Cells were seeded in appropriate amount 18 hours before the

experiment. Final concentration of nanodiamond particles alone or in combination with

PEI800 was 25µg/ml.

4.6.2. Polymer-coated nanodiamonds

As described further in Chapter 4.7, nanodiamonds are aimed to use as a transfection

system. NDs were coated by polyethylenimine (PEI), a polymeric transfection enzyme,

PEI condenses DNA into positively charged particles, which bind to anionic cell surface

residues and are brought into the cell via endocytosis. PEI with various molecular weight

(800, 1 800 and 60 000 g/mol per unit) was investigated in this study.

In first part of our study, we have tested effect of PEI coating on cellular uptake of NDs.

We have observed that PEI of lower molecular weight (800 and 1800 g/mol) bound to the

surface of nanodiamond increased the cellular uptake, when the most of the cells

(approximately 95 %) contained nanodiamonds. Figure 48 clearly show an improved

incorporation of nanodiamonds coated with PEI 800 in HT29 cells. Similar results were

obtained on ND coated with PEI 1 800. ND coated with PEI 60 000 showed increased cell

toxicity (see Chapter 4.7.3 for more detailed description) so the cellular uptake was not

monitored on these samples.

The incorporation of nanodiamonds was studied also on NDs coated with PEI and DNA.

The cellular uptake on the ND-PEI-DNA complexes was comparable to the one observed

on plain ND-PEI coated samples as is clearly visible from Figure 48.

- 85 -

Figure 48: FNDs coated with PEI800 (a-d) and PEI800 and DNA (e-h) intake by HT29 cell line. a) bright field; b) FNDs fluorescence; c) Hoechst stained nuclei; d) overlaid. Figure

show improved cellular uptake after coating with PEI.

HT-29 cells seeded on glass bottomed 6-well plate (In Vitro Scientific, USA) were

incubated with nanodiamond particles alone or in different combinations with PEI800 and

beta-actin cDNA for 24 hours. Images were recorded by confocal microscope Olympus

FV1000 SIM (objective 40X/0.95 and up to 3x digital zoom) and analyzed with Olympus

FLUOVIEW 2.0a software. Cell nuclei were stained with Alexa Fluor 405 and detected

with filters: excitation 405nm and emission 422nm; nanodiamond particles were detected

using filters: excitation 559 and emission 619nm.

In conclusion, cellular uptake of NDs by cells that are not specialized on attacking of

foreign substances is lower than for macrophages. There are several possible ways to

improve the relatively low uptake observed in HT29 adenocarcinoma cell line on untreated

(oxidized) nanodiamonds. Here we presented a simple method of coating nanodiamonds

with PEI that is known to improve the ability of cellular uptake. However, this method is

not specific, i.e. PEI coated NDs are likely to be incorporated to various cells. Other

possibility is a surface functionalization by specific structures for targeting the cancer cells,

such as functionalization by bombezine. Our preliminary results (data not shown) indicate

an increase uptake of bombazine coated nanodiamonds in HT29 cells.

- 86 -

4.7. NANODIAMOND TRANSFECTION SYSTEM ENABLING DETECTION OF

DNA DELIVERY

In this section we discuss the possibility of using ND particles as active biological sensor,

based on the luminescence changes induced by transfer doping. One of the highlights in

applications of NDs in medicine is the use of NDs for drug delivery systems. Recent

studies performed by the group of prof. Dean Ho showed that when doxorubicin was

bound to detonation nanodiamonds, the toxicity of such complexes was dramatically

reduced compared to unmodified doxorubicin [50]. The same group recently showed the

ability of ND to deliver therapeutic nucleic acids.

There are several approaches of the DNA/RNA delivery to cells. In general, the methods

can be divided to two basic approaches: viral and non-viral methods. Viral delivery is the

more conventional approach because viruses have evolved to infect cells with high

efficacy. However, some clinical studies showed safety risks such as cancer or death. Non-

viral transfection systems are using liposomes and polymers that bond the DNA and

delivery it through the cell membrane. These systems are easily produced and reduce the

risk of cytotoxicity, however their great disadvantage is the low transfection efficiency.

The usage of hybrid polymer-nanoparticle materials was reported to enhance the

transfection efficiency Error! Reference source not found., where the nanodiamond

particles showed up to be the perfect candidate for such transfection systems [101] as being

highly biocompatible, chemically stable and easily adaptable for biomolecular attachment

[15]. Polyethylenimine (PEI) coated ND showed increased efficiency of siRNA

transfection in comparison with plain PEI of the same molecular weight, while remaining

biocompatible [101]. The other advantage of ND particles is the content of stable

luminescence centres (NV centers) that allow the long term optical tracking of the ND

delivery platform. Here we show the construction of such transfection system and the

ability of this system to monitor the binding of the transfection polymer and DNA to the

ND particle that enables us to optically monitor the successful gene delivery.

Based on the possibility of charged transfer from electrostatically bounded polymers, one

can use as well PEI (that is cationic polymer) for such charge interactions and develop

optical sensors for subsequent monitoring of the DNA attachment/release. This sensor is

- 87 -

based on a simple fact that the DNA or RNA molecules are strongly negatively charged

and usage in combination with the PEI (positively charged) would lead to reversible charge

switching upon attachment/release of nucleic acids.

Figure 49: The schematic figure of the formation of ND-PEI-DNA complex. This process is based on electrostatic interactions

4.7.1. Monitoring of the formation of the ND-transfection system

Figure 50 shows chemical reaction pathway for attachment of PEI, a cationic polymer used

as a transfection agent. The positive charged PEI condenses the DNA into a complex that

can be taken up by the cell via endocytosis. The use of ND makes the transfection

mechanism more effective, which allows the use of less cytotoxic PEI of lower molecular

weight. The formation of PEI-ND complexes is based on the electrostatic attraction of

positively charged PEI and oxidized ND surface. Polycationic NDs were successfully used

for the grafting of oligonucleotides (see Figure 50). Cationic transfection polymer,

polyethylenimine (PEI), was purchased in three molecular weights, 800, 1800 and 60,000

from Sigma-Aldrich (Prague, Czech Republic). 2 mM PEI was mixed with ND (2 mg/ml)

in the ration 1:1 (vol:vol), sonicated using high-power sonication horn (Hielscher UP400S,

Sonotrode H3) using 400 W at a 1:1 (on/off) cycle for 2 hours under liquid cooling.

- 88 -

DNA isolated from a tail of DBA-2 mouse was obtained from laboratory depository. 7µg

of DNA per 100µl PCR reaction was amplified using HotStarTaqDNA Polymerase

(Qiagen) and iCycler5 (BioRad). Primers for beta-actin used in the reaction were

following: forward 5´agagggaaatcgtgcgtgac 3´, reverse 5´ acggccaggtcatcactattg 3´and

resulted in the amplification of 137bp product. PCR product was purified using QIAquick

PCR purification Kit (Qiagen), eluted into water and the specifity was checked by agarose

electrophoresis. 160ng of DNA was homogenized by sonication in water bath for 1hour

and then combined with a complex of nanodiamond ND-PEI800 (12µg ND) following by

further sonication for two hours.

