University of Groningen Integration and modification of photosystem I for bio-photovoltaics Gordiichuk, Pavlo IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below. Document Version Publisher's PDF, also known as Version of record Publication date: 2016 Link to publication in University of Groningen/UMCG research database Citation for published version (APA): Gordiichuk, P. (2016). Integration and modification of photosystem I for bio-photovoltaics. [Groningen]: University of Groningen. Copyright Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons). Take-down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim. Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum. Download date: 03-10-2020
Transcript
University of Groningen
Integration and modification of photosystem I for bio-photovoltaicsGordiichuk, Pavlo
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.
Document VersionPublisher's PDF, also known as Version of record
Publication date:2016
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):Gordiichuk, P. (2016). Integration and modification of photosystem I for bio-photovoltaics. [Groningen]:University of Groningen.
CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.
5.2 Covalent attachment of dyes to PSI protein complex ........................................................... 63
5.3 Time-resolved spectroscopy study of energy transfer from attached dyes to PSI protein complex ....................................................................................................................................... 64
5.4 Oxygen consumption of modified PSI protein complexes ................................................... 65
5.5 Study of PSI-dye conjugate activity with the help of Kelvin spectroscopy ......................... 68
In this general chapter, a brief introduction to energy consumption on our planet is given. Then it is described how energy are collected by photosynthesis and artificial solar cells. The different working mechanisms and man-made photovoltaic devices are described and the basic principles of natural energy conversion is highlighted. Special emphasis is paid to the important building blocks of photosynthesis. Different proteins performing light harvesting are described followed by the protein complex that carries out charge separation. Furthermore, it is detailed how these building blocks are biophysically investigated on substrate surface out of their natural environment. Finally, it is outlined how these protein machinery is integrated into various types of artificial solar cells. This chapter ends with a motivation of this thesis.
1
Introduction
2
1.1 Energy consumption
In 2007 at the MRS Spring Meeting in San Francisco, Nathan S. Lewis and Daniel G. Nocera gave overview that human energy consumption on Earth is equal to 13 Trillion Watts (13 TW)1, 2. They also reported that 85% of this energy is produced from fossil fuels such as oil, gas and coal, in approximately equal measure, with these natural resources being unequally distributed among different countries. The total unsustainable biomass used equals 1.2 TW. Renewable energy represents only 2% of total consumption of energy produced in 2007, where biomass represented 0.1 TW, wind power 0.1 GW. Solar-generated power from photovoltaic devices represent a small fragment of energy generated. However, this sector is growing strongly each year3.
The consumption of electrical energy is proportional to the Earths’ population, which is still growing. It is predicted that the world’s population will hit 10 billion people in 2050 and, as a result, energy consumption will more than double due to requirements for economic growth. Following the use of fossil fuels, the concentration of CO2 has increased significantly, causing changes in the Earth’s atmosphere and creating a global warming effect. At the same time, there is no clear solution for eradicating excess CO2, which is continuously rising in our atmosphere as a result of the combustion of fossil fuels. This accumulated carbon dioxide can circulate for a long time before it is broken down naturally and as yet there are no known quick and efficient industrial uses for it, except the natural process of photosynthesis. It is for this reason that photosynthesis is a key future strategy for exploiting CO2 to produce green energy for sustainable life. The idea of combining living houses with solar panels to produce energy can now be observed in the modern architectural trend towards green engineering, which makes use of all possible renewable energy sources3. The sun is a natural resource that provides a significant amount of energy. When light from the sun reaches the atmosphere, its solar radiation is similar to that of a black body with a temperature of ~5800 K with a maximum radiation intensity within the blue-green spectral range ( ~ 550 nm). On a sunny day, energy from the sun can reach up m-2 as it enters the atmosphere and ~1-0.5 m-2 at the Earth’s surface, which is absorbed by plants. In comparison to the Earth’s energy sources, the sun provides our planet with 120000 TW of light energy4, far greater than required by people for their daily lives. In just one hour, the sun provides the same amount of energy as required by humans for an entire year of energy consumption. It is for this reason that the trend for developing devices to harvest solar energy and convert it into electrical power is a significant and long-standing aim for mankind. Plants have a tremendous potential in the harvest of solar energy and electrical energy production, either through biomass production or manufacturing bio-photovoltaic devices. To achieve efficient use of solar energy, it is essential that we learn how to harvest sunlight as plants do through photosynthesis, and how to reproduce the effective charge separation mechanisms of photosynthetic reaction centres.
Chapter 1
3
Photosynthesis is the process by which sunlight is directly harvested and converted into energy through the formation of chemical bonds. However, the overall efficiency of the photosynthesis process is below ~1%5, because of several limitations that cause moderate independent growth6, 7. Furthermore, less than 10% of full sunlight penetrates the photosynthetic apparatus’ capacity8. As such, it is important to distinguish between primary reactions, such as energy transfer between light harvesting elements, from reactions with the centre and resulting charge separations, which are highly optimized with almost 100% efficiency. However, the processes involved in the final synthesis of organic material are very inefficient. Photosynthesis is limited by several factors: The existence of light-regulated mechanisms that have been evolutionally developed to
achieve moderate growth, i.e. such bio-systems cannot increase their output mass with an increase of light power. It has a poor absorbance spectrum, and as such some solar light passes through plants and
is not absorbed. Biological growth is a process that is dependent on temperature.
If it were possible to optimise photosynthesis, some plants could complete their growth cycles several times in a single year, which would be a ground-breaking means of efficiently producing bio mass for use in the bio-diesel industry, thereby simultaneously reducing CO2 in the atmosphere. Applying the Willie Sutton principle, one can use photosynthetic proteins and apply them in bio-solar cells. One sustainable aspect of this strategy is that the complex biological machines can be grown and proliferated in bacteria9. Fabrication of bio-solar cells using biological materials such as photosystems is a very promising prospect due to the high internal quantum efficiency of the protein complex and their stability under demanding environmental conditions experienced by extremophile organisms. There are a few important questions that we need to answer before we can do this, however: How do we stack and order these kinds of photo-active protein complexes on
semiconducting surfaces to fabricate devices and efficiently extract charges? Which materials should be selected for electrodes in order to efficiently extract the
charges generated from photo-active proteins to anode and cathode, without reducing their natural functionality? How do we broaden the limited absorption spectrum, of Nature’s chlorophyll molecules?
The physical characteristics of photosynthetic proteins and their molecular structures inspired researchers to fabricate artificial solar cells featuring different working principles found in photosynthesis10-12. The fabrication of such systems via a self-assembly strategy is not easy because several functions have to be integrated in one system, such as light harvesting, electron transfer, proton-coupled electron transfer and catalysis mechanisms, which might influence each other13, 14.
Introduction
4
For example, dye-sensitized solar cells (DSSCs) use a principle, which is found in photosynthesis, to generate electrical energy from harvested light. Sunlight is absorbed by pigment molecules that are covalently attached to a metal-oxide surface, which transfers the excited electrons to the metal-oxide surface. An electrolyte solution acts as counter electrode15, 16. The effective surface area of DSSCs has been increased by fabricating 3D metal-oxide electrode structures. The best DSSCs have an external quantum efficiency as high as 13%17. But these cells still suffer from low stability. However, DSSCs do not contain self-repair systems against photo oxidation caused by singlet oxygen formation as found in natural photosystems.
The idea of using of biologically-grown photosynthetic material to generate power is not a new concept. Marine species such as Elysia Chlorotica18, 19 and sacoglossa Elysia pusilla20 , which can be found in saltwater marshes from South Florida to Nova Scotia, produce chlorophyll molecules and use photosynthesis to support own life by using the “kleptoplasty” method21. These organisms have evolved to use sunlight for food production via intermediate photosynthesis of algae that they have internalized. The chloroplasts (Figure 1.1a) can survive and continue to function inside Elysia Chlorotica for up to nine or even ten months18. How these organisms are able to support the survival of the “stolen” chloroplasts for so long has been the subject of extensive studies22. The long-term functionality of chloroplasts inside Elysia Chlorotica indicates that it must have photosynthesis-supporting genes within its own genome, which may have been realized through a horizontal gene transfer mechanism.
1.2 Photosynthesis: harvesting solar energy
Photosynthesis is the main source of energy for a variety of bacteria, unicellular eukaryotes (algae) and vascular plants. All these different organisms systems utilise the photosynthetic principle. The equation for photosynthesis is described as an oxidation-reduction reaction in which H2O donates electrons to reduce CO2 to produce the carbohydrate glucose, which the organism requires as its primary food source. This reaction produces the by-product O2: 6 + 6
6 + .
The photosynthetic reactions of light-dependent carbon-assimilation take place in chloroplasts (Figure 1.1a), which are intracellular organelles of several micrometres in diameter23. These structures are surrounded by two membranes, known as the outer and inner membrane. Small molecules and ions can penetrate the outer chloroplast membrane, while the inner membrane is packed with thylakoids, vesicles arranged into ribbons named granum (Figure 1.1a). The site of the thylakoid membranes where light active enzymes for photosynthesis are embedded is called lamellae (Lumen). The opposite side is named Stroma and contains enzymes needed for the carbon-assimilation reaction.
Chapter 1
5
a)
b)
Figure 1.1. (a) Schematic structure of a chloroplast. Thylakoids are arranged into granum stacks placed inside the double walled membranes (inner and outer membranes). The inside of thylakoids is known as Lumen and the outside as Stroma. (b) Transmission electron micrograph of the cross-section of a chloroplast24 Photosynthetic proteins in the thylakoid membrane absorb light in the narrow range of visible light between 400 and 700 nm. Therefore, only a small part of the overall solar electromagnetic spectrum is exploited for energy conversion. Sensitivity to red light might have originated in evolutionary processes. The first living bacteria on our planet originated at the bottom of the oceans, where photons associated with the corresponding light were applied in photosynthetic processes.
a)
b)
Figure 1.2. (a) The molecular structure of chlorophyll a and chlorophyll b. (b) The absorbance spectrum of chlorophyll a and chlorophyll b molecules with corresponding Soret, QX and QY bands respectively. Chlorophyll and carotenoid molecules are the main absorbing components in thylakoid membranes. Chlorophylls are green in colour and contain a photoporphyrin with an Mg2+ metal ion occupying the central position. Moreover, chlorophylls possess a long phytol side chain
Introduction
6
(Figure 1.2a). There are two types of chlorophyll present in chloroplasts, chlorophyll a and b, which differ by one methyl and carboxyl group, respectively (Figure 1.2b). They have slightly different absorption spectra, which allow to absorb a broader wavelength range when they act together. Most of the plants have twice as much chlorophyll a than chlorophyll b (Figure 1.2b). Some cyanobacteria such as Thermosynechococcus elongatus (T. Elongatus) contain only chlorophyll a in photosystem I (PSI), one important component in the primary process of photosynthesis. The PSI multi protein complex is able to form trimer structures to function efficiently in harsh conditions such as hot springs, which represent their natural habitat.
For efficient transfer of the absorbed light chlorophyll molecules are arranged into ring-shaped complexes in light harvesting complexes 1 (RC-LH1) and light harvesting complexes 2 (LH2) (Figure 1.3a). While LH2 is just able to harvest light and transfer it further, RC-LH1 additionally contain a reaction center (RC) with the additional function of performing charge separation (Figure 1.3b). Energy transfer between chlorophyll molecules of the same complex occurs within 1-2 ps, while energy transfer between different complexes requires up to 5 ps. Energy transfer between LH2 and RC-LH1 takes up to 35 ps. The energy absorbed by light harvesting complexes can be transferred over long distances to the reaction centres (RC) with highest efficiency. These energy transfer processes between light harvesting complexes and the corresponding charge separation processes inside the RC takes place with an internal quantum efficiency of nearly one11.
a)
b)
Figure 1.3. (a) Structure of light harvesting complex 2 (LH2). (b) Arrangement of the light harvesting complexes 1 (RC-LH1) and LH2 together. The corresponding time-scales of energy transfers are indicated by the numbers11.
Due to the that fact that chlorophyll molecules have a limited absorption spectrum, some organisms living under low light conditions, such as cyanobacteria and red algae, have additional harvesting complexes known as phycobilins. Phycoerythrobilin and phycocyanobilin containing chromophores that are linked to phycobiliproteins, which absorb in the green part of the spectrum25, 26. Other light harvesting systems use carotenoid molecules found in photosynthetic
Chapter 1
7
proteins, which may be yellow, red or purple in colour27, 28. They have both light-harvesting and stabilising functions.
1.3 PSI structure, light harvesting and charge separation mechanisms
In RC-LH1 in plants, the light harvesting function is combined with charge separation. Another example with the same properties is PSI from cyanobacteria. The PSI protein complex extracted from the cyanobacterium T. Elongatus consists of 12 protein subunits (Figure 1.4): PsaA, PsaB, PsaC, PsaD, PsaE, PsaF, PsaI, PsaJ, PsaK, PsaL, PsaM and PsaX29. One PSI protein complex contains 127 ligands: 22 ß-carotenes, 1 calcium ion, 96 chlorophyll a molecules self-assembled in the protein backbone, 3 molecules of 1,2-dipalmitoyl-phosphatidyl-glycerole, 1 molecule of 1,2-distearol-monogalactosyl-diglyceride, 2 phylloquinone molecules and 3 iron-sulphur (Fe4S4) clusters. The average mass of the ligand molecules is 30% of the total mass of the complex. PSI can trimerize under certain conditions and the PSI trimer structure has a threefold rotation C3 symmetry axis, oriented perpendicular to the membrane. The PsaL subunit interconnects the monomers within the trimer configuration. PsaA and PsaB are the biggest subunits, containing 11 transmembrane helices and coordinating the electron transport chain (ETC) inside the protein complex structure and harbouring a large number of chlorophyll molecules. The subunits PsaC, PsaD and PsaE form a crescent shape with PsaD being closest to the C3 axis. PsaC orients the two Fe4S4 clusters of FA and FB in the ETC and offers space to dock ferredoxins (double Fe4S4 clusters)29.
Solar energy harvested by chlorophyll molecules is transferred to the reaction centre P700, where the charge separation takes place within a timeframe of around one picosecond (Figure 1.4b). Within PsaA and PsaB most chlorophylls a separated by a distance of 7 Å to 16 Å (Mg2+-Mg2+ spacing), which allows rapid Föster energy transfer. The majority of chlorophyll molecules is coordinated with the side chains of histidine, tyrosine, aspartate, glutamate and glutamine residues to the protein backbone or with water molecules. There are two chlorophylls that connect the ETC with the light harvesting antenna, which serve as the energy transfer bridge from the light-harvesting ring-shaped chlorophyll arrays to the P700 centre29. Without the help of these chlorophyll couple, the light harvesting antenna would be separated from the ETC of the PSI.
At P700, the special pair of chlorophylls form a excited state with a 1.3 V potential difference and the electron is transferred via a built-in electron acceptor chain (Figure 1.4b), hosted by PsaA or PsaB. The corresponding positive charge generated at the centre of P700 remains at the opposite side of the protein complex scaffold and is re-charged by electrons from cytochrome c6 (and plastocyanin). Different organisms have different reaction centre properties that are adapted to their specific living conditions. For example, the structure of P700 in T. Elongatus is different to bacteriochlorophylls’ reaction centre (PbRC). In PbRC, the reaction centre’s chlorophylls are spaced at a larger distance of 7.6 Å compared with 6.3 Å in P700 of PSI. It is presumed that in the latter protein, - the greater overlap of their aromatic rings. PbRC has a homodimeric special pair of chlorophylls being coordinated by the same amino acid, whereas the hetero dimer of P700 in T. Elongatus is in contact with different amino acids at the
Introduction
8
Mg+2 site. The presence of histidine in and around the ETC may explain the pH effect observed in PSI’s functionality30, 31, which is not the case for PbRC. These two examples show that photo-synthetic proteins can adopt different amino acid configuration to cope with different external conditions. For T. Elongatus the optimum growth temperature is 55 ºC due to its occurrence in hot springs.
a)
b)
Figure 1.4. PSI structure of T. Elongatus. (a) Model of the PSI structure with 2.5 Å resolution29. (b) A P700 reaction centre with its electron acceptor chain29. Diagram showing the kinetics of electron transfer in PSI. After light excitation of P700 to P700*, electrons travel downstream in energy to cofactors A0, A1, FX, FA and FB
32-35.
1.4 Immobilized photosystem I proteins on metal surfaces can rectify tunnelling currents
Besides studying PSI and other photosynthetic proteins in their native membrane, a lot of biophysical investigations deal with the immobilization of such proteins on surfaces. Purified photosynthetic proteins can be self-assembled on chemically modified metal surfaces with the help of short linker molecules via electrostatic interactions and hydrogen bonding. Greenbaum et al. reported the possibility of orienting the PSI Reaction Centre (RC) on mercaptoacetic acid with 83% of PSI RC directed parallel to the surface36-38. In this case, parallel is defined as that the axis along the ETC is arranged alongside of the substrate. The linker 2-mercaptoethanol resulted in an orientation of PSI with the protein complex pointing upwards. This means, the ETC of the protein is arranged perpendicular to the surface with FB directing “upwards”. When employing 2-dimethylaminoethanethiol as directing molecule for PSI, the protein complex is pointing “downwards” in other words, the FB – side is positioned close to the surface and the ETC axis is again arranged perpendicular to the substrate. The rectification of the electron tunnelling current measured with STM was performed on a single reaction centre and used as a means of
Chapter 1
9
determining the physical orientation of PSI RCs protein complexes. Symmetrical semiconducting I-V curves were reported for the PSI RC being in the “parallel” arrangement. This interpretation was rationalized by the assumption that the ETC is effecting the tunnelling current for positive and negative biases equally (Figure 1.5a, upper panel). Rectifying I-V curves were observed for cases where PSI RCs were oriented perpendicularly on the surface in an “upwards” or “downwards” direction (Figure 1.5a, middle and bottom panel).
So far this is the first attempt to measure the orientation of PSI RC on differently functionalized surfaces. The assignment of the different orientation of PSI RC is based on a theoretical model that states that tunnelling through PSI RC proceeds preferentially via the redox elements of the natural ETC36. According to this model, electron transport is facilitated in the direction P700 to FB while tunnelling should be hindered in the opposite direction. Platinization experiments were the key to link the protein orientation to the different I-V characteristics observed in STM experiments.
a)
b)
Figure 1.5. (a) Possible orientations of the RC on a gold surface functionalized with different linker molecules36 (top panel). The PSI RC is located on the side, (middle panel) PSI with the iron-sulphur cluster facing upwards and (bottom panel) the PSI RC with iron-sulphur cluster facing down on the substrate. Different orientations resulting in varying I-V rectification measured by Scanning Tunnelling Spectroscopy (STS)36. (b) Torque moment in platinized RC on a mercaptoacetic acid modified gold surface38. Top, middle and bottom panel show the initial PSI RC orientation, electrostatically-induced torque moment, and the final change in orientation after application of a platinisation procedure. When PSI RC is incubated with the Pt-ion solution and irradiated with light, small Pt-nanoparticles are produced. It is assumed that this can be only happen at one side of the protein, i.e. the FB side where electrons are expelled due to illumination. In the case where PSI RC was oriented parallel to the surface (Figure 1.5b, top panel), platinization lead to a change in the I-V characteristics38.
Introduction
10
The symmetrical I-V curves transformed into asymmetric ones. This change was assigned to a rearrangement of the protein. This reorientation was related to electrostatic interactions. Due to platinization, the FB side gets neutralized because of the deposited metal layer on the positively charged protein surface. It was postulated that this charge unbalance induced electrostatic torque that then induces the upward movement of the FB side38 (Figure 1.5b, middle and bottom panel).
Other factors that might influence the behaviour of tunnelling currents are electrical dipoles that are induced by the protein backbone. In the case of the PSI trimer of the cyanobacterium T. Elongatus, dipole moment was observed in chemical experiments using electric birefringence and electric dichronism. This PSI permanent dipole moment is directed parallel to the C3 symmetry axis with a positive side at the P700 centre (lumen) and the negative side at the FB side (stroma)39. Another example is the inert bovine serum albumin (BSA) protein, which induced the rectification of tunnelling currents40, 41.
Apart from this permanent dipole moment, PSI has a light-induced electrical dipole moment. 100 layers of PSI that are crystallized have a well-oriented internal arrangement of protein complexes relative to each other in certain directions42. Under illumination, light-induced charge separation causes the formation of a charged dipole inside each PSI complex, which generates a big electrical dipole moment measuring up to hundreds of Debye. Light-induced photovoltage generation was measured using Kelvin Probe Force Microscopy (KPFM)42, where light intensity of 1.1 -2 generated a voltage of up to 50 V with corresponding electric fields of 100 kVcm-1. This photo voltage value is even larger than that found in inorganic materials such as LiNbO3: Fe as reported earlier42. Therefore, illumination might also influence tunnelling current through PSI proteins.
1.5 Application of photosynthetic proteins in photovoltaic devices with different materials
Besides studying photo-active proteins on a surface, they were incorporated into many complex devices such as bio-solar cells and bio-fuels cells4, 43-67. The application of the PSI reaction centre extracted from Rhodobacter sphaeroides has been widely studied in different photovoltaic systems68-76. Extracted PSI complexes can be self-assembled on metal and metal oxide surfaces for fabrication of bio-photovoltaic devices. The application of a platinisation procedure to the PSI is an efficient method for hydrogen production, as described previously77-79,54. For the fabrication of bio-inspired solar cells, the PSI RC was self-assembled on a gold electrode using a Ni2+-nitrilotriacetic acid (NTA) linker. The protein scaffold was stabilized with amphiphilic peptides and the PSI monolayer was covered with organic semiconductors. Such a bio-photovoltaic cell exhibited an internal quantum efficiency of approximately 12 %65. Another device containing PSI with both light-harvesting function and charge separation was constructed80, 81. For this study, the PSI with a molecular weight of 330 000 Da was immobilized on 3D metal oxide (TiO2 and ZnO) nanostructures80. In this case, the large PSI complex was immobilized with the help of a modified
Chapter 1
11
PsaE subunit that was able to bind to the metal oxide surface. Compared to the previous example, a liquid electrolyte was employed to supply electrons to P700+· (Figure 1.6a). Increasing the surface area of the electrode by TiO2 nanorods helped to increase the performance of bio-voltaic solar cells with an open circuit photo voltage of 0.5 V, a fill factor of 71% and a photocurrent density of 362 μA/cm2 (Figure 1.6b)80. A similar type of electrolyte cell was fabricated with graphene as a 2D electrode. PSI was extracted from spinach leaves and assembled on a graphene surface without any linker. Such a cell produced a current density of up to 550 (nA/cm2) 55.
a)
b)
Figure 6. (a) Application of PSI inside bio-photovoltaic devices80. (b). Single photocurrent PSI detection32 Instead of using 2D- and 3D-PSI layers, a nano biophotovoltaic device was constructed with a single PSI entity. This was realized with the help of a scanning near-field optical microscope (SNOM). The photocurrent of a single PSI protein complex was measured to be 10 pA under 3.5 kW/cm2 illumination (Figure 1.6b)32. The I-V curves of a single PSI showed no asymmetry but ohmic behaviour, which represents an important difference to the data observed by STM without light irradiation36, 37, 82. In order to understand the photoactive behaviour like in the SNOM experiment described above, PSI monolayer was investigated with different techniques. The effect of light on changes of the work function of self-assembled PSI on gold electrodes was studied using Kelvin Force Probe Microscopy (KFPM) and Surface Photo Voltage (SPV) techniques (Figure 1.7a)62.
