The label-free detection of biomolecules by means of fluorescence spectroscopy and imaging is topical. The developed surface-enhanced fluorescence technique has been applied to achieve progress in the label-free detection of biomolecules including deoxyribonucleic acid (DNA) bases. In this study, the effect of a strong enhancement of photoluminescence of 5'-deoxyadenosine-monophosphate (dAMP) by the plasmonic nanocavity metasurface composed of the silver femtosecond laser-induced periodic surface structure (LIPSS) and gold nanorods or nanospheres has been realized at room temperature. The highest value of 1220 for dAMP on the Ag-LIPSS/Au nanorod metasurface has been explained to be a result of the synergetic effect of the generation of hot spots near the sharp edges of LIPSS and Au nanorod tips together with the excitation of collective gap mode of the cavity due to strong near-field plasmonic coupling. A stronger plasmonic enhancement of the phosphorescence compared to the fluorescence is achieved due to a greater overlap of the phosphorescence spectrum with the surface plasmon spectral region. The photoluminescence imaging of dAMP on the metasurfaces shows a high intensity in the blue range. The comparison of Ag-LIPSS/Au nanorod and Ag-LIPSS/Au-nanosphere metasurfaces shows a considerably higher enhancement for the metasurface containing Au nanorods. Thus, the hybrid cavity metasurfaces containing metal LIPSS and nonspherical metal nanoparticles with sharp edges are promising for high-sensitive label-free detection and imaging of biomolecules at room temperature.
The label-free detection of biomolecules by means of fluorescence spectroscopy and imaging is topical. The developed surface-enhanced fluorescence technique has been applied to achieve progress in the label-free detection of biomolecules including deoxyribonucleic acid (DNA) bases. In this study, the effect of a strong enhancement of photoluminescence of 5'-deoxyadenosine-monophosphate (dAMP) by the plasmonic nanocavity metasurface composed of the silver femtosecond laser-induced periodic surface structure (LIPSS) and gold nanorods or nanospheres has been realized at room temperature. The highest value of 1220 for dAMP on the Ag-LIPSS/Au nanorod metasurface has been explained to be a result of the synergetic effect of the generation of hot spots near the sharp edges of LIPSS and Au nanorod tips together with the excitation of collective gap mode of the cavity due to strong near-field plasmonic coupling. A stronger plasmonic enhancement of the phosphorescence compared to the fluorescence is achieved due to a greater overlap of the phosphorescence spectrum with the surface plasmon spectral region. The photoluminescence imaging of dAMP on the metasurfaces shows a high intensity in the blue range. The comparison of Ag-LIPSS/Au nanorod and Ag-LIPSS/Au-nanosphere metasurfaces shows a considerably higher enhancement for the metasurface containing Au nanorods. Thus, the hybrid cavity metasurfaces containing metalLIPSS and nonspherical metal nanoparticles with sharp edges are promising for high-sensitive label-free detection and imaging of biomolecules at room temperature.
The development of various methods, techniques,
and their combinations
for high-sensitive detection and imaging of biomolecules in different
biological systems at room temperature remains a topical objective
for modern research. Fluorescence spectroscopy is one of the most
sensitive spectroscopic techniques used for studying physical processes
in living systems. It is even capable of identifying and imaging single
molecules.[1−3] High-sensitive fluorescence spectroscopy is successfully
applied for the determination of deoxyribonucleic acid (DNA) sequencing,
which is important for the study of physical and chemical processes
occurring in living systems, in particular, genetic transcription/translation
processes, etc.[4,5] However, it should be noted that
a high quantum yield of intrinsic fluorescence and phosphorescence
of main nucleotides is achieved only at low temperatures, while at
room temperature, it is of 10–5–10–4.[6−10] The quantum yields of fluorescence of ∼0.4 and phosphorescence
of ∼0.6 were reported for 5′-adenosine-monophosphate
(AMP) at low temperatures.[11] Meanwhile,
at room temperature, the quantum yield for 5′-deoxyadenosine-monophosphate
(dAMP) is known to be 7 × 10–5.[6,7] Thereby, the task to enhance the fluorescence of nucleic acids and
nucleotides at higher temperatures is of utmost relevance. To achieve
an effective visualization of DNA sequencing and molecular and cellular
bioimaging over different optical windows, additional labeling techniques
of molecules with specific fluorescent dyes are frequently used.[12−14] Nonetheless, this method has some disadvantages concerning its labor-
and time-consuming character. Therefore, the goal of label-free detection
of biomolecules with fluorescence spectroscopy and imaging is highly
possible.Some approaches have been elaborated to achieve an
increase in
the total quantum yield of fluorophores, namely, by its placing in
close vicinity with metal nanostructures.[15,16] The nanoscale metal structures could increase the excitation efficiency
due to the enhancement of electromagnetic (EM) field near the metal
surface caused by the excitation of surface plasmon resonance (SPR).
