Cationic and anionic surface binding sites on nanocrystalline zinc oxide: Surface influence on photoluminescence and photocatalysis

Article abstract of DOI:10.1021/ja808663b

To probe the influence of the surface on the overall nature of zinc oxide nanocrystals (ZnO NCs) this paper examines the effects of surface modifiers: cobalt Co(II) and trimethylsilanolate, on the properties of ZnO NCs. A clear relationship between the su

Full text of DOI:10.1021/ja808663b

Published on Web 03/05/2009
Cationic and Anionic Surface Binding Sites on
Nanocrystalline Zinc Oxide: Surface Influence on
Photoluminescence and Photocatalysis
D. Scott Bohle* and Carla J. Spina
Department of Chemistry, McGill UniVersity, 801 Sherbrooke Street West, Montreal H3A 2K6,
Quebec, Canda
Received November 4, 2008; E-mail: scott.bohle@mcgill.ca
Abstract: To probe the influence of the surface on the overall nature of zinc oxide nanocrystals (ZnO
NCs) this paper examines the effects of surface modifiers: cobalt Co(II) and trimethylsilanolate, on the
properties of ZnO NCs. A clear relationship between the surface, photocatalytic (PC), and photoluminescent
(PL) character of ZnO is observed. With potassium trimethylsilanolate and cobalt(II) acetate we have
determined that anionic binding sites occupied by silanolate contribute to 40% of the PL character of the
defect emission (500-550 nm). Photodegradation of 4-nitrophenol was used as a probe to evaluate the
effect of surface modification on the PC performance of ZnO NCs. At maximum silanolate modification
the PC activity of ZnO was reduced by 50%. Modification of ZnO NCs with Co(II) resulted in the transfer
of photoexcited electrons to the cobalt center where consequent nonradiative recombination, at energies
lower than required for PC, was observed via a comparable decrease in both PL and PC activity. These
results are critical for using ZnO NCs in sensory, photocatalytic, and electronic applications.
Introduction
versial; there are contradictory assignments, and none explain
all of the phenomena observed.9-11 Identifying the physio-
chemical nature of this deep trap emission in ZnO NCs is of
great importance as the UV and visible emission, as well as
the photoreactivity, are all interrelated.
The features of zinc oxide nanocrystals (ZnO NCs) that make
them unique are also those which are not fully understood. The
surface features of nanomaterials vary from the bulk and
introduce new chemistry and challenges when working with
inorganic materials on the nanoscale. In the photoluminescent
spectrum of colloidal zinc oxide nanocrystals (NCs) two major
components are observed, the ultraviolet (UV) and visible.
Photoexcitation of NCs with energies greater than the band gap
Many sources of the visible emission from NCs have been
proposed (see Table 1). The deep trap emission was first
suggested to result from a recombination event between
chemisorbed O2•- (formed by the reduction of adsorbed O2 by
eCB-) on the surface and an intrinsic trapped state in the ZnO
NC.12 Later, it was proposed to be the recombination of a
shallowly trapped electron with a deeply trapped hole.2 Where
the deep trap level is identified as an oxygen vacancy, Vo•• (of
+
results in the creation of an electron-hole, exciton pair (hVB
+ eCB-). The UV band gap emission results from the radiative
recombination of an excited electron in the conduction band
(eCB-) with the valence band hole (hVB+) and exhibits a
dependence on NC size due to quantum confinement. The visible
or deep trap photoluminescence (PL) is commonly defined as
the recombination of the electron-hole pair from localized states
with energy levels deep in the band gap, resulting in lower
energy emission. These alternate energy levels are usually
attributed to dopants, structural features, or surface defects.1-8
However, the site of deep trap emission still remains contro-
•
+2 charge in relation to O2- state), Vo•• is reduced to Vo through
the interaction of eCB- with a surface system O2-.2 More recent
studies suggest that the surface sites facilitating visible photo-
luminescence are hydroxides (OH- ).6 Although there has been
s
much research on the topic, none of the current theories fully
explain all of the observed phenomena associated with the
luminescent qualities of ZnO NCs. In this paper we examine
(1) Van Dijken, A.; Meulenkamp, E. A.; Vanmaekelbergh, D.; Meijerink,
A. J. Phys. Chem. B 2000, 104, 1715–1723.
(2) van Dijken, A.; Meulenkamp, E. A.; Vanmaekelbergh, D.; Meijerink,
A. J. Phys. Chem. B 2000, 104, 4355–4360.
(3) Bahnemann, D. W.; Kormann, C.; Hoffmann, M. R. J. Phys. Chem.
1987, 91, 3789–3798.
(9) Schmidt-Mende, L.; MacManus-Driscoll, J. L. Mater. Today (Oxford,
U. K.) 2007, 10, 40–48.
(4) Djurisic, A. B.; Leung, Y. H. Small 2006, 2, 944–961.
(5) Lin, P.-F.; Ko, C.-Y.; Lin, W.-T.; Lee, C. T. Mater. Lett. 2007, 61,
1767–1770.
(10) Ozgur, U.; Alivov, Y. I.; Liu, C.; Teke, A.; Reshchikov, M. A.; Dogan,
S.; Avrutin, V.; Cho, S. J.; Morkoc, H. J. Appl. Phys. 2005, 98, 041301/
1-041301/103.
(6) van Dijken, A.; Meulenkamp, E. A.; Vanmaekelbergh, D.; Meijerink,
A. J. Lumin. 2000, 87-89, 454–456.
(11) Zhou, X.; Kuang, Q.; Jiang, Z.-Y.; Xie, Z.-X.; Xu, T.; Huang, R.-B.;
Zheng, L.-S. J. Phys. Chem. C 2007, 111, 12091–12093.
(12) Lin, Y.; Wang, D.; Zhao, Q.; Li, Z.; Ma, Y.; Yang, M. Nanotechnology
2006, 17, 2110–2115.
(7) Norberg, N. S.; Gamelin, D. R. J. Phys. Chem. B 2005, 109, 20810–
20816.
