Superfast Synthesis of Carbon Xerogels

Carbon xerogels (CXs) provide unique opportunities for numerous applications in the areas of adsorption, separation, insulation, catalysis, and electrochemistry, but their use is hampered by time- and energy-intensive synthesis protocols. Synthesis protocols may require several days or more. Here, we report the synthesis of CXs requiring only 5 h using hydrothermal gelation and direct carbonization of the wet polymer. Drying of the polymer gel is of utmost importance for the synthesis of closely related carbon aerogels and is generally considered to be essential for the preparation of CXs. We show that skipping this step has no detrimental effect on the properties of CXs. They are identical to those obtained via conventional routes. With this “superfast” synthesis route, CXs with specific surface areas of about 700 m2 g–1 and total pore volumes up to 1.5 cm3 g–1 are obtained very time-efficiently and without loss of performance.

of CXs with different porous properties, varying amounts of 0.05 M aqueous solution of Na 2 CO 3 (Titripur ®, Merck KGaA Germany) was used to obtain the required RC values.The RC values were varied at 500, 750 and 1000.The solutions were shortly stirred and then poured into Teflon liners, which were inserted into steel autoclaves.The autoclaves were placed in a heating block at 120 °C for 60 min.The solidified RF gel was then crushed with a blade grinder to a powder with particle sizes in the millimeter range.The RF gel granules then were placed in a quartz glass tube that was heated for 2 h at 1000 °C in a Carbolite ® HST split tube furnace for carbonization (heating rate 10 °C.min -1 ).The quartz glass tube was flushed continuously with N 2 .The obtained carbon xerogels are denoted as CX RC SF, RC being the R/C ratio (eq. 1) and M% the mass percentage of the RF educts (eq.2).The obtained carbon yield is comparable with the one of the carbonization of the dried gel (Supplementary Fig. S1).

Hydrothermal Synthesis of Carbon Xerogels
The RF gels were prepared as described for the superfast synthesis with RC variations of 500, 750 and 1000.The obtained gels were then crushed, dried at 80 °C for 24 h, and then placed in a quartz tube and carbonized in a Carbolite ® HST split tube furnace at 1000°C under N 2 flow (with a heating rate of 10 °C.min -1 ).The nomenclature of these carbon xerogels is CX RC HT.

Conventional Synthesis of Carbon Xerogels
The reaction solutions for the normal synthesis were prepared as described above.After 10 min stirring, the solution was poured into glass vials, closed airtight and then placed for 24 h in an oven at 80°C for gelation.The solid RF gel was dried at 80 °C for 24 h and then carbonized under the same conditions as reported above.These carbon xerogels are denoted as CX RC Conv.

Characterization Nitrogen Sorption
For N 2 sorption analyses at liquid nitrogen temperature, the carbon xerogels were degassed under vacuum for 12 h at 300 °C prior to the sorption measurements on a Micrometrics 3Flex instrument.For the data analysis, the MicroActive Software provided by Micromeritics was used.The Brunauer-Emmet-Teller (BET) theory was applied to calculate the specific surface area, S BET .S1 The evaluation of the BET surface area was performed under the Rouquerol criteria.S2 As the materials investigated are not only mesoporous but also microporous, the BET algorithm does not provide correct specific surface areas but rather apparent specific surface areas.S3 For calculation of the micropore volume, V mic , and specific external surface area, S Ext , the t-plot method was used applying the Harkins and Jura equation for calculating the thickness of the adsorbed layer within the capillary pores.S4, S5 The pore width (mode), d pore , in the meso-and macropore range was analyzed with the Barret-Joyner-Halendar (BJH) method using the desorption branches of the isotherms.

Small Angle X-ray Scattering (SAXS)
SAXS analysis was performed on a SAXSess instrument (Anton Paar) with slit-collimation of Cu K α X-ray radiation.Carbon Xerogels were crushed to fine powders and then dried in a convection oven at 80 °C for 3 days.The powders were filled in a slit shaped mask with a width of 10 mm and thickness of 1 mm.The masks were sealed on both sides with transparent tesa® duct tape.The scattering curves were obtained in the q range of 0.13 -7 nm -1 by binning of 10 frames measured with an exposure time of 60 s with a Dectris Mythen 1K strip detector with 10 mm widths.The scattering curves were desmeared with the SAXSquant (V.3.90.2743.35)software algorithm.Using a glassy carbon standard (NIST SRM 3600), the scattering curves were calibrated to an absolute scale according to the specification of NIST.S6 Data evaluation was performed in the Guinier range to gain information on the primary particle dimensions using the Guinier approximation (eq.S3): S7 Eq. S3 contains the parameter, R G , the radius of gyration which is model independent and can be applied to any particle shape.S8 For particles known to be spherical, the mean particle diameter, d part , can be calculated with eq.S4.S9