It has been shown that SPR excitation provides an increase in the
radiative emission rate and, as a consequence, quantum yield of fluorophores.[17] In addition, metal nanostructures characterized
by the high directivity of radiation may cause an increase in collection
efficiency.[18] Therefore, the developed
surface-enhanced fluorescence (SEF) technique[19] has been applied for the label-free detection of biomolecules including
DNA bases.[20−23] However, most investigated metal nanostructures have been quite
inhomogeneous. Respectively, certain imperfections of the obtained
outcomes, namely, weak broad SPRs resulting in moderate signal enhancements,
have been observed.Some new methods have been applied to achieve
a strong enhancement
of biomolecule fluorescence: (i) metal nanostructures with densely
packed sharp tips,[24−27] (ii) periodically ordered metal nanostructures,[28−32] and (iii) plasmonic cavities.[33,34] All of these nanostructures have a similar feature, that is, the
formation of so-called “hot spots”, where the very strong
EM field occurs. For this reason, plasmonic cavities have a very important
place in this matter. In plasmonic cavities, hot spots emerge in a
very thin gap between the components of such a cavity. This phenomenon
originates from the SPR near-field coupling between the produced nanofeatures.
The EM field enhancement supports the processes, which in turn stimulates
SEF and surface-enhanced Raman scattering (SERS) processes.[35] The design of these cavities can be realized
in different variants such as metal NPs on metal substrate–mirror
(NPOM structures), nanopatch optical antennas, metal–insulator–metal
resonators, etc. In particular, in NPOM structures, metal NPs are
located in close vicinity to metal substrate–mirror. NPOM structures
with metal NPs support a new collective plasmonic mode, the so-called
gap mode, emerging due to the hybridization of the plasmonic mode
of metal NP layer and the propagating surface plasmon polariton (SPP)
of the metal substrate.[16,33,36−40] At the excitation of the plasmonic gap mode, the EM field is squeezed
into a very thin gap between the NP layer and the metal film, causing
a strong field enhancement in the nanogap, i.e., the generation of
hot spots. Moreover, the cavities containing nonspherical NPs with
sharp edges should provide a higher signal enhancement for some processes
than the cavities with spherical NPs.Along with the mentioned
structures, a rather high plasmonic enhancement
has been achieved in the structures with gratings on metal surfaces.
In particular, laser-induced periodic surface structures (LIPSS) on
metal substrate represent such a type of plasmonic metasurface.[29−32] The periodic surface structures are formed by the process of interference
between the powerful incident laser plane wave and surface wave. For
metals, this surface wave is determined by the grating-assisted excitation
of SPP. It should be noted that the formed periodic surfaces also
contain some nanofeatures and quasi-gratings of nanoscale dimensions.[41] Nowadays, LIPSS is actively used for the fabrication
of SERS/SEF substrates and plasmonic sensors.[42−47] The application of LIPSS for the formation of plasmonic cavity may
be promising due to the ability of biomolecules to bind to the edges
of nanostructures, which leads to a stronger plasmonic enhancement.[15,41,48,49]Recently, hybrid metal–dielectric micro- and nanocavities
are intensively investigated. The structures of such metal–dielectric
cavities possess the advantages of both metal and dielectric materials.
Indeed, metal NPs highly support subwavelength plasmons with a strong
local EM field; however, they are characterized by relatively high
intrinsic material losses.[50−52] Meanwhile, dielectric structures
support exquisite low-loss versatility, but only moderate confinement.[53−55] Recent attempts toward fabricating metal–dielectric hybrids
were made to combine these two advantages, eliminating the mentioned
limitations.[56−59] It has been shown that the hybrid approach can simultaneously exhibit
a unique combination of plasmonic strong EM field and high confinement
with dielectric small dissipative losses. In such hybrid cavity structures,
the hybridized dielectric–metal resonances arise, leading to
strong and highly tunable optical response, which thereby makes them
suitable for different photonics applications. In the far field, they
exhibit scattering stronger than that for all-metal and all-dielectric
structures. In the near field, they provide high spontaneous emission
enhancement and highly directional emission at visible and infrared
wavelengths. The hybrid cavities of two types have been studied. The
first one consists of the high-refractive-index (HRI) dielectric NPs
located closely to metal plasmonic substrate (metal–dielectric
NPOM structures).[56−59] Typically, the structures consisting of HRI dielectric (rare earth
oxides and nitrides) or semiconductor (Si, Ge, GaP) NPs on the noble-metal
(Ag, Au) substrates have been elaborated. The structures with nonspherical
HRI NPs, e.g., rare-earth oxide (Nd2O3)[56] and semiconductor (Si)[57] nanorods (NRs), were reported to be most promising for surface-enhanced
spectroscopy and sensing applications. The hybrids of the second type
contain dielectric photonic crystals (PCs) combined with plasmonic
metal NPs.[60−62] Metal NPs are used to increase the light-matter coupling
in the PCs. The hybridization of localized SPRs of metal NPs and Bragg
diffraction resonances of dielectric PC leads to a strong and tunable
optical response of such structures, which allows us to use them for
different applications such as surface-enhanced spectroscopy,[63−65] sensing,[66] catalysis,[67,68] nonlinear optics,[69] etc.In this
study, we aim to develop the technique of label-free detection
of biomolecule at room temperature with high sensitivity and reproducibility.
