Dynamic Orientation Control of Gold Nanorods in Polymer Brushes by Their Thickness Changes for Plasmon Switching

Gold nanorods (AuNRs) have unique optical properties such as transverse and longitudinal localized surface plasmon resonance (T‐ and L‐LSPR). As the L‐LSPR absorption depends on the angle of the AuNRs to incident light and polarization, orientational control of AuNRs is a crucial issue. In spite of various techniques to control AuNR orientation, dynamic orientation tuning on a solid substrate remains challenging. Herein, dynamic changes are demonstrated in AuNR orientation in the anionic polymer (DNA) brushes via control of their thickness by salt concentration. AuNRs vertically align toward the substrates when their thickness exceeds the AuNR length. Once their thickness becomes shorter than the AuNR length, the attached AuNRs begin to tilt. The tilt reaches a maximum level close to horizontal when the thickness decreases to half that of the AuNR length. The dynamic control between the vertical (uniform) and tilted (random) orientation of the AuNRs showed not only absorption intensity changes in L‐LSPR but also the switching of side‐by‐side plasmon coupling. The polymer brush‐based system affords a novel platform for the stimuli‐responsive control of AuNR orientation on the substrates via changes in the thickness of polymer brushes for actively tunable plasmonic substrates.


Introduction
Gold nanoparticles have been attractive in the field of material science and various applications owing to their optical properties (e.g., localized surface plasmon resonance: LSPR).Among the various shapes of gold nanoparticles, the anisotropy of gold DOI: 10.1002/admi.202301066nanorods (AuNRs) endows them with unique optical properties as exemplified by two specific plasmonic absorptions; namely, transverse and longitudinal LSPR (T-and L-LSPR). [1,2]These two LSPR modes arise from the difference in excitation from the perpendicular and parallel components of the electric field generated by incident light.The absorbance resulting from L-LSPR can be modulated by changing the angle of the AuNRs in relation to incident light.In addition, the wavelength of the L-LSPR can be tuned based on the aspect ratio of the AuNRs.These L-LSPR properties enable effective and controllable light absorption within a specific wavelength range.10] To realize such applications, there is a growing need for a control system for the L-LSPR properties currently in widespread use in nanoscale devices. [11]uring the last two decades, the development of techniques to control AuNR orientation on substrates has been at the center of much attention.Researchers have established various alignment techniques for AuNRs on substrates;, e.g., evaporation-induced self-assembly, [12][13][14][15] spin-coating, [16] capillary assembly, [17,18] electrophoresis, [19] and mechanical brushing using shear stress. [20]On the other hand, dynamic orientation control is another way to enhance their potential.][22][23] Despite the abovementioned alignment techniques on substrates providing a high degree of order as well as the various advantages to device fabrication, dynamic changes in the orientation of AuNRs on the solid substrate remain challenging.
One potential solution to overcoming this dilemma lies in the use of polymer materials as soft matters.[35][36][37][38] These characteristics, as well as the inherent anisotropy of polymer brushes, are expected to be more suitable to control the orientation of anisotropic nanoparticles.
We have previously achieved vertical alignment of AuNR on DNA-immobilized substrates (DNA brushes). [39,40]Adjustment of the strength of electrostatic interactions between cationic AuNR and DNA via tuning of cationic amount around its surface allowed the AuNRs to be attached along with DNA polymer, resulting in perpendicular orientation to the substrates.Furthermore, following the vertical alignment within the DNA brushes, we demonstrated dynamic orientation changes via changes in interaction strength between the AuNR and DNA. [41]These studies suggest that modulation of AuNR surface properties provides opportunities for further tuning of AuNR motion in DNA brushes.In contrast, as AuNR orientation depends on the conformation of polymer brushes as templates, control of AuNR orientation with the structural changes of the polymer brush can be a versatile system for actively tunable plasmonic substrates via various stimulation by applying to a broad range of polymers.
To realize this concept, in this work, we demonstrate dynamic changes to AuNR orientation via changes in the thickness of polymer brushes (Figure 1).[44][45] Salt concentration changes were used to control the brush thickness.Then, the orientation of the AuNRs embedded in the DNA brushes against brush thickness was investigated for varied DNA lengths, salt types, and AuNR lengths, providing a general insight into the stimuli-responsive control of AuNR orientation on the substrates via changes in the thickness of polymer brushes.

