Bibliometric Evaluation of Research in Hydrochar and Bio-oil

Using biomass to produce an alternative and sustainable source of energy is an attractive possibility. Hydrothermal carbonization is a relatively new technique that is growing in popularity due to the fact that its products have use not only as a source of energy but also in the fields of nanotechnology and material science. The purpose of this study is to examine current research trends and characteristics of hydrochar and bio-oil, the solid and liquid products of hydrothermal carbonization, during the periods of 2009-2018 and 2010-2018, respectively. A bibliometric analysis was performed in order to have a better understanding of the current research being done in the fields of hydrochar and biooil and their different applications. The implication of this study is to shed light on the potential uses of the by-products (hydrochar and bio-oil) produced by the thermochemical technique, hydrothermal carbonization. Using the Web of Science Core Database, 659 articles discussing hydrochar and 71 articles analyzing bio-oil were examined. Furthermore, scientific research produced by various institutions, journals, authors and countries have been recorded and scrutinized. Gephi was used for network analysis of relative co-authors, co-citations and co-keywords used within the two fields.


INTRODUCTION
Two of the main current global challenges are a growing world population and an increasing demand for energy. The current global energy consumption is 524 exajoules per year and this is expected to increase by 65% by 2040, [1] There is a need to shift from fossil fuels to a sustainable and alternative source of energy because fossil fuels are non-renewable resources that contribute towards climate change. Thus, the use of biomass from various wastes is being examined as a potential and sustainable alternative source of energy.
Biomass is an organic material that can be separated into two main categories: lignocellulosic biomass such as woody materials and straw and non-lignocellulosic biomass such as sewage sludge and food waste. When biomass waste decays or is composted, carbon dioxide and methane gas emissions are released into the atmosphere. Thus contributing to an increase in greenhouse gas emissions and climate change. [2] Currently, the abundance of biomass waste that can be turned into energy and chemical products is being regarded as an important sustainable source of clean and renewable energy. Also, energy generated can be used for heat or electricity and thereby help in decreasing the need for fossil fuels. [3] Biomass can be treated thermochemically through pyrolysis and hydrothermal carbonization (HTC). Thermochemical processes when applied to biomass results in solid (biochar in the case of pyrolysis and hydrochar in the case of HTC), liquid (liquor or bio-oil) and gas (little or none in the case of HTC) fractions. Similar to pyrolysis, HTC has been used primarily on lignocellulosic biomass and recently has been applied to non-lignocellulosic biomass and mixtures of different types of biomasses. One of the main differences between the two methods is the moisture content of the feedstock. Pyrolysis treats dry biomass with moisture contents less than thirty percent in an inert or oxygen-free environment at elevated temperatures. [1] HTC converts wet biomass with moisture contents greater than thirty percent. [1] Water acts as both a solvent and a catalyst for HTC. [4] Since drying of the biomass is not required, this would decrease the overall cost of the process. The raw material is treated at temperatures between 150 and 250°C in a closed container and an increase in pressure occurs autogenously. During this process, five main simultaneous reactions are occurring within the biomass. These are hydrolysis, dehydration, decarboxylation, aromatization and re-condensation. [5] The conditions for HTC and the type of feedstock being thermally treated will result in hydrochar and bio-oils with different characteristics which could lead to various potential applications of these products.
However, it is important to note that HTC does have some challenges associated with it. HTC is susceptible to prolonged residence time and non-selective/superficial heating. HTC utilizes an external source of heat and is thus dependent on heat transfer by either convection or conduction. Consequently, the heating process is difficult for users to control. [6] By using microwave energy as the source of heat, these challenges can be alleviated. With microwaves, heating is volumetric and occurs within the product, thereby minimizing limitations of heat transfer into the material and decreasing processing time and power consumption associated with the process.
Woody materials, [7] seafood wastes, [8] agriculture wastes; [9,10] and sewage sludge [11] are just a few examples of the type of biomasses that are currently being used to create hydrochar and bio-oil. Hydrochar and bio-oil can be used as a source of fuel. Moreover, depending on the operating conditions for HTC, the resulting hydrochar can be used as a soil amendment and a low-cost absorbent material for pollutants, toxic metals and biotic contaminants found in soil. [12] Additionally, the bio-oil can be used as a fertilizer for crops.
A bibliometrics analysis can be described as a statistical analysis of publications. [13] It provides evidence on the impact of the research, discovers new and emerging areas of research while identifying potential research collaborators and journals. [14] The field of hydrothermal carbonization and the use of its byproducts from various sources of biomass is rapidly growing. Thus, research activities in the aforementioned fields have increased. The main objective of this bibliometric review is to analyze the literature on hydrochar and bio-oil to provide light on the current state of the research. And also, to highlight hidden information such as publication trends, research areas of current interest and to identify scholars, countries and institutions with maximum productivity during the period of 2009-2018 for hydrochar and of 2010-2018 for bio-oil.

