Accelerat ing t he world's research. Engineering Carbon Materials from the Hydrothermal Carbonization Process of Biomass Magdalena Titirici Advanced Materials Cite this paper Get the citation in MLA, APA, or Chicago styles Downloaded from Academia.edu Related papers Hydrot hermal carbonizat ion: a greener rout e t owards t he synt hesis of advanced carbon mat ... Magdalena T it irici, Mart a Sevilla Black perspect ives for a green fut ure: hydrot hermal carbons for environment prot ect ion and energy s... Magdalena T it irici, Mart a Sevilla Porous Carbohydrat e-Based Mat erials via Hard Templat ing Magdalena T it irici Download a PDF Pack of t he best relat ed papers Engineering Carbon Materials from the Hydrothermal Carbonization Process of Biomass By Bo Hu, Kan Wang, Liheng Wu, Shu-Hong Yu,* Markus Antonietti, and Maria-Magdalena Titirici* 1. Introduction Synthesis and application of carbon materials have a long history and carbon black, fabricated from fuel-rich partial combustion, has been used for ink, pigments, and tattoos for more than 3000 years. [1] Starting with the discovery of fullerenes [2] and carbon nanotubes, [3] the material science related to valuable carbon materials has become a hot area, motivated by its potential applications in carbon fixation, catalyst supports, adsorbents, gas storage, electrode, carbon fuel cells and cell biology. [4–8] Many synthetic methods, such as carbonization, high-voltage-arc electricity, laser ablation, or hydrothermal carbonization have been reported for the preparation of amorphous, carbonaceous, porous, or crystalline carbon materials with different size, shape, and chemical compo- sitions. [9–13] In this Review, we will focus on a more sustainable approach, which relies on low specific energy input and replaces fossil-fuel-based starting products with biomass. Biomass is a qualified carbon raw material for the synthesis of valuable carbon materials because it is available in high quality (e.g., as pure saccharose) and huge amount, and is a environmental friendly renewable resource. An illustration of its potential is the production of bioethanol, which has emerged as a new fuel for vehicles (usually by mixing gasoline with alcohol). In the United States, more than 7 billion gallons bioalcohol were produced in 2007. In Brazil, almost all the light automobiles are running on the blend of gasoline and bioalcohol, and similar scales can be easily envisaged for materials, appropriate carbon products assumed. Even more abundant, waste biomass derived from agricultural resides and forest byproducts has drawn little attention as a raw material, since only simple combustion has been used to elevate the value of waste biomass. Carbon materials fabricated from waste biomass have shown promising applications as sorption materials, hydrogen storage, biochemicals, and others. [14–17] The problem is that there is still no general and satisfactory process for the production of valuable carbon materials from crude biomass to date. In this respect, a hydrothermal carbonization (HTC) process might have the opportunity to turn into a powerful technique for the synthesis of valuable carbon materials from biomass, especially crude biomass (Scheme 1). According to different experimental conditions and reaction mechanisms, two HTC processes can be classified. Based on the pyrolysis of biomass, a high-temperature HTC process is apt to synthesize carbon nanotubes, graphite, and activated carbon materials under high temperature and high pressure. [18,19] A low-temperature HTC process is carried out up to 250 8 C, employing several chemical transformation cascades, and is a more environmentally friendly route. [20,21] Various carbonac- eous materials with different sizes, shapes, and surface functional groups have been synthesized by this process. Furthermore, these REVIEW www.MaterialsViews.com www.advmat.de [ * ] Prof. Dr. S. H. Yu, Dr. B. Hu, K. Wang, L. H. Wu Division of Nanomaterials and Chemistry Hefei National Laboratory for Physical Sciences at Microscale Department of Chemistry University of Science and Technology of China Hefei, Anhui 230026 (P. R. China) E-mail: shyu@ustc.edu.cn Dr. M.-M. Titirici, Prof. Dr. M. Antonietti Department of Colloid Chemistry Max Planck Institute of Colloids and Interfaces MPI Research Campus Golm 14424 Potsdam (Germany) E-mail: magdalena.titirici@mpikg.mpg.de DOI: 10.1002/adma.200902812 Energy shortage, environmental crisis, and developing customer demands have driven people to find facile, low-cost, environmentally friendly, and nontoxic routes to produce novel functional materials that can be com- mercialized in the near future. Amongst various techniques, the hydrothermal carbonization (HTC) process of biomass (either of isolated carbohydrates or crude plants) is a promising candidate for the synthesis of novel carbon-based materials with a wide variety of potential applications. In this Review, we will discuss various synthetic routes