-
April 25, 2017
Dr. Rainer Glaser, Professor of Chemistry
Editor, Inorganic Chemistry
Department of Chemistry, University of Missouri-Columbia
Columbia, MO 65211
RE: REVISED
Carbon Capture, Utilization, and Storage. A New Method for
Carbon Dioxide
Photoreduction Catalyzed by Oxygen Deficient TiO2 Surfaces By:
Melissa Hopkins, Parker Smith, and Ethan Zars
Dear Dr. Glaser,
Thank you for your comments on April 20 and the inclusion of the
peer reviews on our work. We
used their suggestions as guidance for some necessary revisions.
Based on reviewer
recommendation, we have made the following changes:
Major Revisons
[M.1] Methods section is more concise with elimination of some
characterization
descriptions.
[M.2] Discussion is made clearer by reworking of sentences and
increased discussion of
Lewis acid and base theory.
[M.3] Top part of Figure 1 was corrected to match the figure
legend.
[M.4] Scheme 2 is corrected with labeling for bidentated
carbonates.
Melissa Hopkins
Email: [email protected]
Parker Smith
Email: [email protected]
Ethan Zars Email: [email protected]
Department of Chemistry
University of Missouri-Columbia
105 Chemistry Building
601 S. College Avenue
Columbia, MO 65211
USA
mailto:[email protected]:[email protected]:[email protected]
-
2
Response to Reviewer 1
[1.1] See M.3 and M.4.
Response to Reviewer 2
[2.1] See M.2
[2.2] Addition of a few sentences describing how the mechanism
of reduced
intermediates to the products is unknown.
[2.3] Binding structures for the different doped surfaces are
not given in the cited
references.
[2.4] See M.3 and M.4.
Response to Reviewer 3
[3.1] See M.1.
[3.2] Table 1 gives rates of reduction and mass efficiencies for
the discussed catalysts.
[3.3] We don’t find that the addition of color would help convey
the message more
clearly.
Sincerely,
Melissa, Parker, and Ethan
-
1
Carbon Capture, Utilization, and Storage. A New Method for
Carbon Dioxide
Photoreduction Catalyzed by Oxygen Deficient TiO2 Surfaces.
Melissa Hopkins, Parker Smith, and Ethan Zars
Department of Chemistry, University of Missouri-Columbia,
Columbia, MO, USA 65201
Email: [email protected]; [email protected];
[email protected]
mailto:[email protected]:[email protected]:[email protected]
-
2
Abstract
The photocatalytic reduction of carbon dioxide, a known
greenhouse gas, utilizes
properties of semiconducting materials in the formation of
products such as carbon
monoxide and methane. Titanium dioxide (TiO2) surfaces are the
most commonly
reported materials used due to the effective band gap of
molecular orbitals and absorption
in the near-visible region of the electromagnetic spectrum,
allowing for activation via
sunlight irradiation. TiO2 has three polymorphs of anatase,
rutile, and brookite which
have varying properties with regard to the formation of oxygen
vacancies that increase
efficacy. Additionally, TiO2 surfaces can become more effective
catalysts through the
use of dopants such as Ce, Cu and Pd lattices, and Pt, though
cost efficiencies factor into
the feasibility of such compounds for commercial usage.
-
3
Introduction
One of the most pressing issues today is global climate change,
primarily caused
by anthropogenic greenhouse gases.1 One anthropogenic greenhouse
gas that is of
particular concern is carbon dioxide (CO2) because of its role
in global climate change.2
It is, however, possible to remove and utilize CO2 from the
atmosphere by various
capture, utilization, and storage methods.3 The main challenge
that needs to be overcome
in CO2 utilization is the reduction of this highly stable, but
also completely oxidized
molecule, to products such as carbon monoxide (CO) and methane
(CH4) which can be
more easily utilized as chemical feedstock or fuels.