Data Collection
The data for hydrochar and bio-oil were collected from the Web of Science Core Collection database in January 2019 and May 2019 respectively. The Web of Science Core Collection is one of the most dynamic databases that indexes over 20,000 high quality peer reviewed journals in more than 250 categories such as science, social science and humanities. The functionality and sophistication of the Web of Science Core Collection database is superior to other databases such as Scopus and Google Scholar (Norris and Oppenheim 2007). Hence, making it an easy choice for use in this study. No articles were found in this database on hydrochar and biooil before 2009 and 2010, respectively. Therefore, this study examined research publications published solely during the time periods of 2009 to 2018 for hydrochar and 2010-2018 for bio-oil. Information related to publications downloaded from the Web of Science consists of the names of all authors and their affiliations, keywords supplied by the authors, abstracts, cited references and funding information. Among all types of publications such as articles, reviews, proceeding papers, book chapters, editorial materials and discussions, only journal articles were analyzed for this review. The configuration of the assembly for this research and its analysis is shown in Figure 1.
In order to complete a thorough qualitative and quantitative review of the publications in the fields of hydrochar and bio-oil produced by HTC, performance and science mapping analyses were employed in the study.

Performance Analysis
A performance analysis consists of production as well as citation indicators. A production indicator informs the reader about the productivity in any research field, providing information such as the number of articles per year or number of articles per author. Moreover, citation indicators inform the readers about the impact of the specific research field, in terms of number of citations and their given h-indexes. [15] BibExcel software created by Olle Persson was used for the performance analysis of data downloaded as text files from the Web of Science. [16] Apart from calculating various production indicators, BibExcel processes files that could be transported into other software like Pajek, EXCEL or Gephi for further mapping analysis. [16]  BibExcel can process the imported text data to select and examine information related to institutions, author's name, keywords and journals. On-line tutorials by [16] and by other for processing the data are available can be found on software websites 1 .

Science Mapping Analysis
Science mapping helps in the visualization of the structure of the research by demonstrating how they relate to each other through the spatial representation of different fields, discipline or authors. [17] Collaboration networks display the relationships among various authors, institutions and countries working together. Conceptual networks such as co-word networks reveal hot topics in the field that are currently being scrutinized by various scholars. Publication networks, such as co-citation networks reveal a more in-depth understanding about the hot topics being studied through cluster analysis.
A typical network consists of nodes which visually demonstrates an entity and edges that reveals the relationship between the various nodes. In this study, the network analysis was carried out using Gephi, an open source software. Gephi was developed by students at the University of Compiegne in France [18] and can handle larger datasets than other software. Thus, Gephi is the preferred software when compared to others, such as Pajek [19,20] and VosViewer. [21] Network files (.net) generated by BibExcel were imported into Gephi for analysis and graphical visualization of the results. Various statistical algorithms can be initiated to calculate different properties of the network such as its degree, PageRank and modularity class. For proper visualization of the network, formatting of the graphs is necessary. Formatting includes changing the degree range to show which nodes are the most important, editing the layout of the network or altering the color of the nodes, adjusting the size of the nodes depending on their degree or PageRank and editing the weight of each edge.