AuNR Orientation in DNA Brushes with Various Thickness
First, we investigated orientation of the attached AuNRs on DNA brushes of different thicknesses.[41] Briefly, biotinylated ss-DNA was immobilized on streptavidin-immobilized substrates through avidin-biotin interactions.To achieve varied thicknesses, six ssDNA lengths (50, 100, 125, 148, 175, and 200 bases) were selected for each ssDNA brush.The DNA densities were adjusted to a constant value (20 000 chains μm −2 , ≈8 nm as the interchain distance, Figures S1 and S2, Supporting Information) on the basis of their absorption or fluorescent intensity.Subsequently, cationic AuNRs, which surface was modified with a mixture of cationic and nonionic ligands at a 30:70 ratio, were attached to the ssDNA brushes via electrostatic interactions (Length: 36 nm, Width: 10 nm, Figure 2a; Figure S3, Supporting Information).
Figure 2b shows the extinction spectra of AuNRs attached to DNA brushes with various DNA lengths in 10 mm Tris-HCl (pH 7.6) buffer by conventional measurement in which the light was irradiated perpendicular to the substrate.The AuNRs in the 50-, 100-, and 125-base DNA brushes exhibited high L-LSPR intensity, with the shorter DNA brushes showing a particularly heightened L-LSPR peak intensity.As AuNRs show stronger L-LSPR intensity at a higher angle (≈90°) of incident light to the AuNRs, these results suggest that AuNRs were adsorbed at a greater incline in the thinner DNA brushes.In contrast, the extinction spectrum of the AuNRs attached to 148-base DNA brushes showed an approximately five times weaker L-LSPR peak intensity (≈800 nm) than T-LSPR peak intensity.[48] As for longer DNA length (175-and 200-base DNA) brushes, the L-LSPR intensity did not show any significant difference compared to the results for the 148-base DNA brushes, indicating that the AuNRs were also aligned vertically to the substrates in brushes with over 148-base DNA.
To further discuss the effect of brush thickness, we focused on the thickness of the polymer brushes under each set of conditions.For the determination of polymer brush thickness at a relatively low density, an experimental examination is quite demanding due to the ambiguity regarding the brush interface. [49]Therefore, we introduced the Daoud-Cotton model (DCM), which was extensively investigated previously. [45]The modified DCM for a flat surface is given by where k is a constant (close to unity), N is the number of ss-DNA bases, d is the average internucleotide distance, l k is the Kuhn length,  is the ssDNA surface density, and  D is the Debye length.The values of l k and d were taken from reference 45, and  D was calculated from the following equation, applied to Debye-Hückel theory with two ionic species: [50]  where ϵ r is the zero-frequency dielectric constant of the continuous medium, ϵ 0 is the permittivity of free space, R is the gas constant, T is the temperature, e is the elementary charge, z is the number of charges on an ion, N A is the Avogadro constant, and C is the salt concentration.
To quantitatively discuss the difference in AuNR orientation by brush thickness, we utilized the intensity of the L-LSPR peak at a longer wavelength divided by that of the T-LSPR peak at a shorter one (L-/T-LSPR) as an indicator of the orientation of AuNRs. [25,40]The L-/T-LSPR intensity exhibited a value of 3.7 for 12 nm brush thickness, 1.7 for 25 nm brush thickness, and 0.9 for 34 nm brush thickness (Figure S5, Supporting Information).On the other hand, the L-/T-LSPR intensity is significantly lower, ca.0.2, when brush thickness exceeds 40 nm.These results indicate that the orientation of AuNRs shows thickness-dependent and -independent regions with the threshold value of DNA brushes at ≈35 nm, which is comparable with the length of the AuNR.