Figure 50: Achieved non covalent DNA grafting used for design of the fNDPbiosensors operating contacless in cell.

The prepared PEI coated and DNA coated NDs were used with expectation that alternation

of differently charged analytes will results in changes of the NV0/NV

- photo luminescence

ratio. These experiments were performed in cell medium. As expected, the presence of

positively charged PEI led to decreasing or quenching of NV- photoluminescence.

Subsequent addition of negatively charged DNA compensated the positive charge, which

resulted in enhancement or restoration of NV- luminescence. Figure 51 shows a strong

increase of luminescence after DNA attachment on PEI coated ND (Blue curve) compared

to ND with PEI (red curve).

- 89 -

Figure 51: PL spectra of fluorescent ND with various surface terminations (black curve) after binding of PEI (red curve) and DNA (blue curve). All samples except for hydrogenated ND show significant decrease in the NV- luminescence after addition of positively charged

PEI (red). The level of the NV- luminescence increases again with the binding of the negatively charged DNA that compensates the positive charge of the PEI. Large

differences in the NV- luminescence of the fluorinated/hydrogenated ND coated with PEI could be used to monitor the interaction of the particles with strongly negative charged

biopolymers like DNA inside cells.

These first experiments that used variously terminated NDs showed a great advantage of

the partially fluorinated and partially hydrogenated NDs. Due to the mixture of various

affinities at the surface, these particles allowed to distinguish between all three states: plain

NDs, NDs coated with PEI, and NDs coated with PEI and DNA. NDs with other surface

termination showed only “two-states” detection without the possibility to differentiate the

starting untreated material. However, the disadvantages that are related to this surface

treatment, mainly the technological difficulties in the process of fluorination of

hydrogenated surface, but also the hydrophobicity of this surface that lead to the poor

colloidal stability in water or cell medium, forced us to use the well-established oxidized

- 90 -

nanodiamonds for further experiments. We are trying to overcome these difficulties by

developing new techniques of this surface treatment and by the development of additional

surface treatment that would lead to better colloidal stability in water as well as cell

medium. The discovered properties of fluorinated/hydrogenated NDs give further

perspectives for the use of nanodiamonds as an in-cell operating biosensor.

The aggregation of the ND-PEI complex was the main drawback we had to deal with. One

of the options was to use higher molecular weight PEI (instead of previously used PEI800).

The PEI of a molecular weight of 1800 and 60.000 was used. PL spectra and zeta potential

were measured in each step. Even though the use of PEI60000 improved the colloidal

stability, the solution tend to aggregate after the addition of the DNA and the detection

ability of the ND-PEI60000 complex was limited compared to the ND-PEI800 and ND-

PEI1800 as shown in Figure 52.

- 91 -

Figure 52: PL spectra of oxidized nanodiamonds (black), nanodiamonds after addition of PEI800, PEI1800 and PEI60000 (red), and DNA (blue) showing clear decrease in the NV- luminescence after addition of PEI of various molecular weight. The NV- luminescence

quenching (or the shift of the luminescence towards NV0) is more pronounced for higher molecular weight PEI (1800 and 60000). The NV- luminescence is restored to the original

level after the DNA addition for the case of ND-PEI800 and ND-PEI1800 complexes, showing the full compensation of the polymer’s positive charge by the negatively charged

DNA. The sample using ND-PEI60000 also showed an increase in the NV- luminescence after the DNA binding, however, the difference is not as significant in comparison with

lower molecular weight PEI. This indicates only partial charge compensation. The charge differences were monitored by the zeta potential measurements (see Supporting

information) and correspond to the changes in the NV luminescence observed in all samples.

The improved colloidal stability of the ND-PEI800 and PEI1800 was reached by the use of

carboxyl-functionalized NDs that were obtained by the H2SO4 and KNO3 oxidation

procedure (see section 4.3). However, the zeta potential measurements showed that the

long term stability was not achieved and the dispersion tend to aggregate in several days.

The addition of the DNA led, after brief sonication, to the stable water solution that could

be further used for cell experiments.

4.7.2. Detection of the DNA release in cells

In previous section, we have demonstrated the possibility of optical detection of the

formation of the nanodiamond-mediated transfection system based on the electrostatic

interactions between charged biomolecules. The main goal of this thesis is the

demonstration of the DNA delivery to cells and monitoring of the DNA release. Samples

that showed the best sensitivity to the DNA attachment were used for the cell experiments.

- 92 -

Figure 53 show the PL spectra taken from the PEI and DNA coated NDs inside of cells and

outside of cells and the spectra of the plain, oxidized NDs under the same conditions.

Figure 53: The PL of PEI and DNA coated nanodiamonds inside (red) and outside (black) of cells. Spectra were taken after 1 hour of incubation, when some of the nanodiamonds

were taken up by the cells while the others remained in the medium. Spectra measured outside of the cells were measured in the cell medium; spectra showing the luminescence of nanodiamonds inside of the cells were measured on nanodiamonds localized in cells.

The average spectra of 10 measurements are displayed. The biggest relative difference in the luminescence was observed for nanodiamonds coated with PEI1800 and DNA.

If we take the typical spectra of NV0 and NV

- center with their phonon replicas (as for

example in Figure 6), we can deconvolute the signal, model the behavior and calculate the

- 93 -

relative decrease in the NV- luminescence as well as a possible increase in the NV

0 related

PL. This number should correspond to the changes observed using band pass filter on the

confocal microscope. The calculated relative change of the NV- and NV

0 PL is in Figure

54.

Figure 54: Relative changes in the NV-/NV0 inside and outside of the cells.

When we compare the data presented in the Figure 52 and Figure 53, we would expect

higher relative decrease in the NV- luminescence if the DNA would be fully released from

the ND-PEI complex. The relatively low increase in the NV- luminescence could be

caused by two main factors: i) some of the DNA was not released from the ND-PEI

complex, ii) other molecules were attached to the surface of the ND, which resulted in the

compensation of the positive charge of PEI. As shown in the section 4.5.1 the effect of pH

could be neglected in this pH region (6,5 to 7,2).

Figure 55: The PL of PEI coated nanodiamonds inside (red) and outside (black) of cells.

Spectra were taken after 1 hour of incubation. Spectra measured outside of the cells were

measured in the cell medium; spectra showing the luminescence of nanodiamonds inside of

the cells were measured on nanodiamonds localized in cells. The average spectra of 10

measurements are displayed. The spectra show the stability of the ND-PEI complex in cell

- 94 -

medium and inside of the cell.shows the PL spectra of nanodiamonds coated with PEI

outside and inside of cells and the relative change in the NV PL. The low relative change

(that is comparable with the relative change observed for plain oxidized ND) can neglect

the effect of other attached molecules on the PL changes and shows the stability of the

PEI-ND complex in the cell environment.

Figure 55: The PL of PEI coated nanodiamonds inside (red) and outside (black) of cells. Spectra were taken after 1 hour of incubation. Spectra measured outside of the cells were measured in the cell medium; spectra showing the luminescence of nanodiamonds inside of the cells were measured on nanodiamonds localized in cells. The average spectra of 10 measurements are displayed. The spectra show the stability of the ND-PEI complex in cell

medium and inside of the cell.

