Knowledge domain and hotspot trends in coal and gas outburst: a scientometric review based on CiteSpace analysis

Coal and gas outburst is one of the main dynamic disasters in coal mine production. In order to deeply understand the overall research progress in the field of coal and gas outburst, the Web of Science (WOS) database is used as the sample source, and the bibliometric analysis of the literature in the field of coal and gas outburst from 2000 to 2021 is carried out by CiteSpace software. The knowledge maps of the publication quantity, country, institution, and keyword cluster are drawn. The results show that the time distribution has gone through three stages of germination period, development period, and growth period. In terms of the cooperation network, China, Australia, and the USA are the main core research countries, while China University of Mining and Technology, Chongqing University, and Henan Polytechnic University are the main core research institutions, and the main core journals are IGCG, IGRMM, and Fuel. By drawing the knowledge map of keywords timeline, the evolution law of keywords in the field of coal and gas outburst with time is obtained. Taking keywords as the object, cluster labels with gas outburst, gas extraction, gas control, and pore structure as the core are presented. On this basis, from a new perspective, this paper outlines the “scientific research landscape map” in the field of coal and gas outburst, in order to provide a scientific reference for coal and gas outburst research.


Introduction
As the depth of coal mining increases year by year, coal production has gradually entered the stage of deep mining. With the high gas pressure, high gas content, and low permeability of coal seams, gas extraction becomes more difficult, and the risk of coal and gas outburst increases, threatening mine production safety (Cheng and Pan 2020;Fan et al. 2020;Yuan 2016). Therefore, strengthening the effect of coal seam gas (CSG) extraction and improving the level of gas control are important measures and means to reduce the occurrence of coal and gas outburst disasters, which provide strong security for the life and health of underground workers and the normal operation of equipment and facilities.
Meanwhile, the main component of CSG is methane, which is equivalent to the calorific value of natural gas. It is an important unconventional natural gas and can be used as an effective supplement to the national energy structure. According to statistics (Liu et al. 2020a, b), China's current CSG resources shallower than 2000 m are approximately 31.46 Tm 3 , which is equivalent to the total onshore conventional natural gas resources (30 Tm 3 ), of which the recoverable resources are 10.87 Tm 3 . CSG is not only a high-quality, green, clean, and efficient energy source, but it is also the primary source of greenhouse gas emissions, with tens of times the greenhouse effect of carbon dioxide Moore 2012). Therefore, strengthening the effective extraction of CSG is also of great significance to alleviate energy shortages and reduce greenhouse gas emissions.
How to prevent the occurrence of coal and gas outburst is an urgent problem to be solved at present. In this regard, many experts and scholars have carried out a lot of research through theoretical derivation, laboratory experiments, numerical simulation, and field investigation and have achieved fruitful results. Sobczyk (2014) conducted seepage and desorption experiments on coal briquettes and different gasses in the laboratory and analyzed the influence of seepage and desorption processes on the initial conditions of coal and gas mini-outbursts. Wold et al. (2008) measured the gas outburst characteristics and gas pressure changes of outburst-prone coal by using an on-site measurement method, which provides guidance for the division of outburst-prone areas. Wang et al. (2013aWang et al. ( , b, c, 2014 analyzed the stress distribution characteristics of shaft working faces in deep underground coal mines by theoretical analysis and field tests and determined the dominant control factors of gas occurrence and gas dynamic disasters by statistical analysis. Jiang et al. (2011) compared the unaltered and heat-affected coals in order to study the relationship between sill intrusions on coal seams and gas outburst. Wang and Du (2020) summarized and analyzed the occurrence laws and failure characteristics of coal-gas compound dynamic disasters in the past 30 years. Lin et al. (2015), Zou et al. (2014), and Zou et al. (2015) proposed the use of high-pressure water jet slotting technology to increase the permeability of the coal seam, improve gas drainage efficiency, and prevent the occurrence of coal and gas outburst disasters.
Bibliometrics uses mathematical and computer statistical methods to analyze the evolution of disciplines and hot spots and then predicts the development direction of disciplines using the publication quantity, author, institution, cited reference, temporal evolution of keywords, and burstness feature as the object (Chen, 2006;Chen and Song 2019;Fan et al. 2018). In recent years, driven by advanced technologies such as big data, cloud computing, and data visualization, scientific knowledge map analysis technology has been rapidly developed and widely used in many disciplines .
Based on the above analysis, many scholars have carried out a large amount of theoretical and experimental research, which has made crucial contributions to the prevention and control of coal and gas outburst. However, previous research mainly focuses on a scientific problem and adopts corresponding mechanism analysis and program measures, namely the use of point-to-point micro level, and few scholars start from the macro level, combing the evolution process of coal and gas outburst over time and then providing help for people who have just contacted the research field to quickly understand the general development status of this field.
Therefore, this paper uses the CiteSpace knowledge map software to visually analyze the literature in the field of coal and gas outburst based on the WOS core database and draws knowledge maps such as publication quantity, country, institution, and keyword analysis. By means of literature research, bibliometrics, and knowledge maps, the frontier hotspots and trend evolution are excavated, and the "scientific research landscape map" in the field of coal and gas outburst is outlined from a new perspective, in order to provide scientific reference for the research of coal and gas outburst.

