Mesoporous Titanium Dioxide Nanofibers with a Significantly Enhanced Photocatalytic Activity

Article abstract of DOI:10.1002/cctc.201600387

Mesoporous TiO₂ nanofibers with controlled diameter, crystal size, and anatase versus rutile crystal structures are produced by calcination at 500–700 °C of precursor nanofibers of polyvinylpyrrolidone and titanium isopropoxide, obtained from a scalable gas jet fiber spinning process. These TiO₂ nanofibers are used in the photocatalytic oxidation of gas-phase ethanol at room temperature on exposure to UV irradiation. The experimental trends are analyzed using electron–hole (e-h) charge recombination inferred from high-intensity photoluminescence emission spectra, specific surface area, crystallinity, and the proportions of anatase and rutile phases. The nanofibers show a photocatalytic activity that is up to an order of magnitude higher than that of commercial-grade P25 TiO₂ nanoparticles because of slower e-h recombination phenomena in the former. The results show that ethanol undergoes complete oxidation into CO₂ and H₂O on the nanofibers without the accumulation of the intermediate products acetaldehyde and formic acid.

Full text of DOI:10.1002/cctc.201600387

DOI: 10.1002/cctc.201600387
Full Papers
Mesoporous Titanium Dioxide Nanofibers with
a Significantly Enhanced Photocatalytic Activity
Monoj Ghosh,[a] Mehdi Lohrasbi,[b] Steven S. C. Chuang,[b] and Sadhan C. Jana*[a]
Mesoporous TiO2 nanofibers with controlled diameter, crystal
size, and anatase versus rutile crystal structures are produced
by calcination at 500–7008C of precursor nanofibers of polyvi-
nylpyrrolidone and titanium isopropoxide, obtained from a scal-
able gas jet fiber spinning process. These TiO2 nanofibers are
used in the photocatalytic oxidation of gas-phase ethanol at
room temperature on exposure to UV irradiation. The experi-
mental trends are analyzed using electron–hole (e-h) charge
recombination inferred from high-intensity photoluminescence
emission spectra, specific surface area, crystallinity, and the
proportions of anatase and rutile phases. The nanofibers show
a photocatalytic activity that is up to an order of magnitude
higher than that of commercial-grade P25 TiO2 nanoparticles
because of slower e-h recombination phenomena in the
former. The results show that ethanol undergoes complete oxi-
dation into CO2 and H2O on the nanofibers without the accu-
mulation of the intermediate products acetaldehyde and
formic acid.
Introduction
Over the past few decades, indoor air pollution has become
a potential hazard in office buildings, houses, factories, aircraft,
and spacecraft. This has triggered an increasing interest in gas-
solid photocatalytic oxidation research to fight indoor air pollu-
tion caused by volatile organic compounds (VOCs), such as
formaldehyde, acetaldehyde, benzene, chloroform, and etha-
nol, that also irritate the skin, throat, lungs, and eyes and affect
the respiratory system.[1–3] A long-term exposure to some of
these compounds can lead to sick building syndrome (SBS) in
humans.[2,4] Heterogeneous photocatalytic oxidation is one of
the most innovative and promising methods for the degrada-
tion of toxic VOCs at room temperature and, therefore, can be
effective to fight indoor air pollution.[3,5,6] Of the VOCs, ethanol
is a common, widely used chemical in a number of residential
and industrial applications. Its increasing use as a component
of biofuel and a fuel additive in gasoline and continued emis-
sion from industrial processes such as bakeries and breweries
are major contributors of air pollution by ethanol.[7] This work
demonstrates the usefulness of mesoporous TiO2 nanofibers in
conductor metal oxides, TiO2 is one of the most studied photo-
catalysts with well-documented photocatalytic performance,
ready availability, relatively nontoxic nature, self-cleaning prop-
erties on exposure to UV light, and long-term stability.[8] The
photocatalytic activity of TiO2 depends on several parameters,
such as the crystal forms, percent crystallinity, doping materi-
als, surface area, and the density of surface hydroxyl groups.[8]
In this context, a commercial-grade TiO2 nanopowder, P25
TiO2, is recognized as a benchmark photocatalyst for the deg-
radation of organic compounds in aqueous solutions and for
the photo-oxidation of gaseous toxic chemicals.[9] Although
not studied explicitly, the morphology of the crystallites and
the organization of nanometer-size crystals in fibers can play
a significant role in the photocatalytic activity of TiO2.
A detailed study of the literature reveals that P25 TiO2 nano-
particles exhibit high photocatalytic activity if the two poly-
morphs, the anatase and rutile phases, are present at a specific
mass ratio of around 80:20.[10] The photocatalytic activity is at-
tributed to the special electronic states of the two crystal
phases that form semiconductor–semiconductor-type junc-
tions.[8,11,12] This concept was also extended to the doping or
mixing of TiO2 with other semiconducting materials, for exam-
ple, SnO2, ZnO, and V2O5, with metal particles such as V, Pt, Pd,
or Au, or with nonmetal carbon particles, in all cases to tailor
the electronic properties of TiO2 to achieve improvements in
photocatalytic performance.[5,13] Choi et al. doped TiO2 nano-
particles with 21 metal ions by the sol–gel method and ob-
served that metal ion dopants influenced the photoreactivity
because of a change in the charge carrier recombination rates
and interfacial electron transfer rates.[14] However, the metal-
doping of TiO2 nanomaterials does not always lead to higher
photocatalytic activities.[5,15,16] Even carbon doping up to a cer-
tain amount increased the photoactivity of TiO2 for the degra-
dation of organic pollutants under UV and visible light.[17]
the degradation of VOCs by considering ethanol as
representative VOC.
a
Semiconducting metal oxides such as TiO2, ZnO, SnO2, and
V2O5 are useful catalysts for photo-oxidation. Among the semi-
[a] M. Ghosh, Dr. S. C. Jana
Department of Polymer Engineering
The University of Akron
250, South Forge Street, Akron, OH 44325-0301 (USA)
E-mail: janas@uakron.edu
[b] Dr. M. Lohrasbi, Dr. S. S. C. Chuang
Department of Polymer Science
The University of Akron
Goodyear Polymer Center, Akron, Ohio 44325-3909 (USA)
Supporting information for this article can be found under http://
dx.doi.org/10.1002/cctc.201600387.
ChemCatChem 2016, 8, 1 – 12
1
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
Over the past few decades, many research groups explored
the photocatalytic behavior of TiO2 upon exposure to photons
of a suitable energy.[18] TiO2 was given different nanostructural
forms, such as nanowires, nanoparticles, thin films, or nanofib-
ers.[8,19] The majority of TiO2-based photocatalysts in various
morphological forms were prepared by the coprecipitation
method, sol–gel method, ion implantation technique, hydro-
thermal treatment, template-assisted method, and electro-
chemical anodic oxidation method.[20–23] However, all these
methods are complex with a high cost and low TiO2 yield.[24]
Among the various morphological forms reported to date,
the fibrillar structures of TiO2 provide faster charge separation
or slower electron–hole (e-h) recombination and, consequently,
produce higher photocatalytic oxidation rates.[25–27] The long
and continuous fibrillar TiO2 can alleviate concerns associated
with the poor postapplication separation of the catalyst and
the aggregation often observed for ultra-small TiO2 nanoparti-
cles. Qiu et al. reported a higher photoreactivity for flexible
glass-TiO2 composite fibers over commercial P25 TiO2 nanopar-
ticles for the degradation of methyl orange in aqueous solu-
tions.[28] Peng et al. used coaxial nanoscale fibers of SnO2-TiO2
with a tunable internal morphology and showed the faster
photocatalytic degradation of Rhodamine B in water under UV
light in comparison to commercial P25 TiO2 nanoparticles.[27]
The seamless heterojunction between the SnO2 and TiO2
phases provided a higher photoactivity.
hibits a higher photocatalytic activity than the rutile phase.[37]
The composition of the polymorphs (anatase vs. rutile) de-
pends strongly on the calcination temperature.[5,38] The calcina-
tion temperature also influences the grain size and the
porosity of the materials.[5,39]
Herein, we report the enhanced photocatalytic activity of
mesoporous TiO2 nanofibers produced by the GJF process by
considering the photocatalytic oxidation of gas-phase ethanol,
a representative VOC. At the outset, we anticipated that the
TiO2 nanofibers would provide the following advantages. First,
the mesoporous fibers were expected to promote a faster and
greater surface adsorption of the gases. Second, the close con-
tact between the nanocrystals organized in fiber-type mor-
phology was expected to promote faster charge transport.