Cluster Analysis
A cluster analysis collects the nodes into groups in which the nodes have denser connections with each other than with nodes in other clusters. To detect a variety of associated groups or communities in the network, the modularity class value was used. The modularity class value is a scalar value and the network with high modularity values shows that the connections between the nodes of same cluster are denser than the connection of nodes between two different clusters.
In this study, the network analysis of various cited references or publications were combined into clusters depending on their modularity class via Gephi. In order to divide the network 1. https://homepage.univie.ac.at/juan.gorraiz/bibexcel/ analysis into clusters, the modularity class algorithm found in Gephi was used. If two publications are often cited together it is understood that they have a common research interest. [22] The top five papers according to their PageRank were analyzed in order to find the common research area of that particular cluster. Moreover, partitioning the network analysis into clusters allowed for a clearer visual representation of the network analysis and size of the nodes.

PageRank Analysis
The number of citations often determines the importance of a paper. However, this can sometimes be misleading, as the web page is not always properly ranked. PageRank was developed by Google to ensure proper ranking of a web page. [23] In a bibliometric analysis, the PageRank helps to determine the ranking of a publication by taking into account the number of times it has been cited by other highly cited articles. [24] In this study, PageRank was employed to determine the most important articles in the co-citation network. Under the statistics tab in Gephi, the PageRank algorithm was used to calculate the PageRank of all the nodes or publications in the network. The nodes were then sized according to their PageRank. The publications with the highest PageRank were determined to be the most important article in a particular cluster.

h-Index
The h-index is a critical indicator of the impact of a scientist's publications. It takes into account both the number of publications as well as citations count of a scientist, group of scientists, country or institutions in order to rank them. [25] It discusses the productivity as well as the quality of the research carried out. An h-index of X means that X scientific articles have been referenced a minimum of X times. [26] The h-index can be calculated using the data downloaded from the database and BibExcel. The specific steps involved are available on the homepage of the BibExcel software website 2 .
It should be noted that h-index is a dynamic number and thus varies from database to database. This is because the indexing of journals in various databases differs in numbers and only takes into account the number of citations from the journals indexed in their database.

General Statistics
The total number of publications in the field of hydrochar from 2009 to 2018 was recorded to be 659 articles. Original articles excluding conference paper, proceedings, book chapters, editorial material and reviews accounted for 92.4 % of the total number of publications. As depicted in Figure 2, the number of publications remained low until 2011 as the field was still developing. Recently, research on hydrochar has become more active since it is currently being considered both an economical and environmentally friendly substitute for fossil fuels. The total number of publications on hydrochar has continuously increased from 2 publications in 2009 to 177 publications in 2018. The highest average citation per year was 180 in 2009 and the lowest average citation per year of 1 occurred in 2018. The average citation per year during 2009 to 2011 has been found to be higher than the average citation per year in the years to follow. This can be attributed to low initial numbers of publications in the field of hydrochar and the time factor where new articles tends to get fewer citations. Figure 2 demonstrates this more clearly.
The first two published articles that appeared online on the Web of Science Core Collection database in the field of hydrochar was in 2009 focused on the production of hydrochar from hydrothermal carbonization using glucose, sucrose and starch. [27,28] These studies revealed the properties of hydrochar and its potential applications in the field of catalysts, drugs delivery and nanotechnology. And the second study by [29] focused on the production of hydrochar from corn stover.
The total number of articles published on bio-oil was 94 from 2010 to 2018. Original articles in the field accounted for 75.5% of the total articles. The total number of publications on biooil was low until 2013, after which the number of articles has increased incrementally. The highest average citation per year was recorded to be 18 in the year 2010 and the lowest was found to be 1 in 2018.  Table 1a that China produces two times more articles then Germany and about 2.3 times more papers than the USA. China, USA and Malaysia are the top three countries producing articles related to bio-oil.

A B
China and USA are the most prominent countries working on the production of bio-oil from biomass through pyrolysis or hydrothermal carbonization treatment. These two countries account for 69% of total articles in the field of bio-oil.