Many sources of energy are available for the reduction of CO2,
but the process
can only contribute to a decrease in global CO2 levels if the
energy source is renewable
and does not emit CO2. Solar radiation is a particularly
attractive energy source because
it is abundant, renewable, and powerful enough to effectively
reduce CO2 to CO and CH4
among other products.4 Many methods for photocatalytic reduction
of CO2 have been
reported and most involve various forms of TiO2 surfaces in the
three main phases: rutile,
brookite, and anatase. TiO2 surfaces are effective reducers of
CO2 because they form
semiconductors with a HOMO/LUMO bandgap large enough reduce CO2
and almost
small enough to absorb in the visible region of the
electromagnetic spectrum.4 For the
industrial use of TiO2 as a photocatalyst in CO2 reduction, the
cost of the catalyst must
remain low, but the reduction of CO2 must still be effective.
Luckily, anatase-phase TiO2
can effectively reduce CO2, and this activity is enhanced by
partial hydration of the
anatase TiO2 surface.5 Furthermore, creating oxygen vacancies in
brookite-phase TiO2
allows for the photocatalytic reduction of CO2 to CO and CH4
products. 6
-
4
In this paper we will compare the cost7 efficiencies of
brookite-phase TiO2
photocatalysts with oxygen deficiencies, TiO2 photocatalysts
doped with Ce,8 Pd and
Cu,9 and Pt10 (Scheme 1) and we find that the Pd and Cu doped
TiO2 photocatalysts are
the most effective photocatalysts by mass, but Pd is extremely
expensive. When the cost
efficiencies are compared, the brookite-phase TiO2 with oxygen
vacancies proves to be
the most cost efficient.
Scheme 1. Representation of the Four TiO2 Surfaces Tested.
Materials and Methods
Preparation of TiO2 Photocatalyst
TiO2 crystals were synthesized by hydrolysis and hydrothermal
methods to form
their three phases: rutile, anatase, and brookite. The rutile
phase was synthesized by
combining TiCl4 with ethanol and water. The solution was stirred
and then kept in an
oven for 24 hours. The white precipitate was collected, washed
with water, and dried in
an oven. Anatase was synthesized by combining aqueous
Ti(NH3Lac)2·(OH)2 with urea
and then diluting the solution in water. This solution was then
sealed in an autoclave and
placed in an oven at 160 ºC for 24 hours. Next the precipitate
was collected, washed with
water, and dried overnight in an oven at 60 ºC. Brookite was
prepared following the
same method as rutile except using a much higher concentration
of urea. All three
powders were calcined at 400 ºC. To prepare oxygen-deficient
TiO2 crystals the as-
prepared TiO2 crystals were treated with helium gas at an
elevated temperature.
-
5
Characterization of TiO2 Crystals
X-ray Diffraction (XRD), Raman, Transmission electron
spectroscopy (TEM),
UV-Vis, and DRIFTS-IR spectroscopy were used to characterize the
TiO2 surfaces.
XRD (Figure S1) is a useful analytical technique for the
characterization and
identification of crystals because it produces data that is
unique to certain materials and
those data are available for comparison in a database.11
Raman spectroscopy (Figure S1) was also used to characterize and
identify TiO2
polymorphs. The spacing between Raman absorbance peaks will
always be the rotational
constant multiplied by four (except in the case of symmetric
rotors) and therefore the
bond length or reduced mass of the bonding atoms can be
determined. Raman
spectroscopy is used to identify TiO2 crystal phases because
Raman spectroscopy only
detects vibrational or rotational modes that result in a change
of polarizability of the
bond. The different phases of TiO2 will have a unique set of
Raman active modes
(brookite has 36 active modes, anatase has 5, and rutile has 3)
that can be used to identify
and characterize the crystal structures.
Transmission electron microscopy (TEM) is a further technique
that is used to
identify and characterize TiO2 (Figure S2) and is especially
useful because it allows the
user to visualize the crystal structure. TEM allows for finer
resolution of small objects
than traditional visible light microscopy because the size of
objects that can be discerned
is limited by the wavelength of the radiation interacting with
them. The wavelength
associated with electrons transmitted through a sample is
smaller than the wavelengths of
visible light.
While the above techniques allow for the characterization and
identification of the
gross crystal structure, they are not sufficient for detecting
oxygen vacancies or other
-
6
defects in the crystals. UV-Vis spectroscopy was used to detect
oxygen vacancies in the
TiO2 polymorphs (Figure S3). In TiO2 crystals with oxygen
vacancies there will be an
increased absorption in lower frequency radiation (as compared
to TiO2 crystals without
oxygen vacancies) because these oxygen vacancies have compressed
and inconsistent
electron energy level gaps.
