Optomechanical tuning of nonlinearity in graphene resonators
Extracted source text · arXiv 2607.29096
\documentclass[journal=jacsat,manuscript=article]{achemso}
\usepackage{graphicx}
\usepackage{hyperref}
\usepackage[version=3]{mhchem} % Formula subscripts using \ce{}
\author{Alberto Martín-Pérez}
\affiliation{Department of Precision and Microsystems Engineering, Faculty of Mechanical Engineering, Delft University of Technology, Netherlands}
\email{A.MartinPerez@tudelft.nl}
\author{Peter G. Steeneken}
\affiliation{Department of Precision and Microsystems Engineering, Faculty of Mechanical Engineering, Delft University of Technology, Netherlands}
\author{Farbod Alijani}
\affiliation{Department of Precision and Microsystems Engineering, Faculty of Mechanical Engineering, Delft University of Technology, Netherlands}
\title[An \textsf{achemso} demo]
{Optomechanical tuning of nonlinearity in graphene resonators}
\abbreviations{}
\keywords{Optomechanics, Nanomechanics, 2D materials, NEMS}
\begin{document}
\begin{abstract}
Graphene resonators exhibit unique mechanical and electrical properties, making them promising candidates for next-generation sensing applications. Here, we introduce an optomechanical approach that enables active, reversible, in situ tuning of the nonlinear nanomechanical response of graphene resonators. By exploiting photothermal heating and the resulting optomechanical interaction between the mechanical oscillations and the optical cavity formed by a graphene membrane suspended over a reflective substrate, we achieve dynamic control over both the linear restoring force and the nonlinear dynamic response of the resonator. We show that the strength and sign of the Duffing nonlinearity can be continuously tuned through the optical probe power and cavity depth, enabling transitions between hardening and softening behavior as well as near-complete suppression of geometric nonlinearities. Our results establish a versatile platform for on-demand engineering of nonlinear dynamics in graphene resonators and provide a general strategy for active control of the mechanical response of two-dimensional nanoelectromechanical systems.
\end{abstract}
Nanomechanical resonators are excellent tools for probing the nanoworld providing high-sensitivity devices whose versatility has enabled applications spanning from mechanical computation, including logic gates \cite{Tadokoro2021HighlySensitive}, to nanoscale probes for condensed-matter phenomena \cite{bachtold2022mesoscopic}. In this respect, nanoscale sensing is one of the most prominent applications of nanomechanical resonators, which have demonstrated ultrahigh sensitivity in the detection of various physical quantities, including forces\cite{Binnig1986AFM}, temperature\cite{Mertens2003TemperaturePressure}, pressure\cite{ 10.1063/1.4889744, MARTINPEREZ2025117161}, mass\cite{MARTINPEREZ2025117161}, stiffness \cite{Malvar2016MassStiffness} and optical absorption\cite{Kirchhof2023NanomechanicalAbsorption,Chien2018SingleMolecule}. Their exceptional sensitivity has enabled detection limits down to individual molecules \cite{Naik2009SingleMoleculeMS,Chaste2012Yoctogram} and even single electron spins \cite{Rugar2004SingleSpin}. Moreover, the introduction of low-dimensional materials such as graphene or carbon nanotubes as nanomechanical systems \cite{Steeneken2021Dynamics2D, xu2022nanomechanical} offers the prospect of further enhancing the sensitivity of these devices thanks to the unique features of these materials and their exceptional aspect ratio.
Regardless of the target application, the performance of a nanomechanical resonator relies on precise control of its mechanical stiffness, which governs both its linear and nonlinear dynamics. Several strategies have been developed to tune this response. Geometric design can be used to tailor the cubic stiffness and even reverse its sign, as demonstrated in resonators made of bulk silicon nitride \cite{Li2024StrainEngineering} and graphene \cite{Yousuf2025GrapheneTrampoline}. Electrostatic actuation provides post-fabrication control of the resonance frequency and can also modify the nonlinear response \cite{sajadi2017experimental, samanta2018tuning}, while thermal tuning changes the mechanical tension through thermal expansion \cite{davidovikj2018chip, vsivskins2025nonlinear}. These approaches offer control over individual devices, but they do not by themselves ensure reproducible dynamics across nominally identical resonators. Furthermore, the response depends on material and fabrication parameters, including Young’s modulus, mass density, built-in tension, thickness, and membrane flatness. Variations in these quantities can therefore shift both the linear and nonlinear stiffness and obscure the behavior intended through geometric tuning. For established bulk materials such as silicon and silicon nitride, mature nanofabrication processes provide comparatively tight control over these parameters. The fabrication of resonators based on low-dimensional materials remains less reproducible, however, with substantial device-to-device variations in tension, thickness, and static shape \cite{sarafraz2024quantifying}. Understanding and controlling additional perturbations introduced during measurement and actuation is therefore essential for the reliable implementation of these materials in NEMS.
Nevertheless, even in a scenario in which all fabrication challenges are addressed, low-dimensional material based resonators face an additional reproducibility challenge: transduction techniques, such as electrostatic or optothermal readout, can introduce additional stress, altering the stiffness of the resonator. This transduction-induced stiffness change is the consequence of the energy exchange between the probe and the device\cite{Aguila2022PhotothermalResponsivity, barton2012photothermal}. Consequently, the reproducibility of these resonators depends not only on fabrication but also on understanding and quantifying transduction-induced perturbations.
Here, we theoretically and experimentally quantify how optical interferometric transduction and actuation modify the linear and nonlinear stiffness of suspended graphene membranes. We show that this optomechanical effect arises from the combined influence of photothermal heating and membrane thermal expansion. Building on this understanding, we develop a model that obtains active optical control of both the linear and cubic stiffness, allowing the membrane response to be switched from nonlinear hardening to nonlinear softening by tuning the incident light.
Optical interferometry is a widely used scheme for transducing the motion of nanomechanical resonators based on 2D materials. In this approach, a suspended membrane is positioned above a highly reflective substrate so when a laser beam is directed onto it, a standing electromagnetic wave is obtained through the interference of the incident and reflected light fields\cite{Steeneken2021Dynamics2D}, forming an optical cavity (Fig. \ref{fig:fig1}a). The intensity of this standing wave is determined by the membrane-substrate separation (i.e. cavity depth) and the reflectivities of the different interfaces of the device. In this configuration, nanomechanical vibrations of the 2D material modulate the cavity depth, encoding the amplitude of these oscillations in the intensity of the light reflected by the cavity, which forms the basis of interferometric motion transduction (see Supplementary Information for details).