Besides bio-solar cells, photodetector-type devices were constructed containing PSI. Therefore, a field-effect transistor with AlGaN/GaN active layer and top coating of PSI was prepared83. Charge redistribution under illumination was detected, which demonstrates the functionality of such bioelectronic devices83. Moreover, PSI can be attached covalently to carbon nanotubes (CNTs) enabling a change of photo conductance of CNTs under light84-86 (Figure 1.7b). These hybrid structures are promising for optoelectronic devices with self-assembly and self-powering properties on the nanoscale86.
Introduction
12
a)
b)
Figure 1.7. (a) KPFM measurement of PSI self-assembling monolayer activity under illumination62. (b) PSI can be directly attached to single-wall nanotubes (SWNTs) without covalent bonding by electrostatic interactions of the polar stroma/lumen sides or hydrophobic forces between the membrane sides of the proteins and tube (Figure 1.7b shows unpublished results).
The combination of single-wall carbon nanotubes (SWNTs) with photosynthesis was realized by Strano et al., who combined semiconducting SWNTs with membrane functionality to increase the photosynthetic activity of chloroplasts87. Light-induced activity of chloroplasts was monitored by changes in absorbance spectrum at a wavelength of 600 nm in the electron acceptor dye dich ml 1)87. The semiconducting SWNTs can be used as electron transport material in photosynthetic processes88-90. More than a three-fold increase in chloroplast activity over 6 hours was observed, which indicates that the natural photosynthesis process can be improved by synthetic additives containing SWNTs.
1.6 Scope and outline of the thesis
The major goal of this thesis is the integration of the stable Photosystem I (PSI) protein complex from T. Elongatus into solid-state biophotovoltaic cells. However, this goal cannot be reached directly. Firstly, an immobilization strategy needs to be established to cover metal and metal oxide surfaces with PSI. Strongly connected to this activity is the orientation of the PSI protein complex on the surface since this directly influences the performance of the resulting bio-solar cells. In the chapter 2, different linker molecules will be tested to direct the megadalton PSI trimer complex on a substrate. Then different techniques including atomic force microscopy (AFM), conducting AFM and liquid metal contacts of the eutectic Gallium-Indium (EGaIn) alloy were utilized to determine the absolute configuration of PSI.
Chapter 1
13
In the second chapter, PSI was incorporated into a conventional bulk heterojunction organic solar cell. The devices were constructed of a transparent ITO electrode with a PSI monolayer. On top of the biological layer, a semiconducting polymer blend with a C60-derivative was established followed by evaporation of different top electrodes resulting in conventional and inverted cells. Such devices were used to study the orientation of PSI, to determine the dipole of PSI and demonstrate PSI’s activity in an unnatural environment that is very different from the thylakoid membrane.
While in these cells, the photocurrent is determined by the organic semiconductors, in the fourth chapter, a photovoltaic devices was fabricated with PSI as the only photoactive unit. This required treatment of a PSI layer with chlorinated organic solvent and the fabrication of a semiconducting top polymer layer prepared by floating a conjugated polymer on a water surface. The latter was employed in a control device that did not involve organic solvent treatments.
In the final chapter, efforts are undertaken to increase the absorption of PSI that then might improve biophotovoltaic devices as detailed in the chapter before. The carotenoids and chlorophyll a cofactors of PSI do not absorb in the green region of the visible spectrum and therefore PSI does not fully exploit its light-harvesting and charge separation properties. In chapter four, it will be discussed how modification of PSI with organic dyes allows to broaden the absorption properties of PSI. The activity of PSI in solution was studied and the energy transfer process from the organic fluorophore to the biomolecule is proven by various optical spectroscopic techniques.
Introduction
14
1.7 References
1. Lewis, N.S. Powering the planet. MRS Bull 32, 808-820 (2007). 2. Lewis, N.S. & Nocera, D.G. Powering the planet: Chemical challenges in solar energy utilization. P. Natl. Acad. Sci. USA 103, 15729-15735 (2006). 3. Kutal, C. Photochemical conversion and storage of solar energy. J. Chem. Education, 60, 10, (1983). 4. Blankenship, R.E. et al. Comparing Photosynthetic and Photovoltaic Efficiencies and Recognizing the Potential for Improvement. Science 332, 805-809 (2011). 5. Terashima, I., Fujita, T., Inoue, T., Chow, W.S. & Oguchi, R. Green Light Drives Leaf Photosynthesis More Efficiently than Red Light in Strong White Light: Revisiting the Enigmatic Question of Why Leaves are Green. Plant Cell Physiol. 50, 684-697 (2009). 6. Bolton, J.R. Photochemical Conversion and Storage of Solar-Energy. J. Solid State Chem. 22, 3-8 (1977). 7. Bolton, J.R. The Photochemical Conversion and Storage of Solar-Energy - an Historical-Perspective. Sol. Energ. Mat. Sol. C 38, 543-554 (1995). 8. Zhu, X.G., Long, S.P. & Ort, D.R. Improving Photosynthetic Efficiency for Greater Yield. Annu. Rev. Plant. Biol. 61, 235-261 (2010). 9. Swingley, W.D. et al. Niche adaptation and genome expansion in the chlorophyll d-producing cyanobacterium Acarylochloris marina. P. Natl. Acad. Sci. USA 105, 2005-2010 (2008). 10. Wilhelm, C. & Selmar, D. Energy dissipation is an essential mechanism to sustain the viability of plants: The physiological limits of improved photosynthesis. J. Plant. Physiol. 168, 79-87 (2011). 11. Scholes, G.D., Fleming, G.R., Olaya-Castro, A. & van Grondelle, R. Lessons from nature about solar light harvesting. Nat. Chem. 3, 763-774 (2011). 12. Ham, M.H. et al. Photoelectrochemical complexes for solar energy conversion that chemically and autonomously regenerate. Nat. Chem. 2, 929-936 (2010). 13. Alstrum-Acevedo, J.H., Brennaman, M.K. & Meyer, T.J. Chemical approaches to artificial photosynthesis. 2. Inorg. Chem. 44, 6802-6827 (2005). 14. Lewis, N.S. Toward cost-effective solar energy use. Science 315, 798-801 (2007). 15. Gratzel, M. Photoelectrochemical cells. Nature 414, 338-344 (2001). 16. Oregan, B. & Gratzel, M. A Low-Cost, High-Efficiency Solar-Cell Based on Dye-Sensitized Colloidal Tio2 Films. Nature 353, 737-740 (1991).
Chapter 1
15
17. Mathew, S. et al. Dye-sensitized solar cells with 13% efficiency achieved through the molecular engineering of porphyrin sensitizers. Nat. Chem. 6, 242-247 (2014). 18. Rumpho, M.E., Summer, E.J. & Manhart, J.R. Solar-powered sea slugs. Mollusc/algal chloroplast symbiosis. Plant. Physiol. 123, 29-38 (2000). 19. Rumpho, M.E., Pelletreau, K.N., Moustafa, A. & Bhattacharya, D. The making of a photosynthetic animal. J. Exp. Biol. 214, 303-311 (2011). 20. Bergh, L.S.R. Malacologische Untersuchungen. In: C.G. Semper, Reisen im Archipel der Philippinen, Wissenschaftliche Resultate. Band 2, Heft 3: 137-176, Pls. 17-20. (1872). 21. Clark, K.B., Jensen, K.R. & Stirts, H.M. Survey for Functional Kleptoplasty among West Atlantic Ascoglossa (= Sacoglossa) (Mollusca, Opisthobranchia). Veliger 33, 339-345 (1990). 22. Mujer, C.V., Andrews, D.L., Manhart, J.R., Pierce, S.K. & Rumpho, M.E. Chloroplast genes are expressed during intracellular symbiotic association of Vaucheria litorea plastids with the sea slug Elysia chlorotica. P. Natl. Acad. Sci. USA 93, 12333-12338 (1996). 23. Soll, J. & Schleiff, E. Protein import into chloroplasts. Nat. Rev. Mol. Cell Bio. 5, 198-208 (2004). 24. http://galleryhip.com/chloroplast-pictures.html. 25. Adolphs, J. & Renger, T. How proteins trigger excitation energy transfer in the FMO complex of green sulfur bacteria. Biophys. J. 91, 2778-2797 (2006). 26. Brettel, K. Electron transfer and arrangement of the redox cofactors in photosystem I. BBA-Bioenergetics 1318, 322-373 (1997). 27. Maxwell, K. & Johnson, G.N. Chlorophyll fluorescence - a practical guide. J. Exp. Bot. 51, 659-668 (2000). 28. Mikhailov, K.M. et al. Femtopicosecond relaxation of zinc porphyrinate trimer linked by the triazole bridge. Russ. Chem. B+ 63, 76-81 (2014). 29. Jordan, P. et al. Three-dimensional structure of cyanobacterial photosystem I at 2.5 angstrom resolution. Nature 411, 909-917 (2001). 30. Liu, J.G. et al. Characterization of photosystem I from spinach: effect of solution pH. Photosynth. Res. 112, 63-70 (2012). 31. Lonergan, T.A. & Sargent, M.L. Regulation of the Photosynthesis Rhythm in Euglena-Gracilis .2. Involvement of Electron Flow through Both Photosystems. Plant Physiol. 64, 99-103 (1979). 32. Gerster, D. et al. Photocurrent of a single photosynthetic protein. Nat. Nanotechnol. 7, 673-676 (2012).
Introduction
16
33. Santabarbara, S., Heathcote, P. & Evans, M.C.W. Modelling of the electron transfer reactions in Photosystem I by electron tunnelling theory: The phylloquinones bound to the PsaA and the PsaB reaction centre subunits of PSI are almost isoenergetic to the iron-sulfur cluster Fx. BBA-Bioenergetics 1708, 283-310 (2005). 34. Muller, M.G., Niklas, J., Lubitz, W. & Holzwarth, A.R. Ultrafast transient absorption studies on Photosystem I reaction centers from Chlamydomonas reinhardtii. 1. A new interpretation of the energy trapping and early electron transfer steps in Photosystem I. Biophys. J. 85, 3899-3922 (2003). 35. Nguyen, K. & Bruce, B.D. Growing green electricity: Progress and strategies for use of Photosystem I for sustainable photovoltaic energy conversion. BBA-Bioenergetics 1837, 1553-1566 (2014). 36. Lee, I., Lee, J.W. & Greenbaum, E. Biomolecular electronics: Vectorial arrays of photosynthetic reaction centers. Phys. Rev. Lett. 79, 3294-3297 (1997). 37. Lee, I., Lee, J.W., Warmack, R.J., Allison, D.P. & Greenbaum, E. Molecular Electronics of a Single Photosystem-I Reaction-Center - Studies with Scanning-Tunneling-Microscopy and Spectroscopy. P. Natl. Acad. Sci. USA 92, 1965-1969 (1995). 38. Lee, J.W., Lee, I. & Greenbaum, E. Platinization: A novel technique to anchor photosystem I reaction centres onto a metal surface at biological temperature and pH. Biosens Bioelectron. 11, 375-387 (1996). 39. Vanhaeringen, B. et al. Simultaneous Measurement of Electric Birefringence and Dichroism - a Study on Photosystem-1 Particles. Biophys. J. 67, 411-417 (1994). 40. Ron, I. et al. Proteins as Electronic Materials: Electron Transport through Solid-State Protein Monolayer Junctions. J. Am. Chem. Soc. 132, 4131-4140 (2010). 41. Mentovich, E.D., Belgorodsky, B., Kalifa, I., Cohen, H. & Richter, S. Large-Scale Fabrication of 4-nm-Channel Vertical Protein-Based Ambipolar Transistors. Nano Lett. 9, 1296-1300 (2009). 42. Toporik, H. et al. Large Photovoltages Generated by Plant Photosystem I Crystals. Adv. Mater. 24, 2988-2991 (2012). 43. Hartmann, V. et al. Redox hydrogels with adjusted redox potential for improved efficiency in Z-scheme inspired biophotovoltaic cells. Phys. Chem. Chem. Phys. 16, 11936-11941 (2014). 44. Gorka, M., Schartner, J., van der Est, A., Rogner, M. & Golbeck, J.H. Light-Mediated Hydrogen Generation in Photosystem I: Attachment of a Naphthoquinone-Molecular Wire-Pt Nanoparticle to the A(1A) and A(1B) Sites. Biochemistry 53, 2295-2306 (2014). 45. Kothe, T. et al. Engineered Electron-Transfer Chain in Photosystem 1 Based Photocathodes Outperforms Electron-Transfer Rates in Natural Photosynthesis. Chem-Eur. J. 20, 11029-11034 (2014).
Chapter 1
17
46. Kothe, T. et al. Combination of A Photosystem 1-Based Photocathode and a Photosystem 2-Based Photoanode to a Z-Scheme Mimic for Biophotovoltaic Applications. Angew. Chem. Int. Edit. 52, 14233-14236 (2013). 47. Schartner, J. et al. Universal Method for Protein Immobilization on Chemically Functionalized Germanium Investigated by ATR-FTIR Difference Spectroscopy. J. Am. Chem. Soc. 135, 4079-4087 (2013). 48. Badura, A. et al. Photocurrent generation by photosystem 1 integrated in crosslinked redox hydrogels. Energ. Environ. Sci. 4, 2435-2440 (2011). 49. Matsumoto, K., Zhang, S.G. & Koutsopoulos, S. Enhanced Electron Transfer Activity of Photosystem I by Polycations in Aqueous Solution. Biomacromolecules 11, 3152-3157 (2010). 50. Baker, D.R., Simmerman, R.F., Sumner, J.J., Bruce, B.D. & Lundgren, C.A. Photoelectrochemistry of Photosystem I Bound in Nafion. Langmuir 30, 13650-13655 (2014). 51. Baker, D.R. et al. Comparative Photoactivity and Stability of Isolated Cyanobacterial Monomeric and Trimeric Photosystem I. J. Phys. Chem. B 118, 2703-2711 (2014). 52. Iwuchukwu, I.J. et al. Optimization of photosynthetic hydrogen yield from platinized photosystem I complexes using response surface methodology. Int. J. Hydrogen Energ. 36, 11684-11692 (2011). 53. LeBlanc, G., Chen, G.P., Gizzie, E.A., Jennings, G.K. & Cliffel, D.E. Enhanced Photocurrents of Photosystem I Films on p-Doped Silicon. Adv. Mater. 24, 5959 (2012). 54. LeBlanc, G., Chen, G.P., Jennings, G.K. & Cliffel, D.E. Photoreduction of Catalytic Platinum Particles Using Immobilized Multilayers of Photosystem I. Langmuir 28, 7952-7956 (2012). 55. Gunther, D. et al. Photosystem I on Graphene as a Highly Transparent, Photoactive Electrode. Langmuir 29, 4177-4180 (2013). 56. LeBlanc, G., Winter, K.M., Crosby, W.B., Jennings, G.K. & Cliffel, D.E. Integration of Photosystem I with Graphene Oxide for Photocurrent Enhancement. Adv. Energy Mater. 4 (2014). 57. Ciesielski, P.N., Cliffel, D.E. & Jennings, G.K. Kinetic Model of the Photocatalytic Effect of a Photosystem I Monolayer on a Planar Electrode Surface. J. Phys. Chem. A 115, 3326-3334 (2011). 58. Ciesielski, P.N. et al. Enhanced Photocurrent Production by Photosystem I Multilayer Assemblies. Adv. Funct. Mater. 20, 4048-4054 (2010). 59. Ciesielski, P.N. et al. Photosystem I - Based biohybrid photoelectrochemical cells. Bioresource Technol. 101, 3047-3053 (2010).
Introduction
18
60. Ciesielski, P.N. et al. Functionalized Nanoporous Gold Leaf Electrode Films for the Immobilization of Photosystem I. ACS Nano 2, 2465-2472 (2008). 61. Ciobanu, M. et al. Electrochemistry and photoelectrochemistry of photosystem I adsorbed on hydroxyl-terminated monolayers. J. Electroanal. Chem. 599, 72-78 (2007). 62. Carmeli, I., Frolov, L., Carmeli, C. & Richter, S. Photovoltaic activity of photosystem I-based self-assembled monolayer. J. Am. Chem. Soc. 129, 12352 (2007). 63. Badura, A., Kothe, T., Schuhmann, W. & Rogner, M. Wiring photosynthetic enzymes to electrodes. Energ. Environ. Sci. 4, 3263-3274 (2011). 64. Greenbaum, E. Platinized Chloroplasts - a Novel Photocatalytic Material. Science 230, 1373-1375 (1985). 65. Das, R. et al. Integration of photosynthetic protein molecular complexes in solid-state electronic devices. Nano Lett. 4, 1079-1083 (2004). 66. Manocchi, A.K. et al. Photocurrent Generation from Surface Assembled Photosystem I on Alkanethiol Modified Electrodes. Langmuir 29, 2412-2419 (2013). 67. Mukherjee, D., May, M., Vaughn, M., Bruce, B.D. & Khomami, B. Controlling the Morphology of Photosystem I Assembly on Thiol-Activated Au Substrates. Langmuir 26, 16048-16054 (2010). 68. Yaghoubi, H. et al. Hybrid Wiring of the Rhodobacter sphaeroides Reaction Center for Applications in Bio-photoelectrochemical Solar Cells. J. Phys. Chem. C 118, 23509-23518 (2014). 69. Tan, S.C., Crouch, L.I., Mahajan, S., Jones, M.R. & Welland, M.E. Increasing the Open-Circuit Voltage of Photoprotein-Based Photoelectrochemical Cells by Manipulation of the Vacuum Potential of the Electrolytes. ACS Nano 6, 9103-9109 (2012). 70. Yaghoubi, H. et al. The Role of Gold-Adsorbed Photosynthetic Reaction Centers and Redox Mediators in the Charge Transfer and Photocurrent Generation in a Bio-Photoelectrochemical Cell. J. Phys. Chem. C 116, 24868-24877 (2012). 71. Trammell, S.A., Wang, L.Y., Zullo, J.M., Shashidhar, R. & Lebedev, N. Orientated binding of photosynthetic reaction centers on gold using Ni-NTA self-assembled monolayers. Biosens. Bioelectron. 19, 1649-1655 (2004). 72. Kondo, M. et al. Photocurrent and Electronic Activities of Oriented-His-Tagged Photosynthetic Light-Harvesting/Reaction Center Core Complexes Assembled onto a Gold Electrode. Biomacromolecules 13, 432-438 (2012). 73. Tan, S.C. et al. Superhydrophobic Carbon Nanotube Electrode Produces a Near-Symmetrical Alternating Current from Photosynthetic Protein-Based Photoelectrochemical Cells. Adv. Funct. Mater. 23, 5556-5563 (2013).
Chapter 1
19
74. Trammell, S.A. et al. Effects of distance and driving force on photoinduced electron transfer between photosynthetic reaction centers and gold electrodes. J. Phys. Chem. C 111, 17122-17130 (2007). 75. Kondo, M. et al. Self-assembled monolayer of light-harvesting core complexes from photosynthetic bacteria on a gold electrode modified with alkanethiols. Biomacromolecules 8, 2457-2463 (2007). 76. Tan, S.C., Crouch, L.I., Jones, M.R. & Welland, M. Generation of Alternating Current in Response to Discontinuous Illumination by Photoelectrochemical Cells Based on Photosynthetic Proteins. Angew. Chem. Int. Edit. 51, 6667-6671 (2012). 77. Iwuchukwu, I.J. et al. Self-organized photosynthetic nanoparticle for cell-free hydrogen production. Nat. Nanotechnol. 5, 73-79 (2010). 78. Lubner, C.E. et al. Solar hydrogen-producing bionanodevice that outperforms natural photosynthesis. Abstr. Pap. Am. Chem. S 246 (2013). 79. Utschig, L.M., Silver, S.C., Mulfort, K.L. & Tiede, D.M. Nature-Driven Photochemistry for Catalytic Solar Hydrogen Production: A Photosystem I-Transition Metal Catalyst Hybrid. J. Am. Chem. Soc. 133, 16334-16337 (2011). 80. Mershin, A. et al. Self-assembled photosystem-I biophotovoltaics on nanostructured TiO2 and ZnO. Sci. Rep. 2 (2012). 81. Kiley, P. et al. Self-assembling peptide detergents stabilize isolated photosystem I on a dry surface for an extended time. Plos Biol. 3, 1180-1186 (2005). 82. Stamouli, A., Frenken, J.W.M., Oosterkamp, T.H., Cogdell, R.J. & Aartsma, T.J. The electron conduction of photosynthetic protein complexes embedded in a membrane. FEBS Lett. 560, 109-114 (2004). 83. Eliza, S.A. et al. Isolated Photosystem I Reaction Centers on a Functionalized Gated High Electron Mobility Transistor. IEEE T. Nanobiosci. 10, 201-208 (2011). 84. Kaniber, S.M., Simmel, F.C., Holleitner, A.W. & Carmeli, I. The optoelectronic properties of a photosystem I-carbon nanotube hybrid system. Nanotechnology 20 (2009). 85. Carmeli, I. et al. A photosynthetic reaction center covalently bound to carbon nanotubes. Adv. Mater. 19, 3901 (2007). 86. Heller, D.A. et al. Multimodal optical sensing and analyte specificity using single-walled carbon nanotubes. Nat. Nanotechnol. 4, 114-120 (2009). 87. Giraldo, J.P. et al. Plant nanobionics approach to augment photosynthesis and biochemical sensing (vol 13, pg 400, 2014). Nat. Mater. 13, 530-530 (2014). 88. Han, J.H. et al. Exciton antennas and concentrators from core-shell and corrugated carbon nanotube filaments of homogeneous composition. Nat. Mater. 9, 833-839 (2010).
Introduction
20
89. Hertel, T., Fasel, R. & Moos, G. Charge-carrier dynamics in single-wall carbon nanotube bundles: a time-domain study. Appl. Phys. Mater. 75, 449-465 (2002). 90. Scholes, G.D. & Sargent, E.H. Bioinspired Materials Boosting Plant Biology. Nat. Mater. 13, 329-331 (2014).
