Impact of synthesis conditions on surface chemistry and structure of carbide-derived carbons

Article abstract of DOI:10.1016/j.tca.2009.09.002

Carbide-derived carbons produced by chlorination of titanium carbide at 600, 800, or 1100 °C were subjected to a post-treatment at 600 °C in Ar, H₂, or NH₃ atmosphere. Experimental results suggest that the chlorination temperature influences the ordering of carbon in a manner that impacts specific surface area and porosity. Higher chlorination temperatures lead to higher total pore volume and increased ordering, but lower microporosity. The effect of post-treatments on surface chemistry is pronounced only for samples chlorinated at 600 °C; post-treatments in Ar are shown to be less effective for chlorine removal than those performed in H₂ or NH₃. Post-treatments in Ar result in a lower total pore volume compared to the ones in H₂ or NH₃ for the same chlorination temperature. Samples chlorinated at higher temperatures contained less oxygen functionalities than samples chlorinated at 600 °C, and showed correspondingly less desorption of H₂O, possibly due to diminished uptake of ambient water.

Full text of DOI:10.1016/j.tca.2009.09.002

Thermochimica Acta 497 (2010) 137–142
Contents lists available at ScienceDirect
Thermochimica Acta
journal homepage: www.elsevier.com/locate/tca
Impact of synthesis conditions on surface chemistry and structure of
carbide-derived carbons
Cristelle Porteta, Dmitry Kazachkinb,c, Sebastian Osswalda,1, Yury Gogotsia,∗, Eric Borguetb
a Department of Materials Science and Engineering, Drexel University, 3141 Chestnut Street, Philadelphia, PA 19104, USA
b Department of Chemistry, Temple University, Philadelphia, PA 19122, USA
c Department of Chemical Engineering, University of Pittsburgh, Pittsburgh, PA 15261, USA
a r t i c l e i n f o
a b s t r a c t
Carbide-derived carbons produced by chlorination of titanium carbide at 600, 800, or 1100 ◦C were sub-
jected to a post-treatment at 600 ◦C in Ar, H2, or NH3 atmosphere. Experimental results suggest that the
chlorination temperature influences the ordering of carbon in a manner that impacts specific surface area
and porosity. Higher chlorination temperatures lead to higher total pore volume and increased ordering,
but lower microporosity. The effect of post-treatments on surface chemistry is pronounced only for sam-
ples chlorinated at 600 ◦C; post-treatments in Ar are shown to be less effective for chlorine removal than
those performed in H2 or NH3. Post-treatments in Ar result in a lower total pore volume compared to the
ones in H2 or NH3 for the same chlorination temperature. Samples chlorinated at higher temperatures
contained less oxygen functionalities than samples chlorinated at 600 ◦C, and showed correspondingly
less desorption of H2O, possibly due to diminished uptake of ambient water.
Article history:
Received 13 February 2009
Received in revised form 26 July 2009
Accepted 9 September 2009
Available online 16 September 2009
Keywords:
Carbide-derived carbons (CDCs)
Temperature programmed desorption
(TPD)
Chlorination
© 2009 Elsevier B.V. All rights reserved.
Surface chemistry
Porous carbon
1. Introduction
CDC synthesis from metal carbides is a multi-step process. The
key step is metal extraction from the carbide lattice via chlori-
Porous carbon materials are widely used for filtration, purifica-
tion, energy storage, and fabrication of biomedical systems [1]. One
of the advantages of carbide-derived carbons (CDC) over conven-
tional activated carbons is the ability to tune the porous structure
with a very high accuracy [2–4]. That is due to a regular spac-
ing between carbon atoms in the carbide lattice, which enables a
narrow pore size distribution, large number of carbide precursors
available for CDC synthesis, as well as a diversity of synthesis and
processing conditions (e.g., temperature, chlorination time, etc.)