Nucleotide dAMP was chosen for our research due to its high photostability
compared to other nucleotides.[70] We expected
to obtain an effect of strong enhancement of photoluminescence (PL)
(fluorescence and phosphorescence) of dAMP precipitated on plasmonic
metasurface containing Ag LIPSS and Au nanorods (NRs) or spherical
Au NPs for comparison. These materials have been used because (i)
Ag supports the sharpest and most intense SPR in the visible and near-UV
spectral ranges and (ii) SPR of Au NRs can be tuned over the visible
and near-IR ranges by changing the NR aspect ratio. Thus, several
types of metasurfaces have been studied: (1) Ag smooth substrate,
(2) Ag LIPSS, (3) Ag smooth/Au NRs, (4) Ag LIPSS/Au NRs, (5) Ag smooth/Au
NPs, and (6) Ag LIPSS/Au NPs (schematically depicted in Figure ). The metasurfaces with both
types of NPs are the plasmonic cavity nanostructures, where the near-field
coupling of the localized SP mode of the layer of Au NRs or NPs and
the propagating SPP mode of Ag surface arises. A quantitative comparative
study of the quantum yield of dAMP PL on these metasurfaces was carried
out, and the results are interpreted and discussed.
Figure 1
Schemes of the studied
structures with dAMP molecules precipitated
on: (a) Ag smooth surface, (b) Ag LIPSS, (c) Ag smooth/Au NRs, (d)
Ag LIPSS/Au NRs, (e) Ag smooth/Au NPs, and (f) Ag LIPSS/Au NPs. The
schemes include the models depicting the formation of plasmonic hot
spots and excitation of SPPs in the systems with (a) and (b) designs,
and the excitation of plasmonic gap mode in the cavity systems with
(c)–(f) designs. The inset to (a) presents the chemical structure
of the dAMP molecule.
Schemes of the studied
structures with dAMP molecules precipitated
on: (a) Ag smooth surface, (b) Ag LIPSS, (c) Ag smooth/Au NRs, (d)
Ag LIPSS/Au NRs, (e) Ag smooth/Au NPs, and (f) Ag LIPSS/Au NPs. The
schemes include the models depicting the formation of plasmonic hot
spots and excitation of SPPs in the systems with (a) and (b) designs,
and the excitation of plasmonic gap mode in the cavity systems with
(c)–(f) designs. The inset to (a) presents the chemical structure
of the dAMP molecule.
Results and Discussion
Structural
Characterization
The procedures of the fabrication
of Au NRs, Au NPs, Ag substrates, and metasurfaces are given in the Experimental Section, which also presents the features
of optical measurements.The shape and size of the fabricated
Au NRs were analyzed by transmission electron microscopy (TEM), as
presented in Figure a. The fabricated Au NRs have a perfect rod shape with an average
length and diameter of 81 and 14 nm, respectively (aspect ratio is
5.8). The electric field simulation for Au NRs was finite-difference
time-domain (FDTD) simulation (Lumerical Solutions).[71] A mesh size of 0.5 nm in the NR region was used. The size
of Au NR was set according to Figure b; the dielectric function of Au NR was taken from
Au-CRC in the material library of the software. A perfectly matched
layer was used as the boundary conditions in the FDTD simulation.
The calculated electric field distribution is presented in Figure c. The fabricated
Au NPs have a perfect spherical shape with an average diameter of
13 nm, as confirmed by the TEM images presented in Figure d.
Figure 2
(a) TEM image of Au NRs,
(b) the representative TEM image with
a near-average size, and (c) the calculated electric field distribution
in its vicinity. (d) TEM image of Au NPs.
(a) TEM image of Au NRs,
(b) the representative TEM image with
a near-average size, and (c) the calculated electric field distribution
in its vicinity. (d) TEM image of Au NPs.The Ag LIPSS was fabricated by texturing the polished Ag thin plate
with ultrashort laser pulses provided by an amplified Ti:sapphire
femtosecond laser system. The morphology of the ultrashort laser-treated
Ag surfaces, LIPSS, was analyzed using a scanning electron microscope
(SEM). The SEM images in Figure a reveal a submicron LIPSS and a set of nanoscale features
on the Ag surface. LIPSS consists of the surface ripples of two characteristic
periods in perpendicular directions: L1 = 0.62 μm and L2 = 2.09 μm.
Figure 3
(a) SEM
image of Ag LIPSS, where L1 and L2 are the characteristic periods.
(b) SEM image of Ag LIPSS/Au NRs metasurface, where the arrows point
to Au NRs, which are mainly located in the LIPSS hollows.
(a) SEM
image of Ag LIPSS, where L1 and L2 are the characteristic periods.