Chapter 2
21
Photo-electrical properties of self-assembled photosystem I monolayers
In this chapter, PSI was immobilized on a gold surface with two linker molecules, i.e. 2-mercaptoethanol (2ME) and 3-mercapto-1-propanesulfonate (MPS). Through AFM measurements it was demonstrated that these molecules orient PSI trimer molecules on average in opposite directions. Via macroscopic electrical transport measurements, the absolute configuration of the PSI molecules on the surface was determined since it was that the mechanism of charge transport occurs via tunnelling. This finding suggests that the electron transport chain is not involved in electron transport. Hence the conductivity through PSI is dominated by the intrinsic dipole of PSI. Besides looking at electron transport in the dark, light experiments with AFM were performed on single PSI trimers. In contrast to experiments in the dark, no semiconducting I-V curves were measured but curves with ohmic characteristics.
2
Photo-electrical properties of self-assembled photosystem I monolayers
22
2.1 Introduction From all light-active proteins found in Nature, protein complex Photosystem I (PSI) is a promising candidate for use in bio-electronics because of its unique property of inducing electron transfer upon light irradiation. This electron transfer process over the protein structure is exploited for biological growth in living organisms such as cyanobacteria, algae and plants. Isolated PSI protein complexes have already been applied in different environments and have showed promising performance and potential for use in solar cells1, hydrogen production units2, and spatial light modulation transmission films3. Physically, a PSI protein complex can be regarded as an electrical current rectifying diode because it features a directional electron transport chain functioning in the electron extraction process from the P700 reaction centre, located at the bottom of the PSI to the top ferredoxin docking side. These feature quality PSI as an unique and attractive protein biomaterial for studying its electrical properties by a variety of experimental techniques4-14. In this research project, we studied the entire PSI complex (not the RCI in isolation) containing electron transport chain, the light-harvesting system of chlorophylls, the reaction center and the protein scaffold that keeps it all together. Isolated PSI complexes taken from the thermophilic cyanobacterium T. Elongatus have evolutionary formed a trimer structure in order to improve light absorption and stability under harsh conditions. The PSI monomer has a polar stroma and lumen as well as an apolar region in contact with the membrane. Its dimensions measure approximately 13 × 8 × 9 nm and it contains 96 light-sensitive chlorophyll a (Chl a) molecules that are densely packed in the protein scaffold to harvest light. In the complex, harvested photons induce charge separation by chlorophylls in P700 (a special pair of Chl a molecules). Then electron transfer occurs from the primary electron donor complex to the primary electron acceptor, to the ferredoxin docking region through the built-in electron transport chain phylloquinone, FX, FA and FB (Fe4S4 clusters) (Figure 2.1a). The complex exhibits a photovoltage of around 1 V under illumination with an internal quantum efficiency close to unity. PSI can be anchored in a “downwards” orientation in which the natural flow of photo-generated electrons is towards the electrode surface (FB down), an “upwards” orientation where electrons flow in the opposite direction (P700 adjacent to the substrate), or with its electron transport vector parallel to the substrate (Figure 2.1b). This level of control over the orientation of the electron transport chain provides an opportunity to determine its role in the tunnelling transport through PSI with the help a self-assembly process rather than modifying the complexes themselves. The method of orientating PSI on metal and metal-oxide surfaces range from surface modification with different functional groups that interact electrostatically with different parts of the protein complex, to direct covalent attachment via introducing of cysteine mutants7 and SAMs with different functional head groups9, 12. One can examine the orientation of the complexes in the monolayer with atomic force microscopy (AFM) images and conducting probe AFM (CP-AFM) I-V curves.
Chapter 2
23
a)
b)
Figure 2.1 (a) Schematic representation of the electron transport chain in PSI (ET chain), FA, FX and FB. (b) Possible orientations of PSI on Au surfaces induced by different chemical modification. Photosystem I can be oriented “upwards” (left) where the flow of electrons is from the surface (P700 adjacent to substrate), a “downwards” (middle) orientation in which the natural direction of the flow of electrons is towards the electrode surface (FB down), or with its electron transport vector parallel (right) to the substrate.
Electron active proteins such as RCI, the PSI reaction centre (extracted from spinach) and full PSI protein complex (extracted from cyanobacterium, containing its own light harvesting system and charge separation reaction centre in the middle) are frequently used for device fabrication and all have different sizes and protein structures. According to previous reports on single RCI complexes, which are much smaller than the large multiprotein complex PSI, asymmetric charge transport is completely dependent on the orientation of RCI on an electrode’s surface. Greenbaum et al.,8-10 were the first to observe and explain this behaviour. They platinised one end of the photosynthetic complex directly on a surface and, during this process, the RCI complexes changed their initial orientation due to the electrostatic repulsion between positive charges on the non-platinized side of the RCI (the platinisation process can be only carried out at FB protein side) and the positive charges present in the self-assembling monolayer on the gold surface9. They used scanning tunnelling spectroscopy (STS) to examine the electronic properties of each RCI, which elicited orientation-dependent I-V asymmetry. Others have subsequently observed this behaviour in RCIs as single complexes11 and in SAMs12. This phenomenon is easily combined with light-driven processes6, which involves hopping transport (to move through the transport chain as the electron changes in energy) and therefore should not play a role in tunnelling measurements. However, the absolute orientation of the RCI and, by extension, the PSI is determined by STS data based on the assumption that the direction of asymmetry (rectification) follows the electron transport chain. It should be possible to understand the role of the electron transport chain in tunnelling currents using light experiments. Light-induced electrical currents should be enhanced in the direction of the electron transport chain or suppressed in the opposite direction. However, such light experiments employing single molecule techniques, in order to prove the role of the “activated electron transport chain”, have not yet been carried out. The first illumination experiments, which deal with the rectification and role of the electron transport chain, demonstrated a new observation of ohmic I-V curves measured in the case of a single PSI photocurrent with the help
Photo-electrical properties of self-assembled photosystem I monolayers
24
of scanning near optical microscopy (SNOM). The photocurrent produced by one PSI protein complex covalently attached to a gold surface was determined as the offset shift of the electrical current at zero applied bias6. However, the I-V characteristics recorded under dark conditions and their characteristic asymmetric behaviour due to their orientations were not reported for these SNOM experiments, which made it difficult to pinpoint the actual role of the active electron transport chain under illumination6. In this study, the electrical current rectification properties of PSI protein complexes have been investigated using CP-AFM and EGaIn. The latter technique uses a small drop of liquid metal consisting of an eutectic mixture of Gallium and Indium to probe the electrical properties of a surface of molecules assembled at a metal substrate15, 16. Both techniques have been employed to investigate the physical properties of the PSI in absence of light irradiation regarding its conducting properties through a single protein and protein monolayers. The temperature dependent tunnelling currents measured, clearly demonstrated its “non-hopping” nature. The conductivity of light-generated electrons via the built-in transport chain should be a hopping process. Rectification as a result of the PSI neutral dipole moment was proposed to explain the asymmetric tunnelling current based on the dipole moment of the electrical field formed locally within the PSI due to the protein scaffold. Additionally, we observed that the I-V characteristics changed from “semiconducting” to “ohmic” behaviours under illumination, which has not been reported before in the literature. The observed transition in conductivity limits the application of the I-V rectification scenario as a tool for determining the orientation of photosynthetic proteins under illumination. The light-generated electrons under illumination do not influence the symmetry of the measured I-V curves in CP-AFM at positive and negative applied biases i.e. its enhancement. This can be explained by the fact that the light induced electrons at the P700 reaction centre of the PSI are powered by an applied external electrical field between the metal surface and the CP-AFM conducting probe, but not by the electron acceptor transport pathway built into the protein scaffold.
2.2 Characterization of PSI proteins with conducting AFM We exposed ultra-flat template-stripped Au (AuTS) substrates to a solution of 1 mM 2-mercaptoethanol (2ME)9 or sodium 3-mercapto-1-propanesulfonate (MPS)17 to form “director SAMs” to bias the orientation of PSI trimers that self-assemble on top of these SAMs.12 These director SAMs differ in length by 1.7 Å, which may affect the magnitude of the tunnelling current. However this magnitude is not used as a benchmark in this paper and small changes in the thickness of the director SAM are unlikely to influence asymmetry. The PSI in cyanobacteria has an asymmetric distribution of surface charges in the stroma and lumen sides (Figure 2.2). Two thirds of all the charged surface residues are concentrated at the “top” of the complex; the stromal part or FB electron acceptor side (Figure 2.2). This difference is likely to be what determines the preferred orientation of PSI during self-assembly on a modified gold surface. We
Chapter 2
25
immobilized PSI on the substrates by drop casting from an aqueous buffer and incubating them (i.e. leaving them in contact) for two hours. We investigated each SAM topographically with AFM and electrically using CP-AFM. Also, we define the asymmetry of current transport
)/I (+)| with respect to the wiring convention for CP-AFM (sample substrate is biased).
Figure 2.2 The structure of PSI protein complex extracted from T. Elongatus has more charges on the top (at Fe4S4) compared to the bottom (at P700 reaction centre). Two-thirds of all charges are concentrated on the top of the PSI protein complex, i.e. the part protruding to the stroma, which represents the internal chloroplast space. Amino acids are depicted as Lysine (red), Arginine (blue), Aspartic Acid (pink) and Glutamic Acid (white).
The recorded AFM height profile images show higher PSI coverage for 2ME than for MPS, with protein heights close to 6 nm on AuTS (Figure 2.3). This value is smaller than the 9 nm thickness derived from crystal structure data because these SAMs are measured under ambient, anhydrous conditions and contact with the AFM tip can compress them somewhat. We imaged individual PSI trimers within SAMs of PSI formed on both directing SAMs by AFM to determine the density. For 2ME, 853 PSI trimmers per μm2 and for MPS, 723 trimers per μm2 were measured. These different densities can be translated into surface coverage values. For 2ME and MPS coverages of 50% and 45% were calculated, respectively. Tapping mode imaging was performed at three different areas to account for surface inhomogeneity of SAMs. Next, CP-AFM electrical measurements were performed using the TUNA extension mode for AFM Multimode 8, where the conducting Pt/Ir coated doped Si probe was grounded and the gold surfaces with self-assembled linkers and PSI were biased with positive and negative voltages respectively for the I-V recording. The force applied to the CP-AFM probe was set as low as possible to minimise protein deformation during measuring, but was kept sufficiently high to record the tunnelling current via selected PSI protein complexes. A force of less than 10 nN was employed.
SIDE TOP BOTTOM
Photo-electrical properties of self-assembled photosystem I monolayers
26
a)
b)
Figure 2.3. (a) and (b) an AFM image of the PSI coverage on the two surfaces modified with the corresponding linker molecules 2ME and MPS respectively.
These measurements allow to determine the orientation of single PSI trimmers according to Figure 2.1b8-10. The percentage PSI orientation defined by I-V rectification of diode-like curves (downwards, upwards and laterally) for each directing SAM molecule was determined by sampling 100 proteins in the dark. The results are summarized in Table 2.1. For 2ME, the majority of PSI is directed “downwards” (57%). In case of the other SAM molecule, MPS, 69% of the trimmers are oriented “upwards”. The “parallel” orientation of PSI molecules is similar for both SAMs. The distribution ratio of I-V characteristics observed for 2ME is very similar to that measured previously in the literature for much smaller RCI8-10. Table 2.1. Percentage of average orientation of PSI depending on the different directing SAMs.
SAM Down (%) Parallel (%) Up (%) 2ME 57 25 18 MPS 11 20 69
2.3 Electrical current rectification based on neutral PSI dipole moment In the previous paragraph, the orientation of PSI was determined on two directors SAMs based on their rectification properties. Here, we study the mechanism of this rectification process. The PSI protein complex has a permanent dipole as a result of the composition of the amino acids8. The value of the PSI dipole moment can be calculated theoretically18, 19 and in the case of the PSI trimer protein (used in our research) its direction was determined experimentally20. This dipole moment may result in the observed rectification as well (asymmetric recorded I-V curves) due to its internal electrical field, which is analogous to the rectification of well-studied self-assembled monolayers of surface-immobilized molecules with electron-withdrawing and electron-donating
Chapter 2
27
groups (Donor-Acceptor)21. However, the implementation of the PSI dipole rectification model contradicts the “evolutionary developed electron transport pathway” model, which has been proposed previously8-10. The positive side of the PSI dipole is located at the P700 side, while the negative side is located at FB side20. Therefore, the electrons need to travel against the dipole direction when moving on the electron acceptor chain after light illumination. That means two opposite rectification mechanisms are present in one PSI protein complex. If the current we measured by CP-AFM are due to the intrinsic dipole of PSI one would expect a charge transport mechanism relaying on the direct tunnelling between CP-AFM probe and Au substrate, which is temperature independent. In contrast, if the currents we measured follow the electron transport chain one would expect a transport via hopping on individual PSI redox centers. This process is temperature depended. Since temperature depended measurements with CP-AFM are hard to realize, EGaIn experiments were performed on the PSI trimer monolayers on the two director SAMs. These experiments were carried out in the group of Prof. R. Chiechi by Olga Castañeda Ocampo22. In Figure 2.4a, semi-logarithmic I-V (J-V, where J – is current density) curves are shown for the two PSI configurations with EGaIn as top electrode. The monolayer of PSI oriented “downwards” with the help of 2ME shows a higher rectification current than the protein layer directed “upwards” by MPS. In these experiments, the substrate was grounded. Both the EGaIn and the CP-AFM measurements can be directly compared because the PSI layers were prepared under identical conditions on the same surfaces. For this comparison, 50 I-V curves were recorded for both orientations with CP-AFM and averaged (Figure 2.4b). The resulting semi-logarithmic plots show the same trend as the EGaIn measurements19, 16, 23, 24. The 2ME directing SAM for PSI shows higher rectification than the MPS layer. Note that in this case the wiring was configured in such a way that the Pt/Ir tip was grounded. The fact that both techniques result in similar findings can also be seen when plotting the I-V curves for both directing SAMs in a conventional way (Figure 2.4c and 2.4d). The I-V curves of averaged CP-AFM measurements and macroscopic EGaIn experiments coincide for PSI on 2ME and MPS, respectively. In the next step, the mechanism of electron transport through PSI trimers was investigated. Since CP-AFM does not allow recording of currents under large temperature variation EGaIn experiments were performed to address this question.
Photo-electrical properties of self-assembled photosystem I monolayers
28
Figure 2.4. Top: Semi-log plots of current and current density versus voltage for junctions measured using EGaIn (a) and CP-AFM (b) for SAMs of PSI on MPS (purple) and 2ME (black). These data are plotted according to the normal wiring of each technique (see the Supporting Information for details). The horizontal, dashed lines are to guide the eye. Bottom: Per-complex J-V curves for SAMs of PSI on 2ME (c) and MPS (d) measured by CP-AFM (black squares) and EGaIn (green circles) plotted with respect to the standard wiring of CP-AFM (shown in the insets). Per-complex values of J for EGaIn were calculated using number densities of PSI measured by AFM and a correction factor for the difference between the measured and effective area of the EGaIn junctions22.
Microfluidic devices fabricated following the protocols found in the literature were used to acquire J-V traces at different temperatures16, 23. The temperature was changed between 298 to 198 K. The plot of the ln |J| at ±0.50 V for PSI on both directing SAMs is described in Figure 2.5. It can be seem that the values of ln|J| do not vary over the broad chosen temperature range. These data are indicative for a charge transport mechanism that solely relies on tunnelling of electrons through PSI for both SAMs consisting of 2ME and MPS. Moreover, these results suggest that hopping is not involved in electron transport. On the other hand this means that the electrons do not travel via the redox elements of the electron transport chain.
Chapter 2
29
Figure 2.5. ln |J| at ±0.50 V as a function of inverse temperature for PSI on directing SAMs of MPS (triangles) and
negative bias (-0.50 V). The linearity indicates that the mechanism of charge transport is dominated by tunneling asno temperature dependence was measured.
a)
b)
c)
Figure 2.6. The direction of the electrical field that arises from the PSI dipole moment (dashed lines) within PSI-
EGaIn devices, which are shown with EGaIn biased positively (with respect to the normal wiring of EGaIn). The
direction of this field goes from positive to negative in the complex. (a) When PSI is oriented “up,” the electric field
from the applied bias opposes the internal electric field of the PSI complexes. (b) When PSI is oriented “down,” the
direction of the internal electric field is the same as the applied bias. Thus, this mechanism predicts that PSI in the
down orientation will give higher values of R. (c) Energy level diagram across AuTS-PSI(P700/FB)//Ga2O3/EGaIn
junctions. The barrier width is defined by the thickness of one oriented PSI complex, which is depicted in the
“down” orientation with respect to the natural direction of electron flow. The green lines are the frontier orbital
energies of the chlorophyll molecules, which are distributed evenly through the thickness of the PSI complex. The
black lines represent the energies of the electron transport chain and their relative spatial positions. Based on the
orientation of the electron transport chain, more current should flow when the EGaIn electrode is biased negatively
than at the equivalent positive bias. That mechanism would translate into higher values of R when the complexes
are oriented “up” (because this figure is drawn with respect to the wiring of EGaIn junctions; Figure 2.4 shows the
data with respect to the wiring diagram of CP-AFM). The distances between co-factors were estimated with the
software PyMOL from a crystal structure of PSI taken from the Protein Data Bank (1JB0).
5.1Au
4.5Ga2O3EGaIn
- V
+ V
Chla
Chla*
P700
A0A1A2
P700*
FXFAFB
1.5 nm 0.6 nm 1.3 nm 0.5 nm 4.1 nm 1.0 nm
Photo-electrical properties of self-assembled photosystem I monolayers
30
After having determined the mechanism of charge transport through PSI, we can now make a statement about the absolute orientation of PSI on both directing SAMs. The permanent dipole of PSI that was theoretically calculated and experimentally determined by van Haeringen et al.20 is characterized by a positive side at the P700 face while the negative side is located at the FB face. The direction of the dipole of PSI is parallel to the C3-symmetry axis of PSI trimer. The electronic field of this dipole moment either enhances or opposes the total field that is generated by biasing the electrodes within the EGaIn experiments (Figure 2.6a and 2.6b). In the case of 2ME PSI is oriented in a “downwards” direction because we measure higher rectification current at positive voltage compared to MPS. On the other hand MPS orients the majority of PSI trimmers “upwards” since we detect lower rectification at positive voltage compared to 2ME. These results can be confirmed by looking at the energy levels of the electrodes, PSI and the redox centers of the electron transport chain (Figure 2.6c). This picture of the latter component was taken from Nakamura’s observations25. It is clearly visible that during applying a bias of ±1V one does not reach the energy levels of the redox centers of PSI belonging to the electron transport chain (Figure 2.6c).
2.4 Conducting properties of PSI under illumination After studying the electron transport through PSI in the dark and establishing a tunnelling mechanism not involving the electron transport chain, the question arises of how transport of the electrons occurs under illumination. Again CP-AFM measurements were performed to solve this question. A halogen lamp (400-700nm) was used as a single light source with a power of 10-30 -2 for short time sample illumination when recording I-V curves. The same lamp and intensity was later employed in this thesis to follow oxygen consumption of PSI (see Chapter 5). For the acquisition of I-V curves by CP-AFM, a PSI trimer was selected by height image and the tip was placed on the top of the protein complex with a force of less than 10 nN. Firstly, trace and retrace curves were measured between -1.5 and +1.5 V. Subsequently, the lamp was switched on and the I-V curves on the same PSI trimer were recorded. For both directing SAMs, 2ME and MPS, and in the dark and under illumination 50 PSI trimmers were sampled individually resulting in a total of 200 I-V curves. For each configuration, the I-V curves were averaged and plotted (Figure 2.7a). In absence of light, the I-V curves adopt a S-shape indicating semiconducting behaviour. Thereby, similar rectification characteristics were observed as mentioned above (Figure 2.4b) for PSI directing SAMs consisting of 2ME and MPS. Once the light was switched on, the I-V curves changed significantly. A transition from semiconducting to ohmic was observed for both director SAMs. Under light illumination, a larger current for 2ME was measured than for MPS. After recording the I-V under illumination, the light was switched off and semiconducting S-shape I-V curves were restored. These experiments clearly show that exciting the chlorophyll a molecules within PSI increases the current flow through the protein complex at both positive and negative bias. To put our results in perspective, we performed similar CP-AFM experiments with bovine serum albumin (BSA),
Chapter 2
31
which represents a non-photoactive protein and does not contain any light-sensitive cofactors or any redox units. Again 50 I-Vs in light and in the dark were recorded on 2ME each. The averaged curves are plotted in Figure 2.7b. Asymmetric I-V curves were obtained for BSA indicating tunnelling through the protein scaffold. However, in contrast to PSI, the effect of illumination on BSA was negligible. Previously, scanning tunnelling microscopy (STM) and spectroscopy was performed on single light harvesting complex 2 (LHC 2), which does not exhibit an electron transport chain but possess light harvesting chlorophylls26. These measurements showed increased tunnelling current upon light irradiation, however, a complete transition from “semiconducting” to “ohmic” behaviour, as found in our experiments with PSI, was not observed26. We believe that the increase in tunnelling current under illumination and the appearance of ohmic behaviour are related to the increased generation of carriers within PSI under illumination in a dry state compared to the dark. Due to the high amount of carriers generated by light within PSI the rectifying properties of the intrinsic dipole of PSI become less dominant and the conductivity solely gets determined by the external electric field. In this interpretation of the results, the contribution of electron acceptors of the electron transport chain of PSI is minor. a)
-1.5 -1.0 -0.5 0.0 0.5 1.0 1.5-4
-3
-2
-1
0
1
2
3
4
I (nA
)
Voltage (V)
2ME Light OFF 2ME Light ON MPS Light OFF MPS Light ON
b)
-1.5 -1.0 -0.5 0.0 0.5 1.0 1.5-2000
-1500
-1000
-500
0
500
1000
1500
2000
BSA Light OFF BSA Light ON PSI Light OFF PSI Light ON
Voltage (V)
I (pA
)
Figure 2.7 Averaged curves of CP-AFM on PSI protein complexes in dark conditions and under illumination on different SAMs. (a) The PSI protein complexes oriented on both linkers 2ME and MPS show enhanced currents for both cases. (b) The reference CP-AFM measurements with BSA proteins did not show an increase in current under illumination under the same measuring conditions as for PSI.
A similar ohmic behaviour of PSI under light illumination was observed in SNOM experiments with single PSI complex. Also the magnitude of absolute current was with 2 nA at ±1 V6 comparable to the values measured here. While in SNOM6 experiments an offset shift of 10 pA was detected that was ascribed to the photocurrent of a single PSI protein complex, here, we do not observed this photo-generated current with our CP-AFM setup.