Data source
The WOS core database was used as the literature retrieval platform for this paper, and the data was retrieved and downloaded. To begin with, the retrieval subject was "coal and gas outburst," and the time span was "January 1, 2000-June 3, 2022," and the literature type was set as "article" in the advanced search to retrieve 1078 journal papers. The retrieved data is then downloaded and imported into CiteSpace software via deduplication processing, yielding 1078 valid papers.

Analysis method
CiteSpace software is a scientific measurement visualization and analysis software developed by Professor Chen C. M. of Drexel University based on Java. After years of development and popularization, it has become one of the most reliable software products in the field of scientific metrology visualization (Shao et al. 2022). The software can realize the visualization functions of author cooperation networks, keyword co-occurrence, organization distribution, literature integration, and so on. Through the content analysis, induction, and summary of the knowledge map display, it provides guidance for further exploring the development trend of discipline frontier and knowledge association.
The node size reflects the number of publications; the larger the node, the greater the number of country or institution publications. The link between the two nodes represents a cooperative relationship, and the thicker the link, the closer the cooperative relationship. The greater the centrality of the node, the greater its influence (Jiang and Fan 2022).
In order to carry out an in-depth analysis of the coal and gas outburst field, its current characteristics and development trend are analyzed. In this paper, CiteSpace 6.1.1.R2 visualization software is used to draw the knowledge map, including country, institution, and keyword clustering and its evolution. Combined with the classic literature, the knowledge map of coal and gas outburst in 2000-2021 is analyzed and summarized.

Trend analysis of annual publication
It is based on a quantitative analysis of WOS data, the annual publication quantity change trend, and literature citations, all of which reflect research activity in the field of coal and gas outburst. Figure 1 shows the trend of annual publication quantity and citation of literature in the field of coal and gas outburst from 2000 to 2021. From the results of the publication trend, the research in the field of coal and gas outburst has gone through three stages: the germination period, the slow development period, and the rapid growth period.
(1) In the germination period (before 2000), the annual average publication quantity in the field is less than two, which has not been widely concerned by researchers. At this stage, the coal seam mining depth is shallow, the gas content is low, and the number of coal and gas outburst accidents is small. At the same time, the understanding of coal mine safety production is still insufficient, and the occurrence of coal and gas outburst accidents is not taken seriously.
(2) In the development period (2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009), the annual average publication quantity increased from 2 in 2000 to 10 in 2009, with an average of 4.8 per year. At this stage, with the increase of mining depth, coal and gas outburst accidents occur from time to time, resulting in certain casualties and equipment damage. Therefore, during this period, people began to pay more attention to coal and gas accidents. However, due to the lack of deep research on the mechanism of coal and gas outburst, the immature development of technical equipment and the insufficient awareness of safety produc-tion, the rapid development of coal and gas outburst is restricted.
(3) In the growing period (2010)(2011)(2012)(2013)(2014)(2015)(2016)(2017)(2018)(2019)(2020)(2021), the number of articles published in journals increased from 21 in 2010 to 196 in 2021, with an average annual increase of 78.6. At this stage, with the development of advanced technology and the improvement of safety production consciousness, many researchers have carried out a lot of research work on the theoretical analysis and physical experiments in the field of coal and gas outburst, which provides strong technical support and theoretical basis for preventing accidents and ensuring the life and health of personnel.
From the statistics of the citation frequency of papers published from 2000 to 2021, the trend of citation frequency is basically consistent with the publication quantity; that is, citation frequency from the germination period through the slow development period and eventually into the rapid growth period. The citation frequency increased from the initial 12 times (2000) to 107 times (2009) and finally increased rapidly to 4148 times (2021), which increased 8.9 times and 345.7 times, respectively. Especially after 2010, the publication quantity and citation frequency both showed an exponential growth trend, reflecting that this stage is the peak period of coal and gas outburst research, which opened a golden age of rapid development in this field and attracted extensive attention from researchers. At the same time, it can be predicted that this topic will still be a hot spot and trend of research in the future, according to the changes in publication quantity and citation frequency.