that can be used to further control porosity. In particular, carbons
derived from titanium carbide (TiC) have not only demonstrated a
record-breaking performance in a variety of energy-related appli-
cations [3,4], but also enabled systematic studies of the effect of
pore size on properties, which could not be performed on activated
or other porous carbons. This led to improved understanding of ion
adsorption in pores smaller than the solvated ion size [3] and other
important scientific breakthroughs.
nation at elevated temperatures with Cl2 [4,5]. After chlorination,
chlorine may be retained on the surface of the carbon having a neg-
ative effect on the performance of energy storage and biomedical
devices, where pure carbon materials are required [6,7]. There-
fore, it is important to eliminate residual chlorine from CDCs and
post-treatments in Ar, NH3 or H2 were suggested for this purpose
[8].
It was previously shown that the surface properties of carbon
black and activated carbons chlorinated at moderate temperatures
are modified due to chlorine bonded to the carbon surface [8,9].
The effect of synthesis conditions on porosity were mentioned in
Refs. [8,9]. In the present paper, we studied the influence of chlori-
nation and post-treatment conditions on the surface chemistry of
CDC, the porosity of CDC, and the structure (ordering) of CDC, top-
ics that have not been investigated in detail previously. The surface
chemistry of CDC is investigated by temperature programmed des-
orption with mass spectrometry (TPD-MS) and energy dispersive
X-ray spectroscopy (EDS). EDS provides information on the bulk
atomic composition of CDC, while TPD-MS simultaneously detects
multiple species desorbing from the surface of CDC upon heating.
Analysis of TDP-MS spectra allows speculating on the nature of the
surface functional groups, their thermal stability, and the presence
of molecules trapped in pores. The porosity and specific surface area
(SSA) are determined based on volumetric adsorption measure-
∗
Corresponding author. Tel.: +1 215 895 6446; fax: +1 215 895 1934.
E-mail address: gogotsi@drexel.edu (Y. Gogotsi).
Current address: Department of Materials Science and Engineering, Mas-
1
sachusetts Institute of Technology, Cambridge, MA 02139, USA.
0040-6031/$ – see front matter © 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.tca.2009.09.002

138
C. Portet et al. / Thermochimica Acta 497 (2010) 137–142
ments. Structural ordering of as-produced and post-treated CDC
is probed by Raman spectroscopy.
Thermogravimetric analysis (TGA) was performed using a SDT
2960 DTA-TGA (TA instruments). Samples were heated from 25 to
700 ◦C at 2 ◦C/min under 40 ml/min air flow.
TPD-MS studies were performed in a high vacuum chamber
(<10−8 Torr) equipped with a residual gas analyzer (RGA 300, SRS)
[13]. The CDC powder was directly pressed between two stainless
steel blocks into a W-grid and a thermocouple was spot-welded to
the W-grid. The sample was then placed into the vacuum chamber
and pumped to <10−8 Torr at room temperature. Before experiment
the sample was cooled to −173 ◦C with liquid nitrogen. TPD-MS
spectra were recorded in the 1–100 amu range while heating sam-
ples at 2 ◦C/s from −173 to 1127 ◦C (Fig. 1). Approximately 100
TPD-MS curves in 1–100 amu range were collected during a single
run, corresponding to 1 data point for a given mass per 10 ◦C. For
further analysis, profiles of masses of interest were extracted from
the full TPD-MS spectrum and plotted as a function of temperature.
2. Experimental
2.1. CDC synthesis
The CDC synthesis has been described in detail elsewhere [10].
TiC with a particle size ranging from 2 to 4 ␮m (Atlantic Equipment
Engineers; 99.9% purity) was used as a precursor. The powder was
placed in a quartz tube and heated to either 600, 800 or 1100 ◦C
in a furnace under Ar flow. Then Ar flow was switched to Cl2 and
chlorination for 3 h at either 600, 800 or 1100 ◦C was used to extract
Ti:
TiC(s) + 2Cl2(g) → TiCl4(g) + C(s)
(1)
After chlorination, the powders were treated for 1.5 h at 600 ◦C
in Ar, H2 or NH3 in order to remove chlorine and volatile
metal chlorides from the pores. The post-treatments were per-
formed at 600 ◦C, the lowest chlorination temperature used, to
minimize changes introduced to the structure of CDC by post-
treatment.