Photo-electrical properties of self-assembled photosystem I monolayers
32
2.5 Conclusions In this chapter, were have immobilized PSI trimers on an Au-surface with two different linker molecules 2-mercaptoethanol (2ME) and 3-mercapto-1-propanesulfonate (MPS). These directing SAMs orient PSI trimers in on average opposite directions. 2ME forces the FB side of PSI towards the surface while MPS induces the P700 side of PSI onto the substrate. This absolute orientation of PSI was determined by a set of different measurements. With CP-AFM the absolute orientation of PSI ratio of PSI protein complexes on the surface was determined by recording characteristic asymmetric I-V curves. These experiments were complemented by macroscopic EGaIn measurements on the same PSI monolayers. Temperature depended conductivity clearly demonstrated that the mechanism of electron transport is tunnelling. This result led to the conclusion that the electron acceptors of the PSI transport chain are not involved in transfer of electrons. For that reason the rectification must be due to the intrinsic dipole of PSI originating from the protein scaffold. It is noteworthy that the dipole induces opposite rectification as one would expect from the electron transport chain of PSI. Finally, we found that under illumination the conducting properties of PSI change completely compare to the dark. In absence of light, asymmetric I-V curves are diagnostic for the rectifying property of PSI. Under illumination, however, ohmic behaviour and a significantly larger conductivity are no longer determined by its intrinsic dipole.
2.6 Methods Fabrication of monolayers. Immobilization of PSI, we immersed the substrates in a solution of 1mM 2-mercaptoethanol (2ME) or sodium 3-mercapto-1-propanesulfonate (MPS) to direct the PSI complexes to adopt a down or up orientation (FB iron-sulfur cluster adjacent or away from substrate). The time of immersion was limited to 2 hours to avoid the formation of multilayers or aggregates. After this step, were rinsed the substrates with MQ water (MPS) or ethanol (2ME), dried them with nitrogen and incubated them in a previously prepared PSI solution. The PSI solution consisted of 1:1 in buffer A (20 mM HEPES (pH 7.5); 10 mM MgCl2; 10 mM CaCl2; 500 mM Mannitol with 0.05% DDM (n-Dodecyl- -D-maltoside) for 2 h. They were then rinsed with MQ water and dried with nitrogen. Monolayer Surface Characterization. We analyzed each substrate by AFM after each fabrication step: on the Au substrate after cleavage, after surface modification, and after incubation on PSI solution. We used the resulting images to measure the surface coverage of PSI. This analysis revealed a true surface coverage of up to 50%. This was calculated by knowing the diameter of the PSI trimer from TEM images and the coverage density for specific area. We obtained the AFM images with MultiMode 8 with ScanAsyst Microscope in tapping mode with TESP probes (Bruker) with spring constant k = 42 N·m 1, resonance frequency f = 320–410 kHz and tip radius of less than 10 nm. The scan rate and resolution were 1 Hz and 640 lines/sample respectively. We analysed the AFM images with the software NanoScopeAnalysis 1.2 from Bruker. We studied the conductivity of the immobilized PSI on the two orienting monolayers with AFM Tunneling
Chapter 2
33
Atomic Force Microscopy (TUNA) contact mode with a conducting probe. This mode was applied for electrical characterisation of single (trimer) PSI complex with Pt/Ir coated Si n-type probe (APPNano), spring constant k = 0.02-0.8 N·m 1, resonance frequency f = 5-25 kHz and tip radius less than 30 nm with contact resistance of 0.01-0.025 ohm/cm. Statistic data was performed over 100 independent measured points for each orienting SAM. The applied force to CP-AFM conducting probe on top of PSI was started from low and step-by-step increased to reach contacting for I-V recording. This approach was used for each measurement point with forces less than 10 nN.
Photo-electrical properties of self-assembled photosystem I monolayers
34
2.7 References
1. Mershin, A. et al. Self-assembled photosystem-I biophotovoltaics on nanostructured TiO2 and ZnO. Sci. Rep. 2 (2012). 2. Iwuchukwu, I.J. et al. Self-organized photosynthetic nanoparticle for cell-free hydrogen production. Nat. Nanotechnol. 5, 73-79 (2010). 3. Carmeli, I. et al. Spatial modulation of light transmission through a single microcavity by coupling of photosynthetic complex excitations to surface plasmons. Nat. Commun. 6 (2015). 4. Golbeck, J.H. Structure and Function of Photosystem-I. Annu. Rev. Plant. Phys. 43, 293-324 (1992). 5. Das, R. et al. Integration of photosynthetic protein molecular complexes in solid-state electronic devices. Nano Lett. 4, 1079-1083 (2004). 6. Gerster, D. et al. Photocurrent of a single photosynthetic protein. Nat. Nanotechnol. 7, 673-676 (2012). 7. Carmeli, I., Frolov, L., Carmeli, C. & Richter, S. Photovoltaic activity of photosystem I-based self-assembled monolayer. J. Am. Chem. Soc. 129, 12352(2007). 8. Lee, I., Lee, J.W., Warmack, R.J., Allison, D.P. & Greenbaum, E. Molecular Electronics of a Single Photosystem-I Reaction-Center - Studies with Scanning-Tunneling-Microscopy and Spectroscopy. P. Natl. Acad. Sci. USA 92, 1965-1969 (1995). 9. Lee, I., Lee, J.W. & Greenbaum, E. Biomolecular electronics: Vectorial arrays of photosynthetic reaction centers. Phys. Rev. Lett. 79, 3294-3297 (1997). 10. Lee, J.W., Lee, I. & Greenbaum, E. Platinization: A novel technique to anchor photosystem I reaction centres onto a metal surface at biological temperature and pH. Biosens. Bioelectron. 11, 375-387 (1996). 11. Stamouli, A., Frenken, J.W.M., Oosterkamp, T.H., Cogdell, R.J. & Aartsma, T.J. The electron conduction of photosynthetic protein complexes embedded in a membrane. FEBS Lett. 560, 109-114 (2004). 12. Mikayama, T. et al. The Electronic Behavior of a Photosynthetic Reaction Center Monitored by Conductive Atomic Force Microscopy. J. Nanosci. Nanotechno. 9, 97-107 (2009). 13. Frolov, L., Rosenwaks, Y., Carmeli, C. & Carmeli, I. Fabrication of a photoelectronic device by direct chemical binding of the photosynthetic reaction center protein to metal surfaces. Adv. Mater. 17, 2434 (2005). 14. Reiss, B.D., Hanson, D.K. & Firestone, M.A. Evaluation of the photosynthetic reaction center protein for potential use as a bioelectronic circuit element. Biotechnol. Progr. 23, 985-989 (2007).
Chapter 2
35
15. Chiechi, R.C., Weiss, E.A., Dickey, M.D. & Whitesides, G.M. Eutectic gallium-indium (EGaIn): A moldable liquid metal for electrical characterization of self-assembled monolayers. Angew. Chem. Int. Edit. 47, 142-144 (2008). 16. Dickey, M.D. et al. Eutectic gallium-indium (EGaIn): A liquid metal alloy for the formation of stable structures in microchannels at room temperature. Adv. Funct. Mater. 18, 1097-1104 (2008). 17. Yamanoi, Y., Terasaki, N., Miyachi, M., Inoue, Y. & Nishihara, H. Enhanced photocurrent production by photosystem I with modified viologen derivatives. Thin Solid Films 520, 5123-5127 (2012). 18. Felder, C.E., Prilusky, J., Silman, I. & Sussman, J.L. A server and database for dipole moments of proteins. Nucleic Acids Res. 35, W512-W521 (2007). 19. Antosiewicz, J. & Porschke, D. Electrostatics of Hemoglobins from Measurements of the Electric Dichroism and Computer-Simulations. Biophys. J. 68, 655-664 (1995). 20. Vanhaeringen, B. et al. Simultaneous Measurement of Electric Birefringence and Dichroism - a Study on Photosystem-1 Particles. Biophys. J. 67, 411-417 (1994). 21. de Boer, B., Hadipour, A., Mandoc, M.M., van Woudenbergh, T. & Blom, P.W.M. Tuning of metal work functions with self-assembled monolayers. Adv. Mater. 17, 621 (2005). 22. Castaneda Ocampo, O.E. et al. Mechanism of Orientation-Dependent Asymmetric Charge Transport in Tunneling Junctions Comprising Photosystem I. J. Am. Chem. Soc. 137, 8419-8427 (2015). 23. Nijhuis, C.A., Reus, W.F., Barber, J.R., Dickey, M.D. & Whitesides, G.M. Charge Transport and Rectification in Arrays of SAM-Based Tunneling Junctions. Nano Lett. 10, 3611-3619 (2010). 24. So, J.H. & Dickey, M.D. Inherently aligned microfluidic electrodes composed of liquid metal. Lab Chip 11, 905-911 (2011). 25. Nakamura, A., Suzawa, T., Kato, Y. & Watanabe, T. Species Dependence of the Redox Potential of the Primary Electron Donor P700 in Photosystem I of Oxygenic Photosynthetic Organisms Revealed by Spectroelectrochemistry. Plant Cell Physiol. 52, 815-823 (2011). 26. Lukins, P.B. Single-molecule electron tunneling spectroscopy of the higher plant light-harvesting complex LHC II. Biochem. Bioph. Res. Co. 256, 288-292 (1999).
Chapter 2
36
Chapter 3
37
Photo-physical properties of photosystem I inside solid-state solar cells
Biomimetic strategies attracted an alluring attention for solar energy conversion in recent research efforts. Physical principles of light harvesting and energy transfer found in photosynthesis were implemented into solar cells and photo-active natural biomacromolecules integrated into existing solar technologies. Here, the large multi-protein complex photosystem I (PSI), which is at the heart of light-dependent reactions in photosynthesis, was integrated into a bulk heterojunction (BJH) solid-state organic solar cell. Thereby, the BHJ serves as a novel biophysical tool to determine the photo-induced dipole and the orientation of PSI on a substrate surface. Moreover, our experiments demonstrated that photoactive megadalton-protein complexes like PSI are compatible with solution processing of organic solar cells.
Gordiichuk, P. I., Wetzelaer, G.-J. A. H., Rimmerman, D., Gruszka, A., de Vries, J. W., Saller, M., Gautier, D. A., Catarci, S., Pesce, D., Richter, S., Blom, P. W. M. and Herrmann, A. Advanced Materials, 26, 4863–4869 (2014).
3
Photo-physical properties of photosystem I inside solid-state solar cells
38
3.1 Introduction
Solar cell technology has undergone dramatic changes over time, producing a vast diversity of approaches for light-to-energy conversion. The transition from inorganic to organic photosensitive materials represents a significant milestone in solar cell evolution. Organic dye-sensitized solar cells (DSSCs) are fabricated by immobilizing Ru-based light-absorbing chelates on a meso-porous TiO2 working electrode in a two-electrode system. When illuminated, the organic dye is elevated to an excited state and injects electrons into the working metal-oxide electrode while an electrolyte solution shuttles electrons from the counter electrode to the dye, providing the voltage gradient required to perform electrical work.1, 2 Another important advance toward practical organic photovoltaic devices has been the realization of “plastic” solar cells, or bulk heterojunctions. The active layer of such devices is composed of a microphase-separated morphology consisting of a p-type semi-conducting polymer and an electron acceptor moiety.3-5 In contrast to DSSCs, these organic solid state devices can potentially be fabricated at low cost by all-solution processing and printing technologies even on flexible substrates.
Most recently, biomimetic strategies inspired the plants and photosynthetic organisms have been utilized for solar energy conversion. Thereby, processes occurring during photosynthesis, such as dynamic self-repair, light harvesting and quantum effects have been integrated into man-made photovoltaic devices.6-9 Similarly, proteins that enable the natural photosynthesis process are integrated into existing solar energy technologies.10-12 One of the most frequently used photo-active building blocks for that purpose is the multi-protein complex photosystem I (PSI). PSI contains a large antenna system in which light is harvested by photosynthetic pigments that absorb at distinct wavelengths and funnel the excitation energy to the special pair of chlorophylls (P700), where charge separation takes place. The high-energy electron travels via the primary electron acceptors A0 (Chla), A1 (phylloquinone), FX, FA and FB (Fe4S4 clusters) within the complex to ferredoxin. From there, the excited electron can either follow the cyclic- or noncyclic phosphorylation pathway to form ATP and redox-equivalents, respectively, finally enabling carbohydrate production.13 The redox cycle is completed by re-reduction of P700+• by cytochrome c6. PSI is characterized by internal quantum efficiency close to 100%, which is one of the reasons why it has been employed in bio-inspired solar-energy conversion systems. Towards this goal, various PSI immobilization strategies have been evaluated with respect to the photoelectric properties of surface-immobilized PSI complexes and isolated reaction centres.14-20 PSI stabilized with peptide detergents showed long-term stability and functionality under dry conditions on solid surfaces.21 Based on these findings bio-photovoltaic devices similar to DSSCs were fabricated by self-assembly of PSI on 3D nanostructured semiconductor electrodes using a liquid electrolyte as redox mediator.22 However, this architecture requires the use of a bio-friendly electrolyte solution that must be confined and sealed. To the best of our knowledge, the realization of a full solid-state bio-photovoltaic device with the favourable characteristics of an organic bulk heterojunction remains elusive.
Chapter 3
39
Here, we introduce the implementation of PSI in organic electronic devices that combine the ease of processing of organic semiconductors with the bio-photovoltaic activity of PSI. The devices enable us to characterize biophysical properties of the photosynthetic multi-protein complex like its orientation on a surface. Such properties can otherwise only be determined by single molecule experiments employing scanning probe microscopy techniques.
3.2 Immobilization of PSI on a metal oxide surface
For our study, PSI from the cyanobacterium T. Elongatus was employed (Figure 3.1a). This photosynthetic complex contains 96 chlorophylls and 22 carotinoids for light absorption and can exist as a trimer in vivo with an overall molecular mass Mr of 1 068 kDa (Figures 3.1b and 3.1c).23, 24
a)
b)
c)
d)
e)
400 500 600 700 8000.001
0.002
0.003
0.004
0.005
0.006
0.007
0.008
0.009
Abso
rban
ce
Wavelength (nm)
ITO/SAM/PSI
Figure 3.1. (a) Structure of a PSI monomer from T. Elongatus showing the polypeptide backbone and photosynthetic pigments.24 (b) Structure of the trimer of PSI with a height of 6 nm and diameter of around 25 nm. (c) TEM image of PSI trimers employed for the preparation of solar cells. (d) AFM picture of a self-assembled monolayer of PSI trimers on an ITO surface measured in tapping mode. The substrate was pre-treated with dihydroxyacetone phosphate for adsorption of PSI. (e) Visible absorption spectra of a self-assembled monolayer of PSI on a ITO surface that was treated with dihydroxyacetone phosphate.
After purification of the trimeric protein complex (see Experimental section), a dense PSI monolayer was prepared in two steps on transparent metal oxide substrates. First, a self-assembled monolayer of dihydroxyacetone phosphate was established on indium-tin oxide (ITO) supported on glass substrates.25 In the second step, the metal oxide surfaces were immersed in a
Photo-physical properties of photosystem I inside solid-state solar cells
40
PSI solution. As a result, a dense monolayer of PSI was formed, as evidenced by AFM measurements (Figure 3.1d and 3.2). Counting of the PSI trimers on ITO was performed on different areas and revealed an average number of 1.7×1015 m-2 PSI monomers corresponding to a surface coverage of 50%. This high coverage allowed recording of an absorption spectrum of a PSI monolayer assembled on the transparent substrates under dry conditions, where the characteristic absorption peaks of PSI were detected at 430 and 660 nm (Figure 3.1e). These measurements indicate that the dihydroxyacetone phosphate linker is well suited for immobilization of PSI by binding to the metal-oxide surface and interacting with the polar stroma and lumen faces of PSI by electrostatic and hydrogen bonds.
a)
b)
Figure 3.2. AFM height profile of ITO surface used for device fabrication before (a) and after (b) PSI immobilization. The z-scale is represented as colour code on the left of each picture with a range of 29 nm.
3.3 PSI stability study under organic solvent treatment
For incorporation of PSI into organic BHJs, it needs to be tested whether the biomacromolecular complex is compatible with the fabrication process. The effect of the organic solvent chlorobenzene on PSI structure and its surface coverage was studied with the help of AFM. PSI was immobilized in a similar way on a gold surface prepared by the template stripping method
with the directing linker molecule 2-mercapto ethanol (concentration 1 mM in ethanol) by incubation for 12 h 26. The hydroxyl groups introduced in this way induce a good PSI coverage after 2 h immobilization time in the dark (Figure 3.3a). The AFM measurements in tapping mode were recorded before and after spin-monolayer at 2000 RPM speed and accelerating 2000 RPM/s (Figures 3.3a and 3.3b). The heights of PSI complexes in an imaging area of 1.5 by 1.5 μm were measured manually and plotted as height histograms shown in Figures 3.3c and 3.3d. The organic solvent treatment induces a minor height reduction of PSI of less than 0.5 nm. Moreover, the spin coating of chlorobenzene results in decreasing surface coverage of PSI by 18% compared to the non-treated
Chapter 3
41
area. From these measurements, it can be concluded that the most of PSI trimers on the surface remain structurally unaffected by the short treatment with organic solvent.
a)
b)
c)
3 4 5 6 7 8 9 1005
10152025303540455055 Gaussian Fit
Before chlorobenzene treatment
Height distribution, nm
N
d)
3 4 5 6 7 8 9 1005
1015202530354045 Gaussian Fit
After chlorobenzene treatment
Height distribution, nm
N
Figure 3.3. The PSI protein complexes denaturation on a gold surface under chlorobenzene treatment studied by AFM. (a) The height profiles with corresponding height histograms (c) of a fabricated PSI film before treatment. (b) The height profile of a PSI film after 1 time spin coating of chlorobenzene solvent with corresponding (d) height histogram.
3.4 Fabrication BJH solar cells with PSI monolayer
After demonstrating that PSI from thermophilic origin resists short organic solvent treatment, PSI was integrated into a solution processed organic semiconductor device. The incorporation of PSI in such devices is motivated by revealing the photo-induced dipole behaviour of PSI and to determine the orientation of PSI on the surface. To address these questions, PSI was introduced
Photo-physical properties of photosystem I inside solid-state solar cells
42
into polymer-fullerene bulk heterojunctions (BHJs). A 1:4 mixture (by wt.) of the conjugated polymer MEH-PPV and the fullerene derivative PCBM dissolved in chlorobenzene were spin-coated on top of the dense monolayer of PSI on ITO.
a)
b)
Figure 3.4. Energy diagrams of bulk heterojunctions containing a PSI electrode-modification layer. PSI is represented as an oriented dipole with corresponding shifts in VOC. (a) The conventional solar-cell structure with LiF/Al as a top electrode. (b) Inverted device structure with MoO3/Al as a top electrode.
In the final step of device fabrication, either electron-extracting (LiF/Al) or hole-extracting (MoO3/Al) top electrodes were deposited atop the MEH-PPV layer by thermal evaporation with the corresponding energetic diagrams described in Figure 3.4. As a result, two types of cells were obtained, characterized by the flow of photogenerated current in opposite directions. In such devices, the current is mainly generated in the 120 nm thick organic BHJ, which absorbs substantially more light than the PSI monolayer. In this way, the PSI layer acts as modifier of the electrode allowing to extract photoelectric properties of the multiprotein complex. It is well known that the open-circuit voltage VOC of an organic BHJ solar cell is proportional to the difference in work function of the electrodes with and without PSI film.27 In the PSI monolayer under illumination, electrons are directed to the acceptor side (FB-Fe4S4 cluster) while positive charges (holes) remain at the P700 donor. This creates a dipole between the ITO and the organic BHJ, effectively modifying the work function of the ITO.28 Since the open-circuit voltage depends on the work function difference between the ITO and the top electrode, the dipole orientation can be extracted.
As shown in Figure 3.5, the open-circuit voltage is lowered upon insertion of a PSI layer in the device with the low work function LiF/Al top electrode, whereas it increases in the cell with the high work function MoO3/Al top electrode. This behaviour implies that the work function of ITO is decreasing by incorporation of the protein monolayer, indicating that the majority of the dipoles in the PSI layer are directed with the iron-sulphur cluster FB towards the ITO (compare Figures 3.1a and 3.4). In such an orientation the photo-generated electrons of PSI are predominantly expelled towards the metal oxide. The obtained solar cell parameters are summarized in Table
Chapter 3
43
3.1. The data show that the incorporation of PSI has a significant effect on the open-circuit voltage while the other parameters are less affected.
Figure 3.5. J-V curves of bulk-heterojunction solar cells under simulated sunlight with and without PSI. (a) Conventional device with ITO anode and LiF/Al cathode and (b) inverted device with ITO cathode and MoO3/Al anode. Arrows indicate the shift in VOC due to the PSI modification layer, i.e., work-function decrease of the bottom electrode. A neutral density filter with a transmission of 0.089 was used, resulting in an incident light intensity of about 89 W/m2.
Table 3.1. Device characteristics of bulk-heterojunction solar cells with sandwich-type structures of ITO/MEH-PPV:PCBM and ITO/PSI/MEH-PPV:PCBM containing different top electrodes.
Bottom electrode
Top Electrode
VOC (V) JSC, (A/m2) FF (-)
VOC (V)
ITO LiF/Al 0.39±0.02 3.09±0.01 0.41±0.08 -0.05
ITO/PSI LiF/Al 0.34±0.05 2.81±0.17 0.41±0.02
ITO MoO3/Al 0.23±0.03 2.95±0.12 0.47±0.02 +0.11
ITO/PSI MoO3/Al 0.34±0.02 3.05±0.12 0.40±0.04
3.5 PSI dipole calculation
The shift in the work function due to PSI incorporation can be used to calculate surface charges using a simple capacitor model. As indicated before, the change in the VOC originates from a potential drop due to the dipole created by the charge separation in the PSI monolayer. Effectively, one can treat the PSI monolayer as a parallel-plate capacitor with capacitance C = r 0A/L, where 0 is the vacuum permittivity, r the relative permittivity, A the surface area,
Photo-physical properties of photosystem I inside solid-state solar cells
44
and L the layer thickness. The total surface charge can now be calculated as Q = C VOC, resulting in a surface charge density, Q/eA, of 1.84×1015 m-2, with e the elementary charge, r = VOC = 0.05 V, and a thickness of a monolayer, L = 6 nm. The value for the surface-charge density compares very well with the density of PSI on the ITO substrate determined by AFM (1.7×1015 m-2), indicating that indeed PSI is mainly oriented with the FB-Fe4S4 cluster pointing towards the metal oxide surface. In the next step, the dipole moment of PSI was calculated. Therefore, we used equation 1: = , (1)
where N is the number of absorbed molecules per surface area, μ is the dipole moment created with PSI, is the angle of dipole tilting on ITO, and 0 are the dielectric constants of the monolayer and dielectric permittivity of vacuum respectively. Formula 1 can be rewritten in the following way expressing the PSI dipole moment μPSI as: = (2) (suppose ~ ). (2)
Thereby, it was assumed that OC linearly depends on . As a result we obtained a value for μPSI of 295 D as calculated for OC of the conventional device configuration. The dipole of PSI was calculated by theoretical model and estimated to be up to 1000 D29. The difference to our calue might be due to the fact that not all PSI trimers were oriented in the same direction. The effect of dipoles on VOC inside organic solar cells has been exhaustively studied before.28, 30 These measurements demonstrate the compatibility of functional multiprotein complexes with organic electronic materials and existing processing strategies in this field like spin coating. Previously, dipole properties of proteins were measured by electric dichroism.31 Here, we present an alternative strategy of assessing the dipole characteristics of PSI that qualities organic electronic devices as a new biophysical tool.
