Label-Free Optical Nanoscopy of Single-Layer Graphene | ACS Nano

Aug 1, 2019 - Label-Free Optical Nanoscopy of Single-Layer Graphene .... Additional experimental results (PDF). pdf. nn9b05054_si_001.pdf (369.95 kb) ...
0 downloads 0 Views 1MB Size
Subscriber access provided by BUFFALO STATE

Article

Label-Free Optical Nanoscopy of Single Layer Graphene Giulia Zanini, Kseniya Korobchevskaya, Takahiro Deguchi, Alberto Diaspro, and Paolo Bianchini ACS Nano, Just Accepted Manuscript • DOI: 10.1021/acsnano.9b05054 • Publication Date (Web): 01 Aug 2019 Downloaded from pubs.acs.org on August 2, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

Label-Free Optical Nanoscopy of Single Layer Graphene Giulia Zanini,1,2 Kseniya Korobchevskaya,1† Takahiro Deguchi,1 Alberto Diaspro,1,2 and Paolo Bianchini1,* 1Nanoscopy and NIC@IIT, Istituto Italiano di Tecnologia, Via Morego 30, 16163 Genoa, Italy 2Department of Physics, University of Genoa, Via Dodecaneso 33, 16146 Genoa, Italy †current address: Kennedy Institute of Rheumatology, University of Oxford, Roosevelt Drive, Oxford OX3 7FY, UK *[email protected] ABSTRACT: The application of ultrafast pulsed laser sources and spectroscopic techniques is enabling label-free, deep-tissue optical microscopy. However, the circumvention of the diffraction limit in this field is still an open challenge. Among such approaches, pump-probe microscopy is of increasing interest thanks to its highly specific non-fluorescent-based contrast mechanisms for the imaging of material and life science samples. In this paper, a custom femtosecond-pulsed nearinfrared pump-probe microscope, which exploits transient absorption and stimulated Raman scattering interactions, is presented. The conventional pump-probe configuration is combined with a spatially shaped saturation pump beam, which allows for the reduction of the effective focal volume exploiting transient absorption saturation. By optimizing the acquisition parameters, such as power and temporal overlap of the saturation beam, we can image single layer graphene deposited on a glass surface at the nanoscale and with increased layer sensitivity. These results suggest that saturation pump-probe nanoscopy is a promising tool for label-free high-resolution imaging. KEYWORDS: single layer graphene, pump-probe microscopy, absorption saturation, superresolution, nanoscopy Optical microscopy has become an essential tool in the study of biological samples. So far, fluorescence microscopy techniques are the most widely used due to their ability to visualize the structures of interest with high contrast, high specificity, and high spatial and temporal resolution. Despite their well-established benefits, conventional fluorescence microscopy techniques rely on exogenous or gene-modified endogenous labels, which may alter the physical properties of the specimen, and often come at the cost of photobleaching and photodamage effects. With the development of ultrashort pulsed laser sources, non-linear optical (NLO) light-matter interactions became accessible and started to acquire a central role in optical microscopy for labelfree imaging. They broaden the range of accessible targets, imaging weakly or non-fluorescent samples and taking the contrast from intrinsic properties of the systems under study.1 Moreover, three-dimensional (3D) imaging can be achieved due to the intrinsic optical sectioning capabilities of NLO phenomena, while the use of longer wavelengths in the near-infrared (NIR) part of the spectrum results in lower absorption and scattering and permits to image deeper inside tissues. The most widely used non-linear contrast mechanisms are based on multiphoton excitation fluorescence (MPEF),2,3 frequency conversion (sum/difference frequency generation, SFG/DFG) and high harmonic generation (HHG),4 coherent Raman scattering (CRS),5 and four-wave mixing (FWM).6 Another type of NLO interaction finding application in optical microscopy is transient absorption, a phenomenon widely used in spectroscopy7,8 to probe the fast dynamics of the excited states of the

ACS Paragon Plus Environment

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 17

molecules or structures of interest. The technique is commonly known as pump-probe microscopy,9–12 where first a pump beam is tuned to excite the electronic transition, while a subsequent probe beam is used to investigate the generated transient excited state and produce a label-free contrast. The probe beam can undergo different types of NLO interactions, such as two-photon absorption (TPA), excited state absorption (ESA), ground state depletion (GSD), and stimulated emission (SE).10 These mechanisms lead to either a loss (TPA, ESA) or a gain (GSD, SE) in the transmitted probe beam, whose intensity change can be detected, pixel by pixel, with a laser-scanning microscope. The probe intensity variation is typically a few percent of the total transmitted intensity and it is buried in the background. To extract this pump-probe signal, a fast intensity modulation is added to the pump beam, introducing an analogous modulation in the transmitted probe, which can be filtered out with high sensitivity using a lock-in amplifier (see Fig. 1). Ultrashort pulsed laser beams with a proper spatial and temporal overlap are needed in order to achieve highly sensitive non-linear imaging, and in order to follow the fast dynamics of the electronic states with sub-picosecond temporal resolution. Transient absorption microscopy has been explored both in material and biological research to characterize carrier dynamics in materials,13–15 to image highly absorbing chromophores like melanin and hemoglobin,11,16–19 or to study the interaction of non-fluorescent nanomaterials as carriers within living cells.20–24 The main drawback of NLO microscopy techniques is their relatively poor spatial resolution, especially when using NIR excitation, which is physically limited by diffraction to approximately half of the excitation wavelength.25 Some super-resolution techniques inspired by fluorescence microscopy were successfully applied to label-free methods, like intensity weighted subtraction,26,27 saturated excitation28 and structured illumination29. A class of nanoscopy approaches is based on the identification of reversible saturable optical fluorescence transitions (RESOLFT) between two specific molecular states, and on the exploitation of a tailored saturation in order to shrink the emission volume down to tens of nanometers.30,31 The generalized RESOLFT concept was proposed for label-free transient absorption microscopy32,33 and recently applied to graphite nanoplatelets, exploiting pump absorption saturation to obtain sub-diffraction imaging capabilities.34 By superimposing a high-intensity non-modulated pump beam having a doughnut-shaped intensity distribution, the transient absorption is brought to saturation at the periphery of the focal spot. In this way the probe modulation is detected only in the central area of the focal spot, while the saturation effect at the periphery prevents the absorption of the probe beam which will be transmitted without any intensity modulation (see Fig. 1(a)). By tuning the intensity of the saturation pulse, the detection volume can be reduced below the diffraction limit. In this paper, a custom-made NIR transient absorption microscope is presented, and its sensitivity in imaging single layer graphene (SLG) and multilayer graphene defects is demonstrated. In order to obtain the optimal imaging parameters and achieve the maximum resolution enhancement, the absorption saturation of graphene is characterized as a function of temporal alignment and power of the saturation beam. Finally, the optimized parameters retrieved from this characterization are applied for imaging SLG defects, demonstrating a /10 resolution improvement and an increase in layer sensitivity.