To investigate how optomechanics affects the membrane motion beyond the linear regime, we model the suspended membrane as a driven Duffing oscillator. Denoting the out-of-plane deflection by \(x\), the mass-normalized equation of motion is
\begin{equation}
\ddot{x}+2\zeta\dot{x}+\omega^2 x+\gamma x^2+\beta x^{3}= F\cos(\omega_\mathrm{d} t),
\label{eqn:0}
\end{equation}
where \(\zeta\) is the damping ratio and \(\omega\) is the resonance frequency. Furthermore, \(\gamma\) and \(\beta\) are the mass-normalized quadratic and cubic nonlinear coefficients, respectively. The term \(F\cos(\omega_\mathrm{d} t)\) represents the mass-normalized harmonic drive, where \(\omega_\mathrm{d}\) is the drive frequency. For tension-dominated membranes, the resonance frequency is directly related to the membrane tension, such that \(\omega^2\propto n\). In addition, the Duffing coefficient (also known as cubic stiffness term) scales with the Young's modulus, \(\beta\propto E\) \cite{davidovikj2017nonlinear}, whereas the quadratic coefficient scales with the static out-of-plane deflection, \(\gamma\propto x_\mathrm{eq}\) \cite{li2026cascade}. Consequently, the quadratic term vanishes for initially flat membranes. When the membrane is buckled or deflected, however, the quadratic nonlinearity also contributes to the amplitude-dependent frequency shift and can be incorporated into an effective cubic coefficient, \(\beta_\mathrm{eff}=\beta-\frac{10}{9}\frac{\gamma^2}{\omega^2}\), which governs the observed hardening or softening nonlinear responses \cite{Steeneken2021Dynamics2D}.
For membranes with finite optical absorption, we note that a fraction of the optical energy is converted into heat, resulting in an increase in membrane temperature,
\begin{equation}
\Delta T = \eta \left( 1+ M \cos{\left[\frac{4\pi x}{\lambda}\right]}\right) P_{\lambda} ,
\label{eqn:1}
\end{equation}
where the temperature rise $\Delta T$ depends not only on the incident laser power, $P_{\lambda}$, but also on the distance of the membrane from the bottom of the cavity (Fig. \ref{fig:fig1}b), which is approximately equal to the cavity depth $d$, through the position-dependent optical intensity. Here, $ \eta $ is the photothermal responsivity of the membrane, $M$ is a factor that depends on the cavity properties and $ \lambda $ is the wavelength of the laser used. A detailed description of $ \eta$ and $M$ can be found in the Supplementary Information.
The temperature increment $\Delta T$ generates thermally induced stresses in the membrane through thermal expansion. Since $\Delta T$ depends on the membrane position within the optical wave, the resulting restoring force becomes displacement dependent, thereby establishing a photothermal interaction between the optical field and the mechanical motion. The resulting photothermal tension can be decomposed into the sum of a static and a dynamic contribution ($n=n_\mathrm{s}+n_\mathrm{d}$), in which the static term is determined by the equilibrium position $x_\mathrm{eq}$, while the dynamic term, is determined by the motion around that equilibrium position. Here, the static component modifies the tension of the membrane at equilibrium and therefore its linear stiffness and resonance frequency, whereas the dynamic component introduces a displacement-dependent restoring force that alters the nonlinear mechanical response. In the limit of small oscillation amplitudes ($\Delta x \ll \lambda/4$), and assuming that only the membrane heats up, as typically encountered in thermally driven motion, the membrane oscillates around $x_\mathrm{eq}$ and the effective membrane static tension becomes
\begin{equation}
n_\mathrm{s}=n_\mathrm{0}-\frac{Eh\alpha \eta}{1-\nu}
\left(1+M\cos{\left[\frac{4\pi}{\lambda}x_\mathrm{eq}\right]}\right)P_{\lambda} ,
\label{eqn:3}
\end{equation}
where $h$ is the membrane thickness, $\alpha$ is the effective thermal expansion coefficient, and $\nu$ is the Poisson ratio. Using Eq.~\ref{eqn:3} we calculate the photothermally modified resonance frequency as (see Supplementary Information for the detailed derivation),
\begin{equation}
\omega^2={\omega_\mathrm{0}}^2 \left[ 1-\frac{Eh\alpha\eta}{n_\mathrm{0} \left(1-\nu\right)}
\left(1+M\cos{\left[\frac{4\pi}{\lambda}x_\mathrm{eq}\right]}\right)P_{\lambda} \right].
\label{eqn:4}
\end{equation}
Equation~\ref{eqn:4} shows that the resonance frequency consists of the intrinsic value $\omega_\mathrm{0}$ and a photothermally induced correction. This correction scales linearly with the optical power and is periodically modulated by the equilibrium position within the cavity. As a result, the resonance frequency can be tuned not only through the incident laser power but also by changing the membrane position within the optical cavity (Fig.~\ref{fig:fig1}c).
For larger oscillation amplitudes, particularly when the membrane is driven into the nonlinear regime, the dynamic component of the photothermal tension ($n_\mathrm{d}$) becomes relevant. Expanding this displacement-dependent contribution around the equilibrium position yields
\begin{equation}
\begin{split}
n_\mathrm{d}(\Delta x)
=
-{\omega_\mathrm{0}}^2\frac{E h \alpha \eta}{n_\mathrm{0}(1-\nu)}M
\Bigg(
&-\frac{4\pi}{\lambda}\sin{\left[\frac{4 \pi x_\mathrm{eq}}{\lambda}\right]} \Delta x \\
&-\frac{8\pi^2}{\lambda^2}\cos{\left[\frac{4 \pi x_\mathrm{eq}}{\lambda}\right]} \Delta x^2
+\cdots
\Bigg) P_{\lambda} ,
\end{split}
\label{eqn:5}
\end{equation}
where $\Delta x=x-x_\mathrm{eq}$. Owing to the high thermal conductivity of graphene, we assume that the photothermal response can be treated as quasi-static, such that the membrane temperature follows its displacement without a measurable delay. Here, the displacement-dependent terms generate nonlinear restoring forces. Restricting the analysis to the leading nonlinear contribution, the effective Duffing coefficient becomes
\begin{equation}
\beta_\mathrm{eff}=\beta_\mathrm{0}+\frac{8 \pi^2}{\lambda^2} \rho M {\omega_\mathrm{0}}^2 \cos{\left( \frac{4 \pi}{\lambda} x_\mathrm{eq} \right)}P_{\lambda} - \frac{10}{9} \frac{ {\left(\gamma_\mathrm{0}+\frac{4\pi}{\lambda} \rho M {\omega_\mathrm{0}}^2 \sin{\left[ \frac{4 \pi}{\lambda} x_\mathrm{eq} \right] } P_{\lambda}\right) }^2}{\omega^2(P_{\lambda},x_\mathrm{eq})}.