4.7.3. Biocompatibility of the ND-transfection system

Biological experiments were performed at Microbiological Institute, AS CR by Veronika

Benson, Jan Richter, and Anna Fiserova.

Cell proliferation assay

5x 103 cells were seeded in triplicates on a 96-well plate and incubated with nanodiamond

particles alone or in a combination with PEI800 and /or DNA. 1% SDS served as a

negative control of proliferation. After 22 hours of incubation, WST-1 reagent (Roche) was

added to the cells, plate was incubated in 37°C for 2 hours and absorbance was measured

using a reader Tecan Spectra (Schoeller Instruments) and filters 450 and 630nm.

- 95 -

All samples showed excellent proliferation comparable to the control sample (KO) except

for the sample treated with ND coated with high molecular weight PEI, PEI600000 that is

cytotoxic.

Figure 56: Cell proliferation of HT-29 proliferation, left: number of cells in t0 – 10.000 cells, right: number of cells in t0 – 5.000 cells. All samples show excellent proliferation, except

for the cytotoxic PEI60000. SDS served as a negative control.

Apoptosis and cell viability assay

To examine the viability and apoptosis of PEI, DNA or nanodiamond treated samples, 105

cells were incubated with nanodiamond particles alone or in a combination with PEI800

and /or DNA for 24 hours. Cell were harvested, incubated for 15 minutes with AnnexinV-

FITC (Life Technologies) and propidium iodide, and analyzed with a flow cytometry BD-

LSRII and FlowJo 7.2.2 software. Campthothecin (5µg/ml; Sigma-Aldrich) served as a

positive control of apoptotic induction.

- 96 -

Figure 57: Apoptosis in HT-29 cells. NTC is a negative control, CAM served as a positive control of apoptotic induction.

The only sample that showed apoptosis was ND coated with PEI60000 (Figure 57). Figure

58 shows the viability of the samples. The viability of sample ND-PEI60000 was reduced.

Figure 58: Viability of HT-29 cells. NTC is a negative control.

Real-time Reverse Transcription Polymerase chain reaction (real-time RT-PCR)

The proliferation of the treated cells was also examined by the gene expression of the Ki-

67 and c-Myc genes. This tells us about the functionality of cells. 106 cells were seeded on

a 6-well plate and incubated with nanodiamond particles alone or in a combination with

PEI800 and /or DNA for 24 hours. Cells were harvested and nucleic acids were isolated

- 97 -

using RNeasy Mini Kit (Qiagen, Hilden, Germany). 20µg/100µl reaction of RNA was

transcribed into cDNA using cDNA Archive Kit (Applied Biosystems, Foster City, CA).

Subsequent real-time RT-PCRs of Ki-67, c-Myc, and reference gene GAPDH were

performed using TaqMan® Gene Expression Assays (Applied Biosystems) and an

iCycler5 (Bio-Rad, Philadelphia, PA). Ct values obtained for each gene were normalized to

the reference GAPDH and quantitative gene levels were determined with Bio-Rad iQ5 2.0

software. All samples were performed in biological as well as in technical replicates. The

expression was not done on the samples PEI60000, as the cells were apoptotic.

All samples showed expression of the KI-67 and c-Myc genes comparable to the negative

controle (Figure 59)

Figure 59: Normalized fold expression of the Ki-67 and c-Myc genes showing that the used samples do not influence the cell function.

The results presented in this chapter show high biocompatibility of the ND-PEI

transfection system when the low molecular weight PEI was used (PEI800 and PEI1800).

The samples that were prepared using PEI60000 showed high cytotoxicity.

4.7.4. Verification of successful transfection

To verify the successful transfection of the DNA and the release of the transfected nucleic

acid from the ND-PEI complex, the transfected beta-actin was detected in HT-29 cells.

- 98 -

Cells incubated with nanodiamond particles alone or in different combinations with

PEI800 and/or beta-actin DNA were harvested and real-time PCR for the detection of

GAPDH and beta-actin was performed using TaqMan® Gene expression Assay (Applied

Biosystems). The Assay, validated for beta-actin, enables to detect human and mouse

sequence within the region of amplification product incorporated in the transfection

complex of ND-PEI800-DNA. Ct values obtained for each gene were normalized to the

reference GAPDH and quantitative gene levels were determined with Bio-Rad iQ5 2.0

software. All samples were performed in biological and technical triplicates.

Figure 60 show the normalized fold expression of beta actin that was transfected to HT-29

cells. The DNA was successfully transfected only by the PEI that was previously attached

to the ND. Neither PEI itself nor nanodiamonds coated with DNA succeeded in the

transfection of the nucleic acid.

Figure 60: Normalized fold expression of beta-actin in HT-29 cells.

This result shows the ability of the PEI coated ND to successfully transport DNA to cells.

In this section, we have demonstrated the construction of the fND biosensor operating

contactless in cells based on the discovered principle of the surface charge interactions.

The direct optical monitoring of DNA intracellular release from transfection system at

single particle level was demonstrated using the ND containing NV centers as a

delivery/sensing system.

- 99 -

5. Conclusions and Perspectives

This work has introduced a method of alternation the luminescence of nanodiamond

particles by surface charge interactions. Luminescent properties of nitrogen-vacancy (NV)

defects engineered in variously terminated nanodiamonds were studied. It was found that

luminescence of NV centers is sensitive to surface treatment, where the NV- luminescence

can be quenched by the hydrogenation of the surface, which leads to the shift of the

luminescence towards shorter wavelength that are characteristic for the NV0 luminescence.

Additionally, it was found that the luminescence of NV- centers in small nanocrystals can

be stabilized by fluorine termination. This result is particularly interesting for the

applications of NV centers in spintronics and quantum computing, in which the desired

state of the center is NV- that is located at the surface proximity.

The mechanism of charge transfer between negative and neutral state of the nitrogen

vacancy center was explained on the model system, flat single crystal diamond plate with

shallow implanted NV centers, and was modeled theoretically using the band bending

model of the hydrogenated diamond surface.

In order extend the proposed mechanism to biomedical application, the interactions with

charged molecules was examined. The luminescence of oxidized and fluorinated

nanodiamonds was sensitive to non-covalently bounded positively charged polymers. The

presence of positively charged polymers lead to the charge transfer from NV- to NV0

center, which is similar to the effects induced by hydrogen termination. The NV-

luminescence was restored upon the interaction with negatively charged polymers. The

proposed effects were used for the optical detection of the formation of the nanodiamond-

mediated transfection system based on the electrostatic interactions between charged

biomolecules, polyethylene imine and DNA. The transfection system was successfully

used to transport DNA to cancer cells. These results indicate the suitability of

nanodiamonds as gene-delivery platform that provide highly efficient transfection while

remaining harmless to cells. Further on, the nanodiamond-mediated transfection was

optically monitored, showing the ability of functionalized nanodiamond to sense the

biomolecular attachment or release.

- 100 -

Additionally, this work has presented the extensive study of the optimization of the

fabrication of highly luminescent nanodiamonds that lead to the development of bright

nanodiamond particles that were small enough to be easily taken up by cells and were

bright enough to be observable under standard commercially available fluorescence

microscope.