3.6 Conclusions
We successfully incorporated large photosynthetic complex trimers with an overall molecular weight of more than 1 000 kDa into bulk heterojunction solar cells that were exclusively prepared by solution processing. It was demonstrated that the biological component and organic semiconducting materials can be integrated without compromising their original optoelectronic properties. Spin coating a blend of donor and acceptor materials dissolved in an organic solvent did not affect the structural integrity of the photoactive protein complex. The type of device described herein represents a new biophysical tool allowing to study the dipole properties by measuring open circuit voltage shifts in conventional and inverted device configuration geometries. From the dipole characteristics, the orientation of PSI trimers on the surface was calculated, which is important for the fabrication of bioelectronic devices wherein PSI is the only photoactive component. Moreover, the devices described here might be employed in the future to
Chapter 3
45
determine the dipole characteristics of other proteins as long as those are compatible with device fabrication conditions.
3.7 Methods
Cell growth: The thermophilic cyanobacterium T. Elongatus BP1 (generous gift from M. Rögner, Ruhruniversität Bochum, Germany) was grown under agitation (150 rpm) in BG11 medium.32 The temperature was kept at 56°C, continuous light applied at 50-60 μEinstein·m-2·s-1 and cell growth pursued until late log phase. At last, the cells were harvested by centrifugation [JLA 9.100 rotor, Beckman; 7500 g; 15 min], re-suspended in Buffer A [20 mM HEPES (pH 7.5); 10 mM MgCl2; 10 mM CaCl2; 500 mM Mannitol], snap-frozen in liquid nitrogen and stored at - 80°C.
Thylakoid membrane preparation: Thylakoid membranes were prepared according to the following protocol, which represents a combination of two previously described preparation methods.33, 34 To this end, fresh or frozen cells were re-suspended in Buffer A and homogenized five times using a Dounce homogenizer. After the addition of lysozyme [final concentration: 0,5% (wt/vol)] and a tip of a spatula of DNase, the cell suspension was incubated under slow agitation for 45 min at 37°C in the dark. Subsequently, the cells were lysed by two passages through a French Press (15000psi; Constant Systems Limited, UK). Membranes were collected by centrifugation [JLA 16.250 rotor, Beckman; 38000 g; 20 min] and washed with Buffer A containing 3 M NaBr. Afterwards, the membrane suspension was washed once with Buffer A and three times with a buffer containing 0,05% DDM (n-Dodecyl- -D-maltoside) in order to remove the phycobilisomes. Finally, the thylakoid membranes were solubilized by incubation in Buffer A and supplemented with 0,6% DDM for 30 min at 20°C in the dark. Non-solubilized material was pelleted by centrifugation [JLA 16.250 rotor, Beckman; 16000rpm (38000 g); 20 min] and the supernatant was subjected to subsequent purification steps.
Photosystem I purification: For PSI purification, fast liquid protein chromatography was applied on solubilized thylakoid membranes. The chromatographic purification was performed on a ÄKTA explorer [GE Healthcare] using an anion exchange column [HiTrapTM Q HP, GE Healthcare]. After column equilibration with Buffer A + 0,03% DDM, the sample was applied and subsequently eluted by a linear gradient of [0-1M] MgSO4. The green fluorescent fractions were collected and desalted with Buffer A + 0,03% DDM using Vivaspin 20 columns [molecular weight cut-off: 100 kDa; GE Healthcare]. Finally, the purified PSI sample was adjusted to a Chl a concentration of 800 μM [with Buffer A + 0,03% DDM], snap-frozen in liquid nitrogen and stored at -80°C.
Determination of chlorophyll a and protein concentration: Chl a determination was performed in 100% methanol as described in Porra et al.35
PSI immobilization on ITO surfaces and absorption spectrum measurements: ITO surfaces were immersed overnight (14 hours) in 1 mM solution of dihydroxyacetone phosphate hemimagnesium
Photo-physical properties of photosystem I inside solid-state solar cells
46
salt hydrate in deionized (DI) water. After washing the surfaces with DI water and drying with a flow of nitrogen, the substrates were ready for formation of PSI monolayers. Therefore, PSI stock solution was diluted twice with Buffer A and the substrates were incubated for 2 hours with this solution in the dark. After the immobilization, PSI substrates were rinsed with DI water and dried with a nitrogen flow. ITO substrates equipped with a PSI layer were used for device preparation. Absorption spectra of PSI monolayers were recorded with a Jasco V-630 spectrophotometer at 25°C under dry conditions. PSI ITO substrates were used for atomic force microscope (AFM) investigations and to assess the protein coverage.
AFM measuremenrts: Images were recorded in tapping mode by a Multimode 8 instrument with ScanAsyst, Controller V [Bruker]. The TESP silicon probe, 42 N/m spring constant, 320 kHz resonance frequency with tip radius of less than 10 nm was used for all measurements. Analysis of recorded height images was performed with NanoScopeAnalysis 1.2 software.
Fabrication of photovoltaic device: Glass substrates, pre-patterned with indium tin oxide, were thoroughly cleaned by washing with detergent solution, ultrasonication in acetone and isopropyl alcohol, followed by UV-ozone treatment. PSI protein were self-assemble on the surface. Subsequently, a 90 nm PTAA layer was spun from a chlorobenzene solution in a nitrogen-filled glovebox. The devices were finished by thermal evaporation of a MoO3(10 nm)/Al(100 nm) top electrode at a base pressure of 1×10-6 mbar. For the devices with a MEH-PPV:PCBM layer, the TiOx layer was omitted. The 120 nm MEH-PPV:PCBM (1:4 by wt.) layer was spin cast from a chlorobenzene solution. Top electrodes, MoO3(10 nm)/Al(100 nm) or LiF(1 nm)/Al(100 nm) were thermally evaporated. Device characterization: Electrical measurements were conducted in a controlled nitrogen atmosphere in the dark and under illumination of a Steuernagel Solar Constant 1200 metal halide lamp, which was set to 1 Sun intensity using a silicon reference cell and correcting for spectral mismatch. EQE spectra were recorded versus a silicon reference, using a custom-built setup comprising a lock-monochromatic beam from a quartz tungsten halogen lamp, and a range of narrow band pass
Chapter 3
47
3.8 References 1. Gratzel, M. Photoelectrochemical cells. Nature 414, 338-344 (2001). 2. Chung, I., Lee, B., He, J.Q., Chang, R.P.H. & Kanatzidis, M.G. All-solid-state dye-sensitized solar cells with high efficiency. Nature 485, 486-U494 (2012). 3. Halls, J.J.M. et al. Efficient Photodiodes from Interpenetrating Polymer Networks. Nature 376, 498-500 (1995). 4. Brabec, C.J., Sariciftci, N.S. & Hummelen, J.C. Plastic solar cells. Adv. Funct. Mater. 11, 15-26 (2001). 5. Yu, G., Gao, J., Hummelen, J.C., Wudl, F. & Heeger, A.J. Polymer Photovoltaic Cells - Enhanced Efficiencies Via a Network of Internal Donor-Acceptor Heterojunctions. Science 270, 1789-1791 (1995). 6. Ham, M.H. et al. Photoelectrochemical complexes for solar energy conversion that chemically and autonomously regenerate. Nat. Chem. 2, 929-936 (2010). 7. Huang, J.S. et al. Polymer bulk heterojunction solar cells employing Forster resonance energy transfer. Nat. Photonics. 7, 480-486 (2013). 8. Weil, T., Reuther, E. & Mullen, K. Shape-persistent, fluorescent polyphenylene dyads and a triad for efficient vectorial transduction of excitation energy. Angewandte Chemie 41, 1900-1904 (2002). 9. Hayes, D., Griffin, G.B. & Engel, G.S. Engineering coherence among excited states in synthetic heterodimer systems. Science 340, 1431-1434 (2013). 10. Yehezkeli, O. et al. Integrated photosystem II-based photo-bioelectrochemical cells. Nat. Commun. 3 (2012). 11. Krassen, H. et al. Photosynthetic Hydrogen Production by a Hybrid Complex of Photosystem I and [NiFe]-Hydrogenase. ACS Nano 3, 4055-4061 (2009). 12. Iwuchukwu, I.J. et al. Self-organized photosynthetic nanoparticle for cell-free hydrogen production. Nat. Nanotechnol. 5, 73-79 (2010). 13. Brettel, K. Electron transfer and arrangement of the redox cofactors in photosystem I. BBA-Bioenergetics 1318, 322-373 (1997). 14. Blankenship, R.E. et al. Comparing Photosynthetic and Photovoltaic Efficiencies and Recognizing the Potential for Improvement. Science 332, 805-809 (2011). 15. Das, R. et al. Integration of photosynthetic protein molecular complexes in solid-state electronic devices. Nano Lett.4, 1079-1083 (2004).
Photo-physical properties of photosystem I inside solid-state solar cells
48
16. Gerster, D. et al. Photocurrent of a single photosynthetic protein. Nat.Nanotechnol. 7, 673-676 (2012). 17. Lee, I., Lee, J.W. & Greenbaum, E. Biomolecular electronics: Vectorial arrays of photosynthetic reaction centers. Phys. Rev. Lett. 79, 3294-3297 (1997). 18. Trammell, S.A., Spano, A., Price, R. & Lebedev, N. Effect of protein orientation on electron transfer between photosynthetic reaction centers and carbon electrodes. Biosens. Bioelectron. 21, 1023-1028 (2006). 19. Trammell, S.A., Wang, L.Y., Zullo, J.M., Shashidhar, R. & Lebedev, N. Orientated binding of photosynthetic reaction centers on gold using Ni-NTA self-assembled monolayers. Biosens. Bioelectron. 19, 1649-1655 (2004). 20. Carmeli, I., Frolov, L., Carmeli, C. & Richter, S. Photovoltaic activity of photosystem I-based self-assembled monolayer. J. Am. Chem. Soc. 129, 12352 (2007). 21. Kiley, P. et al. Self-assembling peptide detergents stabilize isolated photosystem I on a dry surface for an extended time. Plos Biol. 3, 1180-1186 (2005). 22. Mershin, A. et al. Self-assembled photosystem-I biophotovoltaics on nanostructured TiO2 and ZnO. Sci. Rep. 2 (2012). 23. Boekema, E.J. et al. Evidence for a Trimeric Organization of the Photosystem-I Complex from the Thermophilic Cyanobacterium Synechococcus Sp. FEBS Lett. 217, 283-286 (1987). 24. Jordan, P. et al. Three-dimensional structure of cyanobacterial photosystem I at 2.5 angstrom resolution. Nature 411, 909-917 (2001). 25. Hofer, R., Textor, M. & Spencer, N.D. Alkyl phosphate monolayers, self-assembled from aqueous solution onto metal oxide surfaces. Langmuir 17, 4014-4020 (2001). 26. Hegner, M., Wagner, P. & Semenza, G. Ultralarge Atomically Flat Template-Stripped Au Surfaces for Scanning Probe Microscopy. Surf. Sci. 291, 39-46 (1993). 27. Mihailetchi, V.D., Blom, P.W.M., Hummelen, J.C. & Rispens, M.T. Cathode dependence of the open-circuit voltage of polymer : fullerene bulk heterojunction solar cells. J. Appl. Phys. 94, 6849-6854 (2003). 28. de Boer, B., Hadipour, A., Mandoc, M.M., van Woudenbergh, T. & Blom, P.W.M. Tuning of metal work functions with self-assembled monolayers. Adv. Mater. 17, 621 (2005). 29. Felder, C.E., Prilusky, J., Silman, I. & Sussman, J.L. A server and database for dipole moments of proteins. Nucleic Acids Res. 35, W512-W521 (2007). 30. Tseng, C.T., Cheng, Y.H. & Lee, M.C.M. Study of anode work function modified by self-assembled monolayers on pentacene/fullerene organic solar cells. Appl. Phys. Lett. 91 (2007).
Chapter 3
49
31. Antosiewicz, J. & Porschke, D. Electrostatics of Hemoglobins from Measurements of the Electric Dichroism and Computer-Simulations. Biophys. J. 68, 655-664 (1995). 32. Rippka, R., Deruelles, J., Waterbury, J.B., Herdman, M. & Stanier, R.Y. Generic Assignments, Strain Histories and Properties of Pure Cultures of Cyanobacteria. J. Gen. Microbiol. 111, 1-61 (1979). 33. Mukherjee, D., May, M., Vaughn, M., Bruce, B.D. & Khomami, B. Controlling the Morphology of Photosystem I Assembly on Thiol-Activated Au Substrates. Langmuir 26, 16048-16054 (2010). 34. El-Mohsnawy, E. et al. Structure and Function of Intact Photosystem 1 Monomers from the Cyanobacterium Thermosynechococcus elongatus. Biochemistry 49, 4740-4751 (2010). 35. Porra, R.J., Thompson, W.A. & Kriedemann, P.E. Determination of Accurate Extinction Coefficients and Simultaneous-Equations for Assaying Chlorophyll-a and Chlorophyll-B Extracted with 4 Different Solvents - Verification of the Concentration of Chlorophyll Standards by Atomic-Absorption Spectroscopy. Biochim. Biophys. Acta. 975, 384-394 (1989).
Chapter 3
50
Chapter 4
51
Solid-state bio-photovoltaic cells based on single photosystem I active layer
In the previous chapter, PSI was incorporated into a BHJ solar cell to determine the dipole of PSI and to measure the average orientation of the protein complex. Here, PSI was sandwiched between titanium oxide (TiOx) and the optically-transparent hole-conducting polymer polytriarylamine (PTAA). In this solar cell device architecture, PSI acts as the sole active layer, resulting in a high open-circuit voltage of 0.76 V, a fill factor of 45% and a short-circuit current density of 2.9 mA/m2. Our experiments demonstrate the functionality of PSI in a solid-state bio-photovoltaic device and that photoactive megadalton-protein complexes are capable to operate in a dry non-natural environment that is very different from the biological membrane.
Gordiichuk, P. I., Wetzelaer, G.-J. A. H., Rimmerman, D., Gruszka, A., de Vries, J. W., Saller, M., Gautier, D. A., Catarci, S., Pesce, D., Richter, S., Blom, P. W. M. and Herrmann, A. Advanced Materials, 26, 4863–4869 (2014).
4
Solid-state bio-photovoltaic cells based on single photosystem I active layer
52
4.1 Introduction
Photosynthesis can be a great inspiration for chemists and physicists to fabricate artificial light-energy conversion systems. To mimic light harvesting, as found in light harvesting complex II (LHC II), which were introduced in the first chapter, dendrimers were synthesized that contain donor and acceptor dyes distributed over the dendritic scaffold1, 2. Besides dendritic scaffolds, other structures were employed to host dye molecules in a light harvesting configuration like DNA3. While light harvesting has been reported with several scaffolding strategies, the oxygen evolving complex (OEC) that is found in Photosystem II is much harder to mimic with synthetic systems. An artificial OEC was formed that is composed of cobalt and phosphate4. From this material films were produced on surface of irradiated semiconducting materials like ZnO5. In such design, light absorption is combined with water splitting to achieve solar oxygen production.
Here, we aim for a different strategy of biomimetics. We take photoactive natural building blocks and let them perform their function in an artificial device. In this chapter, we combine the easy processing conditions of organic semiconductors and the formation of monolayers with a photoactive protein complex. As a result, a solid-state bio-organic solar cell was obtained with PSI as the only photoactive component. Another key component of this device was a polytriarylamine polymer that is hole conducting and transparent in the visible where light absorption of PSI occurs. Moreover, special attention is paid to the compatibility of PSI with organic solvents during device fabrication and an alternative method is presented to avoid contact of the biological component with organic solvents.
4.2 Integration of PSI inside solid-state solar cells
Similar as for the preparation of a BHJ, a PSI layer needs to be installed on the electrode. The directing linker molecule dihydroxyacetone phosphate was immobilized on titanium suboxide (TiOx, x=1, 2) supported on glass substrates.6 The self-assembled monolayers on the metal oxide surfaces were immersed in a PSI solution as described in the previous Chapter 3 for ITO. Afterwards, they were studied with the help of AFM to determine the surface coverage. On TiOx an average number of 1.7×1015 m-2 PSI monomers corresponding to a surface coverage of 50% was found allowed recording of an absorption spectrum of a PSI monolayer assembled on the transparent substrates under dry conditions. Characteristic absorption peaks of PSI were detected at 430 and 660 nm and the shape of the spectrum resembles the absorption behaviour of PSI in solution (Figure 4.1a). Such absorption measurements indicate that the dihydroxyacetone phosphate linker is well suited for incorporation of PSI by binding to the TiOX surfaces, which is interacting with the PSI by electrostatic and hydrogen bonds.
Chapter 4
53
a)
400 450 500 550 600 650 700
0.0
0.2
0.4
0.6
0.8
1.0
PSI monolayer on TiOx PSI in solution
Wavelength (nm)
Abso
rptio
n (n
orm
alize
d)
b)
-10-9-8-7-6-5-4-3-2-10
-4.3
-9.7
-6.7
-5.3
-2.3
-4.6
-3.5
-7.5
-4.2Al
MoO3
PTAAPSI
TiOx
Ener
gy v
ersu
s va
cuum
(eV)
ITO-4.8
Figure 4.1. (a) Absorbance of PSI monolayer on TiOX surface (solid green line) and compared with PSI absorbance in solution (black dash line). (b) Energy levels of a solar cell of inverted geometry with TiOx cathode and MoO3 anode containing a PSI monolayer. The redox energy levels of PSI were taken from the literature.7
With knowing the dominant orientation of PSI on the metal-oxide substrate (Chapter 3), the fabrication of solid-state solar cells with PSI acting as the active layer for photo-current generation was realized. For these devices, two major modifications were introduced in relation to the cells described in Chapter 3. First, TiOx was placed on the ITO surface because of its electron-selective properties and its transparency. Second, the blend consisting of MEH-PPV and PCBM covering the PSI layer was replaced by a semiconducting polytriarylamine polymer (PTAA). This polymer was chosen because it is transparent to light in the visible region and therefore does not interfere with the major long-wavelength absorption of PSI.8 PTAA was introduced as a layer on top of PSI by spin coating it from chlorobenzene solution. The energy diagram of fabricated device is described in Figure 4.1b.
The overall device structure is schematically depicted in Figure 4.2a and the device function is anticipated as follows: PSI absorbs light and generates electrons that are transported via the TiOx layer to the ITO bottom electrode. At the same time, the generated holes are transported through the p-conducting PTAA layer to the MoO3/Al top electrode. Figure 4.2b shows the current density-voltage characteristics of the biophotovoltaic device under illumination of simulated AM 1.5 sunlight, with an intensity of 1 Sun. To exclude absorption and subsequent photo-generation of charge carriers by the titanium oxide and PTAA, a 435 nm long pass filter was employed. A clear photovoltaic response was obtained under standard solar-cell testing conditions, which can be attributed to photocurrent generation by PSI. The activity of PSI in this solid-state device is undoubtedly confirmed by photocurrent action spectra. A peak between 600 and 700 nm was observed, which coincides with the absorption of PSI, Figure 4.3a. A reference device without the PSI monolayer gave no response in this wavelength region. For further characterization of the PSI bio-photovoltaic cell, the J-V curves were analysed (Figure 4.2b-c), which was not done in previous studies.9-15 The device exhibited an impressive open-circuit voltage of 0.76 V, retaining
Solid-state bio-photovoltaic cells based on single photosystem I active layer
54
a large part of the intrinsic energy-level potential of PSI. Furthermore, a fill factor of 45% was measured. The short-circuit current is low, which can be mainly attributed to the weak light absorption of the thin PSI monolayer. In addition, the energy levels of the hole- and electron-transport layers (PTAA and TiOx) are not fully aligned with the energy levels of PSI, which may limit charge extraction. To initially test the stability of such a device, we performed three measurements within one week. It can be clearly seen that the PSI is stable for this period of time in the dry environment (Figure 4.3b). No decrease in short-circuit current density was observed.
a)
b)
0.0 0.2 0.4 0.6 0.8 1.0 1.2
-4
-2
0
2
4
6
Curre
nt D
ensi
ty (m
A/m
2 )
Voltage (V)
Voc= 0.76 VJsc= 2.9 mA/m2
FF= 45%
c)
-0.9 -0.6 -0.3 0.0 0.3 0.6 0.9 1.21E-3
0.01
0.1
1
10
Curr
ent D
ensi
ty (m
A/m
2 )
Voltage (V)
Figure 4.2. (a) Current density-voltage plots of solar cells with PSI as active layer sandwiched between TiOx and PTAA, in dark (black) and under simulated sunlight with an intensity of 1000 W/m2, 1 Sun (red). (b) The result on a semi logarithmic scale. A 435 nm long pass filter was employed to exclude photo-generation due to PTAA and TiOx, which absorb light in the UV-region.
Chapter 4
55
a)
450 500 550 600 650 700 750 8000.000
0.002
0.004
0.006
0.008
PSI no PSI
Wavelength (nm)
EQE
(%)
b)
0 1 2 3 4 5 6 7 80
1
2
3
4
5
Shor
t-circ
uit c
urre
nt d
ensi
ty (m
A/m
2 )
Time after fabrication (days)
Figure 4.3. (a) External quantum efficiency (EQE) of a device with and without PSI as active layer between TiOx and PTAA. (b) Time-depended stability study of PSI fabricated solid-state cells.
a)
450 500 550 600 650 700 750 8000.000
0.002
0.004
0.006
0.008
No CB wash 1x CB 3x CB 5x CB
Wavelength (nm)
EQE
(%)
b)
400 420 440 460 480 5000.0
0.2
0.4
0.6
0.8
1.0
No CB wash 1x CB 3x CB 5x CB
Wavelength (nm)
EQE
(%)
Figure 4.4. Effect of chlorobenzene (CB) treatment on the external quantum efficiency of PSI photovoltaic devices with a TiOx electron-transporting layer and a PTAA hole-transporting layer. (a) Response of PSI (in the 600-700 nm region) decreases with increasing number of CB washing steps. (b) The response in the short-wavelength regionincreases with increasing number of CB washing steps. The response in this region is due to charge generation at the TiOx/PTAA interface. This indicates that a certain amount of PSI complexes is removed upon each CB washing step, resulting in a larger contact area between TiOX and PTAA.