\label{eqn:6}
\end{equation}
where $ \rho=\frac{Eh\alpha\eta}{n_\mathrm{0}(1-\nu)} $. Equation \ref{eqn:6} shows that the nonlinear stiffness consists of these intrinsic contributions as well as a photothermally induced correction that can be controlled through both the optical power and the equilibrium position. Unlike the resonance frequency shift, the nonlinear correction exhibits a $\pi$-phase offset with respect to the cavity position and can exceed the intrinsic contribution (Fig. \ref{fig:fig1}d). More specifically, Equation~\ref{eqn:6} reveals two mechanisms through which the nonlinear response can be tuned. First, the direct photothermal correction to the cubic stiffness is proportional to $\cos\!\left(4\pi x_\mathrm{eq}/\lambda\right)$ and therefore changes sign as the equilibrium position is varied within the optical cavity. Thus, for cavity positions satisfying $\cos\!\left(4\pi x_\mathrm{eq}/\lambda\right)<0$, the photothermal contribution increases the effective Duffing coefficient, whereas for $\cos\!\left(4\pi x_\mathrm{eq}/\lambda\right)>0$ it promotes softening behavior (please note that for graphene $\rho<0$). A second contribution arises from the quadratic nonlinearity which always provides a negative correction to $\beta_{\mathrm{eff}}$. The magnitude of this term increases with an increase in the effective quadratic nonlinearity, $\gamma=\gamma_\mathrm{0}+ \frac{4\pi}{\lambda}\rho M{\omega_\mathrm{0}}^2 \sin\left(\frac{4\pi}{\lambda}x_{\mathrm{eq}} \right)P_{\lambda}$, or by a reduction in the resonance frequency due to photothermal heating. Consequently, buckled or statically deflected membranes, as well as resonators exhibiting strong photothermal softening, display an enhanced tendency toward softening nonlinear dynamic behavior.\\
To validate the analytical predictions, we measure the mechanical response of circular bilayer graphene membranes suspended over a circular hole in a silicon oxide layer grown on a polished silicon substrate (Fig \ref{fig:fig2}a). The measurements are carried out using a custom-built optical interferometer (Fig.~\ref{fig:fig2}b), where the optical power serves as a control knob to tune the tension and thus $x_\mathrm{eq}$. A red Helium-Neon laser ($\lambda$ = 632.8 nm) is tightly focused onto the membrane to transduce its motion interferometrically. The reflected signal, whose intensity is modulated by the membrane motion, is detected with an avalanche photodetector and analyzed using a spectrum analyzer (Rohde \& Schwarz FPL1003), allowing the acquisition of the thermomechanical (Brownian) motion spectrum of the resonator under different optical powers (Fig.~\ref{fig:fig2}c). In addition, the setup includes a power-modulated blue diode laser ($\lambda$=488 nm), which serves both as a controllable source of photothermal heating and as an actuator for driving the mechanical motion. The blue laser provides a static optical-power component, that determines the average absorbed optical power, and a modulated component, that generates a periodic photothermal drive force. Since the modulated contribution has zero average value and is substantially smaller than the static component, its effect on the average membrane temperature can be neglected. By synchronizing the modulation signal with the interferometric readout using a vector network analyzer (Rohde \& Schwarz ZNB4), we measure the driven frequency response of the resonators to access the nonlinear Duffing regime (Fig.~\ref{fig:fig2}d). At sufficiently large amplitudes, higher harmonics become visible in the optical response (Fig.~\ref{fig:fig2}e). These harmonics are used to extract both the oscillation amplitude and the equilibrium position $x_\mathrm{eq}$ of the membrane using the calibration method introduced in Ref. \citenum{Dolleman2017AmplitudeCalibration}. A detailed description of the procedure is given in the Supplementary Information.\\
To investigate the experimental control over the stiffness as predicted by the analytical model (Eq. \ref{eqn:3} and \ref{eqn:4}) we characterize the dynamics of a 8 $\mu m$ diameter graphene membrane suspended over a 420 nm cavity under the interferometer using different laser powers. To probe the linear stiffness we measure the Brownian motion of the membrane (Fig. \ref{fig:fig3}a), and obtain the resonance frequency by fitting the resulting curve to the harmonic oscillator model (Fig.\ref{fig:fig2}c). Fig. \ref{fig:fig3}a shows that the resonance frequency increases both with red and blue laser power, following the theoretical prediction (Eq. \ref{eqn:4}) with high accuracy ($R^2>0.97$), while assuming a constant value of $x_\mathrm{eq}$. It shall be noted that the resonance frequency increases with the power since graphene has a negative effective thermal expansion coefficient at room temperature $\alpha<0$.
After having qualitatively confirmed the linear tuning of the resonance frequency $\omega$, we analyze the effect of photothermal heating on the cubic nonlinear stiffness. To this end, we drive the membranes by modulating the blue laser and measuring the response using a vector network analyzer under different conditions of the static laser power (non-modulated component). Each mechanical spectrum obtained in this case is fitted to a Duffing resonator model (Fig. \ref{fig:fig2}d), obtaining the effective Duffing coefficient $\beta$. Here, it shall be noted that, changing the power of the red laser also changes the transduction factor (i.e. the proportionality constant relating the measured voltage with the membrane displacement). Nevertheless, in these drums the transduction factor cannot be obtained given the signal to noise ratio is not high enough to measure the third harmonic, which impedes a direct comparison of the cubic stiffness for these results. To enable a consistent comparison, we normalize each spectrum to the power of the red laser (directly proportional to the transduction factor) to eventually obtain the cubic stiffness by fitting. The data obtained for the effective Duffing coefficient (Fig. \ref{fig:fig3}b) reveal that, for lower powers of the red laser, the effective cubic stiffness increases following a linear trend, which agrees well with the analytical model. For higher red laser power ($P> 5$ mW), the change in the effective Duffing constant deviates from the linear trend, which can be explained with the change in the resonance frequency (Fig. \ref{fig:fig3}b).