Diffuse reflectance infrared Fourier transform spectroscopy
(DRIFTS) is used to
characterize the roles of oxygen vacancies (VO) and Ti3+ sites
in He-treated and
unpretreated surfaces (Figure S4). DRIFTS studies are similar to
IR spectral studies, but
more specific detail can be observed in the peaks due to greater
sensitivity. Light is
passed through a non-absorbing material, such as KBr, then
interacts with the sample by
reflecting within the crystal structure. The result is a linear
relationship between spectral
intensity and sample concentration thus giving more specificity
than a regular IR spectra.
Performance Evaluation by DRIFTS-IR Spectroscopy and Gas
Chromatography
DRIFTS-IR spectroscopy can be used to track changes in amounts
of certain
molecules by observing intensity changes of the characteristic
peaks over time.
Following Beer’s Law, an increase in absorbance at certain
characteristic wavenumbers
corresponds to an increased concentration of molecules with that
characteristic absorption
frequency. In order to measure the effectiveness of a catalyst
system the disappearance
of the H2O absorption peak at 1640 cm-1 and the appearance of
the HCO3- peak at 1420
cm-1 over time need to be considered. Differences in the
absorbances at these
wavenumbers will be directly proportional to the intermediates
and reactants they
represent and will be related to CO and CH4 production which is
exemplified in Figure 1.
CO2 gas was passed through a water bubbler to become loaded with
H2O vapor
and then allowed to pass over the photocatalyst under a 150 W
lamp. For each test 100
-
7
mg of TiO2 was used. Identity and concentration determination of
the product gases was
performed using gas chromatography using both a thermal
conductivity detector and a
flame ionization detector. The gas chromatograms were compared
to standard eluents to
identify and quantify the products. The production of CO and CH4
product gases was
calculated over a 6 hour photolumination period. These gas
chromatography data yield
CO and CH4 production in units of µmol·g-1 (µmol product per
gram of catalyst). In
order to compare the cost efficiency of different catalyst
systems the product yield needs
to be scaled to one hour of reaction time and then the price of
the catalyst system ($·kg-1)
is divided by the product yield per hour. The cost12
efficiencies and performance data of
TiO2 photocatalysts with and without oxygen deficiencies, TiO2
photocatalysts doped
with Ce (0.28 mol %),8 Pd and Cu (11.5 mol % and 2.27 mol %,
respectively),9 and Pt (1
wt %)10 are shown in Table 1.
-
8
Table 1. Cost and Mass Efficiencies of Four Selected
Photocatalyst Systems.
Catalyst Co-Catalyst Co-catalyst
abundance
Yield of CH4
per hour
(μmolg-1h-1)
Price of
reagents per
kg catalyst
($·kg-1)
Price of
reagents per
kg catalyst
per yield of
CH4 ($·g·h·
μmol-1·kg-1)
TiO2 - - 0.57 57.00 100
TiO2 Ce 0.28 mol % 0.89 338.43 380.26
TiO2 Pd and Cu
11.5 mol %
Pd
2.27 mol %
Cu
10.00 22,847.46 2,284.75
TiO2 Pt 1 wt % 2.65 1,978.47 746.59
-
9
Figure 1. DRIFTS-IR spectra of CO2 and H2O interaction with (a,
top) unpretreated
anatase (TiA(UP)) and (b, top) brookite (TiB(UP)), (a, bottom)
He-treated anatase
(TiA(He)) and (b, bottom) brookite (TiB(UP)) irradiated by
radiation in the UV and
visible regions as a function of time. The light was turned on
after the samples sat in the
dark for 15 min.
