This paper presents a Named Entity Recognition (NER) approach to extract key information related to the structural components, synthesis techniques, and photovoltaic performance metrics of Perovskite solar cells (PSCs). Since NER is a key component of Information Extraction (IE), which identifies and classifies key elements of a text, we propose the use of the SpaCy library to build an effective and accessible NER model, suitable for environments with limited computational capacity, unlike other previous works in this field, which make use of large-scale models or high computational resources. Our resulting model was evaluated using K-fold cross-validation, obtaining the mean scores of, precision of 89.94%, a recall of 92.47%, and an F1 score of 89.69%. To provide a test of the practical performance of the resulting model, implementing and comparing the obtained results with manual annotations in two Excel reference databases: Odabaşı (2019) [1] and Jacobsson (2022) [2], demonstrating the potential of our work to facilitate and accelerate knowledge extraction, and the possibility of extending this strategy to other scientific fields where automated text extraction is required.