Again, special attention was paid to the effects of the organic solvent processing on the PSI layer. For the preparation of control devices, the PSI monolayer deposited on the TiOx was repeatedly washed with chlorobenzene before spin-coating PTAA. An increasing number of washing steps that accompanies a gradual decrease in the external quantum efficiency (EQE) indicates that chlorophyll a is not liberated from PSI, pigment absorption does not occur on TiOx, and that
Solid-state bio-photovoltaic cells based on single photosystem I active layer
56
charge generation originates exclusively from the intact photosynthetic protein complex (Figure 4.4a). Instead, the decrease in EQE can be attributed to a gradual removal of PSI complexes from the TiOx surface with repeating chlorobenzene washing steps and shows increase in the current caused by formation of direct contact with TiOX surface (Figure 4.4b). On a gold surface, loss of PSI molecules upon washing with chlorobenzene was also observed in Chapter 3.
4.3 Lamination of PTAA on PSI monolayer
In the above paragraph, special attention was paid to demonstrate that the processing conditions of organic electronic devices, especially spin coating of a top layer onto PSI layer employing organic solvent does not harm the protein component. In this paragraph, we add another control experiment that also represents a bio-electronic device fabrication procedure that fully avoids contact between organic solvent and the biological moieties, Therefore, the hole conducting polymer PTAA was spin coated onto a glass surface, which was subsequently lift-off by slow dipping of the substrate with polymer inside deionized water. The release of the PTAA layer was due to water penetration between polymer and glass surface. Next, the floated polymer film was transferred on top of an ITO/TiOx/PSI substrate and dried. This fabrication technique ensures that PSI is not exposed to any organic solvent. The devices are finished by evaporation of a MoO3/Al top contact. The measured J-V characteristics, as presented in the Figure 4.5, are very similar to the characteristics of devices prepared by directly spin coating the PTAA layer on top of the PSI monolayer employing chlorobenzene as solvent. Therefore, we conclude that the photovoltaic response of our devices is due to the intact proteins and not due to leached chlorophylls. An electrical current of 2.1 mA/m2 and a VOC of 0.7 V were measured.
-0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0
-3
-2
-1
0
1
2
Curre
nt D
ensi
ty (m
A/m
2 )
Voltage (V)
Figure 4.5. Current density-voltage characteristics in dark (black line) and under illumination (red line) of a PSI solar cell with a laminated PTAA hole-transport layer.
The measured performance of a single layer of PSI inside solid-state solar cells is lower than the possible theoretical limit16, which can be due to misalignment of energy levels of P700 reaction
Chapter 4
57
center and the used PTAA hole conducting polymer, which can cause efficient hole extraction from PSI. Also limitations can be due to not perfect orientation of PSI proteins and bad contacts between the high molecular weight polymer and PSI protein that differ significantly in their polarities.
4.4 Conclusions
In this chapter, it was demonstrated that a biological component consisting of the photoactive protein complex PSI can be combined with an organic semiconducting material, the hole conductor polytriarylamine to form an all-solid state biophotovoltaic cell. Thereby, both compounds maintain their original optoelectronic properties. Two ways of fabricating such devices are presented. In the first one, a PSI monolayer installed on an electrode surface is spin coated with the semiconductor polymer from the organic solvent. In the second one, contact between the organic solvent and the biological moiety is completely avoided. This is achieved by producing a free standing semiconducting polymer layer and laminating it on the PSI monolayer. The resulting devices were characterized by a large open-circuit voltage and a photocurrent action spectrum exhibiting the typical absorption features of PSI. These successful proof-of-concept studies will fuel further activities of exploiting new material combinations of bioorganic hybrids for solar power generation.
4.5 Methods
PSI immobilization on TiOx surfaces and absorption spectrum measurements: TiOx surface were immersed overnight (14 hours) in 1 mM solution of dihydroxyacetone phosphate hemimagnesium salt hydrate in deionized (DI) water. After washing the surfaces with DI water and drying with a flow of nitrogen, the substrates were ready for formation of PSI monolayers. Therefore, PSI stock solution was diluted twice with Buffer A and the substrates were incubated for 2 hours with this solution in the dark. After the immobilization, PSI substrates were rinsed with DI water and dried with a nitrogen flow. ITO and TiOx substrates equipped with a PSI layer were used for device preparation. Absorption spectra of PSI monolayers were recorded with a Jasco V-630 spectrophotometer at 25°C under dry conditions. PSI ITO substrates were used for atomic force microscope (AFM) investigations and to assess the protein coverage.
Preparation of TiOx layers: TiOx layers were obtained by spin coating of a 1 wt.% solution of titanium isopropoxide in isopropyl alcohol. The layers were converted to TiOx by hydrolysis in air for 1 hour and were subsequently annealed at 150°C for 10 min to obtain layers of ~5 nm.
Fabrication of photovoltaic device: Glass substrates, pre-patterned with indium tin oxide, were thoroughly cleaned by washing with detergent solution, ultrasonication in acetone and isopropyl
Solid-state bio-photovoltaic cells based on single photosystem I active layer
58
alcohol, followed by UV-ozone treatment. For the PTAA-based cells, the cleaned substrates were coated with a TiOx layer, followed by self-assembly of a PSI monolayer. Subsequently, a 90 nm PTAA layer was spun from a chlorobenzene solution in a nitrogen-filled glovebox. The devices were finished by thermal evaporation of a MoO3(10 nm)/Al(100 nm) top electrode at a base pressure of 1×10-6 mbar.
Device characterization: Electrical measurements were conducted in a controlled nitrogen atmosphere in the dark and under illumination of a Steuernagel Solar Constant 1200 metal halide lamp, which was set to 1 Sun intensity using a silicon reference cell and correcting for spectral mismatch. EQE spectra were recorded versus a silicon reference, using a custom-built setup comprising a lock- used, chopped monochromatic beam from a quartz tungsten halogen lamp, and a range of narrow band pass
Chapter 4
59
4.6 References
1. Maus, M. et al. Intramolecular energy hopping and energy trapping in polyphenylene dendrimers with multiple peryleneimide donor chromophores and a terryleneimide acceptor trap chromophore. J. Am. Chem. Soc. 123, 7668-7676 (2001). 2. Dichtel, W.R., Hecht, S. & Frechet, J.M.J. Functionally layered dendrimers: A new building block and its application to the synthesis of multichromophoric light-harvesting systems. Org. Lett. 7, 4451-4454 (2005). 3. Hannestad, J.K., Sandin, P. & Albinsson, B. Self-Assembled DNA Photonic Wire for Long-Range Energy Transfer. J. Am. Chem. Soc. 130, 15889-15895 (2008). 4. Kanan, M.W. & Nocera, D.G. In situ formation of an oxygen-evolving catalyst in neutral water containing phosphate and Co2+. Science 321, 1072-1075 (2008). 5. Steinmiller, E.M.P. & Choi, K.S. Photochemical deposition of cobalt-based oxygen evolving catalyst on a semiconductor photoanode for solar oxygen production. P. Natl. Acad. Sci. USA 106, 20633-20636 (2009). 6. Hofer, R., Textor, M. & Spencer, N.D. Alkyl phosphate monolayers, self-assembled from aqueous solution onto metal oxide surfaces. Langmuir 17, 4014-4020 (2001). 7. Nakamura, A., Suzawa, T., Kato, Y. & Watanabe, T. Species Dependence of the Redox Potential of the Primary Electron Donor P700 in Photosystem I of Oxygenic Photosynthetic Organisms Revealed by Spectroelectrochemistry. Plant Cell Physiol. 52, 815-823 (2011). 8. Veres, J., Ogier, S., Lloyd, G. & de Leeuw, D. Gate insulators in organic field-effect transistors. Chem. Mater. 16, 4543-4555 (2004). 9. Brabec, C.J., Sariciftci, N.S. & Hummelen, J.C. Plastic solar cells. Adv. Funct. Mater. 11, 15-26 (2001). 10. Halls, J.J.M. et al. Efficient Photodiodes from Interpenetrating Polymer Networks. Nature 376, 498-500 (1995). 11. Yu, G., Gao, J., Hummelen, J.C., Wudl, F. & Heeger, A.J. Polymer Photovoltaic Cells - Enhanced Efficiencies Via a Network of Internal Donor-Acceptor Heterojunctions. Science 270, 1789-1791 (1995). 12. Das, R. et al. Integration of photosynthetic protein molecular complexes in solid-state electronic devices. Nano Lett. 4, 1079-1083 (2004). 13. Carmeli, I., Frolov, L., Carmeli, C. & Richter, S. Photovoltaic activity of photosystem I-based self-assembled monolayer. J. Am. Chem. Soc. 129, 12352 (2007).
Solid-state bio-photovoltaic cells based on single photosystem I active layer
60
14. Trammell, S.A., Spano, A., Price, R. & Lebedev, N. Effect of protein orientation on electron transfer between photosynthetic reaction centers and carbon electrodes. Biosens. Bioelectron. 21, 1023-1028 (2006). 15. Trammell, S.A., Wang, L.Y., Zullo, J.M., Shashidhar, R. & Lebedev, N. Orientated binding of photosynthetic reaction centers on gold using Ni-NTA self-assembled monolayers. Biosens. Bioelectron. 19, 1649-1655 (2004). 16. Gerster, D. et al. Photocurrent of a single photosynthetic protein. Nat. Nanotechnol. 7, 673-676 (2012).
Chapter 5
61
Increasing photosystem I activity by covalently attached organic dyes
Photosynthesis is an efficient mechanism of solar energy conversion into chemical energy. One of the key players in this process is the protein photosystem I (PSI) that converts incident photons into separated charges, which makes this protein attractive for applications in bio-inspired photoactive hybrid materials. However, the efficiency of PSI is still limited by its poor absorption in the green part of the solar spectrum. Inspired by the existence of natural phycobilisome light-harvesting antennae, we have widened the absorption spectrum of PSI by covalent attachment of synthetic dyes to the protein backbone. Steady-state and time-resolved photoluminescence reveal that energy transfer occurs from these dyes to PSI. It is shown by oxygen-consumption measurements that subsequent charge generation is substantially enhanced under broad and narrow band excitation. Ultimately, surface photovoltage experiments prove the enhanced activity of dye-modified PSI even in the solid state.
5
P. Gordiichuk, D. Rimmerman, A. Paul, D. A. Gautier, M. Saller, J. W. de Vries, G.-J. A. H. Wetzelaer, M. Manca, W. Gomulya, M. Loznik, P. W. M. Blom, M. Rögner, M. A. Loi, S. Richter, A. Herrmann, Bioconjugate Chemistry, accepted (2015).
Increasing photosystem I activity by covalently attached organic dyes
62
5.1 Introduction During photosynthesis, absorbed sunlight is efficiently converted into chemical energy. The membrane multi-protein complex photosystem I (PSI) plays a key role in this process, as it converts photons into separated charges with 1 V potential difference at near 100% quantum efficiency. The two most important features of PSI are charge separation and light harvesting. Charge separation takes place in the core of the complex within the reaction center at a special pair of chlorophylls named P7001. From there, the excited electrons migrate through the chain of primary electron acceptors A0 (Chla), A1 (phylloquinone), FX, FA and FB (Fe4S4 clusters), thereby traversing almost the entire volume of PSI2. The resulting hole at P700 is refilled by the electron carrier cytochrome c6 from the lumen side resetting PSI for the next photo-excitation. For the second important function, i.e. light harvesting, a network of pigment antennae surrounding the reaction center is responsible3-5. However, an essential part of the solar energy is not exploited by PSI because these photosynthetic pigments including chlorophylls and carotenoids do not absorb light in the wavelength range of 450-600 nm, which is termed the “green gap”. Although the intensity of the sunlight reaches its maximum in this spectral region, PSI harvests the light poorly at these wavelengths. The lack of absorption in the green gap is a consequence of the chlorophyll absorption spectrum with its characteristic Soret- and Q bands with maxima at 430 nm and 665 nm, respectively. The low absorbance between these two peaks significantly reduces the energy conversion efficiency of the multi-protein complex. In nature, partial closing of this gap is realized with the help of other protein complexes, such as phycobilisomes, which are found in cyanobacteria and red algae6. The protein assemblies absorb light in the range between 500 to 650 nm with the help of phycobilin pigments that are contained in special molecular aggregates called phycoerythrin, phycocyanin and allophycocyanin. These complexes funnel absorbed light energy to PSI and PSII for photosynthetic reactions7, 8. To date, the energy transfer mechanisms between these proteins and charge separation centers at a distance of more than 20 Å is not fully understood 9, 10. Another light-harvesting system where energy transfer over a long distance occurs is the Fenna-Matthews-Olson complex in green sulfur bacteria, which live in deep lakes where only a minor fraction of the sunlight is available 7, 11. These examples demonstrate that utilizing sunlight energy outside the main absorption region of photosynthetic complexes is a viable way to increase the efficiency of photosynthetic energy conversion for life in habitats with low sun intensities. In contrast to bridging the green gap with the help of natural building blocks, this goal can be achieved with synthetic chromophores12, 13 14-16. In this chapter, we study the chemical modification of PSI with multiple fluorescent dyes that absorb in the green region. The effect of dye modification on the photophysical properties and the activity of PSI in solution and in the dry state are discussed below.
Chapter 5
63
5.2 Covalent attachment of dyes to PSI protein complex
The protein complex of PSI used in this study was extracted from the thermophilic cyanobacterium T. Elongatus 17. Under the conditions of extraction PSI exists nearly exclusively as trimeric complex, with a size of a= 281.0 Å, b= 281.0 Å, c= 165.2 Å and a mass of about 1.05 megadalton. Moreover, a large number of light harvesting cofactors are incorporated within PSI, i.e. 96 chlorophylls and 22 carotenoids per monomer. We have chosen for a PSI variant from an extremophile since such proteins are characterized by increased robustness which renders them less prone to loss of biological function upon chemical modification. Importantly, the PSI monomer employed in this study contains approximately 70 surface-exposed lysine residues that are available for attachment of dyes. The commercially available fluorescent dye ATTO 590 was chosen because of its large absorption in the green region, its high fluorescence quantum yield of 80% and the lack of overlap with the PSI main absorption peak at 665 nm. The calculated Förster resonance energy transfer radius RATTO590-Chla of 57.3 Å 18, 19 may allow efficient energy transfer from the artificial emitter to PSI chlorophylls from any attachment point on the protein scaffold.
Figure 5.1. (a) Absorption spectra of PSI (black line) and ATTO 590 (red line). The emission spectrum of the dye is shown in green (excitation 590 nm). (b) Absorption spectra of PSI modified with a different number of ATTO 590 dyes acquired after purification. Up to 35 chromophores are attached to PSI on average.
The lysine residues of PSI were modified with the N-hydroxysuccinimide ester derivative of ATTO 590. During the coupling reaction, various dye-to-protein ratios were selected in order to synthesize PSI conjugates exhibiting different numbers of organic dyes. The hybrids were purified by extensive dialysis prior to calculation of the amount of coupled dyes. The main absorption peaks (Figure 5.1a) of the ATTO 590 dye (590 nm) and PSI (665 nm) do not overlap, which allowed us to calculate the precise number of attached dyes by taking into account the absorption spectrum of the modified PSI and using the reported extinction coefficients 20. The average number of dyes attached to PSI was calculated to range on average from 5.2 to 34.8 dyes per PSI monomer for different coupling ratios (from 1x to 9x per 1 available NH2 group in PSI), as depicted in Figure 5.1b. To achieve these different degrees of modification of PSI, one up to
Increasing photosystem I activity by covalently attached organic dyes
64
nine equivalents of activated dye molecules per surface exposed lysine were chosen during the conjugation reaction. The chemical modification did not affect the trimeric structure of the PSI complex, which was revealed by transmission electron microscopy (TEM) (Figure 5.2a). Due to the fact that we identified the surface-exposed lysine residues from a reported crystal structure we expected a homogenous distribution of the dyes at the available positions on the two faces of PSI. To exclude unspecific binding, PSI was treated with the non-reactive carboxylic acid derivative of ATTO 590 following the same coupling and purification procedures. No absorption in the green gap was observed and only unmodified PSI was recovered (Figure 5.2b). This proves the presence of a covalent bond between PSI and ATTO 590 and that the increased absorption is not due to unspecific adsorption of dyes to PSI.
a)
b)
400 450 500 550 600 650 700 7500.0
0.2
0.4
0.6
0.8
1.0
1.2
Native PSI PSI+COOH Atto 590 after purification
Wavelength (nm)
Norm
aliz
ed A
bsor
ptio
n
Figure 5.2. (a) TEM of PSI covalently functionalized with on average 20.5 ATTO 590 dyes per PSI monomer. Modified PSI maintains its trimeric structure. (b) The green line represents the absorption spectrum of native PSI. The black line indicates purified PSI that was incubated with a 7 fold excess of ATTO 590 only containing a carboxylic acid group instead of an activated ester. No additional absorption peak could be detected at 590 nm indicating that no unspecific binding between PSI and the dyes occurs.
5.3 Time-resolved spectroscopy study of energy transfer from attached dyes to PSI protein complex
Next, photoluminescence (PL) measurements were performed to characterize the dye-modified PSI complex, containing 6.0 dyes per PSI monomer on average. The fluorescence of ATTO 590 when attached to PSI was quenched, which could indicate both energy transfer from the dye to PSI other dyes or direct electron transfer to the reaction center. However, when the dye was just mixed with PSI no reduction of ATTO 590 PL was observed in the steady-state measurements (Figure 5.3a). Time-resolved measurements additionally confirmed a bi-exponential PL decay (averaged 1275 ps) when ATTO 590 is connected to the protein scaffold (Figure 5.3b).
Chapter 5
65
a)
550 600 650 700 750 800
600 700 8000.0
0.5
1.0 PSI PSI-ATTO 590
Wavelength (nm)No
rmal
ized
Pho
tolu
min
esce
nce
Ph
otol
umin
esce
nce
(a.u
.)
Wavelength (nm)
ATTO 590 PSI-ATTO 590
b)
0 200 400 600 800 1000 1200 1400
Phot
olum
ines
cenc
e (a
.u.)
Time (ps)
ATTO 590 PSI-ATTO 590 = 3717 ps < >= 1276 ps
Figure 5.3. Steady-state (a) and time-resolved fluorescence (b) measurements of dye-modified PSI. (a) The fluorescence of ATTO 590 (black line) and ATTO 590 covalently attached to PSI (red line). Photo excitation was performed at 380 nm. Insertion shows normalized PL of PSI and PSI-ATTO 590 dyes at fixed PSI concentration (50
(b) Fluorescence decay of ATTO 590-PSI conjugate (red line) and of ATTO 590 dyes in absence of PSI (black g either dissolved in the buffer
or attached to PSI. The ATTO 590-PSI conjugates had an average number of 6 dyes per PSI protein.
In order to exclude an effect of the protein backbone on the fluorescence behaviour of the dye, bovine serum albumin (BSA), a protein lacking any photosynthetic activity, was modified with ATTO 590. Only a small change in steady-state fluorescence spectrum was detected and the emission decay was unaffected (see Figure 5.4). These control experiments confirm that energy transfer from dye to PSI indeed occurs within the dye-PSI conjugate.
a)
550 600 650 700 750 8000
200
400
600
800
1000
1200
ATTO 590 BSA+ATTO 590
Phot
olum
ines
cenc
e (x
1000
)
Wavelength (nm)
b)
0 500 1000 1500 20000.01
0.1
1
ATTO 590 BSA-ATTO 590
Phot
olum
ines
cenc
e (a
.u.)
Time (ps)
Figure 5.4. (a) When ATTO 590 dyes are covalently attached to BSA their fluorescence is only marginally quenched (red spectrum) in relation to pristine chromophores (black spectrum). (b) Optical density of both solutions were equal. Right: Fluorescence decay of ATTO 590 dyes attached to PSI (red curve) is very similar to the ones of the pristine dyes (black curve).
5.4 Oxygen consumption of modified PSI protein complexes To investigate if the energy transferred or electron transfer from the dye to PSI leads to an improved functionality, oxygen consumption experiments were carried out. The rate of electron
Increasing photosystem I activity by covalently attached organic dyes
66
generation under illumination, which occurs through electron transfer from the P700 reaction center to the iron-sulfur cluster via the cascade electron-transfer chain mentioned in the introduction, can be measured directly by oxygen consumption experiments based on the Mehler reaction 21. a)
-200 0 200 400 600 800 10001200140016000.0
0.2
0.4
0.6
0.8
1.0 PSI-ATTO 590 PSI native PSI under coupling conditions BSA-ATTO590
Time (sec)
Norm
aliz
ed O
xyge
n Co
ncen
tratio
n (x
100%
)
b)
0 5 10 15 20 25 30 35
100
200
300
400
500
600
700
Oxyg
en c
onsu
mpt
ion
(m
O 2*mg(
Chla
)-1h-1
)
Number of attached ATTO 590 dyes to PSI
PSI+ATTO 590 Oxygen consumtion
c)
500 600 700 800 9000
2000
4000
6000
8000
10000 Lamp full specrum
PL
Wavelength (nm)
d)
0 500 1000 1500 2000 2500 3000 35000.0
0.2
0.4
0.6
0.8
1.0
Norm
alize
d Ox
ygen
Con
cent
ratio
n (x
100%
)
Time (sec)
PSI under coupling conditions PSI and ATTO590 not coupled
Figure 5.5. Oxygen consumption experiments of PSI covalently functionalized with ATTO 590 dyes using the procedure described in reference. (a) A faster oxygen consumption was observed for PSI that was on average modified with 32.2 ATTO 590 dyes on average (red) compared to the native PSI protein that was exposed to the same coupling conditions but without dye attachment (blue). PSI that is not exposed to basic coupling conditions exhibits increased functionality (green) compared to PSI that was in incubated in a high pH solution (blue). When ATTO 590 dyes are attached to BSA, which is not photoactive, no reaction was observed (grey). (b) The average number of coupled dyes per PSI protein was increased from 5 to 34. The maximum oxygen consumption activity was observed for 32 coupled dyes per PSI monomer. (c) White-light illumination was used as a source with an intensity of 36 mW/cm2. (d) Oxygen consumption of PSI exposed to basic coupling conditions (blue). When carboxylic acid-functionalized ATTO 590 dyes are added to PSI solution without chemical coupling, the oxygen consumption remains unaffected (grey).