The analytical model also predicts a negative trend in the shift of the effective cubic stiffness for certain values of the equilibrium position of the membrane. Consequently, to analyze such cases, we measure a second graphene membrane with the same diameter but suspended over a 285 nm cavity. In this case, we measure the Brownian motion for different values of the laser power following the same procedure as in the previous case (Fig. \ref{fig:fig3}c). The resonance frequency of this device shows a non-monotonic dependence on optical power, first decreasing and subsequently increasing as either the red or the blue laser power is raised. This trend is inconsistent with the analytical model, which assumes an initially flat membrane and therefore predicts a monotonic frequency shift. Instead, the observed behavior closely resembles that previously reported for thermally buckled membranes \cite{Liu_2024}.
To verify the presence of buckling, we therefore measure the equilibrium position of the membranes as a function of optical power for both cavity depths by mapping the reflected intensity in the sample, comparing the reflected intensity in the center of the cavity and outside the cavity (Fig. \ref{fig:fig3}d). A complete description of this procedure is provided in the Supplementary Information. We notice that the membrane suspended above the 285 nm cavity shows a power-dependent shift of the equilibrium position of up to 10 nm (Fig. \ref{fig:fig3}e), confirming a buckled membrane configuration. In contrast, the membrane suspended above the 420 nm cavity shows no measurable change in its equilibrium position over the investigated power range, consistent with the absence of buckling. Interestingly, for the buckled membrane, the measured effective Duffing coefficient as a function of laser power (Fig.~\ref{fig:fig3}f) decreases with increasing optical power, in agreement with the analytical model. This reduction can be particularly strong when the accompanying reduction in resonance frequency further amplifies the negative contribution to $\beta_{\mathrm{eff}}$ (see Eq.~ \ref{eqn:6}). Although buckling invalidates the assumptions used to derive the resonance frequency model (Eq.~\ref{eqn:4}), the analytical expression for the effective Duffing coefficient (Eq.~\ref{eqn:6}) remains applicable provided that the experimentally measured resonance frequency is used in the evaluation of $\beta_{\mathrm{eff}}$.
To further validate the proposed nonlinear stiffness tuning mechanism and explore its limits, we perform measurements on a third membrane of a larger diameter (12 $\mu m$) suspended over a 285 nm cavity. The larger diameter offers two advantages. First, the increased oscillation amplitude facilitates the detection of higher harmonics, enabling calibration of the displacement using the method proposed by Dolleman \textit{et al.}~\cite{Dolleman2017AmplitudeCalibration}. Second, the lower mechanical stiffness makes the resonator more susceptible to photothermal tuning, thereby enhancing the optomechanical effects predicted by the model.
As shown in Fig.~\ref{fig:fig4}a, the resonance frequency exhibits the same non-monotonic dependence on optical power observed for the 8 $\mu$m membrane suspended above a 285 nm cavity (Fig. \ref{fig:fig3}c), confirming that the larger membrane also undergoes photothermally induced buckling. Owing to the enhanced photothermal response, the nonlinear stiffness now undergoes a transition from hardening ($\beta_{\mathrm{eff}}>0$) to softening ($\beta_{\mathrm{eff}}<0$) nonlinear dynamics as the red laser power is increased (Fig.~\ref{fig:fig4}b). At an intermediate power of approximately 2.7 mW, the resonance response becomes linear, indicating that the photothermally induced softening compensates the intrinsic geometric hardening of the membrane, resulting in an almost vanishing effective Duffing coefficient. As in Figs. \ref{fig:fig3}b \& \ref{fig:fig3}f, the effective Duffing coefficient extracted from the nonlinear response curves decreases approximately linearly with increasing optical power (Fig.~\ref{fig:fig4}c), in agreement with the analytical model. At higher powers, however, the measurements deviate from the linear trend. This deviation coincides with the onset of a pronounced shift in the equilibrium position of the membrane (Fig.~\ref{fig:fig4}d), confirming that buckling becomes increasingly important and modifies the resonance frequency, thereby enhancing the photothermal softening predicted by Eq.~\ref{eqn:6}.\\
In summary, we have established both theoretically and experimentally that optical interferometric readout and actuation in suspended 2D material resonators give rise to an optomechanical interaction that directly modifies their mechanical properties through photothermal heating. By controlling the optical power, we achieve active tuning of both the linear stiffness and the cubic nonlinear stiffness of graphene resonators, including continuous transitions between hardening, linear, and softening nonlinear dynamics. Furthermore, we show that photothermal heating can induce or enhance buckling in membranes with low pretension, providing an additional physical knob for tailoring their nonlinear response. Beyond providing an in situ means to compensate for fabrication-induced variations in membrane tension and nonlinear stiffness, the proposed analytical framework allows intrinsic mechanical properties of graphene resonators to be distinguished from photothermal effects introduced by the optical readout and actuation. More broadly, our results establish optical interferometry not only as a transduction technique but also as a versatile tool for engineering and controlling the dynamics of low-dimensional nanomechanical resonators.
\begin{acknowledgement}
A.M. acknowledges the funding received from the European Union through a postdoctoral fellowship within the framework of the Marie Sklodowska-Curie Actions (MSCA-PF, PROPHOTOM, ID: 101104522). P.G.S. acknowledges funding from the European Union under the Horizon Europe Programme, Grant Agreement No. 101136388. F.A. further acknowledges funding from the European Union (ERC Consolidator, NCANTO, 101125458). Views and opinions expressed are, however, those of the authors only and do not necessarily reflect those of the European Union or the European Research Council. Neither the European Union nor the granting authority can be held responsible for them. The authors also acknowledge SoundCell B.V. for providing the graphene samples for the measurements.
\end{acknowledgement}
\section{Data Availability}
All the experimental data included in this paper are published in the open repository \href{https://data.4tu.nl/}{4TU.ResearchData} under the title "Optomechanical tuning of nonlinearity in graphene resonators" and can be freely accessed and used under a \href{https://creativecommons.org/licenses/by-nc-sa/4.0/deed.en}{CC BY-NC-SA 4.0} license.
\section{Authors contribution}
A.M. conceived and designed the work. A.M. and F.A. developed the analytical model. A.M. performed the experiments and wrote the manuscript with input from all authors. All authors analyzed the data, discussed the results and commented on the manuscript.
