3.1 The Basic Status of PG Study
3.1.1 The Annual Trends of PG Publications by Countries
Figure 2 and Fig. 3 respectively plot the annual trends of articles and patents of PG from 1990 to 2020. In general, the development of PG-related studies is uprising. In terms of total amount, the number of PG patents (2580) is far greater than that in the literatures (719). However, from the perspective of research time, although patents were few before 2006 (none before 1993), there has been a significant rapid development since then. As for the output of each country, it is noteworthy that China has a large proportion in both of articles and patents, especially after 2010. Studies have shown that annual output of PG in China has reached 5 hundred million tons(Zeng et al., 2021), which boosts China to put effort into PG research.
Table 1 and Table 2 show the details of the publications of the top ten countries, such as the total number of documents, total citations, citations per article, and records per capita. In addition, the h-index of each country was calculated to supplement the comprehensive judgment of literature quality. Although China has absolute advantages in total article numbers, total citations and h-index, it does not perform well in terms of average citations per publication and articles per capita. Other countries, such as Spain, India, Brazil, and the United States, have significant performance in indicators that reflect the quality of papers. These countries have made many contributions to the research on the properties and utilization of PG (Campos et al., 2017; Lambert et al., 2018; Luis Guerrero et al., 2020; Nayak et al., 2018). Generally speaking, the regions where PG academic papers are produced are distributed all over the world. In terms of the number of patents (as shown in Table 2), Asian countries such as China, Russia, Kazakhstan, and Belarus are more prominent. Studies have indicated that the main phosphate mine producers and phosphate fertilizers are located in the United States, the former Soviet Union, China, Africa and the Middle East (Tayibi et al., 2009). Therefore, the large production of PG has prompted these countries to accelerate the process of resource recovery through innovative technologies.
Table 1
Articles in the field of PG resource utilization by Country/Region
Rank
|
Country/Region
|
TA
|
TC
|
TC without self-citations
|
SP%
|
ACPP
|
ACPP without self-citations
|
H-index
|
Population
|
Records/Population (10− 8)
|
1
|
China
|
151
|
2084
|
1535
|
21.00%
|
13.80
|
10.17
|
24
|
1,397,715,000
|
11
|
2
|
Brazil
|
56
|
974
|
878
|
7.79%
|
17.39
|
15.68
|
18
|
211,049,527
|
27
|
3
|
USA
|
55
|
748
|
701
|
7.65%
|
13.60
|
12.75
|
17
|
328,239,523
|
17
|
4
|
Spain
|
54
|
1089
|
908
|
7.51%
|
20.17
|
16.81
|
21
|
47,076,781
|
115
|
5
|
Poland
|
45
|
449
|
391
|
6.26%
|
9.98
|
8.69
|
12
|
37,970,874
|
119
|
6
|
India
|
37
|
731
|
702
|
5.01%
|
18.94
|
18.17
|
16
|
1,366,417,750
|
3
|
7
|
Tunisia
|
31
|
297
|
262
|
4.31%
|
9.58
|
8.45
|
9
|
11,694,719
|
265
|
8
|
Russia
|
29
|
109
|
102
|
4.03%
|
3.76
|
3.52
|
5
|
144,373,535
|
20
|
9
|
Egypt
|
27
|
459
|
433
|
3.76%
|
17.00
|
16.04
|
12
|
100,388,073
|
27
|
10
|
Canada
|
24
|
426
|
392
|
3.34%
|
17.75
|
16.33
|
12
|
37,589,262
|
64
|
Note: TA = Total Articles; TC = Times cited in Web of Science Core Collection; SP%=share of publications (719); ACPP = Average citations per publication |
Data source of population: World Bank. https://data.worldbank.org/. |
Table 2
Patents in the field of PG resource utilization by Country/Region
Rank
|
Country/Region
|
TP
|
SP%
|
Population
|
Records/Population (10− 8)
|
1
|
China
|
2241
|
86.86%
|
1,397,715,000
|
160.33
|
2
|
Russia
|
175
|
6.78%
|
144,373,535
|
121.21
|
3
|
Kazakhstan
|
32
|
1.24%
|
18,513,930
|
172.84
|
4
|
WIPO
|
31
|
1.20%
|
--
|
--
|
5
|
Belarus
|
18
|
0.70%
|
9,466,856
|
190.14
|
6
|
Korea
|
15
|
0.58%
|
51,709,098
|
29.01
|
7
|
India