If we consider future applications of fluorescent nanodiamond, this thesis concentrates on

the therapeutic application of nanodiamond: a use of the nanodiamond as a drug delivery

probe where the drug delivery event can be optically monitored. Even though the results

presented in the thesis showed high biocompatibility and high efficiency of the DNA

delivery, it is unlikely that nanodiamonds will be clinically used for this purpose in the

near future. The main reason is the unknown fate of nanodiamonds in the human body. As

nanodiamonds are not biodegradable, the accumulation in the liver or kidneys is expected.

This could not be a problem if nanodiamonds were used in the small doses, but it is not the

case for chemotherapeutic applications, when large amounts of drugs are frequently

injected to the body.

Besides the discussed therapeutic applications of the fluorescence nanodiamonds, the

proposed effect of the surface charge induced modulation of photoluminescence could be

used for diagnostic purposes. One of the possible promising applications that are likely to

be used in the near future is detection of an early stage of cancer. Cancer cells in the early

stage of growth can be traced due to the abnormal regulatory processes. The tracing can be

done by alterations in cell signaling mechanisms, for instance those involving cytoplasmic

tyrosine kinases, specific growth factors, the transcription apparatus and/or genes involved

in the cell cycle and regulation of DNA replication. One of the possible mechanisms is

therefore the detection of the DNA hybridization. If the nanodiamond particle was

specifically functionalized by the specific miRNA, the monitoring of the hybridization can

serve as a probe for the targeting of cancer cells. As the hybridization of DNA is related to

the charge changes in the surface proximity, the methods that were described in this work

can be used for the non-invasive, contactless diagnostics.

- 101 -

In general, the overall findings in this work extend the applications of luminescent

nanodiamond particles as a fluorescent marker, allowing them to be used as a long-term

optical biosensor that operates on the single molecular level. The developed probes can

serve in the future as a generally used system for nanoscale imaging. Further experimental

investigations are needed to estimate the optimal experimental conditions to achieve the

sensitivity of the designed nanodiamond probe to single molecules. Additional research

need to be focused also on the mechanism of the DNA release in cells as well as the overall

stability of the transfection complex. Further experiments will be performed using pH

sensitive covalent linkers for the DNA to the surface of nanodiamond.

- 102 -

6. References

[1] J. Lakowicz, Principles of Fluorescence Spectroscopy, 4th Printing., 2006,

Springer, ISBN 978-0-387-31278-1

[2] Yuan Tu, Martin Chalfie, Green fluorescent protein as a marker for gene

expression, Science, 1994, 263, 5148, 802-805

[3] B. Ballou et al., Noninvasive imaging of quantum dots in mice. Bioconjug Chem.,

2004, 15, 79-86

[4] S. Mahendra et al., Quantum dot weathering results in microbial toxicity. Environ

Sci Technol., 2008, 42, 9424-30

[5] T. D. Ladd, F. Jelezko, R. Laflamme, Y. Nakamura, C. Monroe, J. L. O’Brien,

Quantum computers, Nature, 2010, 464, 45-53

[6] D. Ho, Beyond the Sparkle: The Impact of Nanodiamonds as Biolabeling and

Therapeutic Agents, ACS Nano, 2009, 3 (12), 3825–3829

[7] C. C. Fu, H. Y. Lee, K. Chen, T. S. Lim, H. Y. Wu, P. K. Lin, P. K. Wei, P. H.

Tsao, H. C. Chang, W. Fann, Characterization and application of single fluorescent

nanodiamonds as cellular biomarkers. Proc. Natl. Acad. Sci. USA 2007, 104 (3)

727-732

[8] Y. R. Chang, H. Y. Lee, K. Chen, C. C. Chang, D. S. Tsai, C. C. Fu, T. S. Lim, Y.

K. Tzeng, C. Y. Fang, C. C. Han, H. C. Chang, W. Fann, Mass production and

dynamic imaging of fluorescent nanodiamonds, Nat. Nanotechnol. 2008, 3, 284 -

288

[9] K. K. Liu, C. L. Cheng, C. C. Chang, J. I. Chao, Biocompatible and detectable

carboxylated nanodiamond on human cell. Nanotechnology 2007, 18, 325102¨

[10] J. Tisler, G. Balasubramanian, B. Naydenov, R. Kolesov, B. Grotz, R. Reuter, J. B.

Boudou, P. A. Curmi, M. Sennour, A. Thorel, M. Burscht, K. Aulenbacher, R.

Erdmann, P. R. Hemmer, F. Jelezko, J. Wrachtrup, Fluorescence and Spin

Properties of Defects in Single Digit Nanodiamonds. ACS Nano, 2009, 3 (7), 1959–

1965

[11] T. Gaebel, M. Domhan, C. Wittmann, I. Popa, F. Jelezko, J. Rabeau, A. Greentree,

S. Prawer, E. Trajkov, P. R. Hemmer, J. Wrachtrup, Photochromism in single

nitrogen-vacancy defect in diamond. Appl. Phys. B. 2006, 82, 243-246

- 103 -

[12] K. Iakoubovskii, G. J. Adriaenssens, M. Nesladek, Photochromism of vacancy-

related centres in diamond, J. Phys.:Cond. Matter, 2000, 12, 189-199

[13] S. J. Yu, M. W. Kang, H. C. Chang, K. M. Chen, Y. C. Yu, Bright fluorescent

nanodiamonds: no photobleaching and low cytotoxicity J. Am. Chem. Soc. 2005,

127 17604–5

[14] Y. Yuan, Y. Chen, Y. H. Liu, H. Wang, Y. Liu, Biodistribution and fate of

nanodiamonds in vivo. Diamond Relat. Mater., 2009, 18, 95–100

[15] Y. Liang, M. Ozawa, A. Krueger, A General Procedure to Functionalize

Agglomerating Nanoparticles Demonstrated on Nanodiamond, ACS Nano, 2009, 3

(8), 2288–2296

[16] X. Q. Zhang, M. Chen, R. Lam, X. Xu, E. Osawa, D. Ho, Polymer-Functionalized

Nanodiamond Platforms as Vehicles for Gene Delivery. ACS Nano, 2009, 3 (9),

2609–2616

[17] G. Davies, M. F. Hamer, Optical studies of the 1.945 eV vibronic band in diamond,

Proc. R. Soc. Lond. A. , 1976, 348, 285-298

[18] S. C. Lawson, D. Fisher, D. C. Hunt, M. E. Newton, On the existence of positively

charged single-substitutional nitrogen in diamond, J. Phys.: Condens. Matter, 1998,

10, 6171–6180

[19] N. Mohan, Y. K. Tzeng, L. Yang, Y. Y. Chen, Y. Y Hui, C. Y. Fang, H. C. Chang,

Sub-20-nm Fluorescent Nanodiamonds as Photostable Biolabels and Fluorescence

Resonance Energy Transfer Donors, Adv. Mater. 2010, 22, 843–847

[20] H. B. Dyer, F. A. Raal, L. Dupreez, J. H. N. Loubser, Optical absorption features

asociated with paramagnetic nitrogen in diamond, Philos. Mag., 1965, 11, 763

[21] R. Berman, P. R. W. Hudson, M. Martinez, J. Phys. C: Solid State Phys., 1975, 6,

430-434.

