Effects of metal oxide surface doping with phosphonic acid monolayers on alcohol dehydration activity and selectivity

Article abstract of DOI:10.1016/j.apcata.2018.12.009

Controlling the near-surface environment of heterogeneous catalysts is of fundamental importance for high selectivity and activity. Self-assembled monolayers (SAMs) are effective tools to control reaction selectivity and activity for both supported noble metal and metal oxide catalysts. We previously demonstrated tunable dehydration activity of alcohols on phosphonic acid-modified, anatase-phase TiO₂. In this work, we investigated the generality of this approach by studying the modification of other metal oxides including Al₂O₃, CeO₂, CuO, Fe₂O₃, MgO, rutile-TiO₂, SnO₂, V₂O₅, WO₃, ZrO₂, and ZnO. Modification of these materials with phosphonic acids results in the formation of SAMs on the surface, as determined by infrared spectroscopy; studies of the thermal stability on selected catalysts indicated that the SAMs remained intact up to approximately 400 °C in inert environments. Decomposition of alcohols on these native materials resulted in dehydration, dehydrogenation, and condensation. Upon functionalization with phosphonic acid modifiers, the activity of all pathways decreased significantly, except for dehydration on CeO₂, anatase-TiO₂, and SnO₂. We explored the properties of these oxides that may be responsible for this increase in dehydration activity using correlations to bulk properties. This analysis supported the hypothesis that phosphonic acid monolayers act as surface-level dopants for metal oxides of specific metal-oxygen bond strength and oxidation state.

Full text of DOI:10.1016/j.apcata.2018.12.009

Accepted Manuscript
Title: Effects of metal oxide surface doping with phosphonic
acid monolayers on alcohol dehydration activity and
selectivity
Authors: Lucas D. Ellis, Jordi Ballesteros-Soberanas, Daniel
K. Schwartz, J. Will Medlin
PII:
S0926-860X(18)30598-2
DOI:
Reference:
https://doi.org/10.1016/j.apcata.2018.12.009
APCATA 16913
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
19 October 2018
8 December 2018
11 December 2018
Please cite this article as: Ellis LD, Ballesteros-Soberanas J, Schwartz DK, Will
Medlin J, Effects of metal oxide surface doping with phosphonic acid monolayers
on alcohol dehydration activity and selectivity, Applied Catalysis A, General (2018),
https://doi.org/10.1016/j.apcata.2018.12.009
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Effects of metal oxide surface doping with phosphonic acid monolayers on
alcohol dehydration activity and selectivity
1
1
1
Lucas D. Ellis , Jordi Ballesteros-Soberanas , Daniel K. Schwartz , J. Will
Medlin1*
1Department of Chemical & Biological Engineering,
University of Colorado Boulder, Boulder, CO 80309-0596
*Corresponding author, email: Will.Medlin@colorado.edu
Graphical Abstract
Highlights:
Functionalization of metal oxide catalysts with phosphonic acid

modifiers lowers vapor phase dehydrogenation and condensation
activity of alcohols.



Functionalization of TiO
2
-anatase, CeO
2
and SnO with phosphonic
2
acid modifiers enhances vapor-phase dehydration activity of alcohols,
while lowering dehydration activity for all other oxides studied here.
Metal-oxygen bond strength and oxidation state appear to be unique
metal oxide properties that correlate with enhancement of vapor-phase
dehydration activity of alcohols after functionalization with phosphonic
acid modifiers.
Abstract
Controlling the near-surface environment of heterogeneous catalysts is of
fundamental importance for high selectivity and activity. Self-assembled
monolayers (SAMs) are effective tools to control reaction selectivity and activity
for both supported noble metal and metal oxide catalysts. We previously
demonstrated tunable dehydration activity of alcohols on phosphonic acid-
modified, anatase-phase TiO
approach by studying the modification of other metal oxides including Al
CeO , CuO, Fe , MgO, rutile-TiO , SnO , V , WO , ZrO , and ZnO.
2
. In this work, we investigated the generality of this
2 3
O ,
2
2
O
3
2
2
2
O
5
3
2
Modification of these materials with phosphonic acids results in the formation of
SAMs on the surface, as determined by infrared spectroscopy; studies of the
thermal stability on selected catalysts indicated that the SAMs remained intact up
to approximately 400°C in inert environments. Decomposition of alcohols on
these native materials resulted in dehydration, dehydrogenation, and
condensation. Upon functionalization with phosphonic acid modifiers, the activity
of all pathways decreased significantly, except for dehydration on CeO
anatase-TiO , and SnO . We explored the properties of these oxides that may
be responsible for this increase in dehydration activity using correlations to bulk
2
,
2
2
2