|
13
|
0.50%
|
1,366,417,750
|
0.95
|
8
|
Brazil
|
11
|
0.43%
|
211,049,527
|
5.21
|
9
|
Poland
|
9
|
0.35%
|
37,970,874
|
23.70
|
10
|
Uzbekistan
|
9
|
0.35%
|
33,580,650
|
26.80
|
Note: TP = Total Patents; SP%=share of patents (2580) |
3.1.2 The Subject Categories of PG Publications
Based on the subject classification of the ISI Journal Citation Report (JCR), Figure S1 and Figure S2 depict the top ten subjects of PG articles and patents, respectively. Environmental science & Ecology (254), Engineering (248), Material Science (144), are the top 3 disciplines of PG articles. Moreover, Chemistry (2422), Materials Science (1439), Polymer Science (918) are the top 3 subjects of PG patents. The distribution of subject categories suggested the high priority of environmental, chemical, constructional, and agricultural issues in PG field. In contrast, the subject area of the articles pays more attention to environmental studies, while the patents are more focused on chemical material issues.
3.1.3 The Distribution of Published Journals of PG articles
All these 719 articles were published in 260 journals. Table S1 shows the number of articles, journal impact factors, h-index, country of origin, number of citations, etc. of the top 10 journals. The journal with the highest number of publications is "Construction and Building Materials" (40 articles), accounting for 5.56% of the total, which is 1.38 times that of the second-ranked " Journal of Hazardous Materials" and 1.66 times that of the third-ranked "Journal of Environmental Radioactivity". It can be seen that these are the main carriers of literatures in PG field. In terms of the indicators that measure the quality of journals (impact factor, JCR Partition, h index, citations), half of the journals are located in the first zone of JCR, indicating that the quality of the articles is relatively reliable.
3.1.4 The Distribution of Institutes of PG Publications
The top 20 institutions that made the majority of contributions to the total outputs of PG publications are presented in Table S2 and Table S3. Among the top 20 institutions in terms of articles, Universidad de Huelva ranks first, with 41 publications, accounting for 5.7% of the total. At the second position is Kunming University of Science & Technology with 32 articles followed by University of Sevilla (28). Moreover, although Council of Scientific & Industrial Research (India), Instituto de Pesquisas Energeticas e Nucleares (Brazil) and University of Alberta (Canada) do not rank high in the gross, they perform significantly in the indicators such as ACPP without self-citations and H-index.
Table S3 shows the top 20 assignees, types and countries with patents of PG. Guizhou Kailin Group Co.Ltd. has the greatest number of patents with a total of 208, accounting for 8.06%. The other institutions with a high number of patents are Guizhou University (143) and Kunming University of Science and Technology (74). As for the type of institutions, enterprises and universities are the main forces in the research and development of PG patents. Amongst the top 20 assignees, seven are placed in Guizhou province, three in Beijing, and two in Hubei and Henan apiece, with significant regional difference due to the origin of PG is mainly produced in southwest China. Different from the distribution of articles, these institutions almost all come from China. This is because China has been increasing the investment in solid waste management to make significant progress in PG reutilization in recent years (Zhang et al., 2019).
3.2 The Cooperation Situation of PG Study
3.2.1 Social Network Analysis of Academic Research Groups
In order to distinctly demonstrate the cooperative relationship in academic research of PG, we selected the countries, institutions and authors with more than 5 articles successively, and then used Ucinet software to conduct co-occurrence analysis on social networks.