(8) van Dijken, A.; Meulenkamp, E. A.; Vanmaekelbergh, D.; Meijerink,
A. J. Lumin. 2000, 90, 123–128.
(13) Vanheusden, K.; Warren, W. L.; Seagar, C. H.; Tallant, D. R.; Voight,
J. A.; Gnade, D. E. J. Appl. Phys. 1996, 79, 7983–7990.
9
10.1021/ja808663b CCC: $40.75
2009 American Chemical Society
J. AM. CHEM. SOC. 2009, 131, 4397–4404 4397

A R T I C L E S
Bohle and Spina
Table 1. Identification of Recombination Source for the Visible
tion on the PL and photochemistry of ZnO NCs. Through
modification of the surface we are able to determine the types
of surface sites that contribute to the PL and their reactivity,
and to quantify these sites through PL studies and photocatalysis
studies.32,33 Potassium trimethylsilanolate, (Me3SiO-K+), and
cobalt (Co(II)) are used together to investigate electronic and
surface interactions and the consequent effects on the PL
spectrum and photochemical properties of ZnO NCs. In this
paper we report the quenching dynamics of cobalt (Co(II)) and
potassium trimethylsilanolate, the dependence of photocatalytic
processes on surface modification, and the possible ZnO surface
sites involved in both photoluminescence and photocatalysis.
It is clear from our investigations that the surface sites of ZnO
NCs must be considered inhomogeneous with both anionic and
cationic sites of different coordination numbers and geometries,
some of which contribute to the long disputed visible or defect
emission.
Emission of ZnO NCs
visible emission (eV)
proposed sourcea
refs
1.62 -2.0
2.1
1.95
2.1
2.1
2.11
2.2
2.38
2.4
2.48
2.61
2.8
e
CB′ + Vo
14-16
14
2, 15, 17
14
18-20
7
1, 2, 8
21, 22
16
3
23
•
•
Vo + hVB
x
•
Vo + Vo
Zn′′ + VZn
Zni• + VZn
V
′
′
•
OHS′ + hVB
e
e
Vo:Zni + hVB
O2′ + Tr
••
CB′ + Vo
CB′ + OZn
•
•
e
V
CB′ + Vo
x
•
Zn′/VZn + hVB
17
a Kro¨ger-Vink notation used for proposed source of visible emission
in Table 1.17 In the standard notation Xab, X corresponds to the species
of interest, which may be an electron (e), a hole (h), a vacancy (V), or
any atom. Within Xab, the ‘a’ term is the lattice site represents the lattice
site occupied by species X. If the species occupies a surface site, then a
) s. If the occupied site is interstitial, a ) i. The ‘b’ term represents the
value of the electric charge relative to the original site. For a null
charge, b ) x, for a single negative charge b ) ′, and for a single
positive charge, b ) •.
Experimental Procedures
Sample Preparation. Preparation of ZnO NCs. Colloidal
nanocrystalline zinc oxide (ZnO NC) is prepared at room temper-
ature by the slow addition of tetramethylammonium hydroxide
pentahydrate solution in ethanol (45 µmol, 81 µL) over 1.5 h to a
rapidly stirring solution of zinc acetate dihydrate (25 µmol, 250
µL in DMSO) in DMSO (2.5 mL). Once the addition is complete,
the solution is sealed and left to stir overnight to complete ripening
of the nanocrystals. ZnO NC free of dioxygen modifications, are
prepared by performing the reaction in an inert atmosphere box
under N2(g) with thoroughly degassed reagents and solvents.
Ultrapure ZnO NC are prepared by the same procedure, by replacing
the reagent grade zinc acetate with Zn(OAc)2 ·2H2O (99.999% metal
basis). For solid-state analysis nanocrystals were isolated from the
DMSO solution via precipitation with ethyl acetate.
the defect emission though surface modification studies to
expand the fundamental understanding of the source of defect
PL.
Modification of the NC surface has been noted to influence
the PL, catalytic, and electronic properties of ZnO NCs.7,24-28
It is crucial to understand the chemistry occurring at any
nanocrystalline surface in order to control and tune the properties
of the nanomaterial. To date, NC surface characterization and
the correlation of structure with observed properties remain
important unresolved problems. One method used to probe
nanomaterial exterior involves exploring the interactions of
adsorbates on the ZnO NC surface.29-31 Organic modification,
thermal passivation, or annealing the surface under a reducing
atmosphere of hydrogen results in the reduction, or near
elimination, of the visible fluorescence.7 We choose to use
surface modification techniques to probe the effects of adsorp-
Preparation of Me3SiO-K+-Modified ZnO NCs. Modification
of the nanoparticle surface with potassium trimethylsilanolate was
achieved by the post-synthesis addition of various quantities of an
ethanolic solution of 0.24 M Me3SiO-K+ to the colloidal solutions.
Preparation of Co(II)-Modified ZnO NCs. These are prepared
by adding Co(OAc)2 ·5H2O (5.22 mM, in DMSO) to the freshly
prepared colloidal ZnO NC solution. All samples were characterized
by ultraviolet-visible (UV-vis) and photoluminescent spectro-
scopic techniques.
Nanocrystal Reactivity. Oxygen Reactivity. The response of
ZnO NCs to oxygen was investigated utilizing the Me3SiO-K+-
modified and unmodified nanocrystals prepared under an inert
nitrogen atmosphere. Ethanol (2.5 mL) and ZnO NCs (2.3 µmol,
250 µL) were combined in a Schlenk quartz cuvette under inert
atmosphere. Aliquots of O2(g) were added to the cuvette containing
the ZnO NC solution via a microsyringe. During the additions the
photoluminescence of the nanocrystals was monitored spectroscopi-
cally. After each added aliquot of O2(g), the ZnO NC solutions were
monitored until an unvarying emission spectrum was acquired. The
above procedure was repeated using silanolate-modified nanocrys-
tals to observe the effect of surface modification on dioxygen
addition.
(14) Lima, S. A. M.; Sigoli, F. A.; Jafelicci, M., Jr.; Davolos, M. R. Int.
J. Inorg. Mater 2001, 3, 749–754.
(15) Sukkar, M. H.; Johnson, K. H.; Tuller, H. L. Mater. Sci. Eng., B 1990,
B6, 49–59.
(16) Xu, P. S.; Sun, Y. M.; Shi, C. S.; Xu, F. Q.; Pan, H. B. Nucl. Instrum.
Methods Phys. Res., Sect. B 2003, 199, 286–290.
(17) Kro¨ger, F. A. The Chemistry of Imperfect Crystals, 2nd revised ed.;
North-Holland Publishing Co: Amsterdam, 1973.
(18) Ramanachalam, M. S.; Rohatgi, A.; Carter, W. B.; Schaffer, J. P.;
Gupta, T. K. J. Electron. Mater. 1995, 24, 413–19.
(19) Liu, M.; Kitai, A. H.; Mascher, P. J. Lumin. 1992, 54, 35–42.
(20) Riehl, N. J. Lumin. 1981, 24-25, 335–341.
(21) Lin, B.; Fu, Z.; Jia, Y. Appl. Phys. Lett. 2001, 79, 943–945.