ACS Paragon Plus Environment

Page 3 of 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

Figure 1. (a) Transition diagram of GSD (ground state depletion) transient absorption interaction and experimental pulse sequence at the very center of the focal spot (left) and in the presence of the saturation pump in the doughnut region (right). The dashed red line in the probe pulse graph represents the amplitude of the input probe beam. (b) Scheme of the custom-made NIR pump-probe setup. The pump beam optical path is shown in green, the probe one in red, and the saturation pump one in blue. OPO: optical parametric oscillator with pump and signal outputs, BS: beam splitter, EOM: electro-optical modulator, DM: dichroic mirror, VPP: vortex phase plate, HWP: half-wave plate (#1 used for power adjustment, #2 used for polarization control), QWP: quarter-wave plate, PBS: polarizing beam splitter, ND: neutral density filter, PMT: photomultiplier tube, SP: short-pass filter, LP: long-pass filter, PD: photodiode, NA: numerical aperture, sync: synchronization, trig: trigger, ref: reference, in: input, out: output, (R,X): modulus and in-phase component of the demodulated pump-probe signal, respectively. RESULTS AND DISCUSSION Single layer graphene (SLG) transient absorption imaging. A commercial SLG sample (details in Methods section) was characterized with the NIR pump-probe microscope setup shown in Fig. 1(b) and presented in the Methods section using 800 nm and 1030 nm as pump and probe wavelengths, respectively. Due to its particular linear electronic band structure, graphene shows a wavelength-independent, broadband optical absorption (~ 2.3% per layer35) in the NIR part of the spectrum.36 It is then expected that both pump and probe beams at the wavelengths used in these experiments are absorbed to the same extent and can be used to monitor the carrier population of graphene’s excited state. As presented in Fig. 2(a), graphene exhibits a low fluorescence quantum yield, therefore a fluorescence signal is detected mainly from multilayer regions of the sample while single layer regions remain dark. On the contrary, the introduction of a modulated pump beam modifies the sample transmission providing high signal-to-noise ratio (SNR) non-fluorescence-based pump-probe signal (Fig. 2(b,c)). Images were acquired with 4 mW pump beam at 800 nm and 1 mW probe beam at 1030 nm. Fig. 2(b) represents the in-phase component of the pump-probe signal, which is defined as X = R ∙ cos θ, where R is the signal modulus and θ is the output phase respect to the reference modulation. Positive X values denote a signal that is in-phase with the pump modulation (as sketched in Fig 1(a), left). The presence of such transient absorption contrast, together with the fact that the absorption ACS Paragon Plus Environment

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 17

coefficient is the same at both pump and probe wavelengths, suggests that the pump excitation causes an absorption reduction at the probe frequency, which can be attributed to a ground state depletion (GSD) interaction.

Figure 2. (a-c) Multimodal imaging of SLG: fluorescence image (a), pump-probe image demodulated in the in-phase channel X (b) and in the module channel R (c). Scale bar 5 μm. (d) Line profiles along the white dashed arrows in the transient absorption image in (c). (e-f) Log-log plots of the pumpprobe signal from SLG and multilayer defects as a function of pump (e) and probe (f) excitation powers, showing an initial linear behavior followed by signal saturation. The dashed and solid lines show the linear fits at low powers. (g) Time-resolved spectra of SLG and multilayer defects obtained at different delays of the probe pulse with respect to the pump pulse. The inset shows the pulse sequence sketch (pump pulse at t=0 in dashed green, probe pulse in solid red). The pump-probe signal was fitted with a double exponential decay (solid and dashed lines). The inset graph shows the temporal resolution of the pump-probe microscope retrieved with a Gaussian fit. Fig. 2(c), instead, represents the modulus R of the pump-probe signal and provides information about the signal behavior as a function of the density of electronic transition and of the excitation powers. The line profiles in Fig. 2(d) show that the pump-probe signal intensity is linearly proportional to the number of graphene layers. A double layer defect with the characteristic hexagonal shape can be clearly distinguished in the image (contoured region in Fig. 2(c), line profile I), demonstrating the single layer sensitivity of the system. The linearity of the signal intensity with respect to the number of layers is not fulfilled for a multilayer area (line profile II), probably due to the presence of additional layers of sub-resolved lateral size whose real signal intensity is averaged down to smaller values. The pump-probe signal also exhibits a linear relationship with respect to the applied pump and probe powers at relatively low beam intensity regime (Fig. 2(e,f)). The linearity is present in both SLG and its multilayer defects, giving slopes close to 1 in the log-log plots. In particular, the linear fits present slopes of 0.80 ± 0.10 and 0.72 ± 0.03 for the pump power curves in Fig. 2(e) (for single layer and ACS Paragon Plus Environment

