Articles

https://doi.org/10.1038/s41557-018-0159-8

Robust gold nanorods stabilized by bidentate

N-heterocyclic-carbene–thiolate ligands
Michelle J. MacLeod, Aaron J. Goodman, Hong-Zhou Ye   , Hung V.-T. Nguyen, Troy Van Voorhis and
Jeremiah A. Johnson   *

Although N-heterocyclic carbenes (NHCs) have demonstrated outstanding potential for use as surface anchors, synthetic chal-
lenges have limited their application to either large planar substrates or very small spherical nanoparticles. The development
of a strategy to graft NHCs onto non-spherical nanomaterials, such as gold nanorods, would greatly expand their utility as
surface ligands. Here, we use a bidentate thiolate–NHC–gold(i) complex that is easily grafted onto commercial cetyl trimeth-
ylammonium bromide-stabilized gold nanorods through ligand exchange. On mild reduction of the resulting surface-tethered
NHC–gold(i) complexes, the gold atom attached to the NHC complex is added to the surface as an adatom, thereby precluding
the need for reorganization of the underlying surface lattice upon NHC binding. The resulting thiolate–NHC-stabilized gold
nanorods are stable towards excess glutathione for up to six days, and under conditions with large variations in pH, high and
low temperatures, high salt concentrations, or in biological media and cell culture. We also demonstrate the utility of these
nanorods for in vitro photothermal therapy.

N
-heterocyclic carbenes (NHCs) have emerged as promis-
ing ligands for surface chemistry1,2. NHC-stabilized palla-
dium3–8, nickel9,10, platinum11, ruthenium12–16 and gold17–20
nanoparticles have been studied in the context of catalysis, NHC-
stabilized gold nanoparticles21,22 and planar surfaces23–25 have
demonstrated potential for biomedical applications, and related
persistent carbene-stabilized Si surfaces have expanded the scope of
Si-functionalization methods26. Despite these successes, and unam-
biguous evidence that NHC monolayers can be more stable than
thiol-based monolayers24, there remain critical gaps in the chemistry
and scope of NHC-functionalized nanomaterials. For example, non-
spherical NHC-stabilized nanomaterials, such as gold nanorods, are
conspicuously absent in the literature. The installation of NHCs
onto such substrates represents a unique synthetic challenge that,
if overcome, could significantly expand the utility of NHC-based
interfaces. For example, thermally stable NHC-functionalized gold
nanorods could be applied in the context of photothermal therapy
(PTT), wherein laser irradiation of the nanorod induces local heat-
ing and cell death27.
Protocols for the synthesis of gold nanomaterials with precise
sizes and shapes have been finely tuned over decades28. These
methods often involve the use of intermediate stabilizing ligands
such as cetyl trimethylammonium bromide (CTAB) or citrate in
water. Due to the limited stability of CTAB- and citrate-passiv-
ated nanomaterials towards changes in pH and electrolyte con-
centration, a subsequent ligand exchange step, usually with a
thiol-based ligand, is used before application of the nanorod28. In
the context of NHC surface chemistry, it would be ideal to sim-
ply exchange CTAB or citrate from commercial nanorods with
an NHC of interest; however, existing ligand exchange protocols
(Fig. 1a) that involve the formation of free NHCs are unlikely to
prove successful with ligands such as CTAB or citrate in aque-
ous solutions6–8,16,19,29,30. Moreover, direct reduction methods for
the synthesis of NHC-stabilized nanoparticles have so far only
provided relatively small (≤​15 nm) exclusively spherical nanoma-
terials (Fig.  1a)11,12,17,18,21,22,31–33. Thus, a different strategy for the
installation of NHCs onto gold nanomaterials of any arbitrary size
and shape is needed.
Here, we report an approach for the delivery of bidentate NHC–
thiolate ligands to gold nanorod surfaces that breaks the above limi-
tations and yields robust nanorods (NHC@Au nanorods, Fig. 1b).
Our strategy leverages the well-established ligand exchange of thio-
lates onto CTAB-stabilized gold nanorods28 to bring masked NHCs
near the nanorod surface. Subsequent activation of the masked
NHC provides NHC@Au nanorods where the original nanorod
size and shape are maintained, which suggests minimal nanorod
restructuring from the ligand addition34. We hypothesized that
installation of NHCs via reduction of NHC–gold(i) complexes teth-
ered to the nanorod surface would yield stable interfaces without
the need for reorganization of the underlying gold lattice to achieve
optimal NHC surface binding. This approach was inspired by the
often observed etching of gold surfaces by free NHCs24,29,35 as well
as work from Glorius, Fuchs and co-workers, which suggested that
when NHCs are deposited onto planar gold substrates under ultra-
high vacuum they tend to abstract a gold atom from the surface
lattice to generate translationally mobile NHC@Au adatom com-
plexes36. In addition, it may also slow the formation of bis-NHC
gold complexes, which have been observed on planar gold sub-
strates37. Indeed, here we show that this ‘adatom addition’ method
(Fig. 1b) leads to NHC@Au nanorods that are stable under a wide
range of harsh conditions, including those where traditional NHC
and thiol@Au nanorods are unstable. Finally, in vitro cell culture
assays suggest that NHC@Au nanorods are promising materials for
laser-induced PTT.
Results and discussion
Masked bidentate NHC–thiolate design and synthesis. We began
our studies with the synthesis of poly(ethylene glycol) (PEG)-
based imidazolium salt 1-Br​ (Fig.  2a) via alkylation of 5-ethynyl-
1-methyl-1H-imidazole (1​ ) with 2-bromoethyl 2-nitrobenzyl
sulfide (2​) followed by copper-catalysed azide-alkyne cycloaddition
with azido-terminated 2 kDa PEG (3​) (58% yield over two steps;

Department of Chemistry, Massachusetts Institute of Technology, Cambridge, MA, USA. *e-mail: jaj2109@mit.edu

