*3.3. Keyword Analysis*

In order to carry out the analysis of the keywords, the data were previously cured, eliminating irrelevant terms and unifying terms that allude to the same concept. This analysis allowed us to establish the true state of research in a topic, defining trends and tracing possible lines of work for the development of future research. Figure 6 is a cloud-word showing the 32 keywords that appear in more than 300 publications in this area. The size of each keyword represents the relative proportion of each term to the total number of words.

**Figure 6.** Cloud-word with the more representative keywords.

Among the 160 keywords that appear in more than 80 publications, many of them allude to water-related parameters such as pH, temperature, total organic carbon, or organic matter, and many others refer to water pollutant compounds such as phenols, sulfur compounds, aromatic compounds, or antibiotics.

On the other hand, a large number of articles appear around keywords related to three water types—industrial, urban, and drinking water—with 1357, 786, and 468 articles, respectively. Figure 7 shows the relative importance that each of the 20 most important countries, in terms of the number of publications, attaches to each of these types of water.

**Figure 7.** Relative importance given by each country to each type of water.

It can be observed that, on average, the greatest importance is given to industrial water with 50%, then urban water with 30%, and finally drinking water with approximately 20%. However, if we look in detail at the data for each country, there are countries such as the United States in which the main concern is focused on urban water. Research in Australia also places the greatest importance on urban water. In contrast, urban water has very little relative importance in the research of countries such as Poland, whose research focuses on the study of industrial water. Finally, special consideration should be given to the data obtained in Japan, where almost half of the articles focus on drinking water. This fact is especially important when looking at the year 2011, when the Great East Japan Earthquake occurred, leading to the Japanese Prime Minister declaring the state of nuclear emergency. From that moment on, the Ministry of Health, Labour and Welfare, Japan (MHLW) showed its great concern for the healthiness of tap water and defined a series of indications [68].

The analysis of the keywords by means of the detection of scientific communities allowed us to group the publications in clusters of functions of the keywords used. Figure 8 shows the 15 communities that appeared in the analysis carried out, identifying each cluster with a color. The size of each cluster refers to the importance of the keywords around which the cluster is built, and the thickness of the lines of union between two clusters refers to the number of interactions established between two different communities. In each cluster, there is a variable number of keywords, but it was detected that, in all cases, there are always keywords referring to three aspects related to the analysis we carried out: AOP treatment, wastewater, and targets. Thus, in Table 1, each cluster is identified with the color with which it appears in Figure 8 and with the set of keywords that appear grouped around the three aspects mentioned above.

**Figure 8.** Scientific communities grouped in clusters based on the analysis of keywords in publications on wastewater and advanced oxidation.



Table 1 presents 15 clusters identified from the keywords analyzed relating the AOP treatments (alone or combined with biological oxidation) with the type of wastewater and the target pollutant of which removal is intended. From this, several conclusions can be drawn. On the one hand, ozone-based treatments are mainly used for pharmaceutical removal in a wide range of water matrices, from drinking water (cluster 16) to hospital or olive mill wastewater (cluster 1) [69]. Fenton and assisted Fenton processes find application in complex media such as industrial wastewater from different industrial sectors to remove dyes (cluster 4 and cluster 7), COD, and several persistent organic pollutants (cluster 9 and 14). Among the radiation based AOPs, heterogeneous photocatalysis (TiO2) has been widely studied for industrial wastewater treatment (cluster 13), although hydrogen peroxide-based treatments such as photo-Fenton and H2O2/UVC gained attention for municipal wastewater to remove micropollutants (mainly pharmaceuticals) as well as water and wastewater disinfection (cluster 9 and 12).

Finally, if an analysis is carried out on the importance of keywords in time, two periods can be identified. The first period extends from 2005 to 2008 (Figure 9), while the second period extends from 2010 to 2016 (Figure 10).

**Figure 9.** First period of evolution in advanced oxidation for wastewater treatment (2005–2008).

**Figure 10.** Second period of evolution in advanced oxidation for wastewater treatment (2010–2016).

In the first period, in its beginning toward the year 2005, there were numerous articles that presented keywords such as ozone, hydrogen peroxide, or UV irradiation. Then, mid-term was when advanced oxidation processes, solar energy, or biodegradation stood out. Finally, these lost their leading role to keywords related to the processes of Fenton and microwave, photo-Fenton, drinking water, or pharmaceuticals. This trend was confirmed in the second period, with the Fenton gaining prominence as it progressed at the end of this period, highlighting, above all, solar photo-Fenton.