(22) Wand, D.-F.; Liao, L.; Li, J.-C.; Fu, Q.; Peng, M.-Z.; Zhou, J.-M.
Chin. Phys. Lett. 2005, 22, 2084–2087.
(23) Kamat, P. V.; Patrick, B. J. Phys. Chem. 1992, 96, 6829–6834.
(24) Yang, C. L.; Wang, J. N.; Ge, W. K.; Guo, L.; Yang, S. H.; Shen,
D. Z. J. Appl. Phys. 2001, 90, 4489–4493.
Photocatalytic Degradation. The photocatalytic activity of the
silanolate-modified and unmodified ZnO NCs was examined by
monitoring the photocatalytic degradation of 4-nitrophenol. Into a
quartz cuvette were added ethanol (2.5 mL) and a ZnO solution
(100 µL, 1 µmol, in DMSO). To the cuvette was then added
4-nitrophenol (0.17 µmol, 33.5 mM in ethanol), and the stirred
solution was irradiated using a Mercury-lamp (450 W) for one
(25) Chang, M.; Cao, X. L.; Zeng, H.; Zhang, L. Chem. Phys. Lett. 2007,
446, 370–373.
(26) Gong, Y.; Andelman, T.; Neumark, G. F.; O’Brien, S.; Kuskovsky,
I. L. Nanoscale Res. Lett. 2007, 2, 297–302.
(27) Soares, J. W.; Whitten, J. E.; Oblas, D. W.; Steeves, D. M. Langmuir
2008, 24, 371–374.
(28) Tang, E.; Tian, B.; Zheng, E.; Fu, C.; Cheng, G. Chem. Eng. Commun.
2008, 195, 479–491.
(29) Liu, Y.; Tong, Y. J. Nanosci. Nanotechnol. 2008, 8, 1101–1109.
(30) Serpone, N.; Dondi, D.; Albini, A. Inorg. Chim. Acta 2007, 360, 794–
802.
(32) Bowers, G. N., Jr.; McComb, R. B.; Christensen, R. G.; Schaffer, R.
Clin. Chem. 1980, 26, 724–729.
(33) Zhang, W.; Xiao, X.; An, T.; Song, Z.; Fu, J.; Sheng, G.; Cui, M.
J. Chem. Technol. Biotechnol. 2003, 78, 788–794.
(31) Woell, C. Prog. Surf. Sci. 2007, 82, 55–120.
9
4398 J. AM. CHEM. SOC. VOL. 131, NO. 12, 2009

Surface Influence on Photoluminescence and Photocatalysis
A R T I C L E S
Figure 1. Effect of the addition of cobalt (Co(II)) and potassium
trimethylsilanolate (Me3SiO-K+) on the relative photoluminescent intensity
of ZnO NCs λex ) 342 nm.
minute intervals followed by monitoring the photodegradation of
4-nitrophenol with UV-vis spectroscopy (λmax ) 401 nm). The
previous photocatalytic degradation procedure was repeated using
ZnO NCs modified with various concentrations of Co(II) and
Me3SiO-K+.
Figure 2. Effect of the addition of dioxygen to modified (with Me3SiO-K+
- dashed) and unmodified (solid) ZnO NCs anaerobically prepared. PL trends
observed above for the % relative emission intensity.
Physical Measurements. With the exception of the solvents
employed for inert atmosphere syntheses, all chemicals and solvents
were of reagent grade and used without further purification. All
reagents or glassware that were to be used in the glovebox were
placed under vacuum for 24 h and then transferred and stored in
the inert atmosphere box for later use. Deoxygenation of solutions
and solvents was performed by either distillation or freeze-
pump-thaw. ZnO NC reaction solutions were used directly in
further reactions with concentrations of approximately 9 µM.
Absorption spectra of samples were obtained with a HP 8453
UV-visible diode array spectrophotometer. Fluorescent spectra
were recorded using a FluoroMax 2 (ISA) Jobin Yvon-SPEX
spectrofluorometer, with a constant excitation wavelength of 337
nm. Unless otherwise stated, the onset band-edge absorption spectra
of the ZnO nanoparticles was 337 nm, and the observed emission
spectra contained two peaks, a UV peak at 355 nm, and a broad
visible peak at ∼530 nm, the latter of which is dominant in aerated
solution. Photocatalytic experiments were carried out using a Oriel
Instruments Hg-lamp (λ ) 254 nm) model 66033 as the excitation
source.
Comparison of the effect of oxygen on Me3SiO-K+-modified
and unmodified anaerobically prepared ZnO NC PL is shown
in Figure 2. Silanolate modification resulted in a more rapid
restoration and earlier plateau in the visible, compared to the
PL of unmodified ZnO upon dioxygen addition. The plateau
reached in the visible emission spectrum of silanolate-modified
ZnO is attributed to the presence of adsorbed silanolate anions
occupying surface sites or acting on adjacent sites which would
have otherwise contributed to visible luminescence in the
absence of silanolate. The inability of silanolate to interfere with
all potential O2 binding sites reinforces the presence of more
than one type of surface site. Silanolate adsorption on the ZnO
surface modifies and quenches one type of the surface sites,
whose fluorescent contribution is approximately 40% of the total
visible PL. The remaining sites have a higher efficiency for
visible recombination or affinity to oxygen as they exhibit a
more rapid increase of the visible emission of silanolate-
modified ZnO NCs, Figure 2, dashed line. The greater affinity
of oxygen to ZnO after silanolate addition is thought to be due
to variation in the geometry or energetics of the defect sites
adjacent to the modified sites. These stronger affinity sites
contribute to the other 60% of the total visible PL. Initially the
unmodified ZnO NCs require a greater quantity of oxygen for
equal visible luminescent intensity, with respect to modified
NCs. Addition of oxygen to unmodified NCs however, results
in no discrete plateau for the visible emission at lower oxygen
levels. This is indicative of the presence of a more diverse
surface system with dissimilar binding affinities than those
present on the modified ZnO NC surface. The chemistry of sites
occupied by Me3SiO-K+, and those remaining after modification
was further investigated through analysis of the photochemical
activity of the ZnO NCs.