In Figure S3, each node represents a country, and the dot size represents the degree centrality, which reflects the influence and importance of the country in the network. The thickness of the line between nodes indicates the number of cooperation between countries. Table S4 lists the top ten countries by degree centrality and their corresponding number rankings. The degree values of France, USA, Tunisia, China and Spain are all more than 20, indicating that these countries own important positions in the academic research of PG. It is obvious from the thickness of the connection lines in Figure S3 that France and Tunisia, USA and China are two groups of countries that have more cooperation. France and Tunisia have cooperated in the fields of economy, trade, security, and science and technology. Moreover, coastal industries in the Gulf of Gabès in southeastern Tunisia caused hundreds of millions of tons of untreated PG to be discharged into the open sea, causing serious environmental problems (El Zrelli et al., 2018; Sinfort et al., 2019). USA and China are major world powers and have carried out a lot of collaboration in many scientific research areas. It is worth noting that Saudi Arabia and Uzbekistan are also among the top ten countries in Table S4, even if they rank low in the total number of articles. This phenomenon shows that countries with a small volume of publications may also play a role in international cooperation in specific areas.
Consistent with the method in the above, we analyze the institutional cooperation network of research institutions with more than 5 articles. As shown in Figure S4 and Table S5, Universidad de Huelva is the node with the largest degree of centrality, and has a strong cooperative relationship with University of Sevilla, University of Cadiz and Consejo Superior de Investigaciones Cientificas (CSIC). These institutions are located in Spain with a great number of articles, are the main research institutions for PG academic papers. Another obvious cooperating group comes from China, including Guizhou University, Chongqing University, Chinese Academy of Sciences and other institutions. In general, the organizations that cooperate with each other basically belong to the same country. It is not difficult to understand that due to factors such as geographical location and technological level, neighboring institutions have more opportunities to communicate.
Figure S5 and Table S6 respectively show the author's cooperation network and the degree centrality of the top ten authors. There are two main author cooperative groups and several sub-cooperative groups in Figure S5. Similar to the content in Table S5, these authors are from Spain and China. Ranked first is Ma Liping from China, whose degree centrality is as high as 71, accounting for 7.87%, and is the author with the most academic output of PG. Authors ranked second and third place are Yang Jing from China (with 44 degrees, accounting for 4.88%) and Perez-Lopez, Rafael from Spain (with 33 degrees, accounting for 3.66%), respectively. These top-ranked authors led two major collaboration networks, namely the Chinese group and the Spanish group. In addition, there are some scattered uncomplicated cooperation networks in the map, which shows that the academic research of PG is extensive at the individual level, and the cooperation between individuals is close in a narrow range.
3.2.2 Social Network Analysis of Applied Research Group
In order to further explore the cooperation of PG application technologies, we also use ucinet software to plot the cooperation of the assignees sand the inventors of the patents in Figure S6 and Figure S7 severally. There are only a small number of institutions in the assignees cooperation network, as shown in Figure S6. Guizhou Guifu Ecological Fertilizer Co. Ltd., Guizhou Institute of Technology and Guizhou Kexin Chem & Metallurgy Co. Ltd. are the top three assignees in Table S10, with 38 degrees each, accounting for 24.84%. These three institutions constitute the largest group in the cooperation network. In addition, there are two pairs of partnerships in Figure S6, but compared to the largest, these assignees have less degree centrality. Different from the publication of articles, the special feature of patents lies in their exclusiveness and self-protection. Therefore, it is rare that a patent belongs to two institutions at the same time. However, in actual technology research and development, it is necessary to strengthen exchanges between various institutions, even if they cannot be signed in the same patent.
Figure S7 shows the cooperation network of inventors, which is obviously much more complicated than the map of assignees (Figure S6). The top ten inventors all have a high degree centrality greater than 100, which proves that there is a strong communication and connection between inventors. This suggests that many of the PG patents involve multiple inventors, that is, the collaboration between patents is concentrated among inventors within the same institution. Compared with the author's cooperation network of the articles (discussed in Sect. 3.2.1), the inventors' communication is concentrated in China, which is an evidence level shows that China plays a significant role in academic research of PG from the individual.