Third, a unique combination of the anatase and rutile phases
should facilitate interfacial charge transfer and prevent photo-
generated charge recombination. However, the effects of the
calcination temperature on the morphology of TiO2 nanofibers
and the effects of the morphology on photocatalysis were not
known a priori. All photocatalytic oxidation results were com-
pared with those obtained using commercial-grade P25 TiO2
nanoparticles. A detailed analysis of the photocatalytic oxida-
tion of ethanol on TiO2 nanoparticles using in situ IR spectros-
copy method is reported elsewhere.[40,41]
It is evident that the majority of the reports on the photoca- Results and Discussion
talytic activities of TiO2 cover the degradation of liquid-phase
Nanofiber morphology and TiO2 crystal structures
organic dyes under UV or UV/Vis light.[19,25,27,29] However, stud-
ies on the photocatalytic oxidation of a toxic gas on nanofibril-
lar TiO2 are scarce. In this context, a detailed investigation of
the photo-oxidation reaction mechanism and the kinetics of
the decomposition of a representative VOC gas on TiO2 nano-
fiber surfaces is perceived as an important step forward to
fight air pollution. Notably, the fibrous TiO2 materials used in
previous studies were produced primarily by the electrospin-
ning technique with diameters that ranged from a few tens of
nanometers to a few micrometers.[27,30,31] However, lower pro-
duction rates and high voltage requirements limit the scope of
the electrospinning process to meet the demands of moderate
to large quantities of appropriate TiO2 nanofibers. The limita-
tions of the low rate can be overcome by a method of forced
spinning.[32] However, to the best of our knowledge, nanofibers
with core–shell or side-by-side morphologies cannot be fabri-
cated easily by the latter method. Therefore, the development
of a simple and cost-effective method for the fabrication of
high-quality TiO2 nanofibers with enhanced photocatalytic
properties is an attractive and challenging area of research. In
this work, we evaluate TiO2 nanofibers produced from a new,
flexible gas jet fiber (GJF) spinning process.[33–35] The GJF spin-
ning process uses a high-velocity expanding air jet to draw
fibers from a homogeneous solution of a polymer and pro-
vides higher fiber production rates, at least 30 times higher
compared to the rates observed for single-nozzle electro-
spinning.[33–36]
We first evaluated the morphology of nanofibers and the crys-
talline structures of TiO2. The nanofibers of TiO2 were fabricat-
ed using sol–gel chemistry and the GJF spinning process from
spinning solutions that contained polyvinylpyrrolidone (PVP)
and titanium isopropoxide (TTIP) as reported previously.[34]
However, in the previous study,[34] the effect of the calcination
temperature on the nanofiber morphology and crystal struc-
tures of TiO2 was not investigated. A detailed description of
polymer nanofiber fabrication by the GJF process shown in
Figure 1a and b and Figures S1 and S2 is available else-
where.[33,35,36] The individual solid cylindrical precursor polymer
fibers appear smooth (Figure 1c) with diameters in the range
of approximately 200–800 nm and a mean fiber diameter of
approximately (520ꢀ248) nm. The TiO2 nanofibers were of
a much smaller diameter than the precursor fibers (Figure 1c)
because of the loss of PVP and the organic groups from TTIP
in the calcination process and the crystallization of titania.[30,34]
Representative SEM images of randomly distributed TiO2 nano-
fibers derived from calcination at maximum temperatures of
500, 600, and 7008C are shown in Figure 1d–f. The high-reso-
lution images of the cross-sections of single nanofibers are
presented in Figure 1d1–f1. Such nanofibers are designated as
T500, T600, and T700 nanofibers, respectively. It is evident that
the nanofibers retained their integrity during the calcination
process and that they had lengths up to several millimeters as
estimated from the low-magnification SEM images.
TiO2 exhibits two different crystal phases: anatase and rutile.
The anatase phase has a higher charge carrier mobility and ex-
It is apparent from the SEM and corresponding TEM images
presented in Figures 1d1–e1 and 2a and b that the TiO2 nano-
&
ChemCatChem 2016, 8, 1 – 12
www.chemcatchem.org
2
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Figure 1. a) Schematic diagram of the GJF spinning process[33,34] that shows b) liquid jet initiation. c) SEM image of the as-spun precursor polymer fibers. SEM
images of TiO2 nanofibers after calcination at a maximum temperature of d) 500, e) 600, and f) 7008C. Corresponding high-resolution SEM images of the fiber
cross-sections are shown in d1–f1.
fibers (especially, those of T500 and T600) were porous cylin-
ders of 100–550 nm in diameter formed by crystalline grains of
10–30 nm in size as inferred from the TEM images. The data
summarized in Table 1 show that the grain size increased at
a higher calcination temperature.[5,42] Similar observations were
reported by Chuangchote et al. for TiO2 nanofibers obtained
by the electrospinning method.[43] The TiO2 nanofibers calcined
at 7008C exhibit a solid pearl-necklace-like structure (Figur-
es 1 f1 and 2c). The TEM images (Figure S3) of P25 TiO2
powder, however, showed a crystallite size of 8–21 nm. The
average diameter of TiO2 nanofibers in each case was calculat-
ed from 80 different fiber specimens by using Image-J software
(Table 1). The diameter distributions of representative precur-
sor polymer fibers and the corresponding TiO2 nanofibers cal-
cined at 6008C are provided in Figure S4. The images present-
ed in Figure 2a–c demonstrate that the diameter along the
length of a single nanofiber was fairly uniform. The TiO2 nano-
fibers obtained in this work were polycrystalline in nature as
evident from a representative selected area electron diffraction
(SAED) image shown in Figure 2d.
The crystal phase composition (anatase vs. rutile) and the as-
sociated structures of TiO2 nanofibers and P25 TiO2 powder
were further investigated by wide-angle X-ray diffraction
(WAXD; Figure 3a). The XRD patterns of TiO2 nanofibers and
P25 TiO2 powder show the presence of tetragonal anatase and
rutile phases. These patterns are in agreement with JCPDS
card no. 21-1272 for anatase TiO2 and JCPDS card no. 21-1276
for rutile TiO2. The sharp peaks observed at 2q=25.38 (101),
ChemCatChem 2016, 8, 1 – 12
www.chemcatchem.org
3
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
Table 1. Average diameter, grain size, skeletal density of TiO2 nanofibers, and BET surface area from N2 and CO2 adsorption–desorption data.
Sample
Average
diameter
[nm]
Grain size
[nm][a]
Average skeletal
density
Average BET
surface area
Average
pore size
[nm]
Average pore
volume
Micropore volume
[b]
[gccꢁ1
]
[m2 gꢁ1
]
[ccgꢁ1
]
[ccgꢁ1
]
T 500
T 600
T 700
P25 TiO2
305ꢀ71
289ꢀ90
217ꢀ83
21[c]
~13–29
~10–31
~35–80
~8–21
3.29ꢀ0.017
3.30ꢀ0.017
3.44ꢀ0.021
3.82ꢀ0.015
29.50
17.50
3.40
5.30
9.50
22.60
15.50
0.068
0.051
0.010
0.147
0.0080
0.0025
0.0013
0.0050
45.25
[a] Evaluated from TEM images. [b] Determined from CO2 adsorption–desorption isotherms at 08C. [c] Reported primary particle size from the supplier.
anatase and rutile phases (Table 2). Notably, the grain size in-
ferred from the TEM images is close to the crystallite size from
the XRD data except for the samples obtained at 7008C. The
TEM images, however, do not distinguish between the anatase
and rutile phases.