properties. This analysis supported the hypothesis that phosphonic acid
monolayers act as surface-level dopants for metal oxides of specific metal-
oxygen bond strength and oxidation state.
Abbreviations: SAMs = self-assembled monolayers, MPA = methylphophonic
acid, ODPA = octadecylphosphonic acid.
Keywords: metal oxide catalysis, alcohol dehydration, self-assembled
monolayers, phosphonic acids.
1
. Introduction
Understanding and controlling the near-surface environment is of critical
importance in the field of catalysis. One recently developed strategy involves the
use of self-assembled monolayers (SAMs) to tune the surface chemistry of
catalysts.[1,2] Self-assembled monolayers and similar modifiers have gained
recent interest because of their enhancement of catalytic selectivity and
activity.[3–7] Recent studies have sought to extend these SAM strategies to
metal oxide catalysts and supports using silane[8] and phosphonate
modifiers.[9,10] However, these efforts focused only on Al
2
O
3
and TiO .
2
Phosphonic acids have been shown to form SAMs on a wide variety of metal
oxides, providing potential opportunities to expand this SAM strategy to a broad
set of reaction chemistries.
In this work, we explored the effect of phosphonate SAMs on the reactivity
of alcohols on a wide variety of metal oxide catalysts functionalized with alkyl
phosphonic acids. We found that phosphonates were able to form organized and
robust SAMs on many metal oxides. Activity toward decomposition of linear
3

alcohols was promoted on anatase phase TiO
2
, SnO
2
and CeO , but not other
2
oxides. In fact, most oxides demonstrated a significant reduction in catalytic
activity for all reaction pathways. Through correlations with bulk properties, we
propose that phosphonates acted as surface level dopants, tuning the activity of
alcohol dehydration, for a specific set of oxides.
2
. Experimental
2
.1 Catalyst synthesis
All metal oxide powders were purchased commercially: iron (III) oxide,
NanoArc, (CAS: 1309-37-1, LOT: L20Y042, Alfa Aesar), Copper (II) oxide, min.
7%, 99+% Cu (CAS:1317-38-0, LOT: 28673900, Strem Chemicals Inc.), zinc (II)
oxide, 99.7% (CAS: 1314-13-2, LOT: A9680109, Strem Chemicals Inc.), titanium
9
(
IV) oxide, rutile, 99.8% (CAS:1317-80-2, LOT: T24A025, Alfa Aesar), tungsten
VI) oxide, 99.9% (CAS: 1314-35-8, LOT: BCBN7138V, Fluka Analytical, Sigma
(
Aldrich), titanium (IV) oxide, ≥99% trace (CAS: 1317-70-0, LOT: MKCF0054,
Aldrich Chemistry, Sigma Aldrich), cerium (IV) oxide, 99.5% min (CAS: 1306-38-
3
, LOT: D06X036, Alfa Aesar), magnesium (II) oxide, 99+% (metals basis)
CAS:1309-48-4, LOT: M20B021, Alfa Aesar), tin (IV) oxide, 99.9% (metals
basis) (CAS: 18282-10-5, LOT: R20B037, Alfa Aesar), and zirconium (IV) oxide,
9% trace metals (CAS: 1314-23-4, LOT: BCBM6158V, Aldrich Chemistry,
Sigma Aldrich).
Metal oxide powders were functionalized with methylphosphonic acid,
8% (CAS: 993-13-5, LOT: 10206487, Alfa Aesar) and octadecylphosphonic
(
9
9
4