towards such novel carbon-based materials or composites via the HTC process of biomass. Furthermore, factors that influence the carbonization process will be analyzed and the special chemical/ physical properties of the final products will be discussed. Despite the lack of a clear mechanism, these novel carbonaceous materials have already shown promising applications in many fields such as carbon fixation, water puri- fication, fuel cell catalysis, energy storage, CO 2 sequestration, bioimaging, drug delivery, and gas sensors. Some of the most promising examples will also be discussed here, demonstrating that the HTC process can rationally design a rich family of carbonaceous and hybrid functional carbon materials with important applications in a sustainable fashion. Adv. Mater. 2010 , 22, 1–16 ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 1 Final page numbers not assigned REVIEW www.advmat.de www.MaterialsViews.com carbonaceous materials can combine with other components, such as noble metal nanoparticles to form composites with special chemical and/or physical properties. In this article, we will review the concept and history of the HTC process at both high and low temperature. The promising potential of the HTC process for the preparation of carbon materials from biomass will be demonstrated. Finally, we will briefly present some examples of the application of carbonaceous materials from the HTC process in fields such as environment, catalysis, energy storage, biology, and sensors. 2. Hydrothermal Carbonization: A New Way Towards Carbon Materials 2.1. Concept and History Hydrothermal conditions, i.e., application of an aqueous medium over 100 8 C and 0.1 MPa, are widely found in nature, because many minerals form under these circumstances. [22] Since the pioneering work from 1960s to 1980s, the hydrothermal process has been widely used for the synthesis of a vast range of solid-state compounds such as oxides, sulfides, halides, [22–25] molecular zeolites, and other microporous phases. [25] Nowadays, the hydrothermal process has become an important technique for the synthesis of various kinds of inorganic materials, such as functional oxide [26] and non-oxide nanomaterials [27] with specific shapes and sizes, as well as for the synthesis of new solids. [28,29] For the synthesis of valuable carbon materials, the educts of the HTC process usually include carbohydrates, organic molecules, and waste biomass. The treatment of carbon materials under hydrothermal conditions increases or changes solubility, melts crystalline parts, accelerates the physical and chemical interaction between reagents and the solvent, facilitates ionic and acid/base reactions, and finally leads to the precipitation/formation of the carbonaceous structures. Although there is rarely a clear classification, the HTC process could be classified into two main parts by applied temperature. Shu-Hong Yu received his B.Sc. degree at Hefei University of Technology and his Ph.D. from the University of Science and Technology of China (USTC). He joined Prof. Masahiro Yoshi- mura’s laboratory, Tokyo Institute of Technology, as a postdoctoral Fellow. Afterwards, he was as an Alexander-von-Humboldt Research Fellow in the Max Planck Institute of Colloids and Interfaces, Germany, working with Prof. Markus Antonietti and PD Dr. habil. Helmut Co ̈lfen. He joined the Department of Chemistry USTC as a full Professor in 2002, and was appointed the Cheung Kong Professorship in 2006 by the Chinese Ministry of Education. He is now leading the Division of Nanomaterials and Chemistry at the Hefei National Laboratory for Physical Sciences at Microscale, USTC. His research focuses on hydrothermal carbon, bioinspired self-assembly of new nanostructured materials, and hybrids with high performances. Maria-Magdalena Titirici received her basic academic education in organic chemistry and material physics in Bucharest, Romania. After Ph.D. work with B. Selergren in Mainz and Dortmund, she joined the Max Planck Institute of Colloids and Interfaces where she is currently head of the research group ‘‘Sustainable Functional Carbonaceous and Polymeric Materials’’. Her current projects and interests include self-assembly of nanostructured materials, molecular electronics, energy and hydrogen storage, CO 2 sequestration agents, chromatographic stationary phases, and drug-delivery systems. Bo Hu received his B.S. degree from Hefei University of Technology in 2002 and his Ph. D in Inorganic Chemistry from University of Science and Technology of China in 2008 under the supervision of Prof. Shu-Hong Yu. He is interested in the carbonaceous nanostructured materials and other inorganic nanoparticles as well as the self-assembly process for nanodevices. Scheme 1. Schematic illustration of the hydrothermal carbonization (HTC) process as a powerful technique for the synthesis of valuable carbon materials from biomass and the potential applications of the as-produced carbon materials. 