The XRD peaks that correspond to the anatase and rutile
phases were integrated to determine the fraction of each
phase,[45] and the data are presented in Table 2. It is observed
that TiO2 nanofibers calcined at 5008C contained 100% ana-
tase phase, whereas a mixture of anatase and rutile phases
was present in the samples calcined at 600 and 7008C. The
data also confirm that TiO2 nanofibers obtained at 7008C were
mostly rutile (~96%). This trend of calcination temperature
versus rutile fraction is similar to that reported previously,[38] al-
though Chuangchote et al. reported a much lower anatase
phase content (7.2%) for nanofibers calcined at 6008C.[43] The
XRD pattern of a reference sample of P25 TiO2 powder (Fig-
ure 3a) confirms the presence of anatase and rutile phases in
proportions of approximately 78 and 22%, respectively.
An increase in the calcination temperature also resulted in
an increase of the skeletal density of the TiO2 nanofibers
(Table 1), for example, 3.44 gccꢁ1 at 7008C versus 3.29 gccꢁ1 at
5008C. Correspondingly, the specific surface area (SSA) of TiO2
Figure 2. TEM images of TiO2 nanofibers after calcination at a maximum
temperature of a) 500, b) 600, and c) 7008C. d) SAED pattern of TiO2 nanofib-
ers calcined at 6008C.
nanofibers determined from the BET data with N2 shows a de-
crease with the increase of calcination temperature, for exam-
38.23 (004), 48.47 (200), 54.71 (105), and 56.018 (211) in the
XRD pattern of the nanofibers calcined at 5008C are consistent
with the anatase crystalline phase; the additional rutile peaks
appear at 2q=27.58 (110), 36.90 (101), 41.60 (111), and
55.408 (211) for materials calcined at 600 and 7008C. Notably,
the phase transition from anatase to rutile starts at 6008C.[44]
The crystallite size (t) in the calcined specimens was deter-
mined from the XRD data using Scherrer’s equation [Eq. (1)]:
ple, 29.5 m2 gꢁ1 at 5008C compared to 3.40 m2 gꢁ1 at 7008C. In
comparison, the P25 TiO2 powder provided a surface area of
45.25 m2 gꢁ1, which is much higher than 17.5 m2 gꢁ1 for nano-
fibers calcined at 6008C. The N2-BET isotherms and the Barrett–
Joyner–Halenda (BJH) pore size distributions for the nanofiber
samples and the reference P25 TiO2 powder are shown in Fig-
ure 3b. The nanofibers calcined at 500 and 6008C are mesopo-
rous and show Type IV isotherms as per the IUPAC classifica-
tion with an H3 hysteresis loop, which indicates the capillary
condensation of N2 within the non-uniform mesoporous struc-
tures.[46,47] The N2-BET isotherm of P25 TiO2 powder and the
T700 nanofibers calcined at 7008C show Type II isotherms,
which corresponds to non-mesoporous materials. It is ob-
served that the mesopore volume of the calcined TiO2 nanofib-
ers decreased with an increase of the calcination temperature.
In view of the BET data, we infer that the nanofibers calcined
at 500 and 6008C are primarily mesoporous in nature.
t ¼ Kl=ðb cos qÞ
ð1Þ
in which l is the wavelength, q is the Bragg angle, K is the
shape factor (~0.9 for spherical shape), and b is the full width
at half maximum of the peak.[38]
The values of t are listed in Table 2. In addition, Bragg’s dif-
fraction formula was used to determine the d spacing in the
crystal structure. The XRD data indicate that an increase in
crystallite size can be attributed to improved nucleation and
crystallite growth at a higher calcination temperature.[39] The
crystallite sizes are uniquely determined from the XRD data for
The band-gap energy of the TiO2 nanofibers calcined at
5008C (Table 2 and Figure S5) is determined to be an indirect
band gap with an energy of 3.23 eV.[48] This further confirms
&
ChemCatChem 2016, 8, 1 – 12
www.chemcatchem.org
4
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Figure 3. a) WAXD patterns, b) N2 adsorption–desorption data with the size distribution of mesopores in the inset, c) XPS data, and d) high-resolution XPS
data of the Ti2p peaks of TiO2 nanofiber samples calcined at 500, 600, and 7008C with ramp of 0.58Cminꢁ1. Data for reference material P25 TiO2 is also
shown.
Table 2. Anatase and rutile phase content, optical band gap, and XPS data for elemental analysis and comparison of the intensity ratio of the peaks for
OH (IOH) and bulk O2ꢁ (IOxide) of TiO2 nanofibers calcined at different temperatures and P25 TiO2 nanopowder.
Sample
Rutile phase
[%]
Anatase phase
[%]
Crystallite size, t
[nm]
d spacing
[nm]
Band-gap energy
[eV]
C[a]
O[a]
Ti[a]
IOH/IOxide
T500
T600
T700
P25 TiO2
0
36
96
22
100
64
4
12.32A
0.362A
3.23
3.02
2.95
3.08
22.1
13.6
11.8
14.5
56.5
62.2
62.8
61.3
21.4
24.1
25.3
24.2
0.284
0.286
0.271
0.220
13.23A, 12.95R
14.18A, 15.75R
11.24A, 13.05R
0.362A, 0.332R
0.362A, 0.332R
0.357A, 0.328R
78
[a]=Atomic %. A=Anatase (101); R=Rutile (110)
the presence of 100% anatase phase in this material.[49] The
band-gap energy of TiO2 nanofibers decreased gradually from
3.23 to 2.95 eV with an increase of the calcination temperature
(Table 2). We attribute this decrease to an increase of the rutile
phase content, which has a lower band-gap energy. This trend
is in agreement with the data presented previously for other
forms of TiO2 materials.[48]
The precursor polymer fibers turned into ceramic TiO2 nano-
fibers during the calcination process. The oxidation and de-
composition of PVP and the decomposition of TTIP occurred,
which is evident from the TGA data presented in Figure S6.[30,50]
No significant weight loss occurred at temperatures higher
than 6008C, which indicates the complete transformation of
the precursor materials into oxide nanofibers at 6008C. As will
ChemCatChem 2016, 8, 1 – 12
www.chemcatchem.org
5
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
be seen in conjunction with energy-dispersive X-ray spectros-
copy (EDS) and X-ray photoelectron spectroscopy (XPS), the re-
sultant oxide nanofibers contained small fractions of carbon as
well. This is attributed to char formation by the polymer
precursor materials.