At the ferredoxin docking site, the electron is transferred to methyl viologen (MV+2) that is reduced to MV+1, which in turn reacts with molecular oxygen. The removal of oxygen can be easily measured by an oxygen electrode. The positively charged P700+• center is recharged with electrons from a donor molecule, here with 2,6-dichlorophenolindophenol (DCPIP), which was
Chapter 5
67
present in reaction solution. In this way, the activity of PSI under constant illumination can be measured for an extended period of time (see Experimental part)21. The oxygen consumption experiment was performed with native PSI and PSI modified with covalently attached ATTO 590 under white-light illumination (for spectrum see Figure 5.5c) as shown in Figure 5.5a. The time required for full oxygen consumption was reduced by a factor of 4 for PSI coupled with 32.2 dyes as compared to pristine PSI that was exposed to the same conditions as the dye-modified PSI. Moreover, it was observed that with an increasing number of dyes attached to PSI, the oxygen-consumption rate increased (Figure 5.5b). The maximum charge-separation activity of the dye-modified PSI complex was observed when an average of 32.2 dyes per PSI monomer was attached. An additional increase in the number of attached dyes did not lead to an increase in the functionality of the complex. This observation requires a more detailed study, but might originate from internal PSI structural changes when overloaded with attached dyes, or protein aggregation due to an increase in hydrophobicity of the protein surface. Our TEM studies showed that even after coupling of 20.5 dye molecules on average per PSI monomer the protein complex still retains its trimeric structure (Figure 5.2a). To verify that the increased oxygen consumption of PSI-dye complexes is indeed due to energy transfer from covalently coupled dyes to PSI, followed by subsequent charge separation, two control experiments were carried out. In the first one, the dye was covalently attached to a non-photoactive bovine serum albumin (BSA) scaffold following the same coupling procedure as for PSI (see Methods). In this control experiment, no oxygen consumption was detected, showing that the dyes by themselves do not lead to the oxygen consumption (Figure 5.5a). In the second experiment, carboxylic acid-functionalized ATTO 590 dyes (instead of the activated ester derivatives) were present together with PSI in the reaction buffer, ensuring that no covalent bonds could be formed with the PSI backbone. In this control experiment, no change in the oxygen consumption was detected with respect to pristine PSI (Figure 5.5d). These control experiments prove that the boost in functionality of the dye-modified PSI results from additional energy transfer from the organic dyes to PSI. This transfer does only occur if the dyes are covalently attached to the multi-protein scaffold. To further investigate the origin of the increased oxygen consumption, experiments with selective excitation were performed. Therefore, optical filters were introduced into the measurement setup. With a 660 nm band pass filter, thereby exciting the photosystem directly (ATTO 590 dyes are not excited), the oxygen-consumption rate of dye-modified PSI was similar to the rate of pristine PSI, which indicates that the chemical modification did not affect the PSI functionality and that the increased activity is not due to direct photo-reduction of MV+2 from the ATTO 590 dyes. In contrast, when a 580 nm band pass filter was introduced, dye-modified PSI showed significantly increased activity in relation to native PSI under the same conditions (Figure 5.6). Interestingly, the oxygen-consumption rate for modified PSI excited at 580 nm (dye absorption) was measured to be faster than native PSI excited at 660 nm (main absorption peak of PSI). This again indicates the occurrence of efficient energy transfer from the dyes to the main absorption peak of PSI.
Increasing photosystem I activity by covalently attached organic dyes
Figure 5.6. Normalized oxygen consumption experiment employing narrow bandpass filters 580 nm (FWHM: 10 ± 2 nm) and 660 nm (FWHM: 10 ± 2 nm) both exhibiting 50% transmission. Native PSI (black) and ATTO 590-modified PSI (red) show a similar oxygen-consumption rate when the 660 nm bandpass filter was introduced, with 0.96 mW/cm2 light power intensity after filter passing. The ATTO 590-PSI conjugate shows a ~2 fold faster oxygen-consumption rate (orange) in comparison with native PSI (blue) when applying the 580 nm bandpass filter, with 0.84 mW/cm2 light power intensity after filter passing.
5.5 Study of PSI-dye conjugate activity with the help of Kelvin spectroscopy
In addition to measuring charge separation of modified PSI in buffer solution, its activity was assessed in the solid state by means of surface photovoltage spectroscopy (SPV). We estimated that 2/3 of all charged amino acids (lysine, arginine, glutamic and aspartic acid and partly histidine in PSI) are located on the side of the iron-sulfur clusters of the PSI, Figure 5.7a. The corresponding amino acids are available for H-bonding and electrostatic interaction and therefore greatly facilitate self-assembly of this protein into a monolayer on a 2-mercaptoethanol modified gold surface via interactions with the polar hydroxyl groups. The majority of self-assembled PSI is oriented with the iron-sulfur clusters toward the gold surface. Light-induced charge separation in an oriented PSI monolayer assembled results in a dipole formation under illumination22 (Figure 5.7b). Under illumination, the charges are separated along the protein backbone and cannot be compensated by electron-accepting and donating molecules, as in the oxygen-consumption experiment. The light-activated charged dipole modifies the work function of the gold surface under constant illumination, which can be measured as a contact potential difference (CPD) as a function of excitation wavelength. The dipole direction as measured with SPV can be used to determine the average orientation of the protein in an assembled monolayer.
Chapter 5
69
a)
b)
c)
d)
Figure 5.7. (a) 3D structure of PSI-monomer according to a crystal structure analysis.10 Available amino groups of lysine residues for dye coupling on the top and bottom sides of PSI surface are indicated in red. The picture was generated with PyMOL software and ACD/ChemSketch. (b) Schematic picture of light induced charge separation in PSI-monomer with attached dyes (blue circles). (c) SPV spectra of native PSI monolayer on gold surface. The contact potential difference (CPD) under illumination of unmodified PSI showed negative values (green line), which coincide with the absorption spectrum of the unmodified multi-protein complex (inserted graph). (d) SPV spectra of PSI-ATTO 590 monolayer whereas the modified PSI spectrum (red line) showed appearance of a new peak in CPD at 590 nm (indicated by arrow), corresponding to the absorption spectrum of the dye-modified protein complex (inserted graph). PSI was modified with 6 ATTO 590 dyes on average per PSI monomer.
In Figure 5.7c, the surface photovoltage (where SPV =-same features as the PSI absorption spectrum. This indicates that charge separation indeed occurs upon excitation of a self-assembled PSI monolayer. For the dye-modified PSI (6 dyes attached), an additional peak appears in the SPV spectrum at a wavelength of 590 nm, which coincides with the absorption maximum of the coupled dye (Figure 5.7d). This feature is absent in the SPV spectrum of unmodified PSI. The appearance of a peak around 590 nm clearly indicates that the energy absorbed by the dyes is transferred to PSI and subsequently results in charge separation. A control experiment with dye-modified BSA (Figure 5.8b) did not show any discernable features in the SPV spectrum (Figure 5.8a), which confirms that dye coupling to a photoactive protein scaffold is necessary for the additional peak in the SPV spectrum. In order to exhibit changes in the SPV spectrum, the energy absorbed by the attached dyes has to be coupled to a PSI-mediated charge separation reaction, which is absent in BSA. As a result, these experiments prove the
Increasing photosystem I activity by covalently attached organic dyes
70
energy transfer and subsequent charge separation in PSI even in dry conditions as for example in a solid-state biosolar cell23.
a)
400 500 600 7000.00
0.04
0.08
0.12
0.16
BSA-ATTO 590 dyes
Wavelength (nm)
Abso
rptio
n Sp
ectru
m
b)
300 400 500 600 700 800
-302
-300
-298
-296
-294
-292
-290
BSA-ATTO 590
Wavelength (nm)
CPD
(mV)
Figure 5.8. (a) Absorption spectrum of BSA covalently modified with ATTO 590 after purification. This conjugate was used for the SPV experiments. Pristine BSA does not show any absorption in the visible region. (b) BSA modified with ATTO 590 dyes do not show any features in CPD spectra of the substrate at absorbing wavelength in comparison with PSI protein modifies in the same way.
In previous studies, chemical coupling of organic dyes to the small reaction center of the purple photosynthetic bacterium Rhodobacter sphaeroide with a molecular weight of 100 000 Da24 has been achieved and resulted in extended absorption spectra12, 13 14-16. A 2-3 fold enhancement in the charge separation rate was reported in solution when illuminating the bioorganic hybrid system at single wavelengths coinciding with the absorption maxima of the attached dyes15, 16. In contrast, we have demonstrated an up to 4 times improved functionality under white-light irradiation within a molecular complex that is one order of magnitude larger than the previously utilized reaction center. Hence we demonstrate that energy transfer can be easily accomplished within significantly larger dimensions between synthetic dyes and the natural antenna system. Broadening the absorption of a similar photosynthetic complex has been realized by conjugation of metal nanoparticles25. However, the effect of the inorganic nano-objects on the energy conversion of light into separated charges has not been investigated. Finally, we demonstrate here for the first time that functionality of PSI-dye hybrids can also operate with increased functionality in the dry state.
5.6 Conclusions
In conclusion, we have enhanced the activity of PSI by covalently coupling fluorescent dyes. The attached chromophores significantly enhance the absorption of the protein complex in the green region, where PSI does not absorb efficiently. Energy transfer from the organic dyes to PSI was confirmed by steady-state and time-resolved photoluminescence measurements. The dye-modified PSI showed a more than 4-fold enhancement in the oxygen-consumption rate under
Chapter 5
71
white-light excitation, indicating that the biochemical activity of PSI is significantly improved by covalent dye conjugation. The oxygen-consumption rate additionally showed a clear dependence on the number of attached dyes, reaching saturation for around 32 coupled dyes per monomer of PSI. Ultimately, charge separation in PSI upon excitation of the coupled dye was observed in an SPV experiment in a self-assembled monolayer on a solid metal surface. Our study shows that covalently coupled dyes can be used to fill the green gap in PSI and enhance the charge separation in this large megadalton protein complex both in solution and in dry conditions. As such, the hybrid systems consisting of PSI and organic chromophores reported herein represent important conjugates because they might serve as models to study energy transfer processes to and within PSI and they might improve the photovoltaic conversion efficiency of PSI as active component in biophotovoltaic devices and hydrogen production cells.
5.7 Methods
Cell cultivation. The thermophilic cyanobacterium T. Elongatus was grown under agitation (150 rpm) in BG11 medium17. The temperature was kept at 56°C, continuous light was applied at 50-60 μEinstein×m-2 ×s-1 and cell growth pursued until late log phase. At last, the cells were harvested by centrifugation [JLA 9.100 rotor, Beckman; 7500×g; 15 min], resuspended in Buffer A [20 mM HEPES (pH 7.5); 10 mM MgCl2; 10 mM CaCl2; 500 mM mannitol], snap-frozen in liquid nitrogen and stored at -80°C.
Thylakoid membrane preparation. Thylakoid membranes were prepared according to the following protocol, which represents a combination of two previously described preparation methods 26, 27. To this end, fresh or frozen cells were re-suspended in Buffer A and homogenized five times using a Dounce homogenizer. After addition of lysozyme [final concentration: 0.5% (wt/vol)] and a tip of a spatula of DNase, the cell suspension was incubated under slow agitation for 45 min at 37°C in the dark. Subsequently, the cells were lysed by two passages through a French Press (15000 psi; Constant Systems Limited, UK). Membranes were collected by centrifugation [JLA 16.250 rotor, Beckman; 38000×g; 20 min] and washed with Buffer A containing 3M NaBr. Afterwards, the membrane suspension was washed once with Buffer A and three times with Buffer containing 0.05% DDM (n-dodecyl- -D-maltoside) in order to remove the phycobilisomes. Finally, the thylakoid membranes were solubilized by incubation in Buffer A supplemented with 0.6% DDM for 30 min at 20°C in the dark. Non-solubilized material was pelleted by centrifugation [JLA 16.250 rotor, Beckman; 16000 rpm (38000×g); 20 min] and the supernatant was subjected to subsequent purification steps.
Photosystem I purification. For PSI purification, fast liquid protein chromatography was applied on solubilized thylakoid membranes acc. to 26, 27. The chromatographic purification was performed on ÄKTA Explorer 10 (GE Healthcare) using an anion exchange column (HiTrapTM Q HP, GE Healthcare). After column equilibration with Buffer A + 0.03% DDM, the sample was loaded and subsequently eluted by a linear gradient of (0-1 M) MgSO4. The green fluorescent
Increasing photosystem I activity by covalently attached organic dyes
72
fractions representing homogenous trimeric PSI were collected and afterwards desalted with Buffer A + 0.03% DDM using Vivaspin 20 columns (molecular weight cut-off: 100 kDa; GE Healthcare). Finally, the purified PSI sample was adjusted to a Chl a (Chlorophyll a) concentration of 320 μM (with Buffer A + 0.03% DDM), snap-frozen in liquid nitrogen and stored at -80°C. Determination of chlorophyll a and protein concentration. Chl a determination was performed in 100% methanol as described in Porra et al20. Dye coupling to PSI and purification. N-hydroxysuccinimide ester of ATTO 590 was purchased from ATTO-tec. The dye was solubilized in dry DMF at a concentration of 10 mM. The coupling reaction was carried out at slightly alcalic conditions. To this end, the pH of PSI aliquots in Buffer A+ 0.03% DDM (50 ml) was adjusted to 9.2 by NaOH addition (aq). The 10 μL of ATTO 590 was added together with 400 μL of PSI sample (with Chl a concentration of 320 mM) into coupling buffer for 1 hour at 4°C. Various stoichiometries of dye to surface exposed amine groups within PSI ranging from one to nine were adjusted. To separate the PSI-dye conjugates from the unbound dyes, a 100.000 MWCO viva spin centrifugal dialysis device was used multiple rounds at 4°C (between 20 to 50 times buffer was added depending on the coupling ratio). The purification procedure was repeated until the removed buffer solution did not show any detectable amounts of ATTO 590 as determined by a UV-vis spectrophotometer (Jasco V-630). Finally, the Chl a concentration was determined by UV-vis measurements to calculate the average number of dyes attached to PSI, taking into account that each monomer contains 96 porphyrin molecules. Oxygen consumption measurement. The electron transport rates of purified PSI complexes were monitored by measuring the light-induced oxygen consumption based on the Mehler reaction21. Measurements were obtained with an OXELP oxygen electrode (World Precision Instruments GmbH, Germany) at ambient temperature. After PSI was diluted with reaction buffer [30 mM HEPES (pH 7.5); 3 mM MgCl2; 50 mM KCl; 330 mM mannitol; 0.03% DDM] to a final Chl a concentration of 4 μg/ml, methyl viologen [final concentration: 6 mM], 2,6-Dichlorophenolindophenol [160 μM] and sodium ascorbate [2 mM] were added to the reaction. In order to ensure that the observed oxygen consumption is due to PSI activity, the PSII specific inhibitor 3-(3,4-dichloro-phenyl)-1,1-dimethyl-urea [10 μM] was present at all times. At first, the solution was stirred in the dark until a stable oxygen concentration was measured (usually 5 min). The assay was started by illumination with actinic light or yellow light (590 nm) and, subsequently, the oxygen consumption was monitored. Calibration of the setup was realized by measuring total currents of oxygen- versus nitrogen-saturated assay buffer.
PSI monolayer preparation. First, an Au surface was modified with 2-mercaptoethanol by exposing a 1 mM linker solution to the substrate overnight. A 20 PSI solution was drop casted on these substrates and kept in a closed jar to prevent PSI solution evaporation. After 2 h of immobilization, substrates were rinsed with MQ water and dried with a nitrogen gun.
Chapter 5
73
AFM study of self-assembled PSI monolayer. Height images were recorded in tapping mode by a Multimode 8 instrument with ScanAsyst and Controller V form Bruker. The instrument was equipped with a TESP silicon probe (42 N/m spring constant and tip radius of less than 10 nm) and operated at 320 kHz resonance frequency. Analysis of height images was performed with NanoScopeAnalysis 1.4 software.
Kelvin microscope characterization. The changes in contact potential difference (CPD) were measured by the Kelvin probe technique using a reference gold electrode (Besocke Deltha Phi, Germany). The measurements were conducted at room temperature in a nitrogen-rich ambient environment (~14% humidity). The samples were placed in a fully darkened Faraday cage and the CPD was monitored until it reached a constant value (±1 mV) before illumination was turned on. For SPV, the CPD signal was monitored as a function of photon energy. The illumination was provided by a 350 W xenon-mercury light source (Oriel Inc., USA) and passed through a double monochromator, model MS257 (Oriel Inc., USA). The light intensity was in the range 10–50 mW/cm2, with wavelength intervals of 1 nm and duration of 6 s per interval. Photoluminescence measurements. The samples were excited at 380 nm by the second harmonic of a mode-locked Ti:Sapphire laser delivering pulses of 150 fs with a repetition frequency of 76 MHz. The steady state PL was recorded using a silicon CCD detector from Hamamatsu, while the time-resolved PL was recorded by a Hamamatsu streak camera working in synchroscan mode. The PL spectra are corrected for the spectral response of the setup. All measurements are performed at room temperature.
Increasing photosystem I activity by covalently attached organic dyes
74
5.8 References
1. Jordan, P. et al. Three-dimensional structure of cyanobacterial photosystem I at 2.5 angstrom resolution. Nature 411, 909-917 (2001). 2. Bassham, J.A., Benson, A.A. & Calvin, M. The Path of Carbon in Photosynthesis .8. The Role of Malic Acid. J. Biol. Chem. 185, 781-787 (1950). 3. H. van Amerongen, L.V., R. van Grondelle Photosynthetic excitons. World Scientific Publishing Co. Pte. Ltd, Singapore (ISBN 981-02-3280-2) (2000). 4. Hu, X.C., Damjanovic, A., Ritz, T. & Schulten, K. Architecture and mechanism of the light-harvesting apparatus of purple bacteria. P. Natl. Acad. Sci. USA 95, 5935-5941 (1998). 5. Van Grondelle, R. Excitation-Energy Transfer, Trapping and Annihilation in Photosynthetic Systems. Biochim. Biophys. Acta. 811, 147-195 (1985). 6. Glazer, A.N. Light Harvesting by Phycobilisomes. Annu. Rev. Biophys. Bio. 14, 47-77 (1985). 7. Adolphs, J. & Renger, T. How proteins trigger excitation energy transfer in the FMO complex of green sulfur bacteria. Biophys. J. 91, 2778-2797 (2006). 8. Brettel, K. Electron transfer and arrangement of the redox cofactors in photosystem I. BBA-Bioenergetics 1318, 322-373 (1997). 9. Liu, H.J. et al. Phycobilisomes Supply Excitations to Both Photosystems in a Megacomplex in Cyanobacteria. Science 342, 1104-1107 (2013). 10. Mullineaux, C.W. Excitation-Energy Transfer from Phycobilisomes to Photosystem-I in a Cyanobacterium. Biochim. Biophys. Acta. 1100, 285-292 (1992). 11. Bryant, D.A. & Frigaard, N.U. Prokaryotic photosynthesis and phototrophy illuminated. Trends Microbiol. 14, 488-496 (2006). 12. Gundlach, K., Werwie, M., Wiegand, S. & Paulsen, H. Filling the "green gap" of the major light-harvesting chlorophyll a/b complex by covalent attachment of Rhodamine Red. BBA-Bioenergetics 1787, 1499-1504 (2009). 13. Roger, C. et al. Efficient energy transfer from peripheral chromophores to the self-assembled zinc chlorin rod antenna: A bioinspired light-harvesting system to bridge the "green gap". J. Am. Chem. Soc. 128, 6542-6543 (2006). 14. Lauterbach, R., Liu, J., Knoll, W. & Paulsen, H. Energy Transfer between Surface-immobilized Light-Harvesting Chlorophyll a/b Complex (LHCH) Studied by Surface Plasmon Field-Enhanced Fluorescence Spectroscopy (SPFS). Langmuir 26, 17315-17321 (2010).
Chapter 5
75
15. Milano, F. et al. Enhancing the Light Harvesting Capability of a Photosynthetic Reaction Center by a Tailored Molecular Fluorophore. Angew. Chem. Int. Edit. 51, 11019-11023 (2012). 16. Dutta, P.K. et al. Reengineering the Optical Absorption Cross-Section of Photosynthetic Reaction Centers. J. Am. Chem. Soc. 136, 4599-4604 (2014). 17. Rippka, R., Deruelles, J., Waterbury, J.B., Herdman, M. & Stanier, R.Y. Generic Assignments, Strain Histories and Properties of Pure Cultures of Cyanobacteria. J. Gen. Microbiol. 111, 1-61 (1979). 18. Du, H., Fuh, R.C.A., Li, J.Z., Corkan, L.A. & Lindsey, J.S. PhotochemCAD: A computer-aided design and research tool in photochemistry. Photochem. Photobiol. 68, 141-142 (1998). 19. Muh, F. & Zouni, A. Extinction coefficients and critical solubilisation concentrations of photosystems I and II from Thermosynechococcus elongatus. BBA-Bioenergetics 1708, 219-228 (2005). 20. Porra, R.J., Thompson, W.A. & Kriedemann, P.E. Determination of Accurate Extinction Coefficients and Simultaneous-Equations for Assaying Chlorophyll-a and Chlorophyll-B Extracted with 4 Different Solvents - Verification of the Concentration of Chlorophyll Standards by Atomic-Absorption Spectroscopy. Biochim. Biophys. Acta. 975, 384-394 (1989). 21. Izawa, S. Acceptors and donors for chloroplast electron transport. Methods Enzymol 69, 413-433 (1980). 22. Kronik, L. & Shapira, Y. Surface photovoltage phenomena: theory, experiment, and applications. Surf. Sci. Rep. 37, 1-206 (1999). 23. Gordiichuk, P.I. et al. Solid-State Biophotovoltaic Cells Containing Photosystem I. Adv. Mater. 26, 4863 (2014). 24. Chang, C.H. et al. Structure of Rhodopseudomonas-Sphaeroides R-26 Reaction Center. FEBS. Lett. 205, 82-86 (1986). 25. Carmeli, I. et al. Broad Band Enhancement of Light Absorption in Photosystem I by Metal Nanoparticle Antennas. Nano. Lett. 10, 2069-2074 (2010). 26. El-Mohsnawy, E. et al. Structure and Function of Intact Photosystem 1 Monomers from the Cyanobacterium Thermosynechococcus elongatus. Biochemistry-US. 49, 4740-4751 (2010). 27. Mukherjee, D., May, M., Vaughn, M., Bruce, B.D. & Khomami, B. Controlling the Morphology of Photosystem I Assembly on Thiol-Activated Au Substrates. Langmuir 26, 16048-16054 (2010).