Page 5 of 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

multilayer defects, respectively), and slopes of 1.1 ± 0.1 in both probe power curves in Fig. 2(f). At higher incident powers, the state filling effect of the higher electronic states is expected to occur as a consequence of the Pauli exclusion principle, and the absorption will saturate.37 Such broadband saturable absorption behavior of graphene-based materials have been previously investigated38 and reported for single and multilayer graphene.39 Looking at the graphs in Fig. 2(e,f), signal saturation starts to appear around an average power density of 2 MW/cm2 for both pump and probe beams, which corresponds to a power around 10 mW for the pump and 7 mW for the probe. Graphene exhibits an ultrafast excited state dynamics,40 with a lifetime that is much shorter (~ps) than the typical lifetime of a fluorescent molecule (~ns). This behavior can be observed by acquiring timeresolved spectra monitoring the pump-probe signal at different pump-probe delays. The pump-probe signal is maximum at zero delay when pump and probe pulses are temporally overlapped at the focus, while at larger probe delays it exponentially decreases. In order to follow this ultrafast dynamics, a sub-picosecond temporal resolution is required. Using the SRS response from lipids, as explained in the Methods section, a temporal resolution of (380 ± 10) fs is obtained (see inset graph in Fig. 2(g)). Looking at the acquired time-resolved spectra of SLG in Fig. 2(g), the double exponential fits (solid and dashed lines) retrieves a fast component with (0.38 ± 0.03) ps and (0.40 ± 0.10) ps lifetime (for single layer and multilayer defects, respectively), together with a slow component of (1.9 ± 0.2) ps and (1.6 ± 0.2) ps lifetime (for single layer and multilayer defects, respectively). The fast time constant cannot be correctly retrieved because it is beyond the temporal resolution of the system, while the slow component lies in the picosecond regime and shows a slightly faster behavior in the presence of multilayer defects. These results are in good agreement with previously published works on graphene,14,34 where the slow decay component was attributed to carrier–phonon interaction and the fast unresolved component to carrier-carrier recombination processes (sub-100 fs scale). The outlined optical properties - the broadband strong and saturable absorption, and the ultrafast carrier dynamics - make graphene a suitable candidate to explore the saturation approach for nanoscopy NIR transient absorption imaging. Absorption saturation efficiency in SLG. As shown in Fig. 2(e,f), the state filling effect at high incident powers leads to saturation of the absorption, which is reflected in the decrease of the detected pump-probe signal. If the saturation is caused by an additional non-modulated pump beam, the pumpprobe signal is brought to zero because the lock-in detection won’t identify any modulation of the probe. This effect can be exploited similarly to the stimulated emission depletion (STED) microscopy principle,30 using a doughnut-shaped non-modulated saturation beam to reduce the point spread function to sub-diffraction dimensions34 (see Fig. 1(a)). To evaluate the pump-probe signal suppression efficiency, the SLG was imaged using a nonmodulated Gaussian beam at the pump wavelength (the so-called saturation beam). The effect was studied as a function of the temporal alignment of the saturation pulse respect to the pump one (whose temporal position defines the “time zero”), and as a function of the power of the saturation beam (Fig. 3).

ACS Paragon Plus Environment

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 17

Figure 3. (a) Pulse sketches of the two experimental acquisition modalities to investigate (i) nonsaturated and saturated time-resolved spectra (results in (b-d)), and (ii) the depletion efficiency at different saturation pulse delays (results in (e)) in SLG. The pump pulse is drawn in striped green, the probe pulse in red, and the saturation pulse in blue. The temporal position of the pump pulse defines the time zero. (b-d) Time-resolved spectra obtained without (○) and with (△) saturation pump at different time delays respect to the pump one and at different powers: (b) saturation beam at -0.4 ps and 10 mW, (c) saturation beam at 0 ps and 5 mW, (d) saturation beam at +0.8 ps and 20 mW. The colored data points in each graph represent the direct ratio between saturated and non-saturated data sets. (e) Pump-probe signal depletion obtained with acquisition modality (ii) and varying the saturation power from 5 to 20 mW, while keeping pump and probe powers fixed at 5 mW and 1 mW, respectively. The error bars on the x-axis reflect the error in moving and placing the manual delay line in the saturation pump optical path. (f-h) Absorption saturation depletion curves, which demonstrate the degree of signal suppression as a function of the saturation power and at different time delays between saturation and pump pulses: (f) -0.4 ps delay, (g) 0 ps delay, (h) +0.8 ps delay. The probe beam was kept temporally aligned with the pump one for maximum signal collection. The exponential decay of the depletion curves was fitted with equation (1). The temporal dependence was studied using two different signal acquisition modalities, as sketched in Fig. 3(a). The (i) investigates the non-saturated and saturated time-resolved spectra with the saturation pulse at three different time points (-0.4 ps, 0 ps, and +0.8 ps with respect to the time zero), in order to resolve the position of the saturation pulse in the dynamics of the excited states (as seen also in Wang et al.34). The (ii) investigates the depletion efficiency delaying the saturation pulse while keeping the probe pulse at time zero, in order to find the optimal temporal alignment of the saturation pulse. Results obtained with acquisition modality (i) are shown as empty black data points in Fig. 3(b-d) (○ for non-saturated and △ for saturated conditions) at saturation delays of -0.4 ps, 0 ps, and +0.8 ps. Data was acquired in SLG with pump at 5 mW, probe at 1 mW, and saturation pump at 10 mW (Fig. 3(b)), 5 mW (Fig. 3(c)) and 20 mW (Fig. 3(d)). Data was normalized with respect to the maximum of the non-saturated one. Data reported with full colored points represents the ratio of saturated and non-saturated signals at each delay, which shows the efficiency of the saturation beam in signal suppression.

ACS Paragon Plus Environment

Page 7 of 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

Looking at the ratio at each delay, it is possible to determine the position of the saturation beam with respect to the characteristics of the excited state dynamics. At the delay position where the signal suppression effect is the highest, and therefore the ratio has its minimum value, the saturation pulse is found. Its temporal position is highlighted by a shaded colored area, whose width is determined by the temporal resolution of the system (about 400 fs). In the condition where the pump-probe signal generation is at its maximum, which is at time zero of the time-resolved spectrum, the achieved signal suppression is demonstrated to be strongly dependent on the saturation delay and power. Notably, 50% of the signal suppression is obtained using just half of the saturation power when moving the saturation pulse from -0.4 ps to 0 ps. Interestingly, at a longer delay of 0.8 ps such depletion efficiency cannot be reached even by quadrupling the saturation power. The dependence of the saturation efficiency on the saturation pulse delay can also be studied using the signal acquisition modality (ii), and the results are shown in Fig. 3(e). The data represents the normalized pump-probe residual signal (at time zero) obtained at different saturation delays and for saturation powers ranging from 5 to 20 mW. Pump and probe beams were kept at 5 mW and 1 mW, respectively. The maximum depletion efficiency is always achieved when all three beams are temporally superimposed at time zero. The power dependence is more pronounced at time zero, where stronger depletion is achieved at higher power. Moving towards negative or positive delays the difference in depletion efficiency at different power gradually diminishes, highlighting a weaker power dependence. The power dependence of the saturation efficiency can be better understood by direct acquisition of the depletion curves at fixed delays of the saturation pulse (Fig. 3(f-h)), and fitting them with the following function derived from Wang et al.34 ∆𝑇𝑠𝑎𝑡