3.3 The research area of PG Study
3.3.1 Basic knowledge structure of PG
When two articles are cited by another article simultaneously, the two articles constitute a co-citation relationship. Co-citation analysis of references is an important method to comprehend the structure, dynamics, and paradigm developments of a given research field (Zupic and Cater, 2015). Figure S8 shows the reference co-citation map of PG literatures with 449 nodes and 1036 links drawn by Citespace software. Each node in the graph represents a cited document, and the size of the node is proportional to the total citation frequency of related documents. The spectra of different colors show the chronological order of the symbiotic relationship between the cited documents: blue indicates the oldest literature, and red indicates the latest. Table S9 lists the top ten articles with co-cited frequency and other details, such as centrality, first author, and publication year. The prominent contributions of the top five co-cited articles in the field of PG were analyzed. The most frequently cited is a study on the radioactivity of PG published by Tayibi, H in 2011. Tayibi (Tayibi et al., 2011) studied the stabilization/solidification process of PG as a building material from the perspective of radioactivity, and obtained a building material with lower radioactivity by mixing PG with a polymeric sulpfur matrix. The second most frequently cited article is researched by Rashad, AM and published in 2015 (Rashad, 2015). The results showed that using a part of PG to replace the fly ash in the alkali-activated fly ash paste can increase the compressive strength and thermal shock resistance of the material. The third study from Yang JK used low autoclave to pretreat the PG, and prepared a kind of load-bearing wall brick that meets the Chinese standard brick size and can replace the traditional fired clay brick (Yang et al., 2009). The fourth place comes from Hentati O’s research focused on the effect of PG as a fertilizer on soil biological toxicity (Hentati et al., 2015). The next article from Perez-Lopez, R is to explore the environmental impact of PG piles from the perspective of geochemistry (Perez-Lopez et al., 2016). Generally, the references with the centrality value greater than or equal to 0.1 in Citespace were considered as key publications (Chen, 2006). In Table S9, only the third-ranked Yang JK's article centrality meets this requirement, so it can be considered as a more turning point study. To sum up, these publications all have paid great attention to the application performance and environmental impact of PG or its products from various fields and perspectives and provide the knowledge bases in the field of PG academic research.
In order to further explore the specific knowledge areas involved in these articles, we used the built-in LLR algorithm of Citespace to cluster these articles with keywords. After discriminating the correlation between the labels obtained by clustering and PG research, we screened 9 clusters and displayed their details in Figure S9 and Table S10. Size represents the number of literatures contained in each cluster, namely the measurement of scales. Silhouette score, an indicator of value of homogeneity to evaluate the clusters, is shown in Table S10. These values are generally greater than 0.8, indicating that the quality of each clustering result is reliable. Mean cited year denotes the average citation time of the documents in the corresponding cluster, which can reflect the time this category has been paid attention to by academia.
It is apparent from Figure S9 and Table S10 that “natural radioactivity”, “calcium sulfate” and “soil” are traditional clusters formed before 2010. Generally, PG has a high content of natural radionuclides, especially the decay series of U 238 and Ra 226. Due to these characteristics, PG is considered to be Technologically Enhanced Naturally Occurring Radioactive Materials (TENORM), which has greatly restricted its resource utilization (Moreira et al., 2018). As an inorganic powder material, calcium sulfate whisker (CSW) which has excellent mechanical properties, environmental protection and low price is used as high value-added products of PG. The preparation process of calcium sulfate whiskers mainly focuses on hydrothermal method and atmospheric acidification method, both of which can produce products with high aspect ratio and high purity(Sheng et al., 2018). Besides, PG can exhibit a great deal of advantages as soil amendment in agriculture, such as improving soil physical and chemical properties, supplementing beneficial trace elements, and promoting crop growth. However, the heavy metals and natural radionuclides in PG have prompted studies to pay more attention to its agricultural environmental risks(Wang, 2020). In addition, information of other clusters shows that the basic academic areas of PG are concentrated in cement backfill materials (cluster 3), the preparation of supersulfated cement (cluster 4) and the extraction of rare earth elements (cluster 12). Furthermore, the crystal morphology of PG (cluster 11) has also received a plenty of attention (Jin et al., 2020; Lu et al., 2019; Ru et al., 2012). It is worth noting that cluster 10 "carbon dioxide" and cluster 15 "syngas" are unfamiliar research areas. Studies have shown that the products of PG at high temperatures can capture carbon dioxide (Cardenas-Escudero et al., 2011; Msila et al., 2016; Zhang, W. et al., 2020). This method of synergistic utilization of solid waste resources and carbon emission reduction provides a new perspective for the recovery of PG. Moreover,some scholars use chemical-looping combustion to prepare syngas by using lignite as a carbon source, PG as an oxygen source, and water vapor as a hydrogen source, which makes full use of the calcium sulfate to realize the utilization of PG (Yang et al., 2017; Yang et al., 2018; Yang et al., 2019).