[22] A. T. Collins, Optical centres produced in diamond by radiation damage, New

Diamond Front. Carbon Technol. 2007, 17, 47-61

[23] P. May, Diamond thin films: a 21st-century material. Philos. Trans. R. Soc.

London, Ser. A., 2000, 358, 473-495.

[24] Bundy, F. P.; Hall, H. T.; Strong, H. M.; Wentorf, R. H., MAN-MADE

DIAMONDS. Nature 1955, 176, 51-55.

[25] Bovenkerk, H. P.; Bundy, F. P.; Hall, H. T.; Strong, H. M.; Wentorf, R. H.,

PREPARATION OF DIAMOND. Nature 1959, 184, 1094-1098.

- 104 -

[26] A. Kruger, F. Kataoka, M. Ozawa, T. Fujino, Y. Suzuki, A. E. Aleksenskii, A. Vul

E. Osawa, Unusually tight aggregation in detonation nanodiamond: Identification

and disintegration. Carbon. 2005, 8, 1722-1730.

[27] K. B. Holt, Diamond at the nanoscale: applications of diamond nanoparticles from

cellular biomarkers to quantum computing, Philos. Trans. R. Soc. London, Ser. A.

2007, 12, 2845-2861

[28] Z. J. S Dangsheng, Z. Aihua. Nanocarbon as Robust Catalyst: Mechanistic Insight

into Carbon-Mediated Catalysis, Angew. Chem. Int. Ed. 2007, 46, 38.

[29] Y. Xing, L. Dai, Nanodiamonds for nanomedicine, Nanomedicine, 2009, 4, 2

[30] M. Ozawa, E. Osawa, Preparation and Behavior of Brownish, Clear Nanodiamond

Colloids. Avenced Materials, 2007, 19, 1201 - 1206.

[31] A. Krueger, T. Boedeker, Deagglomeration and functionalisation of detonation

diamond. Phys. Status Solidi A. 2007, 20, 1367-1370

[32] Angus, J. C.; Hayman, C. C., LOW-PRESSURE, METASTABLE GROWTH OF

DIAMOND AND DIAMONDLIKE PHASES. Science 1988, 241, 913-921.

[33] Gruen, D. M., Nanocrystalline diamond films. Annual Review of Materials Science

1999, 29, 211-259.

[34] A. M. Zaitsev. Optical Properties of Diamond : A Data Handbook. 1 edition.:

Springer, 2001. 519 p. ISBN 978-3540665823

[35] F. Jelezko, J. Wrachtrup, Single defect centres in diamond: A review, Phys. Status

Solidi A, 2006, 203 (13)

[36] L. Rondin, G. Dantelle, A. Slablab et al. Surface-induced charge state conversion of

nitrogen-vacancy defects in nanodiamonds, Phys. Rev. B, 2010, 82, 115449-115454

[37] V. Vaijayanthimala, H-C. Chang, Functionalized fluorescent nanodiamonds for

biomedical applications, Nanomedicine, 2009, 4, 1, 47-55

[38] F. Treussart, V. Jacques, E. Wu et al, Photoluminescence of single colour defects in

50 nm diamond nanocrystals, Physica B, 2006, 376, 926-929

[39] F. Neugart, A. Zappe, F. Jelezko et al, Dynamics of diamond nanoparticles in

solution and cells, Nano Lett., 2007, 7, 3588-3591

[40] V. L. Colvin, The potential environmental impact of engineered nanomaterials, Nat.

Biotech, 2003, 21, 1166-1170

[41] G. Jia, H. Wang, L. Yan et al, Cytotoxicity of carbon nanomaterials: Single wall

nanotube, multi wall nanotube and fullerene, Environ. Sci. Technol, 2005,39, 1378-

1383

- 105 -

[42] A. M. Schrand, S. A. C. Hens, O. A. Shenderova, Nanodiamond Particles:

Properties and Perspectives for Bioapplications, Critical Reviews in Solid State and

Material Sciences, 2009, 34, 1-2, 18-74

[43] R. C. Burns, G. J. Davies, Properties of Natural and Synthetic Diamond. Field JE

(Ed.). Academic Press, London, UK, 1992, 395-422

[44] T. D. Moustakas, Synthetic Diamond: Emerging CVD Science and Technology,

Spear KE, Dismukes JP (Eds), John Wiley, New York, USA, 1994, 145-192

[45] P. Cigler, M. Ledvina, M. Tvrdonova, V. Rezacova, M. Nesladek, I. Kratochvilova,

F. Fendrych, J. Stursa, J. Kucka, J. Ralis, Fluorescent nanodiamonds: new platform

for construction of chemo- and biosensors. Abstr. 2nd International Conference,

Conference proceedings, Nanocon 2010 Olomouc, Czech Republic, October 12-14,

2010

[46] G. Davies, S. C. Lawson, A. T. Collins et al., Vacancy related centers in diamond,

Phys Rev. B, 1992, 46, 13157-13170

[47] N. Gibson, O. Shenderova, A. Puzyr, K. Purtov, V. Grichko, T. J. M. Luo, Z.

Fitgerald, V. Bondar, D. Brenner, Nanodiamonds for detoxication, Technical

Proceedings of the 2007 NSTI NanoTechnology Conference, 2007

[48] V. Grichko, V. Grishko, O. Shenderova, Nanodiamond bullets and their biological

targets, Nanobiotechnology, 2006, 2, 37

[49] H. Huang, E. Pierstorff, E. Osawa, D. Ho, Active nanodiamond hydrogels for

chemotherapeutic delivery, Nano. Lett, 2007, 7, 3305

[50] E. K. Chow, X.-Q. Zhang, M. Chen, R. Lam, E. Robinson, H. Huang, D. Schaffer,

E. Osawa, A. Goga, D. Ho, Sci. Transl. Med. , 2011, 3, 73ra21

[51] V. S. Bondar, I. O. Pozdnyakova, A. P. Puzyr, Applications of nanodiamonds for

separation and purification of proteins, Phy. Solid State, 2004, 46, 758

[52] L.C. Huang, H. C. Chang, Adsorption and immobilization of cytochrome c on

nanodiamonds, Langmuir, 2004, 20, 5879

[53] T. T. B. Nguyen, H. C. Chang, V. W. K. Wu, Adsorption and hydrolytic activity of

lysozyme on diamond nanocrystallites, Diamond Relat. Mater., 2007, 16, 872

[54] Bhushan, Bharat. Springer Handbook of Nanotechnology. s.l. : Springer, 2007.

ISBN: 978-3-540-29855-7

[55] Martin, Donald. Nanobiotechnology of Biomimetic Membranes. s.l. : Springer,

2007. ISBN: 978-0-387-37738-4

- 106 -

[56] Kaushik, Dayal. Non-contact AFM. SPM Techniques. [Online] 2 26, 2010. [Cited:

4 3, 2010.] http://www.mechmat.caltech.edu/~kaushik/park/1-2-2.htm.

[57] Dynamic Light Scattering. Particle size standarts and micropheres. [Online]

BrookHaven

[58] Ferrari, Mauro. BioMEMS and Biomedical Nanotechnology: Biomolecular Sensing,

Processing and Analysis. 2007. Vol. V4. ISBN: 978-0-387-25561-35

[59] Vickerman, John C and Gilmore, Ian. Surface Analysis: The Principal Techniques.

