Skip to main content

Plasmonic Enhancement and Directionality of Emission for Advanced Luminescent Solar Devices

Periodic Reporting for period 3 - PEDAL (Plasmonic Enhancement and Directionality of Emission for Advanced Luminescent Solar Devices)

Reporting period: 2018-04-01 to 2019-09-30

Applying photovoltaic (PV) panels to buildings is an important application for wider PV deployment and to achieving our 20% Renewable Energy EU targets by 2020. PEDAL will develop a disruptive PV technology where record increases in efficiency are achieved and costs dramatically reduced;
(1) Diffuse solar radiation will be captured to produce higher efficiencies with concentration ratios over 3 in plasmonically enhanced luminescent solar concentrators (PLSC). Current LSC efficiency achieved is 7.1%, [1]. This proposal will boost efficiency utilising metal nanoparticles (MNP) tuned to luminescent material type in LSCs, to induce plasmonic enhancement of emission (PI and team have achieved 53% emission enhancement). MNP will be aligned to enable directional emission within the LSC (being patented by PI and team). These are both huge steps in the reduction of loss mechanisms within the device and towards major increases in efficiency.
(2) Plasmonically enhanced luminescent downshifting thin-films (PLDS) will be tailored to increase efficiency of solar cells independent of material composition. MNP will be used, where the plasmonic resonance will be tailored to the luminescent species to downshift UV. MNP will be aligned to enable directional emission within the PLDS layer, reducing losses enabling dramatic increases in a layer adaptable to all solar cells.
(3) These novel systems will be designed, up-scaled and a building integrated component fabricated, with the ability not only to generate power but with options for demand side management.
Previous work has been limited by quantum efficiency of luminescent species, with this breakthrough in both the use of MNP for plasmonic emission enhancement and alignment inducing directionality of emission, will lead to efficiencies of both PLSC and PLDS being radically improved. PEDAL is a project based on new phenomena that will allow far reaching technological impacts in solar energy conversion and lighting.

The objectives are:
• To engineer composites (luminescent species (dye/QD) and MNPs in polymer) for PLSC and PLDS and to determine, validate, and maximize manipulation of the optical properties of luminescent species through modifying the localized electrical boundary condition by exploiting the Plasmonic field.
• To achieve record efficiency in a Luminescent Solar Concentrator by exploiting plasmonic coupling phenomena between metal nanoparticles and luminescent species (dye/QD) to enhance emission and alignment of MNP for directional emission.
• To achieve record efficiencies using Luminescent Downshifting Layers to generate more power from a matched solar cell by exploiting plasmonic coupling phenomena, converting solar radiation outside the bandgap of the cell to within its absorption band, along with alignment of MNP to achieve directional emission.
• To develop the first static building integrated PLSC component capable of achieving a concentration factor of 3 or more in diffuse solar radiation.
• To develop the first static building integrated PLDS layer on a commercial PV module capable of achieving an increase in PV efficiency.
• To investigate the balance of systems options for the building integrated devices to enable compatibility with building demand side energy management.
Period covered by the report: from 01/04/2015 to 30/09/2016
Periodic report: 1st
1. Overview of the action's implementation for this reporting period
The following section outlines the tasks and deliverables that have been completed as outlined in the PEDAL proposal. It provides a (i) global overview of the implementation of PEDAL as well as presenting (ii) the dissemination and training activities by the PI and her team in the Solar Energy Applications Group (SEAG) at Trinity College. Further to this (iii) problems that have arisen have been outlined and the delays explained. These issues have been resolved and the team continues to work hard to achieve the outputs of PEDAL
(i) Overview of PEDAL implementation
Plasmonic Luminescent Down-Shifting (pLDS) and Plasmonic Luminescent Solar Concentrator (pLSC) are new optical approaches to increase PV device efficiency by using plasmonic coupling between luminescent materials and metal nanoparticles (MNP). The optical properties of fluorescent species can exhibit dramatic spectral changes in the presence of metal nanoparticles. It is therefore proposed to achieve record efficiency in pLDS and pLSC layers by exploiting plasmonic coupling phenomenon to enhance absorption/emission and alignment of MNP for directional emission. The optical response of metal nanoparticles and their interaction with luminescent species can be investigated by optical measurement of absorption and emission. The main challenge, however, is controlling the composite structure to achieve maximum enhancement.
The first phase of this research is to prepare stock solutions of MNP, mainly Silver and Gold Nanorods (Ag NRs, Au NRs) to be used in further step with luminescent materials (BASF Lumogen series of dyes, rare earths complex Eu(tta)3phen) and QDs with high LQE).