2 ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Adv. Mater. 2010 , 22, 1–16 Final page numbers not assigned REVIEW www.MaterialsViews.com www.advmat.de The high-temperature HTC process proceeds between 300 and 800 8 C and is therefore clearly beyond the stability of standard organic compounds. Reactive gases and carbon fragments are to be expected from ‘‘thermolysis’’, which enable the synthesis of carbon nanotubes, graphitic carbon materials, and activated carbon materials. [19,30] The low-temperature HTC process per- forms below 300 8 C, and functional carbonaceous materials can be produced according to dehydration and polymerization schemes known from ordinary organic chemistry. For the coalification of biomass, the low-temperature HTC process is presumably close to natural coalification [6] but, of course, at highly accelerated speed, decreasing the reaction time from some hundred million years to the time-scale of hours. In addition, it is a spontaneous, exothermic process, with the vast majority of the carbon of the starting products also found in the final product (the ‘‘carbon efficiency’’, a sustainability issue, is close to 1). HTC in material synthesis is a 100-year-old technique, with increasing interest originating from the charcoal formation. [6] Bergius first described the hydrothermal transformation of cellulose into coal-like materials in 1913. [31] Then, detailed investigations were focused on the biomass source, [32] the formation process, [33] and the identification of the final coal composition. [34] Since the discovery of carbon nanotubes in 1991, [3] the high-temperature HTC process has been developed quickly. At the beginning of the new century, a renaissance in the low-temperature HTC process appeared with the reports on the synthesis of uniform carbonaceous particles from sugar or glucose. [20,21] In the past few years, lots of functional carbonaceous materials from biomass have been produced via the HTC process and these materials have shown great potential in many fields. [5] Nowadays, with the gradual acknowledgement of hydrothermal process and carbonization mechanism, [35] the HTC has been widely used to smartly design novel carbon and carbonaceous materials from biomass with important applications. [5,6] 2.2. Hydrothermal Carbonization at High Temperature 2.2.1. Carbon Nanotubes The HTC process at high temperature is a powerful method for the fabrication of well-crystallized multi-walled carbon nanotubes (MWNTs). Some novel synthesis routes have been reported for the production of MWNTs. [36–39] In particular, Yoshimura co-workers reported a hydrothermal processing for the synthesis of high-quality MWNTs from amorphous carbon without the use of a metal catalyst at a temperature of 800 8 C and a pressure of 100 MPa (Fig. 1a). [40] The stability and evolution of single-walled carbon nanotubes (SWNTs) has been investigated during hydrothermal treatment at temperature between 200 and 800 8 C and a pressure of 100 MPa. [41,42] The SWNTs were stable even after mild and short-term treatment, and could transform to short MWNTs and graphitic nanoparticles in a high-temperture and high-pressure water system. The HTC process could also effectively modify the surface of MWNTs, such as the production of hydroxyl-group modified MWNTs. [43] 2.2.2. Three-Dimensional Carbon Structures The HTC process at high temperature could be used to prepare carbon films and materials with high flexibility. [44,45] A nice case was the formation of carbon films on carbides under hydro- thermal conditions at 300–800 8 C. [44] This simple route enabled to coat the surface of SiC fibers, powders, platelets, and single crystals with carbon films with controllable thickness from nanometer to micrometer. The mechanism is that the surface of the substrate transformed into carbon films as follow: SiC x O y þ n H 2 O ! SiO 2 þ x C þ n H 2 Monodispersed carbon microspheres, [46–49] ellipsoidal carbon microparticles, [50] olivary carbon particles, [51] nanocells, [52] and graphite tubes [53] have been successfully synthesized from carbohydrates [47,49] and organic molecules [46,50,51,52,53] via the high-temperature HTC process (Fig. 1b–d). For the synthesis of uniform, pure, paramagnetic carbon particles (Fig. 1c), Pol and Thiyagarajan [54] have carefully studied the process by measuring the in situ autogenuous pressure and dissociated chemical species as a functions of temperature during the thermolysis of mesitylene. These two parameters are important for the production scale-up of these spherical carbon particles. Further, the high-temperature HTC process was an effective technique for the fabrication of activated carbon materi- als. [18,30,55,56] For instance, Salvador et al. have reported the production of activated carbon materials from oak wood and anthracite by the high-temperature HTC process with a broad distribution of micropores and some mesopores. [19] Compared to steam activation, the HTC