2. edition. s.l. : Wiley, 2009. ISBN: 978-0-470-01763-0.J.W.G. Antonie Visser, O.

J. Rolinski, Basic photophysics, 2010, Glasgow UK, available at

http://www.photobiology.info/Visser-Rolinski.html

[60] Confocal microscope In Wikipedia : the free encyclopedia. [Online] Wikipedia

Foundation, 3 10, 2010. [Cited: 3 22, 2010.]

http://en.wikipedia.org/wiki/Confocal_microscope.

[61] V. Petrakova, A. Taylor, I. Kratochvilova, F. Fendrych, J. Vacik, J. Kucka, J.

Stursa, P. Cigler, M. Ledvina, A. Fiserova, P. Kneppo, M. Nesladek, Adv. Func.

Mater., 2012, 22, 4, 812-819

[62] O. Faklaris, V. Joshi, T. Irinopoulou, P. Tauc, M. Sennour, H. Girard, C. Gesset, J.

C. Arnault, A. Thorel, J. B. Boudou, P. A. Curmi, F. Treussart, Photoluminescent

Diamond Nanoparticles for Cell Labeling: Study of the Uptake Mechanism in

Mammalian Cells, ACS Nano, 2009, 3 (12), 3955–3962

[63] H. Huang, E. Pierstorff, E. Osawa, D. Ho, Protein-Mediated Assembly of

Nanodiamond Hydrogels into a Biocompatible and Biofunctional Multilayer

Nanofilm, ACS Nano, 2008, 2 (2), 203–212

[64] R. Martn, M. lvaro, J. R. Herance, H. Garca, Fenton-Treated Functionalized

Diamond Nanoparticles as Gene Delivery System, ACS Nano, 2010, 4 (1), 65–74

[65] M. Chen, E. D. Pierstorff, R. Lam, S. Y. Li, H. Huang, E. Osawa, D. Ho,

Nanodiamond-Mediated Delivery of Water-Insoluble Therapeutics, ACS Nano,

2009, 3 (7), 2016–2022

[66] K. K. Liu, W. W. Zheng, C. C. Wang, Y. C. Chiu, C. L. Cheng, Y. S. Lo, C. Chen,

J. I. Chao, Covalent linkage of nanodiamond-paclitaxel for drug delivery and

cancer therapy. Nanotechnology, 2010, 21 315106

[67] J. Ristein, M. Riedel, L. Ley, Electrochemical surface transfer doping – The

mechanism behind the surface conductivity of hydrogen-terminated diamond, J.

Electrochem. Soc., 2004, 151, 10,315-321

- 107 -

[68] F. Maier, M. Riedel, B. Mantel, J. Ristein, L. Ley, Origin of surface conductivity in

diamond, Phys. Rev. Lett., 2000, 16, 85, 3472-3475

[69] P. Strobel, M. Riedel, J. Ristein, L. Ley, Nature, 2004, 430, 6998, 439-441

[70] C. E. Nebel, H. Kato, B. Rezek, D. Shin, D. Takeuchi, H. Watanabe, T. Yamamoto,

Electrochemical properties of undoped hydrogen terminated CVD diamond.

Diamond Relat. Mater., 2006, 15, 264–268

[71] C. E. Nebel, Surface transfer-doping of H-terminated diamond with

adsorbates, New Diamond and Frontier Carbon Technology, 2005, 15, 5, 247-264

[72] B. Rezek, D. Shin, H. Watanabe , C. E. Nebel, Sensors and Actuators B: Chemical,

2007, 122, 596-599

[73] J. A. Garrido, A. Hardl, S. Kuch, M. Stutzmann, O. A. Williams, R. B Jackmann,

Appl. Phys. Lett., 2005, 86, 073504

[74] A. Hartl, E. Schmich, J. A. Garrido, J. Hernando,S. C. R. Catharino, S. Walter, P.

Feulner, A. Kromka, D. Steinmuller, M. Stutzmann, Nat. Mater. 2004, 3, 736-742

[75] B. Rezek, M. Kratka, A. Kromka, M. Kalbacova, Biosens. Bioelectron., 2010, 26,

1307–1312

[76] J. P. Goss, P. R. Briddon, R. Jones, S. Sque, Donor and acceptor states in diamond.

Diamond Relat. Mater., 2004, 13, 351–354

[77] J. P. Goss, P. R. Briddon, R. Jones, S. Sque, M. J. Rayson, Vacancy-impurity

complexes and limitations for implantation doping of diamond. Phys. Rev. B, 2005,

72, 035214

[78] I. Kratochvilova, A. Taylor, A. Kovalenko, F. Fendrych, V. Rezacova, V. Petrak, S.

Zalis, J. Sebera, M. Nesladek, Fluoresent Nanodiamonds: Effect of Surface

Termination, Mater. Res. Soc. Symp. Proc., 2010, 1203, J03-05.

[79] K. M. C. Fu, C. Santori, P. E. Barclay, R. G. Beausoleil, Conversion of neutral

nitrogen-vacancy centers to negatively charged nitrogen-vacancy centers through

selective oxidation. Appl. Phys. Lett., 2010, 96, 121907

[80] J. Ristein, M. Riedel, M. Stammler, B. F. Mantel, L. Ley, Surface

conductivity of nitrogen-doped diamond. Diamond Relat. Mater., 2002, 11, 359–

364

[81] L. Allers, A. T. Collins, J. Hiscock, The annealing of interstitial-related optical

centres in type II natural and CVD diamond, Diam. Relat. Mater., 1998, 7 228–32

[82] K. Iakoubovskii, G. J. Adriaenssens, Trapping of vacancies by defects in diamond,

J. Phys.: Condens. Matter, 2001, 13 6015–18

- 108 -

[83] M. Nesladek, L. M. Stals, A. Stesmans, K. Iakoubovskij, G. J. Adriaenssens,

Dominant defect levels in diamond thin films: A photocurrent and electron

paramagnetic resonance study. Appl. Phys. Lett., 1998, 72, 25, 3306-3308

[84] B. Rezek, C. Sauerer, C. E. Nebel, M. Stutzmann, J. Ristein, L. Ley, E. Snidero, P.

Bergonzo, Fermi level on hydrogen terminated diamond surfaces, Appl. Phys.

Letters, 2003, 82, 14

[85] S. Pazzagna, B. Naydenov, F. Jelezko, J. Wratchup, J. Meijer, Creation efficiency

of nitrogen-vacancy centres in diamond, New Journal of Physics, 2010, 12, 065017

[86] C. Santori, P. E. Barclay, K. C. Fu, R. G. Beausoleil, Vertical distribution of

nitrogen-vacancy centers in diamond formed by ion implantation and annealing,

Phys. Rev. B, 2009, 79, 12, 5313-5321

[87] J. P. Boudou, P. A. Curmi, F. Jelezko, J. Wrachtrup, P. Aubert, M. Sennour, G.