sented in the Experimental Section and has also been pub-
lished previously.[40,41]
Photocatalytic oxidation of ethanol
The XPS data indicate the presence of Ti and O in TiO2 nano-
fibers and P25 TiO2 powder clearly (Figure 3c). The symmetrical
peaks for Ti shown in Figure 3d with binding energies (BE) of
457.82 and 463.65 eV correspond to Ti2p3/2 and Ti2p1/2, respec-
tively, of Ti4+. The separation between these two peaks is
~5.8 eV, which is similar to the energy splitting data reported
for TiO2.[51] The peaks caused by Ti3+ in Ti2O3 are absent, which
indicates that all of the Ti was present as TiO2 in the nanofib-
ers. Notably, a change in the calcination temperature did not
affect the oxidation state of Ti. The Ti2p binding energy peaks
for the nanofibers calcined at 5008C shifted by approximately
0.5 eV to 458.26 and 464.09 eV possibly because of the pres-
ence of greater amounts of carbon, which influences the bind-
ing energy of the Ti2p electrons.[52]
The IR absorbance spectra of P25 TiO2 powder and TiO2 nano-
fibers are shown in Figure S11. The values of the IR absorbance
(A) were obtained from the single-beam intensity (ISB) and ref-
erence-beam intensity (I0) given by A=log[I0/ISB]. A single-
beam spectrum is a raw data format that reflects the detector
response of the spectrometer in IR frequencies.[54] The IR spec-
tra of TiO2 catalysts exhibited bands at n˜ =3692, 3667, and
3632 cmꢁ1, which correspond to the surface hydroxyl (ꢁOH)
groups, a broad band at n˜ =3550 cmꢁ1, and a band at n˜ =
1625 cmꢁ1, which corresponds to adsorbed water.[55,56] The
spectra of the T500 and T600 nanofibers showed additional IR
bands at n˜ =2345,1581, 1540, 1442, and 1340 cmꢁ1, which cor-
respond to C=O/C=C-containing species that remain on the
nanofibers from the thermal degradation of PVP.[57]
The deconvoluted high-resolution spectra of the O1s level
presented in Figure S7 consisted of two peaks at BE= ~529
and ~531 eV, which indicates the crystal lattice oxygen (OL)
and adsorbed oxygen or oxygen in C=O and CꢁO (ether), re-
spectively.[52] The lattice oxygen is caused by TiꢁOꢁTi bonds in
the TiO2 crystal lattice. The adsorbed oxygen (OH) is mainly
caused by OH radicals, physically adsorbed O2, or oxygen cavi-
ties that result from chemisorbed water.[51] The proportions of
each oxygen type were calculated from the deconvoluted O1s
spectra and are presented in Table 2.
The IR absorbance difference spectra collected after the ad-
sorption of ethanol on P25 TiO2 and T600 nanofibers in O2 at-
mosphere and after 0, 1, 5, 10, and 60 min of UV irradiation are
presented in Figure 4a and e. The difference spectra were ob-
tained by subtracting the absorbance spectra before UV irradi-
ation (At=0 min) from the absorbance spectra after t min of a UV
irradiation (At min). UV irradiation on TiO2 caused a decrease in
the intensity of the CꢁH stretching bands at n˜ =2971, 2931,
and 2870 cmꢁ1 (Figure 4a and e) and the evolution of new
bands at n˜ =1718 cmꢁ1 (acetaldehyde, CH3CHO), 1442 cmꢁ1
(acetate, CH3COOꢁ), 1581 cmꢁ1 (formate, HCOOꢁ), and
2360 cmꢁ1 (CO2).[58–60]
The content of OH as measured by the IOH/IOxide ratio (Table 2)
is higher for the nanofibers considered in this work than for
the TiO2 P25 powder. Notably, adsorbed oxygen is beneficial
for the enhancement of the photocatalytic activity, which will
be discussed later.[53] An increase of the surface hydroxyl
groups can aid the trapping of more photogenerated holes by
the TiO2 catalyst and, therefore, can prevent e-h recombination
and, consequently, leads to high photoreactivity.
The main species that initiate photocatalytic oxidation have
+ [40,61,62]
C
been identified as OH and h .
The steps that lead to the
generation of these species on TiO2 are given in Equations (2)–
(6):
þ
TiO2þhn $ TiO2þeCBꢁþhVB
H2O ! OHꢁþHþ
ð2Þ
ð3Þ
ð4Þ
ð5Þ
ð6Þ
The high-resolution XPS spectra of the C1s level presented
in Figure S8 shows a large peak centered at BE=284.49 eV at-
tributed to CꢁC or CHn, whereas the other two small peaks at
BE=285.95 and 288.30 eV are possibly from trace amounts of
CꢁOꢁC and carboxyl carbon atoms and the TiꢁOꢁC bonds, re-
spectively.[29] The EDS data (Figure S9) reveal that the nanofib-
ers calcined at 5008C were composed of C, Ti, and O and no
impurities and that its carbon content was relatively higher
than that of the nanofibers calcined at 600 and 7008C
(Table S1). However, the TiO2 stoichiometry in the bulk material
measured by using EDS was slightly different from that on the
surface measured by using XPS.
OHꢁþhþ ! OH
C
ꢁ
O2þeꢁ ! O2
ꢁ
ꢁ
C
2 O2 þ2 H2O ! 2 OHþ2 OH þO2
+
ꢁ
In Equations (2)–(6), hVB is a hole at the valence band, eCB
is an electron at the conduction band, h+ is a hole, and eꢁ is
an electron. The holes produced from UV illumination can
C
react with H2O to produce OH, whereas the electrons must
react in a key step, such as the transfer from TiO2 surface to O2
ꢁ
C
[Eq. (5)], and O2 reacts with H2O to produce OH [Eq. (6)]. The
simultaneous occurrence of these reaction steps is required to
avoid charge accumulation, which can retard the photocatalyt-
ic oxidation rate.[63] The presence of H2O on the TiO2 surface
Photocatalytic oxidation of ethanol on TiO2 nanofibers
The photocatalytic oxidation of ethanol on TiO2 nanofibers cal-
cined at 500, 600, and 7008C was studied by in situ IR spec-
troscopy (see Figure S10 for experimental set-up), and P25
TiO2 nanopowder was used as a reference material. A detailed
experimental procedure for data collection and analysis is pre-
C
suggests that OH would be available to initiate the oxidation
reaction by Pathway A in Figure 5 to generate the formate. In
C
addition, OH can desorb from the TiO2 surface and oxidize the
species that are not adsorbed directly on the TiO2 surface.[64]
&
ChemCatChem 2016, 8, 1 – 12
www.chemcatchem.org
6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
C
Therefore, OH can attack the hydrogen on both the
a- and b-carbon atoms in CH3CH2OH. The abstrac-
C
tion of hydrogen from the b-carbon atom by OH
can lead to the breaking of the CꢁC bond to result
in the formation of the formate. In contrast, the ab-
straction of hydrogen from the a-carbon atom
would lead to the formation of aldehyde (IR band at
n˜ =1718 cmꢁ1) and acetate (IR band at n˜ =
1442 cmꢁ1). These species are key intermediates in
ethanol photocatalytic oxidation on TiO2 and are
eventually converted to CO2 and H2O.
The ethanol photocatalytic oxidation rate was esti-
mated from the variation of the IR intensity of the
CꢁH stretching band over 60 s of UV irradiation by
adopting a first-order reaction kinetic model (Fig-
ure 4b and f).[65] We assumed that (i) ethanol photo-
catalytic oxidation is an elementary reaction be-
tween CH3CH2OH and UV-generated active oxygen-
containing species (AOS) on TiO2 and (ii) the AOS
concentration is constant as the UV light intensity
remained constant during the experiment. Thus, the
ethanol photocatalytic oxidation rate was estimated
from the slope of the plot of ln[CꢁH band intensity]
versus time (insets of Figure 4b and f). The estimat-
ed photocatalytic oxidation rates and the amount of
CO2 product for P25 TiO2 particles and T500, T600,
and T700 nanofibers are summarized in Table 3.
T600 nanofibers produced the highest ethanol
photocatalytic oxidation rate among all the TiO2 cat-
alysts under the same UV light intensity. The T600
nanofibers showed an ethanol photocatalytic oxida-
tion rate of 2.1ꢁ10ꢁ2 sꢁ1, which is almost one order
of magnitude higher than that of ethanol on P25
TiO2 powder and three orders of magnitude higher
than that reported for TiO2 particles synthesized by
a sol–gel process.[66] The rate of ethanol photocata-
lytic oxidation on P25 TiO2 powder was estimated to
be 3.0ꢁ10ꢁ3 sꢁ1 in agreement with the previous
studies.[40] The activity of the T500 and T700 nanofib-
ers also surpassed that of P25 TiO2, although their
ethanol photocatalytic oxidation rate was lower than
that of T600 nanofibers.