Results
As shown in Figure 1, the addition of Co(II) to ZnO NCs
results in an exponential decay of the PL intensity. Modifying
the surface with trimethylsilanolate results in only 40% quench-
ing of ZnO visible PL emission, and 65% of the UV. A red-
shift in the absorption spectrum was also observed for ZnO NCs
in the presence of Co(II), Supporting Information, Figure S2b,
and no significant shift was observed in the absorption spectrum
of silanolate-modified ZnO NCs, Supporting Information, Figure
S2a. The limited quenching is interpreted as resulting from a
limited number of silanolate binding sites on the ZnO surface,
where the partial adsorption of silanolate results in partial
quenching. As trimethylsilanolate is observed to only partially
quench the ZnO visible emission, we chose to utilize the
silanolate-modified ZnO (ZnO:[-OSiMe3], (1:1), with 40% and
65% of the visible and UV PL being quenched, respectively)
system to further investigate the electronic, chemical, and
physical properties of the defect states which contribute the PL
of ZnO NCs.
Photoactivity of nanomaterials inherently relies on surface
properties. Nanomaterials are unique in that they efficiently
separate photoinduced charges, have greater solubility, and have
larger surface to volume ratios, all advantageous attributes for
photocatalysts.23,35-39 Control of photocatalytic activity has
previously been possible through internal and external modifica-
Previous studies have established the effect of oxygen on ZnO
photoluminescence, where dioxygen simultaneously promotes
the visible luminescence while quenching the UV (band gap)
emission2,3,12,34 Although the exact role of oxygen in ZnO PL
remains open, it is nonetheless a functional tool in the
characterization, and quantization of ZnO surface structures.
(35) Bell, A. T. Science 2003, 299, 1688–1691.
(36) Dodd, A. C.; McKinley, A. J.; Saunders, M.; Tsuzuki, T. J. Nanopart.
Res. 2006, 8, 43–51.
(37) Koch, U.; Fojtik, A.; Weller, H.; Henglein, A. Chem. Phys. Lett. 1985,
122, 507–510.
(38) Pearton, S. J.; Norton, D. P.; Ip, K.; Heo, Y. W.; Steiner, T.
Superlattices Microstruct. 2003, 34, 3–32.
(34) Bohle, D. S.; Spina, C. J. J. Am. Chem. Soc. 2007, 129, 12380–12381.
9
J. AM. CHEM. SOC. VOL. 131, NO. 12, 2009 4399

A R T I C L E S
Bohle and Spina
tion of nanomaterials. The inclusion of divalent metal dopants,
including Ni(II), Co(II), and Mn(II), is known to decrease the
photocatalytic (PC) activity of ZnO NCs.40 The electronic
activity of ZnO NCs may also be altered with surface modifica-
tion as previously shown in the case of silane-modified ZnO.27,41
The relationship between the NC surface, PL, and PC activity
is exemplified in the inverse relationship between particle size
and both visible PL intensity and PC activity.8,36,40,42 This
relationship between PC activity and PL and the ZnO NCs
surface was elucidated by monitoring both PL and PC while
varying the surface modifier concentration and identity. Quan-
tification of the photochemical activity of ZnO NCs was
accomplised via photodegradation of 4-nitrophenol (4NP), a
spectrophotometric reference material.32,33,43 Changes in the PC
activity and PL with the addition of trimethylsilanolate and
Co(II) were utilized to investigate the structure and character
of the surface sites.
dependence on the surface to volume ratio as well as electronic
confinement.44-46 Consideration of the defects or strains in the
crystal lattice due to size constraints is essential to explain the
physical and chemical phenomena associated with the nano-
structures. We have demonstrated that the prepared ZnO NCs
have a P63mc wurtzite hexagonal lattice from the synchrotron
X-ray powder diffraction patterns, Supporting Information,
Figure S8. In determining the significance and relationship of
the surface with respect to the emission properties of ZnO, one
must first examine the possible geometries which may be present
on the nanocrystalline surface. The surface is important to
consider, as interaction of adsorbates on the surface significantly
influence the photochemical properties of the ZnO NCs.26,47,48
The most common face terminations of bulk ZnO include the
polar Zn-terminated (0001) and O-terminated (000-1) faces (c-
axis oriented), and the nonpolar (11-20) faces (a-axis), and (10-
10) faces which both contain an equal number of Zn and O
atoms.49 It is known that the faces of [0001], a polar orientation
and [01-10], a nonpolar face, have anisotropic growth rates (ν):
ν [0001] . ν [01-10] > ν [000-1]. The stability of the surfaces
is also anisotropic where the polar faces have greater stability
than the nonpolar faces. An inherent polarity exists in the crystal,
where the character of the Zn-O bond is strongly ionic. The
PC activity of the individual faces has been investigated, and
the (0001) and (000-1) faces are found to be largely responsible
for the majority of the PC activity of ZnO in terms of H2O2
production.50 It has also been noted that defects on the (10-10)
face may promote dissociative adsorption of oxygen where a
Photodegradation of 4-nitrophenol (4NP) by ZnO NCs is a
known first-order PC process, Supporting Information, Figure
S3, Scheme S4.32,33,43 The PC rates of ZnO NCs, modified and
unmodified, were calculated by monitoring the UV-visible
absorbance (λmax ) 401 nm, ε ) 19200 M-1 cm-1) of 4NP,
described by eqs 1 and 2.
-dC/dt ) k1C
ln(Co/Ct) ) k1t
(1)
(2)
The rates of photodegradation and PL intensities were
compared at variable adsorbent concentrations, Figure 3. Cobalt
was found to be considerably more effective in inhibiting ZnO
PC activity. The addition of 1% Co(II) results in about a 40%
decrease in both PC capacity and PL. Similar trends are observed
for both the PC rates and PL of ZnO NCs upon the addition of
Me3SiO-K+ with a 50% and 40% decrease in activity, respec-
tively, after modification with one equivalent of Me3SiO-K+.
Although the PL and PC activity are interrelated, the relationship
is not direct. Whether there is modification of the ZnO NC
surface by trimethylsilanolate anion or Co(II), the result is a
decrease in both PL and PC activity, and a clear correlation is
observed. Understanding this relationship between the PL and
PC activity will provide insight into the origin of the defect
emission and the chemistry behind the photocatalysis which will
advance our ability to utilize ZnO NCs.
spontaneous 2e- charge transfer from Vo to O2 (pπ*) will
x
facilitate dissociation.51 Oxygen dissociation results in “filling”
oxygen vacancies that are present, resulting in a “healing” of
the surface with an expected change in electronic properties.
This type of surface alteration upon photoexcitation has been
investigated by us previously.34 Discrete facets are not the only
thing to consider when dealing with ZnO NC surface structure.