\bibliography{references}
\newpage
\begin{figure}
\centering
\includegraphics[width=0.5\linewidth]{Figures/Fig_1.pdf}
\caption{\textbf{Optomechanical interaction. a)} Schematic of the device made of a suspended graphene membrane over a silicon substrate with its optical and mechanical mode. \textbf{b)} Temperature of the membrane calculated as a function of the position. \textbf{c)} Resonance frequency of the membrane calculated as a function of the position. \textbf{d)} Duffing term of the membrane calculated as a function of the equilibrium position }
\label{fig:fig1}
\end{figure}
\newpage
\begin{figure}[H]
\centering
\includegraphics[width=0.5\linewidth]{Figures/Fig_2.pdf}
\caption{\textbf{Experimental setup.} \textbf{a)} Microscopy image of the devices used. Drum 1: 8 $\mu m$ diameter and 420 nm cavity depth depth. Drum 2: 8 $\mu m$ diameter and 285 nm cavity Drum 3: 12 $\mu m$ diameter and 285 nm cavity depth. \textbf{b)} Schematic of the setup used in the experiments. Legend: HeNe. Helium-Neon laser. NVF. Neutral variable density filter. PBS. Polarizing beam splitter. BPF. Optical bandpass filter. PD. Photodetector. BDL. Blue diode laser. DM. Dichroic mirror. $\lambda /4)$. $\lambda /4$ waveplate. Obj. Objective. SPA. Spectrum analyzer. VC. Vacuum chamber. SA. Sample \textbf{c)} Brownian motion spectra acquired for Drum 1 at different red laser powers (PSD: power spectral density). \textbf{d)} Driven mechanical response for Drum 1 showing a nonlinear response. \textbf{e)} First three harmonics measured for Drum 3.}
\label{fig:fig2}
\end{figure}
\newpage
\begin{figure}
\centering
\includegraphics[width=0.5\linewidth]{Figures/Fig_3.pdf}
\caption{\textbf{Stiffness as a function of power. a)} Resonance frequency shift measured for the 8 $\mu$m diameter membrane suspended over a 420 nm cavity depth (Drum 1) as a function of the power of the red (open circles) and the blue laser (open squares) and their respective fits to the proposed model (solid lines). \textbf{b)} Effective Duffing parameter measured for the previous membrane as a function of the red laser power (open circles) and its fit to a line (solid line) \textbf{c)} Resonance frequency shift measured for the 8 $\mu$m diameter membrane suspended over a 285 nm cavity depth (Drum 2) as a function of the power of the red (open circles) and the blue laser (open squares) and their respective fits to the proposed model (solid lines). \textbf{d)} Reflected light intensity mapped in Drum 2. \textbf{e)} Deflection of both membranes measured at their center as a function of the red laser power analyzing the reflected light intensity. \textbf{f)} Effective Duffing parameter measured for the previous membrane as a function of the red laser power (open circles) for Drum 2.}
\label{fig:fig3}
\newpage
\end{figure}
\begin{figure}[H]
\centering
\includegraphics[width=\linewidth]{Figures/Fig_4.pdf}
\caption{\textbf{Cubic stiffness switching. a)} Mechanical response measured for a 12 $\mu m$ diameter resonator suspended over a 285 nm cavity (Drum 3) as a function of the power of the red laser. White dotted lines represent the data shown in the next plot \textbf{b)} Mechanical response of the membranes for 2 mW, 2.7 mW and 4 mW. \textbf{c)} Cubic stiffness as a function of the power of the red laser obtained from the fitting of the previous data. \textbf{d)} Equilibrium position of the membrane measured as a function of the power of the red laser.}
\label{fig:fig4}
\end{figure}
%\include{Supplementary/supplementary_information}
\end{document}
\providecommand{\latin}[1]{#1}
\makeatletter
\providecommand{\doi}
{\begingroup\let\do\@makeother\dospecials
\catcode`\{=1 \catcode`\}=2 \doi@aux}
\providecommand{\doi@aux}[1]{\endgroup\texttt{#1}}
\makeatother
\providecommand*\mcitethebibliography{\thebibliography}
\csname @ifundefined\endcsname{endmcitethebibliography} {\let\endmcitethebibliography\endthebibliography}{}
\begin{mcitethebibliography}{28}
\providecommand*\natexlab[1]{#1}
\providecommand*\mciteSetBstSublistMode[1]{}
\providecommand*\mciteSetBstMaxWidthForm[2]{}
\providecommand*\mciteBstWouldAddEndPuncttrue
{\def\EndOfBibitem{\unskip.}}
\providecommand*\mciteBstWouldAddEndPunctfalse
{\let\EndOfBibitem\relax}
\providecommand*\mciteSetBstMidEndSepPunct[3]{}
\providecommand*\mciteSetBstSublistLabelBeginEnd[3]{}
\providecommand*\EndOfBibitem{}
\mciteSetBstSublistMode{f}
\mciteSetBstMaxWidthForm{subitem}{(\alph{mcitesubitemcount})}
\mciteSetBstSublistLabelBeginEnd
{\mcitemaxwidthsubitemform\space}
{\relax}
{\relax}
\bibitem[Tadokoro and Tanaka(2021)Tadokoro, and Tanaka]{Tadokoro2021HighlySensitive}
Tadokoro,~Y.; Tanaka,~H. Highly Sensitive Implementation of Logic Gates with a Nonlinear Nanomechanical Resonator. \emph{Physical Review Applied} \textbf{2021}, \emph{15}, 024058\relax