∆𝑇𝑢𝑛𝑠𝑎𝑡 =

1 1+

,

𝑃𝑠𝑎𝑡

(1)

𝑃0

where ∆𝑇𝑠𝑎𝑡 ∆𝑇𝑢𝑛𝑠𝑎𝑡 is the pump-probe signal fraction due to the saturation effect, 𝑃𝑠𝑎𝑡 is the power of the non-modulated saturation pump beam, and 𝑃0 is the saturation power at which the pump-probe signal is dropped to half. 𝑃0 is a characteristic parameter of the sample and can be used to evaluate the depletion efficiency of the process: the lower the value of 𝑃0, the higher the saturation efficiency. Data was acquired keeping the pump beam at 5 mW and the probe beam at 1 mW. By keeping the saturation temporally aligned with pump and probe pulses at time zero (Fig. 3(g)), a value of 𝑃0 = (4.2 ± 0.2) mW was retrieved, corresponding to a pump average power density of (0.89 ± 0.04) MW/cm2. This value is of the same order of magnitude as the saturation power value we obtained from graphs in Fig. 2(e), but it results to be three times higher than the value obtained in Wang et al.34 even if more similar to the values found by Bao et al.39 in graphene containing multiple defects. With the saturation pulse delay of -0.4 ps 𝑃0 increases to (15 ± 1) mW ((3.2 ± 0.2) MW/cm2) (Fig. 3(f)), while at 0.8 ps delay 𝑃0 increases up to a value of (42 ± 2) mW ((8.9 ± 0.4) MW/cm2) (Fig. 3(h)). Additionally, this confirms that the depletion efficiency is the highest when the saturation beam is superimposed with pump and probe beams, while it rapidly decreases with the saturation delay. The absorption saturation due to the non-modulated pump beam proves to be efficient in the signal suppression only if it happens within the sub-picosecond lifetime of the excited state in order to avoid state relaxation and probe absorption. Therefore, the temporal alignment of the saturation pulse at the

ACS Paragon Plus Environment

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 17

time zero needs to be set with a precision of the order of the system temporal resolution to guarantee optimized saturation performances. To further demonstrate the temporal delay dependence of the saturation efficiency, pump-probe images of sub-resolved graphene defects were acquired with overlapped pump and probe pulses and with the superimposition of the doughnut-shaped saturation beam (Fig. S1(a-c)). The pump beam was kept at 5 mW, while the probe beam at 1 mW. The saturation beam was kept at 5 mW and moved to different delays as shown in Fig. S1(f). The obtained images were compared with the conventional pump-probe data set acquired in the same conditions but in absence of the saturation beam (Fig. S1(d)). To evaluate the performance, the intensity line profiles for each condition were taken along the indicated arrows and they are shown in Fig. S1(e). When the saturation pulse is delayed with respect to the pump and probe pulses (Fig. S1(a,c)) the saturation power is insufficient to saturate the absorption, in agreement with the depletion curves in Fig. 3(f,h), and the line profiles (green and red) do not show resolution enhancement. On the contrary, when the delay is at zero (Fig. S1(b)), the saturation power is sufficient to induce saturation effects on the crest of the doughnut beam, which results in a narrowing of the intensity peaks (blue line profile in Fig. S1(e)). Super-resolution imaging of SLG at single layer sensitivity. After defining the optimal conditions for absorption saturation and suppression of the pump-probe signal, super-resolved imaging of SLG defects was performed using the doughnut-shaped saturation beam. The probe pulse was placed at time zero for maximum signal generation, and the saturation pulse was also placed at time zero to maximize signal suppression in the doughnut region. The saturation power was kept around 20 mW (1.5 MW/cm2) to obtain a signal suppression of about 70-80%, and in order to avoid sample damage. Pump and probe powers were kept around 2 mW. An example of non-saturated and saturated pump-probe images of SLG is presented in Fig. 4, and normalized line profiles drawn in two different sample regions (I and II) are compared. Images show a complex structure in the SLG, where a single layer covers the majority of the field of view with a uniform signal. Foldings created by multiple graphene layers produce stronger signals throughout the sample, while no signal is detected from cracks and holes in the structure. It is worth noting that comparing the images with and without the doughnut-shaped saturation beam, the saturated modality shows the expected decrease in signal intensity, together with a significant resolution enhancement. Line profiles across the arrows are drawn and are presented in the graphs as black (non-saturated) and red (saturated) dots. To quantify the spatial resolution enhancement, the line profiles were fitted with peak functions to obtain full-width-at-half-maximum (FWHM) values for both conditions. Intensity profiles were fitted with Lorentzian and Gaussian functions for saturated and non-saturated images respectively. A resolution improvement of almost /10 was obtained imaging with 1030 nm. In I, the folding shows a width of (430 ± 20) nm when imaged with the conventional pump-probe approach, compared with a width of (140 ± 20) nm obtained with the superimposition of the doughnut-shaped saturation beam. The technique allows accessing to higher information content, as presented in II where, through the saturation approach, a multilayer defect is resolved as composed of two subunits, which are resolved with a resolution of (90 ± 40) nm.