process has higher gasification rate and better penetration power into the pore structure of the char. The HTC process can quickly and greatly change the porosity of the oak char, widening the micropore structure homogeneously (Fig. 2). It was also shown that this must be carbon from biomass. For anthracite char, the HTC process has shown only little effect on the pore development. Figure 1. a) Transmission electron microscopy (TEM) image of MWNTs. Reproduced with permission from [40]. Copyright 2001, American Chemi- cal Society. b) TEM image of a chain of connected carbon cells. Reproduced with permission from [52]. Copyright 2001, Elsevier Publishing Group. c) Scanning electron microscopy (SEM) images of spherical carbon particles. Reproduced with permission from [54]. Copyright 2009, American Chemical Society. d) SEM images of olivary carbon particles. Reproduced with permission from [51]. Copyright 2005, American Chemical Society. Adv. Mater. 2010 , 22, 1–16 ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 3 Final page numbers not assigned REVIEW www.advmat.de www.MaterialsViews.com 2.3. Hydrothermal Carbonization at Low Temperature: Synthesis of Highly Reactive Carbonaceous Nanostructures 2.3.1. Morphology-Controlled Synthesis of Carbonaceous Nanostructures The HTC process at low temperature is apt to generate monodispersed colloidal carbonaceous spheres, as shown in Figure 3, from the carbohydrate sources such as sugar, [20] glucose, [21] cyclodextrins, [57] fructose, [58] sucrose, [59] cellulose, [60] and starch. [59] The formation of these materials includes the processes of dehydration, condensation, poly- merization, and aromatization. [61] Compared with other routes, the HTC process has a couple of advantages, including low toxicolo- gical impact of materials and processes, the use of renewable resources, facile instrumen- tation and techniques, and a high energy and atom economy. [20,21,35] Catalysts, especially metal ions, [4,62,63,64] can effectively accelerate the HTC process of carbohydrates and direct the synthesis of various carbonaceous materi- als. Yu and co-workers. [63] have reported that iron ions and iron oxide nanoparticles could accelerate the HTC process of forming the hollow carbonaceous spheres from starch (Fig. 3d). When the controlled-dehydration products of carbohydrates are partially replaced by organic monomers, a new type of hybrid between carbon and polymer latex can be produced by copolymerization and cycloaddition reaction. These latexes have not only the surface properties of the polymers, but also the structural, mechanical, thermal, and electric properties of the carbon framework. [65] Titirici and co-workers have reported the production of carboxylate-rich carbonaceous materials in the presence of acrylic acid by the one-step HTC process of glucose (Fig. 3c). [65] In the HTC process of carbohydrates, the formation process and the final material structures are rather complicated and a clear scheme has not been reported. This is mainly due to the formation of a multitude of furan-type dehydrated intermediates from carbohydrates, the complexity of the chemistry, and the lack of a satisfactory technique for the final structure discrimination, which could allow the identification of all-carbon sites with higher resolution. [35,58] For example, the dehydration and fragmentation of glucose can give rise to different soluble products, such as furfural-like compounds (5-hydroxymethylfurfural, furfural, 5-methylfurfural), organic acids and aldehydes (acetic, lactic, propenoic, levulinic, and formic acids), and phenols. [61,66–68] Then, polymerization or condensation reactions do occur forming the final forming the final carbonaceous material, [61,69] which have been identified to occure at least along three lines simultaneously, namely aldol-condensation, cycloaddition reactions, and a hydro- xymethyl-mediated furan resin condensation. [70] Among the HTC process of diverse biomass (glucose, xylose, maltose, sucrose, amylopectin, starch), hexose-based carbon sources mainly produce 5-hydroxymethylfurfural (HMF) as the reaction-driving dehydration products, while pentose mainly works via the (more reactive) furfural. [35] A LaMer model [71,72] has been proposed for the explanation of the growth of the carbonaceous materials. This assumes that the final carbonaceous materials display a type of core–shell structure composed of a hydrophobic core and a stabilizing hydrophilic shell that is less dehydrated and contains a large number of reactive oxygen functional groups (hydroxyl/phenolic, carbonyl, or carboxylic). [21,61] The as-synthesized carbonaceous materials usually have intrinsic porous structures with controllable morphology and Figure 2. SEM images of oak wood and oak char. a) Oak char gasified with supercritical water at different burnoffs and b) oak char gasified with steam at different burnoffs. The scale bar was 20 m m. Reproduced with permission from [19]. Copyright 2005, American Chemical Society. Figure 3. a) SEM image of monodispersed hard carbon spherules. Repro- duced with permission from [20]. Copyright 2001, Elsevier Publishing Group. b) TEM image of carbon spheres. Reproduced from [21]. c) SEM images of carbonaceous materials. Reproduced with permission from [65]. Copyright 2009, American Chemical Society. d) TEM image of hollow spheres. Reproduced from [63]. 