Figure 4. a, e) IR absorbance difference spectra of P25-TiO2 nanopowder and T600 nano-
fibers during 60 min of photocatalytic oxidation of ethanol in O2 environment. b, f) Varia-
tion of IR intensity of ethanol at n˜ =2971 cmꢁ1 (CꢁH intensity) during 60 min of ethanol
photocatalytic oxidation on P25 TiO2 and T600 nanofibers. Inset: logarithmic plot of the
normalized CꢁH intensity in ethanol, ln[NormI2971], during the first 60 s of UV illumina-
tion. The normalized intensity was obtained from NormI2971 =I(t)/Imax, in which I(t) is the
absorbance intensity of the CꢁH stretching band at time t and Imax is the maximum in-
tensity of the CꢁH stretching band over 60 min. c, g) Variation of the IR intensity at
n˜ =2000 cmꢁ1 (i.e., I2000, background shift) over 60 min ethanol photocatalytic oxidation
on P25-TiO2 and T600 nanofibers. d, h) Variation of CO2 amount with time for P25 TiO2
powder and T600 nanofibers.
The photocatalytic activity of P25 TiO2 powder was
attributed previously to the nanoscale junctions be-
tween the anatase and rutile phases.[10] The anatase–
rutile phase heterojunctions were also identified in
T600 nanofibers from XRD data. The close resem-
blance of the XRD patterns of P25 TiO2 powder and
T600 nanofibers suggests that the crystallite size and
the crystal phases of TiO2 in these two samples were
almost equivalent. Previous work suggests that the
anatase to rutile phase mass ratio in TiO2 should be
in the range of 70:30–80:20 for high photocatalytic
performance.[10] In this context, the T600 nanofibers
were composed of anatase and rutile crystallites at
the mass ratio of 64:36.
Figure 5. Reaction pathways for the photocatalytic oxidation of ethanol on TiO2.[40,41]
ChemCatChem 2016, 8, 1 – 12
www.chemcatchem.org
7
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
using a batch reactor, and the surface area should
play a role as the amount of gas adsorbed per gram
Table 3. Summary of ethanol photocatalytic oxidation on TiO2 nanofibers and P25
TiO2 nanoparticles.
of the nanomaterials depends directly on the surface
Catalyst
Ethanol coverage
CO2 amount
CO2 amount
Initial CꢁH bond
area. The differences in surface area and porosity
on catalyst
breaking rate
values reported in Table 1 cannot be invoked to ex-
ꢁ1
ꢁ1
ꢁ1
ꢁ1
TiO2
[mmolgTiO
]
[mmolgTiO
]
[mmolh
g
]
[sꢁ1
]
2
2
plain the significant differences in the photocatalytic
P25 TiO2
T500
T600
820
550
620
580
460
600
950
700
440
540
855
633
3.0ꢁ10ꢁ3
4.9ꢁ10ꢁ3
2.1ꢁ10ꢁ2
7.2ꢁ10ꢁ3
activities observed for P25 TiO2 powder and T600
nanofibers. The BET specific surface area and the
pore volume of T600 nanofibers (17.5 m2 gꢁ1 and
T700
0.051 cm3 gꢁ1, respectively) are lower than those of
P25 TiO2 powder (45.25 m2 gꢁ1 and 0.147 cm3 gꢁ1, re-
The T600 nanofibers and P25 TiO2 powder show similar pho-
toluminescence (PL) patterns with strong emission bands at
l=467 and 560 nm, respectively (Figure 6). However, the PL
intensity for the T600 nanofibers is lower. This indicates that
the recombination of photoinduced electrons and holes can
be inhibited more effectively in T600 nanofibers.[67–69]
spectively) and T500 nanofibers (29.5 m2 gꢁ1 and 0.068 cm3 gꢁ1,
respectively). T600 nanofibers exhibited the best photocatalytic
performance although they had the smallest pore volume and
lowest surface area among the three nanomaterials. A weak in-
fluence of the surface area on the photocatalytic oxidation of
CH3CHO and reduction of CO2 over P25 TiO2-graphene compo-
site materials has been reported elsewhere.[67,76]
The significantly higher photocatalytic performance of T600
nanofibers than that of P25 TiO2 powder can be justified if we
consider the role of its mesoporous morphology. The faster ini-
tial oxidation rate is tied to faster electron or charge transport
in the fibrous morphology.[25] Specifically, fibrous TiO2 materials
produce faster charge separation and vectorial electron trans-
port than the nanoparticles.[27] The results of the PL study dis-
cussed earlier support that the rate of e-h recombination for
T600 nanofibers is lower than that of P25 TiO2 powder. In addi-
tion, the fibrous morphology plays a significant role in photo-
generated charge accumulation during photocatalytic oxida-
tion of ethanol as evident from the result of the background
shift. The variation of the IR intensity at n˜ =2000 cmꢁ1 (I2000
,
background shift) during 60 min of ethanol photocatalytic oxi-
dation on P25-TiO2 powder and T600 nanofibers is shown in
Figure 4c and g. The variation of I2000 is a manifestation of the
accumulation or depletion of the photogenerated electrons
within the conduction band of TiO2.[41,59,77] The accumulation of
photogenerated electrons proceeds if the rate of the electron-
generating reaction in Equation (2) is greater than the rate of
electron-consuming reactions in Equations (5) and (6) and the
rate of e-h recombination in Equation (4). The I2000 profile for
P25 TiO2 powder exhibited a decrease for 300 s followed by
gradual recovery to nearly its initial intensity level. The de-
crease in the intensity of I2000 indicates that the photogenerat-
ed electrons were either recombined with the holes or were
transferred to oxygen and generated O2ꢁ. Acetate and formate
were formed as the oxygen-containing species during this
period as evident from the evolution of IR bands at n˜ =1442
and 1581 cmꢁ1 in Figure 4a and e. As a result of the inability
of h+ to diffuse away from the TiO2 surface, the acetate and
formate species were more likely formed and accumulated on
the h+-generation sites on P25 TiO2, which blocks them for e-h
recombination. This resulted in an accumulation of photogen-
erated electrons within the TiO2 conduction band of P25 TiO2
powder as evident from the increase in the I2000 profile after
300 s of photocatalytic oxidation reaction.
Figure 6. PL emission spectra of T600 and reference P25 TiO2 nanoparticles
under the excitation at l=325 nm at room temperature.
The lower recombination rate of electrons and holes en-
hanced the photocatalytic oxidation kinetics. The charge or in-
terfacial electron transfer process in T600 nanofibers is the pri-
mary reason for the decrease of the e-h recombination rate
during photoexcitation (Figure S12).[70–72] The experimentally
evaluated band-gap energy of the anatase phase (ꢂ3.23 eV) is
higher than that of the rutile phase (ꢂ2.95 eV). T600 con-
tained an anatase to rutile mass ratio of 64:36, whereas no
rutile phase was identified for T500 and ~96% of the crystals
were in the rutile form in T700. This can explain why the T600
nanofibers exhibited a higher rate for the photocatalytic oxida-
tion of ethanol than T500 and T700.[73] The rutile phase of TiO2
provides a lower photoactivity than the anatase analogue.[74]
Therefore, it is no surprise that T700 showed a lower photoac-
tivity than T600. The T500 nanofibers showed lower photoac-
tivity than T600 as no lower-band-gap rutile-phase material
was present and higher amounts of carbon lowered its photo-
catalytic performance.[75]
The surface area of a photocatalyst is an important factor to
influence the rate of photo-oxidation. In this context, the
photo-oxidation experiments in this work were performed by
However, on T600 nanofibers, the availability of surface hy-
C
droxyl groups indicated that the OH-initiating oxidation path-
&
ChemCatChem 2016, 8, 1 – 12
www.chemcatchem.org
8
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
way (pathway A in Figure 5) appeared to dominate the ethanol
photocatalytic oxidation until a certain accumulation of for-
mate and acetate occurred. The XPS data of T600 nanofibers
rutile phases should be present at a specific ratio to obtain an
enhanced rate for the photo-oxidation of ethanol. The kinetic
C
data indicate that the reaction steps shifted from the OH-initi-
ꢁ
and P25-TiO2 powder shows IOH/IOxide(T600) =0.28 and IOH
/
ated oxidation to the O2 -initiated oxidation of CH3CH2OH mol-
I
Oxide(P25) =0.22 (Table 2). As the formate and acetate accumulat-
ecules, which resulted in the formation of CO2 and H2O after
10 min of reaction, whereas the hole-initiating oxidation of
CH3CH2OH took place on P25 TiO2 nanoparticle surfaces and
caused a slow conversion into CO2 and H2O.
ed, the adsorption of O2 and the transfer of electrons to ad-
sorbed oxygen occurred. This alleviated electron accumulation,
depleted the accumulated acetate and formate, and eventually
converted them to CO2 and H2O.