ACS Paragon Plus Environment

Page 9 of 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

Figure 4. Normalized pump-probe and saturated pump-probe images of SLG foldings and defects. Scale bar 2 μm. Zoomed regions I and II are also presented, and line profiles across the arrows are shown as black dots (non-saturated case, PP) and red dots (saturated case, SPP). Gaussian fits of the non-saturated data are shown as black solid lines, while Lorentzian single-peak (I) and double-peak (II) fits of the saturated data are shown as solid red lines. The dashed red lines in graph II highlight the single Lorentzian peaks retrieved from the analysis. The obtained resolution is marked in the graphs as FWHM of the fitted curves. Moreover, the nanoscopy method allows gathering more information in the layer structure of the sample, e.g. achieving an improved sensitivity in counting the number of layers. In Fig. 5(a,e), conventional and saturated pump-probe images of hexagonal multilayer defects grown on top of SLG are shown. The line profiles across a defect (marked by arrows in Fig. 5(a,e)) are plotted as black dots in Fig. 5(b,f), and show a continuous signal increase towards the center of the defect, suggesting a structure made by the stacking of many layers with decreasing dimensions. The background level is marked in purple in the graphs. The blue areas highlight the intensity range of the pump-probe signal coming from the single graphene layer. In the non-saturated case (Fig. 5(b)), only one additional layer can be distinguished looking at the steps in the intensity profile, and its intensity range is covered by the green area. In the saturated case (Fig. 5(f)), a higher discontinuity of the signal is present, and three more layers can be distinguished on top of the single one (green, yellow and orange areas). The intensity ranges selected in the graphs with the colored areas were then located in the images, and the masks reported in Fig. 5(c) and Fig. 5(g) were created to highlight the position of the layers in the sample. The purple background covers a hole in the graphene structure. The single layer (blue) covers the majority of the field of view, while the additional layers are found in a concentric geometry in the hexagonal defects and it can be seen that, especially in the saturated case, they maintain the hexagonal shape. The average intensity of the single layer was used as

ACS Paragon Plus Environment

ACS Nano

(b)

(c)

31013645

(d)

12011650 414580 0-413

0



2500

(e)

(f)

( g)

int. (a.u.) 31013645

( h)

18111950 13311580



661930 211400 0-210

0

sat. pump-probe

int. (a.u.)



pump-probe

(a)

3750

reference to count the additional layers, and the quantification results are reported in the graphs in Fig. 5(d,h) for the non-saturated and the saturated case, respectively. The graphs report the number of layers present in each colored region in the profile and in the image, as a function of the intensity ranges selected from the line profiles and written in the color bars of Fig. 5(c) and Fig. 5(g). The data follow a linear behavior, consistent with the theory of the pump-probe signal (see Fig. 2(d)). In the saturated case, the grey data points in Fig. 5(h) mark the second and the ninth layers, which are not revealed in Fig. 5(f) by the presence of steps in the line profile, but which can be deduced by the quantization of the adjacent layers. The results of the layers number quantification are reported in Table 1. In the saturated case more layers can be identified and correctly quantified with an overall lower error.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 17

Figure 5. (a) Pump-probe and (e) saturated pump-probe images of SLG hexagonal multilayer defects. Scale bar 5 μm. The line profiles across the arrows are shown as black dots in the graphs in (b) for the non-saturated case and in (f) for the saturated case. The purple area in both graphs highlight the background intensity values. The other colored areas mark the intensity ranges that were used to identify graphene layers and build the maps in (c) for the non-saturated case and in (g) for the saturated case, as reported in the color bars respectively. The first layer (marked in blue) was taken as reference to count the additional layers, and the quantification results are reported in the graphs in (d) and (h) for the non-saturated and saturated case, respectively.

ACS Paragon Plus Environment

Page 11 of 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

Graphene layers quantification pump-probe av. int. (a.u.)  SE color (bkg subtracted) blue 270  40

SE: standard error saturated pump-probe # layers  # layers  av. int. (a.u.)  SE color SE SE (bkg subtracted) blue 1.0  0.2 210  10 1.0  0.1 320  10 1.6  0.1 green green 1150  40 4.3  0.7 680  10 3.3  0.2 yellow 1310  10 6.3  0.4 orange 1740  20 8.4  0.5 1890  20 9.1  0.5 red red 3080  80 12  6 2060  20 10  1 Table 1. Comparison between layer quantification in the conventional and in the saturated pumpprobe imaging of SLG, as presented in Figure 6 and in particular in the graphs in Figure 6(d,h). CONCLUSION In this work, a custom-made multimodal NIR pump-probe microscope was presented. The setup is based on an OPO, pumped by a femtosecond Ti:sapphire laser, whose output pump and probe pulses are spatially and temporally overlapped in order to investigate the carrier population of the structures of interest. In addition, the generalized RESOLFT concept, widely exploited in fluorescence nanoscopy,30,31 was implemented in the pump-probe setup in order to achieve label-free subdiffraction imaging capabilities. In particular, the pump absorption saturation was identified as the optimal transition to be exploited for this purpose, using a doughnut-shaped saturation beam.34 The saturation properties and the consequent suppression of the pump-probe signal were extensively studied in single layer graphene (SLG). The suppression efficiency was shown to be exponentially dependent on the applied power, which can be kept as low as few MW/cm2 (10-20 mW). Since graphene exhibits an ultrafast excited state dynamics, also the temporal alignment of the saturation pulse plays a key role in obtaining efficient signal suppression at low powers. The precise temporal overlap between the pulses is a critical point for achieving maximum efficiency, as a displacement of the order of the system temporal response (which is in the sub-picosecond regime) dramatically decreases the signal suppression capabilities. The demonstrated ability to minimize the illumination powers by optimizing the temporal overlap is especially important for imaging samples with low damage threshold, and enable the use of this technique for biological studies. The engineering of the saturation beam in a doughnut shape permitted to achieve a resolution improvement up to /10 using 1030 nm probe wavelength and with limited applied powers, increasing the information content on the graphene defects structure. The saturation-based technique keeps the quantization of the pump-probe signal with the number of layers, increasing not only the spatial resolution but also the layer sensitivity. Moreover, since this imaging technique is implemented on a conventional laser scanning microscope, super-resolved imaging can be obtained on a large scale with short acquisition time (see Fig. S2). The presented results demonstrate the great potential of the pump-probe technique. Its high sensitivity, chemical specificity and high temporal and spatial resolution allow imaging of a variety of non-fluorescent chromophores and nanomaterials, which were previously inaccessible. Moreover,

