Revealing the Molecular Identity of Defect Sites on PbS Quantum Dot Surfaces with Redox-Active Chemical Probes

Article abstract of DOI:10.1021/acs.chemmater.1c00520

Defects arising on the surfaces of semiconductor quantum dots (QDs) limit the applications of these otherwise promising materials. Efforts to rationally passivate these sites using chemical methods, however, are limited by a lack of molecular-level understanding of surface defects. Herein, we report the application of redox-active chemical probes (E - ′ = -0.48 to -1.9 V vs Fc+/0) coupled with spectroscopic tools (nuclear magnetic resonance (NMR), X-ray photoelectron spectroscopy (XPS), and UV-vis-NIR) to gain insight into the molecular-level nature and reactivity of defects at PbS QD surfaces. First, Pb ion-based traps coordinated by oleate ligands are studied by reaction with outer-sphere reductants, wherein reduction of a subpopulation of Pb2+ ions promotes ligand displacement. We observe a correlation between this reactivity and QD size, wherein minimal ligand displacement occurs in small QDs (2.6 nm) but up to ca. 15% of ligands are displaced with larger QDs (>4 nm). The strength of the reductant also has a significant impact; with QD size held constant, more potent reductants induce a higher extent of ligand displacement than mild reductants. Finally, chalcogenide-based defects (disulfides) are interrogated with selective trialkylphosphine reagents. Comparison of QD reactivity with phosphine probes reveals that large PbS QDs possess a greater proportion of native disulfide defects than small QDs. Collectively, this work yields insight into the identities, likely structural environments and reduction potentials of targeted defect sites, thus providing a detailed picture - and roadmap for passivation - of common QD surface defects.

Full text of DOI:10.1021/acs.chemmater.1c00520

pubs.acs.org/cm
Article
Revealing the Molecular Identity of Defect Sites on PbS Quantum
Dot Surfaces with Redox-Active Chemical Probes
Carolyn L. Hartley and Jillian L. Dempsey*
Cite This: Chem. Mater. 2021, 33, 2655−2665
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Defects arising on the surfaces of semiconductor
quantum dots (QDs) limit the applications of these otherwise promising
materials. Efforts to rationally passivate these sites using chemical
methods, however, are limited by a lack of molecular-level under-
standing of surface defects. Herein, we report the application of redox-
active chemical probes (E◦′ = −0.48 to −1.9 V vs Fc+/0) coupled with
spectroscopic tools (nuclear magnetic resonance (NMR), X-ray
photoelectron spectroscopy (XPS), and UV−vis−NIR) to gain insight
into the molecular-level nature and reactivity of defects at PbS QD
surfaces. First, Pb ion-based traps coordinated by oleate ligands are
studied by reaction with outer-sphere reductants, wherein reduction of a
subpopulation of Pb2+ ions promotes ligand displacement. We observe a
correlation between this reactivity and QD size, wherein minimal ligand
displacement occurs in small QDs (2.6 nm) but up to ca. 15% of ligands are displaced with larger QDs (>4 nm). The strength of the
reductant also has a significant impact; with QD size held constant, more potent reductants induce a higher extent of ligand
displacement than mild reductants. Finally, chalcogenide-based defects (disulfides) are interrogated with selective trialkylphosphine
reagents. Comparison of QD reactivity with phosphine probes reveals that large PbS QDs possess a greater proportion of native
disulfide defects than small QDs. Collectively, this work yields insight into the identities, likely structural environments and
reduction potentials of targeted defect sites, thus providing a detailed pictureand roadmap for passivationof common QD
surface defects.
Indeed, QD surface defects are often ill-defined, making
targeted passivation approaches challenging. One tactic to
elucidate the molecular-level details of QD surfaces, with the
aim of controlling these defects sites, is the use of chemical
probes. Chemical probes are molecular reagents that react
predictably with specific sites at the QD surface, and the
resulting reactivity in turn can reveal details about the native
QD surface structure.20−25 Chemical probes may be classified
as either ligand-based probes or redox-active probes depending
on the type of reactivity they engage in at the QD surface
(Scheme 1).19
Substantial insight into the structural and electronic
properties of QD surfaces has been revealed using both
ligand-based and redox-active probes. Redox-active probes in
particular have yielded a new understanding of both
chalcogenide- and metal ion-derived defects that occur natively
1. INTRODUCTION
Colloidal semiconductor quantum dots (QDs) are a promising
class of materials with potential applications in photo-
voltaics,1,2 commercial displays,3 and photoredox catalysis
systems,4,5 to name a few. However, commercialization of
many QD-based optoelectronic devices has seen only limited
success; this is largely due to the low charge transport
efficiencies observed relative to bulk semiconductor materials.
The cause for these low efficiencies can in part be attributed to
performance-limiting defects at QD surfaces, such as under-
passivated metal or chalcogenide ions, and can result in mid-
gap electronic trap states.6,7 Such states can trap excited
electrons or holes, leading to low photochemical quantum
yields and hindered charge transfer between nanocrystals
within a device. Surface defects are present on as-synthesized
QDs, and their occurrence and routes for passivation have
consequently been a major area of study.6−11 For instance,
techniques such as emission spectroscopy and electrochemistry
have traditionally been used to establish the presence and
approximate energies of mid-gap trap states.12−18 Though
useful, such methods are limited by the insight they may
provide regarding the molecular identity and reactivity of
specific defects at the QD surface.19
Received: February 12, 2021
Revised: March 3, 2021
Published: March 16, 2021
https://dx.doi.org/10.1021/acs.chemmater.1c00520
© 2021 American Chemical Society
Chem. Mater. 2021, 33, 2655−2665
2655

Chemistry of Materials
pubs.acs.org/cm
Article
Scheme 1. Approaches to Studying Surface Defects through Reaction of QDs with (i) Ligand-Based (Shown Here for Metal
Ion Defects) or (ii) Redox-Active Chemical Probes
or that may be induced upon reduction or oxidation
(charging) of the surface. For example, Tsui et al.
demonstrated that oxidized diselenide species are a common
defect motif on CdSe surfaces by investigating surface
reactivity with organometallic redox reagents that are known
to selectively reduce chalcogenide dimers.24 Redox probes
were also used to better understand anionic chalcogenide
defects in PbS QDs. Knowles et al. used a molecular acceptor,
tetracyanoquinodimethane (TCNQ), to probe spontaneous
electron transfer from undercoordinated S2− sites on QD
surfaces using Fourier transform infrared (FTIR) and UV−
vis−NIR absorbance spectroscopies.26
Similarly, redox reactivity has been key to elucidating the
nature of metal-based surface defects. A recent study by our
group revealed that the addition of Na[C10H8] (E°′ = −3.1 V
vs Fc+/0) to CdSe QDs led to the reduction of surface Cd2+
ions with a concomitant displacement of charge-balancing
anionic ligands, yielding putative Cd0 sites on the QD
surfaces.23 This example illustrates the redox reactivity of
surface Cd2+ sites and the displacement of anionic ligands via
an electron-promoted X-type ligand displacement mechanism
in response to surface metal ion charging. The reduction of
Cd2+ surface sites was also recently reported by Pu et al. upon
applying an electrochemical bias to CdSe QD thin films
passivated with cadmium carboxylate ligands.27 Density
functional theory (DFT) studies by Voznyy et al. and du
QD reactivity in multicomponent systems, as well as have
major implications in QD charging or remote chemical doping
processes.
To this end, we herein report systematic investigations of
defect sites on PbS QD surfaces using a series of redox-active
chemical probes. PbS was selected as it is a widely studied and
well-understood QD material, and thereby serves as a relevant
platform for redox-active defect sites. We first establish the
reactivity between a moderately reducing organometallic
reductant, cobaltocene, with metal-based surface sites on 4.1
nm PbS QDs and then study the impact of varying QD size
and potency of the reductant. Finally, we explore other redox-
active defect sites at the surface by employing selective
trialkylphosphine reagents to selectively probe chalcogenide
defects. Our studies reveal the presence of both metal and
chalcogenide defect structures at PbS QD surfaces. By studying
the reactivity of these defect sites with redox-active chemical
probes, we elucidate their molecular-level nature and reduction
potentials. In doing so, we deepen our understanding of QD
surface defects and routes to intentionally passivate or induce
them.