Nature Chemistry | www.nature.com/naturechemistry

Articles
a R
R R R R
N N N N
R R R R
Reduction
Au
X ‘CTAB-poor’ gold nanorods were unstable under the same irradia-
NHC-Au(I) NHC-stabilized
complex gold nanoparticle
b R3
N
NO2
CTAB@Au
nanorods, Br
Au N late photogeneration, we compared the stability of the commercial
R3
S hν
N N CTAB Reduction
exchange S S
Au
Br
1-Au 1-Au@Au
Fig. 1 | Strategies for installation of NHCs onto gold nanomaterials. a,
Previous methods involve the reduction of preformed or in situ generated
(the latter from imidazolium tetrahaloaurate salts) NHC–metal (for
example, gold) complexes, or the displacement of weakly bound (for
example, thioether or amine) ligands with free NHCs. So far, it has not yet
been possible to install NHCs onto gold nanomaterials of arbitrary size and
shape with these approaches. X, a coordinating ligand, typically a halogen
such as Cl or Br; R and R2, alkyl, aryl or other substituents. b, Here, we
introduce a bidentate thiolate masked NHC strategy whereby exchange of
CTAB ligands on commercial CTAB@Au nanorods with a photogenerated
thiolate is followed by NHC installation. Shown is the synthesis of thiolate
monolayer 1-Au@Au from photodeprotection of 1-Au​and its subsequent
reduction to generate NHC@Au-I, which introduces a new surface-bound
NHC–gold adatom complex. The resulting bidentate thiolate–NHC-
stabilized gold nanorods have the same size and shape as their parent
commercial nanorods and are robust towards a range of stringent
conditions. R3, triazole-conjugated polyethylene glycol.
see Supplementary Information for details; Supplementary Figs. 1–
4). Conversion of 1-Br​to gold(i) complex 1-Au​was achieved in
85% yield following methods reported by Nolan and co-workers
for related imidazolium salts38. We also prepared the hydrogen
carbonate salt 1-HCO3​(see Supplementary Information, 90%
yield from 1-Br​) to compare how the NHC installation method
impacts nanorod stability39. 1H NMR, 13C NMR and matrix-assisted
laser desorption ionization time-of-flight (MALDI–TOF) mass
spectrometry (MS) supported the structures of these compounds
(Fig. 2b for 1-Au​; Supplementary Figs. 5–9). These bidentate ligands
were designed to enable in situ photogeneration of the thiolate and
subsequent ligand exchange onto commercially available CTAB-
stabilized gold nanorods followed by NHC installation (Fig.  1b).
Although NHC–thiolate bidentate ligands have not been used
in the context of surface functionalization, we reasoned that they
would provide enhanced surface stability, as is known for related
multidentate ligands8,40–44. The ortho-nitrobenzyl protecting group
enables on-demand thiolate generation following irradiation with
ultraviolet light45, while PEG is frequently used to impart aqueous
solubility and biocompatibility to gold nanomaterials46–48.
NHC@Au nanorod synthesis and characterization. Commercially
available CTAB-stabilized (3  mM) 40  ×​ 10 nm gold nanorods
(CTAB@Au) were used to optimize our ligand installation condi-
tions. We first confirmed the stability of CTAB@Au towards long-
wavelength ultraviolet light, which is required for unmasking of the
NATure CHemisTry
R thiolate component of 1-Br​, 1-HCO3​ and 1-Au​. In the presence of
R2 R2
excess CTAB, the nanorods were stable for at least 5 h under 365 nm
R
N N
R light (8 W desktop lamp), as determined by the lack of significant
S changes in the longitudinal surface plasmon resonance (SPR) band in
Ligand exchange the UV–vis spectrum (Supplementary Fig. 10). When excess CTAB
was removed via centrifugation before irradiation, the resulting
Ligand-stabilized tion conditions. However, exposure of ‘CTAB-poor’ gold nanorods
gold nanoparticle
(here, with a to 1-Au​(Supplementary Fig. 11; based on ligand optimization stud-
thioether ligand) ies (Supplementary Fig.  12), 0.1 equiv. of masked bidentate ligand
to CTAB was used throughout this work) for 5 min before irradia-
R3
tion yielded stable nanorods. To validate the importance of thio-
N N CTAB@Au exposed to 1-Br​both with and without ultraviolet irra-
diation to the same CTAB@Au exposed to a previously reported21
PEGylated imidazolium salt that lacks a thioether/thiolate com-
ponent. After dialysing to remove unbound ligand, only 1-Br​ with
ultraviolet irradiation yielded stable, soluble nanorods (Fig. 3a, +​UV,
NHC@Au-I +​S). Aurophilic interactions could potentially stabilize the nanorods
exposed to 1-Au​without the thiolate component49. As shown in
Supplementary Fig. 13, the longitudinal SPR was significantly blue-
shifted for CTAB@Au treated with 1-Au​and not irradiated with
ultraviolet light, while it was much less shifted when 1-Au​and ultra-
violet light were used together. These observations and transmission
electron microscopy (TEM) measurements (Supplementary Fig. 13
and Supplementary Table 1) suggest that 1-Au​on its own induces a
change in the nanorod size and/or aspect ratio, while in situ photo-
generation of the thiolate component of 1-Au​inhibits these unde-
sired alterations.
Direct comparisons of the UV–vis spectra for CTAB@Au before
and after installation of the thiolates of 1-Br​ and 1-Au​ (nanorods
are named 1-Br@Au and 1-Au@Au, respectively) are provided
in Fig.  3b (see Supplementary Fig.  14 for 1-HCO3@Au). A rep-
resentative TEM image of 1-Au@Au is provided in Fig.  3c (see
Supplementary Table 1 for TEM-measured sizes and aspect ratios;
see Supplementary Fig.  14 for TEM images of 1-Br@Au and
1-HCO3@Au). In general, installation of our photogenerated thio-
late ligands produced slightly smaller nanorods compared to CTAB@
Au, as assessed by UV–vis and TEM (Supplementary Table 1). The
same phenomenon was also observed for a traditional PEG-thiol
(PEG-SH) ligand (Supplementary Table  1 and Supplementary
Fig. 14). Fourier-transform infrared spectroscopy provides support
for the presence of our PEG-based ligands (Supplementary Fig. 15);
however, it should be noted that we expect a small amount of CTAB
to remain bound to the nanorods following ligand exchange in each
case, which is commonly observed in ligand exchange protocols
using thiols50,51. Although ligand exchange onto ‘CTAB-poor’ gold
nanorods was possible (vide supra), the process of CTAB removal
produced nanoparticle impurities (Supplementary Fig. 16), which
led us to abandon this approach in this work. In summary, the above
results confirm that in situ photogeneration of thiolates introduces
our ligands onto commercial CTAB@Au in water in a manner that
minimally perturbs the nanorod size and aspect ratio.
Next, we investigated methods for the installation of the NHC
components of our ligands onto these nanorod surfaces. To realize
our adatom addition strategy, 1-Au@Au was treated with borane
tert-butylamine complex (2 equiv. to 1-Au​ ) in THF to provide
NHC@Au-I. For comparison to traditional methods for NHC
installation, 1-Br@Au was exposed to potassium bis(trimethylsilyl)
amide (KHMDS) in THF to provide NHC@Au-II, and 1-HCO3@
Au was stirred for 3 days in THF to provide NHC@Au-III. The
UV–vis spectra for 1-Au@Au and NHC@Au-I were similar
(Fig.  4a) while the longitudinal SPR bands for NHC@Au-II and
NHC@Au-III (Supplementary Fig.  17) were slightly red- and
blueshifted compared to 1-Br@Au and 1-HCO3@Au, respectively,
which suggests that exposure to strong base (KHMDS) and/or the
Nature Chemistry | www.nature.com/naturechemistry
NATure CHemisTry Articles
a O 44 O 44

