Venkatraman et al. J Cheminform (2018) 10:18 https://doi.org/10.1186/s13321-018-0272-0 DATABASE Open Access The dye‑sensitized solar cell database Vishwesh Venkatraman*, Rajesh Raju, Solon P. Oikonomopoulos and Bjørn K. Alsberg Abstract Background: Dye-sensitized solar cells (DSSCs) have garnered a lot of attention in recent years. The solar energy to power conversion efciency of a DSSC is infuenced by various components of the cell such as the dye, electrolyte, electrodes and additives among others leading to varying experimental confgurations. A large number of metal- based and metal-free dye sensitizers have now been reported and tools using such data to indicate new directions for design and development are on the rise. Description: DSSCDB, the frst of its kind dye-sensitized solar cell database, aims to provide users with up-to-date information from publications on the molecular structures of the dyes, experimental details and reported measure- ments (efciencies and spectral properties) and thereby facilitate a comprehensive and critical evaluation of the data. Currently, the DSSCDB contains over 4000 experimental observations spanning multiple dye classes such as tripheny- lamines, carbazoles, coumarins, phenothiazines, ruthenium and porphyrins. Conclusion: The DSSCDB ofers a web-based, comprehensive source of property data for dye sensitized solar cells. Access to the database is available through the following URL: www.dyedb.com. Keywords: Dye sensitized solar cells, Database Background electrons from the dye difuse (fow of current) through With renewable energy-based systems gaining momen- the semiconductor and move on to the back collecting tum, there has been intense focus on harnessing energy electrode. Dye regeneration takes place through electron from the sun while making it useable and cost-efective. donation from the electrolyte aided by the catalyst in the Solar cell technologies that convert sunlight into elec- counter electrode. Te modular architecture of the DSSC tricity include those based on silicon, polymer, quan- thus enables functions such as electron transport, light tum dots, dye-sensitized and more recently, perovskites. absorption and hole transport to be handled separately Among these dye-sensitized solar cells (DSSCs) also [1, 2]. known as Grätzel cells have received a lot of attention [1, Although DSSC efciencies have been improving, the 2]. In recent years, this feld has seen a dramatic increase pace of these improvements has been somewhat slow, ∼ 6% in published research. An ISI Web of Knowledge search with an increase of only [3] from a value of 7.1% for the term “dye-sensitized solar cells” yielded more than in 1991 [4]. Te variations in the power conversion ef- 18,000 articles spanning years 1991–2017 (see Fig. 1) with ciencies (PCE) for diferent DSSCs can be attributed to a signifcant proportion published in the last 5–10 years. changes in the cell architecture and fabrication [5–8]. Te DSSC typically consists of a monolayer of a pho- While a signifcant amount of these eforts have been tosensitive dye that is adsorbed on a mesoporous oxide devoted to molecular engineering of the dye sensitizer layer (such as TiO2, ZnO or SnO2) that is deposited on [9–13], others have focused on the optimization of the a transparent conductive glass substrate, a redox elec- electrodes [14, 15] and electrolytes [16–18] along with trolyte (iodide or cobalt-based) and a platinized counter factors such as the concentrations of the solvent baths electrode. On excitation (absorption of incoming light), during sensitization [19], and the size and thickness of photoanodes [8, 20, 21]. Te DSSC efciency is infuenced by a number of com- *Correspondence: [email protected] Department of Chemistry, NTNU, Høgskoleringen, 7491 Trondheim, ponents/parameters. Other than the dye one can add a Norway cosensitizer to account for the higher wavelength regions, © The Author(s) 2018. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/ publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Venkatraman et al. J Cheminform (2018) 10:18 Page 2 of 9 2500 2000 1500 # Publications 1000 500 0 19911993199419951996199719981999200020012002200320042005200620072008200920102011201220132014201520162017 Fig. 1 Literature growth of dye sensitized solar cells. The graph was produced by performing a Web of Science search for the keywords “dye sensitized solar cells” and restricting the search to articles in English. In addition subject areas such as mechanics, toxicology, pharmacology and educational research were also excluded to yield around 18445 records use agents like chenodeoxycholic acid (CDCA) to reduce demand for methodologies that can accelerate the design aggregation and also change the electrolyte from iodide of molecular materials with tailored properties, chemin- to cobalt which has often resulted in increased PCEs. formatics (or materials informatics) based frameworks Other factors such as the dye bath and concentrations for high-throughput screening of candidate structures also have an impact on the PCEs. Modifcations to the have been proposed: dyes [25], solid state metal oxide structure of the dye sensitizer in particular have been photovoltaic cells [26] and organic photovoltaics [27, 28]. found to be the most widely applied method to improve With a view to understanding how structural/chemical device efciencies. Given that, by introducing system- modifcations impact the solar cell performances, recent atic variations of the substituent groups in the dye can eforts have focused on creating quantitative structure improve the light harvesting and electron injection capa- property relationships [26, 29–36] that establish a mathe- bilities among other properties, various classes of dyes matical relationship between various molecular structure (metal-free and inorganic-based) have been investigated descriptors and a solar cell property of interest such as ranging from coumarins, carbazoles, indolines, tripheny- the PCE. Te models produced in this process have been lamines [10], phenothiazines [9], fulvalenes [22] to ruthe- further used to direct the search for promising dyes/ nium [23] and porphyrin-based [24] sensitizers. photvoltaic materials that satisfy desirable criteria [26, Dye materials discovery has been largely based on ser- 37–39]. Informatics approaches have also been recently endipity or iterative chemical substitution. Given the Venkatraman et al. J Cheminform (2018) 10:18 Page 3 of 9 applied to the identifcation of suitable photocathode 14. Dye class: to indicate the type of the donors or spe- materials [40] and solid state electrolytes [41]. cifc chemical groups in order to enable a keyword- Recently, a number of data repositories such as the based search. Materials Project [42], Khazana [43], the Harvard Organic Photovoltaic Dataset [44], and the Open Spec- Te database is centred around 4 main tables (see tral Database [45] have emerged that facilitate the dis- Fig. 2) refecting the aforementioned details. During the covery of qualitative/quantitative rules, which can be data collection, articles without a valid DOI and those used to guide materials design. Here, we report the Dye with incomplete performance data were excluded. Te Sensitized Solar Cell Database (DSSCDB) consisting of 2D structures of the dyes were drawn using various experimental results compiled from the literature. Te molecular drawing software. For cases where the chemi- database is intended as a central repository for sharing cal names were available, the SMILES formats were gen- photovoltaic performance related data and should be of erated using OPSIN [46], failing which the structures broad interest to scientists in photovoltaics, quantum were drawn by hand. Corresponding InChi keys were chemistry, chemometrics and related disciplines. Search then generated using OpenBabel [47, 48]. Images of the tools have been implemented with both text and struc- structures have been generated using the Indigo Toolkit ture-based functionalities. [49]. Te web interface has been designed using the Django 1.10 MVC framework (https://www.djangopro- Description and utility ject.com) and connected to a PostgresSQL [50] database Information regarding the dyes was manually retrieved and hosted on the Amazon Cloud Platform. Te package from journal articles obtained using keyword (“dye sen- manager Conda 4.3.6 (https://www.conda.io) was used sitized solar cells”, “triphenylamines” etc.) searches on the to include RD-Kit 2017.03.1 [51] and PyBel [52] support. ISI Web of Knowledge. For each dye, the following data Te Docker platform (https://www.docker.com/) was has been recorded: additionally used to facilitate continuous development and ease of deployment. 1. DOI: the digital object identifer for the referenced article Search and retrieval Voc 2. Performance parameters: open circuit voltage2 ( in Te entire database can downloaded as a csv fle which mV), short circuit current (Jsc in mA/cm ), fll factor contains SMILES, InChi, performance data, experimen- (FF), power conversion efciency (PCE)