3. Results and discussion
The impact of the chlorination temperature on pore texture
modification of TiC-CDC has been studied and described else-
where [2,3]. Table 1 reports the modification of specific surface
area (SSA) and pore volumes with annealing treatment. As com-
pared to Ar annealed samples, the SSA and volumes of the H2 and
NH3 annealed ones do not show major changes proving that the
chlorination temperature might have more effect on pore texture
modification than the annealing atmosphere. The SSA increases
with increasing chlorination temperature from 600 to 800 ◦C, but
then slightly decreases after chlorination at 1100 ◦C. The total
pore volume increases with chlorination temperature, while the
micropore volume decreases. Furthermore, it can be seen that the
post-treatments also affect the SSA and pore volume; CDCs post-
treated in H2 and NH3 exhibit a higher SSA and pore volume than
CDCs post-treated in Ar (Table 1).
2.2. Characterization
The SSA and porosity of the CDCs were characterized using vol-
umetric sorption measurements (Quadrasorb from Quantachrome
Instruments) in N2 at 77 K and CO2 at 273 K. The SSA was cal-
culated using Brunauer–Emmett–Teller (BET) analysis using N2
sorption [11]. Pore volumes and pore size distributions (not
shown) were determined using the non-local density functional
theory (NLDFT) model [12] using software provided by the instru-
ment package (QuadraWin), and assuming the pores have a slit
shape.
The elemental composition of CDCs was probed by EDS using
a FEI XL30 scanning electron microscope equipped with an EDAX
detector at an accelerating voltage of 20 kV.
Raman spectroscopy was used to characterize the microstruc-
ture of the CDCs. Spectra were acquired using a Renishaw (UK)
1000/2000 Raman micro-spectrometer with an excitation wave-
length of 514.5 nm (Ar ion laser).
The amount of Ti, Cl, and O present in CDC was determined
by EDS. Fig. 2 shows the amount of Cl present in CDC as a func-
tion of chlorination temperature and post-treatment. For Ar treated
samples, the Cl content decreases with increasing synthesis tem-
perature. CDCs treated in H2 and NH3 contain only small amounts
of Cl (<1 wt.%). However, the post-treatment in NH3 appears to
be slightly more efficient for Cl removal than post-treatment in
Fig. 1. Full TPD spectrum collected from CDC sample chlorinated at 600 ◦C and post-treated with Ar at 600 ◦C. The mass spectrometer detects all species in the 1–100 amu
range while the temperature steadily increased from −173 to 1171 ◦C. Multiple peaks can be seen and are assigned to H2, H2O, CO, Cl, HCl, and CO2.

C. Portet et al. / Thermochimica Acta 497 (2010) 137–142
139
Table 1
SSA, total pore volume and micropore volume of the CDCs produced at different temperatures and annealed in different gases at 600 ◦C.
Post-treatment conditions
BET SSA (m2/g)
Pore volume, vN (cm3/g)
Micropore volume, vCO (cm3/g)
Volume of pores < 1 nm (cm3/g)
2
2
600 ◦C, Ar
600 ◦C, H2
600 ◦C, NH3
1290
1365
1382
0.48
0.51
0.53
0.44
0.46
0.42
0.35
0.38
0.34
800 ◦C, Ar
800 ◦C, H2
800 ◦C, NH3
1468
1480
1556
0.58
0.62
0.65
0.41
0.43
0.42
0.30
0.34
0.32
1100 ◦C, Ar
1100 ◦C, H2
1100 ◦C, NH3
1362
1368
1330
0.61
0.68
0.71
0.33
0.24
0.29
0.26
0.19
0.22
250 ◦C, a weight loss of ∼20% is observed for the Ar post-treated
CDC, presumably corresponding to chlorine containing species
(e.g., HCl) and/or H2O desorption that were observed in TPD spec-
tra for Ar post-treated CDC (Figs. 5a and 6a). The oxidation of
CDCs starts around 450 ◦C and the same oxidation rate is observed
independent of the annealing treatment. A slight increase in the
oxidation temperature is observed for the H2 post-treated CDC
which can be explained by its more ordered structure (Fig. 3)
[14,15].