acid, by mixing metal oxides in a 10mM solution in tetrahydrofuran (THF), HPLC
Grade (CAS: 109-99-9, LOT: 177753, Fisher Chemicals) overnight. Powders
were removed from solution via centrifugation, decanting excess solvent and
annealing at 120°C for 6 hours. Powders were then rinsed with the same volume
as the deposition volume of THF three times before drying the samples overnight
under vacuum.
2
.2 Catalyst Characterization
Atmospheric diffuse reflectance infrared Fourier transform spectroscopy
(DRIFTS) was performed on a Praying Mantis system (Harrick Scientific).
Reported DRIFTS results represent the raw spectra, and are not normalized
between samples. Surface area measurements were performed using nitrogen
physisorption (30% nitrogen, balance helium, Airgas), unless surface areas were
reported by the manufacturer. Inductively coupled plasma mass spectrometry
(
ICP-MS) was performed to measure phosphorous and cation content of native
and functionalized materials. Temperature programmed reaction spectroscopy
TPRS) and temperature programmed desorption (TPD) were performed using a
(
custom quartz tube (6.35-mm inner diameter) vapor phase reactor. The effluent
was monitored with a Pfeiffer mass spectrometer. For TPD experiments,
functionalized catalysts were sandwiched between two 125mg slugs of quartz
wool with either ultra-high purity (UHP) helium (Airgas), or 7 torr water vapor in
UHP helium as a carrier gas. Temperature programmed desorption started at
1
00°C and ended at 550°C with a ramp rate of 10°C/min. For TPRS
experiments, catalysts were sandwiched between 125mg of quartz wool and
5

pretreated in UHP helium for 1 hr at 250°C. After pretreatment, samples were
cooled to 50°C, then saturated with 1-butanol vapor in UHP helium for 10
minutes. The system was then purged for 25 minutes before beginning the
2
0°C/min ramp rate to 550°C.
2
.3 Catalyst Performance
Reactions of 1-butanol were conducted on a custom pyrex-tube (6.35-mm
inner diameter) packed bed vapor phase reactor with an Agilent gas
chromatograph with a thermal conductivity and flame-ionization detector
measuring effluent concentrations of reactants and products. A DB-WAX (30m x
0
.250mm x 0.50μm) column was used to separate components. The reactant, 1-
butanol (≥ 99%, for molecular biology, Sigma-Aldrich), was delivered to the
catalyst by bubbling an inert carrier (helium, ultra high purity, Airgas) through the
liquid phase reactant, which was temperature controlled with a water bath set to
1
4°C. Reactions were performed at 250°C with data reported for data acquired
from 180-210 minutes (time-on-stream) then averaged with two other sample
replicates. Experimental uncertainty is represented as the standard error from
triplicate samples.
3
. Results
3
.1 Catalyst Characterization
Metal oxide powders (Table 1) were functionalized with
octadecylphosphonic acid (ODPA) and methylphosphonic acid (MPA). Here, we
used ODPA modification as a method to test whether relatively close-packed
6

monolayers were formed on the various surfaces. Functionalization of materials
with long-chain alkyl SAMs like ODPA can be thoroughly studied and the
organization of these SAMs is often characterized using asymmetric and
symmetric vibrational modes of methylene functionalities.[10] The frequencies
for these vibrations redshift as the SAM molecules become more organized.
Figure 1 shows the infrared spectra of the ODPA-functionalized metal oxides. All
modified oxides studied here demonstrated an asymmetric methylene vibrational
frequency typical of organized SAMs. Organization does appear to depend to
some extent on the metal oxide; while many metal oxides have clear peaks at
-
1
2
920 cm or lower, V
associated with decreased organization. To our knowledge, this is the first report
of the functionalization of CeO , CuO, MgO, SnO , and V with phosphonic
acids. Functionalization of Al ,[11–13] Fe ,[14] TiO ,[11–13,15] WO ,[16]
ZrO ,[11,12,17] and ZnO [18–21] has been reported previously, however.
An important criterion for the use of SAM-modified catalysts is that the
2
O
5
, ZrO
2
, TiO
2
rutile showed more blue-shifted peaks
2
2
2
O
5
2
O
3
2
O
3
2
3
2
coating must remain stable at reaction temperatures of interest. Few studies
have focused on the thermal stability of phosphonic acid functionalized metal
oxides with a couple exceptions.[13,22] McElwee et al. utilized
thermogravimetric analysis to demonstrate no degradation of ODPA-
functionalized γ-Al
2
O
3
and TiO -anatase powders below 430°C. Bhairamadgi et
2
al. noted no degradation of ODPA-functionalized porous alumina below 500°C.
We expanded the range of oxides investigated and explored the stability of
ODPA functionalized γ-Al
2
O
3
, TiO
2
-anatase, ZrO , and ZnO using DRIFTS of the
2
7