Ex-situ X-ray Diffraction, Total Scattering and Pair Distribution Function (PDF) Analysis
Ex-situ X-ray diffraction data were recorded on a Rigaku SmartLab diffractometer equipped with a rotating anode (9 kW, 45 kV, 200 mA) in Bragg-Brentano geometry (Cu K 1,2 : 1.541862 Å) for qualitative average measurements.Data were collected continuously in the range of 10 -90° 2 in steps of 0.01° and a scan speed of 0.5° min -1 with a HyPix-3000 multi-dimensional detector in 1D mode.For each sample, three scans were collected and summed after data collection.The samples were placed on background-free silicon sample holders.Total scattering experiments were performed on a Stoe STADI P transmission diffractometer (Mo K α1 ) for local structure analysis.The instrument is equipped with a curved Ge (111) monochromator and a Dectris Mythen 1K detector.The powder samples were prepared in borosilicate capillaries having a diameter of 0.7 mm.The program PDFgetX3 was used for processing pair distribution functions (PDFs) from the scattering data and PDFgui v1.1.2was used to visualize the PDFs.S10, S11 For correction of background scattering from air and sample container, an empty glass capillary was measured.PDFs were calculated for Q max of 15 Å -1 .The experimental PDFs were compared with simulated ones for graphite and graphene monolayer.Crystallographic information file (CIF) for graphite (ICSD: 76767) was provided by the Inorganic Crystal Structure Database (ICSD, © FIZ Karlsruhe).The graphene monolayer structure was approximated by setting a very large interlayer spacing value of 500 Å to the graphite structure for eliminating interlayer pair correlations.Si NIST 640e standard data were used for obtaining the instrumental parameters, i.e., Q damp and Q broad as 0.0071 Å -1 and 0.0039 Å -1 , respectively.

Raman Spectroscopy
For determining the degree of disorder in carbon xerogels, Raman spectra were collected with Renishaw inVia Raman spectrometer using a 532 nm laser, collecting 5 acquisitions per point with an exposure time of 10 s and 1% of the total laser power in the range 100 -3200 cm -1 .A total of 3 points per sample were measured to check the homogeneity of samples.Timcal Timrex® SFG150 synthetic graphite powder was measured as a reference.Prior to measurements, calibration was performed with a Si standard wafer.The spectra were then fitted in TOPAS software v5 with the model described by Sadezky.S12

X-ray Photoelectron Spectroscopy (XPS)
XPS was utilized to characterize the surface chemistry and nature of carbon species in the samples.The powder samples were filled in shallow Ti sample holders and measured on a SPECS spectrometer with a hemispherical analyzer (PHOIBOS 150 1D-DLD).The monochromatized Al K α X-ray source (E = 1486.6eV)was operated at 15 kV and 200 W. Analyzer pass energies of 20 eV and 50 eV were applied for the high-resolution and survey scans, respectively.The pressure during the experiment in the analysis chamber was approximately 10 x 10 -10 mbar.To minimize the effect of charging, a flood gun was switched on before measuring each sample.The data were analyzed using the CasaXPS software v2.3.26rev1.1S.S13 All C 1s spectra were calibrated and normalized to the main photoelectron peak of sp 2 hybridized C at 284.5 eV.To determine the nature of surface carbon species (sp 2 / sp 3 ), the C KLL Auger regions were recorded with 100 eV pass energy, 0.5 eV step size for 50 scans.S14 D-parameters were then calculated by differentiating C KLL Auger spectra using Savitsky-Golay Quadratic smoothing width of 9 and SP background type in CasaXPS.

Transmission Electron Microspcopy (TEM)
The carbon xerogels were measured on a Hitachi H-7100 Transmissiopn Electron Microscope equipped with tungsten cathode operating at 100kV.Images were acquired through a side entry CCD camera and processed with the MegaView III software package.For sample preparation, TEM grids were wiped along the inside walls of the glass vials containing the carbon xerogels.No solvent or adhesive was used.

Scanning Electron Microscopy (SEM)
For acquiring SEM images, a Hitachi TM3030 Tabletop Scanning Electron Microscope was used, operating at an acceleration voltage of 15kV.For sample preparation, the carbon xerogel powders were spread (dry) on conductive double sided adhesive carbon tapes which were then placed on the SEM sample holder.

Complementary Information (not mentioned in the main text)
As nitrogen molecules have a quadrupolar momentum, sorption data on porous carbon materials measured with nitrogen as the adsorbate may result in incorrect data with respect to pore sizes/volumes, especially if the surfaces of the given materials is highly polar.Argon is therefore recommended as a nonpolar adsorbate resulting in unbiased data.For checking the validity of our sorption data if using nitrogen as adsorbate, isotherms were measured with nitrogen at 77 K and with argon at 87 K (Fig. S11).The data were measured on a Micromeritics 3Flex instrument, data evaluation was performed using the Quantachrome NovaWin software package, using the NLDFT kernels for adsorption of nitrogen/argon in carbon slit pores at 77/87 K.The comparison of the data obtained from both isotherms shows that the data with respect to micropore volume (V microp ), total pore volume (V tot ), and mean mesopore size (D mesop ) do not differ significantly as illustrated in Fig. S12.As no significant difference was observed using the advanced NLDFT data evaluation method, standard nitrogen adsorption data were used for data evaluation and comparison of the different carbon xerogels throughout the present work.
Fig.S8.The first derivatives of the Auger spectra of CX 750 series and graphite.The difference in binding energies between the maximum and the minimum is denoted as the D-parameter and is indicative for the ratio of sp2 to sp3 hybridization of carbon species.

Fig. S9 .
Fig. S9.Photographs of the crushed RF gel (left) and the carbon xerogel (right) obtained via the superfast synthesis route.

Fig. S10 .
Fig. S10.SEM images of the carbon xerogel obtained via the superfast synthesis route.

Fig. S11 .
Fig. S11.Isotherms of sample CX 750 SF measured at 87 K for argon and at 77 K for nitrogen.