3.3.2 Hot areas of PG research
Figure S10 and Figure S11 respectively plot the co-occurrence network map of PG literature keywords and Derwent Manual Code (DMC) from 2018 to 2020 to reflect the hot areas of PG research. The details of the top 20 keywords and DMC in terms of frequency and centrality are summarized in Table S11 and Table S12.
Obviously, the keywords “phosphogypsum”, “waste”, “gypsum”, “heavy metal”, “rare earth element”, “hydration”, “water”, “cement”, “soil”, “fly ash” are prominent in the frequency rankings. Centrality values present that “phosphate”, “compressive strength”, “aqueous solution”, “kinetics”, “removal”, “waste valorization”, “sewage sludge”, “mortar”, “sulfate” and “set retarder” stand out in the rankings. With these information provided in the figures and tables, we divide the academic hotspot research of PG into the following aspects. At the microstructure level, the mechanism of hydration in the chemical process related to PG, the change of crystal structure, and the physical and chemical properties of the product (compressive strength, water resistance, durability etc.) have attracted attention (Gijbels et al., 2019; Haque et al., 2020; Pinto et al., 2020). These studies can fundamentally clarify the characteristics of PG, so as to better promote its resource utilization. On the other hand, due to the peculiarity of PG containing heavy metals and radioactive materials, its environmental impact and safety management of the stacking yard are still continuously becoming academic hotspots (Attallah et al., 2019; Romero-Hermida et al., 2020; Tsioka and Voudrias, 2020). Furthermore, the utilization of PG focuses on the extraction of rare earth elements(Jalali et al., 2020), high compressive strength cement (Huang et al., 2019), the preparation of cement retarders and cement mortar(Gong et al., 2020), sludge dewatering (Dai et al., 2018), and co-production with other solid waste(Wang et al., 2020). Compared with the previous section (3.3.1), although there are some overlaps between hot areas and knowledge base, the results of keyword co-occurrence in the past three years provide us with more particular perspectives of the research directions about PG.
In the aspect of DMC frequency, L02-C05 (Calcium sulphate cements), L02-A03 (Refractories, ceramics, cement manufacture - shaping, drying) and A12-R01A (Concrete, cement, gypsum, mortar compositions and boards) rank in the top three. As for centrality, A12-R01A (concrete, cement, gypsum, mortar compositions and boards), L02-D01 (mortars and plasters) A03-A04A (cellulose ethers uses) etc. own the high rankings. In general, the result of DMC co-occurrence believes that the patent hot applications of PG are concentrated in the fields of construction, chemical industry and agriculture. To be specific, refractory materials, ceramics, cement manufacturing, cement additives, surface materials, and cement mortar are the concentrated hot areas of PG patents in construction (Liu et al.; Wang et al., 2020; Xu et al.). In terms of chemical material innovation, in addition to the traditional production of inorganic phosphorus compounds and sulfur compounds, the preparation of composite materials by adding acrylate/cellulose ether has also become a hot topic in PG patents (Bu; Gao; Zhou and Shen). Agricultural DMCs, such as C04-D02, C14-T, C04-A10, etc., also account for a part, mainly focused on preparing compound fertilizers and saline soil amendments to promote plant growth (Chen et al.; Huang et al.).
3.3.3 Frontier topics of PG research
Burst keywords represent the words that are cited continually during a certain period and are considered to be indicators of the frontier topics over time. As shown in Table S13 and Table S14, we use Citespace software to detect burst keywords and DMCs from 2018 to 2020 to indicate the frontiers of PG. In particular, we place emphasis on the keywords and DMCs that with longer duration and greater burst strength.