4 ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Adv. Mater. 2010 , 22, 1–16 Final page numbers not assigned REVIEW www.MaterialsViews.com www.advmat.de surface functionality. For example, the monodispersed carbonac- eous spheres after heat treatment (at 250 8 C) have uniform nanoporous structures and specific Brunauer–Emmett–Teller (BET) surface area of 400 m 2 g 1 . Surface modification of these porous carbonaceous materials (e.g., a high concentration of hydroxyl/phenolic, carbonyl, and carboxylic groups [61] ) leads to a high reactivity, which broaden their application in environment, catalyst, electrochemistry, and drug delivery. [73] It is worthy to note that carbonaceous materials with different functionality could be produced by a general route of mixing the carbohydrates with other small organic monomers via the HTC process. [65] Coupling either hard- or soft-templating effects with the HTC process has shown powerful capability in controlling the synthesis of various carbonaceous nanostructures with special morphology. For example, using ultrathin and ultralong Te nanowires as templates, well-defined ultralong carbonaceous nanofibers can be synthesized from glucose by the HTC process (Fig. 4a). [74] Adjusting the reaction time or the ratio of the tellurium and glucose can effectively control the diameters of carbonaceous nanofibers. Hollow carbonaceous spheres with porous walls could be produced from glucose by the HTC process using appropriately functionalized porous silica particles as templates (Fig. 4b). [75] The efficient deposition of a glucose-derived carbon precursor could be advanced by the electrostatic attraction between the positively charged silica and negatively charged carbon precursors. [75] Soft templates, such as sodium dodecyl benzenesulfonate (SDBS), could induce the template synthesis and assembly process to generate carbonaceous nanowires. [76] Using the anionic surfactant sodium dodecyl sulfate (SDS) as a soft template, the HTC process has been developed for preparing hollow carbonaceous capsules with a reactive surface layer and tunable void size and shell thickness. [77,78] Well-aligned and open-ended carbon nanotubes can be synthesized by using anodic aluminum oxide (AAO) films as templates in the HTC process and subsequent carbonization at high temperature (Fig. 4c and 4d). [79] This synthesis method shows obvious advantages over conventional approaches, such as arc discharge, laser ablation, and chemical vapor deposition (CVD), since it requires no catalysts and uses inexpensive raw materials. After being synthesized, both the inner and outer surface of the CNTs were decorated with Pt nanoparticles via the incipient wet method using NaBH 4 as the reductant, resulting in shell–core–shell-like nanotube composites (Pt–CNT–Pt). [79] These hybrid structures exhibited superior catalytic performance compared to a commercial carbon-black-supported Pt electrode and could be used as the anode catalyst in direct-methanol fuel cells (DMFCs). [79] Owing to the low toxicity, low cost, and high stability, functional carbon nanoparticles have been recognized as benign substitutes for conventional quantum dots based on metallic elements. Various chemical methods have been reported for the controlled synthesis of carbon nanoparticles. However, due to the size dependence of their photoluminescence efficiency, most of the synthesized dots show no efficient visible emission. Bourlinos et al. [80] reported different chemical routes, which were utilized to produce carbon dots that are photoluminescent in the visible region with an average size of less than 10 nm through thermal carbonization of citrate and 4-animoantipyrine precursors, respectively (Fig. 5a and 5b). Functionalized hydrophilic nanoparticles were synthesized via the one-step hydrothermal decomposition of 2-(2-aminoethoxy)-ethanol ammonium citrate salt. As to the hydrophilic quantum dots, organic coronas were covalently tethered to the core by the resulting amide linkage (–NHCO–). The quantum dots exhibited visible-range emission when excited at different wavelengths. As the excitation wavelength increased, the emission-band max- imum of the hydrophilic nanoparticles in water shifted to longer wavelengths. Figure 4. a) SEM image of carbon nanofibers. Reproduced with permission from [74]. Copyright 2006, American Chemical Society. b) TEM image of hollow carbon materials. The scale bar corresponds to 200 nm. Reproduced from [75]. Copyright 2007, American Chemical Society. c,d) SEM and TEM image of carbonaceous polymer nanotubes. Reproduced from [79]. Figure 5. a) TEM image of the hydrophilic nanoparticles. b) Absorption spectrum of the hydrophilic nanoparticles in water. Inset: the corresponding normalized fluorescence spectra at different excitation wavelengths. Repro- duced from [80]. c) TEM image of the submicrotubes. d) SEM images of these tubes. The inset shows a TEM image of the end group of the tube structure. Reproduced with permission from [81]. Copyright 2008, American Chemical Society. Adv. Mater. 