ACS Paragon Plus Environment

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 17

the saturated pump-probe technique gives access to higher spatial information beyond the diffraction limit and can be applied to any samples showing a saturable absorption transition. In principle, this approach can also be extended to coherent Raman imaging of cellular components, exploiting the saturation properties of the stimulated Raman scattering process.41 With the implementation of NIR illumination, saturated pump-probe microscopy allows to image in scattering media and tissues, and it can be integrated into a multimodal non-linear microscopy platform, together with 2PEF, SHG, and SRS label-free techniques, in order to broaden its use in biology and material sciences. METHODS Pump-probe nanoscopy setup. The custom NIR pump-probe setup is realized by coupling a tunable mode-locked femtosecond pulsed Ti:sapphire laser (680-1080 nm, 80 MHz, 140 fs, Chameleon Ultra II, Coherent Inc., Santa Clara, CA, USA) with a laser scanning Nikon scan head and a Nikon microscope body (Nikon Instruments, Yokohama, Japan) (Fig. 1(b)). The Ti:sapphire laser pumps an optical parametric oscillator (Chameleon Compact OPO, Coherent Inc., Santa Clara, CA, USA) whose synchronized outputs (the pump at the Ti:sapphire wavelength in the range 740-880 mm, and the signal tunable in the range 1000-1600 nm) are locked at 80 MHz, and have a typical pulse width of 200 fs. The pump output is used as pump beam, and its intensity is modulated at 2 MHz by an electro-optic modulator (EOM, LM 0202, Qioptiq, Göttingen, Germany). The electronic synchronization between pulse rate and modulation is controlled and generated by a digital delay generator (DG645, Stanford Research Systems, Sunnyvale, CA, USA). The signal output is used as probe beam, and its pulses are temporally synchronized with the pump pulses using a delay line made by a motorized high-precision linear stage (M-521.PD, Physik Instrumente, Karlsruhe, Germany, delay up to 1.3 ns with 3 fs accuracy) equipped with a retroreflector (UBBR2.5-2I, Newport Corp., Irvine, CA, USA). The two laser beams are spatially combined with a dichroic mirror (DMSP950, Thorlabs, Newton, NJ, USA), and focused by a 60x 1.27NA water immersion objective (CFI Plan Apo IR, Nikon Instruments, Yokohama, Japan). The probe beam is collected in transmission by a 40x 0.6NA objective (CFI S Plan Fluor ELWD, Nikon Instruments, Yokohama, Japan), filtered by two long-pass filters (RG850, Schott, Mainz, Germany, and FEL1000, Thorlabs, Newton, NJ, USA) to reject pump light, and detected by an amplified InGaAs detector (PDA20CS, Thorlabs, Newton, NJ, USA). A phase-sensitive lock-in amplifier (HF2LI, Zurich Instrument, Zurich, Switzerland) is used to demodulate the probe signal and extract the pump-probe signal at the pump modulation frequency. Both module (R) and in-phase component (X = R ∙ cos θ, where θ is the output phase respect to the reference modulation) of the demodulated signal can be extracted from separate channels of the lock-in amplifier. The demodulated signals are then acquired by the Nikon controller and images are constructed using NISElement Advanced Research software (Nikon Instruments, Yokohama, Japan). The saturation (pump) beam is taken before the EOM with a 50/50 beamsplitter (UT-800-50-45UNP-CVI, CVI Melles Griot, Rochester, NY, USA) and sent through a manual delay line (PT1, Thorlabs, Newton, NJ, USA, delay up to 170 ps with 60 fs accuracy), which is used to adjust its temporal alignment with respect to pump and probe pulses. The beam is then sent to a vortex phase plate (VPP-1a, RPC Photonics, Rochester, NY, USA) to obtain the doughnut shape, and is spatially superimposed with the other beams by a polarizing beam splitter (PBS, PBS203, Thorlabs, Newton,

ACS Paragon Plus Environment

Page 13 of 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

NJ, USA). For the optimal performance the circular polarization of the doughnut beam42 is maintained at the objective plane using a pair of half- and quarter-wave plates. Pump and saturation pump laser powers are controlled by two half-wave plates, while probe power is controlled by a continuously variable neutral density filter (NDC-50C-4M, Thorlabs, Newton, NJ, USA). Two-photon excitation fluorescence (2PEF) is collected with a photomultiplier tube (PMT) on the Nikon detector unit, using a 680 nm short pass filter (FF01-680/SP-25, Semrock, Rochester, NY, USA) and open pinhole. Image analysis is performed with ImageJ/Fiji (NIH, Bethesda, MD, USA).43 Data is analyzed and graphed with Origin (OriginLab Corporation, Northampton, MA, USA). Defining system’s temporal resolution. The pump-probe signal generation is strongly determined not only by the spatial alignment of pump and probe pulses at the focus, but also by their temporal synchronization. Besides being a microscope platform, this system can be used to perform pumpprobe spectroscopy and acquire time-resolved spectra monitoring the pump-probe signal as a function of the pump-probe delay. When considering this kind of experiment the temporal resolution of the system is an important parameter to assess its performance and identify the shortest lifetime that can be detected. The temporal resolution is given by the cross-correlation of pump and probe pulses, and it can be measured using the stimulated Raman scattering (SRS) process, another pump-probe type of interaction which exploits vibrational resonances rather than transient absorption phenomena. Being a scattering process, this interaction can be considered instantaneous, and its time-resolved spectrum reflects the system temporal response. In order to excite the strong vibrational resonance of lipids around 2845 cm-1,44 an olive oil drop was imaged using pump and probe wavelengths tuned to 800 nm and 1036 nm, respectively. Monitoring the SRS signal at different time delays the temporal resolution can be retrieved with a Gaussian fit of the data (see inset in Fig. 2(g)). Sample preparation. Single layer graphene (SLG) sample was purchased from Graphene Supermarket (Graphene Laboratories Inc., Calverton, NY, USA). It consists of a monolayer graphene film grown by chemical vapor deposition (CVD) processing onto copper foil45 and then transferred onto a 0.7-mm thick glass substrate.46 The side with the deposited SLG was mounted on a 0.17-mm coverglass to be placed under a water immersion objective. The graphene film is mostly continuous, with occasional holes and cracks, and has a polycrystalline structure, made of grains with different crystallographic orientations. The sample transmission is reported by the supplier to be above 97%, which makes it suitable to collect the signal efficiently in transmission.

Acknowledgments. This work was partially funded by the European Community's Seventh Framework Programme (FP7/20012-2015) under grant agreement n° 280804 in the LANIR project framework. We thank Amira El Merhie, Silvia Dante, Antonio Esaù Del Rio Castillo (Fondazione Istituto Italiano di Tecnologia, Genova, Italy), and Fumihiro Dake (Nikon Corp., Yokohama, Japan) for the scientific discussion; Nicholas Anthony (Fondazione Istituto Italiano di Tecnologia, Genova, Italy) for proofreading the manuscript; the Nikon Imaging Center at the Fondazione Istituto Italiano di Tecnologia for help with light microscopy.