NO2
N N
(i) Br N N
S O2N O2N
N N
2 Br Au(SMe3)Cl
N N S N N S
N N K2CO3
O
(ii) N3 Au
44
1 3 1-Br Br 1-Au

b m c
O
O o
n 43
l [M – Br]+ = obs. 2,449.9
k m, n calc. 2,449.2
N
N j o
O2N
N
i d e
c
N N S f
a b h g 44 (C2H4O) }
Au
a
Br
i + CHCl3 CH2Cl2

d
j
k b l c
e gh f

8.4 8.0 7.6 7.2 6.8 6.4 6.0 5.6 5.2 4.8 4.4 4.0 3.6 3.2 2.8 1,500 2,000 2,500 3,000 3,500
Chemical shift (ppm) m/z

Fig. 2 | Synthesis and molecular characterization of PEGylated masked bidentate ligands. a, Synthesis of 1-Au​. Alkylation of commercially available 1​
followed by copper-catalysed azide-alkyne cycloaddition (CuAAC) to install azido-polyethylene glycol 3​yields imidazolium salt 1-Br​. Exposure of 1-Br​
to Au(SMe2)Cl in the presence of K2CO3 produces the macromolecular gold complex 1-Au​. In this work, 1-Br​is used as the ‘traditional’ NHC source for
subsequent surface studies (that is, the free NHC is generated by deprotonation of 1-Br​). In contrast, 1-Au​is used for adatom addition. b,c, 1H NMR
(CDCl3, 500 MHz) spectrum (b) and MALDI–TOF spectrum (c) for 1-Au​. The data corroborate the proposed structure of 1-Au​.

a b c
CTAB@Au
+UV, +S
1-Au@Au
–UV, +S
1-Br@Au
+UV, –S
Absorbance (a.u.)
Absorbance (a.u.)

500 600 700 800 900 1,000 500 600 700 800 900 1,000
Wavelength (nm) Wavelength (nm)

Fig. 3 | Installation of masked NHC–thiolate ligands onto commercial CTAB@Au nanorods. a, UV–vis spectra for CTAB@Au exposed to 1-Br​before (–UV,
+​S) and after (+​UV, +​S) exposure to ultraviolet light. The results demonstrate that photogeneration of the thiolate component of 1-Br​ (+​UV, +​S) is critical
for generating stable gold nanorods (that is, the thioether of 1-Br​doesn’t stabilize the nanorods on its own). Also shown are data collected for a non-thiol-
containing PEGylated imidazolium salt where both N substituents are methyl (+​UV, –S), affirming the importance of the thiolate component. b, UV–vis
spectra of CTAB@Au, 1-Br​ and 1-Au​thiolate-exchanged nanorods. Thiolate exchange leads to minimal changes in the transverse and longitudinal SPR
bands compared to the UV–vis spectra of the parent CTAB@Au, suggesting that this process does not dramatically impact the nanorod size and shape. c,
TEM image of 1-Au@Au. Scale bar, 30 nm. The image confirms the presence of uniform nanorods following ligand exchange.

free NHC induced some reorganization of the nanorod structure. Stability and density functional theory studies reveal the impact
Nonetheless, TEM indicated that the NHC installation method had of adatom addition. To assess the stability of bidentate NHC–thi-
little obvious effect on the size and aspect ratio of the nanorods olate-stabilized nanorods NHC@Au-I-III, we exposed each sample
(Fig. 4b, Supplementary Fig. 17 and Supplementary Table 1). and its precursor (1-Au@Au, 1-Br@Au, 1-HCO3@Au) to a wide