Synthesis of Ag NRs:
• Preparation of Ag NRs stock solution
Plasmonic coupling with an optical emitter molecule is a function of several parameters and one of them is the surface plasmon resonance (SPR) frequency of the MNPs, which is a function of size and shape. Therefore it is required to synthesize Ag NRs/Au NRs with precise control over shape & size consequently the aspect ratio to investigate & optimize the Plasmonic coupling. The bottom up approach synthesis procedure was used to synthesize Ag NRs. In this process, Hexadecyltrimethyl Bromide (CTAB) was used as a template to facilitate the growth of NRs at room temperature. The procedure was a seed mediated approach and a similar recipe will be used in the future to synthesize Au NRs. Silver ions were reduced with strong reducing agent (sodium borohydride) in the presence of trisodium citrate, as capping agent stabilizer. Then, the prepared seeds were added to a solution containing more metal salts (Na OH), a weak reducing agent (ascorbic acid) and to the CTAB.
• Optimisation of Ag NRs growth solution
Getting a stable and high yield of Ag NRs is crucial as it insures lesser amount of impurities in the solution. Temperature, pH, size of the seeds, impurities are some of the factors that influence the aspect ratio and yield of the NRs. Optimisation of Ag NRs has been achieved through optical measurements and Microscopic measurements. UV-Vis spectrometer was used to measure the absorbance of Ag NRs. Figure 1 shows UV-Vis spectra for Ag NRs prepared in water. The graph in the left shows UV-Vis spectra for Ag NRs grown with 60 µl seed solution and has a longitudinal peak at 700nm but the yield of NRs is quite low as can be seen by the intensity of the longitudinal peak of the NRs. The graph on the right however, shows UV-Vis spectra of Ag NRs after optimisation of the synthesis procedure, with a higher yield of the NRs. The peak at 440 nm in both graph spectra are attributed to the presence of spherical or/and triangles spheres.

The morphology and size distribution of the Ag NRs were characterised by SEM images using ZEISS Ultra PLUS at an accelerating voltage of 5 kV. After the centrifugation process, samples were re-suspended in distilled water. Thereafter the solution of the Ag NRs diluted 2:10000, followed by ultra-sonication for 60 seconds. A 30 ml of the colloidal solution was deposited on a silicon wafer substrate by spin coating for 30 seconds at 1000 rmp with acceleration time of 3 seconds and left overnight to dry prior to measurements.

It revealed and confirmed that spherical and triangles spheres are present in Ag NRs solution were prepared. The length and width of the NRs for 50 µl are 617nm and 20 nm respectively.
A stock solution of Ag NPs has been papered for the pLDS and pLSC development. Thickness optimisation of LDS layers and mould optimisation for LSC are completed.
The stock solution however contained the Ag spheres/ triangles along with the NRs. It is important to separate these particles from the NRs for directional emission. Having done so, the NRs will be transferred to an organic solvent such as Toluene which has a good compatibility with PMMA and epoxy.

Optimisation for LDS thickness and LSC mold construction
An optimisation process was carried out for two different thicknesses 10 µm and 100 µm of LDS layer. For the 10 µm Spin Coating technique was used for thickness optimisation. Different layers were fabricated with different RPM and different spin duration. From table 1 thickness of ~10 µm was found to be achieved at time duration between 150 -210 seconds when 1000 RPM was used. 2 ~10 µm was achieved at RPM between 1000 -1500. Therefore 180 sec at 1000 RPM was chosen as the optimum spin duration time. This has been verified by making 5 different samples with 1000 RPM and 180 sec.

For the 100 µm casting technique was used for layer fabrication. Clear PMMA solution with 50wt% was prepared and five different layers were drop cast on glass substrates (size of 25 ×25 ×1 mm) and cured for 72 hours at 25 oC. Layers were found to have excellent uniformity of size which is confirmed by the thickness measurement at various positions across the glass substrate.

