VSU Small Atomically Precise Copper Clusters Article Questions

Break down the title

Go through the abstract

What’s the main focus?

What’s happening and why?

What’s my take on the article?

pubs.acs.org/JPCA
Article
Computational Comparative Analysis of Small Atomically Precise
Copper Clusters
Adebola Adeagbo, Tao Wei, and Andre Z. Clayborne*
Cite This: J. Phys. Chem. A 2020, 124, 6504−6510
Downloaded via VIRGINIA STATE UNIV on January 25, 2023 at 16:37:58 (UTC).
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
ACCESS
Metrics & More
Read Online
Article Recommendations
sı Supporting Information
*
ABSTRACT: Atomically precise copper clusters (APC) have attracted attention for
their promise in sensing, water remediation, and electrochemical technologies.
However, smaller-sized APCs and the evolution of their properties as a function of
size and composition are not clearly understood. Here, we have performed an
investigation into the electronic structure, geometry, and optical properties of small
atomically precise copper clusters using density functional theory (DFT) and timedependent DFT. Through comparative analysis, we show that the electronic
structures of the experimentally characterized clusters, Cu4(PN(C6H5)2CH)4 and
Cu4(SN2C7H11)4, are similar with the closed-shell superatom character 1S21P2. By
changing the ligand on Cu4(PN(C6H5)2CH)4 and Cu4(SN2C7H11)4, there were no
major changes observed in the tetrahedral Cu4 core geometry, electronic structure, or
optical spectra. However, a change in the anchor atom causes an increase in the
electronic gap and induces a hypochromic shift in the onset peak in the optical
spectrum of the small clusters. Increasing the copper core size showed small changes Cu−Cu bond lengths, lower electronic gap
values, and a bathochromic shift in the optical spectra. Computational results not only provide detailed physical insight into APCs
but also aid in identifying compound compositions of small atomically precise nanoclusters from data collected in the experiment.
Kacprzak et al.13 used DFT calculations to theoretically
characterize cyclic thiolated copper clusters. Nanoparticles
such as Cu13{S2CNnBu2}6(acetylide)4+ and
[Cu61(StBu)26S6Cl6H14]+ have been described as superatoms,
which rely on the delocalization of metallic states over the
nanoparticle core.11,14 However, the thiolated copper-hydride
nanocluster Cu25H10(SPhCl2)183− lacks superatom states, but it
is reported to be stable due to atom-ligand interactions.15 In a
more recent study, Han and co-workers synthesized an
atomically precise copper cluster that incorporates a neutral
tetrahedral copper core.16 Interestingly, while researchers have
focused on larger atomically precise Cun nanoparticles (i.e., n >
8), smaller-sized ligated copper nanoclusters are often neglected
theoretically though synthesized readily. A detailed study of
smaller-sized atomically precise copper clusters (APCs) could
provide key structural and electronic information as a starting
point to understand the formation of larger APCs. These details
could lay the foundation toward determining the preferential
copper core size or geometry to be used in nanoscale devices
with varying applications.
1. INTRODUCTION
From the size of their metal cores to their organic ligands,
atomically precise nanoparticles offer the tailor ability needed to
achieve desired electronic and optical properties. Of particular
interest, many studies have focused on atomically precise ligated
nanoparticles of gold and silver. The biocompatibility,
tunability, and wide range of applications for sensing, drug
delivery, and catalysis have excited researchers for decades.1−6
Similar to gold and silver, atomically precise copper
nanoparticles have experienced a renewed interest for
researchers. Copper has significant biological and industrial
importance and is more affordable than gold and silver, which
has sparked researchers to pursue a better understanding of
these nanoparticles. Though copper nanoparticles exhibit a
sensitivity to the surrounding environment that can lead to
difficulty in their characterization,7 recent years have shown that
these can be overcome and showed promises in catalytic,
optoelectronic, and sensing technologies.8−11 For example,
ligated Cu catalysts have successfully converted CO2 to formic
acid. Wang et al.12 illustrated that copper can reduce methylene
blue dye. Gao and co-workers7 showed that Cu6(SC7H4NO)6−
could be used for the electrochemical detection of H2O2. Tang
and co-workers illustrated using experiment and density
functional theory (DFT) that the electrochemical reduction of
CO2 occurs on the ligated copper nanoparticle via a latticehydride mechanism.
In other works, researchers have strived to understand the
geometry and stability of Cu nanoparticles. For example,
© 2020 American Chemical Society
Received: May 4, 2020
Revised: June 29, 2020
Published: July 21, 2020
6504
https://dx.doi.org/10.1021/acs.jpca.0c03992
J. Phys. Chem. A 2020, 124, 6504−6510
The Journal of Physical Chemistry A
pubs.acs.org/JPCA
Article
Herein, our purpose is to gain insight into the structural,
electronic, and optical properties of small atomically precise
ligated copper clusters (i.e., Cu4(SN2C7H11)4 and Cu4(PN(C9H11)2CH)4). We not only probe the intricacies of their
fundamental properties using DFT and time-dependent DFT
but also explore how these properties evolve due to ligand
composition and anchor atom choice. These results are
compared to those obtained through UV−vis experiments.