Balasubramanian, R. Reuter, A. Thorel, E. Gaffet, High yield fabrication of

fluorescent nanodiamonds, Nanotechnology, 2009, 20, 23, 5602-5613

[88] M. V. Hauf, B. Grotz, B. Naydenov, M. Dankerl, S. Pezzagna, J. Meijer, F. Jelezko,

J. Wrachtrup, M. Stutzmann, F. Reinhard, J. A. Garrido, Chemical control of the

charge state of nitrogen-vacancy centers in diamond Phys. Rev. B, 2011, 8, 83

[89] V. Petráková, M. Nesládek, A. Taylor, F. Fendrych, P. Cígler, M. Ledvina, J.

Vacík, J. Štursa, J. Kučka, Luminescence properties of engineered nitrogen vacancy

centers in a close surface proximity, Phys Status Solidi A, 2011, 9, 208, 2051-2056,

DOI: 10.1002/pssa.201100035

[90] J. Zeigler, The stopping range of ions in matter, SRIM-2008, online: http://srim.org

[91] B. Grotz, M. V. Hauf, M. Dankerl, B. Naydenov, S. Pezzagna, J. Meijer, F. Jelezko,

J. Wrachtrup, M. Stutzmann, F. Reinhard, J. A. Garrido, Nat. Commun., 2012, 3,

729

[92] S. Osswald, G. Yushin, V. Mochalin, S. O. Kucheyev, Y. Gogotsi, Control of

sp2/sp3 Carbon Ratio and Surface Chemistry of Nanodiamond Powders by

Selective Oxidation in Air, J. Am. Chem. Soc., 2006, 128, 11635-11642

[93] G. Davies, S. C. Lawson, A. T. Collins, et al., Phys Rev B, 1992, 46, 20, 13157-

13170, DOI: 10.1103/PhysRevB.46.13157

[94] S. C. Lawson, D. Fisher, D. C. Hunt, et al., J. Phys. Cond. Matter, 1998, 10, 27,

6171-6180, DOI: 10.1088/0953-8984/10/27/016

[95] F. C. Waldermann, P. Olivero, J. Nunn, et al., Diamond relat matter, 2007, 16, 11,

1887-1895, DOI: 10.1016/j.diamond.2007.09.009

- 109 -

[96] J. Botsoa, T. Sauvage, M. P. Adam, et al., Phys. Rev. B., 2011, 84, 12, 125209,

DOI: 10.1103/PhysRevB.84.125209

[97] G. Davies, A. T. Collins, Diamond relat. Matter., 1993, 2, 2-4, 80-86, DOI:

10.1016/0925-9635(93)90035-Z

[98] G. Davies, M. F. Hamer, Proc. R. Soc. Lond. A. , 1976, 348, 285-298

[99] A. T. Tiwari, J. P. Goss, P. R. Briddon, et al., Phys. Rev B, 2011, 84, 24, 245305,

DOI: 10.1103/PhysRevB.84.245305

[100] M. Mahmoudi, I. Lynch, M. Reza Ejtehadi, M. P. Monopoli, F. Baldelli

Bombelli,S. Laurent, ProteinNanoparticle Interactions: Opportunities and

Challenges, Chem. Rev. 2011, 111, 5610-5637

[101] Mark Chen, Xue-Qing Zhang, Han B. Man, Robert Lam, Edward K. Chow, and

Dean Ho, The Journal of Physical Chemistry Letters, 2010, 1 (21), 3167-3171

- I -

Appendix

A. SUMMARY OF AUTHOR’S PUBLICATIONS RELATED TO THE TOPIC

PUBLICATIONS

V. Petrakova, M. Ledvina, M. Nesladek, Surface doping of diamond, Chapter 7 in Optical

engineering of diamond (edited by Mildren), Wiley-VCH, accepted in press

V. Petrakova, A. Taylor, I. Kratochvilova, F. Fendrych, J. Vacik, J. Kucka, J. Stursa, P.

Cigler, M. Ledvina, A. Fiserova, P. Kneppo, M. Nesladek, Luminescence of

Nanodiamond Driven by Atomic Functionalization: Towards Novel Detection

Principles, Adv. Func. Mater., 2012, 22, 4, 812-819, DOI: 10.1002/adfm.201101936

V. Petráková, M. Nesládek, A. Taylor, F. Fendrych, P. Cígler, M. Ledvina, J. Vacík, J.

Štursa, J. Kučka, Luminescence properties of engineered nitrogen vacancy centers in

a close surface proximity, Phys Status Solidi A, 2011, 9, 208, 2051-2056, DOI:

10.1002/pssa.201100035

V. Petrakova, A. Taylor, I. Kratochvilova, F. Fendrych, P. Cigler, M. Ledvina, J. Kucka, J.

Stursa, J. Ralis, J. Vacik and M. Nesladek, On the mechanism of charge transfer

between neutral and negatively charged nitrogen-vacancy color centers in diamond.

MRS Proceedings, 2011, 1282, mrsf10-1282-a07-05 doi:10.1557/opl.2011.450

S. D. Janssens, P. Pobedinskas, J. Vacik, V. Petrakova, B. Ruttens, J. D’Haen, M.

Nesladek, K. Haenen and P. Wagner, Separation of the intra- and intergranular

magnetotransport properties in nanocrystalline diamond films on the metallic side of

the metal-insulator transition, New. J. Phys., 2011, 13, 083008 doi: 10.1088/1367-

2630/13/8/083008

Kratochvilova, A. Kovalenko, F. Fendrych, V. Petrakova, S. Zalis, M. Nesladek, Tuning of

nanodiamond particles' optical properties by structural defects and surface

modifications: DFT modelling, J. Mater. Chem., 2011, 21 (45), 18248 - 18255

Taylor, F. Fendrych, L. Fekete, J. Vlcek, V. Rezacova, V. Petrak, J. Krucky, M. Nesladek,

M. Liehr, Novel high frequency pulsed MW-linear antenna plasma-chemistry: Routes

towards large area, low pressure nanodiamond growth, Diam. Relat. Mater., 2011,

20, 4, 613-615

- II -

F. Fendrych, A. Taylor, L. Peksa, I. Kratochvilova, J. Vlcek, V. Rezacova, V. Petrak, Z.

Kluiber, L. Fekete, M. Liehr, M. Nesladek, Growth and characterization of

nanodiamond layers prepared using the plasma-enhanced linear antennas microwave

CVD system, J. Phys. D, 2010, 43, 37, 374018

Kratochvilova, A. Kovalenko, A. Taylor, F. Fendrych, V. Rezacova, J. Vlcek, S. Zalis, J.

Sebera, P. Cigler, M. Ledvina, M. Nesladek, The fluorescence of variously terminated

nanodiamond particles: Quantum chemical calculations, Phys Status Solidi A, 2010,

207, 9, 2045-2048

Kratochvílová, A. Taylor, A. Kovalenko, F. Fendrych, V. Řezáčová, V. Petrák, S. Zalis, J.

Šebera and M. Nesládek, Fluorescent Nanodiamonds: Effect of Surface Termination.