2. EXPERIMENTAL METHODS
General Considerations. All syntheses were performed under a
nitrogen environment using standard Schlenk techniques. All
experiments were performed or set up in a dry nitrogen-filled
glovebox (MBraun) unless otherwise noted. Solvents used in the
glovebox (toluene, tetrahydrofuran (THF), acetonitrile) were purified
over an alumina column and dispensed from a dry argon-atmosphere
solvent system (Pure Process Technologies) and then dried over
activated 3 Å molecular sieves. Nuclear magnetic resonance (NMR)
solvents were similarly stored over activated 3 Å molecular sieves, and
in the case of THF-d8 further dried over an alumina plug.
́
Fosse et al. have explored the impact of charging at the surfaces
of semiconductor QDs and predicted that the dimerization of
surface metal ions induced by charging serves as a source of
mid-gap trap states.28,29 Collectively, these examples highlight
how redox-active chemical probes may be employed to better
understand QD surface structure, as well as inform on
accessible redox-active surface moieties.
Chemicals. Lead(II) oxide (PbO) (99.999% trace metals basis),
oleic acid (OA) (90%), 1-octadecene (ODE) (90%), hexamethyldi-
silathiane ((TMS)2S) (synthesis grade), tri-n-butylphosphine (Bu3P)
(mixture of isomers, 97%) and 1,3,5-trimethoxybenzene (TMB)
(Reagent Plus ≥ 99%) were purchased from Sigma-Aldrich and used
as received. Sodium hydroxide was purchased from Fisher Scientific
and used as received. Cobaltocene (CoCp2) (≥98%) was purchased
from VWR and purified by dissolving in toluene in a glovebox in the
dark, filtering off insoluble particulates, and removing the solvent
under vacuum, followed by sublimation of the solid in the dark prior
to use. Purified cobaltocene was stored in a −30 °C freezer under an
inert atmosphere. Decamethylcobaltocene (CoCp*2) was purchased
from Sigma-Aldrich and recrystallized from cold pentane before use
and then stored in a −30 °C freezer in a glovebox. Toluene-d8 and
THF-d8 were purchased from Cambridge Isotope Laboratories. All
other solvents used were purchased from Fisher Scientific or VWR.
While the combined efforts of the aforementioned studies
have yielded insight into QD surface structure, there remains a
need for a systematic investigation of the molecular nature
(atomic identity, bonding environment, oxidation state, or
reduction potential) and reactivity of these surface defects.
Indeed, many of the examples noted above focused on
identifying the structure or oxidation state of likely defect sites
but stopped short of exploring their redox reactivity, an
important indicator of the specific environment of a given
defect, in tandem. Additionally, QD size is known to impact
which bonding environments, and therefore which possible
defects, may exist at the surface, yet this variable has been
underexplored to date.22,30,31 Importantly, the redox properties
of defect sites are expected to have a significant influence on
2656
https://dx.doi.org/10.1021/acs.chemmater.1c00520
Chem. Mater. 2021, 33, 2655−2665

Chemistry of Materials
pubs.acs.org/cm
Article
PbS Synthesis and Purification. PbS QDs were synthesized
using a previously reported literature procedure.22 QDs ranging in
diameter from 2.6 to 4.9 nm were prepared by varying the molar ratio
of OA to PbO used to prepare the Pb(oleate)2 precursor as well as the
injection temperature of (TMS)2S. In general, higher OA/PbO ratios
and higher injection temperatures resulted in larger nanocrystals
(Table S1). The QDs were purified rigorously by repeated
precipitation and centrifugation cycles to isolate the nanocrystals
from excess ligand, solvent, and reaction byproducts. For QDs used in
direct size comparison studies, purification procedures were the same
across all batches. For these batches (2.6, 3.8, 4.1, 4.9 nm, featured in
Section 3.3), the crude reaction mixture was precipitated with acetone
and divided among centrifuge tubes and centrifuged. The supernatant
was decanted, and the QD solid was dispersed in 1−2 mL of pentane
or toluene and precipitated with 12 mL of acetone, sonicated, and
centrifuged. This step was repeated once. The resulting solid was
dispersed in 1 mL of pentane and precipitated with a 1:1 mixture of
acetone and methanol, sonicated, and centrifuged. This step was
repeated a total of three to four times. The QDs were dispersed in 1
mL of pentane, precipitated with 12 mL of methanol and centrifuged.
The QDs were then dispersed in 1 mL of pentane and precipitated
with 9 mL of a 1:2 mixture of acetone/methanol, sonicated, and
centrifuged. This step was repeated a total of two to three times.
Finally, the QDs were dispersed in 1 mL of pentane, precipitated with
12 mL of acetone, and centrifuged a total of three times. For the
largest batch of QDs (4.9 nm) prepared with a large excess of oleic
acid, we found that filtering the QD solution over a glass fiber filter
paper plug between purification steps further helped to get rid of
excess Pb(oleate)2. The purified solid was then dispersed in pentane,
transferred to a 20 mL vial, and dried via evaporation.
residual carbon species. Solutions of QDs in toluene (typically ∼10
μM) reacted with 0, 50, 100, 500, or 1000 equiv cobaltocene for 21 h
and then the QDs isolated by precipitation with acetonitrile. The
isolated nanocrystals were redispersed in toluene and were drop cast
onto the wafers until a brown film was visible by eye. The XPS
samples were loaded onto a sample holder in a nitrogen-filled
glovebox and transported to the XPS facility in a sealed glass tube,
which was then loaded onto the instrument in an inert-environment
glovebag. X-ray photoelectron spectroscopy (XPS) was performed
using a Kratos Axis Ultra DLD X-ray photoelectron spectrometer with
a monochromatic Al Kα X-ray source. Survey scans and high-
resolution scans were obtained with pass energies of 80 and 20 eV,
respectively. Spectra were corrected to the C 1s peak at 284.6 eV.
Transmission Electron Microscopy (TEM) Sample Prepara-
tion. Samples for transmission electron microscopy (TEM) imaging
were prepared by drop-casting the dilute solutions of QDs in pentane
that were filtered through 2 μm poly(tetrafluoroethylene) (PTFE)
syringe filters onto TEM grids (Ultrathin carbon film on lacey carbon
support film, 400 mesh, copper) under ambient conditions. The grids
dried in air and were then conditioned overnight under a vacuum to
remove any trace volatiles. Images were collected on a Thermo
Scientific Talos F200X S/TEM at an accelerating voltage of 200 kV
and with a 70 μm objective aperture.
3. RESULTS AND DISCUSSION
3.1. System for Studying Redox-Active Defect Sites
on PbS Surfaces. Oleate-capped PbS QDs ranging in size
from 2.6 to 5.1 nm in diameter provide a versatile platform for
interrogating redox-active defects on QD surfaces. The size
range explored here provides a means to access QDs with a
range of band edge potentials as well as different morphologies,
surface facets, and stoichiometries, which are known to vary as
a function of QD size for PbS.22,30,31,33 Varying QD size in our
studies thereby provides additional insight into the molecular
nature of reactive surface species.
For samples used in other studies, purification proceeded with
minor variations to the procedure described above. The QDs were
stored as a solid in an inert-atmosphere glovebox. QD size and
extinction coefficients at 400 nm were determined using the sizing
curve reported by Moreels et al.32
UV−vis−NIR Absorbance Studies. UV−vis−NIR absorbance
data was collected on an Agilent Cary 60 or a Cary 5000
spectrophotometer. Samples of QDs were prepared in toluene (1−5
μM) and ∼3 mL of this solution were added to a custom-made quartz
cuvette with an adapted 14/20 ground-glass joint. The cuvette was
equipped with a micro stir bar and then capped with a rubber septum
that was secured with electrical tape and copper wire. A solution of
cobaltocene was prepared in toluene (15 mM) and then loaded into a
gas-tight locking Hamilton syringe. The charged needle was locked
and the end stuck into a separate rubber septum to avoid exposure to
air. The cuvette and syringe were then brought out of the glovebox
and the cobaltocene was added incrementally by syringe, stirring for
approximately 30 s after each addition to ensure thorough mixing
before collecting a spectrum. Samples for long-timescale studies were
prepared in cuvettes sealed with Kontes valves and shielded from
light.