(b) SEM image of Ag LIPSS/Au NRs metasurface, where the arrows point
to Au NRs, which are mainly located in the LIPSS hollows.The next stage of sample preparation was the formation of
plasmonic
cavity metasurface of Ag smooth/Au NRs (Figure c) and Ag LIPSS/Au NRs (Figure d). These cavity metasurfaces
were prepared by drop-coating of the Au NR colloid on each of the
substrates. It was previously shown[29,32] that the drop-coating
method could provide a uniform spread of molecules and NPs over the
substrate. In that way, Au NRs are distributed over the substrate
surface, forming a monolayer. The SEM morphology of the Ag LIPSS/Au
NRs metasurface has confirmed the presence of Au NRs on the substrate
(Figure b). Overall,
Au NRs are spread quite evenly on the Ag surface with a density of
about 0.8 μm–2. It is seen that Au NRs are
mainly located in the LIPSS hollows. The same procedure was used for
the fabrication of Ag smooth/Au NPs (Figure e) and Ag LIPSS/Au NPs (Figure f) metasurfaces containing
spherical Au NPs.Then, the method of drop-coating was used
for the precipitation
of dAMP solution on the real">spective dried metasurface substrate. In Figure , we schematically
depicted the structure of obtained metasurfaces along with the physical
mechanisms explaining dAMP PL enhancement by such metasurfaces.
Surface Plasmon Modes of Metasurfaces: Absorption and Reflection
Experiments
The absorption spectra of the water colloid of
bare Au NRs and mixture colloid of Au NRs with dAMP molecules are
presented in Figure in comparison to that of spherical Au NPs. The spectrum of Au NRs,
with an average length of 81 nm and a diameter of 14 nm, contains
two peaks at 517 and 935 nm corresponding to the transversal (T) and
longitudinal (L) localized surface plasmon (LSP) modes of NR. In contrast,
spherical Au NPs, with an average diameter of 13 nm, show the LSP
mode peaked at 520 nm. The absorption band at 260 nm is assigned to
dAMP.
Figure 4
Absorption spectra of the colloidal aqueous solution of Au NRs
and NPs as well as the mixed solution of Au NRs and NPs with dAMP.
NR:L and NR:T denote the absorption peaks corresponding to longitudinal
and transverse LSP modes of Au NRs; NP denotes the absorption peak
of LSP mode of Au NPs. The absorption spectra of the dAMP solution
is presented to show the absorption peak of dAMP molecules.
Absorption spectra of the colloidal aqueous solution of Au NRs
and NPs as well as the mixed solution of Au NRs and NPs with dAMP.
NR:L and NR:T denote the absorption peaks correal">sponding to longitudinal
and transverse LSP modes of Au NRs; NP denotes the absorption peak
of LSP mode of Au NPs. The absorption spectra of the dAMP solution
is presented to show the absorption peak of dAMP molecules.
The nonradiative character of propagating SPP modes
on the smooth
metal surfaces should be emphasized. Indeed, the excitation of such
surface polaritons requires the fulfillment of matching conditions
for photon and surface polariton wave vectorswhere k is
the module of
the SP polariton and q is the incident photon wave vector projection on the metal surface.
Herewhere ε and ε′
are the
real parts of permittivities of the metal and dielectric surrounding
medium, respectively, φ is the incidence angle, ω is the
frequency of incident light, and c is the light velocity.
Since the surface EM waves are excited, when the condition ε
≤ −ε′ is fulfilled, we obtain thatTherefore,
in the case of a smooth metal surface, k > q, i.e.,
condition (1) cannot be fulfilled and, respectively, the SP polaritons
cannot be excited on the smooth surface by the plane light wave. However,
in the case of nonflat metal surface with high curvature or in the
case of incident light wave with high curvature wavefront, the projection
of incident photon wave vector on the metal surface is effectively
increased by Δq: . This makes the fulfillment of
condition
(1) possible. In the case of periodic surface structurewhere d is the structure
period and m is a natural number. In the case of
the rough surfacewhere ⟨d⟩ is
the mean size of the surface texture features. Thus, in the case of
quasi-periodic submicron and nanosized surface texture, the grating
excitation of SPPs takes place (schematically shown in Figure b). Meanwhile, on the smooth
Ag substrate, the conditions of excitation of SPP modes are fulfilled
due to the waves with large curvature wavefront. These waves arise
due to the scattering of incident plane wave by Au NRs or NPs dispersed
in the proximate vicinity of metal surface (Figure c,e).In the case of Ag LIPSS/Au NRs
or Ag LIPSS/Au NPs, depicted in Figure d,f, respectively,
both the above-mentioned mechanisms take part under SPP mode excitation.
LIPSS consists of submicron surface ripples in the perpendicular direction.