LSC moulds were constructed by securing an acrylic plate with the required LSC plate dimensions between two solid acrylic sheets. The middle and top plates were both cut to the required dimensions. A clear cast epoxy resin was used for testing the mould. Epoxy resin contains two parts, resin and hardener was mixed in weight ratio 100: 42. The mixed resin was then poured into the mold at room temperature (20-24 oC) and left for 120 hours to cure. Tow plate have been fabricated and was observed to be quite uniform when removed from the mould.
Optical measurement for the LSC resin plates were obtained using an integrating sphere UV-VIS/NIR spectrometer to confirm the uniformity of the plates through the absorption and transmission spectra. The measurement has been carried out at four different positions on the fabricated LSC plates. The absorbance and transmittance spectra

• Shape separation using centrifugation: (in progress)
The Ag NPs mother solution (as-synthesized) contains spherical and triangles spheres. There is a need to separate them even if crudely based on the shape. Settling velocity is shape and size related therefore, nanoparticles with different shape behave differently under same centrifugal force. Nanoparticles with higher aspect ratio sediment more slowly than the nanoparticles with same length but lower aspect ratio. Centrifugal process was carried out using Eppendorf Centrifuge 5804R. Method described was used for NRs separation process however the speed (2000rpm/30) mentioned for shape separation in the paper did not work in our case. Optimisation of centrifugal force is therefore needed.
2000rpm/30min: The pellet formed at this speed had not solidified enough to separate it from the solution. Therefore the UV-Vis obtained of the solution contained the Ag spheres/ triangles along with the NRs. Repeated centrifugation resulted in the disintegration of the NRs i.e. a massive blue shift in the longitudinal peak.

8000rpm/30min: The solution obtained in this case is colourless and shows nothing in the UV-vis. There is a formation of a pellet along with a residue at the sidewalls of the centrifuge tube. The UV-vis spectra was taken of the (pellet+ side residue) and it shows no shape separation. It is possible that the long NRs are on the sidewall of the centrifuge tube while the spheres have been pushed down into the pellet at the bottom.
Shape separation experiments are in progress and different centrifugal process will be examined.
• Phase Transfer of Ag NRs (in progress)
The NRs need to be transferred to an organic common solvent such as Toluene. To achieve that we have to remove the CTAB from the Nanorods surface and replace it with a group such as Thiol which has a strong affinity for Gold and silver surface. After the Nanorods are thiolated they can be dispersed into toluene, by a common medium that is soluble in both Toluene and Water such as acetone.

Task 1.2 and 2.2: Modelling and characterisation
Deliverables 1.2.1 & 2.2.1 PLSC & PLDS developed model
Deliverables 1.2.2 & 2.2.2 PLSC & PLDS Validated model
Task 1.2: Modelling PLSC
The Plasmonic Luminescent Solar Concentrator (PLSC) devices employed the plasmonic coupling between optical emitter (quantum dots (QDs) and dye) and metal nanoparticles (Gold (Au) and Silver (Ag)). Hence, to optimize the PLSC device using modelling required to develop a model that can accommodate this demand. The development of PLSC device modelling is mainly divided in two parts. In the first part; two individual models were developed, optical ray-tracing and Finite Difference Time Domain (FDTD). Optical ray-tracing model to predicted the optical properties of optical emitter (quantum dots and dye) and their Luminescent Solar Concentrator (LSC) and FDTD modelling to model the optical properties of metal nanoparticle (Ag, and Au). In the second part; these two individual models are combined to model PLSC device.

1.2.1 PLSC Developed Model
The ray-tracing model for LSCs device developed in the groups earlier work was used. The ray–tracing model algorithm was successfully designed. In the ray-tracing model, the each ray (photon) is represented by mathematical vector, which turns out a number and that number decided the iteration for numerical simulation. The algorithm is based on the Monte Carlo principle, where at each branching point in the flow diagram, randomly generated numbers are tested against the respective calculated probabilities to determine whether the event (reflection, absorption, scattering, emission) ensues or not.

