Solution-processed Mg doped ZnO as buffer films for thin film photovoltaics

Abstract

The subject of this thesis is the development of a chemical bath deposition (CBD) route for magnesium doped zinc oxide (ZnMgO) - a wide band, n-type semiconductor material. ZnMgO is relevant for a wide range of applications, among others for optical coatings, optoelectronic devices (such as touch screens and light emitting diodes) and thin film photovoltaics (TFPV). In the last case, where this thesis focuses, ZnMgO is mainly investigated as buffer layer, combined with chalcogenide (Cu-In-Ga-S(Se) and CdTe), kesterite (Cu-Zn-Sn-S(Se)) or oxide (e.g. Cu2O) absorbers [1], [2]. The motivation to use ZnMgO does not only arise from its appropriate and tunable electronic properties (such as the band-gap width), but also from the fact that it consists exclusively of earth-abundant and non-toxic elements. To achieve high power conversion efficiency in a solar cell, the band-offset between buffer layer and absorber is a critical parameter. The commonly used buffer layer material that matches absorbers of well-established thin film technologies, namely cadmium telluride (CdTe) and Cu-In-(Ga)-S(Se) (CIGS), is the toxic cadmium sulfide (CdS). The replacement of this enviromentally harmful material with alternatives that possess similar or better properties, is of outmost importance, as it will render TFPV more attractive for large scale implementation. ZnMgO has the benefit of a tunable band gap (3.37 - 4 eV) [3] depending on the Mg content in the ZnO lattice, which can be used to optimize band offsets with various absorber materials. So far, ZnMgO films were mainly deposited by vacuum techniques. In order to lower the production costs, simple methods without expensive equipment, like chemical bath deposition (CBD), are desired. The available literature on solution-processed Zn- MgO films is very limited. This thesis provides a fundamental investigation on the CBD of ZnMgO films. The solution chemistry was theoretically examined via speciation modelling. Experimentally it was found that deposition is only possible in a narrow pH window, due to the pH-dependent supersaturation of Zn(OH)2 as a precursor for ZnO. The highest amount of magnesium incorporated into the ZnO lattice was 2.1%, quantified via inductively coupled plasma - optical emission spectroscopy measurements, suggesting that further incorporation is not possible, due to the thermodynamic solubility limit of 4 mol% under equilibrium conditions [4]. Although the achieved band-gap variation of 3.43-3.55 eV for the deposited ZnMgO films is moderate, the smooth, orientated and highly transparent (>80%) films, have implementation potential as buffer layers in solar cells . It was further shown by X-ray diffraction, that the amount of certain Mg species in the solution defines the ZnO film morphology. The cause of this, is most likely faceselective adsorption of the bulky [MgCit]-, as well as Mg2+ and Mg(OH)+ species onto 11 specific ZnO faces. Film morphologies with dominant (100), (101) or (002) ZnO crystal faces were observed depending on the solution composition. Further, a correlation between Mg incorporation and the growth mechanism was proposed. The precise control of the surface morphology allows to optimize the absorber/buffer interface and therefore the conversion efficiency. The ZnMgO films were finally implemented into a Cu2O/ZnMgO type of solar cell, showing rectification characteristics and leading to measurable photovoltaic performance. 12