Further, we address the progression of optical response
properties through changes in copper core size. To the best of
our knowledge, there are no computational studies that address
how the properties of atomically precise copper clusters,
sometimes referred to as ligated superatom clusters, change
due to the number of copper atoms in the core.
2. COMPUTATIONAL METHODS
The calculations in this work were performed using density
functional theory (DFT) as implemented in the GPAW17−19
code at the PBE20 level. Geometry optimizations were
performed with a grid spacing of 0.2 and 7.0 Å of vacuum
surrounding each nanoparticle system. In each simulation, all
atoms were allowed to move freely without constraints until the
forces on all atoms were below a threshold of 0.05 eV/Å. In
previous studies by our group and others, we have found this
method produces results within a reasonable amount of error for
gold and copper nanoparticle systems.7
Initial geometries for Cu4(SN2C7H11)421 (1a), Cu4(PN(C9H11)2CH)4,22 and Cu6(SC7H4NO)6−7 were obtained from
previous experimental reports. Prior to geometry optimization,
we reduced the mesitylene group in the ligand of Cu4(PN(C 9 H11) 2 CH) 422 to phenyl groups to obtain Cu4(PN(C6H5)2CH)4 (2a). Additionally, the R-groups in the ligands
of 1a and 2a were further reduced to methyl/hydrogen groups to
obtain Cu4(SN2C3H7)4 (1b), Cu4(SN2CH3)4 (1c), Cu4(PN(CH3)2CH)4 (2b), and Cu4(PNH2CH)4 (2c). We obtained
Cu4(SNH2CH)4 (2d) by replacing the phosphorus atom in 2c
with a sulfur atom. Similarly, we performed geometry
optimization on the neutral Cu6(SC7H4NO)6− system (3).
Though the Cu6(SC7H4NO)6 system is found in the experiment
as an anion, we chose to compare closed-shell systems. Though
the Cu25(SH)18− (4) nanoparticle has not been synthesized, we
replaced the gold atoms of the Au25(SH)18−23 structure from
previous studies with copper atoms. All nanoparticle systems
were allowed to relax as explained above.
Once all systems were in their lowest ground state, we
performed analyses on their electronic structure and optical
properties. A Bader analysis was performed to determine the
charge on each atom. The spectra of each nanoparticle were
determined using the linear response time-dependent DFT (LrTDDFT)24 module as implemented in GPAW. The oscillator
strengths were obtained and tabulated for the most prominent
transitions for each cluster through the Lr-TDDFT module.
Spectra were rendered through gnuplot.
Figure 1. Comparison of Cu4 nanocluster structures. (A) Total
structures of Cu4 nanoclusters. (B, C) Cu4 metal cores with their
bridging anchor atoms. (D) Isolated metal cores of Cu4 nanoclusters.
Legend: orange sphere, Cu; pink sphere, P; yellow sphere, S; blue
sphere, N; red sphere, O; black sphere, C; and gray sphere, H.
bond length at 3.00 Å that agrees well with the experimental
value. While our results agree with experimental structures, we
wanted to understand the role of the ligand and anchor atom on
the geometry to understand the possible observable effects on
the geometry. Therefore, we altered the R-group and anchor
atom of the ligand.
The geometric changes to the core accompanied by changing
the R-group of the ligands provide some insight into the
importance of the ligand choice and effect on the core. First, we
reduced the R-group on each ligand (Figure S7). For the
Cu4(SN2C7H11)4 system, reducing the R-group from C7H11 to
C3H7 decreased the core bond lengths by 0.37%. By reducing it
further to Cu4(SN2CH3)4, Cu bonds increased by 1.12% as
compared to the Cu4(SN2C3H7)4 system. The R-group
reduction of Cu4(PN(C6H5)2CH)4 to Cu4(PN(CH3)2CH)4
decreased the core Cu bond lengths by 0.05 Å, resulting in a
1.67% decrease. Further reducing it to Cu4(PNH2CH)4
increased the core Cu bond lengths by 0.11 Å, resulting in a
3.73% increase. The noticeable increase by reducing the ligand
to CH3 results from the reduced steric effect of the ligand. Upon
closer investigation of the core, by reducing the ligand to CH3,
the core reduces/increases the distortion in the core. This
illustrates the importance of the ligand interaction and the steric
effect of the ligand on the core geometry. Interestingly, the
changes in core metal−metal distances observed here are similar
to those observed in previous studies on thiolated gold
nanoparticles.25
In addition to investigating the effect of reducing the R-group,
we changed the anchor atom of the ligand to observe any change
in geometry. The anchor atom is defined as the atom bridging
the metal core to the ligand similar to our previous study.25 To
further investigate this observation, an anchor atom change was
performed on the Cu4(PNH2CH)4 system. Changing the
anchor atom to a sulfur atom resulted in a 0.40 Å decrease in
the average core bond lengths. The resulting Cu4(SNH2CH)4
system has an average core bond length of 2.66 Å, which is
within 0.04 Å of the 1a system. Therefore, the anchor atom has a
significant influence on the geometry of these copper systems.