C-H stretching region (supplementary information, Figure S1) and temperature
programmed desorption (TPD) (supplementary information, Figure S2).
Temperature programmed desorption in helium and 1% water vapor in helium
demonstrated no production of volatile decomposition products below 400°C.
We also investigated the organization of the ODPA tail functionality using
temperature- and atmosphere-controlled DRIFTS. In a helium atmosphere, little
degradation of asymmetric or symmetric methylene vibrations was measured
after heating the sample to ~400°C and returning to ambient conditions. These
results were in line with previous studies of reactions of alcohols on phosphonic
acid functionalized TiO
2
- and Al
2
O -based catalysts, which showed no loss of
3
functionality after exposure to reaction conditions.[9,23,24] We noted that there
were subtle shifts in peak frequency and intensity after the thermal treatment; as
shown in Figure S1, for example, the peak intensity for methylene stretching in
some cases increased after heating. Such changes have previously been
interpreted in terms of reorganization of the monolayers to adopt different
orientations (e.g., with different degrees of tilt) during annealing.[25] Though
these SAMs were found to be stable to 400°C in inert environments and in the
presence of low levels of water vapor, it should be noted that harsher (especially
oxidizing) conditions may lead to degradation.
Although the DRIFTS results are generally consistent with a relatively full
monolayer, some coverage variations between the different oxides are expected.
Inductively coupled plasma (ICP) techniques have been used to estimate the
coverage of several phosphonic acids on Al
2
O to be approximately 3-4
3
8

-2
molecules nm , and analysis of benzyl phosphonic acids on TiO have indicated
2
a similar coverage.[9,13,24] Attempts to accurately measure coverage changes
across different oxides with ICP in this study were hampered by the low surface
areas of many of the tested materials. In addition to the possibility of changes in
surface coverage, it is important to note that different oxides have been reported
to yield different modes of phosphonate attachment.[17]
3
.2 Catalyst Performance
Previously, we demonstrated that there was a weak inverse dependence
of alkyl tail length and dehydration activity, where shorter moieties, like MPA
functionalized anatase phase TiO had higher dehydration activity compared with
decylphosphonic- and ODPA-functionalized TiO .[9] With these findings in mind,
2
2
we sought to determine which oxides demonstrated maximal dehydration activity,
by focusing on MPA functionalized materials.
Metal oxide powders functionalized with MPA were compared to their
native counterparts in the vapor-phase decomposition of 1-butanol at 250°C
using a helium gas carrier. Primary alcohols are known to react via dehydration,
dehydrogenation, and condensation on metal oxides. Condensation is often a
minor pathway, with selectivities typically below 15%. However, γ-Al
2
O is a
3
notable exception to this trend, since the ether made up nearly 60% of the
products at these reaction conditions, as expected.[26] After functionalization
with MPA, significant selectivity changes (>15%) occurred for five materials: γ-
9

Al
2
O
3
, TiO
2
-anatase, CeO
2
, V
, TiO
demonstrated increases in dehydration selectivity (>50%) and decreases in
exhibited decreases in
condensation (>30%) and increases in dehydration (>30%). Ceria and V
saw a respective 13.6% and 20.5% decrease in dehydrogenation selectivity and
5.6% and 17.1% increase in dehydration selectivity, with a difference between
2
O
5
, and SnO
2
(Figure 2). The greatest changes
(>30%) occurred for γ-Al
2
O
3
2
-anatase, and SnO
2
. Anatase-phase TiO and
2
SnO
2
dehydrogenation selectivity (>50%), while γ-Al
2
O
3
2
O also
5
1
these pathways coming from a change in condensation selectivity (Figure 2 and
supplementary information Figure S3). All other oxides showed little to no
change in reaction selectivity beyond experimental error.
The selectivity trends show that γ-Al
2
O
3
, TiO
2
-anatase, V
2
O
5
, SnO
2
and
CeO
2
were outliers in performance after functionalization with MPA. However,
upon closer examination of reaction rates, a new and more surprising trend
emerged (Figure 3). Gamma-alumina demonstrated a decrease in activity for
dehydration, dehydrogenation, and condensation. In fact, activities for all
pathways decreased by nearly 99%, indicating that the phosphonates were
highly effective surface poisons. Activity for each pathway also decreased for
coated V
2
O
5
, but to a lesser extent. For CeO
2
, SnO
2
, and TiO -anatase,
2
dehydration activity increased, while dehydrogenation and condensation activity
decreased. In fact, dehydration activity increased by 110%, 210% and 810%
while dehydrogenation activity decreased by 26%, 92% and 99% for CeO
anatase, and SnO , respectively. Because of the high selectivity to the
dehydrogenation product for uncoated CeO and SnO , the significant increase in
2
, TiO
2
-
2
2
2
1
0