Nanoscale features, namely, tips, nanocones, and other nanoprotrusions,
are observed on the tops and slopes of the hills and on the slopes
of the hollows. Thus, the analysis of the morphology of Ag LIPSS allows
us to assume the role of the observed features in the interaction
of such a modified surface and external excitation. LIPSS supports
a rather efficient radiative excitation of the propagating SPP modes
on its surface. Moreover, the observed nanoscale structures form additional
plasmonic hot spots on ripples that will contribute to some physical
processes.The spectral manifestation of the excitation of propagating
plasmon
modes on the studied structures has been demonstrated by means of
reflection spectroscopy (Figure ). The reflection spectra were measured for P-polarized
light, since in this case, the vector of electric field of incident
beam lies in the plane formed perpendicular to the metal surface and
in the direction of SP polariton propagation. Then, the electric field
vector has both components perpendicular to the metal surface and
parallel to the direction of SP polariton propagation, necessary for
SP polariton excitation. For S-polarized light, the electric field
vector is perpendicular to this plane, which makes the SP polariton
excitation impossible. The distinct dips in the reflection spectra
give evidence confirming the excitation of plasmonic propagating modes
on Ag LIPSS, Ag smooth/Au NRs, Ag LIPSS/Au NRs, and metasurfaces with
Au NPs.
Figure 5
Reflection spectra of Ag LIPSS and plasmonic cavity metasurface
Ag smooth/Au NRs and Ag LIPSS/Au NRs. Reflection spectra of Ag smooth/Au
NPs and Ag LIPSS/Au NPs are presented for comparison. The spectra
were measured for P-polarized light.
Reflection spectra of Ag LIPSS and plasmonic cavity metasurface
Ag smooth/Au NRs and Ag LIPSS/Au NRs. Reflection spectra of Ag smooth/Au
NPs and Ag LIPSS/Au NPs are presented for comparison. The spectra
were measured for P-polarized light.The analysis of the reflection spectra (Figure ) shows that the dip depths are the largest
for the metasurfaces containing both LIPSS and layer of Au NRs or
NPs, intermediate for bare Ag LIPSS, and the smallest for the hybrid
metasurface based on smooth Ag surface and Au NRs or NPs. On the one
hand, these data argue for the evidence of considerably higher efficiency
of propagating plasmonic mode excitation for the structures that combines
laser-treated surfaces and Au NRs or NPs, compared to the close levels
of efficiencies for the metasurfaces with smooth Ag surface. The specified
fact is caused by the synergetic effect of plasmonic propagating mode
excitation by the quasi-grating on LIPSS surface and by waves scattered
by Au NRs or NPs. On the other hand, the propagating mode excitation
efficiency is higher for the metasurface containing nonspherical NRs
than spherical NPs. This indicates that nonspherical particles are
better for efficient cavity mode excitation than the spherical ones.
As we have mentioned above, such plasmonic cavity mode is a plasmonic
collective gap mode of the array of Au NRs or NPs and Ag substrate.
Thus, a high-intensity plasmon gap mode is excited due to the plasmonic
coupling in the system of Ag LIPSS Au NRs, as shown in Figure d. The space between the Au
NRs and the LIPSS surface can be approximately evaluated by the thickness
of the capping agent layer on Au NRs. Hereby, the gap between NRs
and LIPSS of around 1 nm can provide a strong EM field that forms
an extended hot spot in the gap. These hot spots determine the enhancement
phenomena for the dAMP molecules discussed below.
Enhancement
of dAMP Photoluminescence by Metasurfaces
The PL spectra
of the dAMP molecules deposited on a polished quartz
plate and on different metal metasurfaces under study are presented
in Figure . The high-energy
band peaked at 335 nm, marked as F, corresponds to the UV fluorescence
of the dAMP molecules. The complex two-component band with maxima
at 410 and 433 nm in the blue region, marked as P, is assigned to
the dAMP phosphorescence.[11] To confirm
the nature of these PL bands, the PL excitation spectra at wavelengths
of 335, 410, and 433 nm for the dAMP/Ag LIPSS/Au NRs structure have
been measured (Figure ). All of the excitation spectra contain the band centered at 260
nm. This specific feature confirms that the obtained emission spectra
belong to the PL of dAMP molecules.
Figure 6
PL spectra of dAMP molecules deposited
on different substrates:
quartz plate, Ag smooth substrate, Ag LIPSS, plasmonic cavity metasurfaces
of Ag smooth/Au NRs, and Ag LIPSS/Au NRs and the respective PL imaging
in the blue range. The PL spectra of dAMP on Ag smooth/Au NPs and
Ag LIPSS/Au NPs are presented for comparison. PL was measured at room
temperature.
Figure 7
PL excitation spectra at wavelengths of 335
nm (F peak) and 410
and 433 nm (two peaks of P band) for dAMP on the Ag LIPSS/Au NR metasurface
measured at room temperature.
PL spectra of dAMP molecules deposited
on different substrates:
quartz plate, Ag smooth substrate, Ag LIPSS, plasmonic cavity metasurfaces
of Ag smooth/Au NRs, and Ag LIPSS/Au NRs and the respective PL imaging
in the blue range. The PL spectra of dAMP on Ag smooth/Au NPs and
Ag LIPSS/Au NPs are presented for comparison. PL was measured at room
temperature.PL excitation spectra at wavelengths of 335
nm (F peak) and 410
and 433 nm (two peaks of P band) for dAMP on the Ag LIPSS/Au NR metasurface
measured at room temperature.One can see that the PL of dAMP molecules deposited on all of the
studied Ag metasurfaces reveals a strong enhancement since the PL
intensity of the dAMP deposited on a quartz plate (black curve in Figure ) is fairly low.