Figure 2 shows the structures of Cu6 and Cu25 systems, and
their average bond lengths are presented in Table 2. Though
both the Cu6 and Cu25 nanoparticles have been studied
3. RESULTS AND DISCUSSION
3.1. Geometry. Figure 1 shows the structures of the Cu4
ligated clusters. The average bond lengths of their core copper
atoms are presented in Table 1. The copper cores in all
structures are tetrahedral or tetrahedral-like (Figure 1). The
average Cu−Cu bond lengths of the relaxed Cu4(SN2C7H11)4
(1a) is 2.70 Å, which is within 0.02 Å of the experimental value.
The Cu4(PN(C6H5)2CH)4 (2a) system has the largest Cu−Cu
6505
https://dx.doi.org/10.1021/acs.jpca.0c03992
J. Phys. Chem. A 2020, 124, 6504−6510
The Journal of Physical Chemistry A
pubs.acs.org/JPCA
Article
Table 1. Average Relaxed and Experimental Core Copper to Copper Atom, Copper to Anchor Atom, Copper to Nitrogen, Sulfur
to Carbon, Phosphorus to Carbon, and Nitrogen to Carbon Atom Bond Lengths (Å) of the Cu4 Systems in This Study and
Experimental Crystal Structures
system
Cu−Cu
Cu−XA
Cu−N
S−C
P−C
N−C
Cu4(SN2C7H11)4 (1a)
Cu4(SN2C3H7)4 (1b)
Cu4(SN2CH3)4 (1c)
Cu4(PN(C6H5)2CH)4 (2a)
Cu4(PN(CH3)2CH)4 (2b)
Cu4(PNH2CH)4 (2c)
Cu4(SNH2CH)4 (2d)
Expt. values21 (1)
Expt. values22 (2)
2.70
2.69
2.72
3.00
2.95
3.06
2.66
2.72
3.00
2.27
2.27
2.27
2.27
2.26
2.27
2.30
2.27
2.27
2.00
2.02
1.99
2.01
2.00
2.01
1.98
1.99
2.01
1.77
1.79
1.78
n/a
n/a
n/a
1.74
1.77
n/a
n/a
n/a
n/a
1.81
1.82
1.79
n/a
n/a
1.81
1.39
1.40
1.34
1.37
1.38
1.30
1.30
1.38
1.37
Figure 2. Ball and stick representation of all ligated copper systems in this study: 1a (Cu4(SN2C7H11)4), 2a (Cu4(PN(C6H5)2CH)4), 3
(Cu6(SC7H4NO)6), and 4 (Cu25(SH)18−). Legend: orange sphere, Cu; pink sphere, P; yellow sphere, S; blue sphere, N; red sphere, O; black sphere, C;
and gray sphere, H.
motif typically observed in atomically precise gold nanoclusters,
which binds only through the sulfur atom. The bonding
interactions are different than other nanoclusters studied here
because there are two ligand interactions with the copper cores,
that is, the nitrogen and either the sulfur or phosphorus atom.
3.2. Electronic Structure. The gap between the highest
occupied molecular orbital (HOMO) and the lowest
unoccupied molecular orbital (LUMO), the HOMO−LUMO
gap (HL gap), was determined for each of the cluster systems
(Table 3). The smallest sized copper clusters (i.e., Cu4R4)
Table 2. Average Bond Lengths (Å) of Cu6 and Cu25 Systems
system
Cu−Cu
Cu−XA
Cu−N
XA−C
N−C
Cu6(SC7H4NO)6 (3)
Cu25(SH)18− (4)
Expt. values26 (3)
2.83
2.62
2.64
2.27
2.24
2.38
2.05
n/a
2.07
1.73
n/a
1.71
1.36
n/a
1.36
previously, we provide their results for comparison to the smaller
copper clusters.7,21 The average Cu−Cu bond lengths of
Cu6(SC7H4NO)6 is 2.83 Å. The average bond length here is
larger than the previously calculated bond lengths and 7.20%
larger than the experimental crystal structure. The source of the
discrepancy may be attributed to the charge differences of the
cluster. In the previous study, the cluster was negatively charged;
here, the cluster is neutral.7 The Cu6 core still retains an
octahedral polyhedron with eight faces. The bonding interaction
of the ligand is similar to the Cu4 systems with interactions
through the nitrogen, carbon, and sulfur atoms. The
Cu25(SH)18− system has the smallest average Cu−Cu bond
length at 2.62 Å. While this structure is theoretical, it has a
geometry similar to that of the Au25(SH)18− cluster with an
icosahedral core.25
Comparing the geometries of the systems here, we note a few
significant similarities and differences. While the number of
copper atoms increases, the bond lengths of the average Cu−Cu
bond do fluctuate only slightly ranging from 2.62 to 2.83 Å.