TPD-MS experiments revealed variations in amount of Cl species
desorbing from CDCs pretreated differently (Fig. 5). The only sample
that showed significant desorption of Cl species was TiC-CDC chlo-
rinated at 600 ◦C and post-treated in Ar (Fig. 5a). The release of HCl
from CDC samples post-treated in Ar at 600 ◦C can be explained by
formation of chlorine containing complexes that do not decompose
(volatilize) at 600 ◦C (e.g., surface TiCl). After exposure of samples to
ambient air, those complexes can be hydrolysed with formation of
HCl and surface hydroxides (TiOH). For all other samples, the TPD-
MS results suggest no detectable signal of Cl-containing species.
Thus, in the case of TiC-CDC chlorinated at 600 ◦C, post-treatments
in reductive atmospheres, such as H2 or NH3, effectively remove
Cl. The possible chemical mechanism of chlorine removal could be
described by the following equations:
Fig. 2. Amount of residual Cl, determined by EDS, left in CDCs produced under
different chlorination temperatures (600, 800, and 1100 ◦C) and post-treatment
atmospheres (Ar, NH3, and H2).
H2. The amount of Ti and O found in all CDC samples (data not
shown) was about 0.3 and 7 wt.%, respectively, and did not show
significant changes upon variations in chlorination temperature or
post-treatment gases.
1 H2 + Clsurf → HCl
(2)
(3)
2
4NH3 + 6Clsurf → 2NH4Cl + N2 + 4HCl
Raman spectra of TiC-CDC chlorinated at 600 and 800 ◦C
annealed in Ar and NH3 are shown in Fig. 3a. The post-treatment
has no significant effect on the carbon microstructure. An increase
in the chlorination temperature results in an increased ordering of
the carbon, as indicated by the decreasing ID/IG ratio (Fig. 3b). The
smaller ID/IG ratio obtained after NH3 annealing suggests a more
ordered structure as compared to the Ar annealed samples.
TGA experiments also show a difference between Ar and H2/NH3
treated samples chlorinated at 600 ◦C (Fig. 4). Between 100 and
TiC-CDC synthesized at 1100 ◦C and annealed in Ar shows des-
orption of Cl-containing species of at least two orders of magnitude
lower than TiC-CDC produced at 600 ◦C (Fig. 5a and b). For samples
chlorinated at 800 and 1100 ◦C and post-treated in Ar, the absence
of Cl-containing species leaving the surface upon heating can be
explained by a more effective diffusion of weakly bonded Cl in
larger pores (Table 1) allowing a faster and more complete removal
of Cl during synthesis [9]. Also, some of the Cl could bind chem-
ically to the carbon surface [9,16], requiring significantly higher
Fig. 3. Raman spectra of the CDC synthesized at 600 and 800 ◦C annealed in Ar and NH3 at 600 ◦C (a). Variation of the ID/IG ratio as a function of chlorination temperature (b).

140
C. Portet et al. / Thermochimica Acta 497 (2010) 137–142
moisture removal is arguably the most important step in the
preparation of supercapacitor electrodes. H2O (18 amu) desorp-
tion profiles were compared for samples subjected to different
pretreatment procedures in order to address the question of CDC
affinity for water (Fig. 6). Samples chlorinated at 600 ◦C, display
similar H2O desorption profiles independent of the post-treatment.