The specific details of the reaction kinetics of photo-oxida-
tion of ethanol are beyond the scope of this study and will be
presented in a separate communication. The data indicate that
ethanol degradation followed different pathways on nanofib-
ers and on P25 TiO2 nanoparticles. For example, the reaction
Experimental Section
Materials: PVP (Mw =1300000 gmolꢁ1; Sigma Aldrich) was chosen
as the fiber forming polymer. TTIP (>97% purity; Sigma Aldrich)
was used as a precursor material for TiO2. Acetic acid purchased
from Fischer Scientific was used as a chelating agent to maintain
the viscosity of the precursor spinning solution. Ethanol (>99.5%
pure) was used as a solvent to dissolve PVP. All materials were
used as received without further purification. TiO2 nanoparticles in
powder form as P25 TiO2 (>99.5%) was purchased from Sigma–Al-
drich.
C
sequences in the case of T600 nanofibers shifted the OH-initi-
ꢁ
ated photocatalytic oxidation to the O2 -initiated photocatalyt-
ic oxidation, which thus leads to CO2 and H2O formation with-
out the accumulation of the acetate and formate intermedi-
ates. This is also reflected in the amount of CO2 generated
with time, for example, as shown in Figure 4d and h much
more CO2 was generated in the case of T600 nanofibers com-
pared to P25 TiO2 powder. A similar behavior was observed for
T500 and T700 nanofibers (Figure S13). However, the hole-ini-
tiating oxidation of ethanol occurred on the surfaces of P25
TiO2 nanoparticles resulted in a slower conversion of ethanol
to CO2 and H2O.
Preparation of PVP-TTIP spinning solution: Typically, TTIP (3 g)
was mixed with of ethanol/acetic acid solution (6 mL, 1:1 v/v) and
stirred magnetically for 30 min at RT for homogenization. A
~7.5 wt% PVP solution was prepared by stirring PVP (0.45 g) in an-
hydrous ethanol (7 mL) for 10 min by using a Thinkyꢂ mixer. The
two solutions were mixed together in an air-tight container and
stirred magnetically for 1 h at RT to obtain a homogeneous spinna-
ble sol solution for use in the GJF spinning process.
The significant enhancement of the photocatalytic activity of
TiO2 nanofibers can now be attributed to their unique mor-
phology that allows the transfer of photogenerated electrons
to O2 and prevents the accumulation of intermediate species
such as acetate and formate. Therefore, we infer that the fi-
brous morphology of TiO2 plays a critical role in the enhance-
ment of the rate of photocatalytic oxidation of ethanol. This
study reveals that the charge separation (e-h) and its transpor-
tation on the surface of TiO2 is the dominant factor to deter-
mine the photocatalytic activity in the order of T600@T700~
T500>P25.
Fabrication of TiO2 nanofibers: The spinning sol solution was
taken into a 10 mL HSW plastic syringe, which was linked with
a 19 gauge needle JG19X-0.5 (Jensen Global) with an internal di-
ameter of 1.08 mm. The sol solution was injected with a flow rate
of 0.3 mLminꢁ1 by using a syringe pump (Model-Fusion 200) for
the preparation of fiber samples used in this study. The spinning
experiment was performed with 20 psi air pressure (or volumetric
air flow rate of 0.1339 m3 minꢁ1 at 208C; Figure 1a–c) at RT (~24ꢀ
28C) and a relative humidity of ~30–35% unless otherwise men-
tioned. Air was supplied through a nozzle with an inner diameter
of 1.2 cm. The distance between the nozzle and the needle tip was
~1.5 cm. A detailed description of the GJF spinning process is
available elsewhere.[33,36] The precursor polymer fibers were collect-
ed on a piece of flat nylon sieve kept ~1.8–2.0 m from the tip of
the needle. The GJF-spun fiber mat was kept in air for ~2 h at RT
to complete the hydrolysis process. The precursor fiber mat was
then placed in a muffle furnace (Model: Vulcan 3–130) for pyrolysis
reactions in air at maximum temperatures of 500, 600, or 7008C.
The pyrolysis was performed in several stages for each maximum
temperature. In the case of 6008C, first, the temperature was in-
creased from RT to 1008C at 0.58Cminꢁ1 and kept at 1008C for
1 h. Second, the temperature was raised to 6008C at 0.58Cminꢁ1
rate and maintained at 6008C for 2 h to burn off the organic com-
ponents in the fibers and to allow TiO2 to crystallize. Finally, the
material was allowed to cool naturally to RT. The TiO2 nanofibers
thus prepared are designated as T500, T600, and T700 to refer to
a maximum calcination temperature of 500, 600, and 7008C, re-
spectively.
Conclusion
We demonstrated the use of a facile gas jet fiber spinning pro-
cess in the fabrication of mesoporous TiO2 nanofibers of a con-
trolled diameter using a template polymer (polyvinylpyrroli-
done) and the sol–gel chemistry of the precursor material (tita-
nium isopropoxide). We also performed a systematic investiga-
tion of the role of the calcination temperature on the morphol-
ogy, crystal structure, and surface area of the TiO2 nanofibers
and the roles of anatase versus rutile content, fibrous morphol-
ogy, and surface area on the photocatalytic performance.
Among the nanofibers studied, the T600 nanofibers showed
an order of magnitude higher rate for the photocatalytic oxida-
tion of gas-phase ethanol under UV light compared to a com-
mercial-grade nanoparticles. The unique organization of nano-
crystallites in the fibrous networks facilitated photogenerated
charge (electron–hole) separation and subsequent oxidation
process by the faster transport and lower recombination rate
of the separated charges. This study revealed that anatase and
Photocatalytic oxidation of ethanol on TiO2 nanofibers: Photoca-
talytic oxidation experiments were conducted by using a diffuse
reflectance infrared Fourier transform (DRIFT) cell (Harrick scientific
HVC-DRP) enclosed by a dome with two IR transparent ZnSe win-
dows and a CaF2 window for UV illumination, a 350 W mercury
ChemCatChem 2016, 8, 1 – 12
www.chemcatchem.org
9
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
lamp (Oriel 6286) with a light condenser (Oriel 77800), and a gas
manifold for the admission of ultra-high purity O2 (Praxair,
99.999%) and organic vapors. A schematic of the experimental set-
up for the photo-oxidation of ethanol is presented in Figure S10.
During each experiment, a thin 5 mg layer of TiO2 was placed in
the sample holder of the DRIFT cell. The thin layer of TiO2 powders
was used to minimize the changes in IR intensity caused by species
adsorbed on the dark fractions of the catalyst (i.e., not illuminated).