Dynamic Orientation Change of AuNRs in the DNA Brushes with Varied NaCl Concentration
Next, we demonstrated dynamic orientation change of vertically attached AuNRs induced by the shrinkage of DNA brushes via increase in salt concentration (x mm NaCl + 10 mm Tris-HCl (pH 7.6)).Following the vertical attachment of AuNRs (36 × 10 nm), NaCl concentration was changed by solvent replacement.The L-LSPR peak intensity exhibited a corresponding increase without significant changes in T-LSPR as the NaCl concentration increased from 0 to 40 mm (Figure 3a).This result indicates that the orientation of the attached AuNRs gradually tilted with the increase in NaCl concentration.The calculated thickness of the 148-base DNA brushes was 41.5 nm at 0 mm NaCl, and 30.8 nm at 40 mm NaCl.As AuNR orientation differs depending on brush thickness, this dynamic orientation change occurs due to the thickness changes.With regard to NaCl concentrations of 50 mm or higher, the 30% cationic AuNRs adsorbed in the DNA brushes were detached due to the weakened electrostatic attraction between the AuNRs and DNA.
[48] The vertical alignment of the AuNRs effectively leads to the side-by-side coupling phenomenon.When vertical AuNRs in the DNA brushes were irradiated with p-polarized light at a 45-degree angle, an extinction peak was observed ≈640 nm (highlighted in red), originating from the blueshifted L-LSPR peak by the side-by-side coupling.Meanwhile, configuration change of the AuNRs from a vertical to a tilted state on increases in NaCl concentration induced a decrease in the extinction of blueshifted L-LSPR and an increase in the extinction of the original L-LSPR with longer wavelengths (highlighted in blue), indicating decoupling of their plasmon via AuNR configuration changes.This active plasmonic system, based on the vertical (uniform) / tilted (random orientation) switching of AuNRs, holds promising potential for applications, such as colorimetric display. [11]urther, we investigated the reversibility of this dynamic orientation change.With the decrease in NaCl concentration, the L-LSPR peak intensity decreased and returned to its initial intensity at 0 mm NaCl.This result indicates that the tilted AuNRs returned to a vertical alignment when the NaCl concentration decreased.Upon repeated solvent exchange between 0 and 40 mm NaCl, the L-/T-LSPR intensity of the attached AuNRs changed reversibly for at least ten cycles (Figure 3c).It is noteworthy that no hysteresis was observed when plotting the L-LSPR intensities in relation to the changes in NaCl concentration (Figure S6, Supporting Information).This result supports the notion that the mechanism underlying the orientational change differs from our previously established system of pH-triggered orientational tuning, which relies on the electrostatic interaction strength between DNA and cationic AuNRs. [41]Whereas the interaction strength is indeed an important factor for the AuNR orientation, as we have previously reported, [39] decreases in the interaction strength by increases in pH (from the initial strength to the strength at AuNR detachment) did not induce the AuNR orientation change until AuNR detachment (Figure S7, Supporting Information).Furthermore, dynamic orientation change did not occur in double-stranded DNA (dsDNA) brushes, which are more rigid than ssDNA brushes (persistence length: ≈50 nm for dsDNA, [51] 2-3 nm for ssDNA [52,53] ) (Figure S8, Supporting Information).These observations support the notion that the shrinkage of polymer brushes is a dominant factor in the dynamic orientation change of the AuNRs in this system.