Nature Chemistry | www.nature.com/naturechemistry

Articles
a As an extreme test of the ability of our ligands to protect the
R3
N
R3
Br
Au N Under these conditions, only NHC@Au-I and NHC@Au-II dis-
TBAB
N N
THF
S S
1-Au@Au NHC@Au-I
b
1-Au@Au
NHC@Au-I
Absorbance (a.u.)
500 600 700 800
Wavelength (nm)
c
Fig. 4 | Characterization of NHC@Au-I. a, Synthesis of NHC@Au-I from
1-Au@Au via reduction of the surface-tethered NHC complex to achieve
adatom addition. TBAB, tert-butylamine borane; THF, tetrahydrofuran;
R3, triazole-conjugated polyethylene glycol. b, UV–vis spectra for 1-Au@
Au and NHC@Au-I showing that the size and shape of the nanorods are
maintained on conversion of 1-Au@Au to NHC@Au-I (that is, adatom
addition does not dramatically affect the nanorod optical properties). c,
TEM image of NHC@Au-I confirming the presence of uniform nanorods
following adatom addition (for comparison to 1-Au@Au see Fig. 3c). Scale
bar, 30 nm.
range of pH values (from 2 to 14), high electrolyte concentrations
(up to 1 M NaCl) and gold etching conditions (25, 50 and 100 mM
KCN) (Supplementary Figs.  18–35)48,50,51. In these studies, both
NHC@Au-I and its precursor 1-Au@Au displayed superior stabil-
ity to all of the other samples. Minimal changes in SPR bands for
NHC@Au-I and 1-Au@Au were observed after three days (longest
examination time) in aqueous solution at pH 2 and 14 and after up
to 16 days in aqueous 1 M NaCl (Supplementary Figs. 18–20), while
NHC@Au-II and NHC@Au-III as well as their precursors 1-Br@
Au and 1-HCO3@Au, showed long-term stability (minimal change
in the SPR band over several days) within a narrower pH range of 7
to 11 and in ≤​100 mM aqueous NaCl (Supplementary Figs. 21–32).
NATure CHemisTry
gold nanorod surface, we exposed each sample to aqueous solutions
of KCN at various concentrations (Supplementary Figs.  33–35).
played some resistance to etching in 50 mM KCN for 10 min. These
observations suggest that although aurophilic interactions may sta-
bilize 1-Au@Au, there is additional stability gained by the bidentate
NHC–thiolate ligand structure. In addition, NHC@Au-I appears to
perform the best under the most stringent conditions, suggesting
that the addition of new gold atoms to the surface is advantageous.
To test whether or not increasing the ligand density would
enhance the etching resistance of NHC@Au-I, 1-Au@Au was
treated with an additional 0.034 or 0.085 equiv. (compared to CTAB
in CTAB@Au nanorods) of 1-Au​and exposed to ultraviolet light.
The longitudinal SPR bands for both samples were slightly blue-
shifted compared to the parent 1-Au@Au (Supplementary Fig. 36).
Reduction followed by TEM imaging (Supplementary Figs.  37
and 38) revealed that the nanorod shape was not significantly
modified by exposure to additional 1-Au​. However, the stability
of these nanorods towards KCN etching was markedly improved:
both samples were completely resistant to 50 mM KCN for 10 min
with significant, though greatly redshifted, SPR bands after 25 min
(Supplementary Figs.  39 and 40, respectively). Once again high-
lighting the importance of adding a gold atom to the surface, the
900 1,000 non-reduced samples were rapidly etched under these condi-
tions regardless of the presence of additional thiolate ligand 1-Au​
(Supplementary Figs. 41 and 42, respectively).
Density functional theory (DFT) calculations were used to eval-
uate the binding energetics of analogues of our bidentate ligands to
a fixed 19-gold-atom cluster where the NHC-bound gold atom was
allowed to relax. Docking of the thiyl component of model ligand
1b+ (Fig. 5a) to the model gold surface produced 1b+@Au with a
binding energy of 1.78 eV, which reflects the Au–S bond strength.
Docking of the tethered NHC 1b@Au (Fig. 5b) provided NHCb@
Au-II with an additional binding energy of 1.57 eV. Notably, the lat-
ter value, which reflects the NHC–gold bond strength, is relatively
small, perhaps due to steric strain at the interface. The overall bind-
ing energy for this bidentate ligand that lacks adatom addition capa-
bility was 3.35 eV.
Docking of 1b-Au-Br (Fig. 5c), an analogue of 1-Au​, provided
1b-Au-Br@Au with a binding energy of 1.49 eV. This value is some-
what less than the value for 1b@Au; however, docking of the reduced
ligand, 1b-Au@Au (Fig. 5d), led to NHCb@Au-I with an additional
binding energy of 2.45 eV. This value is within the range of bind-
ing energies calculated for other NHC–gold adatom complexes
on gold substrates (from ~1.7 eV to 2.69 eV)52. Thus, the overall
bidentate ligand binding energy in NHC@Au-I (3.94 eV) is 0.59 eV
greater than in NHC@Au-II (3.35 eV). Moreover, the NHC–gold
bond was strengthened significantly to 2.84 eV in NHCb@Au-I
(Supplementary Fig.  43), compared to 1.57 eV in NHCb@Au-II,
which suggests that the addition of an adatom in the former leads
to less steric strain. Notably, the NHC–gold bond in NHCb@Au-I
(2.84 eV) is significantly stronger than the bond between the NHC–
gold complex and the underlying gold surface (2.45 eV); desorption
of the ligand should involve removal of the newly added adatom.
Glorius, Fuchs and co-workers proposed that on planar gold
substrates, NHC-mediated extraction of gold atoms from the sur-
face to generate new adatoms involved concerted movement of an
underlying lattice atom to the surface36. In such systems with a large
ratio of gold atoms to NHCs, adatom addition may not lead to sub-
stantial increases in binding energy; however, extraction of surface
gold atoms by NHCs on NHC@Au nanorods may leave vacancies
within the bulk nanorod. Previous work has shown that there is a
~0.7–1 eV energy penalty for removal of a bulk gold atom, with sub-
stantial dependence on geometry53,54. To calibrate our simulations,
we calculated that the formation of a such a vacancy in a cluster of
Nature Chemistry | www.nature.com/naturechemistry
NATure CHemisTry Articles
a b

19 Au

–1.78 eV –1.57 eV

1b+ Au atom
extracted
+
1b @Au 1b@Au NHCb@Au-II

c d

19 Au

–1.49 eV –2.45 eV

1b-Au-Br
New Au
atom
1b-Au@Au 1b-Au(0)@Au NHCb@Au-I

Fig. 5 | DFT calculations. a, A non-PEGylated thiyl-imidazolium salt (1b+) binds to a model 19-atom gold substrate with a binding energy of 1.78 eV to
form 1b+@Au. Note that the bromide counterion is neglected for simplicity. b, Deprotonation of 1b+@Au provides the surface-tethered free NHC 1b@Au,
which then binds to the substrate with an additional binding energy of 1.57 eV to yield NHCb@Au-II, which is analogous to NHC@Au-II. A gold atom from
the underlying lattice is extracted to accommodate the NHC (red arrow). The overall binding energy of the bidentate ligand is calculated to be 3.35 eV.
c, A non-PEGylated thiyl-NHC–Au–Br complex (1b-Au-Br) binds to a model 19-atom gold substrate with a binding energy of 1.49 eV to form 1b-Au-Br@
Au. Reduction of 1b-Au@Au provides the surface-tethered NHC–gold(0) complex, which then binds to the substrate with an additional binding energy
of 2.45 eV to yield NHCb@Au-I, which is analogous to NHC@Au-I. Addition of a new gold atom (green arrow) precludes the need for modification of the
underlying lattice. The overall binding energy of the bidentate ligand is calculated to be 3.94 eV.