As an example, the H2O desorption profile for TiC-CDC chlorinated
at 600 ◦C and post-treated in Ar is shown in Fig. 6a. The amount
of H2O desorbing at ∼300 ◦C significantly exceeds the amount of
water released above 800 ◦C (Fig. 6a). The signal observed at higher
temperatures most likely originates from the decomposition of OH-
containing functionalities on the carbon surface.
The decrease in the amount of H2O desorbed with increas-
ing chlorination temperatures (Fig. 6b and c) can be explained by
increased ordering of CDC with temperature (Fig. 3) and decreased
density of functional groups, that are responsible for wetting of
carbon surface. This is consistent with results discussed below
(Fig. 7); the higher the chlorination temperature the fewer oxygen-
containing functionalities are found.
Fig. 4. TGA of CDCs synthesized at 600 ◦C, annealed in Ar, H2 and NH3. Heating rate
2
◦C/min.
To summarize, we found that samples chlorinated at 600 ◦C and
exposed to ambient air retain a significant amount of water that
desorbs with maximum rate at 300–500 ◦C, while samples chlori-
nated at 800 and 1100 ◦C retain less water. Thus, the chlorination
temperature is the prime factor that defines sample affinity to H2O.
This may be explained by increased graphitic ordering that makes
CDCs less reactive towards ambient species (O2 and H2O) that can
form hydrophilic oxygen-containing functionalities.
Analysis of the desorption profiles of CO (28 amu) and CO2
(44 amu) shows that both species desorb from all CDC samples, sug-
gesting the presence of oxygen functionalities. It can be seen that
TPD-MS profiles of CO and CO2 (Fig. 7) desorbing from both CDC
(Cl2 600 ◦C, Ar 600 ◦C) do not follow the H2O desorption profiles
(Fig. 6). The lack of correlation between desorbing H2O, and CO or
CO2 can be explained by independent processes: water desorption
from pores, surface dehydroxylation, and decomposition of surface
functionalities (carbonyl, carboxyl, and other groups).
temperatures for Cl removal. Papirer et al. reported Cl bonded to
carbon blacks in the form of CCl, CCl2 and CCl3 groups, which
had a strong effect on surface properties by exhibiting a greater
hydrophobic behavior [9,17,18]. It has been demonstrated that
covalently bonded Cl cannot be removed completely by heating
to 1200 ◦C [16]; by analogy we think that Cl remains on the surface
of CDC even at higher temperatures. This is consistent with our
experimental observations where only small quantities of desorb-
ing Cl were detected by TPD-MS for samples chlorinated at 800 and
1100 ◦C (Fig. 5), even though Cl was detected by EDS for the same
samples (Fig. 2). TPD-MS detects functionalities that desorb in the
temperature range explored (−173 to 1127 ◦C), while EDS detects
all the species present in the sample even those that do not desorb
upon heating. To summarize, the results of both TPD-MS and EDS
suggest that H2 and NH3 post-treatments remove residual Cl more
effectively than post-treatments in Ar.
It is well known that CO and CO2 can form upon decomposi-
tion of acidic and anhydride groups at temperatures ∼200–600 ◦C
[19,20], while CO desorbing at >600 ◦C can originate from decom-
Along with chlorine, water is released during CDC heating
(Fig. 1). Moisture sorption from air is important, because it can
strongly affect performance of the carbon sorbent. In addition,
Fig. 5. TPD profiles of HCl and Cl evolved from CDCs chlorinated at 600 ◦C (a) and 1100 ◦C (b) and post-treated in Ar, NH3 and H2. Except for the 600 ◦C CDC annealed in Ar,
there is no conservation of ratios characteristic to natural 35Cl and 37Cl (3:1) abundance suggesting limited desorption of Cl and HCl (note the difference in scale for Ar-600 ◦
sample).