Generalization of AuNR Orientation Change by Shrinkage/Extension of Polymer Brushes
In order to establish a general framework for this system, we further investigated the dynamic orientation change of 36 nm AuNRs across a broad range of thicknesses using ssDNA of 50-, 100-, 125-, 175-, and 200-bases in length.The corresponding spectral changes are shown in Figure S9 (Supporting Information).AuNRs attached to ssDNA brushes of 50-or 200-bases in length, which are significantly shorter and longer than the AuNR length, respectively, did not show any spectrum changes in response to varying NaCl concentrations (Figure S9a,e, Supporting Information).When NaCl concentration was increased in 100-base and 175-base ssDNA brushes with AuNRs, the AuNRs exhibited a slight increase in L-LSPR peak intensity corresponding to the NaCl concentration (Figure S9b,d, Supporting Information).As for 125-base ssDNA brushes, whose thickness is slightly shorter than the AuNR length, the L-LSPR peak intensity of the slightly tilted AuNRs showed corresponding increase to the NaCl concentration (Figure S9c, Supporting Information).To estimate the tilt angle in this intricate plasmonic system, the L-/T-LSPR intensity of AuNRs directly attached on bare glass substrates at the same AuNRs density was determined at a 90°angle; that is, parallel to the substrates, as a reference.The SEM image revealed that all of the adsorbed AuNRs were not piled on the substrate, suggesting that they were adsorbed parallel to the substrate, and the extinction spectrum showed an L-/T-LSPR intensity of 3.92 (Figure S10, Supporting Information).To elucidate the relationship between AuNR orientation and brush thickness, all L-/T-LSPR intensities were normalized with respect to the value obtained at a 90°angle (3.92) and plotted against brush thickness (Figure 4).When the brush thickness was below half that of the AuNR length (less than 18 nm), the tilt angle of the AuNRs showed a maximum value on the DNA brushes.The normalized L-/T-LSPR intensity approached 0.9, suggesting that the maximum angle to the substrate was close to 90°.For brush thicknesses within the range between 18 and 36 nm, which are comparable to the AuNR length, the orientation of the attached AuNRs was dependent on the brush thickness, enabling dynamic orientation change of the AuNRs.When the brush thickness exceeded the attached AuNR length, the AuNRs in DNA brushes were almost vertical to the substrate.This relationship between AuNR length and brush thickness indicates that the origin of the motive force for their tilting was energy gain from electrostatic bonding between cationic AuNR and DNA.In other words, AuNRs tilt to reduce energetic loss from exposure of the cationic surface to the solution.
Furthermore, we investigated the effect of other types of salt (KCl, CaCl 2 , and MgCl 2 ), whose shrinkage effect on ssDNA brushes has been extensively explored. [45]The L-LSPR peak intensity of AuNRs in 148-base ssDNA brushes exhibited a corresponding increase as the KCl concentration was raised from 0 to 40 mm (Figure S11a, Supporting Information).In contrast, a slight increase in the number of divalent ions (1-5 mm MgCl 2 and 1-6 mm CaCl 2 ) caused a drastic tilt in the attached AuNRs, resulting in the L-/T-LSPR intensity ranging from 0.4 to 3.4 (the normalized intensity: 0.1-0.9, Figure S11b,c, Supporting Information).When the acquired data were plotted, all the data exhibited an excellent fit to the same sigmoidal curve shown in Figure 4 (Figure S11d, Supporting Information).Although the spectrum changes with the divalent ions were not perfectly reversible, likely due to insufficient ionic exchange in strongly attracted divalent ions, [54] these results show that the dynamic orientation changes can be achieved across a wide range of Finally, we generalized this system by varying AuNR length, using 24 and 45 nm AuNRs (Figure 5).We selected 100-and 175-base ssDNA brushes for dynamic orientation change in 24 and 45 nm AuNRs, respectively.As expected, dynamic orientational change of the 24 nm AuNRs was successfully obtained in 100-base ssDNA brushes (Figure 5a,b).When the NaCl concentration increased stepwise from 0 to 30 mm, the L-/T-LSPR intensity correspondingly increased.Likewise, a similar dynamic orientation change was observed with 45 nm AuNRs attached to 175-base ssDNA brushes (Figure 5c,d).These results revealed that the orientation of attached AuNRs of different lengths can be tuned dynamically by tailoring the brush thickness.
Figure 5e summarizes the orientation (normalized L-/T-LSPR; references are shown in Figure S10, Supporting Information) of these three types of AuNRs relative to brush thickness.All the data were collected using the same methods as for the experiment using 36-nm AuNRs (Figures S12-S14, Supporting Information).The orientation of both the 24 and 45 nm AuNRs follows the same trend as that of the 36 nm AuNRs.All fitted-curves reach their maximum when the "brush thickness/AuNR length" falls below half the length of the AuNRs.These maximum intensities reached ≈0.95, indicating that the AuNRs were nearly parallel to the substrate.In the range of "brush thickness/AuNR length" from 0.5 to 1.0, the orientation of the attached AuNRs depended on brush thickness, enabling dynamic orientation tuning of AuNRs across a wide range of angles.When the "brush thickness/AuNR length" was close to unity or greater, the AuNRs in the DNA brushes were almost vertical to the substrate.The minimum intensity of the 24 nm AuNRs was higher than that of the longer AuNRs due to the spectrum overlap of T-LSPR on L-LSPR.These results indicate that this is a generalized system for the controlling the AuNR orientation using polymer brush thickness changes, as AuNR orientation responds to the ratio of "brush thickness/AuNR length".This work not only clarified the relationship between AuNR orientation in polymer brushes and brush thickness but also offers a framework for novel plasmonic devices using active AuNR orientation control on the solid substrates.