-
10
Results
All the XRD (Figure S1) intensity peaks corresponded well to the
database11
values. Raman spectra (Figure S1) also show the characteristic
Raman modes of
anatase, rutile, and brookite. Brookite has 36 possible Raman
active modes and in the
spectrum many peaks are visible. Rutile shows one small Raman
absorbance peak at 234
cm-1 and two larger peaks at 446.5 and 608.5 cm-1 corresponding
to the three possible
Raman active modes of rutile. Anatase shows Raman absorbance
peaks at 142.4, 192.5,
393.6, 514.2, and 638.1 cm-1 corresponding to the five Raman
modes of anatase. TEM
also was used to characterize the TiO2 crystals. Only the
pretreated samples were
analyzed by TEM because the XRD and Raman data showed no
significant changes
following helium treatment. The TEM images (Figure S2) for the
anatase crystals show
an interplanar spacing of approximately 0.350 nm and irregularly
shaped hexagons,
circles, or rectangles with an average size of about 9 nm. A
smaller interplanar spacing
of 0.322 nm and short nanoellipses and elliptical rods are
observed for the rutile phase
crystals. The brookite crystals exhibit randomly dispersed
nanorods with a length of
approximately 100 nm and diameter of 20 nm. The interplanar
spacing of brookite is
0.544 nm and the lattice spacing is observed to be 0.344 nm. All
of these TEM
characterizations correspond well with previous crystallographic
descriptions of anatase,
rutile, and brookite TiO2 crystals.
Additionally, there was no discernible difference in the XRD and
Raman before
and after helium treatment meaning the total crystal structure
was not significantly altered
by helium treatment. The helium treatment was an effective way
to remove oxygen from
the TiO2 surface and these oxygen deficient defects were
detected by UV-Vis (Figure
S3) and DRIFTS-IR spectroscopy (Figure S4). For the brookite
phase the absorbance
before 400 nm is noticeably reduced after helium treatment and a
broad, gradual decline
-
11
between 400 and 550 nm can be seen. These changes are also
observed in the anatase-
phase UV-Vis spectrum and indicate the formation of an oxygen
hole or vacancy which
can efficiently capture visible light. The near identical UV-Vis
spectrum for the rutile-
phase before and after helium treatment indicates the absence of
oxygen vacancies in
rutile-phase TiO2.
The oxygen vacancies detected in the UV-Vis spectrum allow for
the addition of
water to the TiO2 structure. When water replaces oxygen in this
structure it creates
anomalies in the surface structure that allow for greater
binding and reduction of CO2.5,6
This is because bending of the normally linear CO2 molecule
decreases the
HOMO/LUMO energy gap and allows for an easier addition of an
electron into the
antibonding orbital of CO2. The possible binding forms of CO2 to
TiO2 surfaces are
given in Scheme 2. Simple coordination of CO2 to the TiO2
surface includes two forms
each of linear coordination and non-linear coordination. An
oxygen atom can be
incorporated into the CO2 to form two forms each of monodentated
and bidentated
carbonates. When water is added to the TiO2 surface two forms
each of monodentated
and bidentated bicarbonates result. In all these cases the
electropositive carbon atom
coordinates or bonds with an electronegative oxygen atoms and
the electronegative
oxygen atoms in CO2 coordinate or bind with the electropositive
Ti atoms in the TiO2
surface. Water addition to the TiO2 surface breaks up the
surface and allows for more
stable coordination, especially of the monodentated bicarbonate
species. All of the
bicarbonate and carbonate species shown in Scheme 2 correspond
to the reduction
products CO2-, HCO3-, and bidentate CO3- (b-CO3-) observed in
the DRIFTS-IR spectra.
The CO and CH4 detected in the gas chromatograms used to
quantify the performance of
the TiO2 surface are products of further reactions of CO2-,
HCO3-, and bidentate CO3- (b-
CO3-) by mechanisms that are yet to be elucidated.
-
12
Scheme 2. Possible surface structures of CO2 on TiO2 surfaces.
CO2 can either be a
simple aggregate, a carbonate (C), or a bicarbonate (B). The
reactions to form C
and B are shown.
-
13
Figure 1 shows DRIFTS-IR spectra of TiA(UP) and TiB(UP) with
photo-illumination
and in the dark. Peaks for reduced carbon species and H2O were
nearly identical in both
scenarios indicating that no new carbon species were formed.
However, a DRIFTS-IR
spectrum of TiA(He) in the dark, shows a rapid formation of
three reduced carbon
species in the first 5 minutes: CO2-, HCO3-, and b-CO3-. Over
time, the intensity of the
H2O band increases, the CO2-band disappears, and there is a
gradual decrease in HCO3-
and b-CO3-. This suggests that H2O can replace the adsorbed
b-CO3- as it occupies the
oxygen vacancies, possibly due to the higher binding energy of
TiO2 with H2O compared
to CO2.