In the second section, the FDTD model was developed to predict optical properties (electric field intensity, photon mode density, and surface plasmon resonance peak) of gold and silver nanoparticles of various shape and size. The particular interest is to model the plasmon resonance enhanced electric field intensity and photon mode density in the vicinity of metal nanoparticles because that is used to manipulate the plasmonic coupling between optical emitter (QDs and dye) and metal nanoparticles(Au and Ag) in the PLSC device. The algorithm was based on to solving the Maxwell equation for employed condition.

1.2.2 PLSC Validated Model
Since the ray-tracing LSCs model is based on Monte Carlo and probability theory, hence, it required certain number of iterations. The number of iteration is decided through the number of rays (photon). The model was run for different number of iteration and this optimized the number of iteration required hence simulation duration. After a certain number of iterations the optical efficiency of LSC device remained constant, therefore this is optimized number of iteration and simulation.

The ray-trace model for LSCs was validated using previous work from the group and had an excellent correlation. The FDTD model validation was carried out for gold nanoparticle of spherical and nanorods shape, which is reported in the literature. First the model was validated for the spherical shape of gold nanoparticle and the extinction spectra of 25 nm diameter. The FDTD model was validated for anisotropic gold nanoparticle of nanorods shape with diameter of 25 nm and length of 100 nm.

Task 2.2: Modelling PLDS
The Luminescent dawn-shifting (LDS) layer in principle and technically is similar to thin film LSCs. In the Luminescent down-shifting layer the geometric gain is independent of size and remains at unity and in LSCs it depend on the size of device. Hence the Ray trace model described in the section 2.1 can apply to predict output in of LDS, however , the LSC ray tracing model did not work, hence required a new algorithm , which is called now, thin film LSCs ray tracing model or multilayer ray tracing model.

2.2.1 PLDS Developed Model
The ray-tracing model for LSCs described in the section 1.2.1 is altered to accommodate requirement of thin film LCSs. This model called as thin film LSC ray-tracing or multilayer ray-tracing model, which will used to model both, thin film LSCs and LDS. Then it will combined to FDTD model to develop PLDS model. Though, in principal it similar to LSCs ray-tracing model, however, due to very high absorption coefficient of optical emitter material in thin film. To counter this effect, a new algorithm was written.
This thin film LSCs model has successfully modelled the thin film LSCs and predicted Luminescent down-shifting (LDS). The model validation was carried out. The thin film LSC device of 25×25×1 mm fabricated using spin coating of red dye doped polymer on a clean glass substrate and refractive index of n≈1.5 with varying dye concentration from 0.1 to 1.1 wt%.
During PEDAL, the intrinsic properties of the solar cells will not be modified. Indeed, PEDAL will only allow the development of a new technology that will improve the performance of these solar cells by matching the solar emission with their absorption band. PLSC and PLDS can be considered as photon conversion techniques and are optimized as an independent optical process, dissociated from the particular physical properties of the operating semiconductor material or solar cell architecture. As a result, the photon conversion devices may be combined with all existing solar cells devices. The optical option is much more versatile, allowing independent and unique research; a method of PV system improvement, lowering of thermalisation and easy implementation at industrial level.

This proposal is a completely new approach (based on plasmonics and directional emission in luminescent devices) for solar cell efficiency improvement, which will require long-term and high quality research. Not only is the underlying physical phenomena highly novel and unconventional, but PEDAL also avoids changing the composition of the PV materials, which would be the standard approach to this problem.

The final products of this project will be new, completely portable, adaptable building component (PLSC) or as an efficient plastic matrices (PLDS) that will be suitable for their adaptation onto all types of solar panels.

PEDAL is an ambitious and, at the same time, realistic project, with clear and specific goals both in the scientific and engineering aspects. The solutions suggested by this project will mean not only a radical improvement of solar cell performance, but also an ambitious research at the basic level with a view to industrial applications. The approach of this project represents a great challenge both for the scientific, technological and engineering fields.

In summary, this project will enable highly significant scientific advances in the field of photon energy conversion processes due to its originality, high scientific content and its expected results. Moreover, PEDAL is a project based on new phenomena that will allow very high technological impacts in terms of solar energy conversion. The progress beyond the state of the art are multiple: new knowledge, new advances for Science and new concrete results for plasmonic enhancement of emission in both PLSC and PLDS systems as well as the alignment of MNP enabling directional emission, as well as improved building integrated devices.