Country analysis
The CiteSpace software is used to summarize and collate the imported literature data, and the publication information of the major research countries is obtained, and the global distribution map of the countries is drawn, as shown in Fig. 2. The main countries that carry out coal and gas outburst research in the world include China, Australia, the USA, Poland, and Russia. All of which are important coalcoalbed methane-producing countries; among them, China is an important coal production and consumption country, and it is also one of the countries with the most serious coal and gas outburst disasters. Therefore, in-depth research on coal and gas outburst is of great significance to ensure the safety of personnel, social harmony and stability, and the steady development of the economy.
The node wheel size represents the publication quantity, and the link between nodes indicates a cooperative relationship. The larger the node round, the larger the publication quantity. The nodes of the purple outer ring represent its high centrality, reflecting its strong importance, contribution, and Fig. 1 The trend analysis of annual publication quantity in the field of coal and gas outburst influence in the research field. From a macro perspective, the node type is selected as "country," and the knowledge map is visualized to draw the knowledge map of the cooperation network between countries, as shown in Fig. 3.
There are 33 nodes and 60 links on the map, and the network density is 0.1136. In terms of publication quantity and centrality, China ranks first and takes a prominent advantage; Poland ranks below Australia and the USA in publication quantity, but exceeds them in centrality.
As shown in Table 1, the data information of the top 8 countries is extracted and summarized. In terms of publication quantity, there is only one country with more than 100   (910), and the global total proportion (GTP) is as high as 84.42%, followed by Australia and the USA, which have published 86 (GTP, 7.98%) and 58 (GTP, 5.38%), respectively.
In the process of influence analysis, if the centrality is greater than or equal to 0.1, it can be considered as having strong importance and influence. In addition to China and Poland, Ukraine and the UK have relatively strong centrality, reaching 0.13 and 0.20, respectively, although publication quantity is relatively small, which indicates that the two also have a strong influence.
Therefore, when analyzing the cooperation network, it is necessary to fully combine the two key indicators of publication quantity and centrality, so as to understand the research status of coal and gas outburst more comprehensively and objectively.

Institution analysis
The institution cooperation network can reflect the publication status and the degree of close cooperation among institutions in the research field. From a microscopic perspective, the node type is selected as "institution," and the knowledge map is visualized to draw the cooperation network between institutions, as shown in Fig. 4.
There are 329 nodes and 698 links on the map, and the network density is 0.0129. In terms of publication quantity, China University of Mining and Technology, Chongqing University, Henan Polytechnic University, China University of Mining and Technology (Beijing), and Anhui University of Science & Technology are among the top five institutions. In the meantime, the five nodes all show a purple outer ring with strong centrality. The high publication quantity and strong centrality show that these five institutions are important to core research institutions in the field of coal and gas outburst.
The data information of the top 10 research institutions is extracted and summarized (see Table 2). In terms of publication quantity, there are three institutions that have published more than 100 papers, namely China University of Mining and Technology (363), Chongqing University (150), and Henan Polytechnic University (124), with GTP of 33.67%,13.91%, and 11.50%, respectively. Meanwhile, among the top 10 institutions, nine institutions are from China and one institution is from Australia (University of Wollongong). In terms of centrality, the top 5 institutions are basically in the same order as publication quantity. Among them, Chongqing University and Henan Polytechnic University are both 0.26, with the same centrality, followed by China University of Mining and Technology (Beijing) and Anhui University of Science & Technology, with 0.16 and 0.14, respectively.
From the perspective of publication time, it is primarily concentrated after 2010, accounting for 80% of them, and its occurrence time is just in the rapid growth period of the above-mentioned trend of publication quantity. There are two main reasons for this phenomenon: First, the geological factor. With the development of deep mining in recent years, the characteristics of high crustal stress, high gas pressure, and low permeability increase the risk of geological disasters such as coal and gas outburst, which makes it important and necessary to carry out relevant theoretical research, laboratory tests, and field verifications, thereby providing scientific guidance for the prevention of accidents (Dutta et al. 2011;Jin et al. 2016a, b;Jin et al. 2018;Tu et al. 2016Tu et al. , 2019. Second, the technical factor. With the progress of technology and the development of key equipment, it is possible to carry out physical similarity simulation tests. The maturity of computed tomography (CT), scanning electron microscopy (SEM), nuclear magnetic resonance (NMR), and other technologies increase the diversity of research methods for microscopic pore structure (Ni et al. 2019(Ni et al. , 2018Nie and Li 2012;Nie et al. 2015;Pan et al. 2015).