ACS Paragon Plus Environment

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 17

Supporting Information Available: Additional experimental results. This material is available free of charge via the Internet at http://pubs.acs.org.

REFERENCES (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19)

Cheng, J. X.; Yue, S. H.; Slipchenko, M. N. Multimodal Nonlinear Optical Microscopy. Laser Photon. Rev. 2011, 5, 496–512. Zipfel, W. R.; Williams, R. M.; Webb, W. W. Nonlinear Magic: Multiphoton Microscopy in the Biosciences. Nat. Biotechnol. 2003, 21, 1369–1377. Diaspro, A.; Bianchini, P.; Vicidomini, G.; Faretta, M.; Ramoino, P.; Usai, C. Multi-Photon Excitation Microscopy. Biomed. Eng. Online 2006, 5, 36. Bianchini, P.; Diaspro, A. Three-Dimensional (3D) Backward and Forward Second Harmonic Generation (SHG) Microscopy of Biological Tissues. J. Biophotonics 2008, 1, 443–450. Cheng, J. X.; Xie, X. S. Coherent Raman Scattering Microscopy; CRC Press: Boca Raton, FL, 2013. Wang, Y.; Lin, C.-Y.; Nikolaenko, A.; Raghunathan, V.; Potma, E. O. Four-Wave Mixing Microscopy of Nanostructures. Adv. Opt. Photonics 2011, 3, 1. Megerle, U.; Pugliesi, I.; Schriever, C.; Sailer, C. F.; Riedle, E. Sub-50 Fs Broadband Absorption Spectroscopy with Tunable Excitation: Putting the Analysis of Ultrafast Molecular Dynamics on Solid Ground. Appl. Phys. B Lasers Opt. 2009, 96, 215–231. Berera, R.; van Grondelle, R.; Kennis, J. T. M. Ultrafast Transient Absorption Spectroscopy: Principles and Application to Photosynthetic Systems. Photosynth. Res. 2009, 101, 105–118. Dong, C. Y.; So, P. T.; French, T.; Gratton, E. Fluorescence Lifetime Imaging by Asynchronous Pump-Probe Microscopy. Biophys. J. 1995, 69, 2234–2242. Dong, P.-T.; Cheng, J.-X. Pump – Probe Microscopy : Theory , Instrumentation , and Applications. Spectroscopy 2017, 32, 2–11. Chen, A. J.; Yuan, X.; Li, J.; Dong, P.; Hamza, I.; Cheng, J.-X. Label-Free Imaging of Heme Dynamics in Living Organisms by Transient Absorption Microscopy. Anal. Chem. 2018, 90, 3395-3401. Dake, F.; Taki, Y. Time-Domain Fluorescence Lifetime Imaging by Nonlinear Fluorescence Microscopy Constructed of a Pump-Probe Setup with Two-Wavelength Laser Pulses. Appl. Opt. 2018, 57, 757. Guo, Z.; Manser, J. S. S.; Wan, Y.; Kamat, P. V. V.; Huang, L. Spatial and Temporal Imaging of Long-Range Charge Transport in Perovskite Thin Films by Ultrafast Microscopy. Nat. Commun. 2015, 6, 1–8. Huang, L.; Hartland, G. V.; Chu, L. Q.; Luxmi; Feenstra, R. M.; Lian, C.; Tahy, K.; Xing, H. Ultrafast Transient Absorption Microscopy Studies of Carrier Dynamics in Epitaxial Graphene. Nano Lett. 2010, 10, 1308–1313. Gao, B.; Hartland, G.; Fang, T.; Kelly, M.; Jena, D.; Xing, H.; Huang, L. Studies of Intrinsic Hot Phonon Dynamics in Suspended Graphene by Transient Absorption Microscopy. Nano Lett. 2011, 11, 3184–3189. Matthews, T. E.; Piletic, I. R.; Selim, M. A.; Simpson, M. J.; Warren, W. S. Pump-Probe Imaging Differentiates Melanoma from Melanocytic Nevi. Sci. Transl. Med. 2011, 3. Fu, D.; Ye, T.; Matthews, T. E.; Yurtsever, G.; Warren, W. S. Two-Color, Two-Photon, and Excited-State Absorption Microscopy. J. Biomed. Opt. 2007, 12, 054004. Fu, D.; Matthews, T. E.; Ye, T.; Piletic, I. R.; Warren, W. S. Label-Free In Vivo Optical Imaging of Microvasculature and Oxygenation Level. J. Biomed. Opt. 2008, 13, 040503. Chen, T.; Huang, Y. Label-Free Transient Absorption Microscopy for Red Blood Cell Flow Velocity Measurement In Vivo. Anal. Chem. 2017, 89, 10120–10123. ACS Paragon Plus Environment

Page 15 of 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(20) (21) (22) (23) (24) (25) (26) (27) (28) (29) (30) (31) (32) (33) (34) (35) (36) (37) (38) (39)