2010 , 22, 1–16 ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 5 Final page numbers not assigned REVIEW www.advmat.de www.MaterialsViews.com The HTC process can be combined with other methods for the production of novel carbon materials. Zhan and Yu [81] presented a novel solvothermal treatment of glucose in the presence of a suitable amount of pyridine followed by a successful self- assembly process to produce carbon-rich composite sub- microtubes (Fig. 5c and 5d). The carbon nanoparticles were formed via the pyridine thermal treatment and the surfaces of the nanoparticles were tethered by organic functional groups. After the solvothermal treatment, the obtained black solution contain- ing the carbon nanoparticles was diluted with distilled water and then the self-assembly proceeded gradually. The functional hydroxyl and pyridyl groups tethered to the surface of the sub-microtubes were confirmed by Fourier-transform infrared (FTIR) spectroscopy and X-ray photoelectron spectroscopy (XPS) analysis, which played a vital role in the self-assembly of the nanoparticles in the mixed solution of water and pyridine. [81] Driven by the different affinities between the functional groups and the solvents, the nanoparticles adjusted their mutual position automatically and assembled themselves forming the tubular structures, just like the self-assembly of surfactant molecules to micelles. It is worth noting that the the carbon content of the sub-microtubes was only about 40 wt%, indicating that these structures were just carbon-rich composites, not conventional carbon tubes. The as-synthesized carbonaceous materials with various shape, size, chemical composition, and surface functional groups have shown novel and interesting intrinsic properties, which have been widely studied. These carbonaceous spheres, after calcina- tions, have shown interesting electric properties and have been used as an anode material for lithium ion batteries, showing excellent specific capacitance, area capacitance, cyclic perfor- mance, volumetric capacitance, and reversible capacity. [20,82–86] The materials have been investigated as a counter electrode for dye-sensitized solar cells [87] and even as an efficient fuel for indirect-carbon fuel cells. [78,88] The surface functional groups and porous structures could greatly enhance the efficiency of carbonaceous spheres for removing heavy metals from aqueous solutions. [62] Carbonaceous spheres also possess an unexpected intrinsic fluorescence, which makes them valuable as marker particles. [89] 2.3.2. Carbon-Based Nanocomposites: Encapsulation and In situ Efficient Loading with Metal Nanoparticles The HTC process of clean carbohydrates or organic monomers also provides a favorable reaction environment for twinned, only slightly coupled reaction schemes. When other reagents were added into the HTC process of carbohydrates or organic monomers, novel carbon-encapsulated core–shell composites, nanocables, and hybrids were successfully synthesized by one-step processes, such as one-pot syntheses of Ag@C [4,64,90–94] and Cu@C [95] nanocables (Fig. 6a and 6b). Other similar core–shell structures have also been synthesized. [96–101] Controlling the reagents and reaction conditions effectively adjusted the diameter, length, and thickness of carbonaceous coating. The individual Ag@C nanocables showed excellent conductivity, which is ideal for constructive interconnects in nanoscale devices. [102] Other carbon-encapsulated core–shell composites have also been synthesized in a similar one-step HTC process by mixing metal or metal oxide nanoparticles, such as Ag, [103,104] Pd, [105] Se, [96,106] Fe 3 O 4 , [107] and SnO 2 , with a carbon source. [108] For example, Pd@C core–shell nanoparticles were found to be selective catalysts for the batch partial hydrogenation of hydroxyl aromatic derivatives. [105] Such one-step HTC process can also be used to synthesize hybrid materials with more complex structures and specific properties. [109–115] For example, using the HTC process core–shell Pt@C nanoparticles embedded in mesoporous carbon were successfully synthesized, which had shown excellent stability and high catalytic activity for metha- nol-tolerant oxygen electroreduction. [116] In addition, well-defined monodisperse Ag@phenol formaldehyde resin (PFR) core–shell spheres can be synthesized by a facile one-step method (Fig. 6c and 6d). [117] In the HTC process, carbonaceous materials effectively and uniformly encapsulated