Journal co-citation analysis
As the main carrier of paper publication, journals provide a platform for communication between scientific technology and theoretical ideas. The citation frequency of a journal represents its influence in this field. The higher the citation frequency, the greater the influence of the papers included in this field, and the higher the attention degree of researchers. In the co-citation map of journals, the larger the node, the greater the frequency of citation keywords. The color reflects the time of co-citation. The inner color of the node ring is dark-gray, indicating that the time of co-citation is relatively early. The outer color is orange-red, representing that the time of co-citation is relatively late.
The node type is selected as "cited journal," and the knowledge map is visualized to draw the journal co-citation network, as shown in Fig. 5. There are 820 nodes and 1940 links in the map, and the network density is 0.0058. The top 5 cited journals are International Journal of Coal Geology, International Journal of Rock Mechanics and Mining Sciences, Fuel, Journal of Natural Gas Science and Engineering, and International Journal of Mining Science and Technology. Meanwhile, the outer color of each node is orange-red, which indicates that the citation time is recent. It reflects that the journals are still active in the field of coal and gas outburst, and the articles published in the journals are widely concerned by current researchers.
The data infor mation from the top 10 jour nals with the most citations is extracted and summarized (see Table 3). In ter ms of publication quantity, there are four jour nals with a citation frequency of more than 500 times. The International Journal of Coal Geology ranks first, with 779. Six of the 10 most cited jour nals are from the Netherlands, accounting for 60% of the total; the remaining four are from the UK, Australia, China, and the USA, respectively.
H-index, also known as h-index or h-factor, is a new method for evaluating academic achievements. h represents "high citations." The h-index of a journal means that no more than h papers have been cited at least h times. The higher the h-index of a journal, the greater the influence of the journal. Although the citation frequency of Fuel ranks third with only 578 times, its h-index is as high as 229, ranking first, followed by Energy & Fuels, which ranks 10th in terms of citation frequency with 243, but has an h-index as high as 198.
Impact Factor (IF) is a data in the Journal Citation Reports (JCR) published by Thomson Reuters. The total number of citations in the previous two years is divided by the total number of citations in the previous two years. It is an important indicator for measuring the influence of academic journals. Generally speaking, the higher the IF, the greater the influence of the journal. The citation frequency of the International Journal of Rock Mechanics and Mining Sciences (664) is less than that of the International Journal of Coal Geology (779), but its IF (7.135) is greater than the latter (6.806).

Keyword co-occurrence analysis
In the keyword co-occurrence map, the larger the node label, the higher the word frequency of the corresponding keyword. The co-occurrence relationship between keywords is connected by a link, and its width reflects the co-occurrence strength between keywords. In order to further analyze the evolution process of each keyword in the field of coal and gas outburst over time, the keyword map type is selected as "Timeline," and the knowledge map is visualized to draw a keyword timeline network map, as shown in Fig. 6. There are 463 nodes and 2808 links on the map, and the network density is 0.0263. Burst keyword can occur suddenly in a certain year or field at different times. In order to  understand the development of burst keywords in a certain period, the red-highlighted way is used to mark, which is more conducive to in-depth analysis of the evolution of hotspots in the field of coal and gas outburst in various periods. It can be seen from the keyword timeline distribution map that high-frequency keywords and burst keywords are mainly concentrated between 2010 and 2015, and less after that. The reasons for the above phenomenon mainly include (1) the time factor. The time interval from 2015 to now is short, so the cited frequency is correspondingly low; (2) the technical factor. Some new concepts have not yet been widely recognized as key technologies and concepts are not yet mature. At the same time, the links between keywords before 2015 are relatively concentrated, while the links after 2015 are relatively dispersed, indicating that more branches have evolved based on the keywords that appear earlier, and the subsequent research is more detailed, microscopic, and in-depth.