Journal Distribution
The top ten journals publishing articles in the field of hydrochar and bio-oil are shown in Tables 2.1 and 2.2 respectively. The top ten journals represent approximately 40 and 63 percent, respectively of the total number of articles published. Tables 2a and 2.b display the number of papers, average number of citations and the h-index of the top ten journals in these two fields. The "Bioresource Technology" journal had the highest numbers of publications on hydrochar, followed by "ACS Sustainable Chemistry and Engineering" and "RSC Advances". The h-index of each of the ten journals varied greatly between 7 and 25. Articles in "Bioresource Technology" often discussed the production and application of hydrochar in green engineering, energy and environmental concerns. However, articles in "ACS Sustainable Chemistry and Engineering" and "RSC Advances" generally dealt with the physico-chemical characterization of hydrochar, such as adsorption of metals or biotic contaminants from the waste, production of carbon nanodots or nanospheres and applications in the field of chemistry. For bio-oil, the top three journals were "Bioresource Technology", "Fuels" and "Energy Fuels". The h-index for these journals varied between 2 and 8. The articles in "Fuels" had more citations than those in "Bioresource Technology" even though it had fewer numbers of publications.

Performance of Authors
The

Institutional Analysis
The top ten institutions along with their total number of publications, total number of citations and their h-index are shown in Table 3a and 3b for hydrochar and bio-oil, respectively. A total of 623 institutions have contributed to   and an h-index of 11. Even though the number of articles for the Chinese Academy of Science and ATB are the same, the citational impact of the latter is more than the Chinese Academy of Science.
The total number of institutes publishing on bio-oil was 102, however within this value, some institutions appeared more than once. For example, the University of Minnesota and the University of Minnesota Twin Cities are the same institute. The total number of publications, h-index and the sum of total citations were presented in Table 3.2. The top three institutes with the highest number of publications in the field of bio-oil are the Chinese Academy of Science, Fudan University and Berge et al. [4] Characterization of hydrochars produced by hydrothermal carbonization of lignin, cellulose, D-xylose and wood meal   216  Industrial and  Engineering  Chemistry Research   2012 Kang, Li, Fan and Chang [5] Conversion of sewage sludge to clean solid fuel using hydrothermal carbonization: Hydrochar fuel characteristics and combustion behavior 204 Applied Energy 2013 [11] Hydrogen peroxide modification enhances the ability of biochar (hydrochar) produced from hydrothermal carbonization of peanut hull to remove aqueous heavy metals: Batch and column tests Biomass and Energy 2011 [29] Chemical, structural and combustion characteristics of carbonaceous products obtained by hydrothermal carbonization of palm empty fruit bunches 145 Bioresource Technology 2013 [7] Chemical and structural properties of carbonaceous products obtained by pyrolysis and hydrothermal carbonization of corn stover.

140
Australian Journal of Soil Research 2010 [29] the University of Minnesota. The Chinese Academy of Science published 13 articles and has an h-index of 11, while Fudan University published 7 articles and has an h-index of 6. Finally, the University of Minnesota published 6 articles and has an h-index of 6.

Citation Analysis
For hydrochar, there were 11,434 citations from 659 publications, while for bio-oil, there were 1,570 citations from 71 publications. The top 10 cited papers for hydrochar and bio-oil are presented in Tables 4.1 and 4.2.
The "Chemical and Structural Properties of Carbonaceous Products Obtained by Hydrothermal Carbonization of Saccharides" by [27] was the article with the highest citation for hydrochar. This article discusses the proprieties of hydrochar made up of microspheres produced from saccharides. [27] state that hydrochar produced from saccharides can potentially be used as nanocomposites or in various vivo applications. The most cited paper on bio-oil was the "Characterization and application of chars produced from pinewood pyrolysis and hydrothermal treatment" by. [30] Hydrothermal char and pyrolytic char were characterized and used as absorbent for the copper removal from aqueous solution.
Using the co-citation network, the citing references were examined for hydrochar and for bio-oil. For proper visualization of the complex citational network, a full graph and zoomed out version of each cluster were stored in the Figure  Four hundred and fifty cited references were studied in the co-citation network of hydrochar. This analysis resulted in a total of 7 clusters, depicted by 7 different colors. Four hundred and seventy-eight references were analyzed to determine the co-citation network for bio-oil. This resulted in five separate and unique colored clusters for bio-oil. Moreover, the modularity class separates each cluster. The nodes vary in size according to their PageRank. Information related to Peng et al. [12] High-yield harvest of nanofibers/mesoporous carbon composite by pyrolysis of waste biomass and its application for high durability electrochemical energy storage   The transformation pathways of nitrogen in sewage sludge during hydrothermal treatment 0.020737 Sewage sludge treatment, Biomass processing for syngas production. Comparative study of thermal properties of bio-coal from aromatic spent with low rank sub-bituminous coals 0.002561