C

C. Portet et al. / Thermochimica Acta 497 (2010) 137–142
141
Fig. 6. H2O desorption profiles from CDCs synthesized at 600 ◦C and post-treated under Ar (a), CDCs synthesized at 800 ◦C (b) and 1100 ◦C (c) post-treated under NH3. The
amount of H2O that can be retained by CDC depends on chlorination temperature: the higher the chlorination temperature the lower is the amount of H2O that can be
retained.
position of neutral and basic groups [19,20]. For the 600 ◦C series,
the CO and CO2 profiles indicate that the post-treatment has no
significant impact on the nature and amount of oxygen-containing
functional groups (Fig. 6a and b). The CO2 desorption profile shows
a broad peak between 150 and 600 ◦C that is likely a result of
the decomposition of carboxylic, anhydride and lactone groups
[19,21,22]. CO that evolves above 600 ◦C can be assigned to the
decomposition of phenol, carbonyl, and quinone groups [22–24].
Because of broad and overlapping profiles of desorbing species the
unambiguous assignment of the observed TPD-MS peaks to cer-
tain functionalities is a challenging task. The main conclusion that
follows from our TPD-MS investigation is the presence of oxygen
functionalities at the surface of CDC. Those results were confirmed
by FTIR. However, quantitative FTIR analysis of oxygen function-
alities in CDC was difficult due to low transparency of CDC and
necessity to dilute CDC with KBr or other salts.
Fig. 7. CO and CO2 profiles of the 600 ◦C CDC post-treated under Ar (a) and NH3 (b) and 800 ◦C CDC post-treated under NH3 (c) and 1100 ◦C CDC post-treated under H2 (d).

142
C. Portet et al. / Thermochimica Acta 497 (2010) 137–142
In general, while there was no effect of post-treatment on
versity of Pennsylvania) for helpful discussions. SEM and Raman
spectrometer were provided by Centralized Research Facilities of
Drexel University. The authors are grateful to the US Department of
Energy (DOE), EERE Program, grant number DE-FC36-04GO14282,
as well as to DOE’s Office of Basic Energy Sciences, for financial
support for this research.
the amount or nature of oxygen functionalities, the increase
in chlorination temperature resulted in decrease of intensity of
both CO and CO2 evolving from CDC samples. The CO2 des-
orption profiles for the samples chlorinated at 800 and 1100 ◦C
are broadened and the maxima are not centered around 300 ◦C
(Fig. 6b and c) as for samples chlorinated at 600 ◦C (Fig. 6a and
b). The peak of CO desorbing at ∼1000 ◦C becomes narrower
and shifts to higher temperatures for samples chlorinated at 800
and 1100 ◦C (Fig. 6b and c). The change in CO desorption pro-
files indicates a modification of the type of oxygen functionalities
[22,23].
According to our experimental data, the formation of oxygen
surface moieties occurs after exposure of CDC samples to ambient
air. The TiC precursor has <2% of oxygen. Moreover, most of the
oxygen-containing surface functionalities [19,20] cannot survive
the temperatures used during CDC synthesis (≥600 ◦C).
References
[1] F. Schuth, K. Sing, J. Weitkamp, in: F. Schuth, K.S.W. Sing, J. Weitkamp (Eds.),
Handbook of Porous Solids, Wiley-VCH, 2002, pp. 1766–1960.
[2] R. Dash, J. Chmiola, G. Yushin, Y. Gogotsi, G. Laudisio, J. Singer, et al., Titanium
carbide derived nanoporous carbon for energy-related applications, Carbon 44
(2006) 2489–2497.
[3] J. Chmiola, G. Yushin, Y. Gogotsi, C. Portet, P. Simon, P.L. Taberna, Anomalous
increase in carbon capacitance at pore sizes less than 1 nanometer, Science 313
(2006) 1760–1763.
[4] G. Yushin, A. Nikitin, Y. Gogotsi, Carbide-derived carbon, in: Nanomaterials
Handbook, 2006, pp. 239–282.