Keyword cluster analysis
The keyword cluster is the use of the metrology method to classify and condense complex keywords and finally present them in the form of several or more nouns. This cluster method is more intuitive and understandable, which helps to quickly understand the important research content and its proportion in the research field. In the process of keyword clustering, the cluster type is selected as "keyword," and the main clustering labels are screened and presented in a visual knowledge map, as shown in Fig. 7.
There are 864 nodes and 3549 links, and the network density is 0.0095. By adjusting the size of the literature's burst strength, the information of authors and time of the article with burst strength is presented above the corresponding cluster label, which makes it easier to quickly understand the important research authors under the corresponding cluster.
The regional divisions among clusters are not completely independent, but are mutually distributed throughout the cluster map. The cluster #3, gas outburst, is located on the left side of the map, and the regional division is relatively independent, but these links between #3 and #0 and #8 in the middle part are numerous and dense, indicating that it is closely connected with the two. Meanwhile, the two clusters in the middle part are also densely connected with other clusters on the right side, and there are overlapping parts between the clusters. Thus, the two clusters in the middle part can serve as a bridge to connect the left and right clusters, making clusters of the whole map interrelated and organic.
From a macro perspective, cluster #3 belongs to the disaster and accident part in the field of coal and gas outburst. In order to prevent accidents and disasters, it is necessary to adopt technical measures for gas extraction and gas control, namely clusters #0 and #8 in the corresponding cluster. The theoretical and experimental research on isotherm adsorption characteristics, pore structure, gas expansion energy, similarity experiments, and evaluation methods are the embodiment of the above-mentioned gas extraction and gas control technical means, which correspond to the other cluster labels on the right. Through an in-depth analysis of the influence of various factors on the occurrence of coal and gas outburst, the corresponding technical methods are proposed. As a result, the cluster map is divided into three levels: (1) disaster and accident, cluster #3; (2) accident prevention, clusters #0 and #8; and (3) research method and means, other clusters.
Theoretical derivation, numerical simulation, laboratory experiment, and field test are important ways and methods to carry out scientific research (Xu et al. 2019a, b;Yang et al. 2019;Zhai et al. 2015;Zhao et al. 2016Zhao et al. , 2019Zou et al. 2020a, b). Therefore, it is also applicable to carry out research on coal and gas outburst. The characteristics of high risk and strong destructiveness make it difficult to carry out the field test of coal and gas outburst. The current field test is mainly obtained from the following two aspects: The first is the method of measuring, investigating, and counting on the relevant data and parameters in the process of accident investigation; the second is the means of sample collection and determination in the prone areas of coal and gas outburst mines.
In order to further analyze the development process and research content in the field of coal and gas outburst, it is classified according to the meaning expressed by keywords, and the results are shown in Table 4. Table 4 classifies the cluster labels in Fig. 7 and extracts the cluster size, silhouette, average year, and main keywords contained in each cluster label.
In the theme of disaster and accident, only one cluster label is included, cluster #3. The main contents of its research include coal-gas outburst, dynamic response, dynamic effect, and gassy coal extrusion. The theme of accident prevention includes clusters #0 and #8. Among them, cluster #0 contains the most keywords and the largest cluster size, up to 207. Meanwhile, as the cluster contains the most keywords, its silhouette is low, at only 0.750. Cluster #8 has the fewest keywords and the smallest cluster size (31); the main research topics include control technologies, mining effect control, and dynamic disaster control.
Most cluster labels are included in the theme of research method and means. It reflects the richness and diversity of methods and means to carry out coal and gas outburst research. The silhouette of the top three clusters is in the theme, ranking first at #5 (0.943), followed by #7 and #6, with 0.918 and 0.902, respectively.
In the average year of publication, the clusters are all after 2010, just in the rapid growth period of publication quantity, reflecting a high degree of consistency between the high occurrence of literature in the field of coal and gas outburst and the trend of publication. Between 2010 and 2021, both publication quantity and citation frequency showed an exponential growth trend; the strong burst of literature and the Gas extraction is an important method and means to prevent coal and gas outburst. Strengthening the effective extraction of CSG can not only avoid disasters and accidents, preventing casualties and property losses caused by accidents, but also reduce greenhouse gas emissions and help to achieve the goal of "carbon peak, carbon neutrality" on schedule Busch and Gensterblum 2011;Cai et al. 2013;Jin et al. 2016a, b). At the same time, CSG, as a high-quality unconventional natural gas, can be used as an important energy substitute. By improving the extraction effect of coal seam gas, the large-scale and commercial production of coal seam gas can be realized, thereby making up for the current dilemma of insufficient supply and demand of conventional fossil fuels such as natural gas and oil in the world.
Through a series of isotherm adsorption experiments, the gas adsorption capacity under different equilibrium pressure conditions is measured, and the Langmuir formula is used for fitting to obtain the limit adsorption capacity of coal samples, so as to provide theoretical guidance for the evaluation of coal gas grade. The connectivity, complexity, and pore size distribution of the pore structure directly affect the migration characteristics of CSG such as adsorption, desorption, diffusion, and seepage. Small pores and micropores with high specific surface area provide a place for gas adsorption and are the main gathering place for adsorbed gas. The connectivity of the pore structure affects the effect of gas extraction and the difficulty of coal and gas outburst. When the connectivity between pores is better, it is conducive to the migration of CSG, and the extraction effect is better. On the contrary, the difficulty of extraction increases, and the effect of extraction is not good. During the mining process, the redistribution of stress and the coal body are destroyed, which leads to the occurrence of coal and gas outburst. At this point, the adsorbed gas in the small pore and micropore is desorbed in large quantities under the action of the pressure difference, which provides a power source and a material source for the maintenance of coal and gas outburst. Therefore, conducting physical similarity experiments of coal and gas outburst in the laboratory and exploring the influence of gas expansion energy (Du and Wang 2019;Du et al. 2018), pore structure (Wang et al. 2013a(Wang et al. , 2013bYang et al. 2019), and isotherm adsorption (Liu et al. 2020a, b;Ni et al. 2019;Tang et al. 2020) on its occurrence and development are important means and methods for further and comprehensive analysis of coal and gas outburst research.