Ethanol vapors are adsorbed on TiO2 at 303 K by exposing the TiO2
samples in the DRIFT cell to an ethanol-saturated O2 stream
(15 sccm) for 10 min and allowing the stream to flow for 10 min
before adjusting the valves in the DRIFT cell to batch mode. The
photocatalytic oxidation reactions were initiated by the UV illumi-
nation of TiO2, and the DRIFT cell was operated in batch mode at
303 K and 1 atm O2 atmosphere. The intensity of the UV light in
each experiment was adjusted to 4 by using a thermopile detector
(818P-001-12, Newport) connected to a hand-held optical meter
(1918-C, Newport). Single-beam IR spectra were recorded by using
a FTIR spectrometer (Thermo Scientific Nicolet 6700) coadding
isotherm data using nonlocal density functional theory (NLDFT). All
BET samples were degassed in vacuum at 858C for 18 h before the
experiment. Powder XRD (Bruker) with Cu-K radiation (l=
0.15418 nm) of all TiO2 samples was performed to derive crystallo-
graphic information and to calculate the percentage of anatase
and rutile phases in each sample. UV/Vis absorption diffuse reflec-
tance and UV/Vis absorption spectra were recorded by using
a spectrophotometer (U-3900 Spectrophotometer) to determine
the band-gap energies of the TiO2 samples. A scanning XPS mi-
croprobe PHI5000VersaProbe II (ULVAC-PHI Inc.) was utilized to
characterize the surface chemical states and the composition of
the TiO2 nanofibers and P25 TiO2 particles. The XPS scan was per-
formed from 0 to 700 eV without any surface coating of the sam-
ples. Binding energies were calibrated relative to the C1s peak
(BE=284.5 eV) to obtain the XPS data. PL spectra of the samples
were obtained by using a spectrophotometer (Horiba Jobin Yvon
Fluoroman-4) at an excitation wavelength of 325 nm at RT. The
skeletal density of the calcined fibers was measured by using a He-
pycnometer (Accupyc 1340 Helium Pycnometer, Micromeritics In-
strument Corp.).
20 scans at a resolution of 4 cmꢁ1
.
Single-beam IR spectra of TiO2 were recorded at 303 K: (i) before
the adsorption of organics and UV illumination (i.e., TiO2 surface
free from adsorbates), (ii) after exposure to organics, and (iii) during
UV illumination. IR spectra were obtained in the form of absorb-
ance (A=log(Io/ISB)), in which ISB represents the single-beam spec-
trum of interest and I0 is an appropriate reference single-beam
spectrum. The absorbance spectra magnify the intensity of each IR
feature showing positive (i.e, concave) and negative (i.e., convex)
bands that indicate the formation and disappearance of species,
respectively, on the TiO2 surface.
Acknowledgments
The authors thank T. N. Burai for help with PL measurements.
Keywords: IR spectroscopy · nanostructures · oxidation ·
photochemistry · titanium
[1] A. P. Jones, Atmos. Environ. 1999, 33, 4535.
[2] J. Mo, Y. Zhang, Q. Xu, J. J. Lamson, R. Zhao, Atmos. Environ. 2009, 43,
2229.
[3] D. T. Tompkins, W. A. Zeltner, B. J. Lawnicki, M. A. Amderson, ASHRAE
Trans. 2005, 111, 60.
[4] S. Wang, H. M. Ang, M. O. Tade, Environ. Int. 2007, 33, 694.
[5] X. Chen, S. S. Mao, Chem. Rev. 2007, 107, 2891.
[6] H. Wu, J. Ma, C. Zhang, H. He, J. Environ. Sci. 2014, 26, 673.
[7] D. S. Muggli, J. T. McCue, J. L. Falconer, J. Catal. 1998, 173, 470.
[8] T. Kawahara, Y. Konishi, H. Tada, N. Tohge, J. Nishii, S. Ito, Angew. Chem.
Int. Ed. 2002, 41, 2811; Angew. Chem. 2002, 114, 2935.
[9] S. A. Larson, J. A. Widegren, J. L. Falconer, J. Catal. 1995, 157, 611.
[10] R. I. Bickley, T. Gonzalez-Carreno, J. S. Lees, L. Palmisano, R. J. D. Tilley, J.
Solid State Chem. 1991, 92, 178.
The initial coverage of ethanol on the TiO2 catalyst before the pho-
tocatalytic reactions was quantified by using temperature-pro-
grammed oxidation (TPO). In each TPO experiment, the DRIFT cell
was loaded with a TiO2 thin layer sample and the ethanol, heated
in batch mode from 30 to 5008C at a heating rate of 108Cminꢁ1
and then held at 5008C for 20 min to facilitate complete oxidation
of the organic reactant to CO2. The total amount of CO2 produced
during TPO was determined by flushing the gas contents of the
DRIFT cell with an O2 stream (15 sccm) into a downstream mass
spectrometer (QMS 200, Pfeiffer Omnistar), recording the intensity
of the mass to charge ratios unique to CO2 (m/e=44). The re-
sponse of the mass spectrometer to CO2 was calibrated by expo-
sure to known amounts of CO2.
[11] D. C. Hurum, A. G. Agrios, K. A. Gray, T. Rajh, M. C. Thurnauer, J. Phys.
Chem. B 2003, 107, 4545.
Materials characterization: The morphology and surface topogra-
phy of nanofibers were inferred from SEM images (JEOL JSM-
7401F) recorded at an accelerating voltage 5–10 kV, emission cur-
rents between 10–20 mA, and magnification in the range 500–
100000ꢁ. The fiber diameter distribution and average diameter
was obtained from measurement of at least 80 different fibers for
each sample using Image-J software (Image J, NIH, USA). TEM
(JEOL JSM-1230) was used to study the crystalline surface morphol-
ogy and the size of grain of the calcined nanofibers. Thermogravi-
metric analysis (TGA; Q200 from TA Instruments) was performed at
a heating rate of 108Cminꢁ1 from RT to 7008C using air flow of
60 mLminꢁ1 to determine the thermal stability and calcination con-
ditions of the polymer precursor fibers. BET surface area analysis
and pore size distribution of TiO2 nanofibers and P25 TiO2 powder
were performed by using N2 adsorption–desorption isotherms
(TriStar II, Micromeritics) at ꢂ77 K. Furthermore, BET adsorption–
desorption at 08C were recorded with CO2 gas for the characteriza-
tion of the surface area of micropores of diameter less than 2 nm.
The micropore size distribution was obtained from CO2-sorption
[12] D. C. Hurum, K. A. Gray, T. Rajh, M. C. Thurnauer, J. Phys. Chem. B 2005,
109, 977.
[13] S. Klosek, D. Raftery, J. Phys. Chem. B 2001, 105, 2815.
[14] W. Choi, A. Termin, M. R. Hoffmann, J. Phys. Chem. 1994, 98, 13669.
[15] S. T. Martin, C. L. Morrison, M. R. Hoffmann, J. Phys. Chem. 1994, 98,
13695.
[16] J.-M. Herrmann, J. Disdier, P. Pichat, Chem. Phys. Lett. 1984, 108, 618.
[17] M. Shen, Z. Wu, H. Huang, Y. Du, Z. Zou, P. Yang, Mater. Lett. 2006, 60,
693.
[18] K. Nakata, A. Fujishima, J. Photochem. Photobiol. C 2012, 13, 169.
[19] E. Stathatos, D. Tsiourvas, P. Lianos, Colloids Surf. A 1999, 149, 49.
[20] M. Bognitzki, H. Hou, M. Ishaque, T. Frese, M. Hellwig, C. Schwarte, A.
Schaper, J. H. Wendorff, A. Greiner, Adv. Mater. 2000, 12, 637.
[21] T. Kasuga, M. Hiramatsu, A. Hoson, T. Sekino, K. Niihara, Langmuir 1998,
14, 3160.
[22] P. Hoyer, Langmuir 1996, 12, 1411.
[23] D. Gong, C. A. Grimes, O. K. Varghese, W. Hu, R. S. Singh, Z. Chen, E. C.
Dickey, J. Mater. Res. 2001, 16, 3331.
[24] H.-H. Ou, S.-L. Lo, Sep. Purif. Technol. 2007, 58, 179.
[25] X. Zhang, V. Thavasi, S. G. Mhaisalkar, S. Ramakrishna, Nanoscale 2012,
4, 1707.
&
ChemCatChem 2016, 8, 1 – 12
www.chemcatchem.org
10
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
[26] J. Fu, S. Cao, J. Yu, J. Low, Y. Lei, Dalton Trans. 2014, 43, 9158.
[27] X. Peng, A. C. Santulli, E. Sutter, S. S. Wong, Chem. Sci. 2012, 3, 1262.
[28] Z. Ma, W. Chen, Z. Hu, X. Pan, M. Peng, G. Dong, S. Zhou, Q. Zhang, Z.