preformed nanoparticles, but also arranged the nanoparticles to controlled superstructures through- out encapsulation (Fig. 7a and 7b), thus, interesting nanocom- posites were formed, or the properties of nanoparticels were improved. [118–124] Such nanocomposites show unique chemical and/or physical properties due to their special structures and components and may find applications in various fields such as catalysis, fuel cells, drug delivery, and bio-imaging. [5] For example, for the carbon-decorated FePt nanoparticles, carbonac- eous shell could not only offer a protective coating for FePt nanoparticles but also provide low coercivity and small magnetic interference from neighboring carbon-coated particles. [125] Carbon-coated Fe 3 O 4 nanospindles could serve as a superior anode material for lithium ion batteries with high reversible capacity, high coulombic efficiency in the first cycle, enhanced cycling performance, and high rate capability compared with bare hematite spindles and commercial magnetite particles. [126] Carbon coating on Si nanodots can be facilely realized by the HTC process and further heat treatment will result in the Figure 6. a,b) SEM and TEM images of nanocables with encapsulated, pentagonal-shaped silver nanowires. Reproduced from [4]. c,d) SEM and TEM images of Ag@phenol formaldehyde resin core–shell nanospheres. Reproduced from [117]. 6 ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Adv. Mater. 2010 , 22, 1–16 Final page numbers not assigned REVIEW www.MaterialsViews.com www.advmat.de formation of Si@SiO x /C nanocomposites with a thin layer of SiO x and carbon. [124] Such nanocomposites have displayed significantly improved lithium-storage performance due to the generation of a passivated layer. [127,128] Further treatment of such core–shell composites can result in more complex structures, especially in hollow structures with tunable void space, as shown in the synthesis of the SnO 2 @C double-shell hollow spheres by removing silica from the original silica@SnO 2 @carbon structures. [129] Such SnO 2 @C double- shell structures were further used as nanoreactors to synthesize tin nanoparticles encapsulated in elastic hollow carbon spheres, while heated at 700 8 C for 4 h under N 2 atmosphere. [130] Also, these materials were tested as promising anode materials for high-performance lithium ion batteries, due to their elasticity and the uptake of mechanical stresses. Using carbonaceous capsules as picoliter containers, novel nanocomposites consisting of inorganic nanoparticles confined within a hollow mesoporous carbon shell had been successfully produced. [131] The as-synthesized core–shell materials contain a significant number of filled inorganic nanoparticles, large surface areas, high pore volumes, and the porosity arises from accessible pores of 2–2.5 nm. [131] It was already stated above that the as-prepared porous carbonaceous spheres are also a favorable support for the inorganic nanoparticles. On one hand, the surface functional groups, such as hydroxyl, carbonyl, and carboxylic groups could in situ reduce noble metal ions into noble metal nanoparticles loaded on the carbon [21,132] as shown in Figure 7c and 7d. On the other hand, the surface functionality acts as a primer or binder to stabilize those hybrid structures. Metal, [133,134] bimetal, [135,136] or metal oxide [137,138] nanoparticles have been successfully deposited on the surface of carbonaceous materials. Again, electrochemical performance of such hybrids is very favorable, as shown in lithium ion batteries [139,140] and methanol electro-oxidation. [59,141–144] 2.3.3. Carbonaceous Materials as Sacrificed Template: Synthesis of a Family of Functional Inorganic Hollow and Complex Nanostructures Because of the facile removal of the carbonaceous materials fabricated by the HTC process, the as-synthesized carbonaceous spheres have been used as sacrificed templates for fabricating hollow spheres, which can be applied in catalysis, sensing, chemical/biological separation, and lithium ion batteries. [20,21,35] The general synthesis method mainly includes the adsorption of metal ions from solution to the surface layer of carbonaceous spheres and subsequent removal of the carbonaceous cores via calcinations. [145–147] Based on this method, some hollow spheres were successfully synthesized (Fig. 8a–c), such as Ga 2 O 3 , [146,148] Figure 7. a) Bare a -Fe 2 O 3 composites in a fully lithiated state. Reproduced from [126]. b) TEM image of glucose-treated FePt nanoparticles. Repro- duced from [125]. c) TEM image of carbon spheres loaded with Ag nanoparticles. Reproduced from [21]. d) TEM image of carbon nanofibers loaded with Au nanoparticles. Reproduced from [132]. Figure 8. a) Schematic representation of the formation of metal oxide hollow spheres by using carbonaceous microspheres as templates. Repro- duced from [146]. b,c) SEM and TEM images of the SnO 2 hollow spheres. Reproduced from [146]. d,e) TEM images of TiO 2 nanotubes. Reproduced with permission from [169]. Copyright 2009, Royal Society of Chemistry. f,g) SEM and TEM images of tin-nanoparticle-encapsulated elastic hollow carbon spheres. Reproduced from [130]. Adv. Mater. 