Redox-active chemical probes were selected to span potency
from strong (E°′ = −1.9 V vs Fc+/0) to weak (E°′ = −0.48 V vs
Fc+/0) (Table 1). As other works have investigated the
Table 1. Chemical Redox Probes Used in This Work34,49
reduction potential (E°′, V
vs Fc+/0
energy vs
d
chemical probe
)
vacuum (eV)
a
decamethylcobaltocene
−1.91
−3.22
(CoCp*2)
b
b
a
cobaltocene (CoCp2)
−1.33
−0.93
−0.48
−3.80
−4.20
−4.65
c
Co(Cp)(dppe)
decamethylferrocene
(FeCp*2)
a
b
1
Reported conditions in CH3CN with [Bu4N][PF6]. Reported
conditions in CH2Cl2 with [Bu4N][PF6]. dppe = 1,2-bis-
(diphenylphosphino)ethane. Conversion to energy from reduction
NMR Studies. H NMR spectra were collected on a 600 MHz
c
Bruker NMR spectrometer with a cryoprobe. Unless noted otherwise,
titration studies were prepared by adding 600 μL of a 30 μM QD
stock solution in toluene-d8 to J-Young NMR tubes. An internal
standard stock solution was then prepared by dissolving ∼15 mg of
1,3,5-trimethoxybenzene in 1.5 mL of toluene-d8, and then 50 μL was
added to each NMR tube. A 50 mM solution of redox reagent was
prepared in toluene-d8 and added in increments of 0, 50, 100, 500,
and 1000 equiv per QD to the NMR tubes at staggered times to
ensure that all samples mixed for the same amount of time (3 h)
before collecting NMR spectra. Spectra were collected with 12 scans
and a d1 delay time of 30 s. The absolute number of bound and free
ligands per QD were determined by spectral fitting with MestReNova
software (Figure S1). 31P{1H} NMR spectra were collected on a 500
or 600 MHz Bruker NMR spectrometer.
d
potential shown in the Supporting Information (SI).
oxidation of PbS QD surfaces,26 we focus here on the impact
of reductive chemical probes. With this wide array of
reductants, we demonstrate that it is possible to vary the
driving force of surface reduction and, in effect, target specific
sites to rationally passivate undesired surface defects.
3.2. PbS QDs Display Surface Reactivity with CoCp2.
The reactivity between a mid-size batch of PbS QDs (4.1 nm
diameter) was first established with the moderate reductant
cobaltocene (CoCp2, E°′ = −1.3 V vs Fc+/0). CoCp2 has been
previously shown to undergo ground-state charge transfer with
PbS QDs and does not display any deleterious side
chemistry.34,35
X-ray Photoelectron Spectroscopy (XPS) Sample Prepara-
tion. Samples for X-ray photoelectron spectroscopy (XPS) analysis
were measured on gold-coated silicon wafers that had been sonicated
for 2 min in 190 proof ethanol and dried in air prior to use to remove
2657
https://dx.doi.org/10.1021/acs.chemmater.1c00520
Chem. Mater. 2021, 33, 2655−2665

Chemistry of Materials
pubs.acs.org/cm
Article
Upon reduction of a colloidal QD by CoCp2, the donated
electron can either occupy a delocalized conduction band
(CB) electronic state or be confined to localized electronic
states at the surface (Figure 1a). To assess whether CoCp2 is
able to reduce the 4.1 nm PbS QD conduction band (CB)
edge state, the electron transfer reaction was monitored using
UV−vis−NIR absorbance spectroscopy. Absorbance spectros-
copy is commonly employed to gauge band edge charging with
excess electrons or holes; upon reduction of the CB or
oxidation of the valence band (VB), a bleach of the excitonic
absorbance feature will be observed due to the lowered
probability of exciting a carrier into an already-filled state.35,36
The absorption spectra of 4.1 nm QDs in toluene before and
after addition of 500 equiv CoCp2 (per mol QD) are shown in
Figure 1b. No excitonic bleach (λmax = 1186 nm) is observed,
indicating that band edge reduction by CoCp2 is energetically
unfavorable even with a large excess of reductant. While
previous studies observed interfacial electron transfer from
CoCp2 into CB states of similarly sized oleate-capped PbS
QDs (≥∼4.2 nm in diameter),35 our observations likely differ
from this prior report as this previous work was performed on
QD thin films rather than in solution.
Although no band edge reduction was observed by UV−
vis−NIR absorbance spectroscopy, we anticipated that CoCp2
was a sufficiently strong reductant to reduce localized surface
states, as has been postulated previously.35 To interrogate the
reduction of the surface directly, titration studies were
1
performed using H NMR spectroscopy. NMR spectroscopy
is a convenient method to probe QD surface chemistry as it
allows for quantification of the capping oleate ligands bound to
surface Pb ions. The alkene proton resonance of the oleate
ligands may be used to differentiate between bound vs. free
ligands in solution. The bound oleate ligand resonance is broad
and appears at 5.66 ppm, whereas the free oleate resonance is
sharper and shifted up-field (5.48 ppm).22,37−39 Character-
ization of as-synthesized 4.1 nm QDs reveals an average bound
ligand coverage of 173 5 oleates per QD.
Addition of excess CoCp2 (0−1000 equiv) to the PbS QDs
in toluene-d8 results in dissociation of a portion of bound
oleate ligands that increases with equivalents of reductant
added (Figure 2). The 4.1 nm QDs lose an average of 14.7
3.2% of initially bound oleate ligands with the addition of up to
1000 equiv CoCp2; this corresponds to approximately 25
oleate ligands displaced per QD. Importantly, the total number
of ligands (bound + free) remains constant throughout the
course of the titration, indicating that the QDs remain stable in
Figure 1. (a) Diagram depicting the optical response observed by
absorbance spectroscopy upon reduction of either the band edge state
or a localized surface state. (b) UV−vis−NIR absorption spectrum of
4.1 nm QDs (2.5 μM) in toluene before (red) and after (blue) the
addition of 500 equiv CoCp2.
1
Figure 2. H NMR titration of 4.1 nm QDs (27.7 μM) in toluene-d8 upon addition of 0 (red), 50 (yellow), 100 (green), 500 (blue), and 1000
equiv CoCp2 (violet). Internal standard of 1,3,5-trimethoxybenzene is indicated by (*) and solvent residual by (+).
2658
https://dx.doi.org/10.1021/acs.chemmater.1c00520
Chem. Mater. 2021, 33, 2655−2665

Chemistry of Materials
pubs.acs.org/cm
Article
solution even in the presence of large excesses of reductant
(Figures S1 and S2). X-ray photoelectron spectroscopy (XPS)
studies confirm that isolated PbS QDs reacted with an excess
of CoCp2 do not have any significant changes in their Pb/S
ratio (Table S2). The observed ligand loss without a
corresponding loss of surface Pb atoms therefore suggests
that the detected free ligand is due to liberation of X-type
anionic oleates, and not Z-type Pb(oleate)2 fragments, upon
reduction. This observation is consistent with the electron-
promoted X-type ligand displacement phenomenon previously
reported for CdSe QDs,23 and these data thereby provide
direct evidence that surface Pb2+ ions are reduced by CoCp2.
Within our system, we propose that oleate-bound Pb2+
surface ions are reduced to Pb0 concurrent with the ejection
of bound anionic oleate ligands; reduction to form surface Pb0
sites is expected to be stable and would be consistent with the
reports of Cd0 formation upon reduction of Cd2+ ions at the
surfaces of CdSe nanocrystals.23,40−43 Though formally
plausible, Pb1+ ions are anticipated to be unstable and, if
formed, would likely dimerize with neighboring Pb ions as has
been postulated in recent computational studies.28,29 We were
unable to detect direct evidence for Pb−Pb dimers or
formation of Pb0 atoms using XPS spectroscopy (Figure S3);
however, this could be due to the limited sensitivity of XPS to
minor surface subpopulations on QDs. Therefore, the
formation of these surface structures upon reduction of Pb2+
ions and displacement of oleate ligands cannot be ruled out.
In addition to monitoring resonances corresponding to the
sites by XPS (Figure S12) and FTIR (Figure S13) spectros-
copies were inconclusive; however, we cannot eliminate the
possibility of small subpopulations of these and similar metal-
based sites on the surface reacting with an added excess
reductant.