dehydration rates was not accompanied by high selectivities. For example,
native SnO had a 99.7% selectivity toward the dehydrogenation product;
2
modification with MPA therefore only improved the dehydration selectivity to
approximately 50%. Overall, these results raised two questions: (1) Why is the
dehydration pathway promoted while dehydrogenation and condensation
decrease in activity? and (2) What oxide properties are associated with an
increase in dehydration activity after MPA functionalization?
3
.3 Influence of phosphonates on surface chemistry
Previous studies of reactions of 1-propanol on anatase TiO demonstrated
2
a 13 kJ/mol decrease in the apparent activation energy for dehydration and a 42
kJ/mol increase for dehydrogenation for the MPA functionalized material.[9] This
suggested a significant shift in the energetic landscape of the dehydrogenation
mechanism with a moderate shift in the dehydration mechanism.
Dehydrogenation is thought to require two Lewis acid sites, the first capable of
binding the alcohol and the second abstracting a hydride alpha to the alcohol
carbon.[27–29] The functionalization with phosphonic acids would be expected
to reduce the number of two adjacent Lewis acids, thus lowering selectivity to
dehydrogenation. The argument is similar for condensation, a bimolecular
reaction. For condensation to occur two alcohols must adsorb near to one
another; having a high concentration of phosphonic acids on the surface likely
reduces the probability for such an occurrence.
1
1

The phosphonic acid surface coverage is consistent with the decrease in
dehydrogenation and condensation activity, but does not explain the varying
response in dehydration activity. More explicitly, dehydration activity appears to
have no correlation with surface coverage. Using temperature programmed
reaction spectroscopy (TPRS) (supplementary information Figures S4-S5) we
noted a significant reduction in the total adsorption of alcohols on all materials,
again, consistent with high surface coverage of phosphonic acids. Moreover, the
available alcohol binding sites appear to have comparable adsorption strength to
the native counterpart. It therefore appears the phosphonic acids must increase
intrinsic activity of the remaining sites for dehydration, but only for specific
oxides. We sought to explain this result by correlations to bulk material
properties.
4
. Discussion
It is useful to consider results from previous studies that have sought to
identify the relationship between metal oxide properties and reaction
performance. Previous studies have provided an in-depth analysis of metal
oxide properties as they impact dehydrogenation activity of simple alcohols like
ethanol.[30,31] Idriss el al. investigated correlations of dehydrogenation activity
with bulk metal oxide properties such as electronegativity difference between the
metal and oxygen, metal oxygen bond strength, oxygen partial charge, oxygen
Madelung potential, and oxygen electronic polarizability, and concluded that the
oxygen electronic polarizability was the best predictor for dehydrogenation
activity. We sought to utilize similar material properties: differences in
1
2