Reference co-citation analysis
Reference is an important part of academic papers and the carrier of academic inheritance and academic ethics. It is not only the embodiment of the author's rigorous academic spirit but also an important basis for evaluating the academic level of papers. By tracking and analyzing the references in the field of coal and gas outburst, a more comprehensive understanding of the current research status in this field is gained. The node type is selected as "reference," and 38,586 references are detected. The top 10 references with the strongest citation bursts are extracted and ranked by using the reference burst tool, as shown in Table 5. Table 5 contains burst strength, title, name, author, and publication year. In order of burst strength, the burst strengths of the 10 extracted references are all greater than 8, indicating that the 10 articles have a high citation frequency in a certain period. Among them, there are three articles with burst strengths greater than 10, with strengths of 13.62, 10.60, and 10.23, respectively. The article with the strongest burst strength was published by Karacan et al. (2011) in the International Journal of Coal Geology in 2011, and the title is "Coal mine methane: A review of capture and utilization practices with benefits to mining safety and to greenhouse gas reduction." According to the classification of journals, the journals with the largest number of publications are Natural Hazards, up to 3, followed by the International Journal of Coal Geology, with 2. It demonstrates that the articles included in the two journals mentioned above are deeply concerned by coal and gas outburst research and have significant reference value for research in this field.
According to the classification of time, the publication years are concentrated in 2011-2014, which are distributes in 2011 (3), 2013 (3), and 2014 (4), respectively. The publication years are at the early stage of the rapid growth of publication quantity. The reasons are as follows: Thanks to the application of advanced instruments, the previous macro analysis has shifted to micro research, from qualitative analysis to quantitative description, and thus laid a solid foundation for the subsequent research on coal and gas outburst.