3.3. QD Surface Reactivity Has a Marked Size
Dependence. Having established measurable reactivity (i.e.,
surface Pb2+ ion reduction and oleate displacement) at the
surfaces 4.1 nm PbS QDs upon addition of excess CoCp2, we
next sought to understand the dependence of surface redox
reactivity on QD size. For PbS, QD diameter is known to
influence the band edge potentials as well as the morphology
and surface facets exposed.22,30,31 To investigate the influence
of these parameters, we prepared several batches of QDs that
varied in diameter. Each batch was handled and purified
similarly for relevant comparison. To that end, we found that
samples prepared by another method46 did not precisely follow
the trends observed below, and we thus restrict our discussion
to these samples prepared by the method reported by Hines
and Scholes.47 Three additional sizes of QDs2.6, 3.8, and 4.9
nmwere prepared and characterized by UV−vis−NIR
1
absorbance spectroscopy, H NMR spectroscopy, TEM, and
XPS methods (Table S3 and Figures S14−S17). QDs larger
than 5 nm are not compared herein due to problems with
solubility and stability of those sizes prepared by the method of
Hines and Scholes.47
To first probe whether CoCp2 is able to directly reduce the
CB states of any of these QDs, each sample was titrated with
aliquots of CoCp2 and the absorbance was monitored via UV−
vis−NIR absorbance spectroscopy (Figures S18−S22). As 4.1
nm QDs did not show any evidence of CB charging, it was not
expected that the smaller QDs (2.6 and 3.8 nm) would show
an excitonic bleach, as their conduction band edge states are
higher in energy. Notably, no excitonic bleach was observed for
any of these four samples, indicating that even for the larger 4.9
nm QDs, CoCp2 is too weak of a reductant to engage in
ground-state electron transfer to the CB edge states of PbS.
Many of the PbS QD samples did display a minor red shift (1−
4 nm) of the excitonic absorbance feature. This red shift is an
indicator of surface charging resulting from Coulombic
repulsion with the delocalized exciton or a Stark effect.48
Titrations of these three additional QD samples with CoCp2
1
surface capping ligands in the H NMR spectra, the reductant
CoCp2 and its oxidized product [CoCp2]+ are also readily
1
detected in the H NMR spectra. At all equivalents added
(50−1000 equiv), the NMR resonance for paramagnetic
CoCp2 is observed at −52 ppm (Figure S4). The presence
of CoCp2 in the QD samples indicates that while a portion of
the reducing agent reacts with the QD (as evidenced by
liberation of X-type oleate ligands), not all of the CoCp2 added
reacts on the hours-long timescale. Furthermore, a new broad
feature is observed at 6.2 ppm after CoCp2 is added (Figure
S1). We tentatively assign this to the NMR resonance of the
cyclopentadienyl protons of the oxidized cobaltocenium ion
([CoCp2]+). In support of this assignment, we isolated the
putative [CoCp2][oleate] from a mixture of PbS QDs with
excess CoCp2 (see the SI). [CoCp2]+ is expected to act as a
counterion to both reduced surface sites as well as to liberated
oleates as a cobaltocenium oleate salt, [CoCp2][oleate]
(Figures S5−S10).44
1
were then monitored with H NMR spectroscopy to compare
the extent of reactivity with an added charge. The addition of
an excess of CoCp2 (0−1000 equiv) resulted in varying
degrees of oleate displacement for different sizes (Figures
S23−S26). The average percentages of bound oleate ligands
(normalized for any minor amounts of a free ligand present in
the as-synthesized QD sample) are presented in Figure 3. The
data show a clear trend of greater oleate loss with increasing
QD size: after mixing for 3 h with 1000 equiv CoCp2, 1.2
1.1% of the bound oleates are displaced from 2.6 nm QDs, 3.5
1.4% from 3.8 nm QDs, 14.7 3.2% from 4.1 nm QDs, and
11.1 2.1% from 4.9 nm QDs. The broad feature at 6.2 ppm
attributed to [CoCp2]+ is observed in all QD samples with
CoCp2 added, as is the presence of paramagnetic CoCp2 at
−52 ppm.
When samples are allowed to react over the course of days,
the relative integration of the paramagnetic CoCp2 resonance
decreases (Figure S11). Concurrently, oleate ligands continue
to dissociate from the QD surface (Figure S11). The slow and
continued reactivity over hours and days point toward a
gradual equilibration between CoCp2 and surface Pb2+ ions.
Therefore, all NMR titrations were carefully timed to ensure
that all samples mixed for exactly the same period of time (3 h)
prior to collecting spectra to enable precise comparisons in the
extent of surface reduction and oleate loss.
1
Additionally, it is important to note that H NMR silent
redox processes are expected to occur in tandem with oleate
displacement. For example, underpassivated Pb2+ sites or
hydroxide-ligated Pb2+ ions have been postulated as PbS
surface defects in the literature, though these would not be
observable by our NMR experiments.28,29,45 Attempts to detect
the presence and reduction of these postulated Pb2+ surface
We interpret the trend of greater amounts of displaced
oleates observed for larger QDs by considering changes in the
PbS QD surface environments with size. Several recent
compelling works have demonstrated that PbS QDs transition
from an octahedral morphology to a cuboctahedral shape with
increasing diameter.22,30,31 This in turn impacts the types of
2659
https://dx.doi.org/10.1021/acs.chemmater.1c00520
Chem. Mater. 2021, 33, 2655−2665

Chemistry of Materials
pubs.acs.org/cm
Article
Figure 3. Comparison of different sizes of QDs with added CoCp2,
1
normalized for minor amounts of free ligand in QD only H NMR
spectra. Data points are the average of three repeated trials with
standard deviation included as error bars. The lines between data
points are a guide to the eye. Inset: Data plotted with smaller y-axis
range for a clearer depiction of trends.
facets and bonding environments at the surface that are
exposed. For example, small octahedral PbS QDs (<3 nm) are
known to be lead-rich with polar (111) facets exposed.
Conversely, larger QDs tend to be more stoichiometric in Pb
and S ions, and it has been proposed that additional neutral
(100) facets begin to appear on QDs larger than ∼3 nm
diameter.22,30,31 Additionally, the larger QD sizes in this study
possess lower ligand packing densities than the smaller QDs
(Table S3), which suggests that the surfaces of larger QDs may
be more accessible to reducing agents.
Figure 4. PbS QDs have been shown to exhibit a size-dependent
morphology where (a) small QDs possess an octahedral shape and
(b) larger QDs a cuboctahedral shape. (100) facets are present on the
surfaces of large PbS QDs, and new Pb environments emerge at these
facets as well as at (111)/(100) facet edge sites.
reaction of 4.3 nm QDs with ca. 500 equiv of each redox agent,
less ligand loss was observed compared with CoCp2 as shown
The size dependence of exposed Pb coordination environ-
ments at the surface likely impacts the extent of ligand
displacement observed upon reduction by CoCp2. Importantly,
surface Pb2+ ions that are reduced by CoCp2 to liberate oleate
ligands are anticipated to be at approximately the same free
energy (as quantified by their reduction potential) regardless of
QD size. Therefore, the greater proportion of oleates displaced
in larger QD sizes suggests that there is a larger relative
number of those specific types of Pb2+ sites in larger QDs.
While the NMR studies themselves cannot indicate distinct
ligand bonding environments, the size dependence observed
here suggest that the Pb2+ sites most likely to be reduced by
CoCp2 may be those at the edges between (111) and (100)
facets as those appear on larger size QDs (Figure 4). Reduction
of Pb2+ ions within the (100) facets present on the larger QDs
is less likely due to the overall charge neutrality of those facets
as well as being less sterically accessible than edge sites.22,45
3.4. Extent of QD Surface Charging Tracks with
Reductant Strength. To gain a more holistic understanding
of the redox reactivity of the surfaces of PbS QDs, we next
probed the dependence of surface reactivity on the potency of
the reductant employed. To do this, both milder and stronger
reductants than CoCp2 (Table 1) were reacted with PbS QDs
1
with H NMR spectroscopy (Figure S27). Specifically, <1%
oleate loss was observed upon addition of excess FeCp*2 and
∼2.5−4% with added Co(Cp)dppe, compared with ∼6%
ligand displaced with added CoCp2. This diminished reactivity
with milder reducing agents illustrates the influence that the
reductant strength has on the extent of surface charging.
To further establish the relationship between surface
charging and reductant strength, we next moved to a reductant
stronger than CoCp2, decamethylcobaltocene (CoCp*2). Due
to the strongly reducing nature of CoCp*2 and estimated
literature values for PbS CB edge potentials, ranging from −3
to ca. −4.2 eV (with respect to vacuum) across sizes from 2 to
8 nm, respectively, we anticipated that CoCp*2 could directly
inject electrons into QD CB edge states.33,35 Addition of 50
equiv CoCp*2 to 4.9 nm PbS QDs in 1:1 THF/toluene
resulted in a significant bleach of the excitonic absorption
feature, as detected by UV−vis−NIR absorbance spectroscopy,
consistent with electron injection into the CB states (Figure
5a). Notably, the excitonic bleach recovers within minutes
(Figure S28). This observation is consistent with a previous
report by Koh et al. that suggested that delocalized CB
electrons in PbS QDs may localize at surface sites over time.35
1H NMR titrations were next performed to assess whether
the more negative reduction potential of CoCp*2 has an
impact on the amount of oleate displacement observed.