The DRIFTS-IR spectrum of TiA(He) in photo-illumination is shown
in Figure 1
and the intensities of the HCO3- and b-CO3- peaks increase with
prolonged irradiation
time and the H2O peak decreases. This contrast to the dark
spectra is evidence of
electrons trapped in defect sites transferring to the adsorbed
CO2. There is also an
appearance of CO2- around 1675 and 1249 cm-1 that occurs by
photo-induced activation
and reduction rather than re-adsorption since there was no
evidence of CO2 dehydration,
the other possible cause of the CO2- presence. The increase in
intensity for the reduced
carbon species, CO2-, HCO3-, and b-CO3- suggests that these
three are the primary
intermediates for CO2 photoreduction on anatase surfaces and
that H2O disappears as it
fulfills its role as an electron donor and fills the holes left
behind in the TiO2 surface.
The DRIFTS-IR spectra of TiB(He) in the dark and shows similar
peaks to
TiA(He) in the dark with the formation of the three carbon
containing species. There is a
remarkable difference in the spectra with photo-illumination,
however, with TiB(He)
showing two new bands at 1716 and 1379 cm-1 attributed to the
formation of formic acid
(HCOOH) rather than the bands at 1675 and 1249 cm-1 for the
formation CO2-. This
difference suggests that the primary intermediates for the
brookite surfaces are CO2-,
-
14
HCO3-, and HCOOH. All these intermediates are subsequently
reduced to CO and CH4
which are the detected and measured reduction products.
Discussion
It is apparent that an electron donor is necessary for the
photocatalytic reduction
of CO2 to CO and CH4. The oxygen already present on TiO2
surfaces can act as an
effective electron donor but the oxygen atom in water is more
effective.5,6 The
intermediates detected by DRIFTS-IR suggest the incorporation of
water into CO2 as the
final reduction products are formed. In fact, almost any
deformation from a uniform
TiO2 surface will increase CO2 reduction because the oxygen
atoms will be made more
reactive. This increase in reactivity could be because oxygen
vacancies have increased
absorption of light in the visible region and allow for the
addition of water into the
vacancy. When the oxygen vacancies are formed the adjacent Ti
atoms are reduced to
Ti3+, a less stable oxidation state than Ti4+ because of the
full octet of Ti4+, which likely
also contributes to the increased photoreduction activity of
oxygen-poor TiO2 surfaces.
The mechanism of CO2 addition to hydrated TiO2 surfaces is shown
in Scheme 3 and
shows that water addition to TiO2 creates two different active
sites for CO2 reduction and
bridging bicarbonates result.
It is also important for a dopant to not be too electronegative
so the electron
donation can occur. For this reason fluorine is not used as a
dopant. Nitrogen would also
have the appropriate electronegativity properties to be used as
a dopant and nitrogen
containing amine-doped TiO2 surfaces are employed in
photocatalytic CO2 reduction.13
In this way the electron donor can be thought of as a Lewis
base. Normally a hard Lewis
base would be ideal for the reduction of the hard Lewis acid
CO2.
-
15
Scheme 3. Addition of water to TiO2 surface followed by
adsorption of CO2 and
formation of bicarbonates.
Oxygen, even from water, as well as other smaller main group
elements are not
the best electron donors due to their small size and lack of
d-orbitals. The most effective
dopants by mass are larger transition metals and lanthanides
which, although formally
considered Lewis acids, will act as soft Lewis bases when
reducing CO2. Palladium is
already known to be an effective donor into π* antibonding
orbital of molecular
oxygen,15,16 and a similar activity can be expected in CO2
reduction. Additionally,
transition metals are stable in a variety of oxidation states
with Pd making especially
frequent use of the Pd0/Pd2+ redox couple. Ce, while much larger
than oxygen, is also not
a particularly effective dopant because its valence electrons
will occupy s-orbitals and
will not be able to donate into the π* antibonding orbital of
CO2 as effectively as valence
electrons in d-orbitals would. It seems that a soft Lewis acid,
such as Pd and Pt in a low
oxidation state, can act as a hard Lewis base even more
effectively than formal hard
Lewis bases. Large transition metals are the ideal dopants but
are very expensive. For
-
16
this reason molecules of main group elements, including water,
are the most cost efficient
(Table 1) dopants for TiO2 surfaces.