Hotspot analysis
By detecting the change of high-frequency keywords and the evolution of burst words, it provides strong technical support for exploring the evolution path of research hotspots in the field of coal-gas outburst and analyzing the frontier trend. With a tool of "burstness," burst words can be detected and tracked, and the top 10 keywords with high burst strength can be extracted, and then the ranking can be rearranged according to the time of burst year, as shown in Table 6. The dark red part represents the year with burst keywords, which is an active state and has received extensive attention from researchers. The table includes the tracking and induction of the strength of each highfrequency keyword, the start and end of the year, and the way of color changes to reflect the evolution trend of its hotspots.
It can be seen from Table 6 that, combined with the changing trend of publication quantity, the research hotspots are divided into two obvious intervals.
The period of interval one is 2000-2009, which corresponds to the development period of publication quantity. There are four burst keywords, including geochemistry, gas outburst, Sydney basin, and mine, and the burst strengths are 3.88, 5.11, 3.22, and 3.39, respectively. Among them, the burst keyword, geochemistry, has remained active for the longest time, up to 13 years; in this interval, the highest burst strength is 5.11, with gas outburst, but it only ranks third in the entire burst strength ranking.
The period of interval two is 2010-2021, which corresponds to the growth period of publication quantity. There are six burst keywords, accounting for 60%. In this interval, the burst strength of keyword, coal and gas outburst, ranks first, as high as 6.24, followed by Huaibei coalfield, with 5.86. The large number and high strength of burst keywords reflects the frequent occurrence of coal and gas outburst accidents in this interval, which has attracted the attention of researchers. Coal and gas outburst accidents pose a serious threat to mine production safety, particularly in the Huaibei coalfield, due to the occurrence of high ground stress, high gas pressure, and low permeability of coal seams (Jiang et al. 2011;Jin et al. 2016a, b;Wang et al. 2014Zhang et al. 2017;Zhao et al. 2019). Therefore, the research surrounding "coal and gas outburst" and "Huaibei coalfield" in this interval has attracted the attention of many researchers, which makes the two keywords have a high burst.
In the process of coal and gas outburst, a large amount of adsorbed gas is desorbed in the pores of the coal body, carrying broken coal blocks, or pulverized coal to the working face, resulting in casualties and equipment damage. Therefore, the relevant desorption experiments of gas-bearing coal are carried out to explore the desorption characteristics of coal and gas outburst, thus to provide guidance for preventing disasters and accidents (Lu et al. 2021(Lu et al. , 2022Xu et al. 2019aXu et al. , b, 2021Xue et al. 2011;Zou et al. 2020a, b).
With the development of deep mining in recent years, coal mining has gradually entered the stage of deep mining. High ground stress and high gas pressure make the types of coal mine disasters more prone to coal-gas compound dynamic disasters. Therefore, in-depth research on compound dynamic disasters has gradually become a hotspot of current attention and will continue in the future.

Conclusions
In this paper, Citespace visualization software is used to perform a bibliometric analysis of the literature in the field of coal and gas outburst from 2000 to 2021. The knowledge maps are drawn and analyzed in terms of publication quantity, country, institution, keyword cluster, and hotspot evolution. The following conclusions are obtained. (1) According to the time distribution, the research in the field of coal and gas outburst has gone through three stages: the germination period, the slow development period, and the rapid growth period. Especially in 2010-2021, the publication quantity increased exponentially, which opened the golden age of rapid development in the field of coal and gas outburst and has been widely concerned with extensive attention from researchers.
(2) Taking the cooperation network as a reference object, this paper analyzes the cooperation network at the national and institutional levels from the macro and micro perspectives, respectively. The main countries that carry out coal and gas outburst research are China, Australia, the USA, Poland, Russia, and other important coal-producing or consuming countries, with the China University of Mining and Technology, Chongqing University, Henan Polytechnic University, China University of Mining and Technology (Beijing), and Anhui University of Science & Technology as the core research institutions. (3) Based on keywords, the keyword maps of horizontal time evolution and vertical label clusters are drawn and analyzed. From a horizontal perspective, the first occurrence year, frequency, burst state, and evolution law of core keywords in the coal-gas outburst field can be obtained. From a vertical perspective, the cluster labels such as gas outbursts, gas extraction, gas control, and pore structure as the main core are formed.