Adapted from Ref. 32. (a) Front cross-sectional view of the plasmonic structure considered for the experimentation (not to scale). The inset shows the coordinate system with the +z-axis coming out of the page. (b) Top view. The lengths of the waveguide and the amplifier section are l and l a , respectively. The pump polarization is indicated by the arrow.
Experimental setup configuration. Broken paths: free-space beam; solid paths: fiber guided. AOM: acousto-optic modulator; APD: avalanche photodetector; BS: 10 dB beam splitter; F: long-pass filter; IRC: infrared camera; L1, L3, L4, and L5: microscope objective lenses; L2: cylindrical lens; M1, M2, and M3: silver mirrors; MC: monochromator; PBS1 and PBS2: polarization beam splitters; PMC: polarization maintaining coupler; PMF: polarization maintaining fiber; PS: plasmonic structure; PS1 and PS2: CW low-power sensors; SW1 and SW2: beam bankers; SW3: fiber-coupling/free-space-coupling selection switch; TEC: thermoelectric cooler; VS: variable slit; WP1 and WP2: λ/2 wave-plates.
Top view of plasmonic structure illustrating (a) end-fire coupling of probe light into the LRSPP using optical fiber and lens and (b) end-fire coupling to the LRSPP using two optical fibers. The image on the far right of (a) shows the LRSPP mode intensity profile at the output facet as captured by the infrared camera.
Stability of insertion loss measurement conducted on the plasmonic structure under consideration using dye solvent (without dye molecules) as upper cladding.
Time averaged intensity distribution at the plasmonic structure's output facet while being pumped as captured by the infrared camera.
Probe on (curves on right) and probe off (curves on left) measurements conducted on the plasmonic structure under consideration for cases (a) where the probe signal is coupled into and out of the LRSPP as shown in Fig. 3(b) and (b) where the waveguide is shifted ∼5μm along the x-axis keeping the fibers and the pump with the original alignment.
The LRSPP mode power gain coefficient as calculated from (a) LRSPP probe stimulated emission measurements and (b) the ASE-LRSPP measurements. The goodness of each fit as given by the Pearson product-moment correlation coefficient, R 2, is indicated. For both cases, amplitude of the error bars corresponds to two times the mean squared error of the fit. Inset: Absolute power measurement of the ASE-LRSPP spectrum obtained with l a = l; its peak is located at λ = 876 nm and exhibits a FWHM spectral linewidth of 8 nm.
Top panel: Mode power gain coefficient obtained via stimulated emission (γSEM, solid circles) and via ASE-LRSPP (γASE, open circles) for a number of experiments. The error bar for γSEM is about the size of the symbol. Center panel: Δγ = γSEM − γASE. Bottom panel: Insertion loss measured in situ for each experiment. For all cases, the amplitude of the error bars correspond to two times the mean squared error of the fit.
Gain and noise measurements of a LRSPP amplifier. Top panels: Amplifier gain, G a , as a function of its length, l a , obtained from LRSPP probe stimulated emission measurements. The dashed lines are fits to the linear gain regions. Bottom panels: Power at the output fiber, P o , for stimulated emission and ASE-LRSPP measurements; probe-on measurements (circles), probe-off measurements (crosses), probe-on minus probe-off (dots). The dashed curves are NLS fits to Eq. (5). The measurements shown in (a)(d), correspond to measurements 1, 2, 7, and 8 in Fig. 8, respectively.
Article metrics loading...
Full text loading...