Other potentially useful dopants would include heavier Group 3
elements such as
Ga and In which will form stable +1 ions because of the
resulting fully filled d-shell.
Phosphorus and sulfur compounds would also be interesting to
study for potential
application as electron donors in TiO2 surfaces because they are
inexpensive and not too
electronegative to disfavor electron donation. Lighter
transition metals that oxidize to
form empty or half-filled d-shells could also be promising. Some
of these metals could
include the Fe and Co groups or the Ti and Sc groups. One
problem that needs to be
considered with lighter transition metals is that the energy
difference between principle
quantum numbers will still be relatively high.
Conclusion
Brookite-phase TiO2 surfaces were synthesized and characterized
by XRD,
Raman spectroscopy, TEM, UV-Vis spectroscopy, and DRIFTS-IR
spectroscopy. It was
found that brookite-phase TiO2 was made a more effective
photocatalyst for CO2
reduction when oxygen vacancies were introduced. This is likely
due to a combination of
factors such as increased visible light absorption, replacement
of oxygen atoms with
more reactive water and hydroxyl deformations, and formation of
more readily reducing
Ti3+ ions.
Catalytic performances and cost efficiencies of oxygen-poor
brookite-phase TiO2
were compared to Ce, Pd and Cu, and Pt-doped TiO2 surfaces and
the oxygen-poor
brookite-phase TiO2 had the lowest catalytic performance but the
highest cost efficiency.
The Pd and Cu-doped TiO2 had the highest catalytic performance
but the high cost of this
material makes it unsuitable for industrial use. Further
research into compounds of main
-
17
group elements and lighter transition metals is necessary to
find dopants that are both low
in cost and high in catalytic performance.
Supplementary Material Available
In the appendix a more detailed description of the synthesis of
the anatase, rutile,
and brookite phases of TiO2 is given along with a more detailed
description of the helium
treatment process that is used to create oxygen deficiencies in
the TiO2 surfaces. Spectral
data characterizing the TiO2 surfaces are also available in the
appendix with XRD,
Raman, TEM, UV-Vis, and IR characterizations reported.
(1) United States Environmental Protection Agency: Climate
Change
https://www.epa.gov/climatechange/climate-change-basic-information
(accessed April
10, 2017)
(2) United States Environmental Protection Agency: Greenhouse
Gas Emissions
https://www.epa.gov/ghgemissions/overview-greenhouse-gases
(accessed April 10, 2017)
(3) International Energy Agency: Industrial Applications of
CCS
https://www.iea.org/topics/ccs/ccssubtopics/industrialapplicationsofccs/
(accessed April
10, 2017)
(4) Li, K.; Peng, B.; Peng, T. Recent Advances in Heterogeneous
Photocatalytic CO2
Conversion to Solar Fuels. ACS Catal. 2016, 6, 7485-7527.
(5) Mino, L.; Spoto, G.; Ferrari, A. M. CO2 Capture by TiO2
Anatase Surfaces: A
Combined DFT and FTIR Study. J. Phys. Chem. C 2014, 118,
25016-25026.
References
https://www.epa.gov/climatechange/climate-change-basic-informationhttps://www.epa.gov/ghgemissions/overview-greenhouse-gaseshttps://www.iea.org/topics/ccs/ccssubtopics/industrialapplicationsofccs/
-
18
(6) Liu, L.; Zhao, H.; Andino, J.; and Li, Y. Photocatalytic CO2
Reduction with H2O on
TiO2 Nanocrystals: Comparison of Anatase, Rutile, and Brookite
Polymorphs and
Exploration of Surface Chemistry. ACS Catal. 2012, 2,
1817-1828.
(7) Sigma Aldrich http://www.sigmaaldrich.com/united-states.html
(accessed April 4,
2017)
(8) Matĕjová, L.; Kočí, K.; Reli, M.; Čapek, L.; Hospodková, A.;
Peikertová, P.; Matĕj,
Z.; Obalová, L.; Wach, A.; Kuśtrowski, P.; Kotarba, A.
Preparation, characterization and
photocatalytic properties of cerium doped TiO2: On the effect of
Ce loading on the
photocatalytic reduction of carbon dioxide. Appl. Catal. B 2014,
152-153, 172-183.