\mciteBstWouldAddEndPuncttrue
\mciteSetBstMidEndSepPunct{\mcitedefaultmidpunct}
{\mcitedefaultendpunct}{\mcitedefaultseppunct}\relax
\EndOfBibitem
\bibitem[Bachtold \latin{et~al.}(2022)Bachtold, Moser, and Dykman]{bachtold2022mesoscopic}
Bachtold,~A.; Moser,~J.; Dykman,~M. Mesoscopic physics of nanomechanical systems. \emph{Reviews of Modern Physics} \textbf{2022}, \emph{94}, 045005\relax
\mciteBstWouldAddEndPuncttrue
\mciteSetBstMidEndSepPunct{\mcitedefaultmidpunct}
{\mcitedefaultendpunct}{\mcitedefaultseppunct}\relax
\EndOfBibitem
\bibitem[Binnig \latin{et~al.}(1986)Binnig, Quate, and Gerber]{Binnig1986AFM}
Binnig,~G.; Quate,~C.~F.; Gerber,~C. Atomic Force Microscope. \emph{Physical Review Letters} \textbf{1986}, \emph{56}, 930--933\relax
\mciteBstWouldAddEndPuncttrue
\mciteSetBstMidEndSepPunct{\mcitedefaultmidpunct}
{\mcitedefaultendpunct}{\mcitedefaultseppunct}\relax
\EndOfBibitem
\bibitem[Mertens(2003)]{Mertens2003TemperaturePressure}
Mertens,~J. e.~a. Effects of temperature and pressure on microcantilever resonance response. \emph{Ultramicroscopy} \textbf{2003}, \emph{97}, 119--126\relax
\mciteBstWouldAddEndPuncttrue
\mciteSetBstMidEndSepPunct{\mcitedefaultmidpunct}
{\mcitedefaultendpunct}{\mcitedefaultseppunct}\relax
\EndOfBibitem
\bibitem[Khan \latin{et~al.}(2014)Khan, Knowles, Dennison, Ghoraishi, and Thundat]{10.1063/1.4889744}
Khan,~M.~F.; Knowles,~B.; Dennison,~C.~R.; Ghoraishi,~M.~S.; Thundat,~T. Pressure modulated changes in resonance frequency of microchannel string resonators. \emph{Applied Physics Letters} \textbf{2014}, \emph{105}, 013507\relax
\mciteBstWouldAddEndPuncttrue
\mciteSetBstMidEndSepPunct{\mcitedefaultmidpunct}
{\mcitedefaultendpunct}{\mcitedefaultseppunct}\relax
\EndOfBibitem
\bibitem[Martín-Pérez and Ramos(2025)Martín-Pérez, and Ramos]{MARTINPEREZ2025117161}
Martín-Pérez,~A.; Ramos,~D. Enhancing cell characterization via hydrodynamic compression in suspended microchannel resonators. \emph{Sensors and Actuators A: Physical} \textbf{2025}, \emph{396}, 117161\relax
\mciteBstWouldAddEndPuncttrue
\mciteSetBstMidEndSepPunct{\mcitedefaultmidpunct}
{\mcitedefaultendpunct}{\mcitedefaultseppunct}\relax
\EndOfBibitem
\bibitem[Malvar(2016)]{Malvar2016MassStiffness}
Malvar,~O. e.~a. Mass and stiffness spectrometry of nanoparticles and whole intact bacteria by multimode nanomechanical resonators. \emph{Nature Communications} \textbf{2016}, \emph{7}, 13452\relax
\mciteBstWouldAddEndPuncttrue
\mciteSetBstMidEndSepPunct{\mcitedefaultmidpunct}
{\mcitedefaultendpunct}{\mcitedefaultseppunct}\relax
\EndOfBibitem
\bibitem[Kirchhof(2023)]{Kirchhof2023NanomechanicalAbsorption}
Kirchhof,~J. N. e.~a. Nanomechanical absorption spectroscopy of 2D materials with femtowatt sensitivity. \emph{2D Materials} \textbf{2023}, \emph{10}, 035012\relax
\mciteBstWouldAddEndPuncttrue
\mciteSetBstMidEndSepPunct{\mcitedefaultmidpunct}
{\mcitedefaultendpunct}{\mcitedefaultseppunct}\relax
\EndOfBibitem
\bibitem[Chien \latin{et~al.}(2018)Chien, Brameshuber, Rossboth, Schütz, and Schmid]{Chien2018SingleMolecule}
Chien,~M.-H.; Brameshuber,~M.; Rossboth,~B.~K.; Schütz,~G.~J.; Schmid,~S. Single-molecule optical absorption imaging by nanomechanical photothermal sensing. \emph{Proceedings of the National Academy of Sciences} \textbf{2018}, \emph{115}, 11150--11155\relax
\mciteBstWouldAddEndPuncttrue
\mciteSetBstMidEndSepPunct{\mcitedefaultmidpunct}
{\mcitedefaultendpunct}{\mcitedefaultseppunct}\relax
\EndOfBibitem
\bibitem[Naik \latin{et~al.}(2009)Naik, Hanay, Hiebert, Feng, and Roukes]{Naik2009SingleMoleculeMS}
Naik,~A.~K.; Hanay,~M.~S.; Hiebert,~W.~K.; Feng,~X.~L.; Roukes,~M.~L. Towards single-molecule nanomechanical mass spectrometry. \emph{Nature Nanotechnology} \textbf{2009}, \emph{4}, 445--450\relax
\mciteBstWouldAddEndPuncttrue
\mciteSetBstMidEndSepPunct{\mcitedefaultmidpunct}
{\mcitedefaultendpunct}{\mcitedefaultseppunct}\relax
\EndOfBibitem
\bibitem[Chaste \latin{et~al.}(2012)Chaste, Eichler, Moser, Ceballos, Rurali, and Bachtold]{Chaste2012Yoctogram}
Chaste,~J.; Eichler,~A.; Moser,~J.; Ceballos,~G.; Rurali,~R.; Bachtold,~A. A nanomechanical mass sensor with yoctogram resolution. \emph{Nature Nanotechnology} \textbf{2012}, \emph{7}, 301--304\relax
\mciteBstWouldAddEndPuncttrue
\mciteSetBstMidEndSepPunct{\mcitedefaultmidpunct}
{\mcitedefaultendpunct}{\mcitedefaultseppunct}\relax
\EndOfBibitem
\bibitem[Rugar \latin{et~al.}(2004)Rugar, Budakian, Mamin, and Chui]{Rugar2004SingleSpin}
Rugar,~D.; Budakian,~R.; Mamin,~H.~J.; Chui,~B.~W. Single spin detection by magnetic resonance force microscopy. \emph{Nature} \textbf{2004}, \emph{430}, 329--332\relax
\mciteBstWouldAddEndPuncttrue
\mciteSetBstMidEndSepPunct{\mcitedefaultmidpunct}
{\mcitedefaultendpunct}{\mcitedefaultseppunct}\relax