It is also important to account for lattice perturbations due to
the inherent curvature that is not described for bulk ZnO. These
lattice truncations and surface curvature must be taken into
greater consideration in nanostructures even though they are
rarely examined in structural detail.9,15,31,52,53 From the observa-
tions described above it is obvious that there is more than one
site contributing to the visible emission. We propose that the
effects on photocatalysis and photoluminescence are due not
only to electronics but also to specific surface interactions. To
investigate the unique surface interactions, silanolate and cobalt
Discussion
As a bulk material the ZnO wurtzite lattice is composed of
three different geometric faces, two polar surfaces of opposite
charge: (0001) and (000-1), and a neutral surface: (11-20),
Supporting Information, Figure S5. However, the size constraints
imposed by nanomaterials causes these faces to become skewed
in nanoparticles and may result in a core and surface structure
different from those of the bulk phase. The dependence on the
PC and PL features based on size of the NCs indicates
(44) Cheng, H.-M.; Lin, K.-F.; Hsu, H.-C.; Hsieh, W.-F. Appl. Phys. Lett.
2006, 88, 261909/1–261909/3.
(45) Fu, Z. D.; Cui, Y. S.; Zhang, S. Y.; Chen, J.; Yu, D. P.; Zhang, S. L.;
Niu, L.; Jiang, J. Z. Appl. Phys. Lett. 2007, 90, 263113/1–263113/3.
(46) Lin, K.-F.; Cheng, H.-M.; Hsu, H.-C.; Hsieh, W.-F. Appl. Phys. Lett.
2006, 88, 263117/1–263117/3.
(47) Hong, R.; Pan, T.; Qian, J.; Li, H. Chem. Eng. J. 2006, 119, 71–81.
(48) Wu, Y. L.; Tok, A. I. Y.; Boey, F. Y. C.; Zeng, X. T.; Zhang, X. H.
Appl. Surf. Sci. 2007, 253, 5473–5479.
(49) Jagadish, C.; Pearton, S. Zinc Oxide Bulk, Thin Films and Nanostruc-
tures: Processing, Properties and Applications; Elsevier: Amsterdam,
2006.
(39) Polyakov, A. Y.; Smirnov, N. B.; Govorkov, A. V.; Kozhukhova,
E. A.; Pearton, S. J.; Norton, D. P.; Osinsky, A.; Dabiran, A. J.
Electron. Mater. 2006, 35, 663–669.
(50) Jang, E. S.; Won, J.-H.; Hwang, S.-J.; Choy, J.-H. AdV. Mater.
(40) Casey, P. S.; Rossouw, C. J.; Boskovic, S.; Lawrence, K. A.; Turney,
T. W. Superlattices Microstruct. 2005, 39, 97–106.
(41) Rohe, B.; Veeman, W. S.; Tausch, M. Nanotechnology 2006, 17, 277–
282.
(Weinheim, Ger.) 2006, 18, 3309–3312.
(51) Ischenko, V.; Polarz, S.; Grote, D.; Stavarache, V.; Fink, K.; Driess,
M. AdV. Funct. Mater. 2005, 15, 1945–1954.
(52) Shen, X.; Pederson, M. R.; Zheng, J.-C.; Davenport, J. W.; Muckerman,
J. T.; Allen, P. B. Los Alamos Natl. Lab., Prepr. Arch., Condens.
Matter 2006, 112; http://arxiv.org/abs/cond-mat/0610002; arXiv:cond-
mat/0610002 [cond-mat.mtrl-sci].
(42) Zheng, Y.; Chen, C.; Zhan, Y.; Lin, X.; Zheng, Q.; Wei, K.; Zhu, J.;
Zhu, Y. Inorg. Chem. 2007, 46, 6675–6682.
(43) Marais, E.; Klein, R.; Antunes, E.; Nyokong, T. J. Mol. Catal. A:
Chem. 2007, 261, 36–42.
(53) Cheng, W. H.; Kung, H. H. Surf. Sci. 1981, 102, L21–L28.
9
4400 J. AM. CHEM. SOC. VOL. 131, NO. 12, 2009

Surface Influence on Photoluminescence and Photocatalysis
A R T I C L E S
Figure 3. Comparing the photoluminescence (PL) with the rate of photocatalysis of 4-nitrophenol (PC) and with varying the percent addition, with respect
to ZnO, of either (a) Co(II) or (b) (Me)3SiO-K+.
(II) ions are used to probe the effects of cationic and anionic
surface sites on the characteristics of ZnO NCs, respectively.
The origin of the green emission, λ ) 530 nm, in ZnO NCs
has long been debated due to its complex nature.1-4,7,8,14,16,37
A variety of assignments for the electronic transition responsible
for the broad visible emission, as shown in Table 1, involve
vacancies and interstitial atoms inherently present in imperfect
ZnO crystals. Other assignments involve non-innate -OH atoms
present on the surface of the nanocrystal. Assigning the
transitions involved in the green emission to a single-point defect
is in contradiction with studies proposing donor-acceptor
transitions and is an unlikely solution to the complex green
emission.54 To better understand the green emission we have
utilized capping agents and dopants to manipulate the ZnO NC
surface, electronic structure, and photoluminescence. In choosing
the anion for surface site binding we opted for a bulky robust
anion with no R or ꢀ C-H bonds. Trimethylsilanolate was
chosen over tert-butoxide due to the relative inertness of the
Si-C bonds and its lower steric bulk. Quenching the visible
emission was achieved by addition of specific external or surface
modifiers. In the dissociated form, anionic trimethylsilanolate
caps cationic ZnO NC surface sites and quenches about half
the visible emission intensity. Incomplete PL quenching indi-
cates a limited interaction between trimethylsilanolate and the
ZnO NC surface sites responsible for PL. The geometry of this
binding is likened to the siloxy-ZnO aggregate structures
prepared by Rell et al.55,56 Silanolate interaction with the ZnO
NC surface is proposed to be through a surface oxygen vacancy
at one, two, or three Zn-coordination sites, Figure 4. Some or
all of these sites are believed to be involved in the recombina-
tion, contributing to ∼40% of the visible PL. As observed by
X-ray powder diffraction, Supporting Information, Figure S8,
the adsorption of trimethylsilanolate on the ZnO NC surface
does not alter the size or crystal structure of the nanoparticles,
where only a surface interaction is inferred to be responsible
for the changes in ZnO character. The involvement of these
sites in PL is reinforced by the effect of silanolate-bound ZnO
NCs under anaerobic conditions, where titration of oxygen into
the closed system results in a more rapid, although incomplete,
restoration of the visible PL, Figure 2. The change in ZnO PL