Incremental addition of CoCp*2 (5−100 equiv) to 4.9 nm
QDs leads to a dramatic decrease in oleate coverage, with up to
60% of bound oleates lost (Figure 5b). Interestingly, the
quantification of displaced oleate ligands corresponds closely
with a 1:1 loss for each equivalent of CoCp*2 added up to 10
equiv, and then slightly a greater ratio of approximately 1.2−
1.4 oleates lost per equiv CoCp*2 added with 25−100 equiv
1
and the extent of reactivity (i.e., oleate displacement in H
NMR studies) compared. For these redox-reagent dependence
experiments, additional batches of PbS QDs ranging in size
from ca. 3 to 5 nm were employed in addition to those studied
in the size-dependence experiments above. These studies
provide insight into the effect of driving force on surface
reduction.
We first employed two reductants milder than CoCp2,
decamethylferrocene (FeCp*2) and Co(Cp)(dppe), and
monitored their reactivity with 4.3 nm PbS QDs. Upon
2660
https://dx.doi.org/10.1021/acs.chemmater.1c00520
Chem. Mater. 2021, 33, 2655−2665

Chemistry of Materials
pubs.acs.org/cm
Article
reduction pathways may occur simultaneously and are not
necessarily mutually exclusive.
Overall, these studies reveal the marked impact of reductant
strength on the extent of surface charging as indicated by the
extent of oleate loss. The lesser degree of reactivity of the QD
surface sites with reductants milder than CoCp2, along with the
more dramatic reactivity upon reaction with stronger
reductants such as CoCp*2, suggests that the slow equilibra-
tion between CoCp2 and PbS QD surface states observed
arises because the reduction potential of CoCp2 is approx-
imately isoenergetic with the reduction potential of a portion
of the surface Pb2+ sites (Figure 6). This equilibration leads to
Figure 6. Depiction of the relative redox potentials of the redox-active
probes used in this study versus band edge or surface-based electronic
states in PbS QDs ranging in size from 2.6 to 5.1 nm.
Figure 5. (a) UV−vis−NIR absorption spectra of 4.9 nm QDs (2.5
μM) in 1:1 THF/toluene before (red) and after (blue) the addition of
ca. 50 equiv CoCp*2. (b) Comparison of the percentage of bound
oleate per QD in 4.9 nm PbS QDs (27.7 μM) with added CoCp2
(red) and CoCp*2 (blue). NMR spectra of CoCp*2 titration are given
in Figure S29.
gradual oleate ligand displacement over time. These Pb2+
surface sites are not energetically accessible by the weaker
reductants employed. By comparison, more potent reductants
increase the driving force for surface Pb2+ ion reduction leading
to more rapid and complete surface charging, as well as
conduction band population.
While the driving force for electron transfer from a reductant
may in theory be predicted by comparing the reduction
potential of the reagent with the reduction potentials of the
localized surface and CB edge states of QDs, precise
measurement of the latter values is quite challenging in
practice. As such, our conclusions are not trivial. Though
reduction potential values for band edge states have been
reported for the thin films of QDs using techniques such as
photoelectron emission spectroscopy and electrochemistry,
these values inherently differ from those describing colloidal
QDs due to differences in the nanocrystal environment.17,33,50
Additionally, the accuracy of reported values across all systems
is questionable: variations in QD synthesis and purification
methods, solvents, and the influence of capping ligands via
surface dipoles directly impact the absolute band edge
positions.51,52 As the reduction potentials of QD band edge
and surface states are difficult to measure directly, the use of
redox-active chemical probes spanning a wide range of
reduction potentials therefore provides an indirect method to
gauge the electronic state potentials based on the extent of
reactivity observed by techniques such as UV−vis−NIR
absorbance or NMR spectroscopy.
(Table S4). The excess oleate ligands displaced per equivalent
of CoCp*2 added suggests that there may be a more complex
ligand displacement mechanism at play in the higher
equivalent regime. For example, CoCp*2 may be capable of
reducing spectroscopically silent surface states, the reduction of
which could promote surface rearrangement leading to some
Z-type ligand dissociation.
Furthermore, unlike for milder reductants and CoCp2, we
observe complete consumption of CoCp*2 as evidenced by the
lack of a paramagnetic resonance for CoCp*2 expected at ca.
45 ppm. Addition of CoCp*2 in greater excess (>100 equiv/
QD) results in rapid QD precipitation, indicating that loss of a
large percentage of passivating surface ligands leads to QD
instability. The dramatic increase in oleate displacement
observed with CoCp*2 addition is likely a direct result of the
greater driving force for the reduction of the same Pb surface
sites that react with CoCp2, as well as reduction of higher
energy populations of Pb ions that are perhaps inaccessible to
milder reductants even at high concentrations. Additionally,
another pathway for surface reduction by CoCp*2 is also
possible wherein electrons initially injected into the CB edge
are subsequently trapped at lower energy surface sites as
evidenced by the recovery of the excitonic bleach on a
minutes-long timescale,35 leading to further oleate ligand
displacement upon localization at the surface. These surface
3.5. Hypothesized Chalcogenide Defects May Be
Studied by Selective Probes. Following the above studies
monitoring surface Pb2+ ion reduction by oleate ligand
displacement, we sought to gain a more complete picture of
2661
https://dx.doi.org/10.1021/acs.chemmater.1c00520
Chem. Mater. 2021, 33, 2655−2665

Chemistry of Materials
pubs.acs.org/cm
Article
other redox-active sites at PbS QD surfaces. The use of redox-
active chemical probes can provide insight into sites that have
been proposed experimentally and computationally.
Chalcogenide ions have been proposed as redox-active sites
on the QD surface, though specific defect subpopulations can
be quite challenging to probe experimentally.7,24,29,53 In
particular, oxidized S1− species are expected to exist as stable
disulfide dimers on as-synthesized QD surfaces.24 To probe the
presence of surface disulfides, studies were performed using a
selective redox-active chemical probe. Tri-n-butylphosphine
(Bu3P) has been established to selectively reduce organic
diselenides and disulfides.24,54 Experiments in our lab confirm
this reactivity for diphenyl disulfide, which is readily reduced to
sulfates upon addition of Bu3P in the presence of NaOH,
yielding tri-n-butylphosphine oxide as a byproduct that may be
monitored by 31P{1H} NMR spectroscopy (Figure S30).
While reactions of Bu3P with organic disulfides support the
theory that disulfide dimers on QD surfaces may be reducible,
to the best of our knowledge the presence of these
chalcogenide dimers on individual QD surfaces has not yet
been directly experimentally confirmed. To study this directly,
PbS QDs were treated with Bu3P in the presence of NaOH and
monitored by 31P{1H} NMR spectroscopy for evidence of
Bu3PO formation as a handle for disulfide reduction (Scheme
Figure 7. 31P{1H} NMR spectrum of (bottom) Bu3P (28 mM) in
THF with NaOH in degassed water. Also, (top) 5.1 nm PbS QDs (29
μM) in THF with ∼1000 equiv Bu3P and 50 equiv NaOH in degassed
water.
minimal change in bound versus free ligand populations
(Figures S34 and S35) of both 3.4 and 4.7 nm QDs with 1000
equiv Bu3P added, suggesting that this is not a dominant
mechanism for the observed reactivity. Overall, the studies
with a selective redox-active probe, Bu3P, provide compelling
evidence for the existence and reactivity of surface disulfides on
as-synthesized PbS QD surfaces. This work therefore serves as
an example of using targeted experimental approaches to
confirm or rule out the existence of hypothesized defect
structures at QD surfaces.
Scheme 2. Reduction of Disulfide Moieties Using Tri-n-
butylphosphine
4. CONCLUSIONS
In conclusion, through the use of a range of outer-sphere
reductants and selective redox reagents, we have gained an
understanding of the native defects that exist on PbS QDs as
well as those that are induced upon surface charging.