\EndOfBibitem
\bibitem[Steeneken \latin{et~al.}(2021)Steeneken, Dolleman, Davidovikj, Alijani, and van~der Zant]{Steeneken2021Dynamics2D}
Steeneken,~P.~G.; Dolleman,~R.~J.; Davidovikj,~D.; Alijani,~F.; van~der Zant,~H. S.~J. Dynamics of 2D material membranes. \emph{2D Materials} \textbf{2021}, \emph{8}, 042001\relax
\mciteBstWouldAddEndPuncttrue
\mciteSetBstMidEndSepPunct{\mcitedefaultmidpunct}
{\mcitedefaultendpunct}{\mcitedefaultseppunct}\relax
\EndOfBibitem
\bibitem[Xu \latin{et~al.}(2022)Xu, Zhang, Zhu, Liu, Eichler, Zheng, Lee, Dash, More, Wu, \latin{et~al.} others]{xu2022nanomechanical}
Xu,~B.; Zhang,~P.; Zhu,~J.; Liu,~Z.; Eichler,~A.; Zheng,~X.-Q.; Lee,~J.; Dash,~A.; More,~S.; Wu,~S.; others Nanomechanical resonators: toward atomic scale. \emph{Acs Nano} \textbf{2022}, \emph{16}, 15545--15585\relax
\mciteBstWouldAddEndPuncttrue
\mciteSetBstMidEndSepPunct{\mcitedefaultmidpunct}
{\mcitedefaultendpunct}{\mcitedefaultseppunct}\relax
\EndOfBibitem
\bibitem[Li \latin{et~al.}(2024)Li, Xu, Norte, Aragón, Steeneken, and Alijani]{Li2024StrainEngineering}
Li,~Z.; Xu,~M.; Norte,~R.~A.; Aragón,~A.~M.; Steeneken,~P.~G.; Alijani,~F. Strain engineering of nonlinear nanoresonators from hardening to softening. \emph{Communications Physics} \textbf{2024}, \emph{7}, 53\relax
\mciteBstWouldAddEndPuncttrue
\mciteSetBstMidEndSepPunct{\mcitedefaultmidpunct}
{\mcitedefaultendpunct}{\mcitedefaultseppunct}\relax
\EndOfBibitem
\bibitem[Yousuf \latin{et~al.}(2025)Yousuf, Wang, Rudawski, and Feng]{Yousuf2025GrapheneTrampoline}
Yousuf,~S. M. E.~H.; Wang,~Y.; Rudawski,~N.~G.; Feng,~P. X.~L. Graphene Trampoline Nanomechanical Resonators with Very High Quality Factors and Broad Dynamic Ranges. \emph{Advanced Functional Materials} \textbf{2025}, \emph{35}, e11158\relax
\mciteBstWouldAddEndPuncttrue
\mciteSetBstMidEndSepPunct{\mcitedefaultmidpunct}
{\mcitedefaultendpunct}{\mcitedefaultseppunct}\relax
\EndOfBibitem
\bibitem[Sajadi \latin{et~al.}(2017)Sajadi, Alijani, Davidovikj, Goosen, Steeneken, and van Keulen]{sajadi2017experimental}
Sajadi,~B.; Alijani,~F.; Davidovikj,~D.; Goosen,~J.~H.; Steeneken,~P.~G.; van Keulen,~F. Experimental characterization of graphene by electrostatic resonance frequency tuning. \emph{Journal of Applied Physics} \textbf{2017}, \emph{122}\relax
\mciteBstWouldAddEndPuncttrue
\mciteSetBstMidEndSepPunct{\mcitedefaultmidpunct}
{\mcitedefaultendpunct}{\mcitedefaultseppunct}\relax
\EndOfBibitem
\bibitem[Samanta \latin{et~al.}(2018)Samanta, Arora, and Naik]{samanta2018tuning}
Samanta,~C.; Arora,~N.; Naik,~A. Tuning of geometric nonlinearity in ultrathin nanoelectromechanical systems. \emph{Applied Physics Letters} \textbf{2018}, \emph{113}\relax
\mciteBstWouldAddEndPuncttrue
\mciteSetBstMidEndSepPunct{\mcitedefaultmidpunct}
{\mcitedefaultendpunct}{\mcitedefaultseppunct}\relax
\EndOfBibitem
\bibitem[Davidovikj \latin{et~al.}(2018)Davidovikj, Poot, Cartamil-Bueno, Van Der~Zant, and Steeneken]{davidovikj2018chip}
Davidovikj,~D.; Poot,~M.; Cartamil-Bueno,~S.~J.; Van Der~Zant,~H.~S.; Steeneken,~P.~G. On-chip heaters for tension tuning of graphene nanodrums. \emph{Nano letters} \textbf{2018}, \emph{18}, 2852--2858\relax
\mciteBstWouldAddEndPuncttrue
\mciteSetBstMidEndSepPunct{\mcitedefaultmidpunct}
{\mcitedefaultendpunct}{\mcitedefaultseppunct}\relax
\EndOfBibitem
\bibitem[{\v{S}}i{\v{s}}kins \latin{et~al.}(2025){\v{S}}i{\v{s}}kins, Ke{\c{s}}kekler, Houmes, Ma{\~n}as-Valero, Koperski, Coronado, Blanter, van~der Zant, Steeneken, and Alijani]{vsivskins2025nonlinear}
{\v{S}}i{\v{s}}kins,~M.; Ke{\c{s}}kekler,~A.; Houmes,~M.~J.; Ma{\~n}as-Valero,~S.; Koperski,~M.; Coronado,~E.; Blanter,~Y.~M.; van~der Zant,~H.~S.; Steeneken,~P.~G.; Alijani,~F. Nonlinear dynamics and magneto-elasticity of nanodrums near the phase transition. \emph{Nature Communications} \textbf{2025}, \emph{16}, 2177\relax
\mciteBstWouldAddEndPuncttrue
\mciteSetBstMidEndSepPunct{\mcitedefaultmidpunct}
{\mcitedefaultendpunct}{\mcitedefaultseppunct}\relax
\EndOfBibitem
\bibitem[Sarafraz \latin{et~al.}(2024)Sarafraz, Liu, Cvetanovi{\'c}, Spasenovi{\'c}, Vollebregt, Manzaneque~Garcia, Steeneken, Alijani, and Verbiest]{sarafraz2024quantifying}
Sarafraz,~A.; Liu,~H.; Cvetanovi{\'c},~K.; Spasenovi{\'c},~M.; Vollebregt,~S.; Manzaneque~Garcia,~T.; Steeneken,~P.~G.; Alijani,~F.; Verbiest,~G.~J. Quantifying stress distribution in ultra-large graphene drums through mode shape imaging. \emph{npj 2D Materials and Applications} \textbf{2024}, \emph{8}, 45\relax
\mciteBstWouldAddEndPuncttrue
\mciteSetBstMidEndSepPunct{\mcitedefaultmidpunct}
{\mcitedefaultendpunct}{\mcitedefaultseppunct}\relax
\EndOfBibitem
\bibitem[Aguila(2022)]{Aguila2022PhotothermalResponsivity}
Aguila,~M. A. e.~a. Photothermal Responsivity of van der Waals Material-Based Nanomechanical Resonators. \emph{Nanomaterials} \textbf{2022}, \emph{12}, 2675\relax