As can be seen from the Table S13, the keywords “recovery”, “coal”, “tailing”, “stabilization/solidification” and “temperature” possess higher strengths between 2019 and 2020. Masmoudi and Salo innovated the method of recovering rare earth elements from PG (Masmoudi-Soussi et al., 2020; Salo et al., 2020). Due to the complex process and high cost of extracting rare earths from PG, the industrialization of such methods is still facing challenges. Low-cost and high-efficiency rare earth recovery methods still need to be developed. Some scholars have also optimized the methods and conditions of ion solidification/stabilization in PG, which is conducive to the reprocessing and recycling of PG (Li et al., 2019). Temperature, namely reaction conditions, plays a pivotal role in the formation of PG products. Exploring the reaction process and mechanism with different temperature conditions will also be the frontier direction of PG academic research (Lu et al.). Besides, the keywords “mechanism”, “durability” and “behavior” also indicates that the research on the physical and chemical properties of substances at the micro level is still the focus of PG in the future (Li et al., 2019; Liu et al., 2020).
Likewise, DMCs such as L02-A08 (Refractories, ceramics, cement - manufacturing methods, equipment - Testing/Control), A12-R07 (Building, civil engineering - Walls, wall coverings and ceilings), L02-D13 (Mortars, concretes – aggregates), A10-E05B (Chemical modification by carbonization) etc. occupy greater burst strengths during 2019–2020. As patents pay more attention to practicability, part of the development trend of PG patented technology is the improvement of equipment manufacturing, such as innovations in crushing sorting devices and calcining and drying devices (Zhou et al.; Zhu). In the preparation of building materials, the innovative wall materials (gypsum plasterboard, gypsum block, cement concrete, silicate concrete, etc.) not only reduce the use of high-energy clay brick to cut down the production costs, but also can solve the problem of large accumulation of PG (Gong and Shi; Jiang et al.; Yin et al.; Zhou et al.). There are also some DMCs (A10-E21A, A10-E12A, A10-E05B) in the table indicating chemical modification of various methods, since the preparation of PG products involves many chemical processes, it is more likely to prepare innovative polymer materials (Shen et al.; Su et al.).
3.4 Prediction of research trend of PG
In order to quantitatively and accurately predict the growth trend of documents in various fields of PG, we merged articles with similar meanings of subject categories (such as classifying architecture, construction & building technology, and engineering as construction engineering), and summarized them as four categories, that are, constructional engineering, chemical materials, environmental science and agricultural science. The technical life cycle curve of the PG articles, the S-curves, drawn by the Loglet Lab4 software, where the blue dots in the figure represents the actual cumulative number of PG articles each year, and the blue line is the fitting curve between the actual and the predicted numbers, as shown in Figure S12. The specific parameters and growth stage details are shown in Table S15 and Table S16. Among the four categories, “Constructional Engineering” has the highest Saturation value (324), which means it will owe a broad development space. On the contrary, the Saturation value of “Agricultural Science” is only 111, which means that the future of PG academic papers in agriculture will not increase drastically. The growth time value and midpoint value can be used to divide the technology life cycle stages, and the results are summarized in Table 18. From the perspective of the timing of the stages, the four types of technology categories are currently in the maturity. The three major categories of “Constructional Engineering”, “Chemical Materials” and “Environmental Science” will enter the decline stage in 2033, 2031, and 2033, respectively. What stands out in Table 18 is the forthcoming decline of the category of “Agricultural Science”, approximately after 2025. To sum up, both in terms of the number of articles and the stage time, “Construction Engineering” has the most application prospects, followed by “Chemical Materials” and “Environmental Science”, but there is not much room for the development of “Agricultural Science”.
Similarly, we also counted the annual distribution of the top four categories of the DMCs in the patents, namely L (Refractories, Glass, Ceramics), A (Polymers, Plastics), C (Agricultural Chemicals) and E (General Chemicals). Figure S13 plots the S-curve of each technical field of the patents. The specific results are shown in Table S17 and Table S18. It can be seen from the value of Saturation that L-category will have the largest number of accumulated patents, followed by A-category, C-category and E-category. While from the value of Growth time, it can be judged that L-category has the longest development time, while A-category has the shortest. In terms of the stage of the technology life cycle (Table S18), it is considered that L-category is in the development period, A-category is in the decline period, and C-category and E-category are in the mature period. Taken together, the future prospect of the distribution of PG patents in L (Refractories, Glass, Ceramics) is the largest and the development time is still very long. C (Agricultural Chemicals) and E (General Chemicals) are basically the same, but there is almost no more space in the A (Polymers, Plastics) category to blossom.