[5] Y. Gogotsi, S. Welz, D.A. Ersoy, M.J. McNallan, Conversion of silicon carbide to
crystalline diamond-structured carbon at ambient pressure, Nature 411 (2001)
283–287.
[6] E. Frackowiak, F. Beguin, Electrochemical storage of energy in carbon nanotubes
and nanostructured carbons, Carbon 40 (2002) 1775–1787.
[7] F.L. Darkrim, P. Malbrunot, G.P. Tartaglia, Review of hydrogen storage by
adsorption in carbon nanotubes, Int. J. Hydrogen Energy 27 (2002) 193–
202.
[8] S.S. Barton, M.J.B. Evans, J.E. Koresh, H. Tobias, The effect of chlorination on
the adsorptive properties of water on carbon cloth, Carbon 25 (1987) 663–
667.
[9] E. Papirer, R. Lacroix, J.-B. Donnet, G. Nanse, P. Fioux, XPS study of the halo-
genation of carbon black. Part 2. Chlorination, Carbon 33 (1995) 63–72.
[10] R.K. Dash, A. Nikitin, Y. Gogotsi, Microporous carbon derived from boron car-
bide, Micropor. Mesopor. Mater. 72 (2004) 203–208.
Both chlorination and post-treatment procedures exclude con-
tact of CDC with oxygen and/or H2O during the synthesis. Thus,
the only possibility for oxygen functionalities introduction to
CDC is the contact of freshly synthesized CDC with ambient air
at room temperature. Zhuang et al. reported the formation of
oxygen-containing functional groups on the surface of carbon after
exposure to air, suggesting a chemical interaction with O2 and/or
H2O [25,26]. Overall, the post-treatments have a minor effect on
the formation of oxygenated functional groups, while it seems
that chlorination temperature is the main factor responsible for
reactivity of CDC with respect to ambient species (oxygen and
water).
[11] S. Brunauer, P. Emmett, E. Teller, Adsorption of gases in multimolecular layers,
J. Am. Chem. Soc. 60 (1938) 309–319.
We believe that the reactivity of the CDCs with respect to oxy-
gen and water, and the nature of oxygen functionalities, depends
mainly, but not exclusively, on the ordering of CDC. As CDC chlori-
nated at 600 ◦C have a less ordered structure compared to the ones
synthesized at higher temperatures (Fig. 4), the CDC with higher
order are likely to be less reactive towards atmospheric species
forming less oxygen functionalities. The information obtained from
TPD-MS is not enough to assign the present surface oxygen func-
tionalities based on desorption temperature. Other experimental
techniques (e.g., Boehm titration) must be involved to identify the
nature of the groups [27–29].
[12] P.I. Ravikovitch, A.V. Neimark, Characterization of nanoporous materials from
adsorption and desorption isotherms, Colloids Surf. A 187–188 (2001) 11–21.
[13] S. Kwon, R. Vidic, E. Borguet, The effect of surface chemical functional groups
on the adsorption and desorption of a polar molecule, acetone, from a model
carbonaceous surface, graphite, Surf. Sci. 522 (2003) 17–26.
[14] H. Nakazawa, T. Kawabata, M. Kudo, M. Mashita, Structural changes of
diamond-like carbon films due to atomic hydrogen exposure during annealing,
Appl. Surf. Sci. 253 (2007) 4188–4196.
[15] N.M.J. Conway, A.C. Ferrari, A.J. Flewitt, J. Robertson, W.I. Milne, A. Tagliaferro,
et al., Defect and disorder reduction by annealing in hydrogenated tetrahedral
amorphous carbon, Diamond Relat. Mater. 9 (2000) 765–770.
[16] B.R. Puri, R.C. Bansal, Surface chemistry of carbon blacks. IV. Interaction of
carbon blacks with gaseous chlorine, Carbon 5 (1967) 189–194.