\mciteBstWouldAddEndPuncttrue
\mciteSetBstMidEndSepPunct{\mcitedefaultmidpunct}
{\mcitedefaultendpunct}{\mcitedefaultseppunct}\relax
\EndOfBibitem
\bibitem[Barton \latin{et~al.}(2012)Barton, Storch, Adiga, Sakakibara, Cipriany, Ilic, Wang, Ong, McEuen, Parpia, \latin{et~al.} others]{barton2012photothermal}
Barton,~R.~A.; Storch,~I.~R.; Adiga,~V.~P.; Sakakibara,~R.; Cipriany,~B.~R.; Ilic,~B.; Wang,~S.~P.; Ong,~P.; McEuen,~P.~L.; Parpia,~J.~M.; others Photothermal self-oscillation and laser cooling of graphene optomechanical systems. \emph{Nano letters} \textbf{2012}, \emph{12}, 4681--4686\relax
\mciteBstWouldAddEndPuncttrue
\mciteSetBstMidEndSepPunct{\mcitedefaultmidpunct}
{\mcitedefaultendpunct}{\mcitedefaultseppunct}\relax
\EndOfBibitem
\bibitem[Davidovikj \latin{et~al.}(2017)Davidovikj, Alijani, Cartamil-Bueno, van~der Zant, Amabili, and Steeneken]{davidovikj2017nonlinear}
Davidovikj,~D.; Alijani,~F.; Cartamil-Bueno,~S.~J.; van~der Zant,~H.~S.; Amabili,~M.; Steeneken,~P.~G. Nonlinear dynamic characterization of two-dimensional materials. \emph{Nature Communications} \textbf{2017}, \emph{8}, 1253\relax
\mciteBstWouldAddEndPuncttrue
\mciteSetBstMidEndSepPunct{\mcitedefaultmidpunct}
{\mcitedefaultendpunct}{\mcitedefaultseppunct}\relax
\EndOfBibitem
\bibitem[Li \latin{et~al.}(2026)Li, Xu, Norte, Arag{\'o}n, Steeneken, and Alijani]{li2026cascade}
Li,~Z.; Xu,~M.; Norte,~R.~A.; Arag{\'o}n,~A.~M.; Steeneken,~P.~G.; Alijani,~F. Cascade of Modal Interactions in Nanomechanical Resonators with Soft Clamping. \emph{Physical review letters} \textbf{2026}, \emph{136}, 097202\relax
\mciteBstWouldAddEndPuncttrue
\mciteSetBstMidEndSepPunct{\mcitedefaultmidpunct}
{\mcitedefaultendpunct}{\mcitedefaultseppunct}\relax
\EndOfBibitem
\bibitem[Dolleman \latin{et~al.}(2017)Dolleman, Davidovikj, van~der Zant, and Steeneken]{Dolleman2017AmplitudeCalibration}
Dolleman,~R.~J.; Davidovikj,~D.; van~der Zant,~H. S.~J.; Steeneken,~P.~G. Amplitude calibration of 2D mechanical resonators by nonlinear optical transduction. \emph{Applied Physics Letters} \textbf{2017}, \emph{111}\relax
\mciteBstWouldAddEndPuncttrue
\mciteSetBstMidEndSepPunct{\mcitedefaultmidpunct}
{\mcitedefaultendpunct}{\mcitedefaultseppunct}\relax
\EndOfBibitem
\bibitem[Liu \latin{et~al.}(2024)Liu, Baglioni, Boix-Constant, van~der Zant, Steeneken, and Verbiest]{Liu_2024}
Liu,~H.; Baglioni,~G.; Boix-Constant,~C.; van~der Zant,~H. S.~J.; Steeneken,~P.~G.; Verbiest,~G.~J. Enhanced sensitivity and tunability of thermomechanical resonance near the buckling bifurcation. \emph{2D Materials} \textbf{2024}, \emph{11}, 025028\relax
\mciteBstWouldAddEndPuncttrue
\mciteSetBstMidEndSepPunct{\mcitedefaultmidpunct}
{\mcitedefaultendpunct}{\mcitedefaultseppunct}\relax
\EndOfBibitem
\end{mcitethebibliography}
\providecommand{\latin}[1]{#1}
\makeatletter
\providecommand{\doi}
{\begingroup\let\do\@makeother\dospecials
\catcode`\{=1 \catcode`\}=2 \doi@aux}
\providecommand{\doi@aux}[1]{\endgroup\texttt{#1}}
\makeatother
\providecommand*\mcitethebibliography{\thebibliography}
\csname @ifundefined\endcsname{endmcitethebibliography} {\let\endmcitethebibliography\endthebibliography}{}
\begin{mcitethebibliography}{3}
\providecommand*\natexlab[1]{#1}
\providecommand*\mciteSetBstSublistMode[1]{}
\providecommand*\mciteSetBstMaxWidthForm[2]{}
\providecommand*\mciteBstWouldAddEndPuncttrue
{\def\EndOfBibitem{\unskip.}}
\providecommand*\mciteBstWouldAddEndPunctfalse
{\let\EndOfBibitem\relax}
\providecommand*\mciteSetBstMidEndSepPunct[3]{}
\providecommand*\mciteSetBstSublistLabelBeginEnd[3]{}
\providecommand*\EndOfBibitem{}
\mciteSetBstSublistMode{f}
\mciteSetBstMaxWidthForm{subitem}{(\alph{mcitesubitemcount})}
\mciteSetBstSublistLabelBeginEnd
{\mcitemaxwidthsubitemform\space}
{\relax}
{\relax}
\bibitem[Bonaccorso \latin{et~al.}(2010)Bonaccorso, Sun, Hasan, and Ferrari]{Bonaccorso2010GraphenePhotonics}
Bonaccorso,~F.; Sun,~Z.; Hasan,~T.; Ferrari,~A.~C. Graphene photonics and optoelectronics. \emph{Nature Photonics} \textbf{2010}, \emph{4}, 611--622\relax
\mciteBstWouldAddEndPuncttrue
\mciteSetBstMidEndSepPunct{\mcitedefaultmidpunct}
{\mcitedefaultendpunct}{\mcitedefaultseppunct}\relax
\EndOfBibitem
\bibitem[Dolleman \latin{et~al.}(2017)Dolleman, Davidovikj, van~der Zant, and Steeneken]{Dolleman2017AmplitudeCalibration}
Dolleman,~R.~J.; Davidovikj,~D.; van~der Zant,~H. S.~J.; Steeneken,~P.~G. Amplitude calibration of 2D mechanical resonators by nonlinear optical transduction. \emph{Applied Physics Letters} \textbf{2017}, \emph{111}\relax
\mciteBstWouldAddEndPuncttrue
\mciteSetBstMidEndSepPunct{\mcitedefaultmidpunct}
{\mcitedefaultendpunct}{\mcitedefaultseppunct}\relax
\EndOfBibitem
\end{mcitethebibliography}