These remain slightly larger than those observed in bulk copper
surfaces (2.556 Å). It is interesting that the sulfur-containing
species tend to have slightly smaller Cu−Cu bond lengths than
its phosphorus counterparts. Yet, there are no noticeable effects
on the Cu−S, Cu−P, or Cu−N bond lengths as the size of the
copper core increases. The shortest Cu−Cu bonds are observed
in the Cu25 structure (2.62 Å), which may be attributed to the
different bonding situation of the ligands. In Cu25, it has a staple
Table 3. HOMO−LUMO Gaps (eV) of the Ligated Copper
Clusters in This Work
system
HOMO−LUMO gap (eV)
1a
1b
1c
2a
2b
2c
2d
3
4
2.29
2.44
2.44
1.75
1.85
2.02
2.34
1.37
0.94
exhibit the largest gap sizes with Cu4(SR)4 clusters having the
largest HL gaps. Both 1b and 1c have the largest gap sizes of 2.44
eV, whereas 1a is smaller by 0.15 eV. The phosphoruscontaining clusters 2a, 2b, and 2c have lower gaps than each of
the sulfur-containing species with gap sizes of 1.75, 1.85, and
2.02 eV, respectively. Though the change in R-group alters the
gap incrementally, when changing the phosphorus to a sulfur
atom in 2, the HL gap increases by 0.32 eV.
6506
https://dx.doi.org/10.1021/acs.jpca.0c03992
J. Phys. Chem. A 2020, 124, 6504−6510
The Journal of Physical Chemistry A
pubs.acs.org/JPCA
Article
Figure 3. Representation of the HOMO and LUMO (HOMO − 1, HOMO, LUMO, LUMO + 1, LUMO + 2) states for the 1a and 2a systems.
As the number of copper atoms in the core increases, there is a
decrease in the HL gap. Contrary to all other nanoclusters, the
Cu25 nanocluster is the only system with a gap of less than 1 eV.
Comparing Cu6 to 1a and 2a, its 1.37 eV value is smaller than
both fully ligated counterparts. This trend follows closely to
those observed in atomically precise gold nanoparticles, where
the gap decreases as the size of the system increases toward bulk
materials.
In many atomically precise clusters, the electronic structure
can provide insight into the stable nature of the system. Often,
atomically precise nanoclusters such as gold and silver (and
others) can be explained via the superatom (or superatom
complex) model.5,27−30 This model relies on the electrons of the
core being iterant on the metal core. When ligands are involved,
these ligands can be either electron withdrawing or electron
donating. The resultant electron count can result in magic
numbers that are indicative of increased stability. For many
ligated copper systems, the superatom analogy can be hit or
miss. Often, it is highly dependent on the oxidation state of the
copper atom. To determine the possibility of 1a and 2a being
superatoms, we investigated the Kohn−Sham orbitals and
Hirschfield charge analysis.31
Figure 3 shows the electron density of the HOMO − 1,
HOMO, LUMO, and LUMO + 1 states for the 1a and 2a. While
the electron density is delocalized on the core, it is actually
delocalized across the entire nanocluster structure (core and
ligand). One may expect this if the ligands withdraw electrons
from the cluster core, creating a zero-valent superatom.
However, performing a Hirschfield charge analysis indicates
that there is no loss of charge from the copper atoms within the
core. This is similar to the findings of the tetrahedral core of the
Cu23 nanocluster system.16 On the other hand, this indicates the
possibility of having a shell closing of 1s21p2 in the superatom
view.28 A close view of the Kohn−Sham orbitals reveals that the
two highest occupied molecular orbitals have P-symmetry
occupied on the core. However, there is a significant amount of
the electron density spread along the ligands (Figure 3;
Supporting Information).
3.3. Optical Properties. Despite having the same number of
core copper atoms, the effects of the contrasting geometries and
HL gaps of the ligated Cu4 clusters manifest themselves in the
spectroscopic profile. The overall spectroscopic profiles 1a, 1b,
and 1c are similar with minimal differences (Figure 4a). Between
400 and 550 nm, there are five peaks observed for 1a (Figure 4).
Figure 4. Theoretical UV−vis spectra of (a) 1a, 1b, and 1c and (b) 2a,
2b, and 2c.
These peaks correspond to transitions between HOMO and
LUMO + 1 (2.45 eV), HOMO − 2 and LUMO (2.60 eV),
HOMO − 3 and LUMO (2.66 eV), HOMO − 2 and LUMO + 3
(2.81 eV), and HOMO and LUMO + 5 (2.93 eV). The intense
peaks observed between 300 and 400 nm have strong oscillator
strengths. These transitions are from the copper core (mostly)
to states where the density resides either solely on the ligand or is
delocalized across the entire cluster system. Once the ligand is
reduced (i.e., 1b and 1c), the first absorption peak observed
exhibits a blue shift, first from 506 (2.450 eV) to 486 nm (2.550
eV). The peak further blue shifts to 471 nm (2.635 eV) as the Rgroup is further reduced (Figure 4). This is not the only effect
due to the ligand reduction. In 1a, there are intense peaks
observed at 370 and 380 nm. However, these are less intense in
1b and 1c.