19 gold atoms would incur a 1.89 eV energy penalty (Supplementary
Fig.  43). Thus, we propose that the enhanced stability of NHC@
Au-I and NHCb@Au-I compared to the traditional NHC-based
ligands is due at least in part to the installation of one adatom per
NHC ligand, which precludes the need for lattice reorganization. To
our knowledge, this adatom addition approach represents a funda-
mentally different way to install NHC ligands onto surfaces.
Stability under bio-related conditions and in vitro PTT studies
using NHC@Au-I. Given the outstanding robustness of NHC@
Au-I, we sought to explore its preliminary biological applications.
Both thiol- and NHC-stabilized gold nanomaterials are susceptible
to degradation by biological thiols such as glutathione (GSH); to our
knowledge, there are no examples of NHC-stabilized gold nanoma-
terials that have shown long-term stability towards 2 mM GSH for
24 h (in vivo, GSH concentration ranges from μ​M to ~10 mM)21,22.
Remarkably, when NHC@Au-I was exposed to 4 mM GSH in water
there was almost no change in the SPR bands over 6 days (Fig. 6a),
which suggests that these nanorods are stable towards GSH.
As discussed above, for use in PTT, nanorods must be stable at
high temperatures55–57. On the other hand, very low temperatures
are often encountered in biological applications (for example, long-
term cell storage). To test the temperature stability of our nanorods,
aqueous solutions of NHC@Au-I were cooled to −​78 °C or heated
to 95 °C for 5 h. Under these conditions, no significant changes
in the SPR bands were observed (Fig.  6b). In contrast, analogous
PEG-SH-stabilized nanorods (PEG-S@Au) experienced extensive
degradation at both low and high temperatures (Fig. 6c). It should
be noted that the temperature stability of PEG-S@Au could poten-
tially be augmented by increasing the amount of PEG-SH used in
the CTAB exchange step44. Nevertheless, our data show that at the
same ligand concentration, NHC@Au-I is significantly more stable
towards temperature variations than PEG-S@Au.
Progressing to preliminary PTT experiments, we irradiated solu-
tions of NHC@Au-I with a near-infrared (830 nm) laser using a flu-
ence of 6 W cm−2 for 8 min. As expected, the solution temperature
increased linearly with NHC@Au-I concentration (Fig. 6d), which
confirms that our nanorods are capable of converting absorbed
light into heat. In addition, NHC@Au-I was stable for at least three
cycles of irradiation (Fig. 6e). Next, MCF7 human breast adenocar-
cinoma cell viability assays were conducted to assess the potential of
NHC@Au-I for in vitro PTT. Cells were irradiated under the same
conditions used above (830 nm laser, 6 W cm−2 for 8 min) to con-
firm that laser irradiation alone did not induce cell death (Fig. 6f).
Incubation of NHC@Au-I in cell culture media for 3 days at 37 °C
led to no change in the longitudinal SPR band (the transverse band
overlaps with cell media), which suggests that the nanorods are
stable in cell media (Supplementary Fig. 44)19. Cell viability assays
(CellTiter-Glo) indicated that doses of 3.8 and 7.5 μ​g of NHC@Au-I
in the absence of laser irradiation were well tolerated, with 91 ±​  16%
(P = 0.59) and 90 ±​  10% (P = 0.46) cell viability, respectively (Fig. 6f
and Supplementary Fig.  45). Laser irradiation of cells exposed
to NHC@Au-I led to significant cell killing: cell viabilities were
24 ±​  30% (P =​ 0.026) for the 3.8 μ​g dose and 6 ±​  1% (P =​  0.0001) for
the 7.5 μ​g dose (Fig. 6f). Thus, the combination of NHC@Au-I and
laser irradiation leads to selective cell killing. These results show
that NHC-functionalized interfaces can be applied in cellular deliv-
ery assays; they also suggest that NHC@Au-I could be a viable can-
didate for translation to in vivo PTT.
Conclusions
In summary, we have demonstrated a bidentate ligand strategy for
the generation of NHC@Au surface linkages that overcomes the size,
shape and solvent limitations of previous methods. Three methods
for the installation of NHCs via this strategy were experimentally
tested; and an ‘adatom addition’ approach involving reduction58 of
thiolate-tethered NHC–Au complexes proved to be the most suc-
cessful, providing robust gold nanorods that were stable towards
wide pH ranges, high salt concentration, gold etching conditions,
GSH, and high and low temperatures. Moreover, these materials
were effective for in vitro PTT. We attribute the unprecedented sta-
bility of these NHC-functionalized surfaces to the adatom addition
installation method, which avoids the need for reorganization of the
nanorod surface to enable optimal NHC binding. This work signifi-
cantly expands both the synthesis and applications of NHC surface
anchors. Moreover, we believe that the concept of adatom addition

Nature Chemistry | www.nature.com/naturechemistry

Articles NATure CHemisTry

a b c
Without GSH RT, 5 h RT, 5 h
+GSH, 0 h –78 °C, 5 h –78 °C, 5 h
Absorbance (a.u.) +GSH, 8 h 95 °C, 5 h 95 °C, 5 h

Absorbance (a.u.)

Absorbance (a.u.)
+GSH, 1 d
+GSH, 6 d

500 600 700 800 900 1,000 500 600 700 800 900 1,000 500 600 700 800 900 1,000
Wavelength (nm) Wavelength (nm) Wavelength (nm)

d 30 e Concentration (μg ml–1): 100 f

NS
NS
100
Δ Temperature (°C)

20 Absorbance (a.u.)

Cell viability (%)

80
50
60 *
10 40
25
20
***
0 0
0 25 50 100 500 600 700 800 900 1,000 Dose (μg): 0 3.8 3.8 7.5 7.5
Concentration (μg ml–1) +hν –hν +hν –hν +hν
Wavelength (nm)

Fig. 6 | Preliminary biological investigations using NHC@Au-I. a, UV–vis spectra for NHC@Au-I dissolved in 4 mM GSH, displaying the stability ofthe
nanorods for up to 6 days (longest time tested). b, UV–vis spectra for NHC@Au-I displaying the stability of the nanorods in water at high and low
temperatures. RT, room temperature. c, UV–vis spectra for PEG-S@Au showing the nanorods’ lack of thermal stability under the same conditions as
b. d, Change in water temperature after irradiation of NHC@Au-I with a near-infrared (830 nm) laser using a fluence of 6 W cm−2 for 8 min. Error bars
correspond to s.d. of three independent measurements. e, UV–vis spectra for NHC@Au-I at various concentrations before and after irradiation for three
cycles; no changes in the spectra are observed with each irradiation cycle, which suggests that the nanorods are stable in these conditions. f, MCF7 in
vitro cell viability studies as determined by CellTiter-Glo. Laser irradiation (dose of 0 μ​g ml−1) or exposure to 100 μ​l of 38 μ​g ml−1 or 75 μ​g ml−1 of NHC@Au-I
had no impact on cell viability. In contrast, laser irradiation of cells exposed to 100 μ​l of 38 μ​g ml−1 or 75 μ​g ml−1 NHC@Au-I induced significant cell killing.
Statistical significance was assessed by a two-tailed t-test (n = 3 independent samples). NS, not significant, *P <​ 0.05, ***P <​ 0.001.