Research Hotspots
In a network analysis, the co-occurrence of keywords was examined for hydrochar and bio-oil and the results are shown in Figures 6(a) and (b), respectively. The nodes correspond to the keywords while edges are the lines that link two keywords. The thickness of the edge's correlate to the relationship that each keyword has with one another. A positive correlation exists between the thickness of the edge and the co-occurrence of two nodes or keywords. Keywords in an article describe its content and thus, can be used to study the hotspots of the research currently being emphasized by various researchers. [31] In the case of hydrochar, the top keywords are "hydrothermal carbonization", "hydrochar", "biochar", "biomass", "pyrolysis", "adsorption", "sewage sludge", "activated carbon", "solid fuel", "food waste" and "lignin". Other important keywords are "miscanthus", "bio energy", "heavy metals", "residence time", "wastewater treatment", "digestate", "combustion" and "microalgae", etc. The top key words are linked to different kinds of inputs (biomass, lignin, sewage waste, food waste), thermochemical processes (pyrolysis and hydrothermal carbonization), products (hydrochar, biochar) and potential uses of the solid product (fuel, adsorption of heavy metals and/or activated carbon). Other important keywords follow a similar trend suggesting different types of feedstock that can be used. These keywords also suggest that various results can be obtained depending on hydrochar operating conditions.
It is important to note that the top keywords for bio-oil correlate to the extraction process. For example, bio-oil is made at the same time as hydrochar by hydrothermal carbonization. Thus, "hydrothermal carbonization" and "hydrochar" are part of the top keywords found in Figure. Similarly, the process of gasification is another keyword. The types of biomass that can be used to create bio-oil as well as the pre-treatments that may be necessary for processing the biomass in an energy efficient matter by HTC are also discussed here.

DISCUSSION AND CONCLUSION
Limitations in bibliometric analyses do exist. Three limitations have been identified here. Firstly, articles in any database may not always be 100% correct. For example, some articles related to hydrochar and bio-oil may have been left out when using specific keywords, while others may have been included in the search even though they are not directly linked to hydrochar or bio-oil. Secondly, only articles from journals indexed in the Web of Science Core Collection were included in this analysis. The citation count used as a quality parameter has some disadvantages. For example, it is dependent on time. Therefore, newer articles tend to get fewer citations even though they are robust. Moreover, the h-index  is a dynamic number as it is dependent on the database and number of journals indexed under it. Yet, not all the citations from the journals are indexed under Web of Science Core Collection. Consequently, the h-index varies. Thirdly, articles are categorized on the basis of their source but sometimes the original content of the article can be different therefore creating potential errors.
In conclusion, a bibliometric analysis and the scientific study of hydrochar and bio-oil was performed for the period of 2009-2018 and 2010-2018, respectively. During this time a total of 659 articles were published on hydrochar and 71 articles were published on bio-oil. Interest in hydrochar and bio-oil gradually increased in popularity over the years, reflecting in the increase in the annual number of publications in these subject areas. This bibliometric analysis on hydrochar and bio-oil has cast light on the current state of the research. A research trend depicts the multidisciplinary applications of the hydrochar in different fields from energy to waste management and chemistry. Many studies focused on the potential use of hydrochar in applications as a viable energy source, soil amendments and as a low-cost absorbent material for pollutants, among others. Many researchers have shown that bio-oil could be a potential liquid fuel and fertilizer.
Multiple keywords are associated with the terms hydrochar and bio-oil. The implications of the keywords identify the current areas of interests while describing the subject. The top keywords associated with the term hydrochar are linked to various inputs, thermochemical processes, by-products of the reaction and its applications in various sectors. The top keywords associated with the term bio-oil describe different extraction processes. Together, these key words (hydrochar and bio-oi) are increasingly being used by scholars, journals and in industries around the world.
The fields of hydrochar and bio-oil are still developing with numerous scholars scrutinizing the use of different and unique types of biomass such as human bio-waste, sewage sludge, food waste and seafood waste. Further research and cost analyses will help with the commercial acceptance of using this technology for processing and managing bio-wastes.