response to oxygen is indicative of the presence of an absorbed
species on the surface modifying O2 adsorption and blocking
potential oxygen binding sites. Changes in the PL intensity upon
silanolate titration into aerobic ZnO NCs were analyzed using
a Stern-Volmer plot, revealing a nonlinear trend. This result
indicates that the quenching of trimethylsilanolate on ZnO NCs,
Supporting Information, Figure S9, does not result from simple
dynamic quenching in support of the above theory of silanolate
binding on the ZnO NC surface. In an attempt to quantify the
surface adsorbents we determined the Langmuir adsorption
isotherms and Stern-Volmer quenching kinetics to relate the
PL intensity to the fraction of occupied surface sites, as
previously described by Munro et al. in the case of CdSe.57
The PL quenching of ZnO NCs is fit to a Langmuir adsorption
isotherm, Supporting Information, Figure S10.57-61 From the
Langmuir fit, assuming homogeneous adsorption, the binding
constant was calculated (Ka,(Me) SiO ) 2500 M-1), representing
-
3
a relatively weak binding between the silanolate and ZnO NC
surface. Utilizing the average ratio of ZnO and trimethylsil-
anolate concentrations at saturation, Supporting Information,
Figure S11, we obtained an approximate value of 0.17 for the
fraction of total ZnO participating in the surface interaction with
trimethylsilanolate. In terms of surface fraction we consider the
surface area/volume ) 0.24 for ZnO NCs (radius ) 2.7 nm),
Supporting Information, Figure S12, where ∼70% of potential
ZnO surface sites are occupied by trimethylsilanolate at satura-
tion. Photocatalytic results may then be explained by the same
adsorption effect. Trimethylsilanolate interferes with the sites
and the quantity of electrons available for photocatalysis. This
is observed in Figure 3b, where in the presence of various
quantities of silanolate, a decreasing trend is observed for both
the photodegradation rate of 4NP, and the relative intensity of
the visible emission. Reduced degradation of the dye is due to
occupation of surface sites by silanolate as observed by a
decrease in both the rate of PC and in the visible emission,
suggesting a relationship between the site responsible for the
defect emission and that associated with photodegradation. The
addition of an anionic surface modifier will inherently only alter
cationic sites, where cationic modifying agents must be used
(54) Djurisic, A. B.; Choy, W. C. H.; Roy, V. A. L.; Leung, Y. H.; Kwong,
C. Y.; Cheah, K. W.; Rao, T. K. G.; Chan, W. K.; Lui, H. F.; Surya,
C. AdV. Funct. Mater. 2004, 14, 856–864.
(55) Driess, M.; Merz, K.; Rell, S. Eur. J. Inorg. Chem. 2000, 2517–2522.
(56) Merz, K.; Hu, H.-M.; Rell, S.; Driess, M. Eur. J. Inorg. Chem. 2003,
51–53.
(57) Munro, A. M.; Plante, I. J.-L.; Ng, M. S.; Ginger, D. S. J. Phys. Chem.
C 2007, 111, 6220–6227.
(58) Kepler, K. D.; Lisensky, G. C.; Patel, M.; Sigworth, L. A.; Ellis, A. B.
J. Phys. Chem. 1995, 99, 16011–16017.
(59) Lorenz, J. K.; Ellis, A. B. J. Am. Chem. Soc. 1998, 120, 10970–10975.
(60) Meeker, K.; Ellis, A. B. J. Phys. Chem. B 1999, 103, 995–1001.
(61) Zhang, J. Z.; Geselbracht, M. J.; Ellis, A. B. J. Am. Chem. Soc. 1993,
115, 7789–7793.
9
J. AM. CHEM. SOC. VOL. 131, NO. 12, 2009 4401

A R T I C L E S
Bohle and Spina
inhibition of the PL and PC character of ZnO indicates that the
sites of silanolate interaction are only a partial contributor of
the PL and PC character of ZnO NCs. The possible geometries
of the site or sites of silanolate binding are shown in Figure 4.
The coordination of silanolate to the surface of ZnO has also
been noted to alter the binding affinity of oxygen to the surface
of ZnO, attributed to chemical or electronic changes of the defect
sites adjacent to the silanolate-bound sites. The action of
silanolate coordination through zinc results in the elimination
of oxygen vacancies that were once present on the ZnO surface
at those sites. The outcome of silanolate occupation of these
oxygen vacancies is a finite decrease in ZnO PL. The contribu-
tion of the oxygen vacancy sites now occupied by silanolate is
assigned to be the origin of the decrease observed in the visible
radiative recombination. Crystal lattice faces that have the
geometry required for silanolate binding are the Zn-capped
(0001), (11-20) (a-axis), and (10-10) faces. We have successfully
shown that the contributions of ZnO defects toward the visible
emission are not due to a sole defect and their identification
may be elucidated through specific modification of the surface.
The contribution of the silanolate-occupied sites toward the
visible PL was calculated to be ∼40% with ∼70% of possible
surface sites on the ZnO NC occupied by trimethylsilanol-
ate anions. Clearly all of the visible emission is not due to
surface defects. The quenching of the visible emission by Co(II)
also occurs through electron transfer mechanisms; additionally,
a residual emission in the visible emission with surface alteration
is always observed.
Figure 4. Potential configurations of the sites occupied by trimethylsil-
anolate on the surface of ZnO NCs. (a) One, (b) two, and (c) three Zn-
coordinate geometries are shown.
Figure 5. Possible coordination geometries for adsorbed Co (II) on the
ZnO NC surface (X ) H, H2, C(O)CH3).
to probe other surface sites. To accomplish this, cationic Co(II)
is utilized to modify the ZnO NC surface.
Co(II) is known to act as a deep trap in II-VI semiconductors,
where excitonic quenching is followed by rapid nonradiative
relaxation to the ground state 4A2(F) f 2E(G) (660 nm), 4A2(F)
62,63
4
4
2
f T1(P) (615 nm), and A2(F) f T1(G) (568 nm).
We
have shown that binding of Co(II) is reversible, as observed in
its effect on ZnO NC PL, Supporting Information, Figure S6.