Chapter 5
76
Summary
77
Summary
In the first chapter of this thesis, an introduction to the primary steps of photosynthesis is given. Moreover, the most important biological photoactive machines involved in this process are highlighted. Besides working in a biological context, these proteins were applied out of the native environment, i.e. the thylakoid membrane. Special emphasis is paid to the photosystem I (PSI) from thermophilic cyanobacteria. This multiprotein complex is able to harvest light and to perform charge separation.
In chapter two, it was shown how the very stable PSI protein complex from Thermosynechococcus elongates was immobilized on a surface. Two different linker molecules, 2-mercaptoethanol (2ME) and sodium 3-mercapto-1-propanesulfonate (MPS), are able to direct PSI trimers on gold surfaces. 2ME positions PSI so that electrons are injected into the metal surface once activated by light. In contrast, MPS directs the majority of PSI in such a way that the electron transfer proceeds in an upward direction according to the electron transport chain. This absolute determination of protein orientation was successful due to employing single molecule techniques like AFM and combining these with macroscopic techniques like using sharp metal tips from an eutectic Ga/In for temperature depended conductivity measurements. Moreover, it was found that the conductivity characteristics of PSI are changing from S-shape semiconducting in the dark to ohmic with a linear I-V curve under illumination.
In the third chapter, PSI was incorporated into a bulk heterojunction solar cell. Thereby, PSI was immobilized on a transparent ITO electrode as a monolayer. With the knowledge of orientation from chapter two, two type of devices were constructed. The PSI monolayer was covered with a blend of a hole conducting semiconducting polymer and the electron acceptor C60. This organic layer was coated with LiF/Al and MoO3/Al to yield a conventional and inverted solar cell configuration, respectively. The PSI monolayer induced a shift in the open-circuit voltage in both devices, which was used to determine average orientation of PSI molecules within the monolayer. With these novel experiments an organic bulk heterojunction was employed for the first time as a biophysical tool to determine an important biomolecular feature, i.e. the dipole moment. Moreover, the device fabrication procedure demonstrates that the PSI complex of thermophilic origin is stable against treatment with hydrophobic, chlorinated, high boiling organic solvents. In this devices, the electricity output was dominated by the phase separated blend of organic semiconductors.
In chapter four, however, a single PSI layer was responsible for photovoltaic activity of an all solid-state bio-solar cell. For the construction of such a device PSI trimers were immobilized on TiOX with a linker molecule. The biological layer was coated with a semiconducting polymer that acts as a p-conductor followed by evaporation of top contacts. The conjugated polymer was chosen because it does not absorb light in the visible range and, therefore, PSI was the only active component in this wavelength region. The resulting devices exhibited an open circuit voltage of
Summary
78
0.76 V, a current density of 2.9 mA/m2 and a fill factor of 45%. It was even possible to achieve constant device performance over seven days indicating that PSI is active and stable in an environment that is very different from the thylakoid membrane.
In an effort to increase the photovoltaic activity of PSI in such artificial environments, the megadalton multiprotein complex was functionalized with organic dyes. The absorption properties of the fluorophores were chosen so that they fill the green gap of the chlorophyll a molecules of PSI, which is the wavelength range between 500 to 600 nm. The attachment of the organic dyes proceeds via simple activated ester coupling. The PSI scaffold was decorated with various amounts of dyes ranging from 5 to 34 artificial chromophores. It turned out that the 96 dyes per PSI trimer is the best ratio of labelling since it resulted in an four-fold increase in photo activity in solution. Since the fluorescence of the dyes overlaps with the absorption of the photosynthetic pigments of PSI, energy transfer from the organic dyes to PSI occurs. These measurements, for the first time, demonstrated that the activity of the native PSI can be significantly boosted with a very simple bioconjugation strategy. Finally, it was proven that the hybrid consisting of PSI trimer and many covalently attached organic dyes functions in the dry state on a metal surface. Therefore, the last chapter is an important step forward in improving the activity of all-solid state bio-photovoltaic devices as presented in chapter four.
Overall, this thesis demonstrates that photo active biological building blocks can be successfully combined with electronic devices without losing the original natural activity. Therefore, the experiments performed herein are important to further integrate the fields of microelectronics and biological systems, which might be important in the future for interfacing the human body with electronic circuits or for bioenergy harvesting and conversion.
Samenvatting
79
Samenvatting
In het eerste hoofdstuk van dit proefschrift wordt een inleiding tot de belangrijkste stappen van fotosynthese gegeven. Verder zijn de meest belangrijke biologische foto-actieve systemen betrokken in dit proces benadrukt. Behalve werkzaam zijn in een biologische verband, werden deze eiwitten toegepast buiten hun natuurlijke omgeving, dat wil zeggen het thylakoïde membraan. Bijzonder aandacht wordt besteed aan het fotosysteem I (PSI) van thermofiele cyanobacteriën. Dit eiwitcomplex is in staat om licht te oogsten en lading scheiding uit te voeren.
In hoofdstuk twee werd getoond hoe het stabiele PSI eiwitcomplex van Thermosynechococcus elongatus geïmmobiliseerd werd op een oppervlak. Twee verschillende linker moleculen, 2-mercaptoethanol (2ME) en natrium-3-mercapto-1- propaansulfonaat (MPS), kunnen PSI trimeren op goudoppervlakken binden. 2ME positioneert PSI zodanig, dat na activatie d.m.v. licht, elektronen geïnjecteerd worden on het metaaloppervlak. Daarentegen stuurt MPS de meerderheid van PSI zodanig dat de elektronen overdracht verloopt in een opwaartse richting volgens elektronentransportketen. Deze absolute bepaling van eiwit oriëntatie was succesvol te bepalen door het gebruik van afzonderlijke-molecuul technieken zoals AFM, en deze te combineren met macroscopische technieken zoals het gebruik van scherpe metalen tips van een eutectische Ga/In voor temperatuur afhankelijke geleidbaarheids metingen. Bovendien werd bevonden dat de geleidbaarheids karakteristieken van PSI verschuift van een halfgeleidende S-vorm in het donker tot een ohmse lineaire I-V curve onder verlichting.
In het derde hoofdstuk werd PSI opgenomen in een heterojunctie zonnecel. Daarbij werd PSI als enkele laag geïmmobiliseerd op een transparante ITO elektrode. Met de kennis van oriëntatie uit hoofdstuk twee, werden twee soorten apparaten geconstrueerd. De PSI monolaag werd bedekt met een mengsel van een gaten-geleidende halfgeleidende polymeer en de elektronenacceptor C60. Deze organische laag werd bekleed met LiF/Al en MoO3/Al om een conventionele en omgekeerde zonnecel configuratie te verkrijgen, respectievelijk. De PSI monolaag induceerde een verschuiving in het open-circuit voltage in beide apparaten, welke gebruikt werden om de gemiddelde oriëntatie van de PSI moleculen in de monolaag te bepalen. Met deze nieuwe experimenten werd een organische bulk heterojunctie voor de eerste keer gebruikt als een biofysisch werktuig om een belangrijke bio-moleculaire functie te bepalen, d.w.z. het dipoolmoment. Bovendien toont de fabricageprocedure van deze apparaten aan dat het PSI complex van thermofiele oorsprong stabiel is tegen behandeling met hydrofobe, gechloreerde hoogkokende organische oplosmiddelen. In deze apparaten werd de elektriciteitsproductie gedomineerd door de fase gescheiden mix van organische halfgeleiders.
Echter, in hoofdstuk vier was een enkele PSI laag verantwoordelijk voor fotovoltaïsche activiteit van volledig vaste-staat bio-zonnecel. Voor de constructie van een dergelijk apparaat werden PSI trimeren geïmmobiliseerd op TiOx met een linker-molecuul. De biologische laag werd bekleed met een halfgeleidend polymeer dat fungeert als een p-geleider, gevolgd door verdamping van
Samenvatting
80
topcontacten. Het geconjugeerde polymeer werd gekozen omdat het geen licht absorbeert in het zichtbare gebied en derhalve was PSI de enige werkzame component in dit golflengtegebied. De resulterende apparaten vertoonden een open-circuit voltage van 0.76 V, een stroomdichtheid van 2.9 mA/m2 en een vulfactor van 45%. Het was zelfs mogelijk om een constante prestatie te verkrijgen van langer dan zeven dagen, wat aangeeft dat PSI actief en stabiel is in een omgeving die zeer afwijkend is van het thylakoïde membraan.
In een poging om de fotovoltaïsche activiteit van PSI in dergelijke kunstmatige omgeving te verhogen, werd het megadalton eiwitcomplex gefunctionaliseerd met organische fluoroforen. De absorptie eigenschappen van deze fluorofoor werd zodanig gekozen dat zij de groene kloof van de chlorofyl a moleculen van PSI zouden vullen, wat het golflengtegebied van 500 tot 600 nm betreft. De binding van de organische fluoroforen verloopt eenvoudig via geactiveerde ester koppelingen. Het PSI complex was met verschillende hoeveelheden fluoroforen bedekt, van 5 tot 34 kunstmatige chromoforen per PSI monomeer. Het bleek dat 96 fluoroforen per PSI trimeer de beste verhouding van labeling is aangezien dit tot een viervoudige toename van de foto-activiteit in oplossing leidde. Omdat de fluorescentie van de fluoroforen overlapt met de absorptie van de fotosynthetische pigmenten van PSI, kan energieoverdracht van de organische fluoroforen op PSI plaatsvinden. Deze metingen hebben voor de eerste keer aangetoond dat de activiteit van het natieve PSI aanzienlijk kan worden versterkt met een zeer eenvoudige bioconjugatie strategie. Tot slot werd aangetoond dat de hybride PSI, bestaande uit trimeren en vele covalent gebonden organische fluoroforen, functioneel is in droge toestand op een metalen oppervlak. Daarom is het laatste hoofdstuk een belangrijke stap voorwaarts in het verbeteren van de activiteit van volledig vaste-staats bio-fotovoltaïsche apparaten zoals beschreven in hoofdstuk vier. Kortom, dit proefschrift laat zien dat foto-actieve biologische bouwstenen succesvol kunnen worden gecombineerd met elektronische apparaten, zonder verlies van de oorspronkelijke natuurlijke activiteit. Daarom zijn de experimenten welke hierin zijn uitgevoerd belangrijk voor de verdere integratie van onderzoeksvelden op het gebied van micro-elektronica en biologische systemen, welke in de toekomst belangrijk kunnen zijn voor het koppelen van het menselijk lichaam met elektronische circuiten of voor het oogsten en de conversie van bio-energie.
Publications
81
Publications
1. Shaohua Liu, Pavlo Gordiichuk, Zhongshuai Wu, Zhaoyang Liu, Wei Wei, Dongqing Wu, Yiyong Mai, Andreas Herrmann, Klaus Müllen, and Xinliang Feng, Patterning Two-Dimensional Free-Standing Surfaces with Mesoporous Conducting Polymers, Nature Communications, 6, 8817 (2015).
2. Vasyl Skrypnychuk, Gert-Jan A. H. Wetzelaer, Pavlo Gordiichuk, Stefan C. B. Mannsfeld, Andreas Herrmann, Michael F. Toney and David R. Barbero, High vertical mobility in an organic semiconductor, Advanced Materials, online (2015).
3. Pavlo Gordiichuk*, Olga E. Castañeda Ocampo*, Stefano Catarci, Daniel A. Gautier, Andreas Herrmann and Ryan C. Chiechi, Mechanism of Orientation-Dependent Asymmetric Charge Transport in Tunneling Junctions Comprising Photosystem I, J. Am. Chem. Soc., 2015, 137 (26), pp 8419–8427 (2015).
4. Pavlo Gordiichuk, Avishek Paul, Daniel A. Gautier, Agnieszka Gruszka, Manfred Saller, Jan Willem de Vries, Gert-Jan A. H. Wetzelaer, Marianna Manca, Widianta Gomulya, Dolev Rimmerman, Paul W. M. Blom, Shachar Richter, Maria Antonietta Loi, Andreas Herrmann, Increasing the activity of a megadalton Photosystem I complex by attachment of organic dyes, Bioconjugate Chemistry, online (2015).
5. Zheng Cao, Pavlo Gordiichuk, Katja Loos, Ernst J. R. Sudhölter, Louis C. P. M. de Smet, The effect of guanidinium functionalization on the structural properties and anion affinity of polyelectrolyte multilayers, Soft Matter, online (2015).
6. Pavlo Gordiichuk, Gert-Jan A H Wetzelaer, Dolev Rimmerman, Agnieszka Gruszka, Jan Willem de Vries, Manfred Saller, Daniel A Gautier, Stefano Catarci, Diego Pesce, Shachar Richter, Paul W M Blom, Andreas Herrmann, Solid-state biophotovoltaic cells containing photosystem I, Advanced Materials, 26(28): 4863-9 (2014).
7. Naureen Akhtar, Alexey O. Polyakov, Aisha Aqeel, Pavlo Gordiichuk, Graeme R Blake, Jacob Baas, Heinz Amenitsch, Andreas Herrmann, Petra Rudolf, Thomas T M Palstra, Self-assembly of ferromagnetic organic-inorganic perovskite-like films, Small, 10(23): c4912-9 (2014).
8. Vladimir Derenskyi, Widianta Gomulya, Jorge Mario Salazar Rios, Martin Fritsch, Nils Fröhlich, Stefan Jung, Sybille Allard, Satria Zulkarnaen Bisri, Pavlo Gordiichuk, Andreas Herrmann, Ullrich Scherf, Maria Antonietta Loi, Carbon nanotube network ambipolar field-effect transistors with 108 on/off ratio, Advanced Materials, 26(34): 5969-75 (2014).
9. Jelena Ciric, Albert J.J. Woortman, Pavlo Gordiichuk, Marc Stuart, Katja Loos, Physical Properties and Structure of Enzymatically Synthesized Amylopectin Analogs, Starch/Stärke, 65: 1061–1068 (2013).
Publications
82
10. Widianta Gomulya, Guadalupe Diaz Costanzo, Elton J. Figueiredo de Carvalho, Satria Zulkarnaen Bisri, Vladimir, Derenskyi, Martin Fritsch, Nils Frohlich, Sybille Allard, Siewert Jan Marrink, Maria Cristina dos Santos, Ulrich Scherf, Pavlo Gordiichuk, Andreas Herrmann Maria Antonietta Loi, Semiconducting single-walled carbon nanotubes on demand by polymer wrapping, Advanced Materials, 25(21): 2948-56 (2013).
11. Satria Zulkarnaen Bisri, Jia Gao, Vladimir Derenskyi, Widianta Gomulya , Igor Iezhokin, Pavlo Gordiichuk, Andreas Herrmann, Maria Antonietta Loi, Advanced Materials, 24, 46, 6147-6152 (2012).
12. Alina Veligura, Paul J. Zomer, Csaba Jozsa, Pavlo Gordiichuk, Nikolaos Tombros, Harry T. Jonkman and Bart J. van Wees, Relating hysteresis and electrochemistry in graphene field effect transistors, Journal of Applied Physics, 110 , 11, 113708 - 113708-5 (2010).
Publications submitted to scientific journals
13. Pavlo Gordiichuk, Vladimir Derenskyi, Widianta Gomulya, Gert-Jan A. H. Wetzelaer, Maria Antonietta Loi, Andreas Herrmann, Current Flow in a Single-Walled Carbon Nanotube Network Field-Effect Transistor, submitted (2015).
14. Solmaz Torabi, Fatemeh Jahani, Pavlo Gordiichuk, Andres Herrmann, Jan C. Hummelen, L. Jan Anton Koster, Doping Effect of LiF Interlayers in Fullerene Based Devices, at submission (2015).
15. Lai-Hung Lai, Widianta Gomulya, Jih-Sheng Yang, Pavlo Gordiichuk, Martin Fritsch, Sybille Allard, Andreas Herrmann, Jih-Jen Wu, Ullrich Scherf, and Maria A. Loi, Polymer Wrapped Semiconducting Single-walled Carbon Nanotubes Toward Efficient H2 Evolving Photocathodes, submitted (2015).
Publications in preparation for submission to scientific journals
16. Pavlo Gordiichuk, Diego Pesce, Alessio Marcozzi, Gert-Jan A. H. Wetzelaer, Olga E. Castañeda Ocampo, Avishek Paul, Mark Loznik, Ryan C. Chiechi, Andreas Herrmann, Orientation and Incorporation of Photosystem I in Bioeletronic Devices mediated by Phage Display, in preparation (2015).
17. Pavlo Gordiichuk, Gert Jan H. Wetzelaer, Dolev Rimermann, Shachar Richter, Andreas Herrmann, Photo-electrical properties of PSI self-assembled monolayer on modified metallic surface, in preparation (2015).
18. Pavlo Gordiichuk, Talieh Ghiasia, Avishek Paul, Shaohua Liu, Widianta Gomulya, Mark Loznik, Maria Antonietta Loi, Xinliang Feng, Andreas Herrmann, Fabrication of bio-photoactive graphene-based material by self-assembly of Photosystem I protein complexes, in preparation (2015).
Acknowledgements
83
Acknowledgements
In the end of my thesis I would like to acknowledge people who supported me during my Ph.D. studies and contributed to this thesis.
First, I give my biggest gratitude to my supervisor Prof. Andreas Herrmann, who offered me the possibility to pursue my Master and Ph.D in his group. I am very thankful for your supervision and all your help during all these five years. Your personal energy and interest in different projects gave me a lot of motivation. I am very grateful to you for teaching me writing of scientific papers and helping me with my English. Thank you for giving me research freedom and the possibility to work on my own ideas and projects. We finished a lot of work successfully and collected a lot of experiences. Thank you for establishing a new AFM machine in your group. I think it was a great investment which helped us to run very exciting and challenging projects. And finally, it helped me to became an advanced user of AFM and to find a postdoc position in Northwestern University.
I want to thank my reading committee: Prof. C.H. van der Wal, Prof. X. Feng, Prof. S. Hecht for reading and approving the manuscript in such a limited period of time.
I would like to thank all my collaborators during my PhD time. First of all, Prof. Maria Loi, Widi, Jia, Satria, Alex, Marianna, Vlad and Lai Hung for the help with the spectroscopic characterization techniques and involving me in a collaboration dealing with carbon nanotubes. It was a great pleasure to work with all of you. My special thanks go to Widi for all your kind help as well as very fast and responsible work. You showed the highest level of research skills and time management capabilities during our work. I want to thank to Prof. Perta Rudolf for being my mentor and for a nice collaboration with Naureen, Alex and Aisha. It was a big pleasure to work with you.
Next, I want to say thank you to Prof. Ryan Chiechi, Olga E. Castaneda Ocampo and Yanxi Zhang. Thank you for the numerous discussions we had, and all your efforts in the PSI project. Your experiments helped us to clarify the mechanism of electron transport through the PSI protein, which is one of the main focus of my thesis. Thank you very much for all our discussions.
I also would like to say a big thanks to Prof. Katja Loos, Cao, Jelena and Qiuyan for our collaborations and your help. It was a great pleasure to work with you and thank you, Katja, for appointing me as a AFM practicum supervisor. It was a good experience for myself to learn more about AFM and polymers. Moreover, thank you for involving me in your research projects, which are finalized in two publications.
Acknowledgements
84
I would like to thank Prof. Paul Blom and Gert-Jan for all our collaborations and discussions. It was very nice to work with you, especially, thank you very much for fabrication of solar cells with PSI.
Additionally, I would like to thank Prof. Jastin Ye, Jie, Jianming and Sasha for all the experiments we did together with 2D TMD materials. I am grateful for all the discussions and work. It was a great introduction for me into science of two dimensional materials.
Next, I would like to say thanks for our collaborations outside the Zernike Institute for Advanced Materials.
Firstly, to Prof. Xinliang Feng and Shaohua from Dresden University of Technology. I enjoyed work with you, and I am very happy that our latest efforts were published in Nature Communications. Shaohua, I am very happy that we are still working together on a current project. I must say, that you are the most courteous person I have ever met.
I would like to say thanks to Prof. David Barbero and Vasyl Skrypnychuk from Umeå University, for a collaboration dealing with P3HT polymer arrangement in vertical direction. I am very thankful for this collaboration and the results we obtained.
I would like to thank to Prof. Shachar Richter, Dolev Rimmerman and Katya Gloukhikh from Tel Aviv University. It was a great pleasure to work with you on the PSI related projects. Thank you very much for our Skype meetings and discussions. I am very grateful for the SPV measurements you performed, which are included in this thesis.
Next, I would like to acknowledge the PCBE group members. I would like to say thank you to Evgenuy Polushkin for SEM measurements. Your SEM pictures have always the best quality. Also, I would like to acknowledge Mark Loznik for helping with the PSI production, Dutch language and for the correction of my papers and chapters. Also, I want to say thanks to Manfred, Diego, Alessio, Danie, Avishekl and Stefano for the collaborative work on PSI modification with dyes.
I would like to thank Dr. Bart Crielaard for our discussions and your kind help with writing papers and my first grant application. It was a great pleasure for me and a lot of experience. Also, I would like to say thanks to Hongyan, Kseniya and Karolin. Karolin, additionally, thank you very much for being my paranymph and for being very nice person. Also, I would like to acknowledge Prof. Patrick van Rijn for useful discussions about surface chemistry. Thank you for all your help and your inspiring style of work.
I want to say thanks to Kai. I am very happy that we are friends and that we can discuss a lot of things. Also, I want to say thank you to Lifei. Discussions with you are always very valuable and fruitful. I was very happy to share the office with you and thank you very much for being my paranymph.
I would like to say thank you to all our PCBE group members: Lei, Jingyi, Zhuojin, Agnieszka, Jan Willem, Tiancai, Feng, Gurudas, Konstantin, Jur, Eliza, Alina, Deepak, Minseok, Konstantin,
Acknowledgements
85
Alberto, Anke, Tobias, Andrew, Wei, Cao, Jing, Qing, Jun, Pei, Xingfei. Additionally, I also would like to thank all my students for their work: Edmond, Mallik and Talieh. Also, a special thanks goes to Albert, Csaba, Yi, Azis, Mai, Anton, Jacob, Dina, Niels, Ivan, Martijn, Jin, Marjot, Dennis and Xu.
I would like to thank all my friends with whom I had great times in Groningen outside the lab: Sasha, Mikalai, Christina, Hanna, Andrii, Oleg, Leonid, Oleksander, Maria, Sabrina and Andreas.
Finally, I want to say thanks to my parents and my brothers: Ivan, Dima and Miroslav with their families for all their support during this time and great thanks to Ilja and Ema.