Conclusion
We have achieved the reversible orientation control of AuNRs in anionic polymer brushes through the structural changes of the polymer induced by salt concentration.The orientation of cationic AuNRs attached to DNA brushes dynamically changes depending on NaCl concentration because of NaCl-induced thickness changes.The dynamic control between the vertical (uniform) and tilted (random) orientation of the AuNRs enables not only absorption intensity changes in L-LSPR but also the switching of plasmon coupling.Furthermore, we generalized the orientation of AuNRs on the anionic polymer brushes with brush thickness as a variable by calculating the DNA brush thickness using the Daoud-Cotton model for a flat surface.The AuNRs on the brushes vertically align toward the substrates when the thickness is longer than the AuNR length.In cases where the thickness is shorter than the AuNR length, the attached AuNRs tilt depending on the thickness with a wide range of angles, and the tilt orientation reaches close to parallel when the thickness decreases to almost half of the AuNR length.Here, salt-triggered thickness changes in anionic polymer brushes were applied as a simple model.The generalized insights obtained from this research lead expansion of this platform to a variety of polymer brushes with different stimulus-induced shrinkage and swelling properties.This platform offers great potential for hybrid materials composed of polymer brushes and anisotropic nanoparticles by design.Density evaluation of DNA brushes by absorption spectrum measurement was performed according to the following calculation formula.
Considering that DNA is immobilized on both sides inside the cuvettes, the relationship between the DNA concentration in the cuvettes and the immobilization density of the DNA polymer brush is as follows: where  is DNA immobilization density, C is the DNA concentration in cuvettes, V is the volume of the solution, N A is the Avogadro constant, and S is surface area of the substrate on which DNA is immobilized.
To calculate the interchain distance of DNA, the study assumed that the DNA molecules were arranged in a hexagonal close-packed pattern (Figure S15, Supporting Information).From the estimation, the interchain distance (D DNA ) can be calculated by the following formula: From the above calculations, the density of DNA brushes and the distance between DNAs were obtained.
Surface Modification of AuNRs with Two Types of Thiol-Ligands: Cationic ligands (NH 2 EG 6 C 11 ) and nonionic ligands (OHEG 6 C 11 ) were dissolved in water.The two types of ligand solutions were mixed, adjusting the total ligand concentration (10 mm).The percentages of cationic ligands were 30%.Before adding ligand solution to the AuNR solution, the AuNR solution (1 mL) were centrifuged (18 000 g, 20 min, 30 °C), and MilliQ-water (940 μL) was added after the supernatant (940 μL) was removed.This centrifugal purification procedure was repeated once (24 × 10 nm AuNRs) or twice (36 × 10 nm, and 45 × 10 nm AuNRs) so that the CTAB concentration was ≈0.36 mm, a concentration suitable for ligand modification.After purification, the mixed ligand solution (50 μL) was added to the purified AuNR solution (950 μL) in a 1.7 mL plastic tube.The AuNR solution with the ligands was vortexed and kept undisturbed at 42 °C for at least 12 h.Next, the AuNR solution (1000 μL) was centrifuged (18 000 g, 20 min, 25 °C), the supernatant (950 μL) was removed, and MilliQ-water (950 μL) was added, and the solution was then vortexed.This purification procedure was performed twice.Finally, to obtain a concentrated AuNR solution, the AuNR solution was centrifuged (18 000 g, 20 min, 25 °C), and the supernatant (700 μL) was removed.
Attachment of Ligand-modified AuNRs onto DNA Polymer Brushes: The Thiol ligand-capped AuNR (cationic AuNR) solution was poured into the inner wall of DNA brush cuvettes.After 10 min incubation, the solution was removed using a pipet and the cuvette was rinsed with a 10 mm

Figure 1 .
Figure 1.Dynamic orientation change of AuNRs inserted into DNA brushes based on shrinkage/extension of the DNA brushes.

Figure 3 .
Figure 3. Salt-triggered spectral changes of AuNRs (36 × 10 nm) in 148-base ssDNA brushes.a) Extinction spectra in a range of 0 to 40 mm NaCl.Non-polarized light was irradiated perpendicular to the substrates in these measurements.b) Extinction spectra under p-polarized light at an angle of 45 degrees relative to the substrate with NaCl concentration.c) L-LSPR/T-LSPR values on repeated changes in NaCl concentration between 0 and 40 mm.

Figure 4 .
Figure 4. L-/T-LSPR values of AuNRs (36 × 10 nm) against calculated brush thickness.Concentration of NaCl was varied (from 0 to 40 mm) to shrink and extend each DNA brush.Black solid line is curve-fitted from the obtained data.