MRS Proceedings, 2009, 1203, 1203-J03-05 doi:10.1557/PROC-1203-J03-05

S. D. Janssens, Pa. Pobedinskas, V. Petráková, M. Nesládek, K. Haenen and P. Wagner,

Influence of methane concentration on the electric transport properties in heavily

boron-doped nanocrystalline CVD diamond films. MRS Proceedings, 2011, 1282,

mrsf10-1282-a15-04 doi:10.1557/opl.2011.454

V. Petráková, M. Nesládek, V. Petráková, M. Nesládek, Luminiscenční nanodiamanty:

Nový marker pro biomedicínu?, Vesmír, 2011, 4, 214

V. Petráková, M. Nesládek, Luminiscenční nanodiamanty: Nový marker pro biomedicínu?,

Vesmír, 2011, 4, 214

PARTICIPATION ON INTERNATIONAL CONFERENCES

Presenting author

Oral presentations

Rezacova V, Nesladek M, Cigler P, Ledvina M, Kratochvilova I, Taylor A, Fendrych F,

Vacik V; Effect of surface termination on photoluminescence of nanodiamond, Abstr. 21th

European Conference on Diamond, 5th-9th September, Budapest, Hungary, O55 (oral),

Abstr No. DIAM2010_0299 – Young Investigator Award for the best oral presentation

V. Petrakova, M. Nesládek: “Surface Termination on Photoluminescence of

Nanodiamond” 1st International Conference on Advances in Cell and Gene Therapy and

Immunotherapy: from basic research to clinical applications: and, 3rd Workshop on

Immunotherapy: September 23-25, 2010, Mikulov Castle, South Moravia, Czech Republic

(oral)

- III -

Rezacova, V.; Nesladek, M.; Cigler, P.; Ledvina, M.; Kucka, J.; Stursa, J.; Ralis, J.; Vacik,

J.; Mojzes, P.; Taylor, A.; Kratochvilova, I.; Fendrych, F.;On the Mechanism of Charge

Transfer between Neutral and Negatively Charged Nitrogen-vacancy Color Centers in

Diamond, MRS Fall Meeting, Boston, Nov.2010, Abstract No. A7.5 Online:

http://www.mrs.org/s_mrs/doc.asp?CID=27791&DID=332936 – Young Investigator

Award for the best oral presentation

V. Petráková, A. Taylor, I. Kratochvílová, F. Fendrych, P. Cígler, M. Ledvina and M.

Nesládek, Optical monitoring of the charge transfer from fluorinated nanodiamonds to

polymer molecules in a liquid environment studied via NV luminescence changes, Hasselt

Diamond Workshop 2011, SBDD XVI, Feb 21th-23th, Hasselt, Belgium, O4.3 (oral), p. 50

V. Petráková, A. Taylor, I. Kratochvílová, F. Fendrych, P. Cígler, M. Ledvina, J. Šturza, J.

Kučka and M. Nesládek, Effect of atomic functionalization on NV charge transfer induced

by interactions with polar molecules, International Conference on New Diamond and Nano

Carbons, 2011, May 16th-20th, Kunibiki Messe, Shimane, Japan, O19-6 (oral), p.130

V. Petráková, A. Taylor, P. Cígler, M. Ledvina, F. Fendrych, M. Nesládek, Study of

luminescence properties of nitrogen-vacancy centres in nanodiamond as a function of size

and surface interactions, 62th De Beers Diamond Conference (oral), 2011, Warvick, UK,

July 4th

-7th

2011, O6

Posters

V. Rezacova, A. Taylor, I. Gregora, Z. Remes, I. Kratochvilova, F. Fendrych, J. Vacik, J.

Poltierova-Vejpravova, M. Nesladek, Raman and luminescence imaging of treated

nanocrystaline diamond films and singe crystal diamond, Hasselt Diamond Workshop

2010, SBDD XVI, Feb 22th-24th, Hasselt, Belgium, P5.51, p. 15

V. Rezacova, V Petrak, J. Krucky, Nanodiamond particles for in cell applications, Poster

2010, Prague, Czech Republic, April 2010 – best poster presentation

V. Petrakova, Optical detection of charged biomolecules, Poster 2011, Prague, Czech

Republic, May 12th

, 2011 – awarded poster

- IV -

Coauthor

Oral:

S. D. Janssens, V. Rezacova, M. Nesladek, K. Haenen and P. Wagner. Influence of

Methane Concentration on the Electrical Transport Properties in Heavily Boron-doped

Nanocrystalline CVD Diamond Films. MRS Fall Meeting, Boston, Nov.2010, Abstract No.

A15.4 Online: http://www.mrs.org/s_mrs/doc.asp?CID=27791&DID=332936

P. Cigler, M. Ledvina, M. Tvrdonova, V. Rezacova, M. Nesladek, I. Kratochvilova, F.

Fendrych, J. Stursa, J. Kucka, J. Ralis, Abstr. 2nd International Conference, Conference

proceedings, Nanocon 2010 Olomouc, Czech Republic, October 12-14, 2010

M. Nesladek, V. Petrakova: Chemical Driving of NV luminescence in ND particles. E-

MRS Spring Meeting 2010 (invited)

M. Nesladek, V. Petrakova: MRS Spring meeting 2011: Chemical Driving of NV

luminescence in anaodiomn for detection in cell (invited)

Poster:

A. Kovalenko, I. Kratochvilova, V. Petrakova, L. Fekete, F. Fendrych, S. Zalis, V. kocka,

M. Ledvina, P. Cigler, J. Stursa, M. Nesladek, Luminescence of variously terminated

nanodiamond particles: Density functional calculations, Hasselt Diamond Workshop 2011,

SBDD XVI, Feb 21th-23th, Hasselt, Belgium, P5.48, p. 15

I. Kratochvilova, A. Kovalenko, V. Rezacova, F. Fendrych, S. Zalis, M. Ledvina, P. Cigler

and M. Nesladek. Parameters Affecting the Fluorescence of Nanodiamond Particles:

Experiment and Quantum Chemical Calculations. MRS Fall Meeting, Boston, Nov.2010,

Abstract No. A5.23 Online: http://www.mrs.org/s_mrs/doc.asp?CID=27791&DID=332936

V. Petrak, P. Cigler, M. Ledvina, V. Rezacova, J. Masek, J. Turanek, S. D. Janssens, K.

Haenen and M. Nesladek Ionic Strength as a Driving Force of Nanodiamond Aggregation

in Aqueous Solution. MRS Fall Meeting, Boston, Nov.2010, Abstract No. A5.12 Online:

http://www.mrs.org/s_mrs/doc.asp?CID=27791&DID=332936

- V -

A. Kovalenko, A. Taylor, I. Kratochvilova, F. Fendrych, V. Rezacova, V. Petrak, S. Zalis,

J. Sebera, Dependence of the energetic structure of the diamond NV centre on

nanodiamond surface termination: quantum chemical calculations, Hasselt Diamond

Workshop 2010, SBDD XVI, Feb 22th-24th, Hasselt, Belgium, P5.41, p. 14

V. Petrak, A. Taylor, V. Rezacova, M. Ledvina, P. Cigler, Z. Remes, F. Fendrych, M.

Nesladek, Optimization of sonication process in monodispesed diamond seeding, Hasselt

Diamond Workshop 2010, SBDD XVI, Feb 22th-24th, Hasselt, Belgium, P5.46, p. 14