This suggests that the same site of Co(II) binding is also
involved in the ZnO PL. The geometry of the bound site is
interpreted from UV-visible spectroscopy, Supporting Informa-
tion, Figure S7, to be a tetrahedral environment; a transition
from the octahedral geometry found in solution phase.63 A range
of possible Co(II) binding configurations on the ZnO NC surface
may be hypothesized, Figure 5. The size and crystal structure
of the ZnO NCs are not affected in any significant manner by
the addition of Co(II), as there is no significant broadening of
the Bragg peaks in the X-ray powder diffraction data, Supporting
Information, Figure S8. External doping of ZnO NCs with
Co(II), as shown in Figure 1, to cap anionic surface sites results
in a significant loss of visible PL. The magnitude of Co(II)
quenching when doped internally is enhanced, indicating a
spatial barrier for electron transfer. The quenching capacity of
surface-bound Co(II) is more limited. Examples of such a limited
quenching radius have been observed previously in CdSe NCs.64
The lesser quenching capacity of surface-bound Co(II) may also
indicate an alternate quenching mechanism on the ZnO NC
surface. The same mechanism which quenches visible PL may
be responsible for the decrease in PC activity. Here the decrease
in catalytic efficiency of Co(II)-doped ZnO NCs is similar to
the trend for the visible emission, Figure 3a, indicative of a
direct correlation between the energy levels involved in pho-
tocatalysis and the visible PL. From our data it is clear that the
anionic and cationic surface sites are unique. To understand the
individual influence of these sites on the overall character of
ZnO NCs we need to consider the basic electronic and structural
components of ZnO. In this discussion we look at the changes
induced by the modifiers on the surface of the ZnO NCs.
Silanolate capping of ZnO NCs results in limited suppression
of the PL and PC activity of ZnO NCs. The incomplete
In previous studies ZnO photocatalytic activity was suggested
to originate on the Zn-free (000-1) face and (01-10) face of the
crystal. Initial reactions that occur on the ZnO NC surface are
summarized by equations 3, 4, and 5.34,50,65-69
ZnO+hυ f ZnO(h+VB+eC-B
Zns-O2abs+eC-B f Zns-O•-
)
(3)
(4)
(5)
(6)
2abs
Zns-O•- +eC-B+2H+aq f Zns+H2O2
2abs
OH-s +h+VB f OH•s
Excitation of ZnO NCs, with an energy below the band gap,
results in the separation of an exciton pair (hVB+ + eCB-) eq 3.
When adsorbed oxygen is present on the ZnO NC, reduction
of the adsorbed species may occur, eqs 4 and 5, resulting in
the production of superoxide and peroxides. Surface hydroxyl
and holes have also been cited as being responsible for
photooxidation reactions, and in photoluminescence, where
surface OH- may undergo oxidation to form a hydroxyl radical,
eq 6. 7,70-74 The proposed role of adsorbed bound oxygen (O2abs
)
(65) Hoffman, A. J.; Carraway, E. R.; Hoffmann, M. R. EnViron. Sci.
Technol. 1994, 28, 776–785.
(66) Kurtz, M.; Strunk, J.; Hinrichsen, O.; Muhler, M.; Fink, K.; Meyer,
B.; Woell, C. Angew. Chem., Int. Ed. 2005, 44, 2790–2794.
(67) Polarz, S.; Strunk, J.; Ischenko, V.; van den Berg Maurits, W. E.;
Hinrichsen, O.; Muhler, M.; Driess, M. Angew. Chem., Int. Ed. 2006,
45, 2965–2969.
(62) Hays, J.; Reddy, K. M.; Graces, N. Y.; Engelhard, M. H.; Shuttha-
nandan, V.; Luo, M.; Xu, C.; Giles, N. C.; Wang, C.; Thevuthasan,
S.; Punnoose, A. J. Phys.: Condens. Matter 2007, 19, 266203/1–
266203/24.
(68) Bahnemann, D. W.; Hoffmann, M. R.; Hong, A. P.; Kormann, C. ACS
Symp. Ser. 1987, 349, 120–132.
(69) Jing, L.; Yuan, F.; Hou, H.; Xin, B.; Cai, W.; Fu, H. Sci. China, Ser.
B 2005, 48, 25–30.
(63) Schwartz, D. A.; Norberg, N. S.; Nguyen, Q. P.; Parker, J. M.;
Gamelin, D. R. J. Am. Chem. Soc. 2003, 125, 13205–13218.
(64) Laferriere, M.; Galian, R. E.; Maurel, V.; Scaiano, J. C. Chem.
Commun. (Cambridge, U.K.) 2006, 257–259.
(70) Richard, C.; Boule, P. New J. Chem. 1994, 18, 547–552.
(71) Richard, C.; Boule, P. Sol. Energy Mater. Sol. Cells 1995, 38, 431–
440.
(72) Sato, S.; Kadowaki, T. J. Catal. 1987, 106, 295–300.
9
4402 J. AM. CHEM. SOC. VOL. 131, NO. 12, 2009

Surface Influence on Photoluminescence and Photocatalysis
A R T I C L E S
in the PC process of ZnO NCs, supports the involvement of a
surface-bound oxygen vacancy in ZnO NC PL and PC
activity.3,68,69 The parallel observations for PL and PC activity
of ZnO NCs with bound silanolate moieties may not be solely
explained by the elimination of oxygen vacancies (Vo) on a Zn-
free crystal face. It has been determined through oxygen addition
to anaerobically prepared ZnO NCs that a combination of sites
contributes to the total visible PL.34 Occupation of surface Vo
by trimethylsilanolate results in the loss of a site of visible
recombination and also loss in the catalytic activity associated
with that site. A relationship between PL and PC activity has
previously been observed. In the comparison of nanobelts, rods,
and films of ZnO it has been shown that trends in UV:visible
emission intensities are comparable to the PC activity in the
order belt > rod > film.75 Anion binding or oxygen vacancies
present on the ZnO NC surface appear to be pivotal for a fraction
of both the PL and PC activity, both of which are affected to
some degree by the adsorption of silanolate on the NC surface.
Investigations of ZnO NCs chemistry (specifically at the NC/
solution interface) and the role of surface structures in the
chemical and PL properties of ZnO will provide enough
evidence to create a new theory to better explain the physio-
chemical nature of ZnO NCs. Determination of the chemical
characteristics responsible for the photophysical properties and
electronic structure of ZnO NCs is a fundamental step toward
optimizing these nanomaterials for use in photocatalytic,
sensory, and electrochemical applications. Diameter, band gap
energy, redox potential, photocatalysis, and photoluminescence
are interrelated properties of nanocrystals.76 The shape, size,
and exposed surface planes of semiconductor nanocrystals all
play a part in the PC and PL nature of ZnO NCs. Few reports
are available on the relationship between the PL and PC of ZnO
NCs. In those investigations the visible emission, surface area,
or defect sites have been implicated as a measure of the
photocatalytic activity of ZnO NCs,42,76-78 although in current
literature, there is no consensus as to the origins and therefore
the relationship between PL and PC character. Photoexcitation
of ZnO NCs with energies greater than the band gap results in
the creation of an exciton pair. A variety of pathways are then
available to this exciton as shown in Figure 6; nonradiative
decay (to either ground state or a trap state) and radiative
recombination produce UV emission or visible emission, and
nonradiative electron or hole transfer to species near or on the
NC surface. It is the sum of these interrelated pathways that
we have observed, culminating in the overall PL and PC
characteristics of ZnO NCs. The PLUV emission results from
the direct recombination of the electron and hole, but this is a
minor emission under ambient conditions. It is understood that
in ZnO NCs there is an enhanced separation of charge carriers,
often observed as an increase in PLVIS and photocatalytic
activity.75,76 The lower energy surface defects promote the
separation of charge carriers, or the transfer of electrons from
the conduction band to a sub-band responsible for visible
emission. The increased lifetime of the charge carriers is
Figure 6. Photochemical pathways available to the exciton pair after
photoexcitation (hν) of ZnO NCs. (a) Nonradiative recombination to ground
state, (b) nonradiative recombination to a trap state or trap state to ground
state, (c) radiative UV or band gap recombination, (d) radiative trap state
recombination, (e) nonradiative redox reactions.