electronegativity between the metal and oxide, polarizability of the cation,
polarizability of the oxide, band gap, and metal oxygen bond strength, as
possible descriptors for the change in dehydration activity of these materials after
functionalization with phosphonic acid modifiers. Differences in metal oxide
electronegativities and polarizability of the cation and oxide did not demonstrate
a significant correlation (supplementary information, Figures S6-S9). Use of the
metal-oxygen bond strength as a descriptor resulted in a “volcano” type
relationship, with strong and weak metal oxygen bond strengths having the most
suppression in dehydration activity after functionalization with phosphonic acids,
while moderate metal-oxygen bond strength materials had improved activity
levels. The use of metal-oxygen bond strength as a predictor does have a
connection to the mechanism of dehydration, since metal-oxygen bond cleavage
is thought to occur during the E2-like surface mediated reaction.[32,33] As
discussed in more detail below, there are a few key outliers in this relationship,
suggesting that other factors likely play a role in dictating how phosphonate
dopants influence catalysis.
The complex response of dehydration activity and functionalization with
phosphonic acids appears to be correlated with electron mobility and energetics.
Though phosphonic acids have not be used for this express purpose (i.e., tuning
catalytic properties of surface atoms), phosphonic acids have been understood to
tune the work function of electronic materials like zinc oxide[19,20] and indium tin
oxide.[35–38] Metiu et al.[39] demonstrated that surface-level cation exchange
in metal oxides can provide a means to control Lewis acid and base properties.
1
3

For example, by replacing a surface level cation with one of higher valance, the
surface can become more of a Lewis base, hypothetically enhancing activity
toward elimination of the beta hydrogen of the alcohol.[32] This newly doped
surface will result in stronger binding of Lewis acids, but weaker binding of Lewis
bases. Interestingly, this type of analysis on our materials highlights one critical
common factor among the three oxides, CeO
2
, SnO
2
, and TiO : all oxides have
2
the same bulk metal oxidation state. Binding of the phosphonic acids on the
surface of these oxides could be seen as a type of surface doping with
phosphorous being the cation of higher oxidation state. With this hypothesis in
mind, according the rules developed by Metiu et al., phosphorous would act as a
high valence dopant, resulting in a surface with weaker binding to Lewis bases.
Since alcohols are Lewis bases, this hypothesis suggests a significant
weakening in the adsorption energy of alcohols. Unfortunately, attempts to
investigate surface acidity using ammonia TPD were complicated by very low
yields observed (supplementary information Table S1); we also were unable to
resolve trends in surface hydroxyl stretching modes on the various oxides
(supplementary information Figure S10).
One obvious outliner in our trends is the comparison of anatase and rutile
phase TiO
2
. Rutile demonstrated a decrease in dehydration activity of 69% while
anatase phase TiO
2
demonstrated an increase of over 200%. Though these two
materials appear to share many similar qualities, there is precedent for significant
variation in catalytic activity of these materials related to electronic structure
1
4

differences, as well as major differences in the interaction of surface
dopants.[40,41]
5
. Conclusions
Phosphonic acid monolayers are capable of forming robust monolayers on
many different metal oxide powders. However, the change in catalytic activity,
specifically for alcohol dehydration, was limited to TiO -anatase, CeO and SnO
2
2
2
.
We speculate the reason is due to electronic doping by the phosphonic acid
monolayers. The bulk properties that seem important to this process relate to
metal-oxygen bond strength of a certain range and oxidation state. Clearly more
studies are required to fully elucidate the mechanistic understanding of this
phenomena, but this study does highlight the use phosphonic acid monolayers
as robust dopants for certain thermochemical conversions of oxygenates by
metal oxide catalysts.
Acknowledgements
The authors would like to acknowledge financial support from the United
States Department of Agriculture, National Institute of Food and Agriculture
predoctoral fellowship program (Award Number #: 2017-67011-26078) and the
Department of Energy, Office of Science, Basic Energy Sciences Program,
Chemical Sciences, Geosciences and the Biosciences Division (Grant No. DE-
FG02-10-ER16206).
1
5