Systematic investigations of the reactivity of the QD surfaces
by NMR spectroscopy reveal oleate ligand displacement in
response to the reduction of surface Pb2+ ions. Comparison of
the extent of ligand displacement across QD batches reveals a
distinct size dependence, with larger QDs exhibiting a higher
degree of ligand loss. This size-dependent surface reactivity
correlates with known changes in PbS QD morphology with
size.22,30,31 Specifically, the emergence of new Pb2+ ion
environments at (100) facets, and the edge sites between
(100) and (111) facets, are proposed to be the primary Pb
species reduced leading to oleate displacement. Furthermore,
we demonstrate that the strength of the reductant employed
has a significant impact on the extent of surface charging and
reactivity; stronger reducing agents result in a greater amount
of overall ligand displacement, whereas weak reducing agents
show minimal reactivity with the QDs. Finally, the reactivity of
chalcogenide-based defects (disulfides) was explored using a
selective trialkylphosphine reagent. Similar to the reduction of
Pb ions, disulfide reactivity displays a distinct size dependence
with more reactivity observed for disulfide reduction in larger
QDs.
2). Notably, as demonstrated by a lack of broad features in the
31P{1H} NMR spectra, Bu3P does not appear to bind to the
PbS QD surface (Figure S31). Addition of approximately 1000
equiv Bu3P to solutions of PbS QDs in THF in the presence of
50 equiv NaOH led to differing results for small (3.4−3.7 nm)
versus large (4.7−5.1 nm) QD samples. Smaller QDs show
very little reactivity; little to no Bu3PO formation was observed
even over several days (Figure S32). By comparison, in larger
QDs, approximately 1% of the added Bu3P (ca. 10 equiv/QD)
was converted to Bu3PO, indicating reactivity of the phosphine
with surface disulfides (Figure 7). This size dependence may
arise from the influence of morphology and surface
stoichiometry on the relative population of these disulfide
traps, where larger QDs may reveal disulfides on exposed
(100) facets that are not present on smaller, octahedral
nanocrystals. These results suggest that the (100) facets and
(100)/(111) facet edges may be the most likely sites for
disulfide defects to arise. Importantly, under the conditions
employed we did not detect any reaction between Bu3P and
Pb(oleate)2, confirming that the observed reactivity was due to
surface sulfide species (Figure S33).
Lewis bases such as trialkylphosphines are known to displace
Z-type ligands from QD surfaces at high concentrations; as
such, it was important to rule out Pb(oleate)2 displacement as
a primary cause for the disulfide reactivity detected.21,22,30
Specifically, we sought to determine whether the reactivity
observed was caused by Bu3P first displacing Pb(oleate)2 to
Taken together, these studies suggest that there is a larger
relative proportion of redox-active defect sites (both metal-
and chalcogenide-based) present on larger PbS QDs. We argue
that this size dependence is not simply because of increased
surface area and the total number of atoms in larger QDs;
1
reveal underlying disulfide structures. H NMR spectra show
2662
https://dx.doi.org/10.1021/acs.chemmater.1c00520
Chem. Mater. 2021, 33, 2655−2665

Chemistry of Materials
pubs.acs.org/cm
Article
rather, changes in morphology correlated with increasing QD
diameter lead to the formation of the redox-active defect sites
probed herein. Overall, we present an example of using a
powerful tool, chemical probes, in tandem with NMR
spectroscopy to study the identity and reactivity of
subpopulations of defect sites on QD surfaces.
under Award Number DE-SC0021173. This work made use of
X-ray photoelectron spectroscopy instrumentation at the
Chapel Hill Analytical and Nanofabrication Laboratory
(CHANL), a member of the North Carolina Research Triangle
Nanotechnology Network (RTNN), which is supported by the
National Science Foundation (Grant ECCS-1542015) as part
of the National Nanotechnology Coordinated Infrastructure
(NNCI). A portion of this work was performed at the
Analytical Instrumentation Facility (AIF) at North Carolina
State University, which is supported by the State of North
Carolina. This work made use of instrumentation at AIF
acquired with support from the National Science Foundation
(DMR-1726294). We thank the University of North Carolina’
s Department of Chemistry NMR Core Laboratory for the use
of their NMR spectrometers with support from the National
Science Foundation under Grant Nos. CHE-0922858 and
CHE-1828183. The authors thank Christian Y. Dones Lassalle
for contributing a PbS QD sample to this project and Michael
Mortelliti for assistance in TEM imaging. The authors thank
Dr. Carrie Donley and Dr. Catherine McKenas for assistance
in XPS measurements. C.L.H. thanks Melody Kessler, Michael
Mortelliti, and Dr. Kedy Edme for helpful discussions.
Importantly, this work is a stepping stone toward establish-
ing a detailed, molecular-level picture of QD surfaces by
employing multiple complementary techniques and targeted
reactions.19 We hope that future works will build upon these
findings to further understanding of the structural nature of
defect sites and their reactivity with an added charge. The QD
field is now, more than ever, poised to do so through the
recent emergence of new tools to study QD surface structure,
including advanced multinuclear NMR spectroscopy methods
(e.g., probing metal or chalcogen nuclei directly, or using
dynamic nuclear polarization or cross-polarization magic-angle
spinning enhancement methods, etc.)55−61 and innovative
computational tools to explore surface structure motifs.6,7,28,62
Such an approach is critical to gaining a molecular-level
understanding of QD surfaces and reaching the full potential of
QD-based systems.
REFERENCES
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.chemmater.1c00520.
■
■
sı
(1) Kovalenko, M. V.; Manna, L.; Cabot, A.; Hens, Z.; Talapin, D.
V.; Kagan, C. R.; Klimov, V. I.; Rogach, A. L.; Reiss, P.; Milliron, D. J.;
et al. Prospects of Nanoscience with Nanocrystals. ACS Nano 2015, 9,
1012−1057.
(2) Carey, G. H.; Abdelhady, A. L.; Ning, Z.; Thon, S. M.; Bakr, O.
M.; Sargent, E. H. Colloidal Quantum Dot Solar Cells. Chem. Rev.
2015, 115, 12732−12763.
(3) Kershaw, S. V.; Jing, L.; Huang, X.; Gao, M.; Rogach, A. L.
Materials aspects of semiconductor nanocrystals for pptoelectronic
applications. Mater. Horiz. 2017, 4, 155−205.
(4) Kodaimati, M. S.; McClelland, K. P.; He, C.; Lian, S.; Jiang, Y.;
Zhang, Z.; Weiss, E. A. Viewpoint: Challenges in Colloidal
Photocatalysis and Some Strategies for Addressing Them. Inorg.
Chem. 2018, 57, 3659−3670.
(5) McClelland, K. P.; Weiss, E. A. Selective Photocatalytic
Oxidation of Benzyl Alcohol to Benzaldehyde or C−C Coupled
Products by Visible-Light-Absorbing Quantum Dots. ACS Appl.
Energy Mater. 2019, 2, 92−96.
Additional synthesis and characterization details; addi-
tional experimental data including TEM, NMR, XPS,
and UV−vis−NIR absorbance spectra (PDF)
AUTHOR INFORMATION
Corresponding Author
Jillian L. Dempsey − Department of Chemistry, University of
North Carolina, Chapel Hill, North Carolina 27599-3290,
United States; orcid.org/0000-0002-9459-4166;
Email: dempseyj@email.unc.edu
■
Author
Carolyn L. Hartley − Department of Chemistry, University of
North Carolina, Chapel Hill, North Carolina 27599-3290,
United States
(6) Giansante, C.; Infante, I. Surface Traps in Colloidal Quantum
Dots: A Combined Experimental and Theoretical Perspective. J. Phys.
Chem. Lett. 2017, 8, 5209−5215.
(7) Houtepen, A. J.; Hens, Z.; Owen, J. S.; Infante, I. On the Origin
of Surface Traps in Colloidal II−VI Semiconductor Nanocrystals.
Chem. Mater. 2017, 29, 752−761.
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.chemmater.1c00520
(8) Thomas, A.; Sandeep, K.; Somasundaran, S. M.; Thomas, K. G.
How Trap States Affect Charge Carrier Dynamics of CdSe and InP
Quantum Dots: Visualization through Complexation with Viologen.
ACS Energy Lett. 2018, 3, 2368−2375.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
(9) Brawand, N. P.; Goldey, M. B.; Vörös, M.; Galli, G. Defect States
and Charge Transport in Quantum Dot Solids. Chem. Mater. 2017,
29, 1255−1262.