6507
https://dx.doi.org/10.1021/acs.jpca.0c03992
J. Phys. Chem. A 2020, 124, 6504−6510
The Journal of Physical Chemistry A
pubs.acs.org/JPCA
The optical spectra for 2a, 2b, and 2c are in Figure 4b. The
first observed peak at 641 nm is a result of a transition from the
HOMO to LUMO + 1. Between 500 and 600 nm, there are two
peaks observed for 1a. These transitions are from the HOMO −
3 to LUMO + 2 (516 nm) and HOMO − 2 to LUMO + 1 (552
nm), respectively. Between 400 and 500 nm there are three
observed peaks corresponding to transitions from the HOMO to
LUMO + 6 (467 nm), HOMO − 1 to LUMO + 6 (421 nm), and
HOMO − 5 to LUMO + 3 (405 nm). Similar to 1a, the peaks
between 300 and 400 nm have strong oscillator strengths. All of
the transitions originate and end in states where the electron
density is delocalized across the entire cluster. When the system
is reduced to 1b and 1c, the first observed peaks exhibit a blue
shift. The first peak of 1b exhibits a blue shift from 641 to 570
nm. Likewise, the first peak of 1c exhibits a blue shift from 570 to
520 nm.
By changing the anchor atom, i.e., phosphorous (2c,
Cu4(PNH2CH)4) to sulfur (2d, Cu4(SNH2CH)4), there are
various noticeable changes in the UV−vis spectra (see Figure S8
for structures). First, 2d has an initial peak at 512 nm, which is
absent once the anchor atom is changed (Figure 5). Second, the
Article
Figure 6. Theoretical UV−vis spectra of ligated copper systems 1a
(Cu4(SN2C7H11)4), 2a (Cu4(PN(C6H5)2CH)4), 3
(Cu6(SC7H4NO)6), and 4 (Cu25(SH)18−).
theoretically and from experiment.7 The discrepancy, i.e.,
experimentally unobserved optical peaks between 700 and 100
nm in theoretical spectra, can be explained by the low intensity
found for the peak transitions. In general, the increased intensity
observed around 400 nm (3.1 eV) for all systems is due to the
involvement of ligand with the electron−hole pair (Supporting
Information), while those transitions between 500 and 700 nm
tend to involve more states mainly localized on the metal core.
It is interesting to ascertain how the systems calculated here
compare to experimentally obtained spectroscopy data. A
comparative analysis of the theoretically simulated optical
absorption of 1a was performed with previous experimental
results for Cu4(SN2C7H11)4. Though the previously reported
UV−vis spectrum is suggested to have the formula [Cu(SN2C7H11)]n (n = 1), the possibility of multiple species could
be expected (Figure 7). Therefore, we also performed
simulations on the Cu(SN2C7H11) monomer as the previous
report suggested that the species may be the monomer instead of
Figure 5. Theoretical UV−vis spectra of Cu4(PNH2CH)4 and its
anchor atom change to Cu4(SNH2CH)4. The inset highlights the
observed peaks between 430 and 500 nm.
first observable peak for 2c resides at 465 nm and is a transition
from HOMO − 1 to LUMO. This transition is similar for the
peak at 449 nm in 2d. Thus, change in the anchor atom from
phosphorus to sulfur results in a hypochromic shift of the
electronic transition from the HOMO − 1 to LUMO (Figure 5).
The remaining spectra of 2c and 2d vary significantly. This may
be due in part to electronic level rearrangement based on the
change of the anchor atom that causes a large change in the
geometry of the copper core as evidenced in Table 1.
Figure 6 provides a comparison of the electronic spectra for
1a, 2a, 3 (Cu6), and 4 (Cu25). The initial absorption peaks for
Cu6 and Cu25 reside between 700 and 1100 nm. While these are
prominent peaks, these are bathochromic compared to the onset
electronic transition in 1a and 2a. Interestingly, there are three
transitions contributing to the initial peaks of systems 3 and 4. In
system 3, the transitions occur between the HOMO → LUMO,
HOMO − 3 → LUMO, and HOMO − 1 → LUMO. In system
4, the transitions occur between the HOMO → LUMO,
HOMO − 2 → LUMO, and HOMO − 1 → LUMO + 1.
Interestingly, though the simulated spectra indicate that there
may be some types of observable UV−vis peaks between 1100
and 700 nm, their oscillator strengths indicate that these may not
be observable in the experimental spectra (Tables S5 and S6).
For example, previous reports on the anion of 3 indicated that
there was a discrepancy between the spectra obtained
Figure 7. (a) Theoretical UV−vis of Cu(MPP) (blue line), Cu2(MPP)
(red, dashed line), and [Cu(MPP)]4 (black, bold line), and (b) the
UV−vis spectra of [Cu(MPP)]n (assuming n = 1) (orange line) and
Cu2(MPP*)1 (red, dashed line). (Reprinted with permission from ref 1.