6. Richter, C., Schaepe, K., Glorius, F. & Ravoo, B. J. Tailor-made N-heterocyclic

to install NHC ligands onto metallic surfaces may provide a general
route to highly stabilized interfaces.
Reporting Summary. Further information on experimental design
is available in the Nature Research Reporting Summary linked to
this article.
Data availability
All data supporting the findings of this study are available within the Article and
its Supplementary Information, and/or from the corresponding author upon
reasonable request.
Received: 3 April 2018; Accepted: 14 September 2018;
Published: xx xx xxxx
References
1. Zhukhovitskiy, A. V., MacLeod, M. J. & Johnson, J. A. Carbene ligands in
surface chemistry: from stabilization of discrete elemental allotropes to
modification of nanoscale and bulk substrates. Chem. Rev. 115, 11503–11532
(2015).
2. Hopkinson, M. N., Richter, C., Schedler, M. & Glorius, F. An overview of
N-heterocyclic carbenes. Nature 510, 485–496 (2014).
3. Ranganath, K. V. S., Kloesges, J., Schäfer, A. H. & Glorius, F. Asymmetric
nanocatalysis: N-heterocyclic carbenes as chiral modifiers of Fe3O4/Pd
nanoparticles. Angew. Chem. Int. Ed. 49, 7786–7789 (2010).
4. Ranganath, K. V. S., Schäfer, A. H. & Glorius, F. Comparison of
superparamagnetic Fe3O4-supported N-heterocyclic carbene-based catalysts
for enantioselective allylation. ChemCatChem 3, 1889–1891 (2011).
5. Serpell, C. J., Cookson, J., Thompson, A. L., Brown, C. M. & Beer, P. D.
Haloaurate and halopalladate imidazolium salts: structures, properties, and
use as precursors for catalytic metal nanoparticles. Dalton Trans. 42,
1385–1393 (2013).
carbenes for nanoparticle stabilization. Chem. Commun. 50, 3204–3207 (2014).
7. Ferry, A. et al. Negatively charged N-heterocyclic carbene-stabilized Pd
and Au nanoparticles and efficient catalysis in water. ACS Catal. 5,
5414–5420 (2015).
8. Rühling, A. et al. Modular bidentate hybrid NHC–thioether ligands for the
stabilization of palladium nanoparticles in various solvents. Angew. Chem. Int.
Ed. 55, 5856–5860 (2016).
9. Soulé, J.-F., Miyamura, H. & Kobayashi, S. Copolymer-incarcerated nickel
nanoparticles with N-heterocyclic carbene precursors as active cross-linking
agents for Corriu–Kumada–Tamao reaction. J. Am. Chem. Soc. 135,
10602–10605 (2013).
10. de los Bernardos, M. D., Pérez-Rodríguez, S., Gual, A., Claver, C. & Godard,
C. Facile synthesis of NHC-stabilized Ni nanoparticles and their catalytic
application in the Z-selective hydrogenation of alkynes. Chem. Commun. 53,
7894–7897 (2017).
11. Lara, P., Suárez, A., Collière, V., Philippot, K. & Chaudret, B. Platinum
N-heterocyclic carbene nanoparticles as new and effective catalysts for the
selective hydrogenation of nitroaromatics. ChemCatChem 6, 87–90 (2014).
12. Lara, P. et al. Ruthenium nanoparticles stabilized by N-heterocyclic carbenes:
ligand location and influence on reactivity. Angew. Chem. Int. Ed. 50,
12080–12084 (2011).
13. Gonzalez-Galvez, D. et al. NHC-stabilized ruthenium nanoparticles as
new catalysts for the hydrogenation of aromatics. Catal. Sci. Tech. 3,
99–105 (2013).
14. Lara, P. et al. NHC-stabilized Ru nanoparticles: synthesis and surface studies.
Nano-Structures Nano-Objects 6, 39–45 (2016).
15. Martínez-Prieto, L. M. et al. Long-chain NHC-stabilized RuNPs as versatile
catalysts for one-pot oxidation/hydrogenation reactions. Chem. Commun. 52,
4768–4771 (2016).
16. Ernst, J. B., Muratsugu, S., Wang, F., Tada, M. & Glorius, F. Tunable
heterogeneous catalysis: N-heterocyclic carbenes as ligands for supported
heterogeneous Ru/K-Al2O3 catalysts to tune reactivity and selectivity. J. Am.
Chem. Soc. 138, 10718–10721 (2016).

Nature Chemistry | www.nature.com/naturechemistry

NATure CHemisTry
17. Song, S. G. et al. N-Heterocyclic carbene-based conducting polymer–gold
nanoparticle hybrids and their catalytic application. Macromolecules 47,
6566–6571 (2014).
18. Crespo, J. et al. Ultrasmall NHC-coated gold nanoparticles obtained through
solvent free thermolysis of organometallic Au(i) complexes. Dalton Trans. 43,
15713–15718 (2014).
19. Cao, Z. et al. A molecular surface functionalization approach to tuning
nanoparticle electrocatalysts for carbon dioxide reduction. J. Am. Chem. Soc.
138, 8120–8125 (2016).
20. Ye, R. et al. Supported Au nanoparticles with N-heterocyclic carbene ligands
as active and stable heterogeneous catalysts for lactonization. J. Am. Chem.
Soc. 140, 4144–4149 (2018).
21. MacLeod, M. J. & Johnson, J. A. PEGylated N-heterocyclic carbene anchors
designed to stabilize gold nanoparticles in biologically relevant media. J. Am.
Chem. Soc. 137, 7974–7977 (2015).
22. Salorinne, K. et al. Water-soluble N-heterocyclic carbene-protected gold
nanoparticles: size-controlled synthesis, stability, and optical properties.
Angew. Chem. Int. Ed. 56, 6198–6202 (2017).
23. Zhukhovitskiy, A. V., Mavros, M. G., Van Voorhis, T. & Johnson, J. A.
Addressable carbene anchors for gold surfaces. J. Am. Chem. Soc. 135,
7418–7421 (2013).
24. Crudden, C. M. et al. Ultra stable self-assembled monolayers of
N-heterocyclic carbenes on gold. Nat. Chem. 6, 409–414 (2014).
25. Crudden, C. M. et al. Simple direct formation of self-assembled
N-heterocyclic carbene monolayers on gold and their application in
biosensing. Nat. Commun. 7, 12654 (2016).
26. Zhukhovitskiy, A. V. et al. Reactions of persistent carbenes with hydrogen-
terminated silicon surfaces. J. Am. Chem. Soc. 138, 8639–8652 (2016).
27. Mackey, M. A., Ali, M. R. K., Austin, L. A., Near, R. D. & El-Sayed, M. A.
The most effective gold nanorod size for plasmonic photothermal therapy:
theory and in vitro experiments. J. Phys. Chem. B 118, 1319–1326 (2014).
28. Grzelczak, M., Pérez-Juste, J., Mulvaney, P. & Liz-Marzán, L. M. Shape control
in gold nanoparticle synthesis. Chem. Soc. Rev. 37, 1783–1791 (2008).
29. Hurst, E. C., Wilson, K., Fairlamb, I. J. S. & Chechik, V. N-Heterocyclic
carbene coated metal nanoparticles. New J. Chem. 33, 1837–1840 (2009).
30. Möller, N. et al. Stabilization of high oxidation state upconversion
nanoparticles by N-heterocyclic carbenes. Angew. Chem. Int. Ed. 56,
4356–4360 (2017).
31. Vignolle, J. & Tilley, T. D. N-Heterocyclic carbene-stabilized gold
nanoparticles and their assembly into 3D superlattices. Chem. Commun.
2009, 7230–7232 (2009).
32. Ling, X., Roland, S. & Pileni, M. P. Supracrystals of N-heterocyclic
carbene-coated Au nanocrystals. Chem. Mater. 27, 414–423 (2015).
33. Bridonneau, N. et al. N-Heterocyclic carbene-stabilized gold nanoparticles
with tunable sizes. Dalton Trans. 47, 6850–6859 (2018).
34. Torrelles, X. et al. Solving the long-standing controversy of long-chain
alkanethiols surface structure on Au(111). J. Phys. Chem. C 122,
3893–3902 (2018).
35. Rodríguez-Castillo, M. et al. Reactivity of gold nanoparticles towards
N-heterocyclic carbenes. Dalton Trans. 43, 5978–5982 (2014).
36. Wang, G. Q. et al. Ballbot-type motion of N-heterocyclic carbenes on gold
surfaces. Nat. Chem. 9, 152–156 (2017).
37. Jiang, L. et al. N-Heterocyclic carbenes on close-packed coinage metal
surfaces: bis-carbene metal adatom bonding scheme of monolayer films on
Au, Ag and Cu. Chem. Sci. 8, 8301–8308 (2017).
38. Collado, A., Gomez-Suarez, A., Martin, A. R., Slawin, A. M. Z. & Nolan, S. P.
Straightforward synthesis of [Au(NHC)X] (NHC =​  N-heterocyclic carbene,
X =​ Cl, Br, I) complexes. Chem. Commun. 49, 5541–5543 (2013).
39. Fèvre, M. et al. Imidazolium hydrogen carbonates versus imidazolium
carboxylates as organic precatalysts for N-heterocyclic carbene catalyzed
reactions. J. Org. Chem. 77, 10135–10144 (2012).
40. Li, Z., Jin, R. C., Mirkin, C. A. & Letsinger, R. L. Multiple thiol-anchor capped
DNA-gold nanoparticle conjugates. Nucleic Acids Res. 30, 1558–1562 (2002).
41. Oh, E. et al. Colloidal stability of gold nanoparticles coated with multithiol-
poly(ethylene glycol) ligands: importance of structural constraints of the
sulfur anchoring groups. J. Phys. Chem. C 117, 18947–18956 (2013).
Nature Chemistry | www.nature.com/naturechemistry
Articles
42. Oh, E., Susumu, K., Goswami, R. & Mattoussi, H. One-phase synthesis of
water-soluble gold nanoparticles with control over size and surface
functionalities. Langmuir 26, 7604–7613 (2010).
43. Man, R. W. Y. et al. Ultrastable gold nanoparticles modified by bidentate
N-heterocyclic carbene ligands. J. Am. Chem. Soc. 140, 1576–1579 (2018).
44. Cao, Z. et al. Chelating N-heterocyclic carbene ligands enable tuning of
electrocatalytic CO2 reduction to formate and carbon monoxide: surface
organometallic chemistry. Angew. Chem., Int. Ed. 57, 4981–4985 (2018).
45. Pillar, V. N. R. Photo-removable protecting groups in organic-synthesis.
Synthesis 1980, 1–26 1980).
46. Dreaden, E. C., Alkilany, A. M., Huang, X., Murphy, C. J. & El-Sayed, M. A.
The golden age: gold nanoparticles for biomedicine. Chem. Soc. Rev. 41,
2740–2779 (2012).
47. Otsuka, H., Nagasaki, Y. & Kataoka, K. PEGylated nanoparticles for biological
and pharmaceutical applications. Adv. Drug Deliv. Rev. 64, 246–255 (2012).
48. Locatelli, E., Monaco, I. & Franchini, M. C. Surface modifications of gold
nanorods for applications in nanomedicine. RSC Adv. 5, 21681–21699 (2015).
49. Liu, H.-X., He, X. & Zhao, L. Metallamacrocycle-modified gold nanoparticles: a
new pathway for surface functionalization. Chem. Commun. 50, 971–974 (2014).
50. Schulz, F. et al. Effective PEGylation of gold nanorods. Nanoscale 8,
7296–7308 (2016).
51. Khlebtsov, N. & Dykman, L. Biodistribution and toxicity of engineered gold
nanoparticles: a review of in vitro and in vivo studies. Chem. Soc. Rev. 40,
1647–1671 (2011).
52. Tang, Q. & Jiang, D. E. Comprehensive view of the ligand gold interface from
first principles. Chem. Mater. 29, 6908–6915 (2017).
53. Schaefer, H. E. Investigation of thermal-equilibrium vacancies in metals by
positron-annihilation. Phys. Status Solidi A 102, 47–65 (1987).
54. Qi, W. H. & Wang, M. P. Vacancy formation energy of small particles. J.
Mater. Sci. 39, 2529–2530 (2004).
55. Huang, X. H., El-Sayed, I. H., Qian, W. & El-Sayed, M. A. Cancer cell
imaging and photothermal therapy in the near-infrared region by using gold
nanorods. J. Am. Chem. Soc. 128, 2115–2120 (2006).
56. Huang, X. H. & El-Sayed, M. A. Gold nanoparticles: optical properties and
implementations in cancer diagnosis and photothermal therapy. J. Adv. Res. 1,
13–28 (2010).
57. Yang, X., Yang, M. X., Pang, B., Vara, M. & Xia, Y. N. Gold nanomaterials at
work in biomedicine. Chem. Rev. 115, 10410–10488 (2015).
58. Doud, E. A. et al. In situ formation of N-heterocyclic carbene-bound
single-molecule junctions. J. Am. Chem. Soc. 140, 8944–8949 (2018).
Acknowledgements
The authors thank the National Science Foundation (CHE-1351646) for support of this
work.
Author contributions
M.J.M. and J.A.J. conceived the idea. M.J.M. conducted all synthesis and characterization
studies. A.J.G. conducted laser irradiation experiments. H.Y. and T.V.V. conducted
DFT calculations. H.V.-T.N. conducted cell culture assays. M.J.M. and J.A.J. wrote the
manuscript. All authors read and revised the manuscript.
Competing interests
The authors declare no competing interests.
Additional information
Supplementary information is available for this paper at https://doi.org/10.1038/
s41557-018-0159-8.
Reprints and permissions information is available at www.nature.com/reprints.
Correspondence and requests for materials should be addressed to J.A.J.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.
© The Author(s), under exclusive licence to Springer Nature Limited 2018
 nature research | reporting summary
Corresponding author(s): Jeremiah A. Johnson