ACS Nano

Chen, T.; Lu, F.; Streets, A. M.; Fei, P.; Quan, J.; Huang, Y. Optical Imaging of NonFluorescent Nanodiamonds in Live Cells Using Transient Absorption Microscopy. Nanoscale 2013, 5, 4701–4705. Chen, T.; Chen, S.; Zhou, J.; Liang, D.; Chen, X.; Huang, Y. Transient Absorption Microscopy of Gold Nanorods as Spectrally Orthogonal Labels in Live Cells. Nanoscale 2014, 6, 10536– 10539. Tong, L.; Liu, Y.; Dolash, B. D.; Jung, Y.; Slipchenko, M. N.; Bergstrom, D. E.; Cheng, J.-X. Label-Free Imaging of Semiconducting and Metallic Carbon Nanotubes in Cells and Mice Using Transient Absorption Microscopy. Nat. Nanotechnol. 2012, 7, 56–61. Liu, J.; Irudayaraj, J. M. K. Non-Fluorescent Quantification of Single MRNA with Transient Absorption Microscopy. Nanoscale 2016, 8, 19242–19248. Li, J.; Zhang, W.; Chung, T.; Slipchenko, M. N.; Chen, Y. P.; Cheng, J.-X.; Yang, C. Highly Sensitive Transient Absorption Imaging of Graphene and Graphene Oxide in Living Cells and Circulating Blood. Sci. Rep. 2015, 5, 12394. Abbe, E. Beiträge Zur Theorie Des Mikroskops Und Der Mikroskopischen Wahrnehmung. Arch. für Mikroskopische Anat. 1873, 9, 413–418. Korobchevskaya, K.; Peres, C.; Li, Z.; Antipov, A.; Sheppard, C. J. R.; Diaspro, A.; Bianchini, P. Intensity Weighted Subtraction Microscopy Approach for Image Contrast and Resolution Enhancement. Sci. Rep. 2016, 6, 25816. Liu, N.; Kumbham, M.; Pita, I.; Guo, Y.; Bianchini, P.; Diaspro, A.; Tofail, S. A. M.; Peremans, A.; Silien, C. Far-Field Subdiffraction Imaging of Semiconductors Using Nonlinear Transient Absorption Differential Microscopy. ACS Photonics 2016, 3, 478–485. Fujita, K.; Kobayashi, M.; Kawano, S.; Yamanaka, M.; Kawata, S. High-Resolution Confocal Microscopy by Saturated Excitation of Fluorescence. Phys. Rev. Lett. 2007, 99, 1–4. Massaro, E. S.; Hill, A. H.; Grumstrup, E. M. Super-Resolution Structured Pump-Probe Microscopy. ACS Photonics 2016, 3, 501–506. HELL, S. W.; WICHMANN, J. Breaking the Diffraction Resolution Limit By StimulatedEmission - Stimulated-Emission-Depletion Fluorescence Microscopy. Opt. Lett. 1994, 19, 780–782. Bianchini, P.; Peres, C.; Oneto, M.; Galiani, S.; Vicidomini, G.; Diaspro, A. STED Nanoscopy: A Glimpse into the Future. Cell Tissue Res. 2015, 360, 143–150. Silien, C.; Liu, N.; Hendaoui, N.; Tofail, S. a M.; Peremans, A. A Framework for Far-Field Infrared Absorption Microscopy beyond the Diffraction Limit. Opt. Express 2012, 20, 29694– 29704. Bianchini, P.; Korobchevskaya, K.; Zanini, G.; Diaspro, A. Pump-Probe Nanoscopy by Means of Transient Absorption Saturation. In 2018 20th International Conference on Transparent Optical Networks (ICTON); IEEE, 2018; pp 1–4. Wang, P.; Slipchenko, M. N.; Mitchell, J.; Yang, C.; Potma, E. O.; Xu, X.; Cheng, J.-X. X. Far-Field Imaging of Non-Fluorescent Species with Subdiffraction Resolution. Nat. Photonics 2013, 7, 449–453. Nair, R. R.; Blake, P.; Grigorenko, A. N.; Novoselov, K. S.; Booth, T. J.; Stauber, T.; Peres, N. M. R.; Geim, A. K. Fine Structure Constant Defines Visual Transparency of Graphene. Science. 2008, 320, 1308. Mak, K. F.; Shan, J.; Heinz, T. F. Seeing Many-Body Effects in Single- and Few-Layer Graphene: Observation of Two-Dimensional Saddle-Point Excitons. Phys. Rev. Lett. 2011, 106, 046401. Klimov, V. I. Spectral and Dynamical Properties of Multiexcitons in Semiconductor Nanocrystals. Annu. Rev. Phys. Chem. 2007, 58, 635–673. Vasko, F. T. Saturation of Interband Absorption in Graphene. Phys. Rev. B - Condens. Matter Mater. Phys. 2010, 82, 1–6. Bao, Q.; Zhang, H.; Ni, Z.; Wang, Y.; Polavarapu, L.; Shen, Z.; Xu, Q. H.; Tang, D.; Loh, K. ACS Paragon Plus Environment

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(40) (41) (42) (43)

(44) (45) (46)

Page 16 of 17

P. Monolayer Graphene as a Saturable Absorber in a Mode-Locked Laser. Nano Res. 2011, 4, 297–307. Dawlaty, J. M.; Shivaraman, S.; Chandrashekhar, M.; Rana, F.; Spencer, M. G. Measurement of Ultrafast Carrier Dynamics in Epitaxial Graphene. Appl. Phys. Lett. 2008, 92, 042116. Betin, A. A.; Pasmanik, G. A.; Piskunova, L. V. Stimulated Raman Scattering of Light Beams under Saturation Conditions. Sov. J. Quantum Electron. 1975, 5, 1309–1313. Galiani, S.; Harke, B.; Vicidomini, G.; Lignani, G.; Benfenati, F.; Diaspro, A.; Bianchini, P. Strategies to Maximize the Performance of a STED Microscope. Opt. Express 2012, 20, 7362. Schindelin, J.; Arganda-Carreras, I.; Frise, E.; Kaynig, V.; Longair, M.; Pietzsch, T.; Preibisch, S.; Rueden, C.; Saalfeld, S.; Schmid, B.; Tinevez, J. Y.; White, D. J.; Hartenstein, V.; Eliceiri, K.; Tomancak, P.; Cardona, A. Fiji: An Open-Source Platform for Biological-Image Analysis. Nature Methods 2012, 9, 676-682. D’Arco, A.; Ferrara, M. A.; Indolfi, M.; Tufano, V.; Sirleto, L. Label-Free Imaging of Small Lipid Droplets by Femtosecond-Stimulated Raman Scattering Microscopy. J. Nonlinear Opt. Phys. Mater. 2017, 26, 1750052. Li, X.; Cai, W.; An, J.; Kim, S.; Nah, J.; Yang, D.; Piner, R.; Velamakanni, A.; Jung, I.; Tutuc, E.; Banerjee, S. K.; Colombo, L.; Ruoff, R. S. Large-Area Synthesis of High-Quality and Uniform Graphene Films on Copper Foils. Science 2009, 324, 1312–1314. Li, X.; Zhu, Y.; Cai, W.; Borysiak, M.; Han, B.; Chen, D.; Piner, R. D.; Colombo, L.; Ruoff, R. S. Transfer of Large-Area Graphene Films for High-Performance Transparent Conductive Electrodes. Nano Lett. 2009, 9, 4359–4363.

ACS Paragon Plus Environment

Page 17 of 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

For table of contents only

ACS Paragon Plus Environment