Copyright 2017 American Chemical Society.).
6508
https://dx.doi.org/10.1021/acs.jpca.0c03992
J. Phys. Chem. A 2020, 124, 6504−6510
The Journal of Physical Chemistry A
pubs.acs.org/JPCA
the Cu4(SR)4 species (Supporting Information). The UV−vis
spectra obtained from the experiment shows one prominent
peak around 300 nm. The simulated spectrum of the monomer
has two prominent peaks at 330 and 270 nm (Figure 7a). On the
other hand, the simulated spectrum at a Gaussian width of 0.15
of 1a has a large prominent peak around 308 nm (Figure 7a).
Interestingly, this prominent peak corresponds to transitions
from deep-lying occupied molecular orbitals to nearly
degenerate LUMO states, i.e., LUMO, LUMO + 3, and
LUMO + 5. These results not only indicate that the
experimental UV−vis spectrum is that of the [Cu(SN2C7H11)]4
instead of the [Cu(SN2C7H11)] monomer but also allow for the
identification of the origin of transitions contributing to
experimentally obtained spectroscopic data for atomically
precise copper nanoclusters.
Relaxed structures for 1a-c, 2a-d, 3, and 4, bond averages
and electron densities for HOMO and LUMO states of
1a, 2a, 3, 4, figures of theoretical optical spectra for 1a,
Cu(MPP), Cu2(MPP), and 2a-c, energies, oscillator
strengths, electron−hole pair transitions and electron
densities for selected peaks for 1a and 2a clusters,
oscillator strengths and electron−hole pair transitions for
Cu6 and Cu25, Information on the Hirschfield analysis and
optical spectra in THF (PDF)
■
AUTHOR INFORMATION
Corresponding Author
Andre Z. Clayborne − Department of Chemistry, Howard
University, Washington, District of Columbia 20059, United
States; orcid.org/0000-0002-0574-0847;
Email: andre.clayborne@howard.edu
4. CONCLUSIONS
The theoretical investigation of geometry, electronic structure,
and optical properties of closed-shell ligated copper nanoclusters were carried out using density functional theory and
time-dependent density functional theory. Geometry optimizations for 1a, 1b, 1c, 2a, 2b, 2c, and 2d were performed. The
influence of the ligand on geometry, electronic structure, and
optical properties was investigated. As one reduces the R-group
on Cu4 nanoclusters, the HOMO−LUMO gap increases by 0.15
eV on average. However, when the anchor atom from
phosphorous to sulfur is changed, there is a large change in
the gap from 2.02 to 2.34 eV. The large change is attributed to
the bond lengths shortening significantly within the copper core.
Based on the evaluation of the Kohn−Sham orbitals and charge
analyses, we determined that nanoclusters have superatom
shells.
A comparative analysis with two larger clusters, Cu6 and Cu25,
was also performed to provide foundational insight about the
variance of properties due to copper core size, which was lacking.
Our results showed that as the size of the copper core increases,
there was no observable trend in the Cu−Cu bond lengths, with
fluctuations between 2.66 and 2.83 Å. However, as the copper
core increases in size, the HOMO−LUMO gap decreases with
1b/c having the largest HL gap (2.44 eV) and Cu25 the smallest
(0.94 eV). The evolution in size of the copper core from n = 4 to
n = 25 led to observable differences in the optical absorption
spectra in part due to the change in electronic structure. The
observable differences in the optical response due to size may be
further enhanced or reduced through molecular interactions that
may occur if these atomically precise nanoclusters are integrated
into optical sensing devices. Further experimental and computational studies into the optical response variance of atomically
precise copper nanoclusters at varying sizes due to (bio)molecular interactions are warranted. Finally, through the
comparison of theoretical and experimental spectra, our study
illustrated how to discern the composition of the structures
observed in the collected spectroscopic profile. These results not
only fill a gap in the knowledge of atomically precise
nanochemistry, but we hope this study encourages more
investigations of ligated copper clusters for sensing devices.
■
Article
Authors
Adebola Adeagbo − Department of Chemistry, Howard
University, Washington, District of Columbia 20059, United
States; orcid.org/0000-0002-4715-5013
Tao Wei − Department of Chemical Engineering, Howard
University, Washington, District of Columbia 20059, United
States; orcid.org/0000-0001-6888-1658
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.jpca.0c03992
Author Contributions
The manuscript was written through the contributions of all
authors. All authors have given approval to the final version of
the manuscript.
Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS
■
REFERENCES
We thank Prof. Trevor Hayton for discussions about copper
nanoclusters. The authors thank the support of the National
Science Foundation (NSF 1831559). Prof. A.Z.C. thanks the
National Natural Science Foundation of China (grant number
21750110448) for partial support. This work used the Extreme
Science and Engineering Discovery Environment (XSEDE),
which is supported by the National Science Foundation grant
number ACI-1548562. Specifically, it used the Bridges system,
which is supported by NSF award number ACI-1445606, at the
Pittsburgh Supercomputing Center (PSC). Prof. T.W. thanks
for computational resources to the Texas Advanced Computing
Center (TACC).