Yang, J. Qiu, ACS Appl. Mater. Interfaces 2013, 5, 7527.
[29] C. H. Kim, B.-H. Kim, K. S. Yang, Carbon 2012, 50, 2472.
[30] D. Li, Y. Xia, Nano Lett. 2003, 3, 555.
[31] J. Mitra, M. Ghosh, R. K. Bordia, A. Sharma, RSC Adv. 2013, 3, 7591.
[32] K. Sarkar, C. Gomez, S. Zambrano, M. Ramirez, E. de Hoyos, H. Vasquez,
K. Lozano, Mater. Today 2010, 13, 12.
[54] Z. Yu, S. S. C. Chuang, Appl. Catal. B 2008, 83, 277.
[55] D. J. C. Yates, J. Phys. Chem. 1961, 65, 746.
[56] K. I. Hadjiivanov, D. G. Klissurski, Chem. Soc. Rev. 1996, 25, 61.
[57] Y. Borodko, H. S. Lee, S. H. Joo, Y. Zhang, G. Somorjai, J. Phys. Chem. C
2010, 114, 1117.
[58] D. A. Panayotov, J. T. Yates, Jr., Chem. Phys. Lett. 2005, 410, 11.
[59] A. Yamakata, T.-a. Ishibashi, H. Onishi, J. Mol. Catal. A 2003, 199, 85.
[60] S. H. Szczepankiewicz, J. A. Moss, M. R. Hoffmann, J. Phys. Chem. B 2002,
106, 7654.
[33] R. E. Benavides, S. C. Jana, D. H. Reneker, ACS Macro Lett. 2012, 1, 1032.
[34] M. Ghosh, S. C. Jana, RSC Adv. 2015, 5, 105313.
[35] J.-H. Shang, R. Benavides, S. Jana, Int. Polym. Process. 2014, 29, 103.
[36] R. E. Benavides, S. C. Jana, D. H. Reneker, Macromolecules 2013, 46,
6081.
[61] G. Ramis, G. Busca, V. Lorenzelli, J. Chem. Soc. Faraday Trans. 1 1987, 83,
1591.
[62] M. J. Backes, A. C. Lukaski, D. S. Muggli, Appl. Catal. B 2005, 61, 21.
[63] J. Peral, D. F. Ollis, J. Mol. Catal. A 1997, 115, 347.
[64] M. C. Lee, W. Choi, J. Phys. Chem. B 2002, 106, 11818.
[65] F. Arsac, D. Bianchi, J. M. Chovelon, C. Ferronato, J. M. Herrmann, J.
Phys. Chem. A 2006, 110, 4213.
[37] T. Luttrell, S. Halpegamage, J. Tao, A. Kramer, E. Sutter, M. Batzill, Sci.
Rep. 2014, 4, 4043.
[38] Y.-F. Chen, C.-Y. Lee, M.-Y. Yeng, H.-T. Chiu, J. Cryst. Growth 2003, 247,
363.
[66] J. M. Coronado, S. Kataoka, I. Tejedor-Tejedor, M. A. Anderson, J. Catal.
2003, 219, 219.
[39] J. Xie, J. Y. Lee, D. I. C. Wang, Chem. Mater. 2007, 19, 2823.
[40] Z. Yu, S. S. C. Chuang, J. Catal. 2007, 246, 118.
[67] Y. Zhang, N. Zhang, Z.-R. Tang, Y.-J. Xu, Phys. Chem. Chem. Phys. 2012,
14, 9167.
[41] F. Guzman, S. S. C. Chuang, J. Am. Chem. Soc. 2010, 132, 1502.
[42] A. H. M. Gonzalez, A. Z. Simoes, M. A. Zaghete, E. Longo, J. A. Varela, J.
Electroceram. 2004, 13, 353.
[43] S. Chuangchote, J. Jitputti, T. Sagawa, S. Yoshikawa, ACS Appl. Mater. In-
terfaces 2009, 1, 1140.
[44] J. F. Porter, Y.-G. Li, C. K. Chan, J. Mater. Sci. 1999, 34, 1523.
[45] R. A. Spurr, H. Myers, Anal. Chem. 1957, 29, 760.
[46] T. J. Barton, L. M. Bull, W. G. Klemperer, D. A. Loy, B. McEnaney, M.
Misono, P. A. Monson, G. Pez, G. W. Scherer, J. C. Vartuli, O. M. Yaghi,
Chem. Mater. 1999, 11, 2633.
[47] X. Wang, J. Choi, D. R. G. Mitchell, Y. B. Truong, I. L. Kyratzis, R. A. Caruso,
ChemCatChem 2013, 5, 2646.
[48] A. Navrotsky, O. J. Kleppa, J. Am. Ceram. Soc. 1967, 50, 626.
[49] H. Choi, A. C. Sofranko, D. D. Dionysiou, Adv. Funct. Mater. 2006, 16,
1067.
[68] S. K. Choi, S. Kim, S. K. Lim, H. Park, J. Phys. Chem. C 2010, 114, 16475.
[69] J. Liu, J. Li, A. Sedhain, J. Lin, H. Jiang, J. Phys. Chem. C 2008, 112, 17127.
[70] Z. Zhang, C. Shao, X. Li, Y. Sun, M. Zhang, J. Mu, P. Zhang, Z. Guo, Y. Liu,
Nanoscale 2013, 5, 606.
[71] Z. Zhang, Y. Huang, K. Liu, L. Guo, Q. Yuan, B. Dong, Adv. Mater. 2015,
27, 5906.
[72] Z. Zhang, K. Liu, Z. Feng, Y. Bao, B. Dong, Sci. Rep. 2016, 6, 19221.
[73] S. Shen, X. Wang, T. Chen, Z. Feng, C. Li, J. Phys. Chem. C 2014, 118,
12661.
[74] N. G. Park, J. van de Lagemaat, A. J. Frank, J. Phys. Chem. B 2000, 104,
8989.
[75] F. Teng, G. Zhang, Y. Wang, C. Gao, L. Chen, P. Zhang, Z. Zhang, E. Xie,
Appl. Surf. Sci. 2014, 320, 703.
[76] Y. T. Liang, B. K. Vijayan, K. A. Gray, M. C. Hersam, Nano Lett. 2011, 11,
2865.
[50] F. Mou, J.-g. Guan, W. Shi, Z. Sun, S. Wang, Langmuir 2010, 26, 15580.
[51] B. Erdem, R. A. Hunsicker, G. W. Simmons, E. D. Sudol, V. L. Dimonie,
M. S. El-Aasser, Langmuir 2001, 17, 2664.
[77] A. Yamakata, T.-a. Ishibashi, H. Onishi, J. Phys. Chem. B 2002, 106, 9122.
Received: April 1, 2016
Revised: May 4, 2016
[52] Z. Yang, G. Du, Q. Meng, Z. Guo, X. Yu, Z. Chen, T. Guo, R. Zeng, J.
Mater. Chem. 2012, 22, 5848.
[53] J. Yu, H. Yu, B. Cheng, M. Zhou, X. Zhao, J. Mol. Catal. A 2006, 253, 112.
Published online on && &&, 0000
ChemCatChem 2016, 8, 1 – 12
www.chemcatchem.org
11
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

FULL PAPERS
M. Ghosh, M. Lohrasbi, S. S. C. Chuang,
S. C. Jana*
Photo finish: Mesoporous TiO2 nanofib-
ers with controlled diameter, crystal
size, and anatase versus rutile crystal
structures are produced by using a scala-
ble gas jet fiber spinning process. These
TiO2 nanofibers are used in the photoca-
talytic oxidation of gas-phase ethanol at
room temperature on exposure to UV ir-
radiation.
&& – &&
Mesoporous Titanium Dioxide
Nanofibers with a Significantly
Enhanced Photocatalytic Activity
&
ChemCatChem 2016, 8, 1 – 12
www.chemcatchem.org
12
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!