2010 , 22, 1–16 ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 7 Final page numbers not assigned REVIEW www.advmat.de www.MaterialsViews.com GaN, [148] WO 3 , [149] SiC, [150] ZnO, [151,152] Mn-doped ZnO, [153] SnO 2 , [154] NiO, [155,156] CoO, [146] In 2 O 3 , [157] TiO 2 , [158,159] SiO 2 , [160] CuO, [147] MgO, [145] CeO 2 , [145] MnO 2 , [161] Mn 3 O 4 , [146] Al 2 O 3 , [146] ZrO 2 , [162] Cr 2 O 3 , [146] layered double hydroxide (LDH), [163] Y 2 O 3 :Eu, [164] La 2 O 3 , [146] Y 2 O 3 , [146] Lu 2 O 3 , [146] MFe 2 O 4 (M ¼ Zn, Co, Ni, and Cd), [165] Fe 2 O 3 , [166] and Bi 2 WO 6 [167] It is worth mentioning that the sizes and structures of different metal oxide hollow spheres are predominantly determined by the templates. These as-synthesized hollow spheres showed high potential applications in the area of gas sensitivity or catalysis. Other kinds of templates have also been used for the synthesis of metal oxide hollow spheres via sacrificial-core techniques, such as colloidal nanoparticles (e.g., Au, Ag, or CdS) and sub-micrometer polystyrene spheres, but these templates are confined to the synthesis of a few particular compositions such as SiO 2 , TiO 2 , SnO 2 , ZrO 2 , and Fe 3 O 4 and cannot be applied as widely as the HTC carbon templates. Similarly, carbonaceous nanofibres also have been used as template for the synthesis of hollow metal oxide fibres (Fig. 8d and e). [168,169] This method can even be extend to the synthesis of uniform ternary oxide nanotubes such as BaTiO 3 [169] In particular, compared to carbon nanotubes, the carbonaceous fibers by the HTC process have higher surface reactivity, making them more suitable for templating production of a variaty of metal oxide nanotubes. [165] The HTC-generated carbonaceous materials can also sacrifice themselves to template some special core–shell-like structures. For instance, synthesis of tin nanoparticles encapsulated within elastic hollow carbon spheres (TNHCs) has been reported (Fig. 8f and g). [130] SnO 2 shells were first deposited on SiO 2 spheres via the hydrolysis of Na 2 SnO 3 . After etching the SiO 2 cores by sodium hydroxide solution, carbon shells were coated on the surface of the SnO 2 hollow spheres via HTC of glucose. Then the samples were heat-treated at 700 8 C for 4 h under N 2 atmosphere, resulting in the TNHCs. Li et al. [170] prepared Fe 3 O 4 @TiO 2 by coating the Fe 3 O 4 core with a glucose-derived carbonaceous layer via HTC, then absorbing the Ti-based precursor onto the surface, and finally calcining the sample under nitrogen. The Ti-precursor was converted to a titania shell, while the carbon layer was simultaneously removed by oxidative heating. This colloidal material was applied for the simple and fast enrichment of phosphopeptides via direct matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF) mass spectrometry. Using a quite similar procedure, a ‘‘jingle-bell’’ structure was obtained with a ferrite hollow sphere and a novel metal core, such as Ag@NiFe 2 O 4 , Ag@CoFe 2 O 4 , Ag@MgFe 2 O 4 , and Ag@ZnFe 2 O 4 [171] Further study on the templating effects of carbonaceous structures still needs to be carried out in more detail in order to realize the goal of controlled templating synthesis. 2.3.4. Porous Carbon Materials With respect to the synthesis of porous carbonaceous materials, the HTC process not only proceeds under facile, low-temperature conditions but also produces porous carbonaceous materials with controllable morphology and surface functional groups as well. The carbonaceous materials collected directly after hydrothermal carbonization have the characteristic to possess only a small number of micropores and, therefore, a small surface area (as compared to activated carbon). However, by further carbonization at higher temperatures the surface areas can reach up to 400 m 2 g 1 due to an increase in microporosity. [20] A disadvantage is that, by increasing the temperature, the surface functionality is partly lost and so is the possibility of further functionalization. This is why a variety of techniques were applied to increase the surface area in the as-synthesized HTC material. For instance, if hydrothermal carbonization of carbohydrates takes place in the presence of various templates or additives, interesting pore systems can be imprinted. The first mesoporous hydrothermal carbons were produced by performing the hydrothermal carbonization in the presence of nanostructured silica templates (Fig. 9). [8] Thus, it was found that it is important to match the polarity of the template surface with the one of the carbon precursor. For mesoporous template, mesoporous carbon shells are obtained after removal of the template, with the whole carbonaceous structure compose