(1) Weerawardene, K. L. D. M.; Häkkinen, H.; Aikens, C. M.
Connections Between Theory and Experiment for Gold and Silver
Nanoclusters. Annu. Rev. Phys. Chem. 2018, 69, 205−229.
(2) Lopez-Acevedo, O.; Kacprzak, K. A.; Akola, J.; Häkkinen, H.
Quantum Size Effects in Ambient CO Oxidation Catalysed by LigandProtected Gold Clusters. Nat. Chem. 2010, 2, 329−334.
(3) Zhao, J.; Jin, R. Heterogeneous Catalysis by Gold and Gold-Based
Bimetal Nanoclusters. Nano Today 2018, 18, 86−102.
(4) Hu, M.; Chen, J.; Li, Z.-Y.; Au, L.; Hartland, G. V.; Li, X.; Marquez,
M.; Xia, Y. Gold Nanostructures: Engineering Their Plasmonic
Properties for Biomedical Applications. Chem. Soc. Rev. 2006, 35,
1084−1094.
ASSOCIATED CONTENT
* Supporting Information
sı
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.jpca.0c03992.
6509
https://dx.doi.org/10.1021/acs.jpca.0c03992
J. Phys. Chem. A 2020, 124, 6504−6510
The Journal of Physical Chemistry A
pubs.acs.org/JPCA
(5) Sharma, S.; Chakrahari, K. K.; Saillard, J. Y.; Liu, C. W. Structurally
Precise Dichalcogenolate-Protected Copper and Silver Superatomic
Nanoclusters and Their Alloys. Acc. Chem. Res. 2018, 51, 2475−2483.
(6) Maity, S.; Bain, D.; Patra, A. Engineering Atomically Precise
Copper Nanoclusters with Aggregation Induced Emission. J. Phys.
Chem. C 2019, 123, 2506−2515.
(7) Gao, X. H.; He, S. J.; Zhang, C. M.; Du, C.; Chen, X.; Xing, W.;
Chen, S. L.; Clayborne, A.; Chen, W. Single Crystal Sub-Nanometer
Sized Cu6 (SR)6clusters: Structure, Photophysical Properties, and
Electrochemical Sensing. Adv. Sci. 2016, 3, No. 1600126.
(8) Yuan, P.; Chen, R.; Zhang, X.; Chen, F.; Yan, J.; Sun, C.; Ou, D.;
Peng, J.; Lin, S.; Tang, Z.; et al. Ether-Soluble Cu53 Nanoclusters as an
Effective Precursor of High-Quality CuI Films for Optoelectronic
Applications. Angew. Chem., Int. Ed. 2019, 58, 835−839.
(9) Chen, A.; Kang, X.; Jin, S.; Du, W.; Wang, S.; Zhu, M. Gram-Scale
Preparation of Stable Hydride M@Cu24 (M = Au/Cu) Nanoclusters. J.
Phys. Chem. Lett. 2019, 10, 6124−6128.
(10) Zall, C. M.; Linehan, J. C.; Appel, A. M. A Molecular Copper
Catalyst for Hydrogenation of CO2 to Formate. ACS Catal. 2015, 5,
5301−5305.
(11) Chakrahari, K. K.; Liao, J.-H.; Kahlal, S.; Liu, Y.-C.; Chiang, M.H.; Saillard, J.-Y.; Liu, C. W. [Cu13{S2CNnBu2}6(Acetylide)4]+: A TwoElectron Superatom. Angew. Chem., Int. Ed. 2016, 55, 14704−14708.
(12) Wang, N.; Ga, L.; Ai, J. Synthesis of Novel Fluorescent Copper
Nanomaterials and Their Application in Detection of Iodide Ions and
Catalysis. Anal. Methods 2019, 11, 44−48.
(13) Kacprzak, K. A.; Lopez-Acevedo, O.; Häkkinen, H.; Grönbeck,
H. Theoretical Characterization of Cyclic Thiolated Copper, Silver, and
Gold Clusters. J. Phys. Chem. C 2010, 114, 13571−13576.
(14) Ghosh, A.; Huang, R.-W.; Alamer, B.; Abou-Hamad, E.; Hedhili,
M. N.; Mohammed, O. F.; Bakr, O. M. [Cu61(StBu)26S6Cl6H14]+: A
Core−Shell Superatom Nanocluster with a Quasi-J36 Cu19 Core and an
“18-Crown-6” Metal-Sulfide-like Stabilizing Belt. ACS Mater. Lett.
2019, 1, 297−302.
(15) Sun, C.; Mammen, N.; Kaappa, S.; Yuan, P.; Deng, G.; Zhao, C.;
Yan, J.; Malola, S.; Honkala, K.; Häkkinen, H.; et al. Atomically Precise,
Thiolated Copper-Hydride Nanoclusters as Single-Site Hydrogenation
Catalysts for Ketones in Mild Conditions. ACS Nano 2019, 13, 5975−
5986.