References
[
[
[
[
1]
2]
3]
4]
S.T. Marshall, M. O’Brien, B. Oetter, A. Corpuz, R.M. Richards, D.K.
Schwartz, J.W. Medlin, Nat. Mater. 9 (2010) 853–858.
C. a Schoenbaum, D.K. Schwartz, J.W. Medlin, Acc. Chem. Res. 47 (2014)
1
438–45.
B. Wu, H. Huang, J. Yang, N. Zheng, G. Fu, Angew. Chem. Int. Ed. Engl.
1 (2012) 3440–3443.
5
S.G. Kwon, G. Krylova, A. Sumer, M.M. Schwartz, E.E. Bunel, C.L.
Marshall, S. Chattopadhyay, B. Lee, J. Jellinek, E. V. Shevchenko, Nano
Lett. 12 (2012) 5382–5388.
[
[
5]
6]
C.-H. Lien, J.W. Medlin, J. Phys. Chem. C 118 (n.d.) 23783–23789.
I. Schrader, J. Warneke, J. Backenköhler, S. Kunz, J. Am. Chem. Soc. 137
(2015) 905–912.
[
[
[
[
[
7]
8]
9]
G. Chen, C. Xu, X. Huang, J. Ye, L. Gu, G. Li, Z. Tang, B. Wu, H. Yang, Z.
Zhao, Z. Zhou, G. Fu, N. Zheng, Nat. Mater. 15 (2016) 564.
L.D. Ellis, S. Pylypenko, S.R. Ayotte, D.K. Schwartz, J.W. Medlin, Catal.
Sci. Technol. 6 (2016) 5721–5728.
L.D. Ellis, R.M. Trottier, C.B. Musgrave, D.K. Schwartz, J.W. Medlin, ACS
Catal. 7 (2017) 8351–8357.
10] T. Van Cleve, D. Underhill, M. Veiga Rodrigues, C. Sievers, J.W. Medlin,
Langmuir 34 (2018) 3619–3625.
11] W. Gao, L. Dickinson, C. Grozinger, F.G. Morin, L. Reven, Langmuir 12
(1996) 6429–6435.
[
[
12] R. Hofer, M. Textor, N.D. Spencer, Langmuir 17 (2001) 4014–4020.
13] J. McElwee, R. Helmy, A.Y. Fadeev, J. Colloid Interface Sci. 285 (2005)
5
51–556.
[
14] C. Yee, G. Kataby, A. Ulman, T. Prozorov, H. White, A. King, M.
Rafailovich, J. Sokolov, A. Gedanken, Langmuir 15 (1999) 7111–7115.
15] G. Guerrero, P.H. Mutin, A. Vioux, Chem. Mater. 13 (2001) 4367–4373.
16] F.H. Li, J.D. Fabbri, R.I. Yurchenko, A.N. Mileshkin, J.N. Hohman, H. Yan,
H. Yuan, I.C. Tran, T.M. Willey, M. Bagge-Hansen, J.E.P. Dahl, R.M.K.
Carlson, A.A. Fokin, P.R. Schreiner, Z.-X. Shen, N.A. Melosh, Langmuir 29
[
[
(
2013) 9790–9797.
17] W. Gao, L. Dickinson, C. Grozinger, F.G. Morin, L. Reven, Langmuir 13
1997) 115–118.
18] B. Zhang, T. Kong, W. Xu, R. Su, Y. Gao, G. Cheng, Langmuir 26 (2010)
514–4522.
19] P.J. Hotchkiss, M. Malicki, A.J. Giordano, N.R. Armstrong, S.R. Marder, J.
Mater. Chem. 21 (2011) 3107.
20] I. Lange, S. Reiter, M. Pätzel, A. Zykov, A. Nefedov, J. Hildebrandt, S.
Hecht, S. Kowarik, C. Wöll, G. Heimel, D. Neher, Adv. Funct. Mater. 24
[
[
[
[
(
4
(2014) 7014–7024.
[
21] R. Quiñones, D. Shoup, G. Behnke, C. Peck, S. Agarwal, R.K. Gupta, J.W.
Fagan, K.T. Mueller, R.J. Iuliucci, Q. Wang, Materials (Basel). 10 (2017).
1
6