[17] C.R. Hall, R.J. Holmes, The preparation and properties of some activated car-
bons modified by treatment with phosgene or chlorine, Carbon 30 (1992) 173–
176.
[18] C.R. Hall, R.J. Holmes, The preparation and properties of some chlorinated
activated carbons. Part II. Further observations, Carbon 31 (1993) 881–
886.
[19] J.L. Figueiredo, M.F.R. Pereira, M.M.A. Freitas, J.J.M. Orfao, Modification of the
surface chemistry of activated carbons, Carbon 37 (1999) 1379–1389.
[20] A.E. Aksoylu, M. Madalena, A. Freitas, M.F.R. Pereira, J.L. Figueiredo, The effects
of different activated carbon supports and support modifications on the prop-
erties of Pt/AC catalysts, Carbon 39 (2001) 175–185.
[21] Y. Otake, R.G. Jenkins, Characterization of oxygen-containing surface com-
plexes created on a microporous carbon by air and nitric acid treatment, Carbon
31 (1993) 109–121.
[22] U. Zielke, K.J. Huttinger, W.P. Hoffman, Surface-oxidized carbon fibers. I. Surface
structure and chemistry, Carbon 34 (1996) 983–998.
[23] J.L. Figueiredo, M.F.R. Pereira, M.M.A. Freitas, J.J.M. Orfao, Characterization of
active sites on carbon catalysts, Ind. Eng. Chem. Res. 46 (2007) 4110–4115.
[24] B. Marchon, J. Carrazza, H. Heinemann, G.A. Somorjai, TPD and XPS studies of O2,
CO2, and H2O adsorption on clean polycrystalline graphite, Carbon 26 (1988)
507–514.
4. Conclusions
A correlation between the synthesis conditions, structure and
surface chemistry of CDC was established. Increase of chlorination
temperature results in higher mesoporosity of CDCs, while microp-
orosity decreases. Post-treatment does not have significant impact
on CDC porosity. The ordering (graphitic character) of the CDC
increases with increasing the chlorination temperature. It results
in a lower reactivity of CDC with respect to ambient species. The
study of the influence of post-treatment under Ar, H2 and NH3 of
TiC-CDC synthesized at different temperatures has shown that the
residual chlorine content can be reduced significantly in reductive
atmosphere of NH3 or H2. On the contrary, post-treatment under
Ar for samples chlorinated at 600 ◦C leaves significant amount of
chlorine.
[25] Q.L. Zhuang, T. Kyotani, A. Tomita, The change of TPD pattern of O2-gasified
carbon upon air exposure, Carbon 32 (1994) 539–540.
Samples chlorinated at 600 ◦C are more reactive towards
atmospheric species, leading to increased formation of oxy-
gen functionalities and increased hydrophilicity in comparison
with samples chlorinated at higher temperatures (800 and
1100 ◦C).
[26] Z.R. Yue, W. Jiang, L. Wang, S.D. Gardner, J.C.U. Pittman, Surface characterization
of electrochemically oxidized carbon fibers, Carbon 37 (1999) 1785–1796.
[27] J.-H. Zhou, Z.-J. Sui, J. Zhu, P. Li, D. Chen, Y.-C. Dai, et al., Characterization of
surface oxygen complexes on carbon nanofibers by TPD, XPS and FT-IR, Carbon
45 (2007) 785–796.
[28] H.P. Boehm, Surface oxides on carbon and their analysis: a critical assessment,
Carbon 40 (2002) 145–149.
[29] X. Feng, N. Dementev, W. Feng, R. Vidic, E. Borguet, Detection of low concen-
tration oxygen containing functional groups on activated carbon fiber surfaces
through fluorescent labeling, Carbon 44 (2006) 1203–1209.
Acknowledgements
We are thankful to Professor Christopher Li (Drexel University)
for providing access to the TGA and Professor J.E. Fischer (Uni-