(16) Han, B.-L.; Liu, Z.; Feng, L.; Wang, Z.; Gupta, R. K.; Aikens, C.
M.; Tung, C.-H.; Sun, D. Polymorphism in Atomically Precise Cu23
Nanocluster Incorporating Tetrahedral [Cu4]0 Kernel. J. Am. Chem.
Soc. 2020, 142, 5834−5841.
(17) Mortensen, J. J.; Hansen, L. B.; Jacobsen, K. W. Real-Space Grid
Implementation of the Projector Augmented Wave Method. Phys. Rev.
B 2005, 71, No. 35109.
(18) Enkovaara, J.; Rostgaard, C.; Mortensen, J. J.; Chen, J.; Dułak,
M.; Ferrighi, L.; Gavnholt, J.; Glinsvad, C.; Haikola, V.; Hansen, H. A.;
et al. Electronic Structure Calculations with GPAW: A Real-Space
Implementation of the Projector Augmented-Wave Method. J. Phys.:
Condens. Matter 2010, 22, No. 253202.
(19) Larsen, A. H.; Mortensen, J. J.; Blomqvist, J.; Castelli, I. E.;
Christensen, R.; Dułak, M.; Friis, J.; Groves, M. N.; Hammer, B.;
Hargus, C.; et al. The Atomic Simulation Environmenta Python
Library for Working with Atoms. J. Phys.: Condens. Matter 2017, 29,
No. 273002.
(20) Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized Gradient
Approximation Made Simple. Phys. Rev. Lett. 1996, 77, 3865−3868.
(21) Nguyen, T.-A. D.; Cook, A. W.; Wu, G.; Hayton, T. W.
Subnanometer-Sized Copper Clusters: A Critical Re-Evaluation of the
Synthesis and Characterization of Cu8(MPP)4 (HMPP = 2-Mercapto5-n-Propylpyrimidine). Inorg. Chem. 2017, 56, 8390−8396.
(22) Rathnayaka, S. C.; Lindeman, S. V.; Mankad, N. P. Multinuclear
Cu(I) Clusters Featuring a New Triply Bridging Coordination Mode of
Phosphaamidinate Ligands. Inorg. Chem. 2018, 57, 9439−9445.
(23) Akola, J.; Kacprzak, K. A.; Lopez-Acevedo, O.; Walter, M.;
Grönbeck, H.; Häkkinen, H. Thiolate-Protected Au25 Superatoms as
Building Blocks: Dimers and Crystals. J. Phys. Chem. C 2010, 114,
15986−15994.
Article
(24) Walter, M.; Häkkinen, H.; Lehtovaara, L.; Puska, M.; Enkovaara,
J.; Rostgaard, C.; Mortensen, J. J. Time-Dependent Density-Functional
Theory in the Projector Augmented-Wave Method. J. Chem. Phys.
2008, 128, No. 244101.
(25) Pohjolainen, E.; Häkkinen, H.; Clayborne, A. The Role of the
Anchor Atom in the Ligand of the Monolayer-Protected Au25(XR)18−
Nanocluster. J. Phys. Chem. C 2015, 119, 9587−9594.
(26) Gao, X.; He, S.; Zhang, C.; Du, C.; Chen, X.; Xing, W.; Chen, S.;
Clayborne, A.; Chen, W. Single Crystal Sub-Nanometer Sized
Cu6(SR)6 Clusters: Structure, Photophysical Properties, and Electrochemical Sensing. Adv. Sci. 2016, 3, No. 1600126.
(27) Reveles, J. U.; Clayborne, P. A.; Reber, A. C.; Khanna, S. N.;
Pradhan, K.; Sen, P.; Pederson, M. R. Designer Magnetic Superatoms.
Nat. Chem. 2009, 1, 310−315.
(28) Häkkinen, H. Electronic Structure: Shell Structure and the
Superatom Concept. In Frontiers of Nanoscience; Elsevier Science, 2015;
Vol. 9, pp 189−222.
(29) Clayborne, P. A.; Lopez-Acevedo, O.; Whetten, R. L.; Grnbeck,
H.; Hkkinen, H. Evidence of Superatom Electronic Shells in LigandStabilized Aluminum Clusters. J. Chem. Phys. 2011, 135, No. 094701.
(30) Lopez-Acevedo, O.; Clayborne, P. A.; Häkkinen, H. Electronic
Structure of Gold, Aluminum, and Gallium Superatom Complexes.
Phys. Rev. B: Condens. Matter Mater. Phys. 2011, 84, No. 035434.
(31) Hirshfeld, F. L. Bonded-Atom Fragments for Describing
Molecular Charge Densities. Theor. Chim. Acta 1977, 44, 129−138.
6510
https://dx.doi.org/10.1021/acs.jpca.0c03992
J. Phys. Chem. A 2020, 124, 6504−6510