[
22] N.S. Bhairamadgi, S.P. Pujari, F.G. Trovela, A. Debrassi, A.A. Khamis,
J.M. Alonso, A.A. Al Zahrani, T. Wennekes, H.A. Al-Turaif, C. van Rijn,
Y.A. Alhamed, H. Zuilhof, Langmuir 30 (2014) 5829–5839.
23] J. Zhang, L.D. Ellis, B. Wang, M.J. Dzara, C. Sievers, S. Pylypenko, E.
Nikolla, J.W. Medlin, Nat. Catal. (2018) 1.
24] P.D. Coan, L.D. Ellis, M.B. Griffin, D.K. Schwartz, J.W. Medlin, J. Phys.
Chem. C 122 (2018) 6637–6647.
25] A.R. Corpuz, S.H. Pang, C. Schoenbaum, J.W. Medlin, Langmuir 30
[
[
[
(2014) 14104–14110.
[
[
[
26] H. Knözinger, A. Scheglila, J. Catal. 17 (1970) 252–263.
27] J.E. Rekoske, M.A. Barteau, J. Catal. 165 (1997) 57–72.
28] H. Idriss, in: R. Rioux (Ed.), Model Syst. Catal., Springer New York, 2010,
pp. 133–154.
[
29] A.M. Nadeem, G.I.N. Waterhouse, H. Idriss, Catal. Today 182 (2012) 16–
2
4.
[
[
[
30] H. Idriss, M.A. Barteau, Adv. Catal. 45 (2000) 261–331.
31] H. Idriss, E.G. Seebauer, Catal. Letters 66 (2000) 139–145.
32] P. Kostestkyy, J. Yu, R.J. Gorte, G. Mpourmpakis, Catal. Sci. Technol. 4
(2014) 3861–3869.
[
[
33] P. Kostetskyy, G. Mpourmpakis, Catal. Sci. Technol. 5 (2015) 4547–4555.
34] Z. Helali, A. Jedidi, O.A. Syzgantseva, M. Calatayud, C. Minot, Theor.
Chem. Acc. 136 (2017).
[
[
35] M. Gliboff, H. Li, K.M. Knesting, A.J. Giordano, D. Nordlund, G.T. Seidler,
J.-L. Brédas, S.R. Marder, D.S. Ginger, J. Phys. Chem. C 117 (2013)
1
5139–15147.
36] S.E. Koh, K.D. McDonald, D.H. Holt, C.S. Dulcey, J.A. Chaney, P.E.
Pehrsson, Langmuir 22 (2006) 6249–6255.
[
[
37] H. Li, P. Paramonov, J.-L. Bredas, J. Mater. Chem. 20 (2010) 2630.
38] S.A. Paniagua, P.J. Hotchkiss, S.C. Jones, S.R. Marder, A. Mudalige, F.S.
Marrikar, J.E. Pemberton, N.R. Armstrong, J. Phys. Chem. C 112 (2008)
7
809–7817.
39] H. Metiu, S. Chrétien, Z. Hu, B. Li, X. Sun, J. Phys. Chem. C 116 (2012)
0439–10450.
[
1
[
[
40] M. Batzill, Energy Environ. Sci. 4 (2011) 3275.
41] C. Hernandez-Mejia, E.S. Gnanakumar, A. Olivos-Suarez, J. Gascon, H.F.
Greer, W. Zhou, G. Rothenberg, N.R. Shiju, Catal. Sci. Technol. 6 (2016)
5
77–582.
1
7

Figure 1 DRIFTS spectrum of ODPA functionalized oxides in the C-H stretching region.
1
8

Figure 2 Selectivity to dehydration and dehydrogenation products for 1-butanol reaction at 250°C. Red
arrows indicated increases in selectivity of >30%, while blue arrows indicate decreases in selectivity of
>
30% after functionalization with MPA.
1
9

Figure 3. Ratio in activity of dehydration, dehydrogenation, and condensation comparing phosphonate
functionalized metal oxide divided by the native performance in the reaction of 1-butanol at 250°C.
2
0

Figure 4 Change in dehydration activity after functionalization with MPA as a function of the difference of
metal oxygen bond strength (EM-O) taken from Helali et al. and Idriss et al.. [31,34]
2
1

Table 1 Summary of crystal structure and BET surface area of metal oxides used in this study.
BET
Surface
Area
Catalysts
Crystal Type
m2/g
γ-Al
CeO
CuO
Fe
MgO
SnO
-anatase
TiO -rutile
2
O
2
3
hexagonal
cubic
monoclinic
cubic
56
60
3.2
30
1.5
5.1
9.0
4.9
3.4
3.0
1.3
3.1
2
O
3
cubic
2
tetragonal
tetragonal
tetragonal
orthorhombic
monoclinic
hexagonal Wurtzite
monoclinic
TiO
2
2
V
2
O
5
WO
3
ZnO
ZrO
2
2
2