This material is based upon work supported by the Air Force
Office of Scientific Research under AFOSR Award No.
FA9550-16-1-0206. C.L.H. acknowledges support from the
National Science Foundation Graduate Research Fellowship
Program (DGE-1650116) and a UNC Chemistry Dobbins
Award. J.L.D. acknowledges support from a Packard Fellow-
ship in Science and Engineering. A portion of this work was
performed using the UV−vis−NIR absorption spectroscopy,
Raman, and FTIR spectrometers in the CHASE Hub
Instrumentation Facility established by the Center for Hybrid
Approaches in Solar Energy to Liquid Fuels (CHASE), an
Energy Innovation Hub funded by the U.S. Department of
Energy, Office of Science, Office of Basic Energy Sciences
(10) Zou, H.; Dong, C.; Li, S.; Im, C.; Jin, M.; Yao, S.; Cui, T.; Tian,
W.; Liu, Y.; Zhang, H. Effect of Surface Trap States on Photocatalytic
Activity of Semiconductor Quantum Dots. J. Phys. Chem. C 2018,
122, 9312−9319.
(11) Kirkwood, N.; Monchen, J. O. V.; Crisp, R. W.; Grimaldi, G.;
Bergstein, H. A. C.; du Fossé, I.; van der Stam, W.; Infante, I.;
Houtepen, A. J. Finding and Fixing Traps in II−VI and III−V
Colloidal Quantum Dots: The Importance of Z-Type Ligand
Passivation. J. Am. Chem. Soc. 2018, 140, 15712−15723.
(12) van der Stam, W.; de Graaf, M.; Gudjonsdottir, S.; Geuchies, J.
J.; Dijkema, J. J.; Kirkwood, N.; Evers, W. H.; Longo, A.; Houtepen,
A. J. Tuning and Probing the Distribution of Cu+ and Cu2+ Trap
2663
https://dx.doi.org/10.1021/acs.chemmater.1c00520
Chem. Mater. 2021, 33, 2655−2665

Chemistry of Materials
pubs.acs.org/cm
Article
States Responsible for Broad-Band Photoluminescence in CuInS2
Nanocrystals. ACS Nano 2018, 12, 11244−11253.
(13) Veamatahau, A.; Jiang, B.; Seifert, T.; Makuta, S.; Latham, K.;
Kanehara, M.; Teranishi, T.; Tachibana, Y. Origin of surface trap
states in CdS quantum dots: relationship between size dependent
photoluminescence and sulfur vacancy trap states. Phys. Chem. Chem.
Phys. 2015, 17, 2850−2858.
(14) van der Stam, W.; du Fossé, I.; Grimaldi, G.; Monchen, J. O. V.;
Kirkwood, N.; Houtepen, A. J. Spectroelectrochemical Signatures of
Surface Trap Passivation on CdTe Nanocrystals. Chem. Mater. 2018,
30, 8052−8061.
(15) Amelia, M.; Lincheneau, C.; Silvi, S.; Credi, A. Electrochemical
properties of CdSe and CdTe quantum dots. Chem. Soc. Rev. 2012,
41, 5728−5743.
(16) Fedin, I.; Talapin, D. V. Probing the Surface of Colloidal
Nanomaterials with Potentiometry in Situ. J. Am. Chem. Soc. 2014,
136, 11228−11231.
(17) Carroll, G. M.; Brozek, C. K.; Hartstein, K. H.; Tsui, E. Y.;
Gamelin, D. R. Potentiometric Measurements of Semiconductor
Nanocrystal Redox Potentials. J. Am. Chem. Soc. 2016, 138, 4310−
4313.
(18) Chen, M.; Guyot-Sionnest, P. Reversible Electrochemistry of
Mercury Chalcogenide Colloidal Quantum Dot Films. ACS Nano
2017, 11, 4165−4173.
(19) Hartley, C. L.; Kessler, M. L.; Dempsey, J. L. Molecular-Level
Insight into Semiconductor Nanocrystal Surfaces. J. Am. Chem. Soc.
2021, 143, 1251−1266.
(20) Saniepay, M.; Mi, C.; Liu, Z.; Abel, E. P.; Beaulac, R. Insights
into the Structural Complexity of Colloidal CdSe Nanocrystal
Surfaces: Correlating the Efficiency of Nonradiative Excited-State
Processes to Specific Defects. J. Am. Chem. Soc. 2018, 140, 1725−
1736.
(21) Anderson, N. C.; Hendricks, M. P.; Choi, J. J.; Owen, J. S.
Ligand Exchange and the Stoichiometry of Metal Chalcogenide
Nanocrystals: Spectroscopic Observation of Facile Metal-Carboxylate
Displacement and Binding. J. Am. Chem. Soc. 2013, 135, 18536−
18548.
(22) Kessler, M. L.; Dempsey, J. L. Mapping the Topology of PbS
Nanocrystals through Displacement Isotherms of Surface-Bound
Metal Oleate Complexes. Chem. Mater. 2020, 32, 2561−2571.
(23) Hartley, C. L.; Dempsey, J. L. Electron-Promoted X-Type
Ligand Displacement at CdSe Quantum Dot Surfaces. Nano Lett.
2019, 19, 1151−1157.
(24) Tsui, E. Y.; Hartstein, K. H.; Gamelin, D. R. Selenium Redox
Reactivity on Colloidal CdSe Quantum Dot Surfaces. J. Am. Chem.
Soc. 2016, 138, 11105−11108.
(25) Drijvers, E.; De Roo, J.; Martins, J. C.; Infante, I.; Hens, Z.
Ligand Displacement Exposes Binding Site Heterogeneity on CdSe
Nanocrystal Surfaces. Chem. Mater. 2018, 30, 1178−1186.
(26) Knowles, K. E.; Malicki, M.; Parameswaran, R.; Cass, L. C.;
Weiss, E. A. Spontaneous Multielectron Transfer from the Surfaces of
PbS Quantum Dots to Tetracyanoquinodimethane. J. Am. Chem. Soc.
2013, 135, 7264−7271.
(27) Pu, C.; Dai, X.; Shu, Y.; Zhu, M.; Deng, Y.; Jin, Y.; Peng, X.
Electrochemically-stable ligands bridge the photoluminescence-
electroluminescence gap of quantum dots. Nat. Commun. 2020, 11,
No. 937.
(31) Beygi, H.; Sajjadi, S. A.; Babakhani, A.; Young, J. F.; van Veggel,
F. C. J. M. Surface chemistry of as-synthesized and air-oxidized PbS
quantum dots. Appl. Surf. Sci. 2018, 457, 1−10.
(32) Moreels, I.; Lambert, K.; Smeets, D.; De Muynck, D.; Nollet,
T.; Martins, J. C.; Vanhaecke, F.; Vantomme, A.; Delerue, C.; Allan,
G.; et al. Size-Dependent Optical Properties of Colloidal PbS
Quantum Dots. ACS Nano 2009, 3, 3023−3030.
(33) Jasieniak, J.; Califano, M.; Watkins, S. E. Size-Dependent
Valence and Conduction Band-Edge Energies of Semiconductor
Nanocrystals. ACS Nano 2011, 5, 5888−5902.
(34) Connelly, N. G.; Geiger, W. E. Chemical Redox Agents for
Organometallic Chemistry. Chem. Rev. 1996, 96, 877−910.
(35) Koh, Wk.; Koposov, A. Y.; Stewart, J. T.; Pal, B. N.; Robel, I.;
Pietryga, J. M.; Klimov, V. I. Heavily doped n-type PbSe and PbS
nanocrystals using ground-state charge transfer from cobaltocene. Sci.
Rep. 2013, 3, No. 2004.
(36) Shim, M.; Guyot-Sionnest, P. n-type colloidal semiconductor
nanocrystals. Nature 2000, 407, 981−983.
(37) Hens, Z.; Martins, J. C. A Solution NMR Toolbox for
Characterizing the Surface Chemistry of Colloidal Nanocrystals.
Chem. Mater. 2013, 25, 1211−1221.
(38) Kessler, M. L.; Starr, H. E.; Knauf, R. R.; Rountree, K. J.;
Dempsey, J. L. Exchange equilibria of carboxylate-terminated ligands
at PbS nanocrystal surfaces. Phys. Chem. Chem. Phys. 2018, 20,
23649−23655.
(39) Knauf, R. R.; Lennox, J. C.; Dempsey, J. L. Quantifying Ligand
Exchange Reactions at CdSe Nanocrystal Surfaces. Chem. Mater.