(9) Long, R.; Li, Y.: Liu, Y.; Chen, S.; Zheng, X.; Gao, C.; He,
C.; Chen, N.; Qi, Z.;
Song, L.; Jiang, J.; Zhu, J.; and Xiong, Y. Isolation of Cu
Atoms in Pd Lattice: Forming
Highly Selective Sites for Photocatalytic Conversion of CO2 to
CH4. J. Am. Chem. Soc.
2017, 139, 4486-4492.
(10) Mao, J.; Ye, L.; Li, K.; Zhang, X.; Liu, J.; Peng, T.; and
Zan, L. Pt-Loading
Reverses the Photocatalytic Activity Order of Anatase TiO2 {001}
and {010} Facets for
Photoreduction of CO2 to CH4. Appl. Catal. B 2014, 144,
855-862.
(11) Hudson Institute of Mineralogy
http://webmineral.com/MySQL/xray.php?lambda=&lambda1=&ed1=&ed2=&ed3=&min
max=10&chem=TiO2&submit=Submit#.WOUc9jvyuUk (accessed
April 5, 2017)
(12) Sigma Aldrich
http://www.sigmaaldrich.com/united-states.html (accessed April
4,
2017)
http://www.sigmaaldrich.com/united-states.htmlhttp://webmineral.com/MySQL/xray.php?lambda=&lambda1=&ed1=&ed2=&ed3=&minmax=10&chem=TiO2&submit=Submit#.WOUc9jvyuUkhttp://webmineral.com/MySQL/xray.php?lambda=&lambda1=&ed1=&ed2=&ed3=&minmax=10&chem=TiO2&submit=Submit#.WOUc9jvyuUkhttp://www.sigmaaldrich.com/united-states.html
-
19
(13) Liao, Y.; Cao, S.-W.; Yuan, Y.; Gu, Q.; Zhang, Z.; Xue, C.
Efficient CO2 Capture
and Photoreduction by Amine Functionalized TiO2. Chem. Eur. J.
2014, 20, 10220-
10222.
(15) Landis, C. R.; Morales, C. M.; Stahl, S. S. Insights into
the Spin-Forbidden
Reaction between L2Pd0 and Molecular Oxygen. J. Am. Chem. Soc.
2004, 126, 16302-
16303.
(16) Popp, B. V.; Wendlandt, J. E.; Landis, C. R.; Stahl, S. S.
Reaction of Molecular
Oxygen with an NHC-Coordinated Pd0 Complex: Computational
Insights and
Experimental Implications. Angew. Chem. Int. Ed. 2007, 46,
601-604.
-
S1
Supporting Information
Carbon Capture, Utilization, and Storage. A New Method for
Carbon Dioxide
Photoreduction Catalyzed by Water-Doped TiO2 Surfaces.
Melissa Hopkins, Parker Smith, and Ethan Zars
Department of Chemistry, University of Missouri-Columbia,
Columbia, MO, USA 65201
Email: [email protected]; [email protected];
[email protected]
mailto:[email protected]:[email protected]:[email protected]
-
S2
Table of Contents
Details of TiO2 anatase, rutile, and brookite syntheses S3
Figure S1 XRD and Raman spectra of TiO2 surfaces S4
Figure S2 TEM images of TiO2 surfaces S5
Figure S3 UV-Vis spectra of TiO2 surfaces S6
Figure S4 DRIFTS-IR spectra of the –OH groups of TiO2 surfaces
S7
Bibliography S8
-
S3
The TiO2 nanocrystals are prepared in the three polymorphs:
rutile, anatase, and
brookite. The rutile crystals are synthesized by adding a 2
molar ratio of titanium
tetrachloride dropwise to a 20 molar ratio of ethanol, which
forms a transparent yellow
solid. The solid is then added to distilled water with a 280
molar ratio. Once combined,
the solution is stirred for thirty minutes, dried in a 50 ⁰C
oven for 24 hours, and
centrifuged to collect the white precipitate. Anatase crystals
are synthesized using 10 mL
of titanium bis(ammonium lactate) dihydroxide aqueous solution,
desired amount of 0.1
M urea, and distilled water to reach a final volume of 100 mL.