advantageous for photochemical reactions as it enhances the
efficiency of interfacial charge transfer. The proposed origin of
the sub-band, the surface defects, must then be present for visible
recombination to occur. When the sites are occupied by
adsorbates, this mechanism of recombination is no longer
available, and a decrease in the visible emission is observed.
The static quenching of silanolate indicates that there is a specific
interaction of the adsorbent with the surface which is responsible
for the decrease in the visible emission. The similar decrease
observed in the rate of photocatalysis upon surface modification
supports the involvement of these same surface defect sites in
the PC process. Excitation of surface-modified ZnO NCs results
in the redirection of the electrons and holes which previously
were involved in PC and PL activity. Nonradiative recombina-
tion processes are then identified to account for the lost electron
recombination pathways as no increase in the UV emission is
observed for either silanolate or Co(II) modification (Figure 6
b). It is clear from our results that there is a relationship between
the PC, PL, and the surface of ZnO NCs. Adsorption of Co(II)
and trimethylsilanolate on ZnO NCs result in changes in the
PL and PC activity, indicative of both anionic and cationic
features on the ZnO surface playing roles in the complicated
surface photochemistry.
Conclusions
There are many current theories about the source of ZnO
visible fluorescence. However, none of the previous theories
explains all of the phenomena observed. As materials approach
the nanoregime, it is well-known that the characteristics can
change and that many of these changes are due to quantum
confinement. However, another feature of nanomaterials is their
large surface to volume ratio. It is not surprising then that
physiochemical properties of ZnO NCs rely heavily on the
surface which plays a large role in the changes observed as
materials are scaled down to the nanoregime. The source of
the controversial visible or defect emission has been theorized
to originate from localized states in ZnO NCs. These states may
either be found in the bulk or at the surface of a semiconductor.
Defect emissions are strongly influenced by the chemical nature
of the defect site and the surrounding environment. Our previous
research indicated a strong influence of the surface features on
the visible emission. Although the precise determination of the
source of the visible emission of ZnO NCs is an ongoing
process, exploitation of carefully chosen surface modifiers has
brought us closer to a complete picture of the emission profile
of ZnO NCs. Then with adsorbates of different charge and
geometry we are able to observe the relationship between
anionic and cationic sites on the nanoparticle surface and the
visible emisson. By isolating these sites we have also observed
(73) Darwent, J. R.; Lepre, A. J. Chem. Soc., Faraday Trans. 2 1986, 82,
1457–1468.
(74) Dodd, A.; McKinley, A.; Saunders, M.; Tsuzuki, T. Nanotechnology
2006, 17, 692–698.
(75) Sun, T.; Qiu, J.; Liang, C. J. Phys. Chem. C 2008, 112, 715–721.
(76) Ye, C.; Bando, Y.; Shen, G.; Golberg, D. J. Phys. Chem. B 2006,
110, 15146–15151.
(77) Jing, L.; Qu, Y.; Wang, B.; Li, S.; Jiang, B.; Yang, L.; Wei, F.; Fu,
H.; Sun, J. Sol. Energy Mater. Sol. Cells 2006, 90, 1773–1787.
(78) Xu, F.; Yuan, Z.-Y.; Du, G.-H.; Ren, T.-Z.; Bouvy, C.; Halasa, M.;
Su, B.-L. Nanotechnology 2006, 17, 588–594.
9
J. AM. CHEM. SOC. VOL. 131, NO. 12, 2009 4403

A R T I C L E S
Bohle and Spina
the same effects on PC activity of ZnO NCs. The surface is
fundamental to consider, as interaction of adsorbates on the
surface significantly influences both the PL and photochemical
properties of the ZnO NCs. This interaction may be utilized to
bring us closer to understanding the true nature of the visible
emission. We observed that trimethylsilanolate binding, with
the possible coordination geometries mentioned above, may
block anion binding surface sites which are responsible for 40%
of the visible emission and 50% of PC activity. Although not
all of the surface anion sites are blocked by silanolate, the
modification of the surface also increases the binding affinity
of oxygen to the surface. There are cation binding sites on the
surface where Co(II) dynamically quenches and where electron
transfer may occur, reducing both the PL and PC activity of
ZnO NCs. Further study is planned for investigation into these
sites with specific interest into the quenching effects of Co(II)
on ZnO NCs. The sites identified may be correlated to specific
faces of wurtzite ZnO structure; however, in the nanoregime,
consideration of unconventional surface sites due to curvature
must be considered. Current theories are less than adequate to
explain all of the observed phenomena; however, with the data
from previous oxygen titration experiments, the aformentioned
reactions, and future experimentation, we hope to attain a more
complete understanding of the surface sites and their relation
to catalytic activity and photoluminescence of ZnO NCs.
Acknowledgment. We acknowledge the NSERC and CRC for
generous support of this research in the form of discovery and a
Canadian Research Chair for D.S.B., and to the CIHR for Chemical
Biology Fellowship for C.J.S.
Supporting Information Available: Spectroscopic data in-
cluding absorption, photoluminescent, and transmission electron
microscopy; photocatalytic degradation detail for the decom-
position of nitrophenol and the calculations of the rates of
photocatalysis; analysis of the quenching dynamics and binding
constants for the cationic and anionc sites on the ZnO NC
surface; quantitative binding site calculations; surface area vs
band gap energy relationships. This material is available free
of charge via the Internet at http://pubs.acs.org.
JA808663B
9
4404 J. AM. CHEM. SOC. VOL. 131, NO. 12, 2009