2016, 28, 4762−4770.
(40) Zhao, J.; Holmes, M. A.; Osterloh, F. E. Quantum Confinement
Controls Photocatalysis: A Free Energy Analysis for Photocatalytic
Proton Reduction at CdSe Nanocrystals. ACS Nano 2013, 7, 4316−
4325.
(41) Rajh, T.; Micic, O. I.; Lawless, D.; Serpone, N. Semiconductor
Photophysics. 7. Photoluminescence and Picosecond Charge Carrier
Dynamics in Cadmium Sulfide Quantum Dots Confined in a Silicate
Glass. J. Phys. Chem. A 1992, 96, 4633−4641.
(42) Shiragami, T.; Fukami, S.; Wada, Y.; Yanagida, S. Semi-
conductor Photocatalysis: Effect of Light Intensity on Nanoscale
Cadmium Sulfide-Catalyzed Photolysis of Organic Substrates. J. Phys.
Chem. A 1993, 97, 12882−12887.
(43) Nedoluzhko, A. I.; Shumilin, I. A.; Nikandrov, V. V. Coupled
Action of Cadmium Metal and Hydrogenase in Formate Photo-
decomposition Sensitized by CdS. J. Phys. Chem. C 1996, 100,
17544−17550.
(44) Pagels, N.; Prosenc, M. H.; Heck, J. An ansa-Cobaltocene with
a Naphthalene Handle: Synthesis and Spectroscopic and Structural
Characterization. Organometallics 2011, 30, 1968−1974.
(45) Zherebetskyy, D.; Scheele, M.; Zhang, Y.; Bronstein, N.;
Thompson, C.; Britt, D.; Salmeron, M.; Alivisatos, P.; Wang, L.
Hydroxylation of the surface of PbS nanocrystals passivated with oleic
acid. Science 2014, 344, 1380−1384.
(46) Hendricks, M. P.; Campos, M. P.; Cleveland, G. T.; Jen-La
Plante, I.; Owen, J. S. A tunable library of substituted thiourea
precursors to metal sulfide nanocrystals. Science 2015, 348, 1226−
1230.
(47) Hines, M. A.; Scholes, G. D. Colloidal PbS Nanocrystals with
Size-Tunable Near-Infrared Emission: Observation of Post-Synthesis
Self-Narrowing of the Particle Size Distribution. Adv. Mater. 2003, 15,
1844−1849.
(48) Houtepen, A. J.; Vanmaekelbergh, D. Orbital Occupation in
Electron-Charged CdSe Quantum-Dot Solids. J. Phys. Chem. B 2005,
109, 19634−19642.
(49) Elgrishi, N.; Kurtz, D. A.; Dempsey, J. L. Reaction Parameters
Influencing Cobalt Hydride Formation Kinetics: Implications for
Benchmarking H2-Evolution Catalysts. J. Am. Chem. Soc. 2017, 139,
239−244.
(50) Brozek, C. K.; Hartstein, K. H.; Gamelin, D. R. Potentiometric
Titrations for Measuring the Capacitance of Colloidal Photodoped
ZnO Nanocrystals. J. Am. Chem. Soc. 2016, 138, 10605−10610.
(28) du Fossé, I.; ten Brinck, S.; Infante, I.; Houtepen, A. J. Role of
Surface Reduction in the Formation of Traps in n-Doped II−VI
Semiconductor Nanocrystals: How to Charge without Reducing the
Surface. Chem. Mater. 2019, 31, 4575−4583.
(29) Voznyy, O.; Thon, S. M.; Ip, A. H.; Sargent, E. H. Dynamic
Trap Formation and Elimination in Colloidal Quantum Dots. J. Phys.
Chem. Lett. 2013, 4, 987−992.
(30) Choi, H.; Ko, J.-H.; Kim, Y.-H.; Jeong, S. Steric-Hindrance-
Driven Shape Transition in PbS Quantum Dots: Understanding Size-
Dependent Stability. J. Am. Chem. Soc. 2013, 135, 5278−5281.
2664
https://dx.doi.org/10.1021/acs.chemmater.1c00520
Chem. Mater. 2021, 33, 2655−2665

Chemistry of Materials
pubs.acs.org/cm
Article
(51) Brown, P. R.; Kim, D.; Lunt, R. R.; Zhao, N.; Bawendi, M. G.;
́
Grossman, J. C.; Bulovic, V. Energy Level Modification in Lead
Sulfide Quantum Dot Thin Films through Ligand Exchange. ACS
Nano 2014, 8, 5863−5872.
(52) Harris, R. D.; Bettis Homan, S.; Kodaimati, M.; He, C.;
Nepomnyashchii, A. B.; Swenson, N. K.; Lian, S.; Calzada, R.; Weiss,
E. A. Electronic Processes within Quantum Dot-Molecule Complexes.
Chem. Rev. 2016, 116, 12865−12919.
(53) Tsui, E. Y.; Carroll, G. M.; Miller, B.; Marchioro, A.; Gamelin,
D. R. Extremely Slow Spontaneous Electron Trapping in Photodoped
n-Type CdSe Nanocrystals. Chem. Mater. 2017, 29, 3754−3762.
(54) Humphrey, R. E.; Potter, J. L. Reduction of Disulfides with
Tributylphosphine. Anal. Chem. 1965, 37, 164−165.
(55) Aebli, M.; Piveteau, L.; Nazarenko, O.; Benin, B. M.; Krieg, F.;
Verel, R.; Kovalenko, M. V. Lead-Halide Scalar Couplings in 207Pb
NMR of APbX3 Perovskites (A = Cs, Methylammonium,
Formamidinium; X = Cl, Br, I). Sci. Rep. 2020, 10, No. 8229.
(56) Piveteau, L.; Morad, V.; Kovalenko, M. V. Solid-State NMR
and NQR Spectroscopy of Lead-Halide Perovskite Materials. J. Am.
Chem. Soc. 2020, 142, 19413−19437.
(57) Chen, Y.; Smock, S. R.; Flintgruber, A. H.; Perras, F. A.;
Brutchey, R. L.; Rossini, A. J. Surface Termination of CsPbBr3
Perovskite Quantum Dots Determined by Solid-State NMR Spec-
troscopy. J. Am. Chem. Soc. 2020, 142, 6117−6127.
(58) Piveteau, L.; Ong, T.-C.; Walder, B. J.; Dirin, D. N.; Moscheni,
D.; Schneider, B.; Bär, J.; Protesescu, L.; Masciocchi, N.; Guagliardi,
A.; et al. Resolving the Core and the Surface of CdSe Quantum Dots
and Nanoplatelets Using Dynamic Nuclear Polarization Enhanced
PASS−PIETA NMR Spectroscopy. ACS Cent. Sci. 2018, 4, 1113−
1125.
(59) Lee, D.; Wolska-Pietkiewicz, M.; Badoni, S.; Grala, A.;
́
Lewinski, J.; De Paëpe, G. Disclosing Interfaces of ZnO Nanocrystals
Using Dynamic Nuclear Polarization: Sol-Gel versus Organometallic
Approach. Angew. Chem., Int. Ed. 2019, 58, 17163−17168.
(60) Zhang, J.; Zhang, H.; Cao, W.; Pang, Z.; Li, J.; Shu, Y.; Zhu, C.;
Kong, X.; Wang, L.; Peng, X. Identification of Facet-Dependent
Coordination Structures of Carboxylate Ligands on CdSe Nanocryst-
als. J. Am. Chem. Soc. 2019, 141, 15675−15683.
(61) Hanrahan, M. P.; Stein, J. L.; Park, N.; Cossairt, B. M.; Rossini,
A. J. Elucidating the Location of Cd2+ in Post-synthetically Treated
InP Quantum Dots Using Dynamic Nuclear Polarization 31P and
113Cd Solid-State NMR Spectroscopy. J. Phys. Chem. C 2021, 125,
2956−2965.
(62) Singh, S.; Tomar, R.; ten Brinck, S.; De Roo, J.; Geiregat, P.;
Martins, J. C.; Infante, I.; Hens, Z. Colloidal CdSe Nanoplatelets, A
Model for Surface Chemistry/Optoelectronic Property Relations in
Semiconductor Nanocrystals. J. Am. Chem. Soc. 2018, 140, 13292−
13300.
2665
https://dx.doi.org/10.1021/acs.chemmater.1c00520
Chem. Mater. 2021, 33, 2655−2665