The solution is placed in a
sealed Teflon-lined autoclave and set in a 160 ⁰C oven for 24
hours. Upon cooling in air,
a precipitate is formed and then separated by centrifugation.
The collected solid is
washed with distilled water until a pH of 7 is achieved then
dried overnight in an oven set
at 60 ⁰C. Brookite is synthesized in a similar manner to
anatase, though 7 M urea is used
in place of the 0.1 M solution. All three solids are calcined at
400 ⁰C for three hours with
a 2 ⁰C/min heating rate.
In addition to these non-pretreated materials, helium treated
samples are prepared
to create an oxygen deficiency in the structures. This is
accomplished by providing a flow
of He at 120 mL/min inside of diffuse reflectance infrared
Fourier transform
spectroscopy (DRIFT) reaction cell or inside of a photo reactor
for 1.5 hours at 220 ⁰C.
The pretreated samples are denoted as Ti(He) and the
non-pretreated samples as Ti(UP).
In sum, the three polymorphs, rutile, anatase, and brookite,
with treated and untreated
samples are denoted as TiR(He), TiR(UP), TiA(He), TiA(UP),
TiB(He), and TiB(UP)
respectively.
-
S4
Figure S1. (a) XRD profiles of unpretreated brookite (TiB(UP)),
rutile (TiR(UP)), and
anatase (TiA(UP)) phase TiO2. (b) XRD profiles of helium-treated
brookite (TiB(He)),
rutile (TiR(He)), and anatase (TiA(He)) phase TiO2. (c) Raman
absorbance spectra of
unpretreated brookite (TiB(UP)), rutile (TiR(UP)), and anatase
(TiA(UP)) phase TiO2.
-
S5
Figure S2. TEM images for (a, b) helium treated anatase
(TiA(He), (c, d) helium treated
rutile (TiR(He)), and (e, f) helium treated brookite (TiB(He))
TiO2 crystals.
-
S6
Figure S3. UV-Vis spectra of unpretreated (TiR(UP), TiB(UP),
TiA(UP)) and He-
treated (TiR(He), TiB(He), TiA(He) rutile, brookite, and anatase
phase TiO2. The inset
picture shows the color differences of the three TiO2 polymophs
after He treatment.
-
S7
Figure S4. DRIFTS-IR spectra of OH groups in the region
3000−3800 cm−1 for (a)
unpretreated anatase (TiA(UP)), brookite (TiB(UP)), and rutile
(TiR(UP)) and (b) He-
treated anatase (TiA(He)), brookite (TiB(He)), and rutile
(TiR(He)) phase TiO2 crystals.
-
S8
Bibliography
Liu, L.; Zhao, H.; Andino, J.; and Li, Y. Photocatalytic CO2
Reduction with H2O on
TiO2 Nanocrystals: Comparison of Anatase, Rutile, and Brookite
Polymorphs and
Exploration of Surface Chemistry. ACS Catal. 2012, 2,
1817-1828.
Long, R.; Li, Y.: Liu, Y.; Chen, S.; Zheng, X.; Gao, C.; He, C.;
Chen, N.; Qi, Z.; Song,
L.; Jiang, J.; Zhu, J.; and Xiong, Y. Isolation of Cu Atoms in
Pd Lattice: Forming
Highly Selective Sites for Photocatalytic Conversion of CO2 to
CH4. J. Am. Chem. Soc.
2017, 139, 4486-4492.
Mao, J.; Ye, L.; Li, K.; Zhang, X.; Liu, J.; Peng, T.; and Zan,
L. Pt-Loading Reverses the
Photocatalytic Activity Order of Anatase TiO2 {001} and {010}
Facets for
Photoreduction of CO2 to CH4. Appl. Catal. B 2014, 144,
855-862.
Matĕjová, L.; Kočí, K.; Reli, M.; Čapek, L.; Hospodková, A.;
Peikertová, P.; Matĕj, Z.;
Obalová, L.; Wach, A.; Kuśtrowski, P.; Kotarba, A. Preparation,
characterization and
photocatalytic properties of cerium doped TiO2: On the effect of
Ce loading on the
photocatalytic reduction of carbon dioxide. Appl. Catal., B
2014, 152-153, 172-183.