Note that only the fluorescence part of dAMP on quartz emission is
observed and the intensity of dAMP phosphorescence in the blue range
is too weak to be reliably measured. Meanwhile, the visible PL microscopy
imaging of dAMP on the studied metasurfaces shows a high intensity
in the blue range (Figure ), while the PL of dAMP on quartz is undetectable. As it is
known from the previous research,[6,7] the quantum
yield of dAMP PL is 7 × 10–5 at room temperature.
It is well known[72−74] that the nature of low dAMP phosphorescence quantum
yield is the capture of the majority of triplet excitations generated
in the dAMP molecule by the surrounding oxygen molecules possessing
the ground triplet state.Now we consider possible physical
mechanisms explaining the PL
enhancement of dAMP molecules for all of the studied metasurfaces.
The observed PL enhancement for the dAMP molecules deposited on the
smooth Ag surface (green curve in Figure ) and Ag LIPSS (dark green curve in Figure ) is caused by the
local EM field enhancement near the corresponding surface, arising
due to the SPP excitation at this surface. At the smooth Ag surface,
SPPs are generated by near-field waves emitted by dAMP molecules.
Meanwhile, an additional mechanism of the SPP excitation at LIPSS
occurs due to its direct excitation on the formed diffraction quasi-gratings,
as shown in Figure . It is well known that the direct SPP excitation on metal gratings
becomes possible due to the fulfillment of the matching condition
for exciting photon and SPP. In the case of a smooth metal surface,
the matching condition cannot be implemented, as SPPs are nonradiative
modes for this case. Naturally, a higher PL enhancement for dAMP deposited
on the Ag LIPSS than on the smooth surface is experimentally observed
in Figure . Moreover,
the phosphorescence intensity is sufficiently higher for dAMP deposited
on the Ag LIPSS than on the smooth Ag surface, while the fluorescence
intensities differ slightly for these surfaces. The most probable
reason for a stronger plasmonic enhancement of the phosphorescence
than the fluorescence is the closer location of the phosphorescence
spectrum to the SPP spectral region (see Figure ). Indeed, the noble-metal (silver and gold)
structures are characterized by surface plasmonic response in the
visible spectral region, where the dAMP phosphorescence is spectrally
located (380–480 nm). Meanwhile, the dAMP fluorescence is located
in the UV region at 280–370 nm, where there is no surface plasmonic
response of noble-metal structures. At the same time, a slight difference
of the level of fluorescence enhancement for dAMP on Ag LIPSS and
smooth Ag surface is explained by a large spectral distance between
the fluorescence and SPP spectra of Ag surfaces.The PL spectra
of dAMP molecules deposited on different metasurfaces
containing Au NRs also reveal a strong PL enhancement (Figure ). Au NRs in the metasurface
structure stimulate the process of the plasmonic near-field coupling
of the Au NR LSP modes with the propagating SPPs of Ag surface (smooth
or LIPSS) that results in the excitation of collective SP mode or
plasmonic gap mode.[33,36−40] Consequently, the plasmonic gap mode generates a
very intense EM field in the cavity, where the dAMP molecules are
located, that causes a stronger enhancement compared to the metasurfaces
without Au NRs. Such an effect is observed experimentally (see Figure and Table ). At the same time, the enhancement
in the structures with spherical NPs was a few times lower, which
indicates the central role of the strong electric field in hot spots
near the tips of nonspherical NRs in the plasmonic enhancement phenomenon
(the calculated electric field near the tips of an NR is presented
in Figure c).[75] Note that the PL spectra of the dAMP deposited
on the bare metasurfaces and the other two containing Au NRs or NPs
demonstrate a significant difference. Namely, in the presence of Au
NRs or NPs, the fluorescence is much more intense than the phosphorescence
(Figure ).