Reporting Summary
Nature Research wishes to improve the reproducibility of the work that we publish. This form provides structure for consistency and transparency
in reporting. For further information on Nature Research policies, see Authors & Referees and the Editorial Policy Checklist.

Statistical parameters
When statistical analyses are reported, confirm that the following items are present in the relevant location (e.g. figure legend, table legend, main
text, or Methods section).
n/a Confirmed
The exact sample size (n) for each experimental group/condition, given as a discrete number and unit of measurement
An indication of whether measurements were taken from distinct samples or whether the same sample was measured repeatedly
The statistical test(s) used AND whether they are one- or two-sided
Only common tests should be described solely by name; describe more complex techniques in the Methods section.

A description of all covariates tested

A description of any assumptions or corrections, such as tests of normality and adjustment for multiple comparisons
A full description of the statistics including central tendency (e.g. means) or other basic estimates (e.g. regression coefficient) AND
variation (e.g. standard deviation) or associated estimates of uncertainty (e.g. confidence intervals)

For null hypothesis testing, the test statistic (e.g. F, t, r) with confidence intervals, effect sizes, degrees of freedom and P value noted
Give P values as exact values whenever suitable.

For Bayesian analysis, information on the choice of priors and Markov chain Monte Carlo settings
For hierarchical and complex designs, identification of the appropriate level for tests and full reporting of outcomes
Estimates of effect sizes (e.g. Cohen's d, Pearson's r), indicating how they were calculated

Clearly defined error bars

State explicitly what error bars represent (e.g. SD, SE, CI)

Our web collection on statistics for biologists may be useful.

Software and code

Policy information about availability of computer code
Data collection for TEM: Microscope User Interface, Tecnai (version 4.5.1 Build 8403) & Gatan Digital Micrograph, Gatan Microscopy Suite (version
2.11.1408.0); for DFT: Q-Chem (version 5.1); for UV-vis measurements: Cary WinUV (version 5.0); for cell viability (plate reader): i-Control
1.10

Data analysis for TEM: ImageJ64 (1.48v); for data plotting: Microsoft Excel (version 16.16.1) and Matlab (version 9.3); for NMR: Mnova (version 11.0)
For manuscripts utilizing custom algorithms or software that are central to the research but not yet described in published literature, software must be made available to editors/reviewers
upon request. We strongly encourage code deposition in a community repository (e.g. GitHub). See the Nature Research guidelines for submitting code & software for further information.

Data
April 2018

Policy information about availability of data

All manuscripts must include a data availability statement. This statement should provide the following information, where applicable:
- Accession codes, unique identifiers, or web links for publicly available datasets
- A list of figures that have associated raw data
- A description of any restrictions on data availability
All data is available from the corresponding author upon reasonable request

1
 nature research | reporting summary
Field-specific reporting
Please select the best fit for your research. If you are not sure, read the appropriate sections before making your selection.
Life sciences Behavioural & social sciences Ecological, evolutionary & environmental sciences
For a reference copy of the document with all sections, see nature.com/authors/policies/ReportingSummary-flat.pdf

Life sciences study design

All studies must disclose on these points even when the disclosure is negative.
Sample size 3 independent biological experiments were conducted, affording statistically significant means and standard deviations

Data exclusions There are no data exclusions.

Replication We have not attempted to replicate these cell viability studies

Randomization No in vivo studies were conducted. For in vitro studies, researcher A seeded cells, dosed cells, and determined cell viability; researcher B
irradiated the cells with test and control samples. Dosing was done in random rows in a fully seeded plate, and irradiation was done in
random rows of cells exposed or not exposed to nanorod samples.

Blinding No in vivo studies were conducted. For in vitro studies, experiments were double-blinded, as cells were blinded by default, and researcher A
was not informed of wells that were irradiated.

Reporting for specific materials, systems and methods

Materials & experimental systems Methods

n/a Involved in the study n/a Involved in the study
Unique biological materials ChIP-seq
Antibodies Flow cytometry
Eukaryotic cell lines MRI-based neuroimaging
Palaeontology
Animals and other organisms
Human research participants

April 2018