Table 1
Enhancement Factors of the Fluorescence
and Total PL (Fluorescence + Phosphorescence) of the dAMP Molecules
Deposited on Different Substrates
quartz
Ag smooth
Ag LIPSS
Ag smooth/Au
NRs
Ag LIPSS/Au NRs
Ag smooth/Au NPs
Ag LIPSS/Au
NPs
fluorescence
1
108
143
850
1050
154
366
PL
1
295
808
938
1220
296
430
To conclude about the
effect of the incorporation of specific nanoscale
components onto the metasurfaces and to evaluate the levels of enhancement
factors of spectroscopic characteristics, we quantitatively analyze
the PL enhancement of dAMP on the studied metasurfaces. An enhancement
factor of PL is determined as the ratio of the total intensity of
the PL spectrum of the dAMP molecules deposited on a metasurface to
the one on the quartz plate. Let us note that because of the transparency
of quartz, the emission distribution pattern of emitters on quartz
can be significantly different from those on the metal metasurfaces.[76] This would lead to different spectrofluorophotometer
collection efficiencies that can cause bias in the quantum efficiency
measurement. Thus, the dAMP PL enhancement factors obtained in our
work are not the enhancement factors of the quantum yield of the dAMP
molecules. Those are the values of PL enhancement of the dAMP species
provided by different metal metasurface substrates compared to the
common nonluminescent dielectric substrate (quartz), which can be
obtained directly by a standard spectrofluorophotometer.The
obtained enhancement factors for total PL (fluorescence + phosphorescence)
and for fluorescence separately are given in Table . The PL enhancement factor of dAMP deposited
on the studied metasurfaces ranges from 295 for dAMP deposited on
smooth Ag surface to 1220 on Ag LIPSS/Au NRs metasurface. It is seen
that the enhancement factor values of fluorescence fit in the range
of 108 for dAMP deposited on the smooth Ag surface to 1050 on the
Ag LIPSS/Au NRs metasurface. The presented data elucidate the crucial
role of the plasmonic gap mode and, accordingly, the plasmonic coupling
in the enhancement of fluorescence. Moreover, a higher enhancement
for the dAMP on the Ag LIPSS/Au NR metasurface in comparison to the
dAMP on the Ag smooth/Au NR one should be emphasized. The nature of
this difference is caused by a higher efficiency of gap mode excitation
and a higher concentration of hot spots formed at the metasurface
based on LIPSS.Finally, it should be accentuated that the highest
value of the
dAMP PL enhancement factor of 1220 achieved in the plasmonic cavity
metasurface is due to the two main components: LIPSS and NRs. The
indicated value of the PL enhancement factor of 1220 is 15 times higher
than the maximum value of 80 previously known from the literature
for Al nanoparticle array.[29] At the same
time, it is similar by value to the 1120-fold PL enhancement for the
dAMP molecules obtained by us hybridizing Ag LIPSS with Ag nanoprisms.[41] The key factors providing these results are
the following: (i) the highest plasmonic near-field coupling for the
LIPSS substrate together with the consequent formation of plasmonic
gap mode and (ii) the highest concentration of hot spots on the textured
metal surface and at the tips of nonspherical NPs.
Conclusions
The levels of plasmonic enhancement for the PL of dAMP have been
analyzed depending on the plasmonic nanocavity metasurface structures,
namely, Ag smooth substrate and LIPSS combined with Au NRs or NPs.
The deposition of Au NRs on different Ag substrates forms the plasmonic
cavity nanosystems with strong near-field coupling of the LSP modes
of Au NRs and the propagating SPP modes of Ag surface (smooth or laser
structured). The obtained values of the PL enhancement factor of dAMP
deposited on the studied metasurfaces fell in the range of 295–1220
depending on the metasurface type. The highest 1220-fold enhancement
has been achieved for the Ag LIPSS/Au NR metasurface. This sufficient
enhancement is shown to arise due to the synergetic effect of the
generation of a high concentration of the hot spots near the sharp
edges of the LIPSS ripples and the tips of NRs, together with the
excitation of the collective gap mode due to the strong plasmonic
near-field coupling in the formed plasmonic cavity. The phosphorescence
intensity is essentially higher for dAMP on the Ag LIPSS than on the
smooth Ag surface, whereas the fluorescence intensities differ slightly
for these Ag metasurfaces. This is explained by a greater overlap
of the phosphorescence spectrum in the SPP spectral region. The visible
PL microscopy imaging of dAMP on the studied metasurfaces shows high
intensity in the blue range, while the PL of dAMP on quartz is undetectable.Thus, we have demonstrated that the combination of laser-textured
metal surface (i.e., LIPSS) and nonspherical metal NPs with sharp
tips (NRs, in particular) forms the plasmonic cavity metasurfaces
promising for rather simple high-sensitive detection and imaging of
biomolecules in ambient air at room temperature. It should be emphasized
that in the case of the presented plasmonic cavity metasurfaces, there
is no need to use any auxiliary dye labels for nucleotide PL imaging.
These findings will be very useful for the development of plasmonic
metasurfaces having the possibility to control the enhancement of
separate PL components by changing the form and material of metal
substrate and NPs. Also, in light of obtained results and recent trends
in the elaboration of hybrid metal–dielectric nanocavity structures,
the next challenge in the future is the design and fabrication of
the metasurfaces consisting of metalLIPSS and HRI dielectric nonspherical
NPs or dielectric photonic crystals for surface-enhanced spectroscopy
applications, high-sensitive detection and imaging of biomolecules
in particular.
Experimental Section
Au Nanorods
The
used Au NRs were synthesized according
to the following procedure.[77]
Seed Solution
Preparation
HAuCl4 (5 mL,
0.5 mM) was mixed with the cationic surfactant hexadecyltrimethylammonium
bromide (CTAB; 5 mL, 0.2 M) solution. Then, 1 mL of fresh 6 mM NaBH4 was injected into the as-prepared solution under vigorous
stirring. After 2 min stirring, the seed solution was aged at room
temperature and under weak room light illumination for 30 min.
Growth
Solution Preparation
CTAB (9.0 g) and 5-bromosalicylic
acid (1.1 g) were dissolved in 250 mL of warm water. After the resulting
solution was cooled to 30 °C, 4 mM AgNO3 solution
was added. The mixture was kept undisturbed for 15 min. Then, 250
mL of 1 mM HAuCl4 solution and 37 wt % HCl were added.
After 15 min of slow stirring, 0.064 M ascorbic acid was added and
the solution was vigorously stirred to colorless for 30 s.Finally,
0.4 mL of the seed solution was injected into the as-prepared growth
solution. The resultant mixture was stirred for 30 s and left undisturbed
for 12 h. Au NRs were collected by centrifugation at 9000 rpm for
20 min, followed by the removal of the supernatant. The precipitates
were redispersed in water to obtain the resultant NR solution with
a concentration of about 0.05 mg/mL.
Au Spherical Nanoparticles
Under vigorous stirring,
5 mL of 1% sodium citrate was added into 100 mL of 0.3 mM HAuCl4 at 100 °C. After 30 min stirring, the mixture was cooled
to room temperature. Au NPs were collected using centrifugation at
13 000 rpm for 20 min, followed by the removal of the supernatant.
The precipitates were redispersed in water to obtain the resultant
NP solution with a concentration of about 0.05 mg/mL.
Ag LIPSS Substrate
A pure Ag thin plate of size 1.5
× 2.5 cm, polished mechanically to the residual roughness less
than 150 nm, was used for the texturing with ultrashort laser pulses
provided by an amplified Ti:sapphire femtosecond laser system (Coherent,
Inc.). The main characteristic parameters of high-intensity femtosecond
laser pulses were as follows: a central wavelength of 800 nm, the
pulse duration of horizontally polarized beam from the amplifier of
140 fs, and the pulse energy of about 0.77 mJ at a repetition rate
of 1 kHz. The experimental setup also included a quartz lens with
a 100.0 mm focal length, a vertically standing sample stage, an electromechanical
shutter to vary the number of laser shots applied to the sample. During
the laser treatment, the sample holder moved at a velocity of 1 mm/s.
The diameter of the laser beam spot on the sample surface was 8.5
× 10–2 cm. The power density of the incident
radiation varied by changing the distance between the focusing lens
and the sample. The laser processing was performed in air under atmospheric
pressure.
Ag LIPSS/Au NR Metasurface
These cavity metasurfaces
were prepared by drop-coating 3 μL of the Au NR colloid on each
of the substrates. The solution spread into a circular area of about
∼5 mm in diameter. We used the Au NR and NP solutions with
concentrations of 0.05 mg/mL, providing approximately the same optical
density in the SP spectral range of absorption spectra.A dAMP
stock solution of 1.7 mM was obtained by dissolving a dAMP powder
in distilled water. For PL measurements, the solution concentration
was reduced to 0.17 mM. Then, the method of drop-coating was used
for the precipitation of 2 μL of the dAMP solution on the respective
dried metasurface substrate. As a result, a circular area of about
3 mm in diameter was obtained. In Figure , we schematically depicted the structure
of the obtained metasurfaces along with the physical mechanisms explaining
dAMP PL enhancement by such metasurfaces.
Optical Characterization
The spectra of the optical
density (light extinction) of Au NR or Au NP water colloids as well
as the mixture of Au NRs or Au NPs and dAMP molecules were recorded
with a Cary 60 UV–Vis spectrophotometer (Agilent Technologies,
Inc.). The reflection spectra of the metasurfaces were measured with
a double-grating spectrometer DFS-12 (LOMO) equipped with a tungsten–halogen
incandescent lamp as the excitation source. The reflection measurements
were carried out at the light incidence angle of 20°. The PL
spectra were recorded with a Shimadzu RF-6000 spectrofluorophotometer
(Shimadzu Corp.). The excitation wavelength of 260 nm was chosen for
the measurements due to its coincidence with the maximum of the dAMP
absorption peak. Let us note that an area covered by dAMP varied in
diameter from about 2.5 to 3.5 mm in different metasurface samples.
However, the cross section of the exciting beam of the spectrofluorophotometer
was quite large and the dAMP area was entirely illuminated by the
exciting beam. The corresponding PL signal was collected from the
same amount of dAMP molecules in experiments with different metasurface
samples. At the same time, the emission was collected over a rather
large angle by a large-scale focusing mirror; thus, the modification
of scattering varied in different sample surfaces might negligibly
affect the collection efficiency. Therefore, no calibration factors
were applied when calculating the dAMP PL enhancement factors for
different metasurface samples. All samples were inspected at the same
parameters of luminescence imaging, exploiting a Nikon Eclipse Ti-U
inverted microscope at 20× optical magnification equipped with
a Nikon Digital Sight DS-Fi2 camera (Nikon). UV excitation over the
whole imaging area of about 0.5 × 0.5 mm and a long-pass 405
nm filter were used. Luminescence imaging was used only for the qualitative
inspection of sample